Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
EAC Agenda 09/01/2010
ENVIRONMENTAL ADVISORY COUNCIL AGENDA SEPTEMBER 1, 2010 COLLIER COUNTY ENVIRONMENTAL ADVISORY COUNCIL AGENDA SEPTEMBER 1, 2010 9:00 A.M. Commission Boardroom W. Harmon Turner Building (Building "F ") — Third Floor I. Call to Order II. Roll Call III. Approval of Agenda IV. Approval of August 4, 2010 meeting minutes V. Upcoming Environmental Advisory Council Absences VI. Land Use Petitions NONE VII. New Business A. Watershed Management Plan Performance Measures and Alternative Evaluation VIII. Old Business A. Update members on projects IX Subcommittee Reports X. Council Member Comments XI. Staff Comments XII. Public Comments XIII. Adjournment Council Members: Please notify Summer Arague, Environmental Services Senior Environmental Specialist no later than 5:00 p.m on Friday, August 27 if you cannot attend this meeting or if you have a conflict and will abstain from voting on a petition (252- 6290). General Public: Any person who decides to appeal a decision of this Board will need a record of the proceedings pertaining thereto; and therefore may need to ensure that a verbatim record of proceedings is made, which record includes the testimony and evidence upon which the appeal is to be based. :1 Technical Memorandum To: Mac Hatcher, PM Collier County From: Moris Cabezas, PBS &J Dave Tomasko, PBS &J Emily Keenan, PBS &J Date: July 9, 2010 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 1, Task 1.2: In- Stream Water Quality 1.0 Objective This Technical Memorandum addresses Element 1, Task 1.2: In- Stream Water Quality. The objective of this task is to characterize the water quality conditions of the following watersheds within Collier County: Cocohatchee- Corkscrew, Golden Gate - Naples Bay, Rookery Bay, Faka Union, Okaloacochee - SR29, Fakahatchee, Marco Island and Naples (Figure 1 -1). This effort focused on characterizing the water quality within Collier County's priority watersheds in the context of the Total Maximum Daily Load (TMDL) impairment conditions as described in the Florida Department of Environmental Protection's (FDEP) verified list of impaired waters. 2.0 Introduction The analysis conducted as part of this project included the following: 1) review of relevant reports from local, regional and state agencies related to water quality conditions, 2) review of relevant water quality data for Collier County's watersheds, 3) an assessment of locations where water quality "impairment" may be a natural condition, 4) determination of the factor(s) likely to be responsible for impairment, 5) determination of factors likely to be responsible for phytoplankton growth, 6) development of recommendations about proposed site - specific water quality criteria, and 7) a conceptual overview of factor(s) that most strongly influence water quality in Collier County's priority watersheds. The reports reviewed to identify impaired and potential waters of concern within Collier County included the water quality impairment analysis completed as part of the FDEP TMDL program implementation and the analysis of water quality conditions conducted by the United States Army Corps of Engineers (USACE) as part of the Southwest Florida Feasibility Study (SWFFS). Results of analyses conducted as part of those studies are presented below. 1 Collier County Watershed Model Update I's and Plan Development July 9, 2010 Figure 1 -1. Collier County Watersheds HENCiRY CCU Cocohatchee- Corkscrew �Y _ e OkaloacocheeSR29 LEE CO. Golden Gate Naples Bay Naples Fakahatchee Faka Union 1 Rookery Bay J Y ff ✓�a� y� Marco id and „ Legend Watershed Boundaryt 8 .3� COLLIER CO. 0 r County Boundary <ti3y� K 0.40hJR0E CO pNp 0 2 4 Miles x N 2 Collier County Watershed Model Update and Plan Development July 9, 2010 2.1. The FDEP TMDL Impairment Analysis For implementation of the statewide TMDL program, the FDEP divided the state into five groups. Each group is comprised of multiple basins. All water bodies within Collier County are in the Everglades West Coast Group 1 Basin. Per TMDL guidelines, every five years (cycle) each WBID is evaluated to determine if available data indicate that water quality parameters exceed the limits defined by FDEP in the Impaired Waters Rule JWR). The list of verified impaired WBIDs for each group and cycle is available from the FDEP website. After the compilation of all impaired WBIDs from Cycles 1 and 2, a total of fifteen impairments have been designated by FDEP in the freshwater discharge portion of the study area. The freshwater discharge WBIDs of concern in the study area are listed in Table 2 -1. The water quality impairment parameters include dissolved oxygen, nutrients, fecal coliform bacteria, iron, un- ionized ammonia, and mercury (Figures 2 -1 to 2 -6). The majority of impairments (9 of 15) were due to low dissolved oxygen concentrations, which was observed mostly in the Cocohatchee- Corkscrew watershed and also in the Golden Gate- Naples Bay and Okaloacochee -SR29 watersheds. Nutrients, un- ionized ammonia and mercury have been considered impairment parameters in WBID 3259W (Lake Trafford) within the Cocohatchee- Corkscrew watershed. Presently, large -scale restoration projects including sediment removal are underway to improve water quality in Lake Trafford. As such, the current water quality conditions may not reflect the impaired water quality status identified by FDEP. Only WBID 3278G (Fakahatchee Strand) was identified as impaired for fecal coliform concentrations. No water quality impairments were identified by the FDEP TMDL program in the freshwater portion of the Rookery Bay or Faka Union watersheds. 3 Collier County Watershed Model Update ITS] and Plan Development July 9, 2010 Table 2 -1. List of FDEP Impaired Waters from Group 1 Cycles 1 and 2 for the freshwater discharge WBIDs of each watershed W610# W13 "" Mo jmoo» w' Pa m ier Water h6d 3259W Lake Trafford Dissolved Oxygen Coco hatchee-Corkscrew 3259W Lake Trafford Mercury Coco hatchee- Corkscrew 3259W Lake Trafford Nutrients Coco hatchee-Corkscrew 3259W Lake Trafford Un- ionized Ammonia Coco hatchee-Corkscrew 3278D Cocohatchee Inland Dissolved Oxygen Coco hatchee - Corkscrew 3278F Corkscrew Marsh Dissolved Oxygen Cocohatchee- Corkscrew 3278L Immokalee Basin Dissolved Oxygen Coco hatchee- Corkscrew 3278K Gordon River Extension Dissolved Oxygen Golden Gate - Naples Bay 3278S North Golden Gate Dissolved Oxygen Golden Gate - Naples Bay 3278S North Golden Gate Iron Golden Gate - Naples Bay 3278G Fakahatchee Strand Dissolved Oxygen Fakahatchee 3278G Fakahatchee Strand Fecal Coliform Fakahatchee 3261 C Barron River Canal Iron Okaloacochee -SR29 3278T Okaloacoochee Dissolved Oxygen Okaloacochee -SR29 3278W Silver Strand Dissolved Oxygen Okaloacochee -SR29 4 Collier County Watershed Model Update ITsi and Plan Development July 9, 2010 Figure 2 -1. WBIDs within priority watersheds that were verified impaired for Dissolved Oxygen by FDEP Quality of Discharge Dissolved Oxygen Group 1- Cycles 1 and 2 b y O f✓ LEE CO Wim Legend ` WBID C3 Imnair.d c3 Watershed Boundary Sub -Basin Boundary County Boundary �n 0 2 4 Miles 3259E r tlz �a F ,259= 1 dh a ' COLLIER CO MONROE CC) 5 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 2 -2. WBIDs within priority watersheds that were verified impaired for Nutrients by FDEP Quality of Discharge Nutrients Group 1- Cycles 1 and 2 W LEE CO. 3278p OM Legend WBiD ILI impaired `? WBID Boundary OWatershed Boundary County Boundary 0 2 4 Miles 11007 3 1� 32590 Ems' 0 HENDRY CC) 3278E 3278T 32594: 32781. 3.784, 82591 3278H ~11 / 3,ftC 3 ?76G i 77-C o LJER CO. A4ONRC'E CO N c� JJ /,VV 6 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 2 -3. WBIDs within priority watersheds that were verified impaired for Fecal Coliform Bacteria by FDEP Quality of Discharge Fecal Coliform Group 1- Cycles 1 and 2 72sas 3Zi8F LEE CO. 325BZ .....,..,.J 3T78D ','785 32'Lt I., 3278 Y` 3278'v Legend , WBID C-3 Impaired WBID Boundary Watershed Boundary County Boundary 0 2 4 Miles HENDPY CO 32,aE 7378T 3 "78L 3278w 7 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 2 -4. WBIDs within priority watersheds that were verified impaired for Iron by FDEP 8 Collier County Watershed Model Update ' and Plan Development July 9, 2010 Figure 2 -5. WBIDs within priority watersheds that were verified impaired for Un- ionized Ammonia by FDEP Quality of Discharge Ili Un- ionized Ammonia HENDRY CO. Group 1- Cycles 1 and 2 i 25eB 278E 327eT 3z7sL 3zrx+� t LEE CO. 325 t 3278E 3278D 'f2591 32785 3278H 3278K �5 3278Y 3278V �t 32781 .3278E Legend WBID 1y COLLIER CO. Impaired +' M �) , W WBID Boundary Watershed Boundary e � r` 3 "6 County Boundary MONROE CO. N 0 2 4 Miles A M11 9 Collier County Watershed Model Update Vj and Plan Development JF July 9, 2010 Figure 2-6. WBIDs within priority watersheds that were verified impaired for Mercury by FDEP Quality of Discharge Mercury Group I- Cycles I and 2 3279F LEE CO. R� Legend WBID C3 Impaired C3VVBID Boundary 0 Watershed Boundary County Boundary F--7----1 0 2 4 Miles HENDRY C,',-) ,327SE 3178T 3 ".19 W 32 TBH 1278I I 3278G I -,COLLIER CC) 10 Collier County Watershed Model Update and Plan Development PBS1 July 9, 2010 2.2. The Southwest Florida Feasibility Study Water Quality Evaluation The SWFFS water quality analysis was conducted in 2004 by Tetra Tech, Inc. and Janicki Environmental, Inc. The report was entitled "Compilation, Evaluation, and Archiving of Existing Water Quality Data for Southwest Florida" to the USACE. Task 7 of that report focused on the identification of waters of concern within the SWFFS area using a modification of the IWR. The boundaries of the watersheds reviewed for that report include the Collier County area included in the current study. WBID boundaries were not considered for their analysis. The SWFFS identified a total of 318 parameter- specific waters of potential concern and 296 waters of verified concern. Figures 2 -7 to 2 -11 show the location of potential waters of concern by parameter. Consistent with FDEP's evaluation, dissolved oxygen is the dominant parameter of concern. All of the priority watersheds in Collier County were identified as potential waters of concern for dissolved oxygen even those with little development pressure. Additionally, fecal coliforms, un- ionized ammonia, and iron were reported as elevated in the majority of watersheds. Discrepancies were found between FDEP impairment analysis and SWFSS evaluation. They are likely due to the variations in water quality databases, spatial scale of analysis (WBID vs. watershed) and the type of analysis JWR vs. modifications to the IWR). Figure 2 -7. Potential Waters of Concern for Dissolved Oxygen as determined by Tetra Tech, Inc and Janicki Environmental, Inc (2004). SWFFS Waters of Potential Concern ` DISSOLVED OXYGEN . (............� W aten of Pawtia Corer, fur Dissolved Cuygen S NFFS Watarirody 5egments J Emphas aed C—Ay Boundary Florida County Boundary 11 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 2'8. Potential Waters of Concern for Nutrients as determined by Tetra Tech, Inc and Janioki Environmental, Inc (2OO4). -- --- SWFFS Waters "If PoterltvM Concern NUTRIEUTS - IX Figure 2'8. Potential Waters of Concern for Fecal Co|ifmrrn as determined bv Tetra Tech, Inc and JanioW Environmental, Inc (DDO4). arm "" =�o ` 12 Collier County Watershed Model Update � and Plan Development July 9.2O1O Figure 2 -10. Potential Waters of Concern for Iron as determined by Tetra Tech, Inc and Janicki Environmental, Inc (2004). i SWFFS Waters of Potential Concern t { CL��R s Q �• inn SWF:S`N.wt od' Srprn u HI zed (, tY 6.undary �� Figure 2 -11. Potential Waters of Concern for Unionized Ammonia as determined by Tetra Tech, Inc and Janicki Environmental, Inc (2004). ShTFS Waura of Potential _ C—e- t ' I FM It ;M 17 Fr7 A NNI O N A r m _ _ - -,4���a � R„ l ninnworv. A mn a °,,.t .NFF �'Jt'.otcrM+rly 4ap<ne7r•; F,ph -ad 1--b 13 u+da .„.J — �i� t 3 Collier County Watershed Model Update and Plan Development rx July 9, 2010 3.0 Stream Water Quality Analysis Methodology Two methods were utilized to evaluate the potential waters of concern for each of the Collier County watersheds: a) A review of long -term water quality data was conducted to identify potential parameters of concern at the watershed level. b) A review of water quality data within each WBID was performed to compare results with FDEP impairment determinations. The data used for the analyses included the IWR Run 39 data (supplied by FDEP), as well as data from Florida STORET, Collier County, City of Naples, and the Rookery Bay National Estuarine Research Reserve (RBNERR). This resulted in an updated and comprehensive database of water quality data. All analyses were conducted using the most recent ten year time period (2000 to 2009) to minimize the effect of temporal variations. Also, it was determined that the majority of water quality data available was collected during this ten year period. To eliminate potential errors due to duplicate data entry via multiple agencies uploading the same data, median values were calculated by station, date, and parameter. To allow for a direct comparison between lab parameters (i.e., nutrients) and field parameters (i.e., temperature, dissolved oxygen) samples were restricted to those collected from less than one meter depth. Since lab parameters are typically from surface grab samples, this ensures that comparisons between various parameters are from samples taken from the same general water depth. Using GIS and the station descriptions, the locations of water quality stations were reviewed in order to identify locations where multiple stations were sampled. Data were merged when more than one water quality station was sampled at the same location and a unique merged station name was assigned to that location. Appendix A lists all water quality stations and assigned merged station names. Each parameter in the database was screened to identify outliers or entry errors due to unit inconsistencies. Identified inconsistencies were reviewed and corrected. When Total Nitrogen (TN) species were not listed, TN was calculated as the sum of Total Kjeldahl Nitrogen (TKN) and Nitrate + Nitrite (NOx). To ensure consistency with IWR guidance, corrected chlorophyll a was preferentially used over uncorrected chlorophyll a for samples collected in 2006 and earlier. After 2006, IWR guidance from FDEP is that only corrected chlorophyll a data should be used. A description of the two methods used to conduct the analyses conducted as part of this study is presented below. 14 Collier County Watershed Model Update M64 and Plan Development July 9, 2010 3.1. Watershed Analysis This analysis was conducted using only data from the long -term water quality stations consistently sampled throughout the ten -year time period 2000 -2009. The use of long -term water quality stations accommodates variability in water quality due to irregular sampling and temporary monitoring efforts. The long -term data also provided a means of evaluating the watershed as a whole, rather than characterizing it using short-term "snapshots" of water quality from individual subbasins Figure 3 -1 shows the long -term freshwater discharge water quality sampling stations used for analysis. For this analysis, summary data for each watershed were compared to the Criteria for Surface Water Quality Classifications (F.A.C. 62- 302.530) for water quality parameters based on their water body classification. Table 3 -1 shows the regulatory class for each watershed. Class is defined as the associated designated use of the water body provided by FDEP. All freshwater bodies examined here are classified as class III freshwater (3F). Table 3 -3 shows the regulatory standards for a Class 3F water body for selected parameters. Regulatory standards have been vetted by the scientific community and provide a biologically relevant basis for comparison. The FAC Chapter 62 -303: "Identification of Impaired Surface Waters ", provides a list of the minimum number of samples not meeting a water quality criterion for a range of sample sizes in order for the water to be included on the FDEP Verified list. The same criteria was used herein to classify the watersheds as a "watershed of concern" when an appropriate regulatory standard was exceeded. In terms of chemical parameters, chlorophyll a, dissolved oxygen, iron, and fecal coliforms are parameters used by FDEP to classify WBIDs as impaired water bodies. In contrast, color, total phosphorus, total nitrogen, and total suspended solids cannot be quantitative assessed to identify impaired water bodies. Total nitrogen and total phosphorus are valuable parameters providing indicators of eutrophication. Both chlorophyll a and dissolved oxygen levels can be directly impacted by the nutrient loads. Color has the ability to affect chlorophyll a and dissolved oxygen concentrations. Additionally, total suspended solids provide an indication of sediment erosion, which occurs frequently in storm water run -off. To further evaluate potential water quality impairments at the watershed level, data were compared to screening level standards, which can provide an indication of water quality concerns when no numeric state standards exist. Screening level standards are available for total nitrogen (TN) and total phosphorus (TP) based on the 701h percentile of all available data, a technique first used by Friedman and Hand (1989). Using IWR Run 39, a similar screening level was calculated by water body type for color and total suspended solids, in which the 70`h percentile of all data available from 2000 to 2009 by water body type was calculated. Table 3 -4 shows the screening level standard for selected parameters by water body type (stream or lake). 15 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 3 -1. Long term stations for watershed in- stream water quality analysis 16 Collier County Watershed Model Update PW and Plan Development July 9, 2010 3.2. Evaluation of WBID Impairment Using methods similar to MR, PBS &J analyzed the water quality data for each WBID in a watershed. As opposed to the watershed analysis that used data only for the long -term water quality stations, for this analysis it was decided that all data available for the period 2000 -2009 would be used because that is consistent with FDEP's approach for impairment evaluation. Dissolved oxygen, iron, fecal coliforms, un- ionized ammonia, and copper concentrations were compared to the appropriate state regulatory standard to determine impairment status (Table 3 -3). It should be noted that a modification to the FDEP method for determining chlorophyll a impairments was utilized. Each chlorophyll a value was compared to the state regulatory standard and the percent exceedance was calculated. In contrast, FDEP calculates an annual average using data from each quarter for comparison with the regulatory standard. The results of PBS &J analysis for each WBID within a watershed were compared to the FDEP impaired WBID list for those water bodies in the study area. Table 3 -1. WBID name and corresponding watershed designation WE3�C1.. 3259W " - 3F tat+tied. Coco hatchee- Corkscrew Wa�Q 11a LAKE TRAFFORD 3259Z 3F Cocohatchee- Corkscrew LITTLE HICKORY BAY 3278D 3F Coco hatchee- Corkscrew COCOHATCHEE (INLAND SEGMENT) 3278C 3F Coco hatchee- Corkscrew COCOHATCHEE GOLF COURSE DISCHARGE 3278F 3F Cocohatchee - Corkscrew CORKSCREW MARSH 3278E 3F Coco hatch ee - Corkscrew COW SLOUGH 3259B 3F Cocohatchee- Corkscrew DRAINAGE TO CORKSCREW 3278L 3F Cocohatchee - Corkscrew IMMOKALEE BASIN 3278H 3F Faka Union FAKA UNION (NORTH SEGMENT) 32781 3F Faka Union FAKA UNION (SOUTH SEGMENT) 3278G 3F Fakahatchee FAKAHATCHEE STRAND 32591 3F Fakahatchee CAMP KEAIS 3278K 3F Golden Gate Naples Bay GORDON RIVER EXTENSION 3278S 3F Golden Gate Naples Bay NORTH GOLDEN GATE 3261C 3F Okaloacochee -SR29 BARRON RIVER CANAL 3278T 3F Okaloacochee -SR29 OKALOACOOCHEE SLOUGH 3278W 3F Okaloacochee -SR29 SILVER STRAND 3278V 3F Rookery Bay ROOKERY BAY (INLAND EAST SEGMENT) 3278Y 3F Rookery Bay ROOKERY BAY (INLAND WEST SEGMENT) '3F: Recreation, Propagation and Maintenance of a Healthy, Well- Balanced Population of Fish and Wildlife (Predominantly Fresh Waters) lm!$f 17 Collier County Watershed Model Update and Plan Development July 9, 2010 Table 3 -2. List of Water Quality Parameters ra�stet�►r Unit Pai remoter - 1000 Salinity ppt Conductivity µmhos /cm Total Nitrogen mg /l Nitrate - Nitrite mg /I Total Phosphorus mg /I Orthophosphate mg /I Total Kjeldahl Nitrogen mg /I Unionized Ammonia mg /I Chlorophyll a µg /I Fecal Coliform # /100ml Color PCU Copper Pg /I Total Suspended Solids mg /I Turbidity NTU Dissolved Oxygen mg /I Biochemical Oxygen Demand mg /I Iron µg /I Hardness mg /I Secchi Depth m Table 3 -3. List of regulatory standards for selected water quality parameters Dissolved Oxygen (mg /1) 5 Iron (Ng /1) 1000 Fecal Coliform ( # /100ml) 400 Chlorophyll a (µg /1) 20 Copper (µg /1) a ^(0.854[InH1- 1.702) Un- ionized Ammonia (mg /1) 0.02 '3F: Recreation, Propagation and Maintenance of a Healthy, Well - Balanced Population of Fish and Wildlife (Predominantly Fresh Waters) Table 3 -4. List of screening levels for selected water quality parameters krar»eter Le#te - _ : Stream Color (PCU) 80 111.5 TSS (mg /1) 13 7 TN (mg /1) 1.7 1.6 TP (mg /I) 0.11 0.22 PW 18 Collier County Watershed Model Update and Plan Development July 9, 2010 4.0 Results and Discussion This section presents the results of both the evaluation of watershed conditions and the review of impaired WBIDs for each of the priority watersheds. In general, using the methodology described previously, five parameters were identified as parameters of concern, color, dissolved oxygen, TN, iron, and fecal coliforms. Chlorophyll a, total phosphorus, and total suspended solids concentrations were within range of the regulatory standards and screening levels for all six watersheds. While un- ionized ammonia was identified by FDEP as a parameter of concern for WBID 3259W (Lake Trafford), elevated levels of un- ionized ammonia were not observed in the Cocohatchee- Corkscrew watershed based on data from the long -term sampling stations. Based on an evaluation of fish tissue, FDEP identified Lake Trafford as impaired for mercury. Since the mercury impairment was not determined through water quality measurements, a review of mercury contamination was not conducted herein. Table 4 -1 lists the parameters of concern and the number of watersheds for which that parameter is of concern. The majority of watersheds frequently show low dissolved oxygen concentrations (Figure 4 -1). Two of the six watersheds showed elevated fecal coliform bacteria levels: Fakahatchee and Cocohatchee- Corkscrew (Figure 4 -2). The Okaloacochee /SR29 watershed had elevated total nitrogen concentrations (Figure 4 -3). Only the Rookery Bay watershed was not identified as having elevated color (Figure 4 -4). Data from a number of watersheds indicated elevated iron concentrations (Figure 4 -5). Table 4 -1. Total number of Watersheds of Concern identified for each parameter Paiiin�r Watersh8d of Cgncern Chlorophyll a 0 Color 5 Dissolved Oxygen g Fecal coliform 2 Iron 2 Total Nitrogen 1 Total Phosphorus 0 Total Suspended Solids 0 Un- ionized Ammonia 0 Mercury Unable to evaluate Description of the analysis results of water quality conditions by watershed are presented below. pftsi 19 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 4 -1. Watershed of Concern for Dissolved Oxygen In- Stream Water Quality Dissolved Oxygen I I LEE CG Legend Watershed of Concern 0 Watershed Boundary _I I County Boundary m 0 1.5 3 Miles HENDRY CO. /alas COLLIER CCCI 20 Collier County Watershed Model Update M91 and Plan Development July 9, 2010 Figure 4 -2. Watershed of Concern for Fecal Coliform Bacteria In- Stream Water Quality Bacteria (Fecal Coliform)l HENDRYCO Golden Gate Naples Bay Legend Watershed of Concern I 0 Watershed Boundary L� ? County Boundary r—_T_7 0 1.5 3 Miles Rookery Bay /arya C't7LLIER CO. N 21 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 4-3. Watershed of Concern for Total Nitrogen Golden Gate Naples Bay Legend 0 "1 Watershed of Concern 0 Watershed Boundary County Boundary 0 1.5 3 Miles Fakahatchea Rookery Say Faka Union 0 22 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 4 -4. Watershed of Concern for Color Legend 1 "�"✓ Watershed of Concern 6� rBt Vvatershed Boundary �Sah COLLIER C-0 County Boundarys 0 1.5 3 Miles ro + MC =NROfi CO 23 Collier County Watershed Model Update ff and Plan Development jo July 9, 2010 fakahaic hee Rookery Bay Faka Union Legend 1 "�"✓ Watershed of Concern 6� rBt Vvatershed Boundary �Sah COLLIER C-0 County Boundarys 0 1.5 3 Miles ro + MC =NROfi CO 23 Collier County Watershed Model Update ff and Plan Development jo July 9, 2010 Figure 4 -5. Watershed of Concern for Iron In- Stream Water Quality Iron h.. _ Cocohatchee- Corkscrew 24 Collier County Watershed Model Update and Plan Development July 9, 2010 4.1. Coco hatc hee-Co rkscrew Following is a description of the results of the watershed analysis and WBID impairment condition. 4.1.1. Watershed Analysis The summary statistics for the Cocohatchee- Corkscrew watershed are provided in Table 4 -2. Based upon the evaluation of the long term stations within the watershed, three potential parameters of concern were identified: color, dissolved oxygen, and fecal coliform bacteria. Chlorophyll a and nutrients were not elevated in the Cocohatchee- Corkscrew watershed. While mercury as a parameter of concern by FDEP in Lake Trafford (WBID 3259W), the impairment was determined based on fish tissue concentrations not water quality data. An analysis of mercury was not completed. 4.1.1.1. Chlorophyll a and Nutrients FDEP declared Lake Trafford (WBID 3259W) impaired for both chlorophyll a (nutrients) and un- ionized ammonia. However, elevated chlorophyll a and nutrient values were not observed on the watershed - scale. It is important to note that due to the poor water quality conditions observed in Lake Trafford by FDEP, Collier County, and the SFWMD, a large -scale restoration project is underway. Water quality conditions that existed and created impairment are complex and no longer exist as they did during the timeframe used for the Lake Trafford TMDL report. Phase I of the Lake Trafford sediment removal project has been completed and observed water quality improvements have been documented (i.e., PBS &J 2009). Phase II of the Lake Trafford sediment removal project is currently underway, and further water quality improvements are possible. Much of the data for Lake Trafford is from the period prior to implementation of Phase I of the Lake Trafford sediment removal project and may not accurately represent post - project conditions. Therefore, the impairments indicated by FDEP (un- ionized ammonia and chlorophyll a) are likely not indicative of current water quality conditions in the Lake. 4.1.1.2. Dissolved Oxygen The majority of the Cocohatchee- Corkscrew watershed is comprised of natural areas (47 %) predominantly due to the Corkscrew Swamp. Increased color and decreased dissolved oxygen values have previously been documented from forested wetlands in Florida (PBSJ 2009). It is likely that the low dissolved oxygen concentrations are a function of natural seasonal fluctuations which occur in wetland environments not due to anthropogenic modifications of the landscape. This topic will be discussed in greater detail in section 4.2, and it holds particular relevance for all the priority watersheds within Collier County. 25 Collier County Watershed Model Update M"51 and Plan Development July 9, 2010 Table 4 -2. Water Quality Summary Statistics for the Coco hatchee-Corkscrew Watershed indicating potential parameters of concern tera BOD, mg/1 125 Nlin qii 1.6 I 1 " 2.4 �i 2.0 ;CUlex 6.8 j� 1 F%r ncom Chlorophyll-a, u /I 449 3.0 9.1 4.3 246.3 9 N Color, PCU 437 5 72 60 300 12 Y Conductivity, umhos /cm 495 317 6718 1064 49624 Copper, u /1 153 0.30 5.28 2.25 178.00 Dissolved Oxygen, mg/1 499 0.42 5.45 5.10 16.74 47 Y Fecal Coliform, # /100ml 442 1 259 88 4500 13 Y Iron, u /1 140 100.0 352.3 325.0 1100.0 1 N Nitrate - Nitrite, m /I 435 0.01 0.08 0.05 0.78 Orthophosphate as P, m /I 347 0.004 0.042 0.024 0.290 Salinity, ppt 447 0.2 4.2 0.5 32.4 Secchi Depth, m 471 0.10 1.16 1.10 2.50 TKN, mg/1 394 0.05 0.86 0.83 4.30 Total Nitrogen, m /I 407 0.005 0.790 0.860 4.300 3 N Total Phosphorus, m /I 428 0.004 0.079 0.055 0.563 6 N TSS, m /I 365 1.3 4.5 2.0 102.0 8 N Turbidity, NTU 276 0.4 2.4 1.8 24.0 Unionized Ammonia, m /I 1 266 1 0.0000 0.0013 1 0.0010 1 0.0082 0 N Based upon the applicable dissolved oxygen criteria, in the Cocohatchee- Corkscrew watershed (and elsewhere in Collier County; see later sections) dissolved oxygen concentration levels were consistently below the regulatory standard of 5.0 mg /1 for freshwater water bodies. An evaluation of the causative factor for these low dissolved oxygen levels was completed to determine the factor(s) likely responsible for depressed dissolved oxygen values. The decomposition of detritus, associated with color, and phytoplankton biomass, associated with TN and TP, can affect levels of dissolved oxygen. However, low dissolved oxygen concentrations due to high levels of color (aka. tannins) can be an entirely natural phenomenon, whereas low levels of dissolved oxygen due to elevated nutrient levels can be indicative of anthropogenic nutrient loads. It is therefore important to determine the most likely causative factor, tannin levels or nutrients, that explain the low dissolved oxygen concentrations. Based upon regressions between dissolved oxygen and TN, TP and color, the primary causative factor for the low dissolved oxygen discharge in the watershed was identified as color (Table 4- 3). For each regression the best -fit curve was selected when comparing exponential, linear, and power relationships. The identification of color as the primary causative factor supports the finding that low dissolved oxygen concentrations are commonly found in the portion of the watershed that includes a large amount of undeveloped land, the Corkscrew Swamp. A site specific alternative criteria (SSAC) for dissolved oxygen is highly recommended for the Cocohatchee- Corkscrew watershed. 26 Collier County Watershed Model Update M$l and Plan Development July 9, 2010 Table 4 -3. Identification of causative factor in the Cocohatchee- Corkscrew watershed for low dissolved oxygen values Using the Fakahatchee Strand as the reference watershed, the proposed dissolved oxygen standards for the Cocohatchee- Corkscrew watershed (as well as other Collier County watersheds) would reflect the wet and dry season median dissolved oxygen values (4.3 and 2.1 mg /L, respectively). However, "violation" of these proposed alternative DO standards does not necessarily denote impaired water quality, but could be used as guidance levels below which the possibility exists that DO levels are lower than expected, based on natural influences. A more detailed discussion of the dissolved oxygen concentration issues in Collier County is included in Appendix B. A guidance document for developing a DO SSAC is included in Appendix C. 4.1.1.3. Fecal Coliform Bacteria The numeric criteria of fecal coliform bacteria concentrations for Class 3 waters as established by Rule 62 -302, F.A.C., states that "The Most Probable Number (MPN) shall not exceed a monthly average of 200, nor exceed 400 in 10% of the samples, nor exceed 800 on any one day. " No WBIDs were identified as verified impaired for bacteria as evaluated by FDEP. In contrast, the Cocohatchee- Corkscrew watershed reported 13% of the 442 values exceeded the 400 # /100mL criteria established for Class 3 waters. As such, the watershed was classified as a "watershed of concern" for bacteria based on the analysis of the long -term sampling stations. Though values exceed the regulatory standard for Class 3 water bodies, fecal coliform bacteria may not be an appropriate indicator for pathogenic diseases in sub - tropical climates. In subtropical environments such as South Florida, the specificity of the fecal coliform test is compromised by the more constant and warmer ambient water temperatures of sampled water bodies. The inability to specifically identify humans as a source of bacteria using traditional indicator bacteria testing protocols has been noted by Fujioka (2001) and Fujioka et al. (1999) for various tropical locations. Further source identification efforts are warranted to verify the impairment. 4.1.2. Evaluation of WBID Impairment Using all of the available water quality data over the 10 -year period for each WBID, PBS &J evaluated the impairment status determined by FDEP for parameters identified in the Cocohatchee- Corkscrew watershed. Table 4 -4 shows the FDEP impairment as well as the results of the PBS &J analysis. As shown all impairments were confirmed when compared to the State standards. Additionally, six potential impairments were identified. At this time, FDEP has not verified impairments within the Cow Slough, Corkscrew Marsh, Cocohatchee (Inland Segment) or the Drainage to Corkscrew water bodies. The review of mercury impairment in Lake Trafford 27 Collier County Watershed Model Update 1V$1 and Plan Development July 9, 2010 � # . .. Color Exponential regression 0.000 0.08 TN Power regression 0.008 0.02 TP Power regression 0.000 0.07 Using the Fakahatchee Strand as the reference watershed, the proposed dissolved oxygen standards for the Cocohatchee- Corkscrew watershed (as well as other Collier County watersheds) would reflect the wet and dry season median dissolved oxygen values (4.3 and 2.1 mg /L, respectively). However, "violation" of these proposed alternative DO standards does not necessarily denote impaired water quality, but could be used as guidance levels below which the possibility exists that DO levels are lower than expected, based on natural influences. A more detailed discussion of the dissolved oxygen concentration issues in Collier County is included in Appendix B. A guidance document for developing a DO SSAC is included in Appendix C. 4.1.1.3. Fecal Coliform Bacteria The numeric criteria of fecal coliform bacteria concentrations for Class 3 waters as established by Rule 62 -302, F.A.C., states that "The Most Probable Number (MPN) shall not exceed a monthly average of 200, nor exceed 400 in 10% of the samples, nor exceed 800 on any one day. " No WBIDs were identified as verified impaired for bacteria as evaluated by FDEP. In contrast, the Cocohatchee- Corkscrew watershed reported 13% of the 442 values exceeded the 400 # /100mL criteria established for Class 3 waters. As such, the watershed was classified as a "watershed of concern" for bacteria based on the analysis of the long -term sampling stations. Though values exceed the regulatory standard for Class 3 water bodies, fecal coliform bacteria may not be an appropriate indicator for pathogenic diseases in sub - tropical climates. In subtropical environments such as South Florida, the specificity of the fecal coliform test is compromised by the more constant and warmer ambient water temperatures of sampled water bodies. The inability to specifically identify humans as a source of bacteria using traditional indicator bacteria testing protocols has been noted by Fujioka (2001) and Fujioka et al. (1999) for various tropical locations. Further source identification efforts are warranted to verify the impairment. 4.1.2. Evaluation of WBID Impairment Using all of the available water quality data over the 10 -year period for each WBID, PBS &J evaluated the impairment status determined by FDEP for parameters identified in the Cocohatchee- Corkscrew watershed. Table 4 -4 shows the FDEP impairment as well as the results of the PBS &J analysis. As shown all impairments were confirmed when compared to the State standards. Additionally, six potential impairments were identified. At this time, FDEP has not verified impairments within the Cow Slough, Corkscrew Marsh, Cocohatchee (Inland Segment) or the Drainage to Corkscrew water bodies. The review of mercury impairment in Lake Trafford 27 Collier County Watershed Model Update 1V$1 and Plan Development July 9, 2010 was not conducted because it was based on fish tissue. It should be noted that the evaluation of the WBID impairment provided similar results to the long -term station watershed analysis by which the Cocohatchee- Corkscrew watershed was identified as having two parameters of concern, dissolved oxygen and fecal coliforms. As indicated previously, both impairments are likely a reflection of the natural condition. Table 4 -4. Impaired WBID comparison for Cocohatchee- Corkscrew watershed W�ii Water Segrr►ent Name FOE lm�iI ad Parameter PBSJ Analy.61S 3259W Lake Trafford Dissolved Oxygen Confirm FDEP assessment 3259W Lake Trafford Mercury N/A 3259W Lake Trafford Nutrients Confirm FDEP assessment 3259W Lake Trafford Un- ionized Ammonia Confirm FDEP assessment 3278D Cocohatchee Inland Dissolved Oxygen Confirm FDEP assessment 3278F Corkscrew Marsh Dissolved Oxygen Confirm FDEP assessment 3278L Immokalee Basin Dissolved Oxygen Confirm FDEP assessment 3278E Cow Slough Nutrients (Chlorophyll a) New 3278E Cow Slough Dissolved Oxygen New 3278F Corkscrew Marsh Fecal Coliform New 3278D Cocohatchee (Inland Segment) Fecal Coliform New 3259B Drainage to Corkscrew Dissolved Oxygen New 3259B Drainage to Corkscrew Fecal Coliform New 4.2. Golden Gate Naples Bay Watershed Following is a description of the results of the watershed analysis and WBID impairment condition. 4.2.1. Watershed Analysis The summary statistics for the Golden Gate - Naples Bay watershed are provided in Table 4 -5. Based upon the evaluation of the long term stations within the watershed, three parameters were identified as being of "potential concern"; color, dissolved oxygen, and iron. The majority of the watershed is comprised of urban development (61%) which suggests the anthropogenic modifications in the watershed may have resulted in a decline in water quality conditions. 28 Collier County Watershed Model Update ffl�f and Plan Development July 9, 2010 4.2.1.1. Dissolved Oxygen An evaluation was completed to determine the causative factor likely responsible for the depressed dissolved oxygen concentrations. The decomposition of detritus, associated with color, and phytoplankton biomass, associated with TN and TP have been shown to be inversely correlated with dissolved oxygen. Based upon regressions between dissolved oxygen and TN, TP, and color, the causative factor for the low dissolved oxygen discharge in the watershed was identified as TP (Table 4 -6). For each regression the best -fit curve was selected when comparing exponential, linear, and power relationships. While TP was identified as the causative factor for low dissolved oxygen values, nutrient and chlorophyll a concentrations remain low in the watershed. Increased nutrient input leading to elevated phytoplankton production typically lead to the decomposition of dead algal cells resulting in depressed dissolved oxygen concentrations. However, it does not appear that TP loads have resulted in excessive phytoplankton biomass. In contrast to the Cocohatchee- Corkscrew watershed, levels of dissolved oxygen appear to be more strongly influenced by nutrient levels - phosphorus - than color. These results suggest that nutrient loading could be a factor that adversely affects levels of dissolved oxygen within the Golden Gate - Naples Bay watershed. However, since levels of dissolved oxygen do not meet state criteria in the much -less developed Fakahatchee watershed, a site specific alternative criteria for dissolved oxygen is still warranted for this watershed. Table 4 -5. Water Quality Summary Statistics for the Golden Gate Naples Bay Watershed indicating potential parameters of concern Parameter N Min Mean Median Max Part ant Fxce ad Parameter of Concern BOD, mg/1 119 0.7 2.0 2.0 5.7 Chloro h II-a, u /1 558 1.0 5.4 3.0 83.0 3 N Color, PCU 553 5 1 93 80 1 800 26 1 Y Conductivity, umhos /cm 558 184 2348 616 40222 Copper, u /I 151 0.15 1.24 1.00 4.90 Dissolved Oxygen, m /I 570 0.17 5.30 5.27 16.10 45 Y Fecal Coliform, # /100m1 502 1 128 32 5400 6 N Iron, u /l 153 100.0 554.6 500.0 1500.0 14 Y Nitrate - Nitrite, m /I 545 0.00 0.05 0.04 0.33 Orthophosphate as P, mg/1 450 0.004 0.015 0.007 0.222 Salinity, ppt 443 0.0 1.7 0.3 25.6 Secchi Depth, m 535 0.00 1.20 1.10 6.00 TKN, m9/1 510 0.04 0.81 0.75 3.30 Total Nitrogen, mg/1 518 0.005 0.750 0.770 3.330 4 N Total Phosphorus, mg/1 525 0.006 0.034 0.025 0.270 0 N TSS, m /I 478 2.0 3.7 2.0 94.0 5 N Turbidit , NTU 394 0.2 2.3 1.9 19.5 Unionized Ammonia, m /1 478 0.0000 0.0008 0.0006 0.0099 0 N 29 Collier County Watershed Model Update PMj and Plan Development July 9, 2010 Table 4 -6. Identification of causative factor in the Golden Gates Naples Bay watershed for low dissolved oxygen values nir Itld p Color Power regression 0.0000 0.062 TN Exponential regression 0.0005 0.024 TP Power regression 0.0000 0.29 4.2.1.2. Iron Iron concentrations in the Golden Gate - Naples Bay watershed are sufficiently elevated to classify the watershed as of "potential concern ". Fourteen percent of the 153 samples were greater than the 1,000 ug /L Class 3 regulatory standard. Similarly, FDEP identified WBID 3278S (North Golden Gate) as impaired for iron. However, in Southwest Florida, groundwater is a more likely source of iron found in surface waters for those areas without sources such as mine drainage, sewage treatment plant outfalls, or landfill leachate from industrial scrap yards (e.g., junkyards for cars). In the absence of industrial sources, the most common nonpoint source of iron is from the weathering of iron bearing minerals and rocks. Therefore, this impairment is also likely a natural condition. 4.2.2. Evaluation of WBID Impairment Using all of the water quality data for each WBID, PBS &J confirmed the impairment status of all three FDEP impaired WBIDs (Table 4 -7). An additional potentially impaired condition for fecal coliforms in the Gordon River Extension was also identified. However, fecal coliforms was not identified as a parameter of concern at the watershed level using long -term station data. Further source identification efforts would be warranted if FDEP verifies the impairment. Table 4 -7. Impaired WBID comparison for Golden Gate - Naples Bay watershed N1i311# 3278K Wa#er S" wrlit °� Gordon River Extension Irn 'a1, d�P rain�at r Dissolved Oxygen PBSJ Jl aly�is Confirm FDEP assessment 3278S North Golden Gate Dissolved Oxygen Confirm FDEP assessment 3278S North Golden Gate Iron Confirm FDEP assessment 3278K Gordon River Extension Fecal Coliform New 4.3. Rookery Bay Following is a description of the results of the watershed analysis and WBID impairment condition. 30 Collier County Watershed Model Update and Plan Development July 9, 2010 4.3.1. Watershed Analysis The summary statistics for the Rookery Bay watershed are provided in Table 4 -8. Based upon the evaluation of the long term stations within the watershed, one parameter (dissolved oxygen) was identified as being of "potential concern", although none of the WBIDs that comprised the Rookery Bay watershed were identified by FDEP as impaired waters for dissolved oxygen. Consistently elevated chlorophyll a and nutrient values were not observed in the watershed. Similar to the Cocohatchee- Corkscrew watershed, the majority of the Rookery Bay watershed is comprised of natural areas (69 %) predominantly in the northern and central portions. Table 4 -8. Water Quality Summary Statistics for the Rookery Bay Watershed indicating potential parameters of concern framstr Mlet ,1?"ramr�rof BOD, mg /I 35 0.8 2.1 2.0 4.7 Chlorophyll -a, ug /I 147 3.0 5.2 3.2 24.6 2 N Color, PCU 144 20 58 50 240 8 N Conductivity, umhos /cm 143 182 1565 810 24400 Copper, ug /I 50 0.30 3.33 1.00 54.00 Dissolved Oxygen, mg /I 147 1.41 5.59 5.69 11.42 39 Y Fecal Coliform, # /100ml 131 1 107 40 2600 1 6 N Iron, ug /I 45 0.1 249.3 220.0 770.0 0 N Nitrate - Nitrite, mg /I 139 0.00 0.04 0.02 0.25 Orthophosphate as P, mg /I 120 0.004 0.008 0.005 0.067 Salinity, ppt 137 0.1 0.8 0.4 14.7 Secchi Depth, m 138 0.20 1.01 1.00 1.80 TKN, mg /I 129 0.24 0.70 0.63 4.30 Total Nitrogen, mg /I 132 0.010 0.631 0.645 4.300 2 N Total Phosphorus, mg /I 129 0.007 0.029 0.022 0.220 0 N TSS, mg /I 122 2.0 3.6 2.0 56.0 6 N Turbidity, NTU 88 0.4 1.6 1.4 7.5 Unionized Ammonia, mg /I 124 0.0000 0.0009 0.0006 0.0088 0 N 4.3.1.1. Dissolved Oxygen As in the other watersheds within Collier County, an evaluation of potential causative factor(s) was completed to identify the reasons for suppressed dissolved oxygen values in Rookery Bay. The Rookery Bay watershed shows dissolved oxygen levels consistently below the regulatory standard of 5.0 mg /l for fresh water bodies. Based upon regressions between dissolved oxygen and TN, TP and color, a potential causative factor for the low dissolved oxygen concentration in the watershed was identified as TP (Table 4 -9). For each regression the best -fit curve was selected when comparing exponential, linear, and power relationships. While TP maybe a causative factor, the low r value (0.11) associated with the correlation between TP and dissolved oxygen concentrations suggests that multiple influences are contributing to dissolved oxygen 31 Collier County Watershed Model Update and Plan Development July 9, 2010 fluctuations. The r2 value can be interpreted as indicating that 89 percent of the variation in levels of dissolved oxygen is not explained by TP levels. Depressed dissolved oxygen concentrations are likely influenced by natural conditions associated with the forested landscape in the upstream portions of the watershed. A site specific alternative criterion for dissolved oxygen is recommended to address the impairment. Table 4 -9. Identification of causative factor in the Rookery Bay watershed for low dissolved oxygen values Cau"Itiv F for llathOd p rz Color Power regression 0.0004 0.085 TN Rookery Bay (Inland West Segment) >0.05 New TP Power regression 0.0002 0.11 4.3.2. Evaluation of WBID Impairment No impaired WBIDs have been identified by FDEP within the Rookery Bay watershed. However, PBS &J identified two WBIDs with potential dissolved oxygen impairments (Table 4- 10). Dissolved oxygen was also identified as a potential parameter of concern when evaluating the long -term monitoring stations for the entire Rookery Bay watershed. As indicated previously, depressed dissolved oxygen concentrations are likely influenced by natural conditions associated with the forested landscape in the upstream portions of the watershed. Table 4 -10. Impaired WBID comparison for Rookery Bay watershed W31t?# Natei*, tttet�t Na me EP,,mpaired ararrieter POSJ:Ar>�alysis 3278V Rookery Bay (Inland East Segment) Dissolved Oxygen New 3278Y Rookery Bay (Inland West Segment) Dissolved Oxygen New 4.4. Faka Union Watershed Following is a description of the results of the watershed analysis and WBID impairment condition. 4.4.1. Watershed Analysis The summary statistics for the Faka -Union watershed are provided in Table 4 -11. Based upon the evaluation of the long term stations within the watershed, two parameters were identified as being of "potential concern"; color and dissolved oxygen. The majority of the Faka -Union watershed is comprised of natural areas (86 %). As previously discussed, dissolved oxygen ns] 32 Collier County Watershed Model Update and Plan Development July 9, 2010 values in the Faka Union watershed can be attributed to high color resulting from discharge from the adjacent natural landscape. In terms of dissolved oxygen, concentrations were consistently below the regulatory standard of 5.0 mg /l for fresh water bodies. An evaluation was completed to determine the causative factor likely responsible for the depressed dissolved oxygen concentrations. Based upon regressions between dissolved oxygen and TN, TP and color, the causative factor for the low dissolved oxygen discharge in the watershed was identified as color (Table 4 -12). For each regression the best -fit curve was selected when comparing exponential, linear, and power relationships. The identification of color as the causative factor further supports the explanation of the tendency for low dissolved oxygen values in watersheds dominated by undeveloped landscapes. While, the altered hydrologic processes in the Faka Union watershed (i.e., the Southern Golden Gate Estates drainage canals) suggest caution is in order, a site specific alternative criterion for dissolved oxygen is recommended. Table 4 -11. Water Quality Summary Statistics for the Faka Union Watershed indicating potential parameters of concern Oaram to IV Min Megn IWWNO Max Percent Exceed ''Parameter, of nc+rn BOD, mg /I 132 1.2 2.2 2.0 8.5 Chlorophyll -a, ug /I 524 1.0 6.3 3.0 206.0 5 N Color, PCU 509 5 1 62 50 1 240 12 Y Conductivity, umhos /cm 528 211 2046 569 62047 Copper, ug /I 166 0.15 1.37 1.00 17.70 0 N Dissolved Oxygen, mg /I 542 1.02 6.02 5.96 14.54 37 Y Fecal Coliform, # /100ml 456 1 135 23 3850 8 N Iron, ug /I 179 100.0 309.1 220.0 1390.0 2 N Nitrate - Nitrite, mg /I 514 0.00 0.03 0.01 1.31 Orthophosphate as P, mg /I 418 0.004 0.007 0.005 1 0.099 Salinity, ppt 522 0.0 1.2 0.3 41.7 Secchi Depth, m 319 0.30 1.19 1.20 2.50 34 Y TKN, mg /I 463 0.04 0.60 0.52 4.90 Total Nitrogen, mg /I 473 0.005 0.516 0.470 5.030 1 3 N Total Phosphorus, mg /I 496 0.004 0.023 0.015 0.435 0 N TSS, mg /I 441 1 2.0 1 .1 2.0 62.0 6 N Turbidity, NTU 331 0.1 1.8 1.3 7.1 Unionized Ammonia, mg /I 449 0.0000 0.0006 0.0003 0.0127 0 N 33 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Table 4 -12. Identification of causative factor in the Faka Union watershed for low dissolved oxygen values C�aitveial- *6d p, Anaryl;s Color Power regression 0.000 0.28 TN Power regression 0.028 0.01 TP Power regression 0.000 0.06 4.4.2. Evaluation of WBID Impairment No impaired WBIDs have been identified by FDEP within the Faka Union watershed. However, PBS &J identified three potential impaired water bodies (Table 4 -13). The Faka Union (South segment) had low dissolved oxygen values and elevated fecal coliform concentrations. Additionally, the north segment also was identified with low dissolved oxygen. In line with the watershed analysis results, dissolved oxygen was identified as a potential parameter of concern in the Faka Union watershed. However, low concentrations are likely a natural condition. If fecal coliforms become a FDEP impairment parameter, source identification work would be necessary. Table 4 -13. Impaired WBID comparison for Faka Union watershed w� uVaterrr;�nt i>me tretl P raxr - Anaryl;s 32781 Faka Union (South Segment) Dissolved Oxygen New 32781 Faka Union (South Segment) Fecal Coliform New 3278H Faka Union (North Segment) Dissolved Oxygen New 4.5. Fakahatchee Watershed Following is a description of the results of the watershed analysis and WBID impairment condition. 4.5.1. Watershed Analysis The summary statistics for the Fakahatchee watershed are provided in Table 4 -14. Based upon the evaluation of the long term stations within the watershed, three parameters were identified as being of "potential concern"; color, dissolved oxygen, and fecal coliform. The Fakahatchee Strand (WBID 3278G) was declared verified impaired by FDEP for both dissolved oxygen and fecal coliforms. However, the majority of the Fakahatchee watershed is comprised of natural areas (85 %). In fact, the Fakahatchee watershed has been identified by FDEP as a reference area due to the limited hydrologic impacts. 34 Collier County Watershed Model Update and Plan Development July 9, 2010 Table 4 -14. Water Quality Summary Statistics for the Fakahatchee Watershed indicating potential waters of concern Parameter , N Min Mean Median Max Pest Exwed Parameter of Concern BOD, mg/1 107 1.5 2.3 2.0 9.8 0.000 0.06 Chlorophyll-a, u /1 435 3.0 9.3 3.0 404.5 9 N Color, PCU 418 5 79 75 350 23 Y Conductivity, umhos /cm 436 197 5599 604 72958 Copper, u /l 133 0.15 1.16 1.00 8.00 Dissolved Oxygen, mg/1 448 0.24 3.80 3.34 12.77 75 Y Fecal Coliform, # /100m1 387 1 201 50 5450 12 Y Iron, u /1 147 0.1 213.8 150.0 1300.0 1 N Nitrate - Nitrite, mg/1 428 0.00 1 0.02 0.01 0.22 Orthophosphate as P, m /l 351 0.004 0.020 0.006 0.368 Salinity, ppt 441 0.0 3.4 0.3 50.3 Secchi Depth, m 361 0.20 1.03 1.00 2.80 TKN, mg/1 395 0.04 0.88 0.74 5.19 Total Nitrogen, mg/1 393 0.005 0.716 0.650 5.320 7 N Total Phosphorus, m /I 1 407 1 0.004 1 0.047 1 0.020 1 1.180 1 3 1 N TSS, mg/1 368 1 2.0 4.8 2.0 97.0 10 N Turbidity, NTU 281 0.1 1 1.0 0.7 5.9 Unionized Ammonia, mg/1 353 0.0000 1 0.0007 0.0003 0.0162 1 0 1 N 4.5.1.1. Dissolved Oxygen Based upon the current regulatory criteria for the Fakahatchee watershed, dissolved oxygen levels were consistently below the regulatory threshold of 5.0 mg /l for fresh water bodies. An evaluation of the causative factor was completed to determine the causative factor likely responsible for the depressed dissolved oxygen concentrations. Based upon regressions between dissolved oxygen and TN, TP and color, the causative factor for the low dissolved oxygen discharge in the watershed was identified as color (Table 4 -15). For each regression the best -fit curve was selected when comparing exponential, linear, and power relationships. The identification of color as the primary causative factor further supports the explanation of the tendency for low dissolved oxygen values in this mostly undeveloped landscape. A site specific alternative criterion for dissolved oxygen is appropriate for the Fakahatchee watershed. Table 4 -15. Identification of causative factor in the Fakahatchee watershed for low dissolved oxygen values causstive is ar Method p , p� Color Power regression 0.000 0.17 TN Linear regression 0.033 0.01 TP Power regression 0.000 0.06 35 Collier County Watershed Model Update IMS) and Plan Development July 9, 2010 4.5.2. Fecal Coliform Bacteria Fecal Coliform bacteria was identified as a potential parameter of concern in the Fakahatchee watershed based on the analysis of the long -term sampling stations. Twelve percent of the 387 values exceeded the 400 # /100mL criteria established for Class 3 waters. WBID 3278G (Fakahatchee Strand) was identified as verified impaired for bacteria as evaluated by FDEP. As was previously discussed, fecal coliforms are used as an indicator of pathogenic organisms and are used to identify potential health threats. However, fecal coliform bacteria may not be an appropriate indicator for pathogenic diseases in sub - tropical climates. Further source identification efforts are warranted. 4.5.3. Evaluation of WBID Impairment Using all of the water quality data for each WBID, PBS &J confirmed the impairment status of both FDEP impaired WBIDs (Table 4 -16). WBID 3289I (Camp Keais) was also identified as potentially impaired for dissolved oxygen. Similarly, both dissolved oxygen and fecal coliforms were identified as potential parameters of concern in the Fakahatchee watershed upon evaluation of the long -term monitoring stations. Table 4 -16. Impaired WBID comparison for Fakahatchee watershed $I 1 lairlr r118nt +{81118 FD19P pa re . Pamm 6 ps 1 I'�is 3278G Fakahatchee Strand Dissolved Oxygen Confirm FDEP assessment 3278G Fakahatchee Strand Fecal Coliform Confirm FDEP assessment 32591 Camp Keais Dissolved Oxygen New 4.6. Okaloacochee - SR29 Watershed Following is a description of the results of the watershed analysis and WBID impairment condition. 4.6.1. Watershed Analysis The summary statistics for the Okaloacochee /SR29 watershed are provided in Table 4 -17. Based upon the evaluation of the long term stations within the watershed, four parameters were identified as being of "potential concern "; color, dissolved oxygen, iron, and total nitrogen. Similar to the watershed evaluation of the parameters of potential concern in the Okaloacochee- SR29 watershed, iron and dissolved oxygen were found to be impaired by FDEP. As the majority of the Okaloacochee -SR29 watershed is comprised of natural areas (58 %), it is likely that some impairments represent a natural condition, as described below. 36 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Table 4 -17. Water Quality Summary Statistics for the Okaloacochee /SR29 Watershed indicating potential waters of concern �R7�At11 i�l Min `.1 hifi ... Maw iy Q. - � BOD, mg /I 38 1.6 2.3 2.0 5.1 Chlorophyll -a, ug /I 266 1.0 6.5 3.0 69.4 11 N Color, PCU 255 5 90 80 450 31 Y Conductivity, umhos /cm 297 103 491 502 905 Copper, ug /I 73 0.15 1.35 1.10 6.29 Dissolved Oxygen, mg /I 299 0.12 2.57 2.36 8.60 91 Y Fecal Coliform, #/100ml 243 1 112 33 3050 5 N Iron, ug /I 49 0.1 478.2 250.0 1910.0 18 Y Nitrate - Nitrite, mg /I 295 0.00 1 0.02 0.01 0.37 Orthophosphate as P, mg /I 254 0.002 0.019 0.010 0.312 Salinity, ppt 154 0.0 0.2 0.2 0.5 Secchi Depth, m 262 0.10 1.28 1.25 2.60 TKN, mg /I 282 0.04 1.23 0.90 35.35 Total Nitrogen, mg /I 280 0.005 1.124 0.811 35.353 16 Y Total Phosphorus, mg /I 290 0.006 1 0.049 0.026 0.470 2 N TSS, mg /I 238 2.0 4.6 4.0 174.0 5 N Turbidity, NTU 210 0.2 1.4 0.7 20.0 Unionized Ammonia, mg /I 267 0.0000 1 0.0018 0,0003 0.3241 1 N 4.6.1.1. Dissolved Oxygen FDEP determined that WBEDs 3278T (Okaloacoochee) and 3278W (Silver Strand) were impaired for dissolved oxygen. Based upon the current dissolve oxygen criteria for the Okaloacochee -SR29 watershed, dissolved oxygen levels were consistently below the regulatory standard of 5.0 mg /l for fresh water bodies. An evaluation of the causative factor was completed to determine the causative factor likely responsible for the depressed dissolved oxygen. Based upon regressions between dissolved oxygen and TN, TP and color, the causative factor for the low dissolved oxygen discharge in the watershed was identified as color (Table 4 -18). For each regression the best -fit curve was selected when comparing exponential, linear, and power relationships. The identification of color as the causative factor is statistically significant, but it has a very low r value, suggesting other factors may be influencing dissolved oxygen levels. The relatively high levels of TN and TP, compared to the Faka Union and Fakahatchee watersheds, suggest that anthropogenic loading could be a concern in this watershed. Regardless of the influence of nutrient loading, the finding that dissolved oxygen levels in the Fakahatchee watershed typically do not meet state criteria supports the need for a site specific alternative criteria within the Okaloacochee -SR29 watershed. 37 Collier County Watershed Model Update M!$I and Plan Development July 9, 2010 Table 4 -18. Identification of causative factor in the Okaloacochee -SR29 watershed for low dissolved oxygen values COO, 1W Factor Method p r2 Color Exponential regression 0.0001 0.06 TN Silver Strand >0.05 Confirm FDEP assessment TP Silver Strand >0.05 New 4.6.1.2. Iron FDEP identified WBID 3261C (Barron River Canal) as impaired for iron. The Okaloacochee- SR29 watershed iron concentrations were sufficiently elevated to classify the watershed as of "potential concern" in regards to elevated iron concentrations as 18 percent of the 49 samples show concentrations higher than the 1,000 ug /L Class 3 regulatory standard. Groundwater is a likely source of iron found in surface waters. Its origin is likely the weathering of iron bearing minerals and rocks. 4.6.2. Evaluation of WBID Impairment Per the evaluation of the water quality data for each WBID, PBS &J confirmed the impairment status of all three FDEP impaired WBIDs. As shown in Table 4 -19, three additional potential impairments were identified. Dissolved oxygen and iron were both identified as parameters of concern in the watershed analysis. However, the copper and chlorophyll a impairments resulted from the analysis of data for each WBID. In regards to the potential copper impairments for WBID 3278W (Silver Strand), four water quality locations provide data within the waterbody. However, all of the copper data was collected at the 21FLSFWMIMKBRN location. It is recommended that the water quality sampling location be reviewed to ensure that the water samples is not collected near boardwalks and pilings constructed from pressure- treated lumber as it may be subject to copper leaching. This would result in potentially elevated copper concentrations. In terms of Chlorophyll a, a review of the data analyzed for Okaloacoochee WBID (3278T) showed that five water quality stations exist in the WBID. However, 76 of the 78 chlorophyll a data points came from one station (Okala858). A preliminary investigation indicates that TP may be the causative factor resulting in elevated phytoplankton production at this location. However, a more detailed evaluation of the data and sampling station location is necessary. Table 4 -19. Impaired WBID comparison for Okaloacochee /SR29 watershed i�D#i " 3261C Water grn s #Fl!Ialoe - Barron River Canal �, .. Iron t,A Eysis Confirm FDEP assessment 3278T Okaloacoochee Dissolved Oxygen Confirm FDEP assessment 3278W Silver Strand Dissolved Oxygen Confirm FDEP assessment 3278W Silver Strand Copper New 3278T Okaloacoochee Chlorophyll a New 3261 C Barron River Canal Dissolved Oxygen New 38 Collier County Watershed Model Update PB�f and Plan Development July 9, 2010 5.0 Conclusions The FDEP has identified multiple impairments of individual WBIDs for many water quality parameters, although the most widespread "impairments" or "parameters of concern" appear to be those for dissolved oxygen and color. However, high levels of color appear to be related to influences of high- tannin water from the extensive forested landscapes in areas such as the Corkscrew Swamp and the Fakahatchee Strand. In turn, increased tannin -rich waters during the wet season appear to result in depressed levels of dissolved oxygen; the existing criterion for freshwater water bodies appears to be overly restrictive. In response, Collier County should consider the development of a site specific criterion for dissolved oxygen for its freshwater water bodies, perhaps using the approach outlined in Section 4.2.1. This proposed approach would be based on observed levels of dissolved oxygen in the Fakahatchee Strand, with different thresholds of "impairment" for the dry vs. wet seasons. A number of exceedances of the existing criterion for iron might actually reflect influences of groundwater, rather than an anthropogenic influence from the landscape. With the exception of Lake Trafford, freshwater water bodies in Collier County are not characterized by consistently high levels of TN or TP. The water quality benefits that seem to be occurring in response to the dredging project for Lake Trafford should be considered prior to implementing any water quality "fixes ", as water quality may already be improved sufficiently that further activities are not needed. While many of the freshwater water bodies within the watersheds of Collier County are designated as "impaired" for fecal coliform bacteria, these indicator organisms do not specifically identify humans as a source of contamination ( Fujioka 2001, and Fujioka et al. 1999). Additional efforts aimed at source identification are appropriate. No discrepancies were found between FDEPs and PBS &J impaired WBID designation. However, PBS &J identified 16 new potential impairments. The incorporation of additional data with the IWR Run 39 dataset as well as differences in the analysis period are likely responsible for the identification of potentially new impaired WBIDs. Results of water quality analyses at the watershed level suggest that DO is the primary impairment in Collier County in the context of FDEP's TMDL program. Analyses of individual WBIDS are consistent with these results, although additional water quality impairments are apparent within some WBIDS. 39 Collier County Watershed Model Update PII!q and Plan Development July 9, 2010 6.0 References APHA (ed) (1995) Standard Methods for the Examination of Water and Wastewater, Vol. American Public Health Association, Inc., Washington DC Black, Crow, and Eidsness, Inc. 1974. Hydrologic Study of the G. A. C. Canal Network. Gainesville, FL. Project no. 449- 73 -53. FDEP. 2004. Everglades Marsh Dissolved Oxygen Site Specific Alternative Criterion Technical Support Document. Final Report to Florida Department of Environmental Protection. Tallahassee, FL. 61 pp. FDEP. 2008. Dissolved Oxygen TMDL for the Gordon River Extension, WBID 3278K (formerly 3259C). Final Report to Florida Department of Environmental Protection. Tallahassee, FL. 40 pp. Fujioka, R.S. 2001. Monitoring coastal marine waters for spore - forming bacteria of faecal and soil origin to determine point from non -point source pollution. Water Science and Technology. 44: 181 -188. Fujioka, R.S., Stan- Denton, C., Borja, M., Castro, J., and K. Morphew. 1999. Soil, the environmental source of Escherichia coli and enterococci in Guam's streams. Journal of Applied Microbiology. (Symposium supplement) 85: 83S -89S. Harwood VJ, Whitlock J, Withington V (2000) Classification of antibiotic resistance patterns of indicator bacteria by discriminant analysis: use in predicting the source of fecal contamination in subtropical waters. Applied and Environmental Microbiology 66:3698- 3704 Klein, H., W.J. Schneider, B.F. McPherson and T.J. Buchanan. May 1970. Some Hydrologic and Biologic Aspects of the Big Cypress Swamp Drainage Area, Southern Florida. United States Geologic Survey Open -file Report 70003. McCormick, P.V., Chimney, M.J. and D.R. Swift. 1997. Diel oxygen profiles and water column community metabolism in the Florida Everglades, U.S.A. Archives die Hyrobiologie. 140: 117 -129. Tetra Tech, Inc and Janicki Environmental, Inc. 2004. Compilation, Evaluation, and Archiving of Existing Water Quality Data for Southwest Florida. Contract No. DACW 17 -02 -D- 0009. Final Report submitted to Department of Army, Jacksonville District Corps of Engineers. Weiss. R.F. 1970. The solubility of nitrogen, oxygen and argon in water and seawater. Deep -Sea Research. 17: 721 -73.5. 40 Collier County Watershed Model Update MISI and Plan Development July 9, 2010 Appendix A Water Quality Station List Collier County Watershed Model Update PBSI and Plan Development July 9, 2010 Appendix 6 Dissolved Oxygen Concentration Issues in Collier County Collier County Watershed Model Update M1 and Plan Development July 9, 2010 Appendix C Guidance Document for Developing a DO SSAC Collier County Watershed Model Update and Plan Development July 9, 2010 PBS140 Technical Memorandum To: Mac Hatcher, PM Collier County From: Moris Cabezas, PBS &J Dave Tomasko, PBS &J Ed Cronyn, PBS &J Date: July 9, 2010 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 1, Task 3.2 Functional Assessment 1.0 Objective This Technical Memorandum addresses Element 1, Task 3.2: Functional Assessment. The primary objective of this task is to develop and conduct a landscape -level functional assessment of native wetland and upland communities for six watersheds: • Cocohatchee- Corkscrew • Golden Gate - Naples Bay • Rookery Bay • Faka Union • Okaloacoochee /SR 29 • Fakahatchee In addition to ecological functional assessment, this task includes analysis of potential hydrological storage on undeveloped lands and summary of non - native invasive vegetation coverage in these watersheds. 2.0 Introduction This Technical Memorandum summarizes the development and application of a landscape - level functional assessment method that will be used to determine the ecological value of existing conditions in six watersheds within the County. This method will also be utilized later in developing ecological -based performance measures for proposed restoration projects. In addition to the development and application of an ecological functional assessment method, this memorandum presents two additional analyses: estimated ecological capacity for additional water storage on undeveloped lands, and coverage of non - native invasive species. These additional data sets, though not utilized directly for the functional assessment method, provide related information that may be used to further assess watershed conditions. Results of the functional assessment and the other two sets of data include an overview of the watersheds' existing functional value and identification of 1 Element 1 PMT 3.2 Functional Assessment areas within each watershed where projects are most likely to result in improved functional values. 3.0 Methodology The Uniform Mitigation Assessment Method (UMAM, Chapter 62 -345 Florida Administrative Code) served as a template from which to design the functional assessment method for this project, due to the widespread use and acceptability of UMAM for ecological assessment. Modifications from UMAM were necessary in order to implement the functional assessment at the watershed level for this project, rather than the site - specific level for which UMAM was designed. The overall concepts and design, however, remain similar to UMAM. The methods of analysis for the other items (hydrological storage and non - native invasive vegetation) are described after the functional assessment methodology. Similar to UMAM, the optimal condition for this functional assessment is defined in terms of the landscape position, vegetation, and hydrology of the ecological community in a pre- determined reference condition. Scores for each of these three parameters are assigned via indices based on the degree of ecological change between the reference condition and existing condition. The reference condition and scoring methodology are further described below. The first, necessary element of the functional assessment method consisted of establishing reference conditions, similar to the Part I "frame of reference" proscribed in UMAM. Discussions with Collier County staff, other agencies, and non - profit conservation organizations quickly led to the conclusion that the data set most applicable to this project is the pre - development vegetation map (PDVM) developed for the Southwest Florida Feasibility Study (Duever, 2004). This data layer, more thoroughly described in the Technical Memorandum for Reference Period Comparison (Element 1 Task 3.1), consists of a seamless five - county map (including Collier County) of fifteen vegetation associations defined by common vegetative composition and hydrological characteristics, in approximately the locations where they would have occurred prior to development. The functional assessment method developed for this project consists of three indices: • Landscape Suitability Index • Vegetation Score • Hydrological Score Similar to UMAM, each of the three indices includes scores ranging from 0 (no functional value) to 10 (optimal functional value), based on comparison to a reference condition. The landscape position of a wetland, upland or other natural community determines the opportunity for that community to provide ecosystem functions. For instance, a wetland with an intact natural hydroperiod and undisturbed vegetation in the middle of a highway interchange has limited opportunity to provide habitat support functions to wetland- 2 Element 1 r 3.2 Functional Assessment dependent wildlife, and would therefore score low for landscape position. The functional assessment method developed for this project utilizes a scoring system developed by the University of Florida known as Landscape Suitability Index (LSI) in order to evaluate landscape position. The LSI is described in greater detail in Section 3.1 below. The species, stratum (e.g., forested vs. herbaceous), and type of ecosystem (e.g., upland vs, wetland) play a large role in determining ecological functions (wildlife nesting, foraging, etc.). UMAM evaluates this set of characteristics in a parameter known as Community Structure. The vegetation scoring index developed for this project evaluates these characteristics via comparison between 2007 FLUCCS data and the PDVM. The Technical Memorandum for Task 3.1 (Reference Period Comparison) included analysis of land use changes, and the results of that analysis are incorporated into the vegetation scoring index for this project. The vegetation scoring index is described in greater detail in Section 3.2 below. The depth and duration of inundation help to determine how wildlife utilize a natural community. The hydrological scoring index developed for this project assigns values by comparing the hydrological characteristics of existing and pre - development communities, which is based on the vegetation and land use change analysis conducted for Task 3.1. The hydrology scoring index is further described in Section 3.3 below. In accordance with UMAM procedures, a score of 10 for existing condition is appropriate where a site retains optimal value (100 percent of the value present in the pre - development condition), 7 for moderate value (70 percent of the value present in the pre - development condition), 4 for minimal value (40 percent of the value present in the pre - development condition), 0 for no value, and other whole- number scores between 1 and 9 as appropriate. Due to the overall goal of developing a method primarily applicable at the watershed level, as well as the regional characteristics of this analysis, the method exclusively relies on available GIS data to determine functional value. The results presented herein should therefore be utilized only for watershed -level assessment rather than as a substitute for on- site analysis typically required for permitting purposes. 3.1 Landscape Suitability Index (LSI) The scoring system used for this project to assign value to landscape position is known as the Landscape Suitability Index (LSI), developed by the Center for Wetlands at the University of Florida (UF) (Bardi et al; Reiss et al, 2009; Brown and Vivas, 2005). The LSI is a land use based index of human disturbance, which serves as a proxy for the degree to which land uses support or inhibit ecological functions of adjacent areas. An area surrounded by high- intensity developed land uses will provide relatively few ecological functions, while an area surrounded by undeveloped natural lands will tend to have intact ecological functions. For ease of evaluation on a watershed level, the application of LSI to this project began with converting the 2007 FLUCCS coverage into a grid composed of 750 by 750 -feet (approximately 230m) cells. Use of a grid rather than irregular polygons enables uniform 3 Element 1 • 3.2 Functional Assessment calculation of the area of adjacent land uses, since the adjacent land uses are determined from the primary land use within each of the eight adjoining cells. Also, the selected cell size represents 1/4 of the MIKE SHE grid cell size and will facilitate analysis of model results, in terms of impacts and benefits to the natural vegetation. LSI for a given cell is calculated spatially based on coefficients applied to the weighted extent of land use distribution in the adjoining cells. The coefficients are shown in Table 3- 1. 0 is the lowest possible coefficient (e.g., for an area dominated by a central business district with buildings averaging 4 stories) and 10 is the highest (for natural lands and/or natural water bodies). It is to be noted that the land use distribution within a cell under consideration do not enter in the calculation of its LSI because the focus of this index is location, not internal characteristics. The LSI value for each cell is thus represented by the following equation (adapted from Brown and Vivas, 2005): LSItotal =E %LUi • LSI„ where LSIt,ai = LSI score for a 750 by 750 -feet cell %LUi = percent of the adjoining cells dominated by land use i LSIi = landscape suitability index coefficient from Table 3 -1 for land use i The LSI value can also be estimated for areas larger than a cell, such as a WBID or the entire watershed by simply averaging the LSI values of the cells within that area. PM 4 Element 1 3.2 Functional Assessment Table 3 -1. Landscape Suitability Index Coefficients for Land Use /Land Cover Classes in Florida (from K. Reiss. uers. comm.) GIB tend Natural System 1 rtes 10.00 Natural Open water 10.00 Pine Plantation 9.36 Recreational / Open Space (Low-intensity) 9.08 Woodland Pasture (with livestock) 8.87 Pasture (without livestock) 8.03 Low Intensity Pasture (with livestock) 7.32 Citrus 7.02 High Intensity Pasture (with livestock) 6.96 Row crops 6.07 Single Family Residential (Low-density) 3.57 Recreational / Open Space (High-intensity) 3.42 High Intensity Agriculture (Dairy farm) 3.33 Single Family Residential (Med- density) 2.81 Single Family Residential (High-density) 2.72 Mobile Home (Medium density) 2.56 Highway (2 lane) 2.43 Low Intensity Commercial 2.22 Institutional 2.14 Highway (4 lane) 1.91 Mobile Home (High density) 1.90 Industrial 1.87 Multi-family Residential (Low rise) 1.49 High Intensity Commercial 0.91 Multi-family Residential (High rise) 0.90 Central Business District (Average 2 stories) 0.64 Central Business District (Average 4 stories) 0.00 3.3 Vegetation Score Similar to UMAM, the vegetation scoring index for this project assumes that the pre - development vegetation community provides optimal functional value for native wildlife (e.g., for food, cover, and breeding), and changes reduce the functional value. The vegetation scoring index developed for this project therefore measures the degree of change in vegetation between the PDVM and existing FLUCCS. The greater the difference in vegetation between PDVM and existing FLUCCS, the lower the vegetation score. The input data for this evaluation come from the vegetation and land use change analysis conducted for Task 3.1 (Reference Period Comparison). Existing FLUCCS indicating a shift both in dominant stratum type (e.g., forested to herbaceous) and ecosystem (e.g. wetland to upland) in comparison to the PDVM were M 5 Element 1 3.2 Functional Assessment assigned low scores, while those areas with a similar dominant stratum and ecosystem (e.g., freshwater marsh in the reference and existing condition) were assigned the highest scores. Changes in land use, which establish the vegetation scores according to the index below, were determined by overlaying the existing GIS FLUCCS polygons directly on the PDVM polygons. The vegetation scoring index, based on comparison between PDVM and existing FLUCCS vegetation, is summarized below: • Polygons whose existing FLUCCS designation indicates the same dominant vegetation as in the PDVM (e.g., hydric flatwoods pre - development and existing) received a score of 10; • Polygons that retained the same dominant stratum and ecosystem type (e.g., freshwater forested wetland to freshwater forested wetland) also received a score of 10; • Polygons that shifted from one dominant stratum to another but retained the same ecosystem type (e.g., freshwater forested wetland to herbaceous freshwater wetland) received a score of 8; • A shift between mesic to hydric flatwoods or vice -versa received a score of 8; • Vegetation that shifted between natural ecosystem types and stratum (e.g., herbaceous freshwater wetland to forested native upland) received a score of 6; • Polygons that have been converted to an artificial water body received a score of 3; • A natural system that has been converted to a developed land use class (e.g., agriculture, urban development, golf course, pasture) received a score of 0. Vegetation scores in the Results section (Section 4.0) are mapped by polygon and aggregated as a percentage of the watershed or WBID land area with scores falling within a certain range. 3.4 Hydrology Score The hydrology scoring index developed for this project estimates the functional value provided to native wildlife based on depth and duration of inundation. Similar to the vegetation scoring index, the primary source of input data are the PDVM and current GIS FLUCCS layers. Shifts in vegetation that represent a change in depth and duration of inundation within a polygon indicate low functional value, while a polygon with current vegetation that represents similar depth and duration of inundation in comparison to PDVM indicate high functional value. The depth and duration of inundation for PDVM and current FLUCCS vegetation are inferred from the average typical hydrological characteristics of vegetation communities in Southwest Florida, using a table developed by Mike Duever at SFWMD (Table 3 -2). Changes in land use, which establish the hydrology scores according to the index below, were determined by overlaying the 2007 FLUCCS polygons directly on the PDVM polygons. The hydrology scoring index, determined from the FLUCCS/PDVM overlay, is summarized below. 6 Element 1 3.2 Functional Assessment • For those polygons where the existing hydroperiod is shorter than the pre - development hydroperiod (e.g., long - hydroperiod wetland to short- hydroperiod wetland or upland), the simplest and most direct score is the percentage decline in wet season water level and shortened hydroperiod. For example, a 30% reduction in depth and duration of inundation would result in a score of 7 for that polygon. This calculation covers approximately 1/3 of the potential change scenarios. • Where the hydroperiod is longer than in pre - development, the polygon is assigned the same value as a change in the opposite direction (e.g., an increase in hydroperiod and wet season water level is deemed to have the same effect on the function of the wetland than the same magnitude of decrease in hydroperiod and water level). This is consistent with the overall premise in functional assessment methods like UMAM —that the pre - development condition of the wetland provides the highest value of functions. • A third group of scenarios includes those where the system has changed from a freshwater system to a brackish system (e.g., freshwater wetland to estuarine wetland). In those instances, a moderate score (6) is assigned, representing change in hydrological functions due to change in timing and quantity of flows and, consequently, salinity. • Where uplands have become wetlands, a hydrological score of 10 is assigned the polygon, due to providing an entirely new set of hydrology -based wildlife functions not present in the pre - development condition. • Where a PDVM natural upland or wetland system has been converted to an artificial system (e.g., agriculture, development), a score of 0 is assigned, representing 100 percent loss of the value of hydrological functions provided to wildlife in the natural system. • Where the pre - development condition was a natural open water body that was subsequently converted into an artificial water (e.g., drainage canal, mining lake), a score of 3 is assigned, representing that few hydrological functions remain from the original system. Hydrology scores in the Results section (Section 4.0) are mapped by polygon and aggregated as a percentage of the watershed or WBID land area with scores falling within a certain range. 7 Element 1 3.2 Functional Assessment Table 3 -2. Hydrologic Regimes of Major Southwest Florida Plant Communities (from Duever, pers. comm. SW Florida Plant Gotdmunities: ydxorit (Months) Serial mater, Levi (des) Wet Dry (1,10)* Xeric Flatwood 0 < -24 -60,-90 Xeric Hammock Mesic Flatwood <1 <2 — -46,-76 Mesic Hammock H dric Flatwood 1-2 2-6 -30,-60 H dric Hammock Wet Prairie 2 - 6 6-12 -24,-54 Dwarf Cypress Freshwater Marsh 6-10 12 - 24 -6,46 Cypress 6 - 8 12 - 18 -16,-46 Swamp Forest 8-10 18 - 24 -6,-36 Open Water >10 >24 < 24, -6 Tidal Marsh Tidal Tidal Tidal Mangrove Beach * 1 = average year low water 10 = 1 in 10 year drought July 2002 An important caveat when using this table, as noted by Duever as well as others (e.g., Bales et at, 1997) is that the historic hydrologic conditions are inferred from soils survey data and the conditions under which these same ecological communities presently occur. Also, the values represent long -range average conditions for each of the ecological communities and vary from year to year. 8 Element 1 3.2 Functional Assessment y 3.5 Average Functional Assessment Value The average functional assessment value for each polygon was calculated using the same approach as UMAM: all three parameters were given equal weight. The functional assessment score is therefore the average of the three individual scores, resulting in a single score ranging from 0 to 10 for each polygon. Average Functional Assessment Value = (LSI + Vegetation Score + Hydrology Score) 13 Functional values in the Results section are mapped by polygon. For the purpose of watershed -level and WBID -level assessment, the functional values are aggregated by calculating the percentage of the total land area represented by polygons within a particular scoring range. 3.6 Data Sources The primary data sources for this project are described in greater detail in the Technical Memorandum for Element 1 Task 3.1: Reference Period Comparison. Other sources are described below. Duever, M. 2002. Hydrologic Regimes of Major Southwest Florida Plant Communities. South Florida Water Management District. Table 3 -2, developed by Mike Duever, represents a summary of typical, average hydrological conditions of natural ecological communities in Southwest Florida, based on Duever's and others' literature review and professional knowledge. The 15 communities are the same as those used in the PDVM, and this data table is one of the data inputs into the SWFFS Natural Systems Model (NSM). The NSM was used to predict hydrological characteristics based on several variables, including vegetation, under a scenario in which canals and control structures have been removed. The table represents long -term typical conditions for each of the listed ecological communities, although it should be noted that these conditions vary from year to year. Table 3 -2, contained or cited in several publications by and for SFWMD, was obtained directly via email from Mr. Duever by PBS &J staff. Florida Natural Areas Inventory. March 2010. Florida Invasive Plants Geodatabase (FLInv). FNAI has developed FLInv to map exotic plant infestations on public conservation lands in Florida. The FLInv (htti)://www.fnii.org /IiivasiveSl)ecies.cfril) is populated with existing data submitted by resource managers and supplemented by FNAI field surveys. The non- native invasive species therein are those listed by the Florida Exotic Pest Plant Council as being invasive species. State (Florida Department of Agriculture and Consumer Services) and federal (United States Department of Agriculture's Animal and Plant Health Inspection Service) rankings are also listed for the EPPC - listed species. Each polygon identifies at a minimum the species and date that the observation was made. Other factors such as acreage, plant distribution, reference, and eradication efforts taking place at that location 9 Element 1 3.2 Functional Assessment are included as available. The FLInv is updated quarterly and was last updated on 03/31/2010. The University of Georgia - Center for Invasive Species and Ecosystem Health. August 2009. Early Detection and Distribution Mapping System (EDDMapS). EDDMapS is a web -based mapping system for documenting invasive species distribution. EDDMapS was launched in 2005 by the Center for Invasive Species and Ecosystem Health at the University of Georgia and was originally designed as a tool for state Exotic Pest Plant Councils to develop more complete distribution data of invasive species. EDDMapS combines data from other databases and organizations as well as volunteer observations to create a national network of invasive species distribution data that is shared with educators, land managers, conservation biologists, and others. An interactive Web interface allows participants to submit their observations or view results through interactive queries into the EDDMapS database. All data is reviewed by state verifiers to ensure data is accurate. 4.0 Results and Discussion The functional assessment scores for the three parameters and average functional assessment value in each of the watersheds are summarized in Figures 4 -1 through 4 -4 and Tables 4 -1 through 4 -5. Hydrological storage capacity and coverage of non - native invasive species are summarized in Figures 4 -5 through 4 -8 and Tables 4 -6 and 4 -7. 4.1. Cocohatchee- Corkscrew Watershed The functional assessment of the Cocohatchee- Corkscrew Watershed (see Figures 4 -1 through 4 -4) reveals trends in three distinct areas: the central part of the watershed dominated by the Corkscrew Swamp system maintains a high functional value; the northern and eastern portions dominated by agricultural lands are moderate to low; and the western urbanized portion other than preserved coastal systems are low functional value. The LSI remains high (seven or greater) throughout the eastern 2/3 of this watershed due to low intensity land uses. Vegetation and hydrology scores are lower due to converting 48 percent of the watershed to agricultural or urban uses. Overall, approximately 40 percent of this watershed (52,000 acres) has a functional value of 7 or higher, and over 50 percent of the watershed (71,000 acres) has a functional value of 4 or less (Table 4 -1). LSI scores (Table 4 -2) reflect the relative dominance of low- intensity land uses within this watershed. Almost 75 percent (nearly 100,000 acres) of the Cocohatchee- Corkscrew Watershed has an LSI of 7 or greater. The largest portion of this high -LSI area (nearly 50,000 acres) occurs in WBID 3278F (Corkscrew Marsh). At the other end of the spectrum, the western portions of the Cocohatchee- Corkscrew Watershed (WBIDs 3259Z, 3287C, and 3278D) contain the bulk of the watershed with LSI scores less than 7 (just under 20,000 acres). The distribution of vegetation scores within the Cocohatchee- Corkscrew Watershed (Figure 4 -2) reflects conversion of natural habitat other than within the Corkscrew Swamp system. More than 50 percent (over 70,000 acres) of this watershed has a vegetation score mow% 10 Element 1 iiJ 3.2 Functional Assessment of 0, resulting primarily from the loss of most of the mesic and hydric flatwoods in this watershed (as documented in the Technical Memorandum for Task 3.1 Reference Period Comparison). By contrast, less than 35 percent of WBID 3278F has a vegetation score of 0, and over 60 percent (nearly 33,000 acres) of that WBID has a vegetation score of 8 or higher (Table 4 -2). Hydrology scores in the Cocohatchee- Corkscrew Watershed follow a pattern similar to the vegetation scores, with nearly 72,000 acres (53 percent) of the watershed scoring 0, and nearly 50,000 acres (37 percent) of WBID 3278F scoring higher than 7 (Tables 4 -1 and 4- 2). Reviewing these results, the greatest opportunities for improvement of ecological value occur north and east of Corkscrew Swamp and Lake Trafford. This portion of the watershed is characterized by moderate to low overall functional value, with substantial potential improvement via restoration of natural hydrology and/or vegetation. Projects immediately adjacent to Corkscrew Swamp and Lake Trafford, in particular, would benefit from the LSI values provided by these two natural systems, and restoration projects would also lead to an improvement in the LSI values of other nearby lands. WBIDs with a combination of high LSI and low vegetation and hydrology scores (Table 4 -2) include 3259B (Drainage to Corkscrew), 3278E (Cow Slough), and 3278L (Immokalee Basin). The western portion of the watershed, on the other hand, provides relatively little opportunity for measurable ecological improvement. Urban development in this portion of the watershed significantly limits the degree to which any projects would improve ecological values outside of the footprint of the projects themselves. 4.2. Golden Gate - Naples Bay Watershed More than 75 percent of the Golden Gate - Naples Bay watershed (almost 70,000 acres) has scores less than 4 for all three functional assessment indices (Table 4 -1). The low scores reflect the conversion of almost 70 percent of this watershed to urban or agricultural development as summarized in the Technical Memorandum for Task 3.1 (Reference Period Comparison). The area with the highest functional value is the less - developed portion of Northern Golden Gate Estates (WBID 3278S), east of the CR951 Canal (see Figures 4 -1 through 4 -4 and Table 4 -3). Reflecting the relatively lower intensity of land uses in this portion of the watershed, almost 25 percent of this WBID (17,000 acres) has LSI scores of 7 or higher and vegetation scores of 8 or higher, and over 15 percent (13,000 acres) has a hydrology score of 10. Overall, this watershed presents relatively few opportunities for large -scale improvement in ecological value. Urban and suburban development throughout the watershed limits the degree to which restoration projects would improve functional values beyond the footprint of the project itself. In relation to other portions of the watershed, the eastern portion of WBID 3278S (Northern Golden Gate Estates) presents the greatest opportunity for ecological restoration. The relatively lower intensity land uses in this portion of the 11 Element 1 3.2 Functional Assessment watershed may allow restoration projects to improve ecological values on a wider scale, particularly in the northern areas adjacent to the Corkscrew Swamp system. 4.3. Rookery Bay Watershed The functional assessment values of this watershed reflect low scores in the portions of the watershed surrounding Belle Meade and Tamiami Trail, but overall relatively higher functional values than in the other two primary watersheds (Table 4 -1). More than 50 percent (50,000 acres) of the Rookery Bay Watershed has an overall functional value greater than 7. This is primarily because less than 30 percent of the watershed has been converted to urban or agricultural uses, as reported in the Technical Memorandum for Task 3.1 (Reference Period Comparison). Of the three scoring criteria, the hydrological scores are the relatively lowest, due to ecosystem changes in the area north of Belle Meade. LSI scores in the Rookery Bay Watershed (Figure 4 -1) reflect high- intensity land uses along the Tamiami Trail corridor and moderate - intensity land use in the Belle Meade area. At the WBID level (Table 4 -4), this results in relatively lower LSI values in WBID 3278Y (Rookery Bay - Inland West Segment) than in other WBIDs in this watershed. Almost 75 percent (11,000 acres) of WBID 3278Y has an LSI less than 7, while slightly more than 10 percent (3,000 acres) of WBID 3278U (Partial— Rookery Bay Coastal Segment) and just under 20 percent (10,000 acres) of WBID 3278V (Rookery Bay - Inland East Segment) have LSI scores less than 7. Vegetation score distribution (Figure 4 -2) likewise reflects the relatively high proportion of undeveloped lands in this watershed other than the Belle Meade area and Tamiami Trail corridor. Over 65 percent (65,000 acres) of this watershed has a vegetation score of 8 or higher. Among WBIDs (Table 4 -4), 70 percent (over 10,000 acres) of 3278Y has a vegetation score of 3 or less, while the other two WBIDs each have vegetation scores of 8 or higher for over 70 percent (58,000 acres) of their area. Hydrology scores (Figure 4 -3) are similar to vegetation scores, with low scores in the Tamiami Trail corridor and Belle Meade area. A notable difference from the geographic distribution of vegetation scores is the presence of relatively lower hydrology scores in the area north of Belle Meade south of Alligator Alley /I -75. Comparing the PDVM to current FLUCCS data shown in the Technical Memorandum for Task 3.1 (Reference Period Comparison), this portion of the watershed once supported swamp forest and is now dominated by shorter - hydroperiod hydric flatwoods. As a result of this shift, almost 60 percent (30,000 acres) of WBID 3278V has a hydrology score of 3 or lower. The high existing functional value of lands throughout much of this watershed results in limited opportunities for ecological improvement. The areas with greatest opportunity for improvement in functional assessment values include the northern Belle Meade area and areas bordering the western and eastern edges of the Tamiami Trail development corridor. F%1 12 Element 1 P 3.2 Functional Assessment 4.4. Faka Union, Okaloacoochee/SR 29, and Fakahatchee Watersheds These watersheds, individually and as a whole, retain relatively high functional value, particularly the areas south of I -75 Alligator Alley dominated by undeveloped lands within and adjacent to the Big Cypress Preserve. Vegetation shifts and related hydrological changes north of Alligator Alley are responsible for lower scores in the northern portion of these watersheds. Overall, almost 65 percent (335,000 acres) of the land within these watersheds has a functional value of 7 or higher (Table 4 -1). Almost 95 percent (470,000 acres) of the land in these watersheds has an LSI value of 7 or higher. The relatively lowest -LSI value WBID is 3278H (Faka Union North Segment), with just under 40 percent of that WBID having an LSI less than 7. Relatively lower scores in this area reflect low- density rural development north of Alligator Alley in the eastern portion of Golden Gate Estates. Each of the other WBIDs have LSI scores of 7 or higher for 80 percent or more of their area. Vegetation and hydrology scores (Figures 4 -2 and 4 -3, Table 4 -5) generally follow a north to south and west to east pattern, with highest scores in the southern and eastern portions of this area. WBID 3278H, the northern portion of the Okaloacoochee -SR29 Watershed (WBIDs 3278T- Okaloacoochee Slough and 3278W- Silver Strand), and the northern portion of the Fakahatchee Watershed (3259I -Camp Keais) comprise the area of lowest scores. These WBIDs all have vegetation scores less than 4 for at least 40 percent of their area and hydrology scores of 3 or less for at least 50 percent of their area. These scores reflect lower functional value due to widespread conversion of these areas to agricultural and low- density rural land uses. As with the other watersheds, the greatest opportunity for measurable improvement in functional value occurs in areas with relatively high LSI and low vegetation and hydrology scores. In this trio of watersheds, WBIDs with this combination of scores (Table 4 -5) include 3259I (Camp Keais), 3278H (Faka Union North Segment), 3278T ( Okaloacoochee Slough), and 3278W (Silver Strand). Projects in these WBIDs that target restoration of natural hydrology and vegetation, particularly in areas adjacent to lands with existing high functional value, will result in the greatest large -scale improvements in functional value. 13 Element 1 � 3.2 Functional Assessment Table 4 -1. Functional Assessment of Collier County Watersheds LSI Cocohatchee- Corkscrew Golden Gate Naples Bay Rookery Bay Faka Union, Fakahatchee, OK -29 Acres of Watershed Acres % of W'shed Acres % of W'shed Acres % of Watershed < 4 11,648 9 25,142 28 7,089 7 1,627 0 4 - < 7 23,257 17 1 46,475 51 18,246 18 30,101 6 7 - < 10 68,221 51 18,827 21 27,454 28 193,156 38 r 10 31,237 23 723 1 45,945 47 277,776 55 otal Ac 134,362 91,167 98,735 502,660 Vegtn. Score Cocohatchee- Corkscrew Golden Gate Naples Bay Rookery Bay Faka Union, Fakahatchee, OK -29 Acres of Watershed Acres % of W'shed Acres % of W'shed Acres % of Watershed 0 70,183 52 67,859 74 27,596 28 135,240 27 3 1,511 1 1 1,885 2 2,131 2 2,996 1 6 4,378 3 1,756 2 4,119 4 17,149 3 8 17,472 13 8,665 10 29,481 30 137,720 27 10 40,819 30 11,002 12 35,408 36 209,556 42 Total Ac 134,362 91,167 98,735 502,660 Hydro. Score Cocohatchee- Corkscrew Golden Gate Naples Bay Rookery Bay Faka Union, Fakahatchee, OK -29 Acres / of Watershed Acres % of W'shed Acres /o of W'shed Acres % of Watershed 0 71,759 53 69,783 77% 29,907 30% 140,522 28% 1 504 0 13 0% 0 0% 1,627 0% 2 5,927 4 4,042 4% 13,946 14% 25,297 5% 3 3,616 3 2,286 3% 4,132 4% 20,687 4% 5 2,479 2 52 0% 620 1% 48,657 10% 6 362 0 39 0% 1,730 2% 4,558 1% 10 49,716 37 14,954 16% 48,399 49% 261,312 52% Total Ac 134,362 91,167 98,735 502,660 Avg Functn Cocohatchee- Corkscrew Golden Gate Naples Bay Rookery Bay Faka Union, Fakahatchee, OK -29 Acres of Watershed Acres % of W'shed Acres % of W'shed Acres % of Watershed 0 - < 1 3,603 3% 9,155 10% 2,492 3% 258 0% 1 - < 4 1 67,549 50% 60,253 66% 25,762 26% 136,389 27% 4 - < 7 11,054 8% 7,012 8% 18,427 19% 41,090 8% 7 - < 10 27,027 20% 14,282 16% 27,686 28% 162,151 32% 10 25,129 19% 465 1% 24,367 25% 162,771 32% Total Ac 134,362 91,167 98,735 502,660 14 Element 1 * 3.2 Functional Assessment Figure 4 -1. LSI Value of Collier County Watersheds 15 Element 1 FIMIX 3.2 Functional Assessment 0 Figure 4 -2. Vegetation Score of Collier County Watersheds 16 Element 1 WII& 3.2 Functional Assessment Figure 4 -3. Hydrology Score of Collier County Watersheds Hydrology - Functional Assessment LEE CO. 'ACC CO. JI Pffil 17 Element 1 3.2 Functional Assessment Figure 4 -4. Average Functional Assessment Value 18 Element 1 3.2 Functional Assessment Table 4 -2. Functional Assessment of Coco hatchee- Corkscrew Watershed. by WBID 19 Element t �, 3.2 Functional Assessment 3259A 32596 3259W 32592 32780 3278D 3278E 3278F 3278L LSI Acres Percent of WBID Acres Percent of WBID Acres. Percent of WBID Acres Percent of WBID. Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID < 4 69 2 168 1 0 0 499 79 796 37 7,966 31 91 1 13 0 1,338 15 4-<7 776 25 2,788 13 0 0 129 20 1,034 48 8,692 34 1,792 15 3,343 6 3,105 36 7 - <10 1,920 62 17,649 82 214 14 7 1 325 15 5,878 23 9,440 80 25,110 47 4,274 49 10 323 10 972 5 1,276 86 0 0 0 0 3,302 13 454 4 24,448 46 28 0 Total Acres 3,088 21,576 1,490 635 2,155 25,837 11,777 52,914 8,745 3259A 3259B 3259W 3259Z 3278C 3278D 3278E 3278E 3278L Veg Score Acres Percent of WBID Acres Percent of WBID Acres Percent I of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 758 25 14,757 68 0 0 597 94 1,503 70 16,241 63 7,906 67 18,158 34 6,331 72 3 52 2 116 1 0 0 26 4 48 2 840 3 26 0 102 0 189 2 6 191 6 618 3 46 3 0 0 61 3 674 3 408 3 1 1,736 3 425 5 8 248 8 3,153 15 1 0 11 2 321 15 2,452 9 1,783 15 7,878 15 985 11 10 1,839 60 2,931 14 1,443 97 0 0 221 10 5,631 22 1,653 14 25,040 47 815 9 Total Acres 3,088 21,576 1,490 635 2,155 25,837 11,777 52,914 8,745 3259A 3259B 3259W 3259Z 3278C 3278D 3278E 3278F 32781 Hydro Score Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID ...Acres Percent of WBID Acres Percent of WBID 0 810 26 14,978 69 0 0 623 98 1,555 72 16,906 65 8,057 68 18,376 35 6,441 74 1 0 0 148 1 0 0 0 0 3 0 245 1 13 0 65 0 13 0 2 11 0 503 2 0 0 2 0 13 1 1,578 6 697 6 2,538 5 411 5 3 120 4 1,245 6 0 0 0 0 38 2 300 1 429 4 1,098 2 204 2 5 0 0 948 4 0 0 0 0 0 0 13 0 295 3 1,003 2 142 2 6 204 1 7 0 0 46 3 0 0 0 0 5 0 70 1 1 0 0 0 10 1,943 1 63 3,754 17 1,444 97 9 1 545 25 6,789 26 2,215 19 29,833 56 1,533 18 Total Acres 3,088 21,576 1,490 635 2,155 25,837 11,777 52,914 8,745 3259A 32598 3259W 32592 3278C 3278D 3278E 3278F 3278L Avg Func Val Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID < 1 0 0 39 0 0 0 303 48 162 8 2,053 8 0 0 0 0 846 10 1 -<4 810 26 14,731 68 0 0 321 51 1,380 64 14,739 57 7,932 67 18,233 34 5,595 64 4 -<7 215 7 2,101 10 6 0 2 0 80 4 2,320 9 1,310 11 3,914 7 676 8 7 - <10 1,778 58 4,230 20 226 15 9 1 532 25 4,211 16 2,209 19 10,802 20 1,615 18 10 284 9 475 2 1,257 84 0 0 0 0 2,514 1 10 326 3 19,965 1 38 13 0 Total Acres 3,088 21,576 1,490 635 2,155 25,837 11,777 52,914 8,745 19 Element t �, 3.2 Functional Assessment Table 4 -3 Functional Assessment of Golden Gate - Naples Bay Watershed, by WBID LSI 3278k 3278R 32785 Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID <4 3,413 63 5,474 59 15,136 21 4-<7 1,818 34 2,773 30 40,850 56 7-<10 181 3 1,065 11 16,242 22 10 0 0 1 0 555 1 Total Acres 1 5,412 9,313 72,784 Veg Score 3278k 3278R 32785 Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 4,646 86 7,661 82 53,421 73 3 67 1 347 4 1,445 2 6 116 1 2 452 5 1,130 1 2 8 277 5 174 2 7,663 11 10 306 6 680 7 9,125 13 Total Acres 5,412 9,313 72,784 Hydro Score 3278k 3278R 3278S Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 4,700 87 7,943 85 54,947 75 1 0 0 0 0 13 0 2 199 4 18 0 3,732 5 3 26 0 510 5 1,521 2 5 0 0 0 0 52 0 6 0 0 31 0 0 0 10 487 9 811 9 12,521 17 Total Acres 5,412 9,313 72,784 Avg Func Val 3278k 3278R 32785 Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID < 1 1,624 30 3,157 34 3,662 5 1-<4 3,050 56 4,825 52 50,934 70 4-<7 315 6 527 6 5,810 8 7-<10 422 8 803 9 12,017 17 10 0 0 1 0 362 0 Total Acres 5,412 9,313 72,784 20 Element 1 3.2 Functional Assessment Table 4 -4. Functional Assessment of Rookery Bay Watershed, by WBID LSI 3278U* 3278V 3278Y Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID < 4 778 3 1,530 3 4,307 29 4 - <7 2,445 9 8,873 16 6,568 44 7 - <10 6,681 26 16,422 30 3,337 22 10 1 16,267 1 62 1 27,166 1 50 1 843 1 6 Total Acres 1 26,171 1 53,991 1 15,055 Veg Score 3278U* 3278V 3278Y Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 4,036 15 13,293 25 9,506 63 3 243 1 834 2 1,054 7 6 1,252 5 2,365 4 353 2 8 7,160 27 19,773 37 1,900 1 13 10 13,480 52 17,726 1 33 2,242 1 15 Total Acres 26,171 53,991 15,055 Hydro Score 3278U* 3278V 3278Y Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 4,372 17 14,967 28 9,751 65 1 0 0 0 0 0 0 2 293 1 12,319 23 1,034 7 3 208 1 3,329 6 459 3 5 0 0 607 1 0 0 6 1,534 6 117 0 24 0 10 19,764 76 22,653 42 3,787 25 Total Acres 26,171 53,991 15,055 Avg Func Val 3278U* 3278V 3278Y Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID <1 126 0 544 1 1,557 10 1-<4 3,988 15 13,059 24 8,220 55 4-<7 847 3 15,217 28 1,874 12 7-<10 10,873 42 12,874 24 2,926 19 10 10,338 40 12,298 23 478 3 Total Acres 26,171 53,991 15,055 * The data for this WBID only includes the portion within the Rookery Bay Watershed 21 Element 1 PWIII 3.2 Functional Assessment Table 4 -5. Functional Assessment of Faka Union, Okaloacoochee/SR29, and Fakahatchee Watersheds, by WBID * Note: The data for this WBID only includes the portion that lies within the Faka Union, Fakahatchee and Okaloacoochee /SR 29 Sub- basins 22 Element t 4*, 3.2 Functional Assessment 32591 3259M* 32610 327BG 3278H 32781 3278T 3278W 151 Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID < 4 103 0 0 0 0 0 0 0 909 3 0 0 39 0 431 1 4 - <7 10,039 18 405 1 32 0 111 0 9,640 35 52 0 4,181 3 5,015 9 7 - <10 29,842 54 2,321 5 3,427 10 4,449 5 15,269 56 3,545 6 84,637 67 46,111 86 10 15,721 28 41,211 94 29,907 90 89,934 95 1,631 6 55,853 94 37,123 29 2,273 4 Total Acres 55,706 43,938 33,365 94,494 27,449 59,450 125,980 53,830 32591 3259M* 32610 3278G 3278H 32781 3278T 3278W Veg Score Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 23,076 41 413 1 309 1 445 0 11,981 44 434 1 57,736 46 38,332 71 3 981 2 108 0 105 0 188 0 395 1 86 0 121 0 978 2 6 2,252 4 160 0 2,578 8 1,294 1 1,068 4 1,994 3 5,070 4 2,377 4 8 9,763 18 8,766 20 20,091 60 25,927 27 6,523 24 34,946 59 24,417 19 5,083 1 9 10 19,633 35 34,491 79 10,282 31 66,641 71 7,482 27 21,990 37 38,636 31 7,060 1 13 Total Acres 55,706 43,938 33,365 94,494 27,449 59,450 125,980 53,830 32591 3259M* 3261C 3278G 3278H 32781 3278T 3278W Hydro Score Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID 0 23,829 43 559 1 453 1 870 1 13,016 47 871 1 58,895 47 39,451 73 1 232 0 0 0 406 1 33 0 13 0 47 0 709 1 172 0 2 3,949 7 0 0 446 1 2,180 2 2,303 8 10,859 18 3,832 3 1,179 2 3 3,777 7 0 0 970 3 2,859 3 1,718 6 3,464 6 6,386 5 1,345 2 5 444 1 j 1,156 3 16,575 50 15,869 17 65 0 3,454 6 8,839 7 1,249 2 6 25 0 1,930 4 0 0 1,441 2 0 0 786 1 2 0 11 0 10 23,449 42 40,292 92 14,514 44 71,243 75 10,334 38 39,969 67 47,318 38 10,424 19 Total Acres 55,706 43,938 33,365 94,494 27,449 59,450 125,980 53,830 32591 3259M* 3261C 3278G 3278H 32781 3278T 3278W Avg FuncVal Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID Acres Percent of WBID <1 0 0 0 0 0 0 0 0 0 0 0 0 13 0 187 0 1 - <4 23,150 42 508 1 388 1 585 1 12,236 45 455 1 57,775 46 38,830 72 4-<7 6,734 12 90 0 1,216 4 2,762 3 4,834 18 11,605 20 9,967 8 3,109 6 7 -<10 15,500 28 11,057 25 22,399 67 29,724 31 9,616 35 26,479 45 33,708 27 10,654 20 10 10,322 19 32,282 73 9,362 28 61,423 65 762 3 20,911 35 24,517 19 1,050 2 Total Acres 55,706 43,938 33,365 94,494 27,449 59,450 125,980 1 53,830 * Note: The data for this WBID only includes the portion that lies within the Faka Union, Fakahatchee and Okaloacoochee /SR 29 Sub- basins 22 Element t 4*, 3.2 Functional Assessment 4.5 Ecological Capacity for Additional Water Storage Methodology and Results The potential for hydrological storage provides information useful for evaluating watershed conditions and restoration opportunities. Restoration of an area that is currently an upland but was previously a wetland, for instance, would require development of storage capacity to a depth and duration typical of the pre - development wetland. Storing water in restored wetlands could also cleanse and attenuate freshwater flows to downstream estuaries, depending on the location and morphology of those wetlands. This section describes both the methodology and results of calculating potential water storage. The difference between the existing depth and duration of inundation (based on FLUCCS vegetation type) and the pre - development depth and duration of inundation (based on PDVM vegetation type) may be considered as the ecological capacity for additional water storage. Adding water beyond this amount would potentially exceed the hydrological tolerance of the pre - development vegetation community and result in a transition to a different type of wetland or open water system that was not previously present on that site. For example, adding too much water for too much time to a current upland site that was a wet prairie could result in creation of a deepwater marsh or unvegetated pond rather than restoration of the pre - development wet prairie. This scenario would be ecologically and logistically risky. For the purposes of this project, the change in depth of inundation is estimated in inches based on comparisons between the typical water levels of existing vegetation and PDVM vegetation. Table 4 -6 provides input data for the equation used for this portion of the technical analysis. The calculation developed for this purpose of defining the available ecological capacity for additional storage is: Capacity for Additional Storage = WSWLPOVM- WSWL2oo7, where WSWL is the long -range average wet season water level typical for the type of ecological community. Similarly, the change in duration of inundation between PDVM and current FLUCCS vegetation is calculated in months, based on the hydroperiods duration per Table 3 -2. Due to the close relationship between vegetation community and hydrology, the results of the calculations for depth and duration of inundation are displayed and summarized together. Table 4 -7 summarizes the application of the equations. A comparison of the hydrological characteristics of pre - development and 2007 vegetation communities (Figure 4 -5) suggests areas for potential additional wet season water storage (Figure 4 -6). Overall, approximately 44,000 acres of undeveloped lands (including over 10,000 acres in Rookery Bay watershed) have capacity for additional wet season storage of at least 0.5 feet up to over 2.5 feet (Table 4 -7). The largest opportunity for storage, based strictly on the difference in hydrological characteristics between pre - development and 2007 vegetation communities, is the portion of the Rookery Bay watershed north of Belle Meade. Restoration of hydrology in these areas, in combination with ecological restoration within the northern portion of Belle 23 Element 1 3.2 Functional Assessment i Meade itself, could lead to large -scale improvements in both functional value and hydrological storage. Not included in this assessment is the potential benefit tc downstream estuaries as a result of attenuating freshwater flows. To the extent that improved storage in northern Belle Meade would restore healthier salinity regimes in downstream estuaries, this would further contribute to the ecological value of such projects. Table 4 -6. Ecological Capacity for Additional Storage s � ' > FLV Ad it ta4 i torw, Cap city Open Water Freshwater marsh, cypress, or > 1 foot swamp forest — Open Water Wet prairie, dwarf /scrub > 1.5 feet — cypress Open Water Hydric flatwood, hydric > 2 feet hammock — Open Water Mesic flatwood, mesic > 2.5 feet hammock — Open Water Xeric flatwood, xeric hammock > 4 feet Any Developed 0 Freshwater Marsh, Cypress, or Swamp Wet prairie, dwarf cypress 0.5 -1 foot Forest Freshwater Marsh, Hydric flatwood, hydric Cypress, or Swamp hammock 1 -1.5 feet Forest Freshwater Marsh, Mesic flatwood, mesic Cypress, or Swamp hammock 1 -2 feet Forest Freshwater Marsh, Xeric flatwood, xeric Cypress, or Swamp hammock >_ 3 feet Forest Any natural system Same system 0 FW 24 Element 1 3.2 Functional Assessment Table 4 -7. Ecological Capacity for Additional Water Storaae potential Addthonal Sttc> cock "'ecc Co drew Wden Gate Na lei "Ba lloo)�ery. �Vatexsleis Total 0.5 - 1 foot 277 75 694 2,042 3,087 0.5 -1 ft 285 7 42 1,919 2,254 0.5 - 1 ft 571 14 84 3,839 4,508 1 - 1.5 feet 2,071 2,026 7,673 8,612 20,381 1 - 2 feet 677 472 1,611 3,935 6,695 < =0.5 ft 292 21 219 6,304 6,837 > =1 foot 7 2 5 80 94 > =2 feet 1 0 0 0 1 > =2.5 feet 0 1 5 3 10 n/a (developed) 50,200 55,029 21,619 74,047 209,030 4.7 Non - Native Invasive Vegetation — Methodology and Results The presence of non - native invasive vegetation can significantly degrade wildlife habitat functions, as documented by many studies, including studies specific to southwestern Florida (e.g., Myers, 1975). Due to the potentially significant impact of non - native invasive species at a watershed level, several data sources, government agencies, and non- profit organizations were consulted to determine the availability of comprehensive, County -wide, accurate GIS coverages of non - native exotic vegetation. However, no GIS data layers were found that provide a sufficiently comprehensive and accurate coverage of the six watersheds to incorporate these into the functional assessment methodology. The two best sources of identified data are the Florida Natural Areas Inventory (FNAI) Florida Invasive Plants Geodatabase (FLInv) for public lands and the Early Detection and Distribution Mapping System (EDDMapS) for private lands. Due to the limited extent of both of these data layers, non- native invasive vegetation was not included in the calculation of watershed -wide functional values. Instead, data from these two sources are mapped and discussed separately from the functional assessment, as well as suggestions for obtaining additional GIS data for this purpose. The data presented in Table 4 -8 and Figures 4 -7 and 4 -8 represent the most up -to -date and accurate GIS sources available at this time. Due to the lack of comprehensive non- native invasive species data on private lands, the most suitable use of these GIS data sources is to evaluate the ecological effects of non - native invasive species on publicly managed lands, in combination with the other factors described earlier in this Technical Memorandum. The public lands with the greatest extent of non - native invasive species on these maps are the Belle Meade and western Corkscrew Swamp areas. Comparing the non - native invasive species maps to the functional assessment and hydrological storage data for these two areas, the greatest opportunity for multi- function improvement on public lands occurs in northern Belle Meade. Projects in this area would achieve improvements in overall functional value (particularly if coupled with restoration of adjacent private lands), large 25 Element 1 J 3.2 Functional Assessment potential improvements in hydrological storage, and improvements in natural vegetation communities. A more thorough analysis and comparison that incorporates non - native invasive species coverage is only possible with the development of additional GIS coverages over private lands. The primary options include remote sensing via multi - spectral imagery coupled with unsupervised classification and a more detailed mapping via hyperspectral imagery, LiDAR, and supervised classification based on existing known non - native invasive vegetation data points. Multispectral imagery and unsupervised classifications can be expected to achieve overall accuracy of 60 — 70 percent. A more detailed and accurate mapping of non - native invasive vegetation can be acquired through use of hyperspectral imagery, LiDAR and supervised classifications. Table 4 -8. Gross Acres of Non - native invasive Species on Publicly Managed Lands (Source: FNAI PR%R 26 Element 1 3.2 Functional Assessment Downy (old Wat��d Brazf�fan Cogon Rese- Mel�tr�ea World " p er brass . C100bing Fern Cocohatchee- Corkscrew 16,052 3,041 3,747 13,246 11,942 Golden Gate Naples Bay 985 37 828 829 Rookery Bay 1,674 1 166 8,438 421 Faka Union, Fakahatchee, Okaloacoochee -SR29 6,415 271 0 206 106 Total Area 25,125 3,313 3,950 22,719 13,298 PR%R 26 Element 1 3.2 Functional Assessment Figure 4 -5. Hydrology of Pre - Development and 2007 Vegetation (Source Data from M. Duever and SFWMD) Pre - Development Based Hydrology 2007 Vegetation -Based 4 27 Element 1 3.2 Functional Assessment ir Legend ar 25 5Mk, Hydrology Hydropenod Wet Season (months) Water Level (inches) 0 <_ -24 x 1 -2 2 -6 2-6 6 -12 6 -10,4„ 12 -24 8 -10 18 -24 -10 24 Tidal . Tidal Urban /Agriculture OWatershed Boundary a county Boundary A 4 27 Element 1 3.2 Functional Assessment ir Figure 4 -6. Ecological Systems' Wet Season Water Storage Potential Potential Ecological Tolerance for Wet Season Water Storage Legend —0.5 ft 0.5 -1 ft >=1 ft 1 -1.5ft i 1 -2ft it > =2 ft 'f —2.5 ft Watershed Boundary I� County Boundary m 0 1.5 3 Miles - (DI -I .IER C , � MONROE (�O. r N �a As 28 Element 1 PM) 3.2 Functional Assessment HF-.NDRV r 4^. y j � xj m 9 a• � S Cocohatc`hee- C.,k -rew a �• z Okaloacochee -SR29 - S 2. 4` g .am w Golden GateNaples;B ay Nap les`��.�: Legend —0.5 ft 0.5 -1 ft >=1 ft 1 -1.5ft i 1 -2ft it > =2 ft 'f —2.5 ft Watershed Boundary I� County Boundary m 0 1.5 3 Miles - (DI -I .IER C , � MONROE (�O. r N �a As 28 Element 1 PM) 3.2 Functional Assessment Figure 4 -7. Non - native invasive Species on Public Lands (Source: FNAI) 29 Element 1 * 3.2 Functional Assessment Figure 4 -8. Non - native invasive Species Observation —Point Data (Source: EDDMapS) i Exotic Vegetation - Observed HENDRY CO f IN ,'-,w a L.EE.. CO. - .. "...... i d :yam Legend Exotic Veg Brazilian Pepper Cogon Grass Downy Rose -Myrtle « Melaleuca Old World Climbing Fern Watershed Boundary County Boundary IT1 0 2 4 Miles COt_LJER CO p MONROE CO N A 30 Element 1 3.2 Functional Assessment 5.0. Conclusions The results of the functional assessment indicate significant degradation of ecological values in the Golden Gate - Naples Bay watershed, as well as the western Cocohatchee- Corkscrew watershed, Belle Meade portion of Rookery Bay watershed, and northern portions of the other basis. In all cases, the low functional value of these areas is due to conversion of natural lands to development or agricultural land uses. One potential application of the functional assessment method developed for this project is identifying projects in locations with a moderate to high LSI value but low vegetation and hydrology value. A low LSI value indicates locations that are surrounded by land uses that substantially interfere with ecological functions; whereas higher LSI values occur on lands that are adjacent to low- intensity or natural lands that present less of an obstacle to ecological restoration. Reviewing the watershed -wide maps and data, some areas with this combination of factors include lands north and east of Corkscrew Swamp and Lake Trafford; portions of Northern Golden Gate Estates; the northern Belle Meade area; and eastern areas of the County north of Alligator Alley such as the Immokalee area. Additional synergistic benefits would occur through restoring wetlands to improve wet season water storage and removing non- native invasive species. With the currently available GIS data, one area with potentially significant restoration opportunities is the northern Belle Meade area, both on the public lands as well as the northern part of the private lands. Development of additional GIS data layers, particularly non - native invasive vegetation, would help to further pinpoint areas of greatest multi- function restoration potential. The methodologies and results described in this memorandum also serve as ecological performance measures. The LSI, hydrological scoring index, vegetation scoring index and overall average functional assessment value each provide existing baseline data, as well as numerical methods for evaluating and comparing the effects of proposed projects. These measurements can be applied on a small -scale (i.e., individual polygons) or larger watershed level, and further supplemented by the data layers for ecological capacity for water storage and non - native invasive vegetation for portions of each of these watersheds. ■ %J 31 Element 1 3.2 Functional Assessment 6.0. Bibliography Bales, J.D., J.M. Fulford, E Swain, 1997. Review of Selected Features of the Natural System Model, and Suggestions for Applications in South Florida. U.S. Geological Survey Water- Resources Investigations Report 97 -4039. Bardi, E., M. T. Brown, K. C. Weiss, and M. J. Cohen. Uniform Mitigation Assessment Method. Web -based training manual for Chapter 62 -345, FAC for Wetlands Permitting. htto: /!www.dep.state.fl.us/ water /wetlands /mitidation!umarn.htm Brown, M.T., and M.B. Vivas. 2005. Landscape Development Intensity Index. Environmental Monitoring and Assessment (2005) 101: 289 -309. Duever, M. 2004 Southwest Florida Pre - Development Vegetation Map. South Florida Water Management District. Myers, R.L. 1975. The Relationship of Site Conditions to the Invading Capability of Me/aleuca Ouinquenervia in Southwest Florida. Masters Thesis, University of Florida. Reiss, K.C., E. Hernandez, M. T. Brown, 2007. An Evaluation of the Effectiveness of Mitigation Banking in Florida: Ecological Success and Compliance with Permit Criteria. Final Report Submitted to the Florida Department of Environmental Protection Under Contract #WM881 and United States Environmental Protection Agency Region Four Under Contract #CD 96409404 -0. Reiss, K.C., M.T. Brown, C.R. Lane, 2009. Characteristic community structure of Florida's subtropical wetlands: the Florida wetland condition index for depressional marshes, depressional forested, and flowing water forested wetlands. Wetlands Ecol Manage DOI 10.1007/sl 1273- 009 - 9132 -z. Springer Science +Business Media B.V. 2009 FM 32 Element 1 3.2 Functional Assessment :1 Technical Memorandum To: Mac Hatcher, PM Collier County From: Moris Cabezas, PBS &J Peter deGolian, PBS &J Date: August 11, 2009 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 1, Task 4: Initial Model Comparison and Estimate of Flow to Estuaries 1.0 Introduction The purpose of this technical memorandum is to provide a preliminary assessment of historical fresh water discharges from the County watersheds into the receiving estuaries. It addresses two major topics, a literature review and a comparison of simulation results of MIKE SHE computer models developed for the Big Cypress Basin Project Implementation Report (PIR). The three MIKE SHE models were developed for the PIR to evaluate the potential benefits of restoring the Southern Golden Gates Estates (SGGE) area of Collier County. This project is now referred to as the Picayune Strand Restoration Project (PSRP). The three models include an existing conditions model that is based on year 2000 land use, a future conditions models that is based on year 2050 land use, and a pre - development (or natural systems) model developed for the Southwest Florida Feasibility Study (SWFFS). Each of the PIR models was originally developed using the software version 2000 and were later updated to run with version 2003. For this analysis, PBS &J ran each of the models using version 2009 of the software. 2.0 Literature Review In order to adequately define future water management strategies, it is necessary to understand the history of water management in Collier County. Task 1 of Element 1 of the Collier County Watershed Model Update and Plan Development is to complete a review of literature related to flows and discharges from Collier County watersheds. For this task, PBS &J reviewed more than 50 documents that are listed in the bibliography. This section summarizes 11 documents that were found to provide the most information in describing the historical hydrology and flow conditions in Collier County. It is noted that in many of the older documents, the Faka Union Canal Basin is referred to as the Fahka Union Basin 1 Element 1 Pffil Initial Model Comparison and Estimate of Flow to Estuaries 2.1 Summary of Relevant Literature Following are summary descriptions of the relevant documents identified as part of this task. Davis, John H. October 1943. The Natural Features of Southern Florida, Especially the Vegetation, and the Everglades. Florida Geological Survey Bulletin No. 25. This bulletin describes some of the cultural history and the main physical and biological features of South Florida prior to major development and construction of the existing drainage network, although it does not provide quantified estimates about historic flows or water levels in Collier County. In this document, Collier County is described as consisting of three physiographic regions; the Flatlands, the Big Cypress Swamp, and the Southwest Coast and Ten Thousand Islands (Figure 2 -1). Davis states that the county is 2,025.5 square miles in size, making it the largest land mass county east of the Mississippi River. Figure 2 -1 Physiographic Regions of Collier County, Florida (from Davis 1943) N The Flatlands region is described as consisting mainly of low, nearly flat to gently rolling land with some rivers dissecting the plains. There are many small ponds, sloughs and other depressions. The Collier County portion of the Flatlands regions is less well drained and of lower elevation than portions of the Flatlands region to the north. Another feature of the Flatlands region is the great number of marsh, swamp, and open -water depressions including Lake Trafford and the Corkscrew marsh. The Big Cypress Swamp region was described as covering about 1,200 square miles most in Collier County with small areas in southeastern Hendry County and northern Monroe County. Davis describes the chief characteristics of the Big Cypress as "vegetational with an abundance of the cypress and mixed swamps of large trees, open elongated forest of 2 Element 1 1%) Initial Model Comparison and Estimate of Flow to Estuaries cypress and medium sized trees, are large areas of scrubby stunted cypress trees growing in marsh -like seasonally wet prairies. The region is of low elevation, low relief and very confused drainage. Most of it lies between elevations of 5 and 20 feet. A number of sloughs drain the Big Cypress, some draining to the Gulf of Mexico, and others into the Everglades. Most of the west part drains toward the south through the Fathahatchee Swamp." The Southwest Coast and Ten Thousand Islands regions is described as a very low -lying coastal region of small shoal -water islands, It is one of the most dissected coastal regions of Florida and one of the least accurately known due to dense mangrove swamps. These mangrove swamps and salt -water marshes are among the largest in the world. The relatively high tidal range causes the tidal inundation of large areas far inland and forces salt water far up the estuaries. Kenner, W. E., 1966, "Runoff in Florida ", Map Series No. 22, U.S. Geologic Survey In 1966, the United States Geologic Survey and W. E. Kenner produced Map Series No. 22 titled, "Runoff in Florida." This map suggests that the total runoff from the Collier County area at that time was between 0 — 10 inches annually. Klein, H., W.J. Schneider, B.F. McPherson and T.J. Buchanan. May 1970. Some Hydrologic and Biologic Aspects of the Big Cypress Swamp Drainage Area, Southern Florida. United States Geologic Survey Open -file Report 70003. In May 1970, the United States Geologic Survey and specifically, H. Klein, W.J. Schneider, B.F. McPherson and T.J. Buchanan published Open File Report 70003 entitled, "Some Hydrologic and Biologic Aspects of the Big Cypress Swamp Drainage Area, Southern Florida." The prime purpose of the report was to determine the importance of the Big Cypress in maintaining an adequate water supply for (1) the Everglades National Park, for (2) the expanding population of southwest Florida, and for (3) the adjacent estuaries, which constitute nurseries for fish. For this report, the Big Cypress was divided into three subareas as shown in Figure 2 -2. Each subarea has a reasonably distinct internal drainage determined largely by topographic configuration and man -made drainage. Subarea A lies northeast of a low ridge and drains southeastward into Conservation Area 3 of the Central and Southern Florida Flood Control District. Subarea B includes approximately 550 square mile at the west edge of the Big Cypress. It is characterized by an extensive system of canals, which drain southward and westward into the Gulf Coast estuaries. This canal system includes primarily the Golden Gate Estates canal system. Subarea C occupies the central part of the Big Cypress and drains toward the Everglades National Park. It consists of about 1,450 square miles. 3 Element 1 Initial Model Comparison and Estimate ,i of Flow to Estuaries Figure 2 -2 Map of the Big Cypress showing the delineation of the drainage area and the subareas as defined by Klein 91070) 43 1 30y N� r. ` 1Y n t \ � IS I I 2{Q0 ASS i i >mSa� � —1 ti EXPLANATION *OAS CANA' LEYEE COUNIY Bt1UNOARY PAR. BoUNOMY 506WA -4Y OF 810 CYPRESS ORAINAO[ AMA BOUNDARY OFTWFFN SUSAPEA A AREA Ot SIGNA"ON i!' 30' 30' A0' 3d x 21U �1 rt i • tv{AGt AR 6f 13' A0` 46' Klein stated that during the rainy season, shallow depressions fill with water and, because of the poor drainage, water stands on the land until it evaporates or slowly drains off. Thus, as much as 90 percent of the undrained part of the Big Cypress is inundated to depths ranging from a few inches to more than three (3) feet at the height of the rainy season. Klein stated that in southern Florida, land development usually began with the construction of canals to drain swampy land and to assure protection from high water during the rainy seasons. Significant development affecting the Big Cypress region began in the early 1920s, when two major roads were built. First was the north -south road (U.S. Highway 29) from Everglades City to Immokalee, completed in 1926. Second was the completion of the Tamiami Trail in 1928. Both were constructed of borrow material from continuous pits adjacent to the roads. The borrow pits became canals. The Everglades Parkway (Alligator Alley) was completed in 1967. Numerous bridges along the parkway permit southward flow of water. Land development for housing in the 4 Element 1 ffl�l Initial Model Comparison and Estimate of Flow to Estuaries 188- square mile Golden Gate Estates area in western Collier County began in the 1960s. Drainage canals, most notably the Golden Gate Main Canal and the Cocohatchee River Canal were dug to drain the western part of the estates. The Fahka Union Canal was completed to drain the southern portion in November 1969. The Golden Gate Canal system is described as extending inland from the Gulf about 20 miles. The bottom of the canal is excavated to about five (5) feet below sea level near the coast and to 6 -8 feet above sea level in the interior. The shallow depth of the canal and the distribution of weirs in the canal network limit the drainage of water from the shallow aquifer in inland areas. Prior to construction of the drainage network, the area inland from Naples was inundated each year during the rainy season. In 1968, construction was started on the Fahka Union Canal. Klein reports that when completed, this canal will extend northward nearly to Lake Trafford. Weirs will be distributed throughout this canal system to limit the drainage of water from the shallow aquifer and to maintain water levels in conformance with the general slope of the land surface. Canals will connect the Fahka Union system with the Golden Gate system. This canal system was subsequently completed in the early 1970's. Klein reported that of the various canals in Collier County, the Golden Gate Canal has been most frequently monitored and studied. Surface water has flowed continuously over the Golden Gate Canal outlet weir since its completion in August 1963. The northern most weirs in the system were completed between mid -1969 and mid -1970. Flow over the primary weir of the Golden Gate Canal (measured from 1965 through 1968) ranged from a high of 2,390 cubic feet per second (cfs) on July 1, 1966 to a low of 28 cfs on May 27, 1967. The average flow over the weir during the period was 350 cfs. Figure 2 -3 shows hydrographs of discharge for the 1966 and 1968 water years. 5 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Figure 2 -3 Hydrographs of Discharge for the Golden Gate Canal for the 1966 and 1968 Water Years (From Figure 18 of Klein, 1970) 2000 'i 965 ~— —_1966- 0 1000 Z p 500 ¢ 10 -- i a 50, OCT NOV DEC 'JAN FEB MAR APR MAY JUN JUL AUG SEPT F- w W ir v m c� w cr Q x v N n MI WV Utf- JAN r re MAR APR MAY JUN JUL AUG SEPT WATER YEAR McCoy, Jack. 1972. Hydrology of Western Collier County, Florida. State of Florida, Department of Natural Resources, Division of Interior Resources, Bureau of Geology Report of Investigations No. 63. This project was a study of the Hydrology of Western Collier County and was completed at the request of the County. The driving issue was development of additional freshwater supplies to meet the demands of the rapidly growing population. McCoy states that although the water supply potential of western Collier County is large, water problems exist in that the 54 inches of annual rainfall are not evenly distributed throughout the year. In addition, salt -water intrusion threatens the Naples well field during prolonged dry periods, and contamination of existing and future ground water supplies is possible by man related activities. The study focused on the areas drained by the Golden Gate and Faka Union Canal systems and included Henderson Creek. McCoy states that prior to construction of the canal system, much of Collier County was inundated each year during the rainy season. McCoy (1972) describes the canal system as follows: ■ The Golden Gate Canal extends about 20 miles inland from the Gordon River. The bottom of the canal is 5 feet below mean sea level (msl) at its outlet to Gordon River and 6 to 8 feet above msl in the interior. The design plans for the Fahka Union Canal call for similar bottom elevations. Distributed throughout the canal system are about 30 weirs, with increase in elevation toward the interior. The elevations of the coastal weirs on the Golden Gate and Fahka Union Canals are 3 and 2 feet above msl. The elevation of the highest interior weir (near Immokalee) is 17 feet above msl (it is assumed to mean NGVD29). 6 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries ■ The function of the canals is to lower annual peak water levels to prevent flooding during the rainy season. The function of the weirs is to control the canal flow and reduce the possibilities of over drainage. During the rainy season, when water levels in the interior are high, water moves from aquifer storage into the canals and downstream over the weirs. At the beginning of the dry season, flow over the inlandmost weirs ceases but continues over the downstream weirs. Flow over the weirs ceases in succession downstream, as the dry season continues, until flow occurs only at coastal weirs on the primary canals. By limiting drainage from aquifer storage, regional water levels near the coast are not lowered excessively, and therefore, the problem of sea -water intrusion is not magnified. ■ The Golden Gate Canal is about 100 feet wide, less than 8 feet deep and has several fixed weirs throughout its reach of 26 miles; the Fahka Union Canal is similar in width and depth and about 30 miles long; the Henderson Creek and Cocohatchee River Canals are 25 feet wide, less than 5 feet deep, and 7 to 13 miles in length respectively. The Henderson Creek Canal is uncontrolled except for a constriction at Alligator Alley which acts as a surface water divide most of the time. However, at the peak of the rainy season, the Henderson Creek Canal probably receives some flow from the Golden Gate Canal. The Cocohatchee River Canal has a control a short distance upstream from the gaging station. Farmers regulated the control according to irrigation needs. The Cocohatchee River Canal drains most of the area southwest of Lake Trafford, but it also helps drain the Golden Gate area during peak wet periods. McCoy reports that during 1970, the average discharge at each of the four monitoring stations was: 250 cfs from the Golden Gate Canal, 270 cfs from the Fahka Union Canal, 25 cfs from the Henderson Creek Canal, and 15 cfs from the Cocohatchee River Canal. It was further noted that during the dry season of 1971, discharge at the Golden Gate Canal outlet reached a record low of less than 20 cfs. This was approximately twice the average daily pumpage of the Naples water system in 1970. Freeberger, H.J. 1972. Stream Flow Variation and Distribution in the Big Cypress Watershed during Wet and Dry Periods. Map Series 45. Bureau of Geology, Florida Dept. of Natural Resources, Tallahassee, FL. In 1972, the Florida Bureau of Geology and Herbert Frieberger published Map Series No. 45 to present the Streamflow Variation and Distribution in the Big Cypress Watershed during wet and dry periods. This was based on measured flows from 1969 through 1971. Figure 2 -4 shows post -canal construction flow paths as estimated by Frieberger. This figure indicates that the overland sheet flow is reduced when compared to the natural system. The majority of flow is intercepted by the canal system and carried to tide via the Cocohatchee, Golden Gate, Henderson Creek, and Fahka Union Canals. Figure 2 -5 provides a comparison of average measured flows at the end of the rainy season in 1969 versus measured flows during the dry season of 1971. 7 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Figure 2-4 Map of the Big Cypress Basin Showing Direction of Overland Flow for the Period November 18 — 20, 1969 (From Figure 1 in Frieberger 1972) r jz t Figure 2 -5 Average Measured Flow Data (From Frieberger 1972) I �_ ii ou�+a• i 4 � �£ I _ � t r3!'( ?II t F i!f h' 1, A i! it _ "•� 1 �...,i t t l l t i! Y ! ti. t x ,Y . ,. ... i I.na.n'tit r Top line = November 18-20,1969 (cfs) Bottom line = March 9, 1971 (cfs) 1%) 8 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Black, Crow, and Eidsness, Inc. 1974. Hydrologic Study of the G. A. C. Canal Network. Gainesville, FL. Project no. 449- 73 -53. In 1974, Black, Crow, and Eidsness (BCE) completed a Hydrologic Study of the G. A. C. Canal Network. This study investigated the changes in the historical watersheds of Collier County and the resulting increase in wet season inflows through the Golden Gate Canal system into Naples Bay. BCE presented a diagram of pre -canal construction basin boundaries of western Collier County. This diagram is shown in Figure 2 -6. In the pre -canal time period, surface water in the Belle Meade Basin, which includes the existing Golden Gate basin, was integrated with the Corkscrew Swamp to the north and the Fakahatchee Strand to the east. Historical outlets from the Golden Gate Watershed were the Cocohatchee River, Gordon River (Naples Bay), Rock Creek, Henderson Creek, and the Fakahatchee Strand. Figure 2 -7 shows the post -canal construction drainage basins as defined by BCE (1974). Figure 2 -6 Pre -Canal Construction Basin Boundaries in Western Collier County (From Figure 2.3 in BCE 1974) WE$YFRN COCL!LR CDUNTV + .. COCOHAIGNEF RIVER KAIERSNED 1 ¢, - ULIN•A rC YkE RYxI.H'_' + r tt �1F'�r +_tN�NkN ie.tE WIVYM � Y i P1 T 8ELLE MEADE 8ASIN- y k ��- �•dAUtU6l � •�- � Ij CASW KEAi1S SAsm YYb 9 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Figure 2 -7 Post -Canal Construction Basin Boundaries in Western Collier County (From Figure 2.2 in BCE 1974) GOLDEN DATE CANAL WAYERSNEO WE SIC kN COL of R COO, Yv F it ryi ry ...,F� 4L UYJN•r ...F�S GOLDEN DATE CANAL WAYERSNEO FAHKALRAON CANAL WATERSHED BCE presented the following conclusions concerning changes in the surface water drainage patterns which are attributed to construction of the canal network: • The Cocohatchee River Watershed has been reduced in size. This is due to construction of a system of canals which drain the Southern portion of Corkscrew Swamp. The main flow from these canals is directed to the Golden Gate Canal system. • The Gordon River Watershed has also been reduced in size from approximately 25 square miles to approximately 8 square miles. Flows from a major portion of this watershed are now directed to the Golden Gate Canal system. • Substantial portions of the Rock Creek Watershed have been incorporated into the Golden Gate Canal Watershed. • Most of the area north of Alligator Alley (State Route 84) and east of State Route 951, which was once tributary to the Henderson Creek estuary, is now part of the Golden Gate Canal system. This is the single most significant change from pre - construction conditions. • The Faka Union Canal Watershed has increased in drainage area by a small amount. • Observed mean annual runoff for the four outlets of the G.A.C. Canal Network is nearly 500,000 acre feet per year, which is equivalent to 24 inches of water. This is probably 2 to 3 times greater than the pre- construction runoff value. 10 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries + i V iAVI FG: FAHKALRAON CANAL WATERSHED BCE presented the following conclusions concerning changes in the surface water drainage patterns which are attributed to construction of the canal network: • The Cocohatchee River Watershed has been reduced in size. This is due to construction of a system of canals which drain the Southern portion of Corkscrew Swamp. The main flow from these canals is directed to the Golden Gate Canal system. • The Gordon River Watershed has also been reduced in size from approximately 25 square miles to approximately 8 square miles. Flows from a major portion of this watershed are now directed to the Golden Gate Canal system. • Substantial portions of the Rock Creek Watershed have been incorporated into the Golden Gate Canal Watershed. • Most of the area north of Alligator Alley (State Route 84) and east of State Route 951, which was once tributary to the Henderson Creek estuary, is now part of the Golden Gate Canal system. This is the single most significant change from pre - construction conditions. • The Faka Union Canal Watershed has increased in drainage area by a small amount. • Observed mean annual runoff for the four outlets of the G.A.C. Canal Network is nearly 500,000 acre feet per year, which is equivalent to 24 inches of water. This is probably 2 to 3 times greater than the pre- construction runoff value. 10 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Table 2 -1 summarizes the general data related to the major drainage basins of western Collier County as defined by BCE in 1974. Each of these basins is monitored by the United States Geologic Survey and /or the South Florida Water Management District. Table 2 -1. General Data of Major Drainage Basins of Western Collier County' Drainage Basin Drainage Area Total Length of Number of Drainage Density sq. miles Canals (miles) Weirs miles /sq. mile Cocohatchee River 18.7 8 None 0.428 Golden Gate Canal 130 102 13 0.785 Henderson Creek 7.4 4 None 0.541 Canal Faka Union Canal 234 88 12 0.376 All values are based on the watershed defined by the location of the USGS stream gages. 2 Also serves as an overflow outlet for Golden Gate Canal during periods of high flow. Effective drainage area and drainage density are actually indeterminate. From this table, it appears that in 1974, the majority of flow in western Collier County was routed through the Golden Gate and Faka Union Canals as these basins incorporate more than 90 percent of the drainage area, all of the weir structures, and more than 90 percent of the constructed canals. Flow control structures have subsequently been installed on both the Cocohatchee and Henderson Creek Canals. BCE reported that the Golden Gate Canal drains about 1/3 of the area served by the western Collier County drainage network, yet accounted for approximately 50 percent of the total runoff. This is shown in Table 2 -2, which lists estimated annual runoff volumes from 1965 to 1973. To address potential reductions in discharge to the estuary system, BCE considered several alternatives including: • Fill the existing canal network. • Enlarge the present canal system to create additional storage. • Redistribute canal flows to natural areas and enlarge the canal network to create additional storage. ■ Element 1 !� Initial Model Comparison and Estimate of Flow to Estuaries Table 2 -2. Annual Runoff at Stream Gaging Stations Water Year Annual Runoff in acre -feet Water Year (Oct. — Sept.) Cochatchee River Golden Gate Canal Henderson Creek Canal Faka Union Canal 1965 -- 164,800 -- -- 1966 -- 302,400 -- -- 1967 -- 222,200 -- -- 1968 -- 323,600 -- -- 1969 19,470 221,400 13,050 -- 1970 25,540 278,000 23,400 -- 1971 18,010 197,100 13,310 247,400 1972 22,460 239,900 16,230 177,000 1973 39,590 294,600 17,740 195,300 Mean Annual Runoff 25,014 249,333 16,746 206,600 The final alternative suggested major enlargements to the existing canal system to allow a raising of the weirs to within 2 feet of the ground surface. Wet season flows in the Golden Gate Canal System would be redistributed to former historical patterns. The estimated cost was more than $18 million dollars for the Golden Gate Canal system alone. Comparable funds would be required for the Belle Meade and Faka Union Canal systems. 12 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries McPherson, B.F., G.Y. Hendrix, Howard Klein, and H.M. Tyus. 1976. The Environment of South Florida, A Summary Report. Geologic Survey Professional Paper 1011. Department of the Interior, Resource and Land Investigations Program. This project was triggered by the planned construction of an international jetport in the Big Cypress Swamp by the Dade County Port Authority. This report summarizes the effort to develop the scientific information base required by land resource managers to make informed decisions affecting the economy and environment of south Florida. Much of the information presented in this report is for areas located in the easternmost portions of Collier County and the western portions of Dade County. However, McPherson does reiterate that the purpose of the canal system in western Collier County was to lower groundwater levels, making the land suitable for urbanization, and reduce flooding. McPherson states that "the Golden Gate Canal system and the Faka Union Canal system are cut into the highly permeable limestone of the shallow aquifer. Because of the high permeability, ground water drains rapidly to the canals and thereby lowers annual peak groundwater levels in the watershed. Where ever ponding occurs within those drainage areas during the rainy season, it is likely to be local and short lived. Thus, the pattern of slow prolonged southward sheet flow of freshwater through the west part of the Big Cypress to the Gulf estuaries was changed to one of accelerated and shortened- period runoff, primarily through the canal systems." The report also states that water levels in the watershed were lowered approximately two (2) feet or more over a span of 4 or 5 years as a result of construction of the Golden Gate Canal network. Before the area was drained, it was inundated during most of the rainy season and for 2 or 3 months afterward. The Faka Union Canal network has also lowered water levels. McPherson concluded that accelerated flow through the canal systems tends to increase the opportunity for transport of pollutants and water of poor quality to be discharged to the estuaries. It was suggested that the weir elevations in the Golden Gate and Faka Union Canal systems be raised by l to 2 feet. McPherson postulated that "the reduction in runoff would salvage for potential use a large part of the flow to the sea. The resulting rise in water levels would tend to reduce damage to the environment and the possibility of saltwater intrusion and would probably reinundate some of the sloughs that became dry as a result of drainage. The possibility of environmental changes in the Fakahatchee Strand, and in the Corkscrew Marsh northeast of Naples, would be reduced because diversion of freshwater toward canals would be reduced." CH2M Hill. February 1980. Gordon River Watershed Study. Engineering Report. South Florida Water Management District. In 1980, CH2M Hill completed a study of the existing conditions within the Gordon River Watershed. The study evaluated the flood hydrology of the basin during the 25- and 100- 13 Element 1 Initial Model Comparison and Estimate i of Flow to Estuaries year storm events, determined the water surface profiles, and evaluated the economic impact of flooding on existing and potential development. In the report, CH2M Hill stated that "Historically, the Gordon River Watershed was over 25 square miles in size, extending northeast from Naples Bay beyond the present intersection of S.R 551 and S. R. 846. With the development that has occurred in the area — specifically the construction of Airport Road (S. R. 31) and the Golden Gate Canal system, the watershed has been significantly reduced in size to about 8.5 square miles. The main conclusions of the report are: • Flooding in the watershed does not vary significantly between the 25- and 100 - year storms. • Flooding is generally limited to natural low -lying mangrove areas, golf courses, and portions of the area north of Pine Ridge Road (S.R. 896). • Except for the area north of Pine Ridge Road, flooding is limited to areas which experience either no or moderate use. Economic impacts due to flooding south of Pine Ridge Road were considered negligible. • Shallow flooding – up to one foot in depth – occurs over large portions of the area north of Pine Ridge Road. This flooding does affect buildings, equipment, and materials in the area. • Economic impacts due to flooding in the industrial park area were estimated at $4,667 per year and possibly as much as $14,000 per year at full development • Large improvements within the Gordon River Watershed consistent with the Master Plan for Water Management District No. 7 were recommended to benefit surface water management within the basin. These primarily consisted of culvert replacements. Johnson Engineering, Inc. December 1981. Golden Gate Water Management Study. Big Cypress Basin Board, South Florida Water Management District. This study (Golden Gate Water Management Study) was completed on behalf of the South Florida Water Management District. The goals were to determine the feasibility of diverting a portion of the normal outflow from the Golden Gate Canal into other areas for water conservation purposes and /or retaining increased amounts of surface water in the Golden Gate Canal system. Johnson Engineering stated that, in the early 1900's, this watershed was basically a "sheet - flow type system." It was a large flat prairie- cypress area on which water stood much of the year. Johnson Engineering quoted a Naples Bay study completed in 1979 indicating that the greatest concern for Naples Bay was not the quality of water discharged from the Golden Gate Canal, but the increase in quantitative surges during the wet season. Johnson Engineering considered several alternative approaches for this project. Including: 14 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries • Diversion of water between basins to promote storage. • Alteration of proposed land uses to promote wetland protection and groundwater recharge. • Increased retention in the canal system. • Increased operable flexibility in the canal system. • Maintain the status quo. Recommendations included increasing the operable flexibility and retention in the canal system. It was also suggested that purchasing low lying areas along the canal for retention and increasing the open space along major waterways would provide significant benefit to the environment and water quality. United States Army Corps of Engineers — Jacksonville District. February 1986. Golden Gate Estates Collier County, Florida — Draft Feasibility Report. In 1986, the United States Army Corps of Engineers completed a Draft Feasibility Study for Golden Gate Estates. The primary study objective was to assess the feasibility of modifying the existing, privately constructed water control works within the Faka Union basin of Golden Gates Estates for protection and enhancement of the basin's resources. This effort considers the restoration of the basin's wetland environmental values and other natural resources to the extent possible, while maintaining and protecting compatible human resources within the basin. In describing flows within the canal network, the USACE states: • Although weirs were placed within the canals to retard canal discharge and prevent overdrainage during periods of low flow, the canal system has more than doubled the pre -canal surface water runoff. The total mean annual surface run -off from the Golden Gate Estates Canal network is 497,693 acre -feet or 162,115 million gallons of water. Over 90 percent of the observed runoff is discharged through the Golden Gate Canal (50 percent) and the Faka Union Canal (42 percent). • Under natural conditions, there was a lag of several months between peak rainfall and peak runoff and the magnitude of season variation in runoff was dampened by storage in the basin. The pattern of canal discharge more closely approximates the rainfall pattern by responding quickly to rainfall events. An ecological assessment of Faka Union Bay concluded that canal discharge affects abundance of some estuarine organisms by affecting salinity distributions. After detailed review of six proposed management strategies for the Faka Union basin, including the proposal by BCE and proposals suggested by the Golden Gate Estates Study Committee, the USACE concluded that, "after review of current Federal policies and guidelines, there is no basis for Federal implementation of modifications to the Faka Union Basin portion of the existing Golden Gate Estates water control system." However, it was 15 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries suggested that the conceptual information in the report could be used by State and local interests to determine long -term solutions to local water management problems within the basin. South Florida Water Management District. January 2007. Naples Bay Surface Water Improvement and Management (SWIM) Plan. This plan, prepared by the South Florida Water Management District focuses on strategies to improve the health and habitat of Naples Bay. Key strategies consider initiatives on water quality, stormwater quantity, watershed master planning, and implementation, and habitat assessment, restoration and improvement. With regards to the flow and timing of discharges from the Golden Gate Estates Canal system, this report states that, "the results of 60 years of canal drainage and urban development activities have reduced water clarity, increased concentrations of contaminants and nutrients, increases in freshwater and reduced dissolved oxygen levels in the NBW. The Watershed now collects surface water input from approximately 120 square miles, over a ten -fold increase from the historic drainage condition. Extensive areas of mangroves and salt marsh have been replaced by canals, seawalls and bulkheads. Development activities in the watershed have altered the volume, quality, timing and mixing characteristics of freshwater flows reaching Naples Bay. Natural tributaries, Gordon River, Rock Creek, and Haldeman Creek, have been altered by urban infrastructure which has significantly changed the historic flowways to Naples Bay and impacted its biology. Seasonal influxes of freshwater from the Golden Gate Canal system have altered the natural salinity regime of the Bay, resulting in declines in seagrass beds, and harmful impacts to all levels of flora and fauna in the aquatic ecosystem." 2.2 Summary and Conclusions The literature review was unable to identify any flow monitoring data for the period prior to development of the canal system in Collier Canal. However, it has been estimated that flows from western Collier County were typically between 0 — 10 inches annually prior to construction of the canal network. It has been documented that construction of the canal network has significantly changed the flow regime into the receiving water bodies. The combined current annual flow from the primary canals in western Collier County averages approximately 36 inches. This is approximately 3.5 times the maximum annual volume of runoff estimated by Kenner (1966). The percentage of rainfall that discharges to tide has increased from approximately 17 percent (10 inches of runoff /57 inches of rainfall) prior to construction of the canal network to more than 60 percent (36 inches of runoff /57 inches of rainfall) after construction of the canal network. 16 Element 1 Initial Model Comparison and Estimate i of Flow to Estuaries For Naples Bay it was estimated that the volume of freshwater discharge has increased by 20 to 40 percent which has significantly changed the salinity balance in the estuary. Historically, the Gordon and Rock Creek watersheds were the primary sources of inflow to Naples Bay. These two basins had a combined area of approximately 50 square miles. Now, the Golden Gate Canal watershed is the primary source of inflow to Naples Bay. This basin has an area of approximately 130 — 175 square miles. 3.0 Comparison of Computer Models and Model Results As described in the introduction, the three MIKE SHE models that were developed for the Big Cypress Basin Project Implementation Report (PIR) in order to evaluate the methods and benefits of restoring the wetland system within the Southern Golden Gates Estates (SGGE) area of the Big Cypress Basin (BCB) were compared to conduct a preliminary assessment of discharge volumes from the Collier County watersheds. These models were received from the United States Army Corps of Engineers (USACE) for this analysis and represent the existing condition (year 2000 land use), the future condition (year 2050 land use), and the natural system (pre - development) condition. MIKE SHE models The following three sections provide a description of the models and document the differences between the model input files. In addition, comparative results are presented to evaluate basin discharge to the estuary systems and to review predicted water budgets and hydro - periods. 3.1 Computer Model Descriptions Three models, existing conditions, future conditions, and natural systems (pre - development), were received from the USACE. Each of these models is described below. 3.1.1 Existing Conditions Model The original existing conditions MIKE SHE model developed for the Big Cypress Basin is documented in a report titled "Big Cypress Basin Integrated Hydrologic - Hydraulic Model" (DHI, 2002). The model received from the USACE was updated in 2006 and is documented in a reported titled "Southwest Florida Feasibility Study, Hydrologic Model Development, Scope of Work Modification IDC DACW 17 -01 -D -0013, Big Cypress Basin, Final Report (CDM, 2006). This model is referred to as the Existing Conditions Model (ECM). The ECM model is based on year 2000 land use conditions and was updated to the 1988 (NAVD) vertical datum from the 1929 (NGVD) vertical datum in 2006. In addition, the rules that determine structure operations were changed during the model update to reflect the operational guidance specified by the South Florida Water Management District (SFWMD). Figure 3 -1 shows the model domain and canal network used in the ECM simulation. 17 Element 1 Initial Model Comparison and Estimate % of Flow to Estuaries This model was run using meteorological data for 1976 — 1986 in order evaluate the system under a range of hydrologic conditions. The USACE determined that this period of time included wet, dry, and average year conditions in the study area. Figure 3 -1 Modeled Canal Network for the BCB Existing Conditions Model rI 1WM { AVIE.I �c..W W, Legend River Netviork 3.1.2 Future Conditions Model The BCB Future Conditions Model (FCM) is based on a projected year 2050 land use map that was generated by the SFWMD. This model includes the canal network defined in the ECM as well as the pumps and spreader canals recommended for the proposed Picayune 18 Element 1 Initial Model Comparison and Estimate %� of Flow to Estuaries Strand Wetland Restoration Project (PSRP). Figure 3 -2 shows the canal network for the BCB Future Conditions model. This project was formally known as the SGGE. The PSRP considers the installation of canal blocks in the Miller, Faka Union, Merritt and Prairie Canals south of I -75. In addition, it calls for construction of spreader swales and large pump stations to prevent flooding in the Northern Golden Gate Estates north of I -75. Figure 3 -2 Modeled Canal Network for the BCB Future Conditions Model PP II• Ll r 'to awl' �� F� Or Legend Major Roads . -, 19 Element 1 F%) Initial Model Comparison and Estimate of Flow to Estuaries 3.1.3 Natural Systems Model The Natural Systems, or pre - development, model was developed for the entire Southwest Florida Feasibility Study ( SWFFS) area. A full description of the model can be found in the report titled "Final Report, Natural Systems Model (NSM) Scenario Southwest Florida Feasibility Study" (SDI, 2007). The model domain includes the BCB as well as the Caloosahatchee and Estero River Basins, The SWFFS and BCB model domains are shown in Figure 3 -3. Figure 3 -3 also shows the modeled natural systems river network. In the BCB model area, only the Imperial and Cocohatchee Rivers are explicitly modeled. In order to accurately compare the three models, the NSM model was rerun as part of this project using the same BCB model domain as was defined for the ECM and the FCM. Groundwater results were extracted from the larger SWFFS NSM model and used to define a time varying boundary condition for the northern edge of the BCB NSM model. Figure 3 -3 Model Domains and Canal Network for the Natural Systems Model De Soto Rghlands Colh er Legend River Netrwrk BCB Model Domain miles S %FFS Model Domain OCounty Boundanes 20 Element 1 Initial Model Comparison and Estimate % of Flow to Estuaries 3.2 Comparison of Key Model Input Parameters In this section, results of comparisons among several of the input parameters for the three MIKE SHE models are presented. The discussion focuses on model inputs related to overland flow and discharge to the estuarine system because the saturated components of the three models are equivalent. 3.2.1 River Network As discussed above, the NSM includes only the Imperial and Cocohatchee Rivers within the Big Cypress Basin model domain. This representation assumes that no structures exist within the river network and is assumed to be representative of the pre- development time period. The model does not include the Gordon River or Henderson Creek, although both were present in the pre - development time. The ECM and FCM river /canal networks include many of the canals and structures that have been constructed in Collier County since the 1960s. The models are set up using a 1500 -foot grid cell size where the river /canal network consists of the primary drainage canals and structures maintained by the SFWMD and do not explicitly represent the secondary canals maintained by Collier County or within private developments. 3.2.2 Topography The ECM and FCM models utilize the same topographic input data file that is based on a 750 -foot grid. The topographic data input file was prepared by the SFWMD and includes a mixture of data sources, including LiDAR and topographic survey maps. To define topographic characteristics within each of the 1500 -foot grid cells, the model calculates the average of four (4) 750 -foot grid cells from the original data set to determine the value used in a single grid cell. The 1500 -foot grid cell topographic data file was used in the comparative analysis. The NSM report did not clearly define the sources of information used to define the topographic input file used in the NSM. Therefore, it is difficult to determine the level of reliability for the data. The data set was provided to the modeling team by the SFWMD and is based also based on a 1500 -foot grid. Therefore, it was possible to directly compare to the ECM and FCM topographic data files. Plate 1 shows the topographic elevation for the ECM and NSM models. In addition, a map showing the difference for each cell between the ECM and NSM topographic maps is included in Plate 1. Positive values indicate that the ESM topographic elevation is higher than the NSM topographic elevation. Negative values, on the other hand, indicate that the NSM topographic elevation is higher. As shown in Plate 1, there is a significant difference in ground surface elevation between the models. In the Okaloacoochee Slough and Faka Union Canal area south of I -75, the 21 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries ECM and FCM topographic elevation is as much as three (3) feet higher than the NSM. In the Faka Union Canal area south of I -75, this may be reasonable and could be attributed to road building and other development activities. The Okaloacoochee Slough is a natural area that has little development; therefore, it seems that the difference in elevation would be much less three (3) feet. The elevation difference is likely due to the quality of data available when the models were developed. The new LiDAR data that will be used to define the Collier County ECM should be an improvement over the current topographical data set. However, caution is advised when comparing results against the NSM model. 3.2.3 Detention Storage In the MIKE SHE model, detention storage is used to define the volume of water (inches or millimeters) that will be stored in a grid cell before overland flow occurs. The values are typically related directly to land use characteristics. In natural areas, this value is indicative of the volume of storage available in local depressions or micro - topography. In urban areas, this value represents the volume of water stored in ponds or other storm water management features that are not explicitly modeled. Plate 2 shows the detention storage values used in each of the models. The FCM and NSM models used similar detention storage values for the same land uses throughout the model domain. However, the ECM used significantly higher values. The ECM will detain anywhere from 0.8 to 3.8 inches more water in each cell before overland flow will occur. These differences significantly determine model results which may impact the validity of model comparisons. Potential effects would include (but not be limited to) changes in evapotranspiration, infiltration, overland flow and annual hydroperiod. 3.2.4 Overland Manning Values In MIKE SHE, Manning n values are assigned to each grid cell and are typically associated with land use. These values influence the rate of overland flow from cell to cell. It is expected that natural areas will offer more resistance to overland flow; while urbanized areas would offer less resistance to overland flow. Plate 3 shows the Manning n values used in each grid of the models. These maps show inconsistency in the application of Manning values between the models, although the models all utilize the same land use categories. The range of values varies from 0 -1 for the ECM, from 0.5 -100 for the FCM, and from 0.04 -0.59 for the NSM. The Natural Systems Model documentation report (SDI, 2007) provides a table that documents the relationship between the land use classification and the assigned Manning value. The initial Big Cypress Basin Integrated Hydrologic - Hydraulic Model report (DHI, 2002) reported that a uniform value of n = 0.5 was specified for all land uses in the ECM. 22 Element 1 Initial Model Comparison and Estimate j of Flow to Estuaries However, the 2006 modeling report (CDM) does not provide any information describing the basis of the revised Manning values used in the final ECM and FCM models. 3.2.5 Soils For each soil type in the model, retention and conductivity curves are defined based upon soil moisture. In the unsaturated zone soils database, there is a slight difference in the definition of the Plantation soil type between the models. This soil type is observed primarily in the wetland areas of the model. The soils database used for the ECM and FCM models extends the conductivity and retention curves for the Plantation soil. The curves defined for the NSM are not defined to the same extent as for the ECM and FCM. Therefore, the NSM generates a warning for most time steps indicating that calculated soil moisture values are outside the range of values provided for the conductivity curve. For each of those time steps, the conductivity value was subsequently set to zero (0). These warning are not generated for the ECM and FCM models. It is likely that the NSM underestimated infiltration; however, it is not clear what the full effect of this warning had on the overall model results. 3.3 Comparison of Model Results Table 3 -1 provides a summary of average annual rainfall data across the entire model domain. The model input file uses a distributed rainfall pattern, meaning that different rainfall time series are associated with each grid cell in the model. The volume of rainfall applied to each grid cell varies widely across the model domain. Table 3 -1. Average Annual Rainfall Comparison Year Average Model Rainfall Basin -wide (inches) 1976 58.58 1977 55.23 1978 53.62 1979 58.18 1980 53.26 1981 44.29 1982 69.01 1983 76.18 1984 51.53 1985 50.74 1986 52.68 23 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries For comparison purposes, and based on the basin -wide average annual rainfall values, comparative model results were generated for the years 1981 (dry year), 1983 (wet year) and 1986 (average year). Model results are presented in the following sections. Because of the inconsistency among the models, these results and conclusions should be considered preliminary. 3.3.1 Basin Discharges Table 3 -2 provides a summary of discharge to the estuaries from the contributing basin during the simulation period. The values for the NSM model are taken from the total water budget for each basin, and represent the total overland flow out of each basin. The results for the ECM and FCM models were extracted from the results of the canal system portion of the model. These results represent the discharge from the canal system directly into the receiving estuary. When reviewing results, it should be kept in mind that the Cocohatchee Basin discharges to the Cocohatchee Estuary, the Golden Gate Basin discharges to Naples Bay, the Henderson Creek Basin discharges primarily to Rookery Bay, and the Faka Union Basin discharges to the Ten Thousand Islands Estuary. It should also be noted that interbasin flow transfers occur during wet dry periods, which does not allow for a direct correlation between basin and estuary discharge; however, the overall conclusions are still valid. Review of the results indicates that they are consistent with the historical discharges discussed in the Literature Review of this report. Discharge from the NSM model is generally consistent with the average annual discharge value of 10 inches estimated by Kenner (1966). The flow to Naples Bay from the Golden Gate Basin has increased significantly since construction the canal network. On average, the increase is about four (4) times the volume predicted by the NSM, although there were years where the increased flow predicted by the ECM and the FCM for the Golden Gate Basin was more than 10 times the volume predicted by the NSM. This is also generally true for flow to the 10,000 Islands Estuary from the Faka Union Basin. Flow to Rookery Bay from Henderson Creek Basin in the ECM and the FCM is approximately double that predicted by the NSM. The model results also indicate little difference in average annual discharge from the Cocohatchee Basin. This may be due to the fact that comparatively little development has occurred in Corkscrew Swamp that forms the headwaters of this basin. In addition, structural operations in the Cocohatchee Canal are able to route water south into the larger Golden Gate Canal system. 24 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Table 3 -2. Annual Total Discharge per Basin Year Rainfall In Cocohatchee NSM Existing Future In In In Golden Gate NSM Existing Future In In In Henderson Creek NSM I Existing Future In In In NSM In Faka Union Existing Future In In 1977 5523 0.79 4.43 6.62 2.20 44.22 48.29 1.81 35.04 27.76 4.57 38.38 25.29 1978 53.62 0.75 3.26 5.79 2.09 37.95 48.15 4.29 31.10 27.40 5.24 31.04 26.19 1979 58.18 1.69 6.00 8.83 1.93 43.08 51.51 1.73 23.19 20.37 3.98 35.17 31.17 1980 5326 425 6.43 10.27 4.69 1 51.15 57.28 3.07 27.02 23.81 5.47 40.86 35.63 1981 4429 2.91 4.17 6.02 4.06 35.86 45.26 2.09 20.19 18.26 2.83 26.74 21.00 1982 69.01 6.54 8.68 11.26 12.80 55.98 64.44 14.45 39.48 36.18 13.27 59.70 57.74 1983 76.18 19.61 10.82 15.19 36.02 72.54 77.09 54.17 45.54 40.60 37.44 72.57 69.73 1984 51.53 16.06 6.76 10.81 27.01 53.25 59.66 37.44 21.34 18.05 23.74 35.75 31.02 1985 50.74 6.57 4.88 7.71 13.82 42.95 50.79 20.16 28.40 25.22 8.98 30.28 25.70 1986 52.67 3.86 2.81 5.60 9.37 38.90 48.46 12.56 19.71 17.40 6.61 30.96 26.62 Aver a 56.47 1 6.30 5.82 8.81 11.40 47.59 55.09 15.18 1 29.10 25.51 11.21 40.14 35.01 25 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries 3.3.2 Hydroperiods Calculated annual hydroperiod maps for the three modeled conditions are presented in Plates 4 — 6. Hydroperiods were calculated by determining the number of days per year that the depth of water was greater than 0.1 inches above the ground surface. The hydroperiod results appear to be reasonable over most of the BCB model domain. In general, the hydroperiod predicted for the NSM is much longer than that predicted for the ECM and FCM. The maps also demonstrate the effect on the PSRP on the wetland areas south of I -75 between the ECM and the FCM. However, in the Okaloacoochee Slough (northeast portion of the model domain) there appears to be a discrepancy. This is an area that has been kept in its natural state and one would expect that the hydroperiod would be very similar between all of the models. The model results indicate that the hydroperiod predicted by the ECM and FCM is longer than in the NSM in 1981 and 1986. This is unexpected given that the topographic elevation in this area is lower in the NSM than in the ECM or FCM. The discrepancy may be a function of the boundary conditions used in the NSM or the effect of differences in model input parameters. This discrepancy will have to be evaluated if the NSM is to be used as a baseline for evaluating future projects. 3.3.3 Average Water Depth Above the Ground Surface Average depth of water calculations were completed for the wet and dry seasons for each year of the simulation that was analyzed herein. The analysis was made consistent with the USACE definition of the wet season as being from May 1 — October 15 of each year. Therefore, the dry season is from October 16 — April 30. These time periods were used for the average season calculations. The results of the average depth of water calculations for 1981, 1983, and 1986 wet and dry seasons are presented on Plates 7 — 12. Results are consistent with the hydroperiod results described above. 3.3.4 Groundwater Levels Plates 13 — 18 present comparisons of annual groundwater elevations in the surficial aquifer for 1981, 1983, and 1986. Each plate includes three (3) maps. The first map shows the average NSM groundwater elevation in the surficial aquifer. The second map shows the average groundwater elevation in the surficial aquifer associated with either the ECM or the FCM. The third map on each plate presents the difference between the average elevations in the other two maps. A positive value means that the NSM groundwater elevation is higher 26 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries than the ECM or FCM groundwater elevation. A negative value means that the ECM or FCM groundwater elevation is higher. As with the Hydroperiod and Average Depth of Water results, the predicted groundwater elevations for each model appear to be reasonable over most of the model domain. The ECM and FCM model results show a depression in the water surface elevation of the surficial aquifer consistent with the location of the Collier County well field. The difference represents the extent of the surficial aquifer drawdown relative to the NSM model. The groundwater results also show a significant difference in head elevation in the Okaloacoochee Slough area. The difference maps indicate that the average head elevation in Okaloacoochee Slough is as much as five (5) feet higher in the ECM and FCM than in the NSM. This result is consistent with the observed hydroperiod results and may be due to the groundwater boundary conditions defined in the NSM. This issue will have to be investigated if the NSM is to be used as a baseline for evaluating future projects. 3.3.5 Water Budgets The MIKE SHE model provides many options for producing water budgets. Total water budgets can be produced in tabular or graphical format. In addition, detailed water budgets may be produced for each component (overland, groundwater, unsaturated zone, etc.) of the MIKE SHE model. Figure 3 -4 shows the Total Water Budget graphical output produced for 1986 year meteorological conditions in the Future Conditions Model. 27 Element 1 F%J Initial Model Comparison and Estimate of Flow to Estuaries Figure 3 -4 Graphical Water Balance Output for 1986 of the FCM on Accumulated waterbalance from 1MM986 to 12f31i1986. Data type : Storage depth [millimeter 1. Flow Result file: C:% BigC ypressBasin \BC[0BCB_2050_V2009.SHE - Result FilesWB_2060_V2009 Title : BCB - 2050 Future Without Proiect Conditions 1976 to 1986 Text: 2050 FWO Proiect Conditions 1976 to 1986 0 -EEE- Boundary flow 131 0 Brain SMoundary 25 61 Boundary flow '--- 156 Equation I (below) describes the components used to calculate the change in storage for the overland (OL) and unsaturated (UZ) components of the model. Equations 2 and 3 show the components used to calculate the change in storage for the saturated zone (SZ) and the water volume contributed to the MIKE I I model, respectively. OL +UZ Change in Storage = Prec - ET + Irr - OL to Riv + O /UZ In - O /UZ Out — GW Infil + (1) GW Evap SZ Change in Storage = D /SZ in —D /SZ Out +SZ In - SZ Out + GW Infil — GW Evap - Pump - (2) BF to Riv + BF from Riv - Dr to Riv - Dr to Ex Contribution to MIKE 11 = OL to Riv + BF to Riv - BF from Riv + Dr to River (3) 28 Element 1 • Initial Model Comparison and Estimate of Flow to Estuaries Where: Prec = Precipitation ET = Evapotranspiration Irr = Irrigation OL to Riv = Overland Flow to River O/UZ in = Overland /UZ Boundary In O /UZ Out = Overland /UZ Boundary Out D /SZ In = Drain SZ/Boundary In D /SZ Out = Drain SZ/Boundary Out SZ In = SZ Boundary In SZ Out = SZ Boundary Out GW Infil = Infiltration to GW GW Evap = Infiltration from GW Pump = Pumping BF to Riv = Baseflow to River BF from Riv = Baseflow from River Dr to Riv = Drain to River Dr to Ex = Drain to External River Water budgets for the MIKE SHE model were extracted from the results for the entire BCB model domain and for the Golden Gate, Cocohatchee, Henderson Creek and FakaUnion Canal subcatchments (basins). Subcatchment locations are shown in Figure 3 -5. The water budget comparisons for the entire BCB model domain and the four subcatchments are shown in Tables 3 -3 through 3 -7. It is noted that the subcatchment water budgets only consider the hydrologic processes that occur within the subcatchment. They do not consider inflows from outside the subcatchment within the canal /river network. The total contribution from the entire BCB model domain to the estuary system via overland flow can be calculated using the following equation: BCB Flow to Estuaries = OL to Riv + O /UZ Out + Dr to Riv + BF to Riv — BF from Riv Table 3.8 provides a summary of calculated total flow to the estuaries for each of the BCB models during the three rainfall years evaluated during this analysis. Results indicate that the discharge ratio compared to the NSM is largest during average years. During dry years, runoff and baseflow are limited which reduces discharges to the receiving water bodies. In addition, structures during those periods are operated such that flows are retained in the drainage system. During wet years, discharges from both natural and developed areas are large due to high groundwater elevation and soil saturation. It should be noted that values in Table 3 -8 represent the total flow from the entire model domain and may differ significantly from discharge rates from individual sub - basins, as shown in Table 3 -2. 29 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Figure 3 -5 Defined Subcatchments (basins) in the Big Cypress Model Domain N i W�E t S I Cocohat ee -I ial Inter i Cocohatchee Main Canal ial Ri—B sin Barron iver Golden Gate Main Canal 1 Faka Union Canal Fakahatchee Henderson Creek ed Coastal asins s Coastal Basi s Legend BCB Model Domain 2 4 s 12 County Boundaries `c; ies . - El Subcatchments Table 3 -8. Total Runoff from the BCB MIKE SHE Models Year MIKE SHE Model NSM ECM FCM 1981 4.76 14.17 13.70 1983 19.92 35.39 34.96 1986 5.98 23.98 15.75 AVG 10.22 24.51 21.47 30 Element 1 Pffill Initial Model Comparison and Estimate of Flow to Estuaries Table 3 -3. Total Water Budget Comparison for BCB Model Domain BCB Model Domain 1981 1983 1986 Water Budget NSM ECM FCM NSM ECM FCM NSM ECM FCM Component (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) Precipitation 44.29 44.29 44.29 76.18 76.18 76.18 52.68 52.68 52.68 Evapotranspiration 50.35 36.77 36.69 51.18 38.78 38.98 50.39 37.76 37.76 Irrigation 0.00 3.07 2.99 0.00 0.94 0.94 0.00 1.85 1.81 Overland (OL) Flow to River 0.08 -4.17 -3.15 0.94 -1.10 0.08 0.08 3.94 -2.83 OL/UZ Boundary Flow In 0.55 0.00 0.00 0.16 0.00 0.00 0.43 0.00 0.00 OL/UZ Boundary Flow Out 4.41 3.86 4.84 18.58 5.47 9.80 5.67 3.90 5.16 Overland Storage Change -4.69 -0.75 -0.75 6.34 1.38 1.10 -0.28 0.28 0.47 Unsaturated Zone (UZ) Storage -0.39 -0.35 -0.31 0.08 0.47 0.31 0.00 0.35 0.24 Change Infiltration to GW 5.35 19.88 18.54 7.95 41.50 37.17 6.97 24.02 22.40 Evaporation from GW 10.24 7.76 8.54 8.70 9.33 10.31 9.69 7.80 8.66 GW Pumping 0.00 3.90 3.86 0.00 1.85 1.81 0.00 2.76 2.68 Drain to River 0.00 11.93 9.88 0.00 27.83 22.68 0.00 13.35 11.10 Basef low to River 0.28 3.27 2.95 0.39 4.33 3.86 0.24 3.54 3.15 Baseflow from River 0.00 0.71 0.83 0.00 1.14 1.46 0.00 0.75 0.83 GW Boundary Flow In 3.46 6.46 6.38 3.54 5.98 5.98 3.54 6.22 6.14 GW Boundary Flow Out 0.75 2.01 2.05 0.94 3.07 3.27 0.79 2.40 2.40 Drain SZ /Boundary Flow Out 0.00 0.59 0.79 0.00 1.26 1.65 0.00 0.71 0.98 Saturated Zone (SZ) Storage -2.40 -2.44 -2.28 1.46 0.98 0.94 -0.20 0.47 0.35 Change Total Error 0.00 0.08 1 0.08 1 0.00 0.04 0.04 1 0.00 0.04 1 0.04 31 Element 1 PM1 Initial Model Comparison and Estimate of Flow to Estuaries Table 3-4. Total Water Budget Comparison for Golden Gate Basin Golden Gate Basin 1981 1983 1986 Water Budget NSM ECM FCM NSM ECM FCM NSM ECM FCM Component (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) Precipitation 47.52 47.52 47.52 79.33 79.33 79.33 54.57 54.57 54.57 Evapotranspiration 54.29 27.99 24.88 53.11 32.60 30.87 51.57 30.67 28.31 Irrigation 0.00 1.38 1.18 0.00 0.55 0.47 0.00 0.79 0.67 Overland (OL) Flow to River 0.00 -1.30 7.13 0.00 -1.34 13.70 0.00 -0.91 7.36 OL/UZ Boundary Flow In 3.86 0.28 0.71 20.35 1.02 2.32 5.00 0.16 0.43 OL/UZ Boundary Flow Out 4.06 0.04 0.20 36.02 0.08 0.67 9.37 0.00 0.08 Overland Storage Change -5.98 0.00 0.00 7.80 0.28 0.20 -1.89 0.04 0.00 Unsaturated Zone (UZ) Storage -0.51 -1.22 -1.06 0.08 1.26 1.22 0.04 1.22 0.91 Change Infiltration to GW 6.93 24.06 18.46 6.61 48.66 35.71 6.02 24.80 19.09 Evaporation from GW 7.40 0.28 0.12 3.90 0.59 0.12 5.55 0.28 0.08 GW Pumping 0.00 5.00 4.84 0.00 4.37 4.33 0.00 4.57 4.45 Drain to River 0.00 19.02 9.88 0.00 38.03 20.08 0.00 17.01 8.70 Baseflow to River 0.00 6.89 9.37 0.00 10.79 15.20 0.00 8.19 10.31 Basef low from River 0.00 3.46 2.09 0.00 2.72 1.65 0.00 2.95 1.73 GW Boundary Flow In 1.42 3.86 3.90 1.73 4.72 4.61 1.30 4.06 4.02 GW Boundary Flow Out 1.77 1.46 1.26 2.44 1.34 1.34 2.13 1.34 1.18 Drain SZ /Boundary Flow Out 0.00 0.04 0.00 0.00 0.12 0.08 0.00 0.04 0.04 Saturated Zone (SZ) Storage -0.83 -1.34 -1.06 2.01 0.87 0.91 -0.31 0.39 0.08 Change Total Error 0.00 0.00 0.04 0.00 0.00 0.04 1 0.00 1 0.00 1 0.00 32 Bement e M? Initial Model Comparison and Estimate of Flow to Estuaries Table 3 -5. Total Water Budget Comparison for Cocohatchee Basin Cocohatchee Basin 1981 1983 1986 Water Budget NSM ECM FCM NSM ECM FCM NSM ECM FCM Component (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) Precipitation 45.16 45.16 45.16 74.76 74.76 74.76 50.59 50.59 50.59 Evapotranspiration 50.39 39.45 38.46 50.35 39.37 38.46 48.94 38.35 37.32 Irrigation 0.00 6.18 5.94 0.00 2.20 2.09 0.00 3.98 3.78 Overland (OL) Flow to River 0.28 -32.48 -29.76 2.48 -34.61 -30.94 0.43 -33.54 -30.00 OL/UZ Boundary Flow In 0.55 0.47 0.16 5.51 0.87 0.43 0.63 0.59 0.20 OL/UZ Boundary Flow Out 2.60 0.20 0.51 17.09 0.75 1.61 3.39 0.20 0.43 Overland Storage Change -5.59 -0.28 -0.28 6.02 0.63 0.63 -2.68 0.08 0.08 Unsaturated Zone (UZ) Storage -1.06 -0.47 -0.47 0.47 0.55 0.51 0.16 0.39 0.28 Change Infiltration to GW 5.67 47.24 44.84 13.07 73.46 69.41 8.62 51.46 48.46 Evaporation from GW 6.54 1.77 1.93 9.25 2.24 2.32 7.64 1.69 1.93 GW Pumping 0.00 6.22 5.98 0.00 2.36 2.28 0.00 4.09 3.90 Drain to River 0.00 40.28 37.68 0.00 64.49 61.26 0.00 43.35 40.51 Basef low to River 0.04 1.10 1.26 0.04 0.98 1.10 0.04 1.06 1.18 Baseflow from River 0.00 0.67 0.83 0.00 1.22 1.26 0.00 0.63 0.91 GW Boundary Flow In 1.06 4.02 3.82 1.65 3.35 3.46 1.22 3.39 3.19 GW Boundary Flow Out 1.97 4.65 4.41 3.66 6.73 6.02 2.72 5.59 5.28 Drain SZ /Boundary Flow Out 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Saturated Zone (SZ) Storage -1.81 -2.09 -1.81 1.77 1.14 1.14 -0.51 -0.28 -0.28 Change Total Error 0.00 0.12 0.12 0.04 0.08 0.04 1 0.00 1 0.12 1 0.04 33 tiement i Initial Model Comparison and Estimate of Flow to Estuaries Table 3 -6. Total Water Budget Comparison for Henderson Creek Basin Henderson Creek Basin 1981 1983 1986 Water Budget NSM ECM FCM NSM ECM FCM NSM ECM FCM Component (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) Precipitation 46.34 46.34 46.34 77.20 77.20 77.20 55.87 55.87 55.87 Evapotranspiration 55.43 32.20 31.89 53.39 34.49 34.33 53.19 33.43 33.23 Irrigation 0.00 0.16 0.12 0.00 0.08 0.94 0.00 0.08 0.04 Overland (OL) Flow to River 0.00 -3.11 -4.02 0.00 -2.87 0.83 0.00 -2.64 -1.22 OL/UZ Boundary Flow In 6.18 0.12 0.24 43.03 0.63 1.30 13.11 0.08 0.28 OL/UZ Boundary Flow Out 2.09 1.10 1.85 54.17 2.56 4.76 12.56 1.34 2.48 Overland Storage Change -6.46 -0.04 -0.79 7.28 0.51 0.94 -0.87 0.35 0.31 Unsaturated Zone (UZ) Storage -0.08 -1.30 -0.87 0.00 0.63 0.35 0.00 0.55 0.28 Change Infiltration to GW 10.51 20.00 22.20 8.03 46.10 42.52 8.58 24.96 24.61 Evaporation from GW 8.98 2.28 3.54 2.68 3.46 5.12 4.53 1.89 3.46 GW Pumping 0.00 2.44 2.48 0.00 2.32 2.32 0.00 2.32 2.36 Drain to River 0.00 12.48 11.97 0.00 31.81 25.51 0.00 13.86 11.10 Baseflow to River 0.00 0.51 1.30 0.00 0.63 1.69 0.00 0.63 1.42 Baseflow from River 0.00 1.93 1.65 0.00 1.57 1.57 0.00 1.69 1.42 GW Boundary Flow In 3.07 3.23 3.31 3.19 3.43 3.78 2.99 3.19 3.35 GW Boundary Flow Out 6.46 9.57 9.76 7.56 11.57 11.89 7.36 10.20 10.35 Drain SZ /Boundary Flow Out 0.00 0.04 0.16 0.00 0.24 0.67 0.00 0.08 0.28 Saturated Zone (SZ) Storage -1.89 -2.17 -2.09 1.02 1.06 0.71 -0.31 0.91 0.43 Change Total Error -0.04 1 0.00 1 0.08 1 0.00 1 0.04 1 0.08 1 0.00 1 0.04 1 0.04 34 Element 1 ! Initial Model Comparison and Estimate of Flow to Estuaries Table 3 -7. Total Water Budget Comparison for Faka Union Canal Basin Faka Union Canal Basin 1981 1983 1986 Water Budget NSM ECM FCM NSM ECM FCM NSM ECM FCM Component (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) (inches) Precipitation 45.91 45.91 45.91 78.78 78.78 78.78 55.12 55.12 55.12 Evapotranspiration 53.58 32.28 35.67 52.09 35.04 37.56 52.32 33.82 36.54 Irrigation 0.00 1.77 1.61 0.00 0.51 0.47 0.00 1.02 0.91 Overland (OL) Flow to River 0.00 -2.44 -1.42 0.00 -2.17 14.88 0.00 -1.54 0.47 OL/UZ Boundary Flow In 1.50 1.22 1.50 20.91 9.41 11.22 3.90 2.52 2.68 OL/UZ Boundary Flow Out 2.83 -0.12 0.47 37.44 -0.28 2.80 6.61 -0.28 0.55 Overland Storage Change -7.48 -0.12 -0.79 8.94 0.55 1.06 0.16 0.20 0.71 Unsaturated Zone (UZ) Storage -0.28 -0.98 -0.59 0.08 0.91 0.43 0.00 0.87 0.31 Change Infiltration to GW 5.87 21.42 21.61 2.99 56.46 43.03 4.17 26.77 26.61 Evaporation from GW 7.17 1.18 5.94 1.85 1.81 9.29 4.29 1.18 6.46 GW Pumping 0.00 3.46 3.31 0.00 2.24 2.17 0.00 2.72 2.60 Drain to River 0.00 10.12 7.05 0.00 37.83 20.63 0.00 12.01 8.58 Baseflow to River 0.00 11.14 7.48 0.00 17.01 11.22 0.00 13.27 9.17 Baseflow from River 0.00 0.43 1.34 0.00 0.20 0.87 0.00 0.28 1.10 GW Boundary Flow In 0.94 3.50 1.34 0.87 4.29 1.93 0.91 3.94 1.38 GW Boundary Flow Out 0.98 1.26 1.93 1.02 1.14 1.73 0.98 1.18 1.93 Drain SZ /Boundary Flow Out 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Saturated Zone (SZ) Storage -1.38 -1.77 -1.42 0.98 0.94 0.75 -0.16 0.67 0.39 Change Total Error -0.04 0.00 0.00 0.00 0.04 0.04 1 0.00 1 0.00 1 0.00 35 tiemeni i PW Initial Model Comparison and Estimate of Flow to Estuaries 4.0 Summary and Conclusions In this report, the MIKE SHE models used for the BCB PIR were reviewed and the discharge results were evaluated relative to values reported in the literature. The model review indicated that there appears to be inconsistency in how some parameters were defined in the MIKE SHE models. The Manning's "n" values and the detention storage values show the most variation between the models. In general, the model predicted discharge results are consistent with the values identified in the literature. However, the review of the NSM model results raised some questions about input values used in the NSM model in the vicinity of Okaloacoochee Slough. The Okaloacoochee Slough area is mostly undeveloped; however, the NSM predicted groundwater levels and hydroperiod in this area are substantially different from the results predicted by the ECM and FCM models. These differences will have to be investigated if the NSM is to be used as a baseline for evaluating future projects. The model results indicate that the average annual discharge from the NSM model is generally consistent with the average annual discharge value of 10 inches estimated by Kenner (1966). The model comparison results (see Table 3 -2) also indicate that the flow to Naples Bay from the Golden Gate basin and to the Ten Thousand Islands from the Faka Union basin has increased significantly since construction of the canal network. On average, the increase in flow in these basins is approximately four (4) times the volume predicted by the NSM over the simulation period. However, there were years where the increased flow predicted by the ECM and the FCM for these basins was estimated to be more than 10 times the volume predicted by the NSM. These values are also consistent with those reported by BCE (1974) and others as described in the literature review. The model comparisons of individual rainfall years indicate that the discharge ratio compared to the NSM is largest during average years. 36 Element 1 Initial Model Comparison and Estimate �r of Flow to Estuaries 5.0 Bibliography Abbott, G.C. and A.K. Nath. 1996. Hydrologic Restoration of Southern Golden Gate Estates Conceptual Plan. South Florida Water Management District, Naples, Florida. 206 pp. plus appendices. Black, Crow, and Eidsness, Inc. 1974. Hydrologic Study of the G. A. C. Canal Network. Gainesville, FL. Project no. 449- 73 -53. CDM. January 2007. Nutrient Load Assessment, Estero Bay and Caloosahatchee River Watershed. South Florida Water Management District. CH2M Hill. February 1980. Gordon River Watershed Study: Engineering Report. South Florida Water Management District. Collier County. October 1997. Collier County Growth Management Plan — Public Facilities Element, Drainage Sub - Element Davis, John H. October 1943. The Natural Features of Southern Florida, Especially the Vegetation, and the Everglades. Florida Geological Survey Bulletin No. 25. Debra Childs Woithe, Inc. and Sherry Brandt - Williams. February 2006. Naples Bay Surface Water Improvement & Management Plan Reconnaissance Report. DeGrove, Bruce. June 1979. Gordon River /Naples Bay Intensive Survey Documentation. Florida Department of Environmental Regulation. DHI, Inc. January 2002. Big Cypress Basin Integrated Hydrologic - Hydraulic Model Final Report. South Florida Water Management District. Dickson, Kevin G., William M. Helfferich, Michael Brady, and Sharon Hynes. May 1983. The Collier County Water Resource Mapping Program — Technical Report. South Florida Water Management District. Dynamic Solutions, LLC and Camp Dresser & McKee, Inc. January 2008. Task 2 — Model Development, Verification, and Preliminary Sensitivity Analysis, TMDL Data and Model Support Activities in the Caloosahatchee River Basin. Division of Water Resource Management, Watershed Assessment Section, Florida Department of Environmental Protection. Freiberger, H.J. 1972. Stream Flow Variation and Distribution in the Big Cypress Watershed during Wet and Dry Periods. Map Series 45. Bureau of Geology, Florida Dept. of Natural Resources, Tallahassee, FL. Gentile, John H., Paul Montagna, Jeffrey M. Klopatek, and Michael Walters. July 2008. Scientific Peer Review of the Draft Technical Document to Support a Water Reservation Rule for Picayune Strand and Downstream Estuaries. South Florida Water Management District. Google.com. 2009. Public Data — Population. Data Source is United States Census Bureau, Population Division. H. W. Lochner, Inc. December 2004. Immokalee Storm Water Management Plan, Hydrologic and Hydraulic Water Quality Modeling, Collier County. South Florida Water Management District. Heald, Eric J. and Durbin C. Tabb. November 1973. Rookery Bay Land Use Studies, Environmental Planning Strategies for Development of a Mangrove Shoreline, Study No. 6 "Applicability of the Interceptor Waterway Concept to the Rookery Bay Area." The Conservation Foundation. Hydrogeologic, Inc. August 2006. Hydrologic - Hydraulic and Environmental Assessment for the Camp Keais Strand Flowway. South Florida Water Management District. Johnson Engineering, Inc, Agnoli, Barber & Brundage, Inc., and Boylan Environmental Consultants, Inc. July 1999. South Lee County Watershed Management Plan. South Florida Water Management District. Johnson Engineering, Inc. December 1981. Golden Gate Water Management Study. Big Cypress Basin Board, South Florida Water Management District. Johnson Engineering, Inc. September 1983. Hydrologic Effects of Storm of September 1 -3, 1983 in Golden Gate City. Big Cypress Basin Board, South Florida Water Management District. Kenner, W. E., 1966, `Runoff in Florida ", Map Series No. 22, U.S. Geologic Survey Klein, H., W.J. Schneider, B.F. McPherson and T.J. Buchanan. May 1970. Some Hydrologic and Biologic Aspects of the Big Cypress Swamp Drainage Area, Southern Florida. United States Geologic Survey Open -file Report 70003. 37 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries Klein, Howard. 1980. Water- Resources Investigations, Collier County, FL. United States Department of the Interior, Geological Survey and South Florida Water Management District. LaRoe, Edward T. January 1974. Rookery Bay Land Use Studies, Environmental Planning Strategies for Development of a Mangrove Shoreline, Study No. 8 "Environmental Considerations for Water Management District 6 of Collier County. Collier County Conservancy, Inc. McCoy, H. Jack. 1975. Summary of Hydrologic Conditions in Collier County, Florida, 1974. Open File Report FL- 75007. United States Department of the Interior, Geologic Survey. McCoy, Jack. 1972. Hydrology of Western Collier County, Florida. State of Florida, Department of Natural Resources, Division of Interior Resources, Bureau of Geology Report of Investigations No. 63. McPherson, B.F., G.Y. Hendrix, Howard Klein, and H.M. Tyus. 1976. The Environment of South Florida, A Summary Report. Geologic Survey Professional Paper 1011. Department of the Interior, Resource and Land Investigations Program. Metcalf & Eddy, AECOM. February 2006. Reconnaissance of Hydrology and Environmental Conditions in Central Big Cypress Basin, Final Report. South Florida Water Management District. Minerals Management Service and Fish and Wildlife Service — U.S. Department of the Interior. September 1984. An Ecological Characterization of the Caloosahatchee River /Big Cypress Watershed. Missimer and Associates, Inc. December 1982. Groundwater Resources of the Cocohatchee Watershed, Collier County, FL Phase II Geologic Model. Big Cypress Basin, South Florida Water Management District. Parsons. September 2006. Belle Meade Area Stormwater Management Master Plan. South Florida Water Management District. Rosendahl, Peter C. and David A. Sikkema. 1981. Water Management Plan: Turner River Restoration. South Florida Research Center Report M -621. Schmid, Jeffrey R., Kathy Worley, David S. Addison, Andrew R. Zimmerman, and Alexa Van Eaton. February 2006. Naples Bay Past and Present: A Chronology of Disturbance to an Estuary, Final Report. City of Naples and South Florida Water Management District. SDI Environmental Services, Inc., BPC Group Inc. and DHI, Inc. January 2008. Southwest Florida Feasibility Study Integrated Hydrologic Model, Model Documentation Report. South Florida Water Management District, Ft Myers, FL. Simpson, B., R. Aaron, J. Betz, D. Hicks. J. van der Kreeke, B. Yokel. 1979. The Naples Bay Study, The Collier County Conservancy. Naples, Florida. South Florida Water Management District Big Cypress Basin and United States Department of Agriculture and Natural Resources Conservation Service. October 2003, Southern Golden Gate Estates Watershed Planning Assistance Cooperative Study South Florida Water Management District. 2005 -2006 Update. Lower West Coast Water Supply Plan. South Florida Water Management District. 2005 -2006 Update. Lower West Coast Water Supply Plan - Appendices. South Florida Water Management District. 2005 -2006. Consolidated Water Supply Plan — Support Document. South Florida Water Management District. January 1994. Big Cypress Basin, Five Year Plan —1994 — 1998. South Florida Water Management District. January 2007, Naples Bay Surface Water Improvement and Management Plan. Southwest Florida Regional Planning Council. August 1976. Water Quality and Hydrodynamic Sampling Program Design. Taylor Engineering, Inc. June 2005. Evaluation of Naples Bay Water Quality and Hydrologic Data, South Florida Water Management District. United States Army Corps of Engineers — Jacksonville District and South Florida Water Management District. April 1999. Central and Southern Florida Project Comprehensive Review Study; Final Integrated Feasibility Report and Programmatic Environmental Impact Assessment. United States Army Corps of Engineers — Jacksonville District and Florida Department of Environmental Protection. August 2003. Comprehensive Water Quality Feasibility Study; Draft Project Management Plan. 38 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries United States Army Corps of Engineers — Jacksonville District and South Florida Water Management District. September 2004. Comprehensive Everglades Restoration Plan Picayune Strand Restoration (Formerly Southern Golden Gate Estates Ecosystem Restoration), Final Integrated Project Implementation Report and Environmental Impact Statement. United States Army Corps of Engineers — Jacksonville District and CDM. February 2007. Southwest Florida Feasibility Study Water Quality Model Development. United States Army Corps of Engineers — Jacksonville District. February 1986. Golden Gate Estates Collier County, Florida — Draft Feasibility Report. Veri, Albert R., Arthur R. Marshall, Susan Uhl Wilson, James H. Hartwell, Peter Rosendahl and Thomas Mumford. October 1973. Rookery Bay Land Use Studies, Environmental Planning Strategies for Development of a Mangrove Shoreline, Study No. 2 The Resource Buffer Plan: A Conceptual Land Use Study Water Management District No. 6, Collier County, Florida. The Conservation Foundation. Wanielista, Marty. August 2006. An Evaluation of Southwest Florida Basin Rule BMP Efficiencies. Weisberg, Robert H. and Lianyuan Zheng, College of Marine Science, University of South Florida. September 2007. Estuarine Hydrodynamic Modeling of Rookery Bay, Final Report. Florida Department of Environmental Protection. 39 Element 1 Initial Model Comparison and Estimate of Flow to Estuaries L "I ':1 Technical Memorandum To: Mac Hatcher, PM Collier County From: Moris Cabezas, PBS &J Dave Tomasko, PBS &J Emily Keenan, PBS &J Date: July 9, 2010 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 2, Task 2: Quality of Discharge 1.0 Objective This Technical Memorandum addresses Element 2, Task 2: Quality of Discharge. The objective of this task is to characterize the water quality of fresh water discharges delivered to the following four estuaries in Collier County: • Wiggins Pass • Naples Bay • Rookery Bay • Ten Thousand Islands (TTI) Six watersheds were evaluated that discharge fresh water to these four estuaries (Figure 1 -1). Wiggins Pass estuary receives runoff from the Cocohatchee- Corkscrew watershed. The Golden Gate - Naples Bay and Rookery Bay watersheds discharge into Naples Bay and Rookery Bay estuaries, respectively. Three watersheds comprise the drainage area to the Ten Thousand Islands estuary: Faka Union, Fakahatchee, and Okaloacoochee -SR29. 2.0 Quality of the Discharge Analysis Methodology To accurately characterize the water quality of the discharge waters from priority watersheds in Collier County in addition to the review of the available reports, PBS &J analyzed available water quality data for Cocohatchee- Corkscrew, Golden Gate - Naples Bay, Rookery Bay, Faka Union, Okaloacochee - SR29, and Fakahatchee watersheds. From this analysis parameters of concern that could lead to coastal eutrophication, water quality impairment and the loss of natural function were identified. Additionally, data for pH and conductivity Faka Union watershed were reviewed to determine whether the watershed, with its extensive drainage canals, shows a more significant influence of groundwater contributions compared to the other watersheds. This analysis is referred to as source water determination. 1 Collier County Watershed Model Update MS1 and Plan Development July 9, 2010 Figure 1 -1. Collier County Watersheds HEN DRY CO �. .... ..... ....__._......_ a ";t Cocohatchee- Corkscrew i d� t OkaloacocheeSR29 LEE CU i Golden Gate Naples Bay Naples Fakahatchee Fake Union �t Rookery Bay w.. Marco Island 7 � w Legend 0 Watershed Boundary r COLLIER CO County Boundary P` MONROE CO N 0 2 4 Miles A 2 Collier County Watershed Model Update PW and Plan Development July 9, 2010 2.1. Water Quality Data The data used for the analyses included the IWR Run 39 data (supplied by FDEP), as well as data from Florida STORET, Collier County, City of Naples, and the Rookery Bay National Estuarine Research Reserve ( RBNERR). This resulted in an updated and comprehensive database of water quality data. All available water quality data were subjected to a quality assurance / quality control procedure, It should be noted that the analyses were conducted using the most recent ten year time period (2000 to 2009) to minimize the effect of temporal variations. Also, it was determined that the majority of water quality data available was collected during this ten year period. To eliminate potential errors due to duplicate data entry via multiple agencies uploading the same data, median values were calculated by station, date, and parameter. To allow for a direct comparison between lab parameters (i.e., nutrients) and field parameters (i.e., temperature, dissolved oxygen) samples were restricted to those collected from less than one meter depth. Since lab parameters are typically from surface grab samples, this ensures that comparisons between various parameters are from samples taken from the same general water depth. Using GIS and the station descriptions, the locations of water quality stations were reviewed in order to identify locations where multiple stations were sampled. Data were merged when more than one water quality station was sampled at the same location and a unique merged station name was assigned to that location. Appendix A lists all water quality stations and assigned merged station names. Each parameter in the database was screened to identify outliers or entry errors due to unit inconsistencies. Identified inconsistencies were reviewed and corrected. When Total Nitrogen (TN) species were not listed, TN was calculated as the sum of Total Kjeldahl Nitrogen (TKN) and Nitrate + Nitrite (NOx). To ensure consistency with IWR guidance, corrected chlorophyll a was preferentially used over uncorrected chlorophyll a for samples collected in 2006 and earlier. After 2006, IWR guidance from FDEP is that only corrected chlorophyll a data should be used. All water quality stations retrieved from the IWR database or Florida STORET were previously assigned to a WBID by FDEP. Water quality station data provided by Collier County, City of Naples, or RBNERR were assigned to a WBID and watershed based on location coordinates as shown in Figure 2 -1. The figure also shows the stations located on the downstream end of the watershed that better represent the characteristics of the discharge into the estuaries. Table 2 -1 lists the WBIDs that drain into the four estuarine receiving waters being investigated. The parameters summarized for each station and receiving water body are listed in Table 2 -2. 3 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Figure 2 -1. Water quality monitoring station location map 4 Collier County Watershed Model Update PMj and Plan Development July 9, 2010 Table 2 -1. List of fresh water discharge WBIDs combined to create the designated Watershed and corresponding Receiving Water Table 2 -2. List of Water Quality Parameters Parameter .nit; Par�nii� ter' Unit 3259W LAKE TRAFFORD Cocohatchee- Corkscrew Wiggins Pass 3259Z LITTLE HICKORY BAY Cocohatchee- Corkscrew Wiggins Pass 3278D COCOHATCHEE (INLAND SEGMENT) Cocohatchee- Corkscrew Wiggins Pass 3278C COCOHATCHEE GOLF COURSE DISCHARGE Cocohatchee- Corkscrew Wiggins Pass 3278F CORKSCREW MARSH Cocohatchee- Corkscrew Wiggins Pass 3278E COW SLOUGH Cocohatchee -Corkscrew Wiggins Pass 3259B DRAINAGE TO CORKSCREW Cocohatchee- Corkscrew Wiggins Pass 3278L IMMOKALEE BASIN Cocohatchee- Corkscrew Wiggins Pass 3278K GORDON RIVER EXTENSION Golden Gate Naples Bay Naples Bay 3278S NORTH GOLDEN GATE Golden Gate Naples Bay Naples Bay 3278V ROOKERY BAY (INLAND EAST SEGMENT) Rookery Bay Rookery Bay 3278Y ROOKERY BAY (INLAND WEST SEGMENT) Rookery Bay Rookery Bay 3278H FAKA UNION (NORTH SEGMENT) Faka Union TTI 32781 FAKA UNION (SOUTH SEGMENT) Faka Union TTI 3278G FAKAHATCHEE STRAND Fakahatchee TTI 32591 CAMP KEAIS Fakahatchee TTI 3261C BARRON RIVER CANAL Okaloacoochee -SR29 TTI 3278T OKALOACOOCHEE SLOUGH Okaloacoochee -SR29 TTI 3278W SILVER STRAND Okaloacoochee -SR29 TTI Table 2 -2. List of Water Quality Parameters Parameter .nit; Par�nii� ter' Unit Salinity ppt Conductivity µmhos /cm Total Nitrogen mg/1 Nitrate - Nitrite mg/1 Total Phosphorus mg/1 Orthophosphate mg11 Total Kjeldahl Nitrogen mg 11 Unionized Ammonia mg/1 Chlorophyll a µg11 Fecal Coliform # /100m1 Color PCU Copper µg /l Total Suspended Solids mg/1 Turbidity NTU Dissolved Oxygen mg/1 Biochemical Oxygen Demand mg/1 Iron µg /l Hardness mg/1 Secchi Depth m 5 Collier County Watershed Model Update M1 and Plan Development July 9, 2010 2.2. Critical Parameters The water quality data for each station where analyzed to determine if the parameters of concern exceeded either the State Water Quality Standards or screening level standards. Table 2 -3 shows the regulatory standards for a Class 3F water body for parameters that have a numeric criteria. Screening level standards were used to assess water quality for parameters for which no numeric state standards exist. Screening level standards are available for total nitrogen (TN) and total phosphorus (TP) based on the 701h percentile of all available data, a technique first used by Friedman and Hand (1989). Using IWR Run 39, a similar screening level was calculated by water body type for color and total suspended solids, in which the 701h percentile of all data available from 2000 to 2009 by water body type was calculated. Table 2 -4 shows the screening level standard for selected parameters by water body type (stream or lake). Table 2 -3. List of regulatory standards for selected water quality parameters Niameter CGlass 3F " Fecal Coliform ( # /100ml) 400 Dissolved Oxygen (mg /1) 5 Iron (µg/1) 1000 Chlorophyll a (µg/1) 20 Un- ionized Ammonia (mg /1) 0.02 Copper (µg /1)* e ^(0.854[1nH]- 1.702) *Median Hardness (mg/l) value by station from 2000 -2009 was used to calculate standard Table 2 -4. List of screening levels for selected water quality parameters Pa >daaetit take s Color (PCU) 80 111.5 TSS (m /1) 13 7 TN (m 1) 1.7 1.6 TP (m /1) 0.11 0.22 From the data review, five parameters of concern were determined to be critical for the receiving water bodies. These parameters include total nitrogen, total phosphorus, total suspended solids, color, and fecal coliform bacteria. An analysis of the potential impact of the discharge of these parameters into each of the estuaries was conducted. For each of the critical parameters identified, a graphic was created which identifies potential areas of concern based on the location of the water quality station. Additionally, water quality stations were grouped based on the frequency in which the data exceeded the appropriate regulatory standard or screening level for each parameter analyzed. Stations with values in exceedance of the appropriate regulatory standard or screening level in less than 10 percent of the total samples were colored green. Stations with greater than a 10 percent exceedance rate were identified as a potential concern and 6 Collier County Watershed Model Update MI and Plan Development July 9, 2010 colored yellow. The 10 percent exceedance value was used to be consistent with the IWR criteria for placing a water body in the FDEP's planning list. Additionally, stations that had exceedances in 50 percent or more of the samples were identified because they represent locations at which exceedances occur more often than not for the parameter of concern. These stations are colored red. The maps are included as Figures 2 -2 through 2 -6. 2.3. Source Water Determination An analysis of the pH and conductivity data within each watershed was reviewed to evaluate if the water chemistry can be used to detect quantifiable differences among the watersheds at a watershed level in terms of groundwater influence. Expectations are that groundwater contributions (even from the surficial aquifer) might result in waters with a more alkaline pH signature than surface water. Additionally, the conductivity values in groundwater might also be expected to be greater due to elevated mineral content. To show spatial patterns in a mapping view, pH values for each water quality station were categorized into four pH ranges: <6.8 >6.8 to 7.2 >7.2 to 7.8 >7.8 Each station was colored coded to identify which pH range contained the majority of values reported. A similar analysis was generated for conductivity, whereby stations were categorized into the four conductance ranges of: <400 µmhos / cm 400 -600 µmhos / cm 600 -800 µmhos / cm >800 µmhos / cm 7 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 2 -2. Total Nitrogen potential areas of concern by water quality station. 8 Collier County Watershed Model Update pwi and Plan Development July 9, 2010 Figure 2 -3. Total Phosphorus potential areas of concern by water quality station 9 Collier County Watershed Model Update ffl�f and Plan Development July 9, 2010 Figure 2 -4. Total Suspended Solids potential areas of concern by water quality station 10 Collier County Watershed Model Update PMJ and Plan Development July 9, 2010 Figure 2 -5. Color potential areas of concern by water quality station 106ilh- 11 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 2 -6. Bacteria (Fecal Coliform) potential areas of concern by water quality station 12 Collier County Watershed Model Update p"* and Plan Development July 9, 2010 3.0 Results and Discussion This section presents the results and discussion of the water quality characterization for fresh water discharges in each watershed, an evaluation of the critical water quality parameters, and source water determination. The water quality characterization portion is structured such that the water quality of the watersheds to the four priority estuaries of Wiggins Pass, Naples Bay, Rookery Bay and the Ten Thousand Islands are discussed separately. Summary water quality statistics were developed for each water quality station and watershed for the five parameters identified as critical for determining the estuarine water quality condition; total nitrogen, total phosphorus, total suspended solids, color, and bacteria. Fecal coliforms were used as the proxy for bacteria. 3.1. Wiggins Pass Eight WBIDs comprise the Cocohatchee- Corkscrew watershed that discharges fresh water into Wiggins Pass (Table 2 -1). Presently, WBIDs in the watershed are impaired for dissolved oxygen, mercury, nutrients, and un- ionized ammonia. A total of 79 water quality stations contain data for the parameters reviewed from 2000 to 2009 (Table 3 -1). Summary statistics by station are available in Appendix B. The water quality summary statistics are presented in Table 3 -2. Lake Trafford, located in the upstream reaches of the Cocohatchee- Corkscrew watershed, contributes to the freshwater and pollutant loads discharged into Wiggins Pass. It is important to note that due to the poor water quality conditions observed in Lake Trafford by FDEP, Collier County, and the SFWMD, a large -scale restoration project is underway. Water quality conditions that created the impairment are complex and no longer exist as they did during the timeframe used for the Lake Trafford TMDL report. Phase I of the Lake Trafford (WBID 3259W) sediment removal project has been completed and water quality improvements have been documented (i.e., PBS &J 2009). Phase II of the Lake Trafford sediment removal project is currently underway, and further water quality improvements are possible. Much of the data for Lake Trafford (discussed below) are from the period prior to implementation of Phase I of the Lake Trafford sediment removal project and may not accurately represent post - project conditions. 13 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Table 3 -1. List of stations with water quality data from 2000 to 2009 in the discharge waters of Cocohatchee- Corkscrew watershed 14 Collier County Watershed Model Update PW and Plan Development July 9, 2010 3259B 21FLFTM 28030072FTM 3259W 21FLGW 22728 3259B 21FLFTM 28030073FTM 3259W 21FLGW 22729 3259B 21FLSFWMCORKSW 3259W 21FLGW 22730 3259W 21 FLCOLLLKTRAF2 3259W 21FLGW 22731 3259W 2 1 FLCOLLLKTRAF3 3259W 21FLGW 22732 3259W 21FLCOLLLKTRAF4 3259W 21FLGW 22733 3259W 21FLCOLLLKTRAF5 3259W 21FLGW 22734 3259W 21FLCOLLLKTRAF6 3259W 21FLGW 22735 3259W 21FLCOLLLKTRAF7 3259W 21FLGW 22736 3259W 21FLGW 15160 3259W 21FLGW 22737 3259W 21FLGW 15162 3259W 21FLGW 22738 3259W 21FLGW 15165 3259W 21FLGW 22739 3259W 21FLGW 15169 3259W 21FLGW 22740 3259W 21FLGW 15171 3259W 21FLGW 22741 3259W 21FLGW 15181 3259W 21FLGW 22742 3259W 21FLGW 15185 3259W 21FLGW 22743 3259W 21FLGW 15187 3259W 21FLGW 22744 3259W 21FLGW 15188 3259W 21FLGW 22745 3259W 21FLGW 22719 3259W 21FLGW 22746 3259W 21FLGW 22720 3259W 21FLGW 22747 3259W 21FLGW 22721 3259W 21FLGW 3496 3259W 21FLGW 22722 3259W Trafford 3259W 21FLGW 22723 3278C 21FLGW 13716 3259W 21FLGW 22724 3278D 21FLFTM BC14 3259W 21FLGW 22725 3278D 21FLFTM EVRGWC0076FTM 3259W 21FLGW 22726 3278D 21FLFTM EVRGWC0077FTM 3259W 21FLGW 22727 3278D 21FLFTM EVRGWC0079FTM 3278E 21FLCOLLLKTRAFI 3278D 21FLFTMCOC @IBIS 3278E 21FLCOLLLKTRAF8 3278D 21FLFTMCOCPALM 3278E 21FLGWI3732 3278D 21FLFTMEVRGWC008OFTM 3278E 21FLSFWMIMKFSHCK 3278D 21FLGW 14183 3278F 21FLFTM 28030044 3278D 21FLSFWMBC13 3278F 21FLFTM CREW2 3278D 21FLSFWMBC14 3278F 21FLSFWMCORKN 3278D 21FLSFWMCOC @IBIS 3278F 21FLSFWMCORKS 3278D 21FLSFWMCOCAT41 3278F 21FLSFWMCORKSCRD 3278D 21FLSFWMCOCPALM 3278F Corkscrd 3278D BC 15 3278L 21FLSFWMIMK6STS 3278D Coco @Lake 3278L 21 FLSFWMIMKMAD 3278D ECocoRiv 3278L 21FLSFWMIMKSLGH 14 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Table 3 -2. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Cocohatchee- Corkscrew watershed teen BOD, mg/1 491 iii 0.5 Mein 3.6 Ma ... b 3.2 26.7 - Chlorophyll -a, ug/l 953 1.0 23.6 7.5 251.0 33 Color, PCU 748 5 85 80 420 27 Conductivity, umhos /cm 1327 3 3546 624 49624 - Copper, ug/l 290 0.15 4.00 1.65 178.00 2 DO, mg/l 1338 0.0 6.1 5.9 20.4 38 Fecal Coliform, # /100m1 896 1 275 78 8500 15 Iron, ug/l 195 100 425 360 2500 5 Nitrate - Nitrite, mg/l 713 0.000 0.081 0.030 1.328 - Orthophosphate as P, mg/l 880 0.004 0.091 0.035 0.580 Salinity, ppt 938 0.00 2.10 0.25 32.39 - Secchi Depth, m 1291 0.03 0.82 0.65 3.60 - TKN, mg/l 934 0.05 1.58 1.10 7.50 - Total Nitrogen, mg/1 849 0.005 1.362 1.020 6.585 32 Total Phosphorus, mg/l 827 0.004 0.128 0.070 11.000 20 TSS, mg/l 893 1.3 9.6 2.1 323.0 25 Turbidity, NTU 703 0.3 8.8 4.5 107.0 - Unionized Ammonia, mg/l 806 0.0000 0.0058 0.0013 0.2214 6 3.1.1. Nutrients (Total Nitrogen and Total Phosphorus) As indicated for the evaluation of stream water quality, screening levels were used to assess nutrient load characteristics because no numeric state standards exist. Thirty -two percent of the total nitrogen measurements exceeded the Florida screening levels of 1.6 mg/L for streams or 1.7 mg/1 for lakes. The median total nitrogen was 1.02 mg /L with a maximum of 6.59 mg/L. The majority of water quality stations which exceeded the appropriate screening level in > =50% of the samples were located in the most upstream portion of Cocohatchee- Corkscrew watershed in Lake Trafford (Figure 2 -2). Lake Trafford is the only WBID in the study area impaired for nutrients based on elevated chlorophyll a values. Total phosphorus concentrations were greater than the appropriate Florida screening levels of 0.22 mg/L for streams or 0.11 mg /1 for lakes in 20% of the samples. The total phosphorus median concentration was 0.07 mg/L with a maximum value of 11 mg /1. Elevated nutrients are of concern in the watershed. High total phosphorus concentrations (10 -100% exceedance) were found in the most upstream portions of the Cocohatchee- Corkscrew watershed (Figure 2 -3). In contrast, total phosphorus concentrations adjacent to Wiggins Pass were below the screening level, indicating that the load of total phosphorus is likely not of critical concern for the Wiggins Pass estuary. 15 Collier County Watershed Model Update NEI and Plan Development July 9, 2010 3.1.2. Total Suspended Solids Total suspended solids exceeded the appropriate calculated screening level of 13 mg/1 for lakes or 7 mg /l for streams in 25% of the samples. Median and maximum TSS values were 2.1 and 323 mg/l, respectively. Elevated TSS concentrations ( > =50% exceedance) were found in the most upstream portions of the Cocohatchee- Corkscrew in Lake Trafford (Figure 2 -4). Along the Tamiami Trail, TSS concentrations exceeded the screening level in the 10 -49% range. This leads to the conclusion that TSS loads should be considered a parameter of concern for the estuary. 3.1.3. Color Twenty -seven percent of the color measurements were above the appropriate calculated screening levels of 111.5 PCU for streams or 80 PCU for lakes. The median color was 80 with a maximum of 120 PCU. Elevated color ( > =50% exceedance) was found in the upstream portions of the Cocohatchee- Corkscrew but not near the discharge point into the estuary (Figure 2 -5). 3.1.4. Fecal Coliforms Bacteria Wiggins Pass has been declared impaired for fecal coliforms. Fecal coliform concentrations in the watershed exceeded 400 # /100mL in only 15% of the samples indicating that bacterial loading is not of major concern over the time period examined in the Cocohatchee- Corkscrew watershed. The median bacteria concentration was 78 # /IOOmL with a maximum value of 8,500 # /100mL. No stations had fecal coliforms concentrations with consistently greater than 400 # /100mL ( > =50% exceedance; Figure 2 -6). However, the downstream segment of the Cocohatchee- Corkscrew watershed has multiple stations with elevated bacteria loads (10 -49 %) suggesting a potential contamination to the Class 2 Wiggins Pass. It should be noted that Wiggins Pass is subject to more stringent fecal coliform standards due to the Class 2 designation which requires compliance with the 43 # /100mL to maintain healthy oyster populations. 3.2. Naples Bay Naples Bay receives discharges from the Golden Gate - Naples Bay watershed. The watershed includes two WBII)s (Table 2 -1). Those WBIDs have been found impaired for dissolved oxygen and iron concentrations. A total of 26 water quality stations contain data for the parameters reviewed over the period 2000 to 2009 (Table 3 -3). Summary statistics by station are available in Appendix B. The water quality summary statistics for the fresh water portion of the Golden Gate - Naples Bay watershed are presented in Table 3 -4. 16 Collier County Watershed Model Update n6i and Plan Development July 9, 2010 Table 3 -3. List of stations with water quality data from 2000 to 2009 in the discharge waters of Golden Gate- Naples Bay watershed Table 3 -4. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Golden Gate- Naples Bay watershed Mi4 w4etr N Min Mtan 3278S 21FLFTM EVRGWC0038FTM 3278S 21FLGW21746 3278S 21FLFTM EVRGWC0039FTM 3278S 21FLGW21753 3278S 21FLFTM EVRGWC0046FTM 3278K 21FLSFWMBC3 3278S 21FLFTM EVRGW00047FTM 3278S ARS896 -1 3278S 21FLFTM EVRGWC0048FTM 3278S Cork @846 3278S 21FLFTM EVRGWC0049FTM 3278S GC @858 3278S 21FLFTM EVRGWC0050FTM 3278S GGCAT31 32785 21FLFTM EVRGWC0051FTM 3278S GGCAT951 3278S 21FLFTM EVRGWC0052FTM 3278S GG03 @32 3278S 21FLFTM EVRGWC0053FTM 3278K GRE896 -1 3278S 21FLGW 3495 3278K GordonRiv 3278S 21FLGW 37106 3278S I951_Immokalee 3278S 21FLGW14182 3278S Lon shore Table 3 -4. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Golden Gate- Naples Bay watershed Mi4 w4etr N Min Mtan Median Mgt Percent Exceed BOD, mg/1 151 0.6 2.1 2.0 8.1 - Chlorophyll -a, ug/1 646 0.9 6.6 3.0 150.0 5 Color, PCU 589 5 92 80 800 26 Conductivity, umhos /cm 600 0 2228 608 40222 - Copper, ug/1 189 0.15 1.42 1.00 16.00 0 DO, mg/l 612 0.2 5.4 5.4 16.1 45 Fecal Coliform, # /100ml 538 1 131 30 5400 6 Iron, ug/1 185 62 582 503 1520 18 Nitrate - Nitrite, mg/1 588 0.003 0.055 0.035 0.750 - Orthophosphate as P, mgA 489 0.004 0.022 0.008 1.100 - Salinity, ppt 443 0.00 1.66 0.28 25.62 - Secchi Depth, m 603 0.00 1.18 1.10 6.00 33 TKN, mgA 553 0.04 0.85 0.77 5.30 - Total Nitrogen, mg/1 561 0.005 0.803 0.783 5.690 5 Total Phosphorus, mg/l 613 0.006 0.042 0.026 1.500 1 TSS, mg/l 480 2.0 3.7 2.0 94.0 5 Turbidity, NTU 430 0.2 2.5 2.0 19.5 Unionized Ammonia, mg/1 516 0.0000 0.0020 0.0006 0.2406 l 17 Collier County Watershed Model Update misi and Plan Development July 9, 2010 3.2.1. Nutrients (Total Nitrogen and Total Phosphorus) Nutrient concentrations in the watershed were consistently less than the screening standards. Five percent of the total nitrogen measurements were above the appropriate Florida screening levels of 1.6 mg/L for streams. The median total nitrogen was 0.78 mg/L with a maximum of 5.69 mg/L. Total phosphorus were greater than the appropriate Florida screening levels of 0.22 mg /L for streams or 0.11 mg /1 for lakes in only 1 % of the samples. The total phosphorus median concentration was 0.026 mg /L with a maximum value of 1.5 mg /l. Based on the low nutrient concentrations in the watershed, particularly in the downstream section, nutrient loads are expected to have a minimal impact on Naples Bay. 3.2.2. Total Suspended Solids Total suspended solids exceeded the appropriate calculated screening level of 7 mg /1 for streams in 5% of the samples. Median and maximum TSS values were 2 and 94 mg /1, respectively. Along the Tamiami Trail, TSS values were elevated (10 -49% exceedance) prior to discharging into Naples Bay estuary (Figure 2 -4). These elevated values indicate that TSS loads from the watershed are of concerned for the estuary's water quality conditions. 3.2.3. Color Twenty -six percent of the color measurements were above the appropriate calculated screening levels of 111.5 PCU for streams or 80 PCU for lakes. The median color was 80 with a maximum of 800 PCU. Elevated color ( > =50% exceedance) was found in the upstream portions of the Golden Gate - Naples Bay watershed (Figure 2 -5). As described for the stream water quality analysis, low levels of dissolved oxygen in the upstream portion of the watershed are in response to the elevated levels of color. The data analysis indicated that color values decrease downstream and approach screening levels near the confluence with the estuary. Therefore, the impact of color on the Naples Bay estuary is likely minimal. 3.2.4. Bacteria (Fecal Coliforms) Fecal coliform concentrations exceeded 400 # /100mL in 6% of the samples. The median bacteria concentration was 30 # /100mL and a maximum value of 5,400 # /100mL. Bacterial loads are low indicating a minimal health risk within the watershed. Concentrations in the upstream portion of the watershed were consistently below the regulatory standard (Figure 2 -6). The most downstream sampling station exceeded the fecal coliform standard in 10 to 50% of all samples. None of the WBIDs in the Golden Gate - Naples Bay watershed are impaired for elevated bacteria loads. However, Naples Bay estuary is impaired. The estuary is subject to more stringent fecal coliform standards due to the Class II designation which requires compliance with the 43 # /IOOmL to maintain healthy oyster populations. As such, bacterial loads from the upstream watershed could negatively influence estuarine conditions. In contrast to the Cocohatchee- Corkscrew watershed, nutrient concentrations in the Golden Gate - Naples Bay watershed are only infrequently above screening criteria levels. Levels of TN 18 Collier County Watershed Model Update M$#0 and Plan Development July 9, 2010 exceed screening criteria 5 percent of the time, while TP levels exceeded screening criteria 1 percent of the time. These data suggest that nutrient enrichment from the watershed is not likely a large concern for the downstream estuary of Naples Bay; elevated levels of chlorophyll a in Naples Bay are not likely due to nutrient enrichment alone. 3.3. Rookery Bay Two WBIDs comprise the Rookery Bay watershed contributing area (Table 2 -1). Presently, there are no impaired WBIDs in the watershed. A total of five water quality stations contain data for the parameters reviewed from 2000 to 2009 (Table 3 -5). Summary statistics by station are available in Appendix B. The water quality summary statistics for the fresh water portion of the Rookery Bay watershed are presented in Table 3 -6. 3.3.1. Nutrients (Total Nitrogen and Total Phosphorus) Both TN and TP concentrations were consistently below the Florida screening level (Figures 2 -2 and 2 -3). Four percent of the total nitrogen measurements were above the appropriate Florida screening levels of 1.6 mg /L for streams or 1.7 mg/1 for lakes. The median total nitrogen was 0.66 mg /L with a maximum of 4.41 mg /L. Total phosphorus were greater than the appropriate Florida screening levels of 0.22 mg/L for streams or 0.11 mg/l for lakes in 2% of the samples. The total phosphorus median concentration was 0.023 mg /L with a maximum value of 0.63 mg/1. Nutrient loads from the Rookery Bay watershed to the estuary are not a major concern for the estuary. 3.3.2. Total Suspended Solids Total suspended solids exceeded the appropriate calculated screening level of 13 mg /l for lakes or 7 mg /l for streams in 6% of the samples. Median and maximum TSS values were 2 and 56 mg /l, respectively. Total suspended loads to Rookery Bay should be minimal based on the consistently low concentrations observed along Tamiami Trail (Figure 2 -4). 3.3.3. Color Eight percent of the color measurements were above the appropriate calculated screening levels of 111.5 PCU for streams or 80 PCU for lakes. The median color was 50 with a maximum of 240 PCU. Water quality in the Rookery Bay watershed is consistently within regulatory and screening levels (Figure 2 -5). While there are not current impaired WBIDs within the watershed, dissolved oxygen values were below the 5.0 mg /L regulatory standard in 39% of the samples reviewed. Depressed dissolved oxygen values may be a function of the natural landscape (see Technical Memorandum Element 1, Task 1.2: In- stream Water Quality). 3.3.4. Bacteria (Fecal Coliforms) Fecal coliform concentrations exceeded 400 # /100mL in 6% of the samples. The median bacteria concentration was 40 # /100mL with a maximum value of 2,600 # /100mL. Low bacterial 19 Collier County Watershed Model Update PW and Plan Development July 9, 2010 loads are observed in the watershed with values consistently below the regulatory standard (Figure 2 -6). Rookery Bay estuary is impaired for bacteria levels. The estuary is subject to more stringent fecal coliform standards due to the Class II designation which requires compliance with the 43 # /100mL to maintain healthy oyster populations. As such, bacterial loads from the upstream watershed could negatively influence estuarine conditions. Table 3 -5. List of stations with water quality data from 2000 to 2009 in the discharge waters of Rookery Bay watershed Table 3 -6. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Rookery Bay watershed lh 3278V 21FLGW21745 3278V 21FLGW21747 3278V 21FLGW21757 3278V 21FLSFWMBC22 3278Y 21FLSFWMLELY Table 3 -6. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Rookery Bay watershed lh BOD, mg/1 35 0.8 2.1 2.0 4.7 - Chlorophyll -a, ug/1 150 3.0 6.1 3.2 88.0 3 Color, PCU 147 20 58 50 240 8 Conductivity, umhos /cm 146 182 1592 807 24400 - Copper, ug/l 50 0.30 3.33 1.00 54.00 2 DO, mg/l 150 1.3 5.6 5.7 11.4 39 Fecal Coliform, # /100ml 134 1 106 40 2600 6 Iron, ug/1 45 0 249 220 770 0 Nitrate - Nitrite, mg/l 142 0.003 0.038 0.020 0.250 - Orthophosphate as P, mg/1 120 0.004 0.008 0.005 0.067 Salinity, ppt 137 0.09 0.85 0.41 14.74 - Secchi Depth, m 141 0.10 1.00 1.00 1.80 45 TKN, mg/l 132 0.24 0.75 0.64 4.40 - Total Nitrogen, mg/1 135 0.010 0.675 0.660 4.409 4 Total Phosphorus, mg/1 132 0.007 0.037 0.023 0.630 2 TSS, mg/l 122 2.0 3.6 2.0 56.0 6 Turbidity, NTU t F 0.4 2.4 1.4 60.0 - Unionized Ammonia, mg/1 124 0.0000 0.0009 0.0006 0.0088 0 20 Collier County Watershed Model Update PW and Plan Development July 9, 2010 In the Rookery Bay watershed, nutrient are only infrequently above screening criteria levels. Levels of TN exceed screening criteria 4 percent of the time, while TP levels exceeded screening criteria 2 percent of the time in the Rookery Bay watershed. These data suggest that nutrient enrichment from the watershed is not likely a large concern for the downstream estuary of Rookery Bay, and elevated levels of chlorophyll a in Rookery Bay are not likely due to nutrient enrichment alone. 3.4. Ten Thousand Islands (TTI) Three watersheds discharge fresh water into the TTI: Faka- Union, Fakahatchee, and Okaolacoochee -SR29. A total of seven WBIDs comprise the three watersheds that provide the fresh water discharge into TTI (Table 2 -1). Presently, WBIDs in the fresh water discharge portion of the three watersheds are impaired for dissolved oxygen, fecal coliforms, and iron. A total of forty -seven water quality stations contain data for the parameters reviewed from 2000 to 2009 (Table 3 -7). Summary statistics by station are available in Appendix B. The water quality summary statistics for the fresh water portion of the Faka- Union, Fakahatchee, and Okaolacochee -SR29 watersheds are presented in Tables 3 -8 to 3 -10. Table 3 -7. List of stations with water quality data from 2000 to 2009 in the discharge waters to the Ten Thousand Islands iD W ter rieel atne ' B- W Wat r h Name 3278I Faka Union 21FLGW14163 3261C Okaloacoochee -SR29 112WRD 260231081203900 3278I Faka Union 21FLGW14166 3261C Okaloacoochee -SR29 21FLGW 3494 3278H Faka Union 21FLGW 14181 3278T Okaloacoochee -SR29 21FLGW 11114 3278H Faka Union 21FLGW14184 3278W Okaloacoochee -SR29 21FLGWI3713 32781 Faka Union 21FLGW21749 3261C Okaloacoochee -SR29 21FLGWI3736 32781 Faka Union 21FLGW21750 3278W Okaloacoochee -SR29 21FLGWI4159 3278H Faka Union 21FLGW21752 3261C Okaloacoochee -SR29 21FLGWI4162 3278H Faka Union 21FLGW21756 3261C Okaloacoochee -SR29 21FLGW14164 32781 Faka Union 21FLGW21758 3278T Okaloacoochee -SR29 21FLGW14167 3278H Faka Union 21FLSFWMBCIO 3261C Okaloacoochee -SR29 21FLGW14168 3278I Faka Union 21FLSFWMBC20 3278W Okaloacoochee -SR29 21FLGWI5184 3278I Faka Union 21FLSFWMBC7 3261C Okaloacoochee -SR29 21FLGW21744 32781 Faka Union 21FLSFWMBC8 3261C Okaloacoochee -SR29 21FLGW21748 3278H Faka Union 21FLSFWMBC9 3261C Okaloacoochee -SR29 21FLSFWMBC24 32781 Faka Union 21FLSFWMFAKA 3278W Okaloacoochee -SR29 21FLSFWMIMKBRN 3278H Faka Union CR858 3278T Okaloacoochee -SR29 A01_Nbear 32781 Faka Union SGGEIOSW 3278T Okaloacoochee -SR29 BCAP1 32781 Faka Union SGGEI ISW 3278T Okaloacoochee -SR29 OKALA858 32781 Faka Union SGGE16SW 3278G Fakahatchee 21FLSFWMBC19 32781 Faka Union SGGE22SW 3278G Fakahatchee 21FLSFWMBC21 32781 Faka Union SGGE23SW 3259I Fakahatchee 21FLSFWMBC25 3259I Fakahatchee 21FLSFWMBC11 3278G Fakahatchee 21FLSFWMSGGE17SW 3278G Fakahatchee 21FLSFWMBC12 3278G Fakahatchee Chkmate 3278G Fakahatchee 21FLSFWMBCI8 21 Collier County Watershed Model Update Mf and Plan Development July 9, 2010 3.4.1. Nutrients (Total Nitrogen and Total Phosphorus) In the Faka Union watershed, three percent of the total nitrogen measurements were above the appropriate Florida screening levels of 1.6 mg/L for streams or 1.7 mg /l for lakes. The median total nitrogen was 0.473 mg/L with a maximum of 5.03 mg/L. Less than 1% of the total phosphorus concentrations were greater than the appropriate Florida screening levels of 0.22 mg /L for streams or 0.11 mg /l for lakes. The total phosphorus median concentration was 0.015 mg /L with a maximum value of 0.435 mg/l. In the Fakahatchee watershed, seven percent of the total nitrogen measurements were above the appropriate Florida screening levels of 1.6 mg /L for streams or 1.7 mg/I for lakes. The median total nitrogen was 0.67 mg/L with a maximum of 5.32 mg/L. Total phosphorus concentrations were greater than the appropriate Florida screening levels of 0.22 mg/L for streams or 0.11 mg/1 for lakes in 3% of the samples. The total phosphorus median concentration was 0.02 mg/L with a maximum value of 1.18 mg /1. In the Okaloacochee /SR29 watershed, twenty percent of the total nitrogen measurements were above the appropriate Florida screening levels of 1.6 mg/L for streams or 1.7 mg/l for lakes. The median total nitrogen was 0.873 mg /L with a maximum of 35.35 mg/L. Total nitrogen levels are elevated indicating a potential concern for nutrient loads to the TTI. Total phosphorus concentrations were greater than the appropriate Florida screening levels of 0.22 mg /L for streams or 0.11 mg/l for lakes in 9% of the samples. The total phosphorus median concentration was 0.03 mg/L with a maximum value of 0.72 mg /1. In the Faka Union and Fakahatchee watersheds just upstream of the Ten Thousand Islands, stations had slightly elevated TN concentrations (10 -49% exceedance) along the Tamiami Trail indicating the potential for elevated TN loads (Figure 2 -2). High total phosphorus values (10- 100% exceedance) were found in the most upstream portions of the Fakahatchee and Okaolacoochee -SR29 watersheds (Figure 2 -3). However, total phosphorus concentrations adjacent to the priority estuaries were not elevated, indicating that elevated total phosphorus loads are perhaps unlikely to the Ten Thousand Islands. 3.4.2. Total Suspended Solids In the Faka Union watershed, total suspended solids exceeded the appropriate calculated screening level of 13 mg/l for lakes or 7 mg /l for streams in 6% of the samples. Median and maximum TSS values were 2 and 62 mg/l, respectively. Total suspended solids in the Fakahatchee watershed exceeded the appropriate calculated screening level of 13 mg /1 for lakes or 7 mg /1 for streams in 11% of the samples. Median and maximum TSS values were 2 and 97 mg /1, respectively. Okaloacochee -SR29 watershed total suspended solid concentrations exceeded the appropriate calculated screening level of 13 mg/1 for lakes or 7 mg /1 for streams in 10% of the samples. Median and maximum TSS values were 4 and 174 mg /1, respectively. TSS values in the Fakahatchee watershed were consistently high with exceedance values 10 -49% for most stations (Figure 2 -4). Along the Tamiami Trail within the Faka Union and Fakahatchee watersheds, TSS values were elevated (10 -49% exceedance) prior to discharging into the 22 Collier County Watershed Model Update pff$10 and Plan Development July 9, 2010 adjacent estuary. These elevated values indicate the potential for increased TSS loads into the TTI estuary. 3.4.3. Color Twelve percent of the color measurements in the Faka Union watershed were above the appropriate calculated screening levels of 111.5 PCU for streams or 80 PCU for lakes. The median color was 50 with a maximum of 240 PCU. In the Fakahatchee watershed, twenty -three percent of the color measurements were above the appropriate calculated screening levels of 111.5 PCU for streams or 80 PCU for lakes. The median color was 80 with a maximum of 400 PCU. In contrast, thirty -six percent of the color measurements in the Okaloacochee -SR29 watershed were above the appropriate calculated screening levels of 111.5 PCU for streams or 80 PCU for lakes. The median color was 80 with a maximum of 450 PCU. Elevated color ( > =50% exceedance) was found in the upstream portions of the Okaolocoochee -SR29 watershed but not near the confluence with the estuary. Within the Fakahatchee watershed, color was consistently greater than the screening levels (10 -49% exceedance) indicating a constant color source, most likely due to the extensive forested wetlands throughout the landscape (Figure 2 -5). The influence of color on dissolved oxygen and phytoplankton production in the TTI may be impacted by discharge from the Fakahatchee watershed. 3.4.4. Bacteria (Fecal Coliforms) Bacteria concentrations were low in both the Faka Union and Okaolocochee -SR29 watershed when compared to the Class 3 regulatory standard. In the Faka Union watershed, fecal coliform concentrations exceeded 400 # /100mL in 8% of the samples. The median bacteria concentration was 22 # /100mL and a maximum value of 3,850 # /100mL. In the Fakahatchee watershed, fecal coliform concentrations exceeded 400 # /100mL in 12% of the samples. The median bacteria concentration was 52 # /100mL with a maximum value of 5,450 # /100mL. In the Okaolocochee- SR29 watershed, fecal coliform concentrations exceeded 400 # /100mL in 5% of the samples. The median bacteria concentration was 28 # /100mL with a maximum value of 3,050 # /100mL. The Fakahatchee watershed is impaired for fecal coliforms and has several stations with bacteria loads elevated above the regulatory standard yet the TTI are not impaired for bacteria (Figure 2- 6). As such, bacterial loads from the Fakahatchee watershed may negatively impact water quality in the TTI. 23 Collier County Watershed Model Update MJ and Plan Development July 9, 2010 Table 3 -8. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Faka Union watershed Pa"Ift "ter Min Mean Mai" Maz' BOD, mg /1 132 1.2 2.2 2.0 8.5 - Chlorophyll -a, ug/1 538 0.9 6.3 3.0 206.0 5 Color, PCU 518 5 62 50 240 12 Conductivity, umhos /cm 542 211 2185 567 62047 - Copper, ug/1 166 0.15 1.37 1.00 17.70 0 DO, mg/l 561 0.4 5.9 5.9 14.5 38 Fecal Coliform, # /100m1 465 1 133 22 3850 8 Iron, ug/1 190 100 304 210 1390 2 Nitrate - Nitrite, mg/I 531 0.000 0.027 0.010 1.310 - Orthophosphate as P, mg/l 431 0.004 0.007 0.005 0.099 0.368 Salinity, ppt 532 0.00 1.20 0.27 41.65 - Secchi Depth, m 334 0.10 1.16 1.20 2.50 36 TKN, mg/1 479 0.04 0.61 0.52 4.90 - Total Nitrogen, mg /l 488 0.005 0.526 0.473 5.030 3 Total Phosphorus, mg/1 515 0.004 0.024 0.015 0.435 0 TSS, mg/l 455 2.0 3.2 2.0 62.0 6 Turbidity, NTU 340 0.1 1.9 1.3 32.0 - Unionized Ammonia, mg/l 449 0.0000 0.0006 0.0003 0.0127 0 Table 3 -9. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Fakahatchee watershed Parameter Min Mean Media ', Maz' i�ercent Exceed BOD, mg/l 112 1.5 2.4 2.0 9.8 - Chlorophyll -a, ug/1 454 3.0 11.5 3.0 993.0 9 Color, PCU 438 5 80 80 400 23 Conductivity, umhos /cm 456 197 5371 594 72958 - Copper, ug/1 138 0.15 1.14 1.00 8.00 0 DO, mg/l 470 0.2 3.8 3.3 12.8 75 Fecal Coliform, # /100ml 406 1 198 52 5450 12 Iron, ug/1 154 0 212 1 150 1300 1 Nitrate - Nitrite, mg/l 447 0.000 0.016 0.010 0.220 - Orthophosphate as P, mg/l 373 0.004 0.020 0.006 0.368 - Salinity, ppt 463 0.00 3.25 0.29 50.25 - Secchi Depth, m 380 0.10 1.03 1.00 2.80 49 TKN, mg /l 414 0.04 0.90 0.75 6.75 - Total Nitrogen, mg/1 411 0.005 0.721 0.670 5.320 7 Total Phosphorus, mg/l 427 0.004 0.049 0.020 1.180 3 TSS, mg/l 390 2.0 5.0 2.0 97.0 11 Turbidity, NTU 293 0.1 1.5 0.7 150.0 - Unionized Ammonia, mg/l 372 0.0000 1 0.0007 0.0003 0.0162 0 24 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Table 3 -10. Water quality summary statistics from 2000 to 2009 in the discharge waters of the Okaolacochee -SR29 watershed r me BOD, mg11 N 48 M 1.6 a n- 2.2 V'Odian 2.0 MAX._ 5.1 Percent Exceed - Chlorophyll -a, ug/1 277 0.9 7.0 3.0 69.4 13 Color, PCU 301 0 96 80 450 36 Conductivity, umhos /cm 347 103 471 465 905 - Copper, ug /I 103 0.15 2.76 1.50 19.07 6 DO, mg/I 347 0.1 2.8 2.5 9.3 88 Fecal Coliform, # /100ml 254 1 108 28 3050 5 Iron, ug/1 49 0 478 250 1910 18 Nitrate- Nitrite, mg/l 331 0.003 0.053 0.010 1.670 - Orthophosphate as P, mg/1 291 0.002 0.045 0.012 0.627 Salinity, ppt 187 0.00 0.22 0.20 0.45 - Secchi Depth, m 304 0.10 1.19 1.10 4.30 40 TKN, mg/l 326 0.04 1.26 0.97 35.35 - Total Nitrogen, mg11 321 0.005 1.187 0.873 35.353 20 Total Phosphorus, mg/I 331 0.005 0.081 0.030 0.718 9 TSS, mg/I 279 2.0 4.9 4.0 174.0 10 Turbidity, NTU 221 0.2 1.6 0.8 20.0 - Unionized Ammonia, mg/I 300 0.0000 0.0019 0.0003 0.3241 1 In the Faka Union and Fakahatchee watershed, nutrients are more likely to exceed screening criteria levels in the Okaloacochee -SR 29 watershed. Levels of TN exceed screening criteria 20 percent of the time in the Okaloacochee -SR 29 watershed, while TP levels exceeded screening criteria 9 percent of the time. Of the three watersheds that drain into the Ten Thousand Islands estuary, these data suggests that nutrient enrichment from the watershed is not likely a concern for the Faka Union and Fakahatchee watersheds, but nutrient enrichment could be a concern for fresh water discharging from the Okaloacochee -SR 29 watershed. In all three watersheds, dissolved oxygen was consistently lower, while color was higher, when compared with the regulatory standard. However, Technical Memorandum Element 1, Task 1.2: In- stream Water Quality presents recommendations for developing site specific alternate criteria for dissolved oxygen, as low levels of dissolved oxygen are often found in areas with the most natural landscape features in all three watersheds. 3.5. Source Water Determination It has been previously postulated that water sources (groundwater or surface water), at least at various time of the year, might be different when comparing the Faka Union and Fakahatchee watersheds. Due to the extensive network of canals, groundwater discharge from the surficial aquifer has been identified as a more dominant influence in the Faka Union watershed. As can be seen in Figures 3 -6 and 3 -7, no clear pattern in pH or conductivity exists that would support the contention that the Faka Union watershed is mostly characterized by surface waters that are influenced more strongly by groundwater, with its expected higher levels of specific conductance 25 Collier County Watershed Model Update and Plan Development July 9, 2010 and pH, as compared to the Fakahatchee watershed, which is characterized more by surface waters with lower specific conductance and pH. Instead, pH and conductivity values were similar in both the Faka Union and Fakahatchee watersheds. Figure 3 -6. pH values by water quality station pwi 26 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 3 -7. Conductivity values by water quality station 27 Collier County Watershed Model Update and Plan Development July 9, 2010 4.0 Conclusions Water quality data, including pollutant concentrations, of fresh water discharges to estuaries along the coast of Collier County were reviewed and evaluated. Wiggins Pass, Naples Bay, and Rookery Bay estuaries were all found to be impaired for dissolved oxygen by FDEP, while Ten Thousand Islands estuary was not (see Technical Memorandum for Element 2, Task 3: Quality of Receiving Waters). Rookery Bay was also found to be impaired for nutrients, based on elevated levels of chlorophyll a. However, violations of water quality in these estuaries may reflect factors other than anthropogenic nutrient loading. Greater attention should be given to those estuaries where the corresponding watershed has elevated levels of nutrients and /or total suspended solids. Using a screening level for TN of 1.6 mg / liter for streams and 1.7 mg / liter for lakes, only Lake Trafford (in the Cocohatchee- Corkscrew watershed, which drains to Wiggins Pass estuary) had average TN levels that exceed screening thresholds (i.e., concentrations greater than screening levels more than 50 percent of the time)A few stations near Tamiami Trail (in the Faka Union and Fakahatchee watershed) and in the Okaloacochee -SR 29 watershed farther north have TN levels above the screening level between 10 and 50 percent of the time. Most water quality data, including those from the developed portions of the Naples Bay watershed, do not exceed 1.6 mg TN / liter more than 10 percent of the time. The calculated screening level for TP was 0.22 mg / liter for streams and 0.11 mg / liter for lakes. Using these numbers, only Lake Trafford again had TP levels that consistently exceeded screening thresholds (i.e., concentrations greater than screening levels more than 50 percent of the time). A few stations in the headwaters of the Cocohatchee- Corkscrew and Okaloacochee- SR 29 watersheds also exhibited elevated levels of TP. As in TN, the majority of stations, including those within the developed portions of the Naples Bay watershed, do not exceed screening levels of TP more than 10 percent of the time. Based on the results of our analysis, Collier County's estuaries do not appear to be adversely impacted by high TN and TP loadings from their corresponding watersheds. The watersheds are not characterized by have consistently high levels of TN or TP, with the exception of Lake Trafford (in the Cocohatchee- Corkscrew watershed which discharges to Wiggins Pass estuary). Instead, the majority of ecological dysfunction in Collier County's estuaries appears to be related to altered timing and quality of fresh water inflows, rather than eutrophication due to anthropogenic nutrient loading. 28 Collier County Watershed Model Update and Plan Development July 9, 2010 5.0 References Black, Crow, and Eidsness, Inc. 1974, Hydrologic Study of the G. A. C. Canal Network. Gainesville, FL. Project no. 449- 73 -53. Friedemann, M., and J. Hand, 1989, Typical water quality values for Florida's lakes, streams and estuaries: Florida Department of Environmental Regulation, Tallahassee, 31 p. Klein, H., W.J. Schneider, B.F. McPherson and T.J. Buchanan. May 1970. Some Hydrologic and Biologic Aspects of the Big Cypress Swamp Drainage Area, Southern Florida. United States Geologic Survey Open -file Report 70003. PBSJ, 2009. Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318, Element 4, Task 3 Water Quality and Ecological Assessment of Lake Trafford. Submitted To Collier County, 29 Collier County Watershed Model Update ITS? and Plan Development July 9, 2010 Appendix A Water Quality Station List Collier County Watershed Model Update PW and Plan Development July 9, 2010 Appendix B Discharge Water Quality Summary Statistics by Station Collier County Watershed Model Update PW and Plan Development July 9, 2010 ,+ Technical Memorandum To: Mac Hatcher, PM Collier County From: Moris Cabezas, PBS &J Dave Tomasko, PBS &J Emily Keenan, PBS &J Date: July 9, 2010 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 2, Task 3: Quality of Receiving Waters 1.0 Objective This Technical Memorandum will address Element 2, Task 3: Quality of Receiving Waters. The objective of this task is to characterize the water quality conditions in the receiving waters of the four primary estuaries of concern: • Wiggins Pass • Naples Bay • Rookery Bay • Ten Thousand Islands (TTI) Six watersheds were identified as the headwaters to the four estuaries of interest for this project (Figure 1 -1). The Wiggins Pass estuary is located at the outfall from the Cocohatchee- Corkscrew watershed. Naples Bay estuary receives discharge from the Golden Gate - Naples Bay watershed, while the Rookery Bay watershed discharges into Rookery Bay estuary. The Ten Thousand Islands (TTI) estuary is influenced by inflows from three main watersheds: Faka Union, Fakahatchee, and Okaloacoochee -SR29. 2.0 Introduction This Technical Memorandum focused on the downstream, estuarine portions of the above - listed watersheds in Collier County. These estuaries are influenced by the quantity, timing, and quality of inflow from their associated watersheds. Characterization of the quality of water within the watersheds was the focus of the technical memorandum prepared for Phase 2, Element 2, Task 2. The main impact to the Collier County estuaries have resulted from changes in historic fresh water flow patterns over the years due to increased development. These hydrologic changes have adversely impacted the environmental integrity of many of the estuaries, mostly in terms of widely varying salinity patterns (Browder et al. 1998, Shirley et al. 2005) and the reported 1 Collier County Watershed Model Update PW and Plan Development July 9, 2010 introduction of large quantities of organic -rich sediments from accelerated rates of freshwater inflow (Locker 2005). Specifically about Ten Thousand Islands, much of the scientific literature focused on the issue of altered hydrology and the need for a more natural pattern of freshwater inflow (e.g., Browder et al. 1988, Shirley et al. 2005). Figure 1 -1. Collier County Estuaries and Major Features Estuary Locations _ W1g9 {ns pasc ap�•a 8 y 1,, ff t v 't✓ j a, x13 3 Y " • Legend i �, f k. - -•• -• Esluay Boundary T.W Mush County BOL"ay /. 3.0 Methodology In order to accurately characterize the receiving waters of the Collier County estuaries, PBS &J completed a review of the existing impaired water bodies as defined by Florida Department of Environmental Protection (FDEP) and compared these results to available water quality data within the estuarine portion of each watershed: Cocohatchee- Corkscrew, Golden Gate - Naples Bay, Rookery Bay, Faka Union, Okaloacochee - SR29, and Fakahatchee. A discussion of the analysis conducted is presented below. 3.1. FDEP Impaired WBIDs For implementation of the statewide Total Maximum Daily Load (TMDL) program, the FDEP has divided the state into five groups. Each group is comprised of multiple basins. All water bodies within Collier County are located within the Everglades West Coast Group 1 Basin. Per TMDL guidelines, every five years each WBID is evaluated to determine if available water quality parameters exceed the limits defined by FDEP in the Impaired Waters Rule (IWR). The verified impaired list of WBIDs for each group and cycle is available on the FDEP website. 2 Collier County Watershed Model Update ITS1 and Plan Development July 9, 2010 After the compilation of all impaired WBIDs from Cycle 1 and 2, a total of ten impairments have been designated by FDEP for the four WBIDs representing the estuaries. Those WBIDs are listed in Table 3 -1. Table 3 -1. WBID Name and corresponding estuarine receiving water WBID# WBID Name Receiving Water 3259A Cocohatchee River Wiggins Pass 3278R Naples Bay (Coastal Segment) Naples Bay 3278U Rookery Bay (Coastal Segment) Rookery Bay 3259M Ten Thousand Islands Ten Thousand Islands 3.2. Water Quality Analysis As was done for characterizing the quality of water discharging into the estuaries, the IWR Run 39 dataset was supplemented with data from Florida STORET, Collier County, City of Naples, and the Rookery Bay National Estuarine Research Reserve (RBNERR) to create a comprehensive water quality database. To eliminate potential errors due to apparent data duplications for water quality stations (possibly due to multiple agencies uploading the same data, a single agency loading the data more than once with slight variations like rounding errors, etc.), median values were calculated by station, date, and parameter. For field parameters such as water temperature and dissolved oxygen, all data with water depths greater than one meter were analyzed no further. This ensured that any comparisons of field parameters to lab parameters (i.e., nutrients) were from samples taken at the same water depth. Using GIS and the station descriptions, the location of water quality stations were reviewed in order to identify locations where multiple stations were sampled. Data were merged when more than one water quality station was sampled at a location and a unique merged station name was assigned to that location. Appendix A lists all water quality stations and assigned merged station names. Each parameter in the database was screened to identify outliers or entry errors due to unit inconsistencies. Identified inconsistencies were reviewed and corrected. When Total Nitrogen (TN) species were not listed, TN was calculated through the addition of Total Kjeldahl Nitrogen (TKN) and Nitrate + Nitrite (NOx). For chlorophyll a data consistent with IWR, corrected chlorophyll a was preferentially used over uncorrected chlorophyll a when available for samples collected in 2006 and earlier. After 2006, only corrected chlorophyll a data were used. All statistical analysis was completed using the most recent ten year time period (2000 to 2009) to characterize each watershed. All water quality stations retrieved from the IWR database or Florida STORET were previously assigned to a WBID by FDEP. Water quality station data provided by Collier County, City of Naples, or RBNERR were assigned to a WBID based on location coordinates (Figure 3 -1). A list of the parameters analyzed for each station and receiving water is provided in Table 3 -2. 3 Collier County Watershed Model Update MS1 and Plan Development July 9, 2010 Figure 3 -1. Water quality monitoring station location map 4 Collier County Watershed Model Update PMj,k and Plan Development July 9, 2010 Summary data for each estuary were compared to the Criteria for Surface Water Quality Classifications (F.A.C. 62- 302.530) for water quality parameters based on water body classification. All of the receiving waters are Class II water bodies, i.e. their designated use is shellfish propagation and harvesting. In addition, FDEP's anti - degradation policy (62- 302.300 FAC) allows for protection of water quality above the minimum required for a classification and Class II water bodies, i.e. "water quality sufficient for the protection and propagation of fish, shellfish, and wildlife, as well as for recreation in and on the water, is an interim goal to be sought whenever attainable ". Table 3 -3 lists the regulatory standards for a Class 2 water body for the selected parameters previously identified by the FDEP TMDL program as verified parameter. Regulatory standards have been vetted by the scientific community and provide a biologically relevant basis for comparison. To further evaluate potential water quality impairments for chemical parameters for which no numeric water quality standard currently exist, the data were compared to screening level standards, which can provide an indication of potential water quality concerns but do not necessarily constitute an impairment problem. Screening level standards are available for TN and TP based on the 70`h percentile of all available data, as in Friedman and Hand (1989). Using IWR Run 39, a similar screening level was calculated by waterbody type for color, total suspended solids, and secchi depth, in which the 70`h percentile of all data available from 2000 to 2009 by waterbody type was calculated. Table 3 -4 shows the screening level standard for selected parameters by water body type (estuary). 3.3. Impaired WBID Comparison Using methods similar to FDEP IWR, PBS &J analyzed the water quality data for each WBID All analyses were conducted using the most recent ten year time period (2000 to 2009) to minimize the effect of temporal variations. Also, it was determined that the majority of water quality data available was collected during this ten year period. All data collected for each WBID during that period were used to evaluate the parameters previously declared "verified impaired" by FDEP. Dissolved oxygen, iron, fecal coliforms, nutrients (chlorophyll a), and copper concentrations were compared to the appropriate state regulatory standard to determine impairment status (Table 3 -3). A modification to the FDEP method for determining chlorophyll a impairments was utilized. Each chlorophyll a value was compared to the state regulatory standard and the percent exceedance was calculated. In contrast, FDEP calculates an annual average using data from each quarter for comparison with the regulatory standard. The results of PBS &J WBID analysis were compared to the FDEP impaired WBID list for those water bodies in the study area. 3.4. Critical Parameters Four critical parameters were further evaluated for the estuaries to identify those estuaries of concern: chlorophyll a, dissolved oxygen, transparency (secchi depth), and bacteria. For each of the critical parameters identified, potential areas of concern within the estuaries by water quality station were identified. Water quality stations data were used if the sample size was greater than or equal to 12. Data from each of the water quality stations were categorized based on the 5 Collier County Watershed Model Update and Plan Development July 9, 2010 percent of data values that exceeded the appropriate regulatory standard or screening level. Stations which have data with less than 10 percent of the total samples greater than the regulatory standard were shaded green Data from stations with values that exceed the appropriate regulatory standard or screening level in 10 -49 percent of the total samples were shaded yellow. Stations with values that exceed the appropriate regulatory standard or screening level in 50 percent or more of the total samples were shaded red. Table 3 -2. List of Water Quality Parameters Parameter Unit Parameter Unit Salinity t Conductivity mhos /cm Total Nitrogen m /l Nitrate - Nitrite mg/1 Total Phosphorus mg/1 Orthophosphate mg/1 Total K'eldahl Nitrogen (TKN) m Un- ionized Ammonia mg/1 Chlorophyll a µ 1 Fecal Coliform # /100ml Color PCU Copper /l Total Suspended Solids (TSS) m /l Turbidity NTU Dissolved Oxygen (DO) m /l Biochemical Oxygen Demand (BOD) m /l Iron pgA Hardness m /l Secchi Depth m Table 3 -3. List of regulatory standards for selected water quality parameters Para 6wf Mass Dissolved Oxygen (mg /1) 4 Fecal Coliform ( # /100ml) 43 Chlorophyll a (µg/1) 11 Iron (µg /1) 300 Copper (µg /1) 3.7 Table 3 -4. List of screening levels for selected water quality parameters Parat �a �_ Color (PCU) 40 SD (m) 1.38 TSS (m /1) 17 TN (m /1) 1 TP (mg /1) 0.19 6 Collier County Watershed Model Update and Plan Development July 9, 2010 4.0 Results and Discussion This section presents the results and a discussion of the FDEP impaired WBIDs, a water quality characterization for each estuary, and an evaluation of critical water quality parameters. The water quality characterization results and discussion are discussed separately for Wiggins Pass, Naples Bay, Rookery Bay, and the Ten Thousand Islands estuaries. 4.1. FDEP Impaired WBIDs The impairments identified by FDEP in the estuaries include dissolved oxygen, fecal coliform, iron, nutrients, and copper (Figures 4 -1 to 4- 5). Three (Wiggins Pass, Naples Bay and Rookery Bay) of the four estuarine receiving waters are verified impaired for both dissolved oxygen and fecal coliform bacteria by FDEP. Only the Ten Thousand Islands is presently not listed as impaired for any water quality parameters. TTI is the only estuary in which average dissolved oxygen concentrations have remained above the regulatory standard of 4.0 mg /L for marine waters and fecal coliform concentrations have remained below 43 # /IOOmL for a Class 2 water bodies. Rookery Bay was shown to be the only receiving water to have elevated chlorophyll a concentrations attributed to nutrient loads. Naples Bay is also presently verified impaired for copper ( >3.7 pg /L). Data from Naples Bay and Wiggins Pass estuaries indicate elevated iron concentrations ( >300 pg /L) 7 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 4-1. WBIDS verified impaired for Dissolved Oxygen in the estuarine receiving waters of the study area by FDEP Quality of Receiving Waters L� Dissolved Oxygen ,, W1, �,Girot,p'j- Cycles I and 2 LEE CO. Legend WBID C3 Impaired C3 WBID Boundary CWatershed Boundary ,2 �,uunty twufluaty I - I F---T----1 0 1.5 3 Miles 7,-,(;OLLIER CO. MONROE CO N 8 Collier County Watershed Model Update pw� 74 and Plan Development V July 9, 2010 Figure 4 -2. WBIDS verified impaired for Nutrients in the estuarine receiving waters of the study area by FDEP 1Quality of Recei Ntrients Group l- Cycles 1 and 2 , LEE CO. Legend WBID Impaired C3 WBID Boundary Watershed Boundary County Boundary m 0 2 4 Miles �• iriLY Waters dIM - COLLIER CO. A MONROECO. o b' N 9 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 4 -3. WBIDS verified impaired for Fecal Coliform in the estuarine receiving waters of the study area by FDEP Quality of Receiving Waters ;.Fecal Coliform i \'-Group l- Cycles 1 and 2 .......... __......_ .... _ ..... __� } LEE GO) I:bBA N. 33780 327HR t ]2786 aiasa d COLLIER CO. Legend WBID ! 3 - - -_ MGNRUE CO. Impaired WBID Boundary O Watershed Boundary 0 County Boundary r� N 0 2 4 Miles u ` ff 10 Collier County Watershed Model Update wit and Plan Development July 9, 2010 Figure 4 -4. WBIDS verified impaired for Copper in the estuarine receiving waters of the study area by FDEP Quality of Receivi - Cycles 1 and 2 LEE CO, Legend WBID Impaired C3 WBID Boundary V Watershed Boundary County Boundary f---T --- I 0 2 4 Miles Waters COLLIER CO, M' ONROE CO. 11 Collier County Watershed Model Update " and Plan Development July 9, 2010 Figure 4 -5. WBIDS verified impaired for Iron in the estuarine receiving waters of the study area by FDEP ,Quality of Receivi 1- Cycles 1 and 2 LEE CO. Waters 12 Collier County Watershed Model Update and Plan Development July 9, 2010 dam' A COLLIER CO Legend tll WBID Impaired MONROECO WBID Boundary 0 Watershed Boundary County Boundary 0 2 4 Miles 12 Collier County Watershed Model Update and Plan Development July 9, 2010 4.2. Water Quality of the Estuarine Receiving Waters Summary statistics for each water quality parameter, by station and watershed, were calculated for the four estuarine receiving water bodies. Four critical water quality parameters were identified to evaluate the estuarine water quality condition: • chlorophyll -a • dissolved oxygen • transparency and • Fecal coliform bacteria The maps shows as Figures 4 -6 though 4 -9 show water quality conditions by sampling station for each of the parameters. Secchi depth was used as the measure of transparency in the water body. 4.2.1. Wiggins Pass Estuary Wiggins Pass is located within WBID 3259A (Cocohatchee River) and is presently listed as impaired for three water quality parameters; dissolved oxygen, fecal coliforms and iron. Wiggins Pass is the receiving water for the Cocohatchee- Corkscrew watershed. A total of eighteen water quality stations contain data for the parameters reviewed from 2000 to 2009 (Table 4 -1). Summary statistics by station are available in Appendix B. The water quality summary statistics for Wiggins Pass are presented in Table 4 -2. Table 4 -1. List of stations with water quality data from 2000 to 2009 in Wiggins Pass (WBID 3259A) Name Name 21FLFTM 28030071FTM 28030036 21FLFTM EVRGWC0024FTM BFBSP 21FLFTM EVRGWC0026FTM COCEOF31 21FLFTM EVRGWC0041FTM COCORI 21FLFTM EVRGWC0042FTM COCOR2 21FLFTM EVRGWC0081FTM COCORVW 21 FLSFWMROOK467 Canal @ 99thAve 28030009 Coco @ Collier Reserve TURKBAY Coco at SR 865 13 Collier County Watershed Model Update PMI and Plan Development July 9, 2010 s i Figure 4 -6. Chlorophyll a potential areas of concern by water quality station. 14 Collier County Watershed Model Update and Plan Development July 9, 2010 Figure 4 -7. Dissolved Oxygen potential areas of concern by water quality station. 15 Collier County Watershed Model Update 11W and Plan Development July 9, 2010 Figure 4 -8. Transparency (Secchi Depth) potential areas of concern by water quality station. 16 Collier County Watershed Model Update M"Si and Plan Development July 9, 2010 Figure 4 -9. Bacteria (Fecal Coliform) potential areas of concern by water quality station 17 uouief t oumy VVdlt7IJnCU IVIVUUI VpUQ- and Plan Development July 9, 2010 Table 4 -2. Water quality summary statistics from 2000 to 2009 in Wiggins Pass (WBID 3259A) BOD, m /l 108 1.0 2.2 2.0 8.1 - Chlorophyll-a, u /l 209 1.0 4.5 3.0 70.0 6 Color, PCU 149 5 51 50 200 8 Conductivity, umhos /cm 282 305 28135 33990 57059 - Copper, u /1 75 0.51 1.95 1.65 7.60 4 DO, mg/1 340 0.1 5.1 4.9 19.4 29 Fecal Coliform, # /100ml 260 1 187 62 5700 57 Iron, ugA 72 35 290 239 840 40 Nitrate - Nitrite, m /l 213 0.002 0.508 0.027 99 - Orthophosphate as P, mg/1 122 0.004 0.018 0.012 0.140 Salinity, t 167 0.0 22.3 29.0 66.9 - Secchi Depth, in 268 0.10 0.92 1.00 3.50 45 TKN, m /l 150 0.08 0.74 0.78 2.00 - Total Nitrogen, m /l 181 0.05 0.72 0.74 2.09 1 Total Phosphorus, m /l 210 0.004 0.046 0.036 0.310 0 TSS, m /l 90 2.0 10.1 7.5 62.0 50 Turbidity, NTU 1 210 1 0.1 3.5 2.8 18.1 Unionized Ammonia, mg/1 1 34 1 0.0001 0.0008 0.0008 0.0022 4.2.1.1. Impaired WBID comparison Using all of the water quality data for each WBID, PBS &J confirmed the impairment status determined by FDEP for parameters identified in the Wiggins Pass Estuary (WBID 3259A) (Table 4 -3). Table 4 -3. Impaired WBID comparison for Wiggins Pass estuary 4.2.1.2. Chlorophyll a Overall, chlorophyll a concentrations exceeded 11 pg/L in 6 percent of the samples collected. The median chlorophyll a concentration was 3 pg/L with the maximum measuring 70 µg /L. Based on these values, phytoplankton concentrations are within the allowable range for a marine water body. Low TN and TP loads from the Cocohatchee- Corkscrew watershed were predicted 18 Collier County Watershed Model Update PW and Plan Development July 9, 2010 3259A Cocohatchee River Dissolved Oxygen Confirm FDEP assessment 3259A Cocohatchee River Fecal Coliform Confirm FDEP assessment 3259A Cocohatchee River Iron Confirm FDEP assessment 4.2.1.2. Chlorophyll a Overall, chlorophyll a concentrations exceeded 11 pg/L in 6 percent of the samples collected. The median chlorophyll a concentration was 3 pg/L with the maximum measuring 70 µg /L. Based on these values, phytoplankton concentrations are within the allowable range for a marine water body. Low TN and TP loads from the Cocohatchee- Corkscrew watershed were predicted 18 Collier County Watershed Model Update PW and Plan Development July 9, 2010 based on the concentrations observed along the downstream portion of the watershed (see TM 3.1: Quality of Discharge). As such, phytoplankton production has remained consistently below the regulatory standard throughout the estuary (Figure 4 -6). 4.2.1.3. Dissolved Oxygen Dissolved oxygen values were less than 4.0 mg/L in 29 percent of the samples. The median DO concentration was 4.9 mg/L with a minimum value of 0.1 mg/L. Low dissolved oxygen levels appear to be evident in the upstream portions of Wiggins Pass, where stations were below 4 mg/L in more than 50 percent of samples reviewed. Wiggins Pass is impaired for low dissolved oxygen values by FDEP (Table 4 -3). Depressed dissolved oxygen continued mid - estuary but less consistently with stations reporting values below the standard in 10 -49 percent of samples (Figure 4 -7). Those stations with the most interaction with the Gulf of Mexico consistently met regulatory standards for dissolved oxygen. Depressed dissolved oxygen values could be attributed to decomposition of organic material from the upstream landscape. However, elevated color and high total suspended solid concentrations were not observed in the downstream portion of the watershed (see TM 3.1: Quality of Discharge). Therefore, further analyses are necessary to identify the cause of low dissolved oxygen concentrations. 4.2.1.4. Transparency (Secchi Depth) Forty -five percent of the secchi depth measurements were below the calculated screening level of 1.38 in indicating low visibility. The median secchi depth was 1.00 in with a minimum of 0.10 m. No water quality station had consistent secchi depth values greater than the screening level (Figure 4 -8). Low secchi depth values indicate poor light penetration which could lead to degradation of seagrass communities and other photosynthetic biota. It is possible that secchi depth values are low due to the limited flushing of the estuary and resuspension of bottom material. 4.2.1.5. Fecal Coliform Bacteria Fecal coliform concentrations are elevated in Wiggins Pass with values exceeding 43 # /l00mL in 57 percent of the samples. The median bacteria concentration was 62 # /100mL and a maximum value of 5,700 # /100mL. Wiggins Pass is impaired by FDEP for elevated fecal coliform levels (Table 4 -3). The water quality stations within the upper estuary exceeded 43 # /100mL in > =50 percent of all samples reviewed (Figure 4 -9). The frequency of elevated values decreases with proximity to the Gulf of Mexico. The decrease in frequency of elevated bacterial loads is possibly due to the high inactivation ("die-off') rate of both fecal coliforms and E. coli in saline waters (Anderson et al. 2005). Though values exceed the regulatory standard for Class 2 water bodies, fecal coliform bacteria may not be an appropriate indicator for pathogenic diseases in sub - tropical environments such as South Florida where the specificity of the fecal coliform test is compromised by the more constant and warmer ambient water temperatures of sampled water bodies. The inability to specifically identify humans as a source of bacteria using traditional indicator bacteria testing 19 Collier County Watershed Model Update PW and Plan Development July 9, 2010 protocols has been noted by Fujioka (2001) and Fujioka et al. (1999) for various tropical locations. Wiggins Pass is classified for oyster propagation and harvesting; the monitoring of bacterial loads is important for public health concerns. These data and results mostly support the conclusions reached by FDEP, i.e. that dissolved oxygen values in the Wiggins Pass may not meet state criteria. However, levels of TN and TP within Wiggins Pass exceeded screening criteria in only 1 percent or less of the samples; it is possible that that low levels of dissolved oxygen are due to factors other than nutrient enrichment alone. In addition, impairment for fecal coliform bacteria does not necessarily mean that there is an anthropogenic cause; further source identification efforts would be warranted to determine whether anthropogenic factors are the cause of the elevated bacteria concentrations. 4.2.2. Naples Bay Estuary Naples Bay is located within WBID 3278R (Naples Bay - Coastal Segment) and is presently listed as impaired for four parameters; dissolved oxygen, fecal coliforms, copper and iron. Naples Bay is the receiving water for the Golden Gate- Naples Bay watershed. A total of forty water quality stations are available which contain data for the parameters reviewed for the period 2000 to 2009 (Table 4 -4). Summary statistics by station are available in Appendix B. The water quality summary statistics for Naples Bay are presented in Table 4 -5. Table 4 -4. List of stations with water quality data from 2000 to 2009 in Naples Bay (WBID 3278R) Name Name Name 21FLBRA 3259G -B 21FLSFWMBC4 21FLNAPLNBAYBV 21FLBRA 3259G -C AQS8 -1 21FLNAPLNBAYCC 21FLBRA 3259G -D BC2 21FLNAPLNBAYLLO 21FLBRA 3259G -E Ba 20 21FLNAPLNBAYNL 21FLFMRINTK200120 COL8 21FLSFWMBCI 21FLFTM 28030069FTM COLS ROOK464 21FLFTM28030031 ESBAY HaldemanBa 21FLGW14160 GORDIO Ja ceePark 21FLGW21751 GORD30 Na IesBa 22 21FLGW22543 GORD31 Na lesBa 24 21FLNAPLGORDJOE GORD70 Na lesBa 41 21FLNAPLGORDPK Gord60 Na 1esBa 50 21FLNAPLNBAY13 Gord80 HC @Ba shore 21FLNAPLNBAY29 20 Collier County Watershed Model Update PV�f and Plan Development July 9, 2010 Table 4 -5. Water quality summary statistics from 2000 to 2009 in Naples Bay (WBID 3278R) BOD, m /l 500 0.3 2.1 2.0 li 12.0 1 Chlorophyll-a, u /1 842 0.9 6.6 3.7 110.0 14 Color, PCU 719 5 45 40 200 41 Conductivity, umhos /cm 729 449 34163 42751 57220 - Copper, ugfl 513 0.15 3.43 2.90 25.30 30 DO, m A 714 0.6 5.6 5.7 14.0 16 Fecal Coliform, # /100ml 682 1 148 29 4700 43 Iron, u /1 306 29 419 390 2530 65 Nitrate - Nitrite, mgfl 712 0.00 0.05 0.04 0.26 - Orthophosphate as P, m /l 596 0,004 0.021 0.018 0.081 Salinity, t 660 0.2 22.7 27.9 38.2 - Secchi Depth, m 746 0.15 1.13 1.10 3.90 78 TKN, m /I 683 0.04 0.60 0.60 5.90 - Total Nitrogen, m /l 766 0.01 0.63 0.63 17.00 10 Total Phosphorus, mg/1 823 0.004 0.046 0.040 0.310 0 TSS, mg/1 577 2.0 10.4 6.0 270,0 14 Turbidity, NTU 618 0.1 2.7 2.1 63.0 - Unionized Ammonia, mg/1 0 - - - - 4.2.2.1. Impaired WBID comparison Using all of the water quality data for each WBID, PBS &J confirmed the impairment status determined by FDEP for parameters identified in the Naples Bay Estuary (WBID 3278R) (Table 4 -6). Additionally, PBS &J identified Naples Bay Estuary as potentially impaired for chlorophyll a. Table 4 -6. Impaired WBID comparison for Naples Bay estuary 3278R Naples Bay Coastal Copper Confirm FDEP assessment 3278R Naples Bay Coastal Dissolved Oxygen Confirm FDEP assessment 3278R Naples Bay Coastal Fecal Coliform Confirm FDEP assessment 3278R Naples Bay Coastal Iron Confirm FDEP assessment 3278R Naples Bay Coastal Nutrients (Chlorophyll a) New 4.2.2.2. Chlorophyll a Overall, chlorophyll a concentrations exceeded 11 pg/L in 14 percent of the samples collected. The median chlorophyll a concentration was 3.7 pg/L, with the maximum measured value equal 21 Collier County Watershed Model Update M�l and Plan Development July 9, 2010 to 110 µg/L. The chlorophyll a data do not show evidence of excess phytoplankton production in Naples Bay (Figure 4 -6). Low nutrient loading was predicted from the Golden Gate- Naples Bay watershed but not evident in the elevated chlorophyll a concentrations observed in the estuary (see TM 3.1: Quality of Discharge). 4.2.2.3. Dissolved Oxygen Naples Bay is impaired for low dissolved oxygen values by FDEP (Table 4- 6).Dissolved oxygen concentrations in the PBS &J data set were less than 4.0 mg /L in 16 percent of the samples. The median DO concentration was 5.7 mg/L with a minimum value of 0.6 mg /L. Low dissolved oxygen levels appear to be evident in the upstream portions Naples Bay where stations were below 4 mg /L in more than 50 percent of samples reviewed. Depressed dissolved oxygen levels continued mid - estuary but less consistently with stations reporting values below the standard in 10 -49 percent of samples (Figure 4 -7). Those stations with the most interaction with the Gulf of Mexico consistently met regulatory standards for dissolved oxygen. Elevated total suspended solids loads were predicted from the Golden Gate- Naples Bay watershed, which could lead to depressed dissolved oxygen concentrations in the estuary if sufficient organic material is available for decomposition. 4.2.2.4. Transparency (Secchi Depth) Seventy -eight percent of the secchi depth measurements were below the calculated screening level of 1.38 m. The median secchi depth was 1.10 m with a minimum of 0.15 m. No water quality station had consistent secchi depth values greater than the screening level (Figure 4 -8). Transparency appears to be the lowest in the Naples Bay estuary. As previously mentioned, total suspended solid loads are likely a concern for the Naples Bay estuary, which may result in reduced water clarity and decreased secchi depth values. 4.2.2.5. Fecal Coliform Bacteria Fecal coliform concentrations exceeded 43 # /100mL in 53 percent of the samples. The median bacteria concentration was 29 # /100mL and the maximum value was 4,700 # /100mL. Naples Bay estuary was declared impaired by FDEP for elevated fecal coliform concentrations. Water quality stations within this estuary exceeded 43 # /100mL in > =50 percent of all samples reviewed (Figure 4 -9). The majority of consistent exceedances occurred in the upper portion of each estuary. The frequency of elevated values decreases with proximity to the Gulf of Mexico. The decrease in frequency of elevated bacteria loads is possibly due to the high inactivation ( "die -off') rate of both fecal coliforms and E. coli in saline waters (Anderson et al. 2005). None of the WBIDs discharging into Naples Bay have been declared impaired for fecal coliforms. However, bacterial loads from the watershed would provide a source of contamination to the estuary. As in Wiggins Pass, these data mostly support the conclusions reached by FDEP, in that water quality within Naples Bay show levels of both chlorophyll a and dissolved oxygen that do not meet State criteria. However, levels of TP within Naples Bay do not exceed screening criteria, 22 Collier County Watershed Model Update and Plan Development July 9, 2010 and levels of TN exceed screening criteria only 10 percent of the time. As in Wiggins Pass, it is possible that that elevated levels of chlorophyll a and low levels of dissolved oxygen are due to factors other than nutrient enrichment alone. In addition, impairment for fecal coliform bacteria may not necessarily mean that there is an anthropogenic impact ( Fujioka 2001, Fujioka et al. 1999); further source identification efforts are warranted. 4.2.3. Rookery Bay Estuary Rookery Bay is located within WBID 3278U (Rookery Bay - Coastal Segment) and is presently listed as impaired for three parameters; dissolved oxygen, fecal coliforms, and nutrients. Rookery Bay is the receiving water for the Rookery Bay watershed. A total of thirty -nine water quality stations contain data for the parameters reviewed for the sampling period 2000 to 2009 (Table 4 -7). Summary statistics by station are available in Appendix B. The water quality summary statistics for Rookery Bay are presented in Table 4 -8. Table 4 -7. List of stations with water quality data from 2000 to 2009 in Rookery Bay (WBID 3278U) Name Name Name' 21FLFMRINTK200121 21FLFTM EVRGWC0061FTM HendersonCreek 21FLFMRINTK200122 21FLFTM EVRGWC0062FTM HendersonCrk @41 21FLFMRINTK200123 21FLFTM EVRGWC0063FTM JohnsonBa 1 21FLFMRINTK200124 21FLGWI3733 JohnsonBa 2 21FLFMRINTK200129 21FLGW15163 JohnsonBa 3 21FLFTM EVRGWC0027FTM 21FLSFWMHALDCRK NTK200125 21FLFTM EVRGWC0028FTM 21FLSFWMROOK461 NTK200126 21FLFTM EVRGWC0029FTM 21FLSFWMROOK462 NTK200130 21FLFTM EVRGWC0030FTM 21FLSFWMROOK463 PORTAUPR5 21FLFTM EVRGWC0031FTM Bi MarcoRiver ROOK458 21FLFTM EVRGWC0059FTM COL10 ROOK459 21FLFTM EVRGWC0060FTM DollarBa 15 ROOK460 TarponBayl UH TarponBay 23 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Table 4 -8. Water quality summary statistics from 2000 to 2009 in Rookery Bay (WBID 3278U) Parameter N Min Mean 11 ecliaa Max Percent Exceed BOD, m /l 66 0.8 2.2 2.0 5.8 - Chlorophyll-a, u /l 691 0.8 5.9 4.3 74.0 10 Color, PCU 192 15 71 60 277 83 Conductivity, umhos /cm 521 125 29777 1 36345 60964 - Copper, u 84 0.25 5.94 1.71 51.00 38 DO, m /l 771 0.6 4.9 4.8 20.6 31 Fecal Coliform, # /100ml 166 1 136 80 1143 62 Iron, u 79 16 340 240 1440 43 Nitrate - Nitrite, m /l 453 0.0003 0.0170 0.0094 0.1440 - Orthophosphate as P, m /l 315 0.002 0.012 0.008 0.126 - Salinity, ppt 779 0.1 24.3 27.7 41.4 - Secchi Depth, m 267 0.15 1.06 1.04 2.59 82 TKN, m /l 167 0.19 0.83 0.75 2.90 - Total Nitrogen, m 418 0.01 0.50 0.39 2.91 11 Total Phosphorus, m /l 542 0.002 0.043 0.038 0.206 0 TSS, m /l 127 2.0 5.5 2.0 70.0 6 Turbidity, NTU 670 -1.0 5.1 4.1 70.5 - Unionized Ammonia, mg 0 - - - - - 4.2.3.1. Impaired WBID Comparison Using water quality data for each WBID, PBS &J confirmed the impairment status determined by FDEP for two parameters identified in the Rookery Bay Estuary (WBID 3278U), dissolved oxygen and fecal coliforms (Table 4 -9). The evaluation of chlorophyll a data indicated that values were not elevated frequently enough to classify the water body as impaired. The discrepancy in impairment classification could be due to the modified technique used to evaluate chlorophyll a, the data used or time period examined. PBS &J also identified Rookery Bay estuary as potentially impaired for copper and iron. Table 4 -9. Impaired WBID comparison for Rookery Bay estuary 24 Collier County Watershed Model Update PW and Plan Development July 9, 2010 Water Segment l�Tatae FDEP d Padlner PBSJ Analysis 3278U Rookery Bay Coastal Dissolved Oxygen Confirm FDEP assessment 3278U Rookery Bay Coastal Fecal Coliform Confirm FDEP assessment 3278U Rookery Bay Coastal Nutrients (Chlorophyll a) Not confirm 3278U Rookery Bay Coastal Copper New 3278U Rookery Bay Coastal Iron New 24 Collier County Watershed Model Update PW and Plan Development July 9, 2010 4.2.3.2. Chlorophyll a Overall, chlorophyll a concentrations exceeded 11 pg/L in 10 percent of the samples collected. The median chlorophyll a concentration was 4.3 pg/L, with a maximum reported value of 74 pg /L. The majority of the data from water quality stations exceeded the regulatory standard in less than 10 percent of the samples (Figure 4 -6). Three stations in Rookery Bay exceeded the 11 pg/L standard in 10 -49 percent of all samples over the time period reviewed. The Rookery Bay WBID was designated impaired for nutrients based on elevated chlorophyll a values by FDEP. The persistence of elevated chlorophyll a levels indicates the potential for high nutrient loading from the tributaries to Rookery Bay, yet elevated concentrations in runoff were not observed, as discussed in the Phase 2 Technical Memorandum for Element 2, Task 2. 4.2.3.3. Dissolved Oxygen As indicated previously, the estuary was declared impaired for low dissolved oxygen values by FDEP. The PBS &J data set showed that dissolved oxygen concentrations were less than 4.0 mg/L in 31 percent of the samples. The median DO concentration was 4.8 mg /L, with a minimum reported value of 0.6 mg /L. Low dissolved oxygen levels appear to be evident in the upstream portions of Rookery Bay where stations were below 4 mg/L in more than 50 percent of samples reviewed (Figure 4 -7). Depressed dissolved oxygen continued in the northern section of the estuary but less consistently with stations reporting values below the standard in 10 -49 percent of samples. Those stations with the most interaction with the Gulf of Mexico consistently met regulatory standards for dissolved oxygen. Elevated chlorophyll a concentrations could be responsible for the fluctuations in dissolved oxygen in the estuary. 4.2.3.4. Transparency (Secchi Depth) Eighty -two percent of the secchi depth measurements were below the calculated screening level of 1.38 m. The median secchi depth was 1.04 m with a minimum of 0.15 m. Fecal coliform concentrations exceeded 43 # /100mL in 62 percent of the samples. No water quality station had consistent secchi depth values greater than the screening level (Figure 4 -8). It is possible that secchi depth values were low due to the low flushing characteristics of the estuary and resuspension of bottom material. 4.2.3.5. Fecal Coliform Bacteria Rookery Bay is designated impaired by FDEP for elevated fecal coliform levels. For this analysis, the median bacteria concentration was 80 # /100mL, with a maximum value of 1,143 # /100mL. Sixty -two percent of all samples were greater than the regulatory standard of 43 # /100mL for Class II water bodies. Both water quality stations examined within this estuary exceeded 43 # /100mL in > =50 percent of all samples reviewed (Figure 4 -9). While low bacterial loads would be expected from the watershed (see TM 3.1: Quality of Discharge), the more stringent regulatory standards applied to Class II water bodies may result in the impairment condition. 25 Collier County Watershed Model Update Imsi and Plan Development July 9, 2010 These data mostly support the conclusions reached by FDEP, in that water quality within Rookery Bay can exceed guidance levels for chlorophyll a and quite often are below state criteria for dissolved oxygen. However, levels of TP within Rookery Bay never exceeded screening criteria levels, and TN concentrations exceeded screening criteria levels only 11 percent of the time. As in Wiggins Pass and Naples Bay, it is possible that that elevated levels of chlorophyll a and low levels of dissolved oxygen in Rookery Bay are due to factors other than nutrient enrichment alone. In addition, impairment for fecal coliform bacteria may not necessarily mean that it is caused by an anthropogenic impact ( Fujioka 2001, Fujioka et al. 1999). 4.2.4. Ten Thousand Islands Estuary Ten Thousand Islands is located within WBID 3259M (Ten Thousand Islands) and is presently not listed as impaired. The Ten Thousand Islands is the receiving water for the Faka- Union, Fakahatchee, and Okaloacochee /SR29 watersheds. A total of sixty -three water quality stations are available for the parameters reviewed from 2000 to 2009 (Table 4 -10). Summary statistics by station are available in Appendix B. The water quality summary statistics for TTI are presented in Table 4 -11. Table 4 -10. List of stations with water quality data from 2000 to 2009 in Ten Thousand Islands (WBID 3259M) Name Name Name 187 Fakahatchee 21FLSFWMTTI53 SEAS007_Fer uson 21FLA 66011SEAS 21FLSFWMTTI65 SEAS OIO_IndianKe 21FLA 66038SEAS 21FLSFWMTT167 SEAS028_Turtle 21FLFMRISTK200201 21FLSFWMTTI68 SEAS029_Sna Shoal 21FLFMRISTK200205 21FLSFWMTTI69 SEAS034_DismalKe 21FLFMRISTK200208 21FLSFWMTTI70 SEAS035_SantinaBa 21FLFMRISTK200210 21FLSFWMTTI72 SEAS036_Pum kin 21FLFMRISTK200211 21FLSFWMTTI74 SEAS037_Santina 21FLFMRISTK200212 21FLSFWMTTI75 SEAS111_Fakahatchee 21FLFMRISTK200214 21FLSFWMTTI76 SEAS 112 Fakahatchee 21FLFMRISTK200216 BARRIVN SEAS 113 Fakahatchee 21FLFTM EVRGWCOOO I FTM BL_Kwater SEAS 114 Fakahatchee 21FLFTM EVRGWC0002FTM BRMouth SEAS281_FishHawk 21FLFTM EVRGWC0003FTM Brid e030122 SEAS299_Blackwater 21FLFTM EVRGWC0004FTM COL14 SEAS300_Blackwater 21FLGW 13734 COL15 SEAS301_ShellKe 21FLGWI5173 COL16 SEAS 302_Sna Shoal 21FLSFWMROOK451 FAKAUPOI SEAS303_Buttonwood 21FLSFWMTT151 FU SEAS401_FakaUnion Fa- Aunion STK200206 SEAS771_FakaUnion FakahatcheeBay PumpkinBay Seas077 26 Collier County Watershed Model Update 1TS§ and Plan Development July 9, 2010 Table 4 -11. Water quality summary statistics from 2000 to 2009 in the Ten Thousand Islands (WBID 3259M) Parameter N Min Mean Median Max Exceed BOD, m /l 52 0.6 1.9 2.0 8.3 - Chlorophyll-a, u A 1113 0.5 4.0 3.0 47.5 3 Color, PCU 167 10 60 50 200 68 Conductivity, umhos /cm 9150 306 43007 1 48013 64190 - Copper, u /1 63 0.26 1.33 1.00 5.13 5 DO, m /I 7593 0.2 5.0 5.0 20.3 24 Fecal Coliform, # /100ml 431 1 56 1 2300 17 Iron, u 66 65 233 180 980 15 Nitrate - Nitrite, m 918 0.0005 0.0162 0.0108 0.1100 - Orthophosphate as P, m 319 0.002 0.012 0.009 0.054 - Salinity, t 10132 0.2 28.1 31.4 43.4 - Secchi Depth, m 190 0.20 1.52 1.55 2.90 41 TKN, m /l 152 0.04 0.63 0.58 2.26 - Total Nitrogen, mg/1 769 0.01 0.43 0.38 2.27 5 Total Phosphorus, mg/1 926 0.001 0.144 0.033 99 0 TSS, m /l 130 2.0 6.9 2.0 113.0 8 Turbidity, NTU 9735 0.3 9.4 8.0 249.0 - Unionized Ammonia, m /l 0 - - - 4.2.4.1. Impaired WBID comparison No WBIDs were declared impaired by FDEP in the Ten Thousand Island estuary (Table 4 -12). However, PBS &J identified two parameters of potential impairment, dissolved oxygen and fecal coliforms. The discrepancy in impairment condition could be due to the data base used or time period examined. Table 4 -12. Impaired WBID comparison for Ten Thousand Island estuary WBID# Water Segment name FDkP Impaired, Parameter PB3,I Analysis 3259M Ten Thousand Islands Dissolved Oxygen New 3259M Ten Thousand Islands Fecal Coliform New 4.2.4.2. Chlorophyll a Overall, chlorophyll a concentrations exceeded 11 pg/L in 3 percent of the samples collected. The median chlorophyll a concentration was 3.0 µg/L with a maximum reported value of 47.5 pg/L. Only one station in the eastern portion of the estuary near the Tamiami trail indicated values in exceedance of the standard in 10 to 49 percent of the samples (Figure 4 -6). Low TP loads from the discharging watersheds were predicted based on the concentrations observed (see TM 3.1: Quality of Discharge). However, elevated TN loads were expected from the Faka 27 Collier County Watershed Model Update fv!�; and Plan Development July 9, 2010 Union and Fakahatchee watersheds. Regardless of nutrient loads to the estuary, phytoplankton production has remained consistently below the regulatory standard throughout the estuary. Freshwater discharge with elevated color concentrations from the forested watersheds is likely contributing to the suppressed phytoplankton values. 4.2.4.3. Dissolved Oxygen Dissolved oxygen values were less than 4.0 mg /L in 24 percent of the samples. The median DO concentration was 5.0 mg /L with a minimum value of 0.2 mg/L. Low dissolved oxygen levels appear to be evident in throughout the mid - estuary with stations reporting values below the standard in 10 -49 percent of samples (Figure 4 -7). While FDEP has not declared the WBID impaired for dissolved oxygen, the analysis completed by PBS &J indicates that during the time period examined dissolved oxygen is a parameter of concern for the estuary. The elevated color and total suspended solids discharged from the contributing watersheds are likely responsible for the depressed dissolved oxygen values. However, as discussed in technical memorandum 1.2: In- stream water quality, low dissolved oxygen concentrations in this region are likely due to natural conditions associated with the extensive forested wetlands found in the adjacent watershed. 4.2.4.4. Transparency (Secchi Depth) Forty -one percent of the secchi depth measurements were below the calculated screening level of 1.38 m. The median secchi depth was 1.55 m with a minimum of 0.2 m. No water quality station had consistent secchi depth values greater than the screening level. Limited data were available for the TTI, those stations with sufficient data had values less than 1.38 m in 10 -49 percent of all samples reviewed (Figure 4 -8). Total suspended solid and color loads from the adjacent watersheds contribute to the reduced water clarity in the Ten Thousand Islands estuary. It is also likely that secchi depth values were low due to the flushing of the estuary and resuspension of bottom material. 4.2.4.5. Fecal Coliform Bacteria Fecal coliform concentrations exceeded 43 # /100mL in 17 percent of the samples. Fecal coliform values for each station were compared to the regulatory standard of 43 # /100mL for Class II water bodies. The median bacteria concentration was 1 # /100mL with a maximum value of 2,300 # /100mL. While FDEP has not declared the WBID impaired for fecal coliforms, the analysis completed by PBS &J indicates that during the time period examined bacteria is a parameter of concern for the estuary. One water quality station in the eastern portion of the estuary exceeded 43 # /100mL in > =50 percent of all samples reviewed (Figure 4 -9). The bacteria concentrations of the water quality stations upstream of this area of concern were low compared to the Class 3 regulatory standards for fecal coliforms. The remainders of the water quality stations were consistently below the regulatory standard. Additional water quality sampling to identify the potential bacteria sources to the estuary is recommended. PE6f 28 Collier County Watershed Model Update and Plan Development July 9, 2010 As found by FDEP, water quality within the Ten Thousands Islands estuary exceeds chlorophyll a criteria very infrequently, and impairment is not demonstrated. In contrast, levels of dissolved oxygen fall below state criteria approximately 24 percent of the time. However, levels of TP within the Ten Thousand Islands never exceeded screening criteria levels, and TN concentrations exceeded screening criteria levels only 5 percent of the time. It is likely that low levels of dissolved oxygen are due to factors other than nutrient enrichment. In addition, impairment for fecal coliform bacteria may not necessarily mean that there is an anthropogenic impact as the cause ( Fujioka 2001, Fujioka et al. 1999). 5.0 Conclusions Water quality impairments identified by FDEP were generally confirmed by results of analyses completed for the estuaries water quality evaluation. Only one discrepancy was identified in a comparison of the FDEP and PBS &J derived impaired WBIDs. FDEP verified Rookery Bay estuary (WBID 3278U) as impaired for chlorophyll a; however, PBS &J did not conclude the same result. The discrepancy is likely due to the modified method utilized by PBS &J as well as the data set and time period analyzed. A more extensive review of this impairment is recommended. Five additional potential impairments were identified for the estuaries examined. Recommendations were developed based on the results of this evaluation to address dissolved oxygen and chlorophyll a impairments and source identification for fecal coliforms in some cases. It is recommended that Collier County work with FDEP to develop site specific alternative criteria for Wiggins Pass, Naples Bay and Rookery Bay estuaries for dissolved oxygen and chlorophyll a. While these three estuaries have been declared verified impaired by FDEP for dissolved oxygen, it is unlikely that naturally occurring levels of dissolved oxygen from reference sites would pass the state's dissolved oxygen standard, as is more fully discussed in the Technical Memorandum Element 1, Task 1.2: In- Stream Water Quality. In addition, levels of the potentially causative nutrients TN and TP are only infrequently above relevant screening criteria; it is likely that factors other than nutrient enrichment alone influence concentrations of chlorophyll a and levels of dissolved oxygen. Site specific alternative criteria might also be useful for both chlorophyll a and dissolved oxygen as levels of both these parameters often do not meet state criteria without a concurrent link to an anthropogenic causation. Additionally, Collier County should work with FDEP to develop a directed sampling effort focusing on identifying potential sources (including non - anthropogenic ones) of fecal coliform bacteria in Wiggins Pass, Naples Bay, and Rookery Bay, perhaps as part of FDEP's TMDL and /or Basin Management Action Plan (BMAP) programs. Further assessments are needed to determine whether or not levels of iron and copper are indicative of anthropogenic influences in Collier County's estuaries; iron in particular may reflect groundwater sources rather than contamination. 29 Collier County Watershed Model Update HIS) and Plan Development July 9, 2010 6.0 References Anderson, K.L. Whitlock, J.E., Harwood, V.J. 2005. Persistence and differential survival of fecal indicator bacteria in subtropical waters and sediments. Applied and Environmental Microbiology 71:3041 -3048 Black, Crow, and Eidsness, Inc. 1974. Hydrologic Study of the G. A. C. Canal Network. Gainesville, FL. Project no. 449- 73 -53. Browder, J.A., Tashiro, J., Coleman - Duffie, E., and A. Rosenthal. 1988. Comparison of Ichthyoplankton Immigration Rates into Three Bay Systems of the Ten Thousand Islands Affected by the Golden Gate Estates Canal System. Volume I. Final Report to the South Florida Water Management District. Davis, S.M. and Ogden, J.C. 1994. Everglades: The Ecosystem and Its Restoration. St. Lucie Press: Delray Beach, Florida. 826 p. Friedemann, M., and J. Hand, 1989. Typical water quality values for Florida's lakes, streams and estuaries: Florida Department of Environmental Regulation, Tallahassee, 31 p. Fujioka, R.S. 2001. Monitoring coastal marine waters for spore - forming bacteria of faecal and soil origin to determine point from non -point source pollution. Water Science and Technology. 44: 181 -188. Fujioka, R.S., Stan - Denton, C., Borja, M., Castro, J., and K. Morphew. 1999. Soil, the environmental source of Escherichia coli and enterococci in Guam's streams. Journal of Applied Microbiology. (Symposium supplement) 85: 83S -89S. Klein, H., W.J. Schneider, B.F. McPherson and T.J. Buchanan. May 1970. Some Hydrologic and Biologic Aspects of the Big Cypress Swamp Drainage Area, Southern Florida. United States Geologic Survey Open -file Report 70003. Locker, S.D. 2005. Establishing Baseline Benthic Habitat Coverages in Faka Union and Fakahatchee Bays for Present and Future Environmental Studies. Final Report to South Florida Water Management District. Contract No. DG040614. 60 pp. SFWMD. 2007. Naples Bay: Surface Water Improvement and Management Plan. South Florida Water Management District. 47 p. Shirley, M., O'Donnell, P., McGee, V., and T. Jones, 2005. Nekton species composition as a biological indicator of altered freshwater inflow into estuaries. Pp. 351 -364. In: S.A. Bortone (ed.). Estuarine Indicators. CRC Press, Boca Raton, FL, 30 Collier County Watershed Model Update and Plan Development July 9, 2010 Appendix A Water Quality Station List Collier County Watershed Model Update ms) and Plan Development July 9, 2010 Appendix B Receiving Water Quality Summary Statistics by Station 32 Collier County Watershed Model Update and Plan Development July 9, 2010 W` a To: Mac Hatcher, PM Collier County From: David Tomasko, PBS &J Technical Memorandum Date: 8/24/09 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318, Element 4 - Task 1 — Review of IWR data 1.0 Introduction Throughout the state of Florida, every five years each water body (WBID) is evaluated to determine if available water quality parameters exceed the limits defined by the Florida Department of Environmental Protection (FDEP). The Impaired Water Rule (Chapter 62- 303, F.A.C.) provides the methods used by the FDEP for evaluation of the available water quality data. The limits are implemented to assist in the identification of impaired water bodies and the degradation of water resources. All water bodies within Collier County are located within the Everglades West Coast Group 1 Basin at the FDEP TMDL program implementation. To assess impairment conditions, FDEP evaluated each parameter within a WBID over the planning period (January 1995 - December 2004), to determine with 80 percent confidence if a parameter exceeds its established threshold. Based upon this determination, identified parameters and WBIDs were next evaluated over the verified period (January 2000 — June 30, 2007). During that analysis, parameters that exceed the threshold with at least a 90% confidence of impairment were classified as "verified impaired" and would then be required to have a Total Maximum Daily Load (TMDL) developed. The planning period was January 1995- December 2004 The verified period is January 2000 -June 30, 2007 The May 2009 Cycle 2 Verified list contained fourteen WBIDs with a total of twenty -four impairments. The WBIDs of concern (Groups 1 Cycle 2 Verified List) identified by FDEP are shown in Table 1. Final TMDL reports have been developed and adopted by FDEP for fecal coliform bacteria in the Cocohatchee River Estuary (WBID 3259A), for dissolved oxygen in the Gordon River extension (WBID 3278K) and for nutrients, dissolved oxygen and un- ionized ammonia in Lake Trafford (WBID 3259W). PBS &J completed a review of available data within the Impaired Water Rules (IWR) database to validate the impairment classifications. More extensive reviews of WBID 3278K (Gordon River Extension) and WBID 3259W (Lake Trafford) were previously completed for Collier County by PBS &J and technical memorandums were submitted as separate documents. Those water bodies were not included in the current analysis. 1 Collier County vvatersnea iwoaei upudia PWand Plan Development Table 1 Collier County WBID and Parameters on Group 1 Cycle 2 Verified List Yi►BII(� 11Ma1lr 'YYatelr�Sed � � SOUTHWEST COAST GULF 5 COASTAL 3M O pir��b�e _. Bacteria (in Shellfish) 8065 3259A COCO - HATCHEE RIVER ESTUARY 2 Fecal Coliform 3259A COCO - HATCHEE RIVER ESTUARY 2 Iron 3259W LAKE TRAFFORD LAKE 3F Dissolved Oxygen 3259W LAKE TRAFFORD LAKE 3F Nutrients (TSI) 3259W LAKE TRAFFORD LAKE 3F Un- ionized Ammonia 3261C BARRON RIVER CANAL STREAM 3F Iron 3278D COCO - HATCHEE INLAND STREAM 3F Dissolved Oxygen 3278F CORKSCREW MARSH STREAM 3F Dissolved Oxygen 3278G JEF AK A-HATCHEE STRAND STREAM 3F Dissolved Oxygen 3278G FAKA - HATCHEE STRAND STREAM 3F Fecal Coliform 3278K GORDON RIVER EXTENSION STREAM 3F Dissolved Oxygen 3278L IMMOKALEE BASIN STREAM 3F Dissolved Oxygen 3278R NAPLES BAY COASTAL ESTUARY 2 Copper 3278R NAPLES BAY COASTAL ESTUARY 2 Dissolved Oxygen 3278R NAPLES BAY COASTAL ESTUARY 2 Fecal Coliform 3278R NAPLES BAY COASTAL ESTUARY 2 Iron 3278S NORTH GOLDEN GATE STREAM 3F Dissolved Oxygen 3278S NORTH GOLDEN GATE STREAM 3F Iron 3278T OKALOA - COOCHEE STREAM 3F Dissolved Oxygen 3278U ROOKERY BAY COASTAL ESTUARY 2 Dissolved Oxygen 3278U ROOKERY BAY COASTAL ESTUARY 2 Fecal Coliform 3278U ROOKERY BAY COASTAL ESTUARY 2 Nutrients (Chlorophyll -a) 3278W SILVER STRAND STREAM 3F Dissolved Oxygen 2 - Shellfish propagation or harvesting 3F - Recreation, propagation, and maintenance of a healthy, well - balanced population of fish and wildlife in fresh water 3M - Recreation, propagation, and maintenance of a healthy, well - balanced population of fish and wildlife in marine water 2 Collier county vvatersnea iwvaei �N�a« �������� and Plan Development 2.0 Method for IWR Query and Data Summarization for Impaired WBIDs PBSJ reviewed the data retrieved from the Impaired Water Rules (IWR) database to evaluate each WBID and parameter. The criterion for evaluation was obtained from Chapter 62 -303. Run 36 of the IWR SAS dataset (iwr200 _pv.sas7bdat) was queried for the impaired WBIDS within Collier County. This dataset was then further queried for parameters of interest (i.e. reasons for impairment; Dissolved oxygen, Fecal coliform, Iron, TN, TP, Chlorophyll a- uncorrected , Chlorophyll a- corrected, Un- ionized Ammonia, Copper, Color, Conductivity, Salinity, BOD). The data was then classified in terms of whether the sampling date occurred during the Planning Period (January 1995- December 2004) and /or the Verified Period (January 2000 -June 2007). Following is a description of the findings by individual impairment parameters. 2.1 Dissolved Oxygen The criterion for dissolved oxygen is dependent upon the salinity regime of the water body. The Florida's Surface Water Quality Standard (Rule 62 -302, F.A.C.) states that, for Class III freshwater water bodies, the DO concentration... Shall not be less than S.0 (mg /L). Normal daily and seasonal fluctuations above these levels shall be maintained. The standard also states that for Class III marine waters, the DO concentration ... Shall not average less than 5.0 in a 24 -hour period and shall never be less than 4.0. Normal daily and seasonal fluctuations above these levels shall be maintained. Therefore, in order to verify that the proper exceedance thresholds were applied, statistics for conductivity and salinity values (Min, Max, Mean, STD, N) were output for each WBID utilizing the MEANS procedure in SAS. The output was used to identify those WBIDs where the salinity levels were higher than expected based on water body type. These WBIDs were then added to a list of stations where salinity was sufficiently high to suggest they were actually "marine" rather than "freshwater. The following stations were identified as potentially being marine: • WBID 3278D- classified as Stream (3F) o Station 21FLSFWMECOCORIV: salinity maximum of 11.18 ppt). • WBID 3278G — classified as Stream (3F) o 92 observations with elevated salinities. These are in stations: ■ 21FLSFWMBC18 —with maximum salinity of 39.47 ppt, and 14 values greater than 5 ppt ■ 21 FLSFWMBC 19 — with maximum salinity of 40.1 ppt and 16 values greater than 5 ppt 3 Collier county vvatersnea ivivaei upualu I'mi and Plan Development ■ 21FLSFWMBC21 —with maximum salinity of 50.25 ppt and 21 values greater than 5 ppt Both WBID 3278D (Cocohatchee Inland) and WBID 3278G (Fakahatchee Strand) are located along the west coast and discharge into the Gulf of Mexico. These streams are likely influenced by tidal conditions, yet are classified as freshwater streams. Based on the discrepancy in the salinity data from the water quality stations, an evaluation of both the marine and freshwater DO standards was completed as part of this analysis. The available data set was evaluated consistent with I" requirements. It was found that 8 data points out of a total of 6,013 represented samples that were collected at the same location less than 4 days apart. It was concluded that this problem would not significantly affect the data interpretation, even though IWR guidance requires a 4 day interval for individual samples to be counted as separate events. Because many of the DO measurements appeared to be part of a profile sample, the lowest 10`" percentile value was calculated for each date at each station. The lowest l Ot" percent value was always equal to the minimum value as there were usually only 2 -3 samples per profile. The numbers of (minimum) DO concentration values ranging between 4 and 5 mg/L were counted and compared to the number of samples required to list a WBID as not meeting the DO criterion based on the sample size. It was determined that sufficient samples exceeded the criterion for DO to declare impairment in all the WBIDs reviewed. Results are shown in Table 2. Note that the required statistical confidence levels for the planning and verified periods are 80 and 90 percent, respectively; the number of samples needed for a conclusion of impairment will differ with any given sample size between the two periods. Previous work submitted for the Gordon River Extension provided evidence that the reference sites used in the Gordon River TMDL report (FDEP also failed to meet the DO criteria, indicating that the dissolved oxygen criteria may not be reflective of natural conditions, rather than anthropogenic impacts. Consequently, a site specific alternative criterion (SSAC) may be appropriate for deriving appropriate DO targets for Collier County water bodies. 4 Collier County watersnea rAoaei update and Plan Development Table 2 Number of Samples Exceeding the Dissolved Oxygen Concentration Threshold over the TMDL Planning and Verified Periods WBID Sample Size Below Marine Standard Below Freshwater Standard Minimum # of exceedances for impairment Planning Period <4.0 <5.0 3259W 518 63 3278D 342 130 175 40 3278F 69 57 10 3278G 180 117 144 22 3278K 75 69 11 3278L 30 23 5 3278R 276 102 33 32785 365 163 42 3278T 99 86 13 3278U 343 68 40 3278W 16 12 4 Verified Period 3259W 357 52 44 3278D 420 145 203 51 3278F 81 67 13 3278G 223 139 172 29 3278K 73 71 12 3278L 52 38 9 3278R 518 143 32785 401 159 49 3278T 94 81 14 3278U 283 43 36 3278W 26 1 18 6 2.2 Iron The maximum value for iron for each station per date was calculated. These maximum values were then compared to the IWR criteria of > 1,000 µg /l (Class 3F) and > 300 µg /l (Class 2). The number of exceedances based on these criteria, and the number of exceedances indicating impairment for the planning and verified periods were calculated (Table 3). Sufficient samples exceeded the criterion for Iron in all WBIDs reviewed to declare impairment. Although no detailed analyses have been conducted to determine the sources of iron, it is possible that elevated concentrations represent natural conditions. A more detailed examination of potential point sources, as well as an assessment of the potential influence of groundwater sources is warranted, but contaminant sources are not known to exist in these watersheds. 5 Collier County watersneo mooei vpuaia 11Mand Plan Development Table 3 Number of samples which exceeded the Iron threshold over the Planning and Verified Periods. WBID Planning Period Sample Size Class 3F > 1,000 1 Class 2 > 300 µg /1 Minimum # of exceedances for impairment 3259A 3261C 24 14 3 22 5 3 3278R 66 39 10 32785 53 12 8 Verified Period 3259A 73 40 12 3261C 21 6 5 3278R 260 161 33 32785 114 31 17 2.3 Fecal coliform bacteria The standard for fecal coliform impairment is evaluated for IWR impairment by water body classification (i.e., recreational or shellfish harvesting). The maximum criteria of 400 colony forming units (CFU) per 100 ml (Class 3F- recreational) and 43 CFU / 100 ml (Class 2- shellfish harvesting) was used in this analysis of Collier County impairments. Within each water body, the maximum value for fecal coliform for each station per date was determined using a sorting procedure. These maximum values were then compared to the Water Body Class criteria. For this analysis, PBS &J determined the number of times each WBID exceeded its appropriate criterion, and compared that number to the minimum number of exceedances (based on sample sizes and confidence levels) required for a finding of impairment for the planning and verified periods (Table 4). Sufficient samples exceeded the criterion for fecal coliform bacteria to declare impairment in all the reviewed WBIDs. The identification of fecal coliform contamination can be determined through Microbial Source Tracking. The Cocohatchee River Inland water body (WBID 3278D) already has a TMDL for fecal coliform bacteria. Source identification efforts could be incorporated into Basin Management Action Plan (BMAP) efforts. The function of the BMAP is to determine the types of projects required to meet TMDL goals, but BMAPs can also include efforts at further understanding the sources of pollutants in the watershed, as that information is critical to determining what types of projects are needed. This type of work has been conducted by PBS &J for water bodies in Polk, Hillsborough and Duval Counties. 6 Collier County Watershed Model Update fWand Plan Development Table 4 Number of samples which exceeded the Fecal Coliform Criteria During the Planning and Verified Periods WBID N Class 3F Class 2 Minimum # of exceedances for impairment Planning Period > 400 CFU /100 mL > 43 CFU /100 mL 3259A 98 69 13 3278G 162 22 20 327811 253 204 30 3278U 18 7 4 Verified Period 3259A 150 99 21 3278G 191 25 25 3278R 481 285 58 3278U 31 11 6 2.4 Copper According to FDEP, one WBID is impaired for copper: WBID 3278R (Naples Bay Coastal). The maximum value for copper for each station per date was calculated as part of the current analyses. These maximum values were then compared to the criterion of > 3.7 µg /l established for a Class 2 water body. As shown in Table 5, PBS &J determined the number of times WBID 3278R exceeded these criteria. PBS &J also calculated the minimum number of exceedances indicating impairment for the planning and verified periods. Sufficient samples exceeded the criterion for copper in WBID 3278R to declare impairment. Potential causes of copper could include stormwater runoff, which can contain elevated levels of various metals and other contaminants. However, a more detailed analysis of potential sources is necessary. 7 Collier County Watershed Model Update PWand Plan Development Table 5 Number of Samples Within WBID 3278R (Naples Bay Coastal) that Exceeded the Copper Threshold Over the Planning and Verified Periods 2.5 Chlorophyll a WBID 3278U (Rookery Bay Coastal) is an estuarine system and was the only water body declared impaired for nutrients (chlorophyll a) in Collier County, other than Lake Trafford. In order to declare a water body impaired for nutrients (chlorophyll a), only one year during the analysis period is required to exceed the maximum threshold of 11.0 µg /l for an estuarine system, the significance of which is discussed below. For this analysis, data on chlorophyll a (both corrected and uncorrected for phaeophytin) were first queried to determine the frequency of samples. There were insufficient data to calculate the annual average for the years prior to 1999. For the years 1999 -2005, there were uncorrected chlorophyll a data available in every quarter. Therefore, a quarterly average was computed per year (for the two periods) and annual averages were calculated from the quarterly averages (Table 6). Using corrected chlorophyll a data, a quarterly average was computed and an annual average was calculated from the quarterly averages for 2006. In 2006, the annual chlorophyll a- corrected value was 14.0 µg /1, which was used as the basis for the impairment. Table 6 Annual Chlorophyll a Average for WBID 3278U. Year Sample Size Chloro Minimum # of Period Sample > 3 7mg /I exceedances for Size 2000 4 5.8 2001 impairment Plannin 78 42 11 Verified 281 107 36 2.5 Chlorophyll a WBID 3278U (Rookery Bay Coastal) is an estuarine system and was the only water body declared impaired for nutrients (chlorophyll a) in Collier County, other than Lake Trafford. In order to declare a water body impaired for nutrients (chlorophyll a), only one year during the analysis period is required to exceed the maximum threshold of 11.0 µg /l for an estuarine system, the significance of which is discussed below. For this analysis, data on chlorophyll a (both corrected and uncorrected for phaeophytin) were first queried to determine the frequency of samples. There were insufficient data to calculate the annual average for the years prior to 1999. For the years 1999 -2005, there were uncorrected chlorophyll a data available in every quarter. Therefore, a quarterly average was computed per year (for the two periods) and annual averages were calculated from the quarterly averages (Table 6). Using corrected chlorophyll a data, a quarterly average was computed and an annual average was calculated from the quarterly averages for 2006. In 2006, the annual chlorophyll a- corrected value was 14.0 µg /1, which was used as the basis for the impairment. Table 6 Annual Chlorophyll a Average for WBID 3278U. Year Sample Size Chloro h Il a /l corrected Uncorrected 1999 4 4.6 2000 4 5.8 2001 4 4 2002 4 7 2003 4 2004 4 V4.9 0 2005 4 3 2006 4 14.0 The FDEP verified list indicates total nitrogen (TN), total phosphorus (TP) and biological oxygen demand (BOD) were "...causing impairment ". The average concentrations for all measurements of TN and TP over the verified period were 0.43 and 0.05 mg /l, respectively g uouier t ounty vvaiCiancu rvivu— vr-u.- and Plan Development (Table 7). BOD data were not available for the verified period. Under the current recommendations by FDEP, estuarine systems with a Trophic State Index (TSI) below 50 are considered "good ". In order to achieve this classification, average concentrations of TN < 0.7 mg /l, TP < 0.04 mg /1 and chlorophyll a < 10 µg /l are required (based on TSI equations for each variable). Concentrations of TN within Rookery Bay are well below the guidance levels from FDEP for estuarine systems, but levels of TP are higher than the above - mentioned guidance. It is therefore likely that phosphorus may be a more important nutrient for phytoplankton limitation than nitrogen, at least in Rookery Bay. As 2006 included data from I 1 sampling stations within Rookery Bay not considered for the previous years, a review of those stations was completed as part of this analysis to determine if differences in station locations contributed to the chlorophyll a exceedance. Those stations are listed in Table 8. The locations of the stations are shown in Figure 1. Table 7 . Average TN, TP and BOD concentrations for WBID 3278U for Planning and Verified Periods Prior to 2006, most water quality sampling stations within the Rookery Bay WBID were located within the more open water reaches of the bay, close to flushing channels. The 2006 stations are located alongside roadways near boat ramps, and away from the more open waters of the bay. These locations are less effected by tide and other mixing actions than the stations sampled prior to 2006. It is likely that the elevated chlorophyll a levels reported in 2006 reflect changes in characteristics of the sampling locations, rather than a true temporal change in water quality. 9 Collier County Watershed Model Update IMS1 and Plan Development TN TP BOD (m g/1) m /1 m /1 Planning Period 0.32 0.08 2.22 Verified Period 0.43 0.05 Prior to 2006, most water quality sampling stations within the Rookery Bay WBID were located within the more open water reaches of the bay, close to flushing channels. The 2006 stations are located alongside roadways near boat ramps, and away from the more open waters of the bay. These locations are less effected by tide and other mixing actions than the stations sampled prior to 2006. It is likely that the elevated chlorophyll a levels reported in 2006 reflect changes in characteristics of the sampling locations, rather than a true temporal change in water quality. 9 Collier County Watershed Model Update IMS1 and Plan Development Table 8 New Sampling Stations Monitored in 2006 for WBID 3278U Station 21FLFTM EVRGWC0059FTM 21FLFTM EVRGWC0060FTM 21FLFTM EVRGWC0061FTM 21FLFTM EVRGWC0062FTM 21FLFTM EVRGWC0063FTM 21FLFTM EVRGWC0027FTM 21FLFTM EVRGWC0028FTM 21FLFTM EVRGWC0029FTM 21FLFTM EVRGWC0030FTM 21FLFTM EVRGWC003 I FTM 21FLFTM PORTAUPR5 Figure 1 Locations of Sampling Stations Within WBID 3278U Including New 2006 Sampling Locations msi 10 Collier County Watershed Model Update and Plan Development 3.0 Pending Changes in the WBID Water Quality Standard Presently, FDEP is proposing changes to the water quality standards provided in Chapter 62- 302 and 62 -303. The changes could impact Class I and III Fresh Streams and Lakes. The evaluation based upon TSI values will be eliminated and replaced by numerical nutrient standards. The proposed standards would be calculated based on the 95 percent predication limits determined from a regression of log transformed chlorophyll a and TN or TP data for all Florida fresh streams or lakes. The proposed draft rule revisions currently include the following changes: For Class I and III Freshwater Streams: • TP limits From 0.069 mg /I to 0.415 mg /I • TN limits Ranging From 0.82 mg /l to 1.73 mg /l • Nitrate limit of 0.35 mg /1 for spring systems For lakes, the draft rules are outlined in Table 9. Nutrient thresholds would be invoked only if lakes did not meet chlorophyll a targets of 9 and 20 pg / liter for clear and colored lakes, respectively, with clear lakes being those with average color values lower than 40 platinum cobalt units (PCU) and colored lakes being those with average color values higher than 40 PCU. Should a lake exceed these chlorophyll a thresholds, proposed numeric nutrient criteria would be established as shown in Table 9. Table 9 Proposed TP and TN thresholds for lakes (Ken Weaver personal communication) >40 Platinum Cobalt 0.05 1.23 0.157 I 2.25 Units <40 Platinum Cobalt 0.030 1.00 0.087 1.81 Units and > 50 mg CaCO3/L <40 Platinum Cobalt 0.015 Units and = 50 mg CaCO3 L At present, there are no freshwater streams or lakes in Collier County that have been declared verified impaired for nutrients, other than Lake Trafford. However, future evaluations of water body conditions may be impacted by these proposed changes in the IWR methodology. 1VF4 11 Collier County Watershed Model Update and Plan Development 4.0 Conclusions There were no discrepancies in the mathematical calculation of impairment for the previously identified impaired water bodies by FDEP in Collier County. However, a review of the maximum thresholds for each parameter should be completed to ensure that they are reflective of the natural background concentrations of the region. • Eleven WBIDs in Collier County are verified impaired for Dissolved Oxygen. Review of the Gordon River Extension TMDL showed that the reference stations were also impaired. Therefore, a re- evaluation of the DO standard for this region may be necessary, and could be accomplished through the development of site specific alternative criteria for DO for the Collier County region. • Four WBIDs in Collier County are verified impaired for Iron. Groundwater and /or natural sources of iron could be the basis for impairment, as opposed to anthropogenic sources. • Four WBIDs in Collier County are verified impaired for Fecal Coliform Bacteria. Source identification of bacterial loading in the WBIDs can be completed to reduce /eliminate the concentrations, with such efforts potentially initiated via the Cocohatchee Inland TMDL program's BMAP process. • One WBID in Collier County is verified impaired for copper. Sources of copper could be related to stormwater runoff, but additional work to identify potential sources is necessary. • Rookery Bay is the only WBID in Collier County that has been verified impaired for nutrients (chlorophyll a), other than Lake Trafford. It is estimated that the location of the sampling stations in 2006 contributed to the impairment of the water body, as new sampling locations are less effected by tide and other mixing actions than the stations sampled prior to 2006. It is likely that the elevated chlorophyll a levels reported in 2006 reflect the change in characteristics of the sampling locations, rather than a true temporal change in water quality. � _�����,`1 12 Collier County Watershed Model Update E� Y and Plan Development :1 Technical Memorandum To: Mac Hatcher, PM Collier County From: Moris Cabezas, PBS&J Dave Tomasko, PBS&J Date: June 25, 2009 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 4, Task 3: Water Quality and Ecological Assessment of the Gordon River This evaluation covers portions of the work associated with Element 4, Task 3 for the Collier County Model Update and Plan Development. A similar assessment is being conducted for Lake Trafford. This evaluation contains the following components: ■ An overview of the Gordon River extension. ■ Water quality issues in the Gordon River. ■ Concerns related to the TMDL for the Gordon River extension. ■ Guidance on developing site - specific criteria and modeling approaches for the Gordon River and elsewhere. An Overview of the Gordon River Extension The Gordon River watershed, identified by the Florida Department of Environmental Protection (FDEP) as WBID 3278K, is located entirely within Collier County. The watershed also includes portions of the city of Naples. Much of the watershed is modified by a series of artificial canals and lakes which function either as mechanisms to drain the landscape or as locations of fill material for development. The watershed of WBID 2378K covers approximately 8 square miles (FDEP 2008), and is approximately three miles long. Substantial portions of the headwaters of the Gordon River are drained by the Gordon River Extension. The river then flows southward through a portion of the watershed characterized mostly by wetlands, before reaching its confluence with the Golden Gate Canal at the northwest corner of the Naples Municipal Airport. Below the confluence with the Golden Gate Canal, the Gordon River flows south for just over 1 mile before discharging into Naples Bay south of US Highway 41. That portion of the Gordon River just upstream of its confluence with the Golden Gate Canal is characterized by a watershed that is largely in an undeveloped state; areas further upstream are bordered mostly by golf courses, and areas downstream are characterized by intense development. A water control structure is located along the main Collier County Watershed Model update and Plan Development channel at the Golden Gate Parkway crossing. This structure was placed to prevent saltwater intrusion upstream. Figure 1 — Location map and features of the Gordon River watershed (WBID 3278K). Control structures and water quality sampling stations shown. Also shown are groundwater protection zones (image from Collier County). The prevention of saltwater intrusion is necessary to protect water supply wells for the City of Naples that are located along the upper reaches of the Gordon River (see Figure 1; Mac Hatcher, personal communication). There is also a control structure located on the Golden Gate Canal approximately one mile upstream from its confluence with the Gordon River. Consequently, water quality conditions for areas upstream of the control structures on both the Gordon River and the Golden Gate Canal result in those areas being freshwater systems. 2 Collier County Watershed Model Update t and Plan Development 0 Portions of the Gordon River and the Golden Gate Canal downstream from those two control structures have water quality that varies from freshwater conditions during times of high flows to increasingly brackish and even marine conditions during times of reduced rainfall. Water Quality issues in the Gordon River The Gordon River extension (WBID 3278K) was verified as impaired for dissolved oxygen (DO) based on the results of sampling and analysis carried out between January 1, 1995, and June 30, 2002 during the planning period for the first basin assessment cycle (Table 1). These results revealed that 67 percent of the DO concentration values measured during the planning period of the Total Maximum Daily Load (TMDL) program were below the Class III freshwater criterion of 5 milligrams per liter (mg /L). During this period, 46 surface water samples were analyzed for total nitrogen (TN) and 58 surface water samples were analyzed for total phosphorus (TP), with median values of 0.755 and 0.07 mg/L, respectively. TN was considered the causative pollutant for the low DO values observed. Table 1 — Data from Cycle 1 verified sampling period (1995 -2002) for the Gordon River (WBID 3278K). Data from FDEP (2008). Parameter No. of samples No. of Exceedances Exceedances needed for impairment (per IWR guidance) Median value DO 116 76 17 3.89 BOD 6 NA NA 1.05 TP 86 NA NA 0.05 TN 95 NA NA 0.75 Additional data were collected and summarized by FDEP for their Cycle 2 Verified period, as shown in Table 2. These data cover the more recent time period of 2000 to 2007. For both Cycle I and Cycle 2 assessments, the Gordon River failed the freshwater DO standard for Class III waterbodies. Table 2 — Data from Cycle 2 verified sampling period (2000 -2007) for the Gordon River (WBID 3278K). Data from FDEP (2008). Upon verifying that water quality data within the Gordon River failed to meet the established DO standard for freshwater waterbodies, FDEP (2008) then identified the causative pollutants by "...investigating those parameters typically responsible for depressed DO." 3 Collier County Watershed nnoaei upaate and Plan Development i Exceedances Parameter No. of samples No. of Exceedances needed for impairment (per Median value IWR guidance) DO 72 70 12 2.53 BOD 17 NA NA 2.0 TP 66 NA NA 0.07 Upon verifying that water quality data within the Gordon River failed to meet the established DO standard for freshwater waterbodies, FDEP (2008) then identified the causative pollutants by "...investigating those parameters typically responsible for depressed DO." 3 Collier County Watershed nnoaei upaate and Plan Development i The TMDL for the Gordon River (FDEP 2008) showed an inverse relationship between DO and Total Nitrogen (TN), a finding that supported the concept that there was a link between "pollutants" such as TN and the response variable of DO. Figure 2 (same data used as in Figure 3.1 in FDEP [20081) shows the results of the relationship between TN and DO for the reference sites. Reference Sites (median values) 8.0 6.0 E4.0 y - - 6.1148x + 8.7376 2.0 R = 0.4409; p < 0.01 • 0.0 0 0.2 0.4 0.6 0.8 1 TN (mg 1 liter) Figure 2 - Relationship between reference station median TN and DO values (data from FDEP 2008). Based upon the data shown in Figure 2, it was concluded by FDEP that the causative pollutant of TN was responsible for at least a portion of the DO problems in the Gordon River. Upon concluding that levels of TN were a causative pollutant responsible for the DO impairments in the Gordon River, FDEP developed a statistical approach to develop a target concentration for TN. Sampling stations meant to represent water quality from areas with "...relatively low impacts from urban development..." were chosen from throughout the FDEP's Southwest coast Planning Unit. The 75th percentile of the median TN values from these WBIDs was used to derive a reference TN concentration. Using data from the same WBIDs that were used for the DO vs. TN regression (Figure 2), the 75t' percentile for TN was calculated to be 0.74 mg TN / liter. This value was then chosen to represent the target TN concentration value for the Gordon River, which would then represent conditions that would allow the Gordon River to meet state standards for DO. Assessments were also completed for total phosphorus (TP) and biological oxygen demand (BOD), using these same WBIDs. The reference concentration targets developed using this approach are 0.04 mg /L for TP and 1.85 mg /L for BOD Water quality from the Gordon River was then assessed to determine the frequency of occurrence for data exceeding this 0.74 mg TN / liter target. For each exceedance, the concentration reduction required was determined using the following equation: [(exceedance value) - (0.74)] / ( exceedance value) x 100 4 Collier County watersnea ivwaei LJpuniC MIand Plan Development The median concentration reduction calculated (of the 36 exceedances) was 29 percent. The TMDL thus concluded that a 29 percent reduction in TN loads would be needed to bring TN concentrations down to 0.74 mg / liter. The TN concentration of 0.74 mg / liter was thought to represent the "best quartile" (i.e., best 25 %) of water quality conditions for reference sites with minimal development. The 29 percent TN load reduction was thus concluded to be capable of bringing about the TN concentrations appropriate for the DO standard for the Gordon River. Concerns Related to the TMDL for the Gordon River A number of items are problematic with the Gordon River TMDL report (FDEP 2008). A brief examination of these factors is valuable for determining the depth and breadth of issues that should be considered as relates to water quality monitoring and modeling efforts for both the Gordon River, as well as other similar systems (i.e., other surface water features in Collier County). These items include the following: 1) appropriate water quality standard for the Gordon River, 2) reference sites applicability to the Gordon River, and 3) causative pollutant for DO? These items will be discussed below. Appropriate water quality standard for the Gordon River (WBID 3278K)? A primary concern is whether or not the Gordon River extension (WBID 3278K) is in fact a freshwater waterbody, and there is a difference in water quality standards between freshwater and marine systems. Florida Administrative Code 62- 302.200 defines "Predominantly Fresh Waters" as those where "...the chloride concentration at the surface is less than 1,500 milligrams per liter." Most water quality monitoring programs do not report levels of chloride. Instead, they typically report either specific conductance or salinity. The equation used for converting chloride to salinity (Dawes 1981) is: Salinity (ppt) = 0.03 + 1.805 x chlorinity (converted to ppt) Upon conversion, 1,500 milligrams per liter of chloride approximates a salinity of 2.5 ppt. The 1,500 milligrams of chloride standard is a "not to exceed" value; locations where maximum salinities exceed 2.5 ppt would thus be classified as marine waters. Upon examining the I" data base for the Gordon River watershed used for the TMDL report (WBID 3278K), the vast majority of sampling data comes from station 21FLSFWMBC3, which is located downstream of the water control structure at Golden Gate Parkway (see Figure 3). The mean and maximum salinities in IWR 35 for this station are 8.73 and 30.6 ppt, respectively. Consequently, the vast majority of water quality data used in the Gordon River TMDL report (FDEP 2008) come from a waterbody type that should be classified as marine, not freshwater. The marine DO standard, which states that DO levels "Shall not average less than 5.0 in a 24 -hour period and shall never be less than 4.0" is thus the appropriate water quality criteria for this station. From both the Cycle 1 and Cycle 2 sampling periods, DO levels in the Gordon River would still fail the marine DO standard; even though it is less restrictive than the freshwater water body DO standard (see Tables 1 and 2). 5 Collier County Watershed Modei update M) ) and Plan Development Figure 3 - Location of stations and number of sampling events for stations within the Gordon River extension (WBID 3278K). Data from IWR run 35. Reference sites applicability to the Gordon River? The sampling sites used by FDEP to represent reference conditions are shown in Figure 4. While the Gordon River TMDL report (FDEP 2008) was based on the designation of the Gordon River extension (WBID 3278K) as being freshwater, the data from that WBID (3278K) show that the majority of water quality data in that WBID is from a station that is actually marine, not freshwater. As such, it was surmised that a similar assessment should be made of the reference sites used in the TMDL report. Data on salinity for each station used as a reference is shown in Table 3. 6 Collier County Watershed Model Update si and Plan Development Figure 4 - Locations of reference stations for the Gordon River TMDL (from Appendix A of FDEP 2008). Table 3 - Summary of water quality data for salinity for reference sites for the Gordon River TMDL report (data from IWR 35). IWR Station Salinity t Minimum Maximum Mean 21 FLSFWMFAKA858 0.11 0.30 0.22 21 FLSFWMFAKA 0.00 2.90 0.49 21 FLSFWMCHKMATE 0.00 0.27 0.17 21 FLSFWMBC9 0.10 7.28 0.43 21 FLSFWMBC8 0.00 0.32 0.24 21 FLSFWMBC7 0.10 1.25 0.31 21 FLSFWMBC22 0.13 6.29 0.69 21 FLSFWMBC21 0.18 50.25 7.88 21 FLSFWMBC20 0.22 41.65 5.22 21 FLSFWMBC19 0.20 40.10 4.57 21 FLSFWMBC18 0.14 39.47 4.23 21FLSFWMBC12 0.00 0.46 0.30 21 FLSFWMBC10 0.00 1 0.31 1 0.24 pni 7 Collier County Watershed Model Update and Plan Development Converting the freshwater standard (1,500 mg / liter chloride) into a salinity value of 2.5 ppt (using equations from Dawes 1983) it can be seen that seven of the thirteen reference sites should actually be considered marine, rather than freshwater. Four of the thirteen stations are actually hypersaline on occasion, with salinity values higher than typical oceanic values of about 35 ppt. These stations (21FLSFWMBCI8, 21FLSFWMBCI9, 21FLSFWMBC20, and 21FLSFWMBC21) are located along Tamiami Trail, in areas that are actually more consistent, in terms of salinity, with the classification of a saltern, also known as a salt flat (Dawes 1983). These areas are characterized by irregular saltwater intrusions that evaporate in place, and have unique ecologies that only halophytic plants have adapted to. In no sense are they consistent with their use in the TMDL report (FDEP 2008) as reference sites for freshwater systems. Only six sites apparently meet the chloride concentration guidance for freshwater. These sites (21FLSFWMFAKA858, 21FLSFWMCHKMATE, 21FLSFWMBC8, 21FLSFWMBC7, 21FLSFWMBCI2, and 21FLSFWMBCIO) are mostly located considerably north of Tamiami Trail. The Gordon River extension watershed (WBID 3278K) contains both freshwater and marine conditions, depending upon whether the station is located upstream or downstream from the water control structure at Golden Gate Parkway. While sites within the Gordon River fail to meet both marine and freshwater DO criteria, so do the majority of reference sites used in the Gordon River TMDL report (Table 3). . Table 4 - Summary of water quality data for DO for reference sites for the Gordon River TMDL report (data from IWR 35). IWR Station DO m / liter Minimum Maximum Mean 21 FLSFWMFAKA858 1.73 9.20 4.67 21 FLSFWMFAKA 2.92 9.83 6.35 21 FLSFWMCHKMATE 0.32 9.54 2.14 21 FLSFWMBC9 1.75 12.82 4.58 21 FLSFWMBC8 2.65 12.61 6.80 21 FLSFWMBC7 1.61 12.14 7.04 21 FLSFWMBC22 1.78 10.02 5.61 21 FLSFWMBC21 0.83 9.78 4.24 21 FLSFWMBC20 1.00 12.88 3.82 21FLSFWMBC19 0.24 10.16 3.18 21FLSFWMBC18 0.30 8.08 3.17 21 FLSFWMBC12 0.90 12.13 6.20 21 FLSFWMBC10 0.53 9.21 5.63 Regardless of which DO criteria is applied (marine or freshwater) all thirteen reference sites used in the Gordon River TMDL report (FDEP 2008) contain DO values below 4 mg / liter. Thus, the question arises as to whether or not DO levels in Collier County can meet the values outlined in 62.320, as neither impacted nor reference sites meet either freshwater or marine standards. ������� "/ g Collier County Watershed nnoaei update and Plan Development Causative Pollutant for Low DO Concentrations The identification of nitrogen as the causative pollutant for the Gordon River was based on the inverse relationship between DO and TN, as shown in Figure 2 (Figure 3.1 in FDEP 2008). The data used to produce this relationship was based on a comparison of median values for both DO and TN for the reference sites used by FDEP (2008), even though those stations represent a mixture of freshwater, marine, and intermittently hypersaline environments (Table 3). When data from the Gordon River watershed itself are used (IWR 35) there is no statistically significant relationship between TN and DO (Figure 5). Gordon River (WBID 3278K) F- 7 5 as 4 pE 3 G 2 1 0 0.0 0.5 1.0 1.5 2.0 TN (mg / liter) Figure 5 — DO (mg /liter) vs. TN (mg /liter) for the Gordon River (WBID 3278K). Data from IWR 35. These results show that when using data from within the Gordon River itself, there is no evidence that reductions in TN concentrations would bring about a concurrent increase in DO values. In addition, the DO and TN data from the Gordon River suggest that attainment of TN values less than the proposed standard of 0.74 mg TN /liter is no guarantee that DO levels will meet standards (whether marine or freshwater), as described below. On the 44 dates when both TN and DO were both report ed from station 21FLSFWMBC3, TN values were at or below the proposed TMDL target of 0.74 mg TN /liter 16 times (36% of samples). During those 16 times, DO values fell below 4 mg /liter 14 times, for an 88% "failure rate" of the never -to- exceed marine DO standard (which is the appropriate standard for that location, based on its salinity). Levels of DO were below 5 mg /liter 15 of the 16 times when TN values were below the proposed target TN level of 0.74 mg /liter. Clearly, levels of DO in the Gordon River do not track levels of TN in any meaningful way, and maintenance of TN levels at or below the proposed target TN level suggested by FDEP (2008) is no assurance that DO levels will meet what appears to be an inappropriate standard for the Gordon River. 9 Collier County Watershed Model Update iF% and Plan Development Guidance on Developing Site - Specific Criteria and Modeling Approaches for the Gordon River and Elsewhere For Collier County, the desire to protect and restore water quality would likely require the development of site - specific water quality criteria, especially for DO concentrations. In marine systems, levels of DO decline with both increased water temperature and increased salinity (Weiss 1970) an issue that is not fully accommodated by existing state standards. While the DO target for marine waters does allow for values to fall below 5 mg / liter (as long as they do not fall below 4 mg / liter) the marine standard requires all DO values to be above 4 mg / liter. For freshwater systems, DO values must remain above 5 mg / liter. Neither the freshwater nor the marine standards fully account for the influence of water temperatures on DO concentrations. In the absence of any biological activity, the influence of water temperature and salinity alone can significantly reduce DO levels. For example, water temperatures at site 21FLSFWMBC3 in the Gordon River averaged 25.5° C, and reached as high as 31.7° C. Salinities at this station averaged 8.7 ppt, with a maximum salinity of 30.6 ppt. Under average conditions at this site (i.e., water temperature of 25.5° C and salinity of 8.7 ppt) distilled water alone would contain 7.7 mg DO / liter at 100% saturation (from Weiss 1970). At the maximum temperature and salinity combination, distilled water would contain approximately 6.2 mg DO / liter at 100% saturation (from Weiss 1970) a value 19% lower. In addition to the effects of water temperature and salinity, the surface waters of Collier County, including the Gordon River, are characterized by a significant amount of biological activity. The western portion of the Everglades is contained within Collier County, and a considerable amount of water quality characterization has been previously conducted in this area. In their study of DO levels in the Everglades, McCormick et al. (1997) examined both nutrient - impacted sites and reference sites. The reference sites all had levels of total phosphorus below adopted guidance for Everglades restoration efforts, and were characterized by healthy assemblages of emergent plant communities. McCormick et al. (1997) found that "...even at reference sites, 02 was less than the 5 mg 1 -1 water quality standard for other water bodies (State of Florida Class III Standards) from 40 -70% of time." In fact, reference sites used in the Everglades study by McCormick et al. (1997) and also in the Gordon River TMDL report (FDEP 2008) failed the State of Florida's DO standards for both freshwater and marine waterbodies. The reference sites for the Gordon River TMDL report (FDEP 2008) were mostly located in the Picayune Strand, which is within the Big Cypress Basin. The Picayune Strand is comprised of hydric forests (mostly Cypress) in the lower elevation areas, hardwood hammocks in the upland areas, and a mixture of wet prairie and pine flatwoods in the transitional areas. The finding that DO standards are not met in reference sites in either the Everglades or the Picayune Strand suggests that the existing state -wide DO standard may be regionally inappropriate, as the conclusions of the inadequacy of the state -wide DO criteria for the Everglades also appear relevant for the Gordon River. Rather than striving to achieve an inappropriate water quality standard (as is the case with the existing TMDL effort), the issue to be addressed should be about an appropriate DO level for both freshwater and marine systems in Collier County. 10 Collier County Watershed Model Update Mand Plan Development In 2004, FDEP produced the report "Everglades Marsh Dissolved Oxygen Site Specific Alternative Criterion Technical Support Document ". This report was produced in part due to the conclusion by FDEP (2004) that DO levels lower than the Class III standard of 5 mg / liter (Chapter 62 -302, F.A.C.) are common, even in sites "...minimally impacted by nutrient enrichment or cattail invasion." Rather than use a state -wide standard, FDEP produced a site specific alternative criteria (SSAC) for DO that predicted DO levels based on water temperature and the time of day that readings were made. The SSAC for Everglades DO found that the factors that best explained the observed pattern of DO in reference locations were the influence of water temperature and sample collection time. As shown by Weiss (1970) water temperature is inversely related to oxygen concentrations, as warm water has a lower saturation point that colder water. The influence of time of day was also noted by McCormick et al. (1997) and reflects the background conditions wherein periphyton biomass either adds DO to the water via photosynthesis, or removes DO from the water via respiration. These background processes were considered by application of the following equation: DOij = -3.70—[1-50 -sine (2Tr11440 - tip) -(0.30 -sine (4Tr11440 • tij))] + 11(0.0683 + 0.00198 - Cij + 5.24.10 -6 - Cij 2) Where, tij and Cij are the sample collection time (in minutes) and water temperature (in degrees C), respectively, of the ith dissolved oxygen measurement at the jth station. The value of 1440 refers to the number of minutes in a day. This algorithm was developed after determining the most appropriate locations to use as reference sites. The reference sites used for the herbaceous marsh systems in the Everglades DO SSAC (FDEP 2004) were mostly open water and /or marsh systems, while the TMDL reference sites for the Gordon River were located within the forested wetlands of the Picayune Strand; the word "strand" denotes a forested wetland. Therefore, the equations used to develop the SSAC for the Everglades are not likely to be directly transferable for developing a SSAC for the Gordon River. However, the need for a SSAC for the Gordon River, and other locations such as the Gordon River, is as evident as was the case in the Everglades. As the reference sites (chosen for their lack of impacts due to urbanization) do not meet water quality criteria for DO, it is likely that the DO criteria itself needs revision. A further consideration, when developing a SSAC for the Gordon River, is the impacts to DO associated with the influence of groundwater. From 2007 to 2009, groundwater quality of the surficial aquifer was assessed on five separate occasions at eight locations within the Gordon River watershed. Sample depths ranged from 1.33 to 7.05 feet below ground surface. The mean DO level (n = 40) was 0.58 mg / liter, with a minimum of 0.37 and a maximum of 1.17 mg DO / liter. Groundwater infiltrating into the surface waters of the Gordon River would likely depress DO values, as the Gordon River has numerous locations where groundwater inflows have been observed either as "boils" in sandy bottoms of the canal, or in areas where lateral groundwater movement has been observed in the side cuts of the channel walls (Mac Hatcher 11 Collier County Watershed Model update and Plan Development i and Rhonda Watkins, personal communications). This information could be useful as a metric to compare estimates of the hydrologic balance between surface water and groundwater, to determine if DO levels in the Gordon River can be used as a "tracer" of groundwater influence. The MIKE -SHE model currently being developed under this contract will allow development of mass balances based on groundwater and surface water interactions. As there is no evidence that the TN vs. DO relationship used in the Gordon River TMDL report (FDEP 2008) works for the Gordon River itself, there is little evidence to suggest that TN load reductions would have an impact on DO in the Gordon River. And since DO levels in the Gordon River most often do not meet either freshwater or marine DO criteria, even when TN levels are at or below the TMDL target of 0.74 mg TN / liter, there is little reason to believe that attaining a TN target of 0.74 mg / liter would result in DO targets being met in the Gordon River. Recommended steps would be to develop, in close coordination with FDEP, a SSAC for DO in the Gordon River, using methodology that can be extrapolated to both the Golden Gate Canal and other surface water systems in Collier County. This SSAC, in combination with a detailed water balance, determining the percentage of water in the river due to surface water vs. groundwater, would seem more capable of explaining DO levels in the Gordon River than the TMDL approach outlined by FDEP (2008). REFERENCES Dawes, C.J. 1981. Marine Botany. John Wiley and Sons. New York. 628 pp. FDEP. 2004. Everglades Marsh Dissolved Oxygen Site Specific Alternative Criterion Technical Support Document. Final Report to Florida Department of Environmental Protection. Tallahassee, FL. 61 pp. FDEP. 2008. Dissolved Oxygen TMDL for the Gordon River Extensions WBID 3278K (formerly 3259C). Final Report to Florida Department of Environmental Protection. Tallahassee, FL. 40 pp. McCormick, P.V., Chimney, M.J. and U.R. Swift. 1997. Diel oxygen profiles and water column community metabolism in the Florida Everglades, U.S.A. Archives die Hyrobiologie. 140: 117 -129. Weiss. R.F. 1970. The solubility of nitrogen, oxygen and argon in water and seawater. Dee - Sea Research. 17: 721 -735. 12 Collier County Watershed Model update Pffil and Plan Development ati To: Mac Hatcher, PM Collier County From: Moris Cabezas, PhD, PE Technical Memorandum David Tomasko, PBS&J Date: 8/13/09 Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318, Element 4, Task 3 Water Quality and Ecological Assessment of Lake Trafford 1.0 Introduction This evaluation covers portions of the work associated with Element 4, Task 3 for the Collier County Model Update and Plan Development. A similar assessment has been conducted for Gordon River. This evaluation contains the following components: ■ An overview of Lake Trafford. ■ Water quality issues in Lake Trafford. ■ Concerns related to the TMDL for Lake Trafford. ■ Guidance on developing site - specific criteria and modeling approaches for Lake Trafford and elsewhere. 2.0 An Overview of Lake Trafford The Lake Trafford watershed, identified by the Florida Department of Environmental Protection (FDEP) as WBID 3259W, is located entirely within Collier County. It encompasses 18,393 acres, the majority of which are classified as cropland (6,526 acres) or wetlands (6,189 acres) and includes the city of Immokalee. Lake Trafford is the largest lake south of Lake Okeechobee, covering approximately 1,522 acres, with an average water depth of 7 feet. It is located in the headwaters to the Corkscrew Swamp, and it discharges south through the Fakahatchee strand and then into the Southern Golden Gate Estates Critical Project Area (identified by CERP) during high water levels in the wet season. The lake has no defined tributaries and presently there are no hydraulic control structures in the area draining into the lake. Figure 1 shows an overview of Lake Trafford and its immediate watershed; note the lack of any significant shoreline development. Collier County Watershed Model Update and Plan Development Figure 1: Lake Trafford Watershed 3.0 Water Quality Issues in Lake Trafford Lake Trafford (WBID 3259W) was verified as impaired for nutrients, un- ionized ammonia, and dissolved oxygen (DO) based on the results of sampling and analysis carried out between January 1, 2000, and June 30, 2007 during the Florida Total Maximum daily Load (TMDL) program verification period for the second basin assessment cycle (FDEP 2008a; summarized in Table 1). The assessment for nutrient impairment is based on an elevated Trophic State Index (TSI) value, which is calculated using concentrations of total nitrogen (TN), total phosphorus (TP) and chlorophyll -a (Chi -a). The State of Florida's Impaired Waters Rule (IWR; 62- 303.352, FAC) states that a TSI value which exceeds 60 in any one year of the verified period is sufficient to classify a lake with a color level greater than 40 platinum - cobalt units as impaired for nutrients. In addition, to calculate an annual TSI value, at least one sample of TN, TP and Chl -a must be taken within each season (quarter) of the year. Based upon the average color values for Lake Trafford over the verification period, a TSI of 60 is applicable. PW 2 Collier County Watershed Model Update and Plan Development The TSI value was calculated for the verified period (2000 -2007) except 2006 and 2007 due to data limitation. As shown in Figure 2, Lake Trafford TSI values exceeded 60 in all six of the years for which it was calculated (FDEP 2008). Based upon these results, Lake Trafford was declared impaired for nutrients. Figure 1 — Trophic State Index (TSI) calculations by year (from FDEP 2008). 80 60 40 20 2000 2001 2002 2003 2004 2005 Year The Lake Trafford un- ionized ammonia concentrations were high enough to warrant a classification of the lake as impaired based upon a 10% exceedance of the maximum allowable concentration in freshwater lakes. The Class III freshwater criterion of 0.02 mg /1 was violated 44 times out of the 332 samples collected from January 2000 and June 2007 (i.e., 13% of samples exceeded the criterion). An evaluation of dissolved oxygen (DO) concentrations indicated that Lake Trafford did not meet the criteria for impairment during the first cycle verified period of 1994 -2002. However, additional data were collected and summarized by FDEP for their Cycle 2 Verified period, as shown in Table 1. These data cover the more recent time period of 2000 to 2007. Results revealed that 15 percent of the DO concentration values measured during the TMDL verified period were below the Class III freshwater criterion of 5 milligrams per liter (mg/L). During this period, 409 surface water samples were analyzed for TN and 296 surface water samples were analyzed for TP, with median values of 2.43 and 0.103 mg/L, respectively (Table 1). Nutrients were considered the causative pollutant for the low D.O. values observed. PW 3 Collier County Watershed Model Update and Plan Development Table 1 — Data from Cycle 2 verified sampling period (1994 -2002) for Lake Trafford - WRID 2259W_ Data from FDEP (2008). Upon verifying that water quality data within Lake Trafford failed to meet the established nutrient standard for freshwater waterbodies, FDEP (2008a) then "...applied the Hydrologic Simulation Program FORTRAN (HSPF) model to simulate water quality discharges to the lake and eutrophication processes in the lake to determine the appropriate nutrient target ". The HSPF model is designed to assess the water quantity and quality of surface water and groundwater. The external load assessment (point, nonpoint and atmospheric loads) involved the use of land use patterns, soils, topography, hydrography, point sources, service area coverages, climate and rainfall. Multiple modules, parameters and constants are required to simulate the complicated biological processes associated with lake eutrophication analysis. As in all models, the results of model simulations are only as accurate as the quality and spatial and temporal intensity of the datasets available. In order to estimate historical water quality conditions for the lake, the current land use categories for the Lake Trafford watershed were reassigned as water, wetlands and undeveloped rangeland /upland forests. Additionally, the sediment oxygen demand (SOD) rate was adjusted to reflect the reduced organic matter inputs to the bottom sediments that would result from a more natural land use. Using the hydrological conditions that occurred from 1998 to 2007, the average historical TN, TP and Chl -a concentrations were reported to be 1.07 mg /l, 0.018 mg /1 and 17.1 pg /1, respectively. The corresponding historical TSI value for Lake Trafford was calculated to be 51. The TMDL TSI Target for Lake Trafford was determined based upon the modeled historical TSI and nutrient limitation. In order to allow for assimilative capacity within the lake over time, FDEP allows for a 5 unit increase in TSI over historical values without being declared impaired. Therefore, based upon a historical TSI of 51, the target TSI for Lake Trafford was determined to be 56. The current nutrient limitation condition for the lake was attributed to co- limitation by both nitrogen and phosphorus. However, the historical condition of the lake was thought to be that of a phosphorus- limited system. 4.0 Concerns Related to the TMDL for Lake Trafford A number of items are problematic with the Lake Trafford TMDL report (FDEP 2008a). A brief examination of these factors is valuable for determining the depth and breadth of issues that should be considered as relates to water quality monitoring and modeling efforts for Lake Trafford and other similar systems (i.e., other similar surface water features in Collier County). These items include the following: 1) HSPF water quality model parameterization 4 Collier County Watershed Model Update INVand Plan Development Exceedances Parameter No. of samples No. of Exceedances needed for impairment (per Median value IWR guidance) DO 357 44 52 (IWR Run 32) BOD 493 NA NA 4.1 TP 296 NA NA 0.103 TN 409 NA NA 2.43 Upon verifying that water quality data within Lake Trafford failed to meet the established nutrient standard for freshwater waterbodies, FDEP (2008a) then "...applied the Hydrologic Simulation Program FORTRAN (HSPF) model to simulate water quality discharges to the lake and eutrophication processes in the lake to determine the appropriate nutrient target ". The HSPF model is designed to assess the water quantity and quality of surface water and groundwater. The external load assessment (point, nonpoint and atmospheric loads) involved the use of land use patterns, soils, topography, hydrography, point sources, service area coverages, climate and rainfall. Multiple modules, parameters and constants are required to simulate the complicated biological processes associated with lake eutrophication analysis. As in all models, the results of model simulations are only as accurate as the quality and spatial and temporal intensity of the datasets available. In order to estimate historical water quality conditions for the lake, the current land use categories for the Lake Trafford watershed were reassigned as water, wetlands and undeveloped rangeland /upland forests. Additionally, the sediment oxygen demand (SOD) rate was adjusted to reflect the reduced organic matter inputs to the bottom sediments that would result from a more natural land use. Using the hydrological conditions that occurred from 1998 to 2007, the average historical TN, TP and Chl -a concentrations were reported to be 1.07 mg /l, 0.018 mg /1 and 17.1 pg /1, respectively. The corresponding historical TSI value for Lake Trafford was calculated to be 51. The TMDL TSI Target for Lake Trafford was determined based upon the modeled historical TSI and nutrient limitation. In order to allow for assimilative capacity within the lake over time, FDEP allows for a 5 unit increase in TSI over historical values without being declared impaired. Therefore, based upon a historical TSI of 51, the target TSI for Lake Trafford was determined to be 56. The current nutrient limitation condition for the lake was attributed to co- limitation by both nitrogen and phosphorus. However, the historical condition of the lake was thought to be that of a phosphorus- limited system. 4.0 Concerns Related to the TMDL for Lake Trafford A number of items are problematic with the Lake Trafford TMDL report (FDEP 2008a). A brief examination of these factors is valuable for determining the depth and breadth of issues that should be considered as relates to water quality monitoring and modeling efforts for Lake Trafford and other similar systems (i.e., other similar surface water features in Collier County). These items include the following: 1) HSPF water quality model parameterization 4 Collier County Watershed Model Update INVand Plan Development and assumptions, 2) TMDL consistency with recent water quality improvements, and 3) dredging effects on water quality in Lake Trafford. These items will be discussed below. 4.1 HSPF Water Quality Model Parameterization and Assumptions The existing conditions for the Lake Trafford watershed were modeled using a 10 year simulation period of 1998 to 2007. The calibration period was 1998 to 2002 and the validation period was 2003 through 2007. Of special note is that a large -scale sediment removal project was implemented and completed in Lake Trafford between November 2005 and May 2006 as part of a restoration plan to improve fisheries resources of the lake. Thus, the model validation period encompasses time prior to, during, and after the sediment removal project. More importantly, model simulations for this period do not incorporate water quality conditions that might have been altered by the sediment removal activities. As discussed further below, if sediment removal significantly altered water quality, the model is likely not representing current ambient water quality conditions and /or nutrient - phytoplankton relationships. The HPSF model contains modules to simulate biological processes within the water column (i.e., phytoplankton response to nutrient availability, etc.). Several calibration constants were derived to represent rates of plankton production in Lake Trafford. The ratio of Chl -a to phosphorus content of phytoplankton (RATCLP) was assigned a value of 3. However, an evaluation of the relationship between Chl -a and TP from 1998 to 2002 (i.e., the calibration period) indicates a ratio of either 0.24 or 24 depending on whether or not values of Chl -a and TP are "normalized" to the same units or left as is (i.e., µg /l for Chl -a and mg /l for TP). These results (Figure 2) indicate that the assumed value of the coefficient is questionable to represent Chl -a responses to TP concentrations. �l r��� 5 Collier County Watershed Model Update and Plan Development Figure 2 — Chlorophyll -a responses to TP during the model calibration time period (1998 to 2002). Data from IWR -35. a� E R A Gi O t. O A -0.1 6E -16 0.1 0.2 0.3 0.4 0.5 0.6 TP (mg/1) Additionally, the fraction of nitrogen requirements for phytoplankton growth that is satisfied by nitrate (ALNPR) was given a value of 0.75 for the calibration period. Based upon the water quality data over the calibration period, a significant relationship between nitrate and chlorophyll was not evident (neither positive nor negative correlations), which suggests that the HSPF model does not appropriately reflect the role of inorganic nitrogen uptake kinetics on phytoplankton growth in Lake Trafford. A more compelling issue, related to both the TMDL itself (FDEP 2008a) and lake management activities, is that of nutrient limitation in general, and how the TMDL report determined the most likely limiting nutrient for Lake Trafford phytoplankton. In their report, FDEP (2008a) concluded that phytoplankton within Lake Trafford is most likely limited by the abundance of phosphorus, as levels of Chl -a did not exceed 50 µg / liter during those times when the TN: TP ratio exceeded 30 (the level at which FDEP presumes TP to be limiting). While this approach could be an appropriate method for assessing nutrient limitation, it is not the only technique available to assess nutrient limitation. Upon examining the water quality data for the period of record for Lake Trafford, it appears that nitrogen might play a currently unrecognized role in controlling phytoplankton levels. The IWR database was utilized to obtain data for this analysis. As a statewide database, it contains data from multiple agencies and a vast amount of water quality data. Run 35 of the IWR dataset "iwr2002 _pv— Run35.sas7bdat" was queried for all data for Lake Trafford (WBID 3259W). These data were further queried for selected parameters of interest, 6 Collier County Watershed Model Update and Plan Development y = 0.2419x + 0.0076 R2 = 0.2886 p <0.0001 -0.1 6E -16 0.1 0.2 0.3 0.4 0.5 0.6 TP (mg/1) Additionally, the fraction of nitrogen requirements for phytoplankton growth that is satisfied by nitrate (ALNPR) was given a value of 0.75 for the calibration period. Based upon the water quality data over the calibration period, a significant relationship between nitrate and chlorophyll was not evident (neither positive nor negative correlations), which suggests that the HSPF model does not appropriately reflect the role of inorganic nitrogen uptake kinetics on phytoplankton growth in Lake Trafford. A more compelling issue, related to both the TMDL itself (FDEP 2008a) and lake management activities, is that of nutrient limitation in general, and how the TMDL report determined the most likely limiting nutrient for Lake Trafford phytoplankton. In their report, FDEP (2008a) concluded that phytoplankton within Lake Trafford is most likely limited by the abundance of phosphorus, as levels of Chl -a did not exceed 50 µg / liter during those times when the TN: TP ratio exceeded 30 (the level at which FDEP presumes TP to be limiting). While this approach could be an appropriate method for assessing nutrient limitation, it is not the only technique available to assess nutrient limitation. Upon examining the water quality data for the period of record for Lake Trafford, it appears that nitrogen might play a currently unrecognized role in controlling phytoplankton levels. The IWR database was utilized to obtain data for this analysis. As a statewide database, it contains data from multiple agencies and a vast amount of water quality data. Run 35 of the IWR dataset "iwr2002 _pv— Run35.sas7bdat" was queried for all data for Lake Trafford (WBID 3259W). These data were further queried for selected parameters of interest, 6 Collier County Watershed Model Update and Plan Development including those associated with lake eutrophication processes. For data analysis, TN was used from the direct measurement or calculated by summing the TKN and Nitrate values. The IWR database contained nitrogen and phosphorus data from 1967 until December 2004. Additionally, Collier County provided supplemental water quality data collected before and after the initial dredging activities were completed (2004 to 2008). The two datasets were either analyzed separately, or combined for analysis depending upon the question posed. The dataset was further filtered for data collected at water depths no more than I meter below the surface, to ensure a consistency of sample collection efforts would be used for analysis. On those occasions when multiple station locations were sampled on the same day, daily average values were calculated for the lake. Daily averages were used for all statistical interpretations. The pattern of Chl -a concentrations over time in Lake Trafford is shown in Figure 3. Figure 3 - Chlorophyll -a in Lake Trafford. Line is 5 point moving average. 250 200 w 150 ° 100 U h 50 0 Jan -88 Jan -92 Jan -96 Jan -00 Jan -04 Jan -08 Date Based on the IWR database, information on the abundance of Chl -a in Lake Trafford dates back to the late 1980s. Results indicate a pattern of lower levels of Chl -a in the late 1980s, very high levels in the late 1990s, and a recent decline in Chl -a over the past few years. Figure 4 compares the pattern of TP concentrations over time to the pattern for Chl -a concentrations, truncated to the same time frame (nutrient data were collected prior to Chl -a data). 7 Collier County Watershed Model Update and Plan Development i Figure 4 - TP (mg / liter) and chlorophyll -a (µg / liter) in Lake Trafford. Lines are 5 point moving averages. 250 200 C bD 150 a c 100 a U 50 0 Jan -88 * Chi a 1.4 1.2 Gn 0.8 0 a 0.6 a 0.4 F 0.2 0 Jan -92 Jan -96 Jan -00 Jan -04 Jan -08 a TP 5 per. Mov. Avg. (Chi a) 5 per. Mov. Avg. (TP) When comparing patterns of both Chl -a and TP, a temporal disconnect seems to exist. High levels of TP in the late 1980s do not appear to coincide with elevated levels of Chl -a, and high levels of Chl -a in the late 1990s do not appear to coincide with elevated levels of TP. This temporal disconnect does not exist when comparing concentration trends of Chl -a vs. TN (Figure 5). �P��,,.r 8 Collier County Watershed Model Update Mand Plan Development Figure 5 - TN (mg / liter) and chlorophyll -a (µg / liter) in Lake Trafford. Lines are 5 oint movine averages. 250 s 8 F- 7 ■ 0 200 150 -- - -� 5 8 ■ ® ■ 4 0 c. ■ c 100 ■ * 3 2 50 0 Jan -88 Jan -92 Jan -96 Jan -00 Jan -04 Jan -08 Chia ■ TN 5 per. Mov. Avg. (Chi a) 5 per. Mov. Avg. (TN) As opposed to TP, the patterns of TN and Chka concentrations match more closely. Both TN and Chka show temporal patterns of lower levels in the late 1980s, peak abundances in the late 1990s, and a general pattern of decline over the past few years. To test for the relative importance of TN vs. TP as "predictors" of levels of Chi -a, an additional technique to test for the general influence of nutrients on phytoplankton was pursued. This technique involved the use of Spearman's rho to test for the presence of any statistically significant correlations between TN, TP, and Chia. This non - parametric statistical technique was used, rather than linear regression, due to the highly non - normal distribution of data, and the lack of homoscedasticity (aka homogeneity of variance) of the various data sets. The data set used in this analysis was limited to the years prior to the onset of the lake -wide sediment removal project, which started in November 2005, to control for the effects of project construction. 9 Collier County Watershed Model Update PWand Plan Development Table 2 — Spearman's rank correlations for TN, TP and Chl -a for Lake Trafford. Correlation coefficients (first row of each comparison) range between -1 and +1 and measure the strength of the linear relationship between selected variables. Also shown (second row) is the P -value which tests the statistical significance of the estimated correlations. P- values below 0.05 (in red) indicate statistically significant non -zero correlations at the 95.0% confidence level. The results summarized in Table 2 show that TN correlates with levels of Chl -a to a much higher degree than does TP, although both TN and TP have statistically significant correlations with Chka. The correlation coefficient between TN and Chka is more than three times as high as that between TP and Chka (0.593 vs. 0.193, respectively). Interestingly, there is no significant correlation between levels of TN and TP, suggesting the sources of these two nutrients may differ. Based upon the results of this analysis, it is apparent that Lake Trafford is dominated by nitrogen - fixing phytoplankton (discussed in more detail below). Nitrogen fixers can access a pool of nitrogen (di- nitrogen gas derived from the atmosphere) that other species of phytoplankton cannot incorporate. Recorded concentrations of TN within Lake Trafford (e.g., 2 to 6 mg / liter) are elevated well above the median concentration for Florida lakes (1.1 mg /1). In addition, the range of high TN concentrations found in Lake Trafford includes numerous readings well above those that can be explained by even the worst -case scenarios for stormwater loads considered in both the Lake Trafford and the Gordon River TMDL reports (FDEP 2008a and 2008b, respectively). These extremely high TN values are consistent with prior assessments in Lakes Jesup and Hancock (PBS &J 2006 and 2008, respectively), which found significant rates of nitrogen fixation by cyanobacteria (aka blue -green algae) in lakes with similarly elevated levels of TN. In contrast, reported total phosphorus concentrations are similar or lower than levels used to quantify baseflow and stormwater loads into both Lake Trafford and the Gordon River from their surrounding watersheds (FDEP 2008a and 2008b, respectively). Consequently, the HSPF representation of nitrogen uptake may not be accurate; thereby invalidating the conclusions reached for TMDL development. 4.2 TMDL Consistency with Recent Water Quality Improvements A large - scale, multi -year habitat restoration project involving sediment removal has already been initiated for Lake Trafford, and Phase I efforts have already been completed. The 10 Collier County Watershed Model Update and Plan Development TP TN Chl -a TP 0.347 0.330 ;t it 0.( 004 11 0.+?0 0 7 TN 0.347 0.388 1' OMW4 1' o.Ooo I Chl -a 0.330 0.388 1' = O.00( i7 1' 0.000 I The results summarized in Table 2 show that TN correlates with levels of Chl -a to a much higher degree than does TP, although both TN and TP have statistically significant correlations with Chka. The correlation coefficient between TN and Chka is more than three times as high as that between TP and Chka (0.593 vs. 0.193, respectively). Interestingly, there is no significant correlation between levels of TN and TP, suggesting the sources of these two nutrients may differ. Based upon the results of this analysis, it is apparent that Lake Trafford is dominated by nitrogen - fixing phytoplankton (discussed in more detail below). Nitrogen fixers can access a pool of nitrogen (di- nitrogen gas derived from the atmosphere) that other species of phytoplankton cannot incorporate. Recorded concentrations of TN within Lake Trafford (e.g., 2 to 6 mg / liter) are elevated well above the median concentration for Florida lakes (1.1 mg /1). In addition, the range of high TN concentrations found in Lake Trafford includes numerous readings well above those that can be explained by even the worst -case scenarios for stormwater loads considered in both the Lake Trafford and the Gordon River TMDL reports (FDEP 2008a and 2008b, respectively). These extremely high TN values are consistent with prior assessments in Lakes Jesup and Hancock (PBS &J 2006 and 2008, respectively), which found significant rates of nitrogen fixation by cyanobacteria (aka blue -green algae) in lakes with similarly elevated levels of TN. In contrast, reported total phosphorus concentrations are similar or lower than levels used to quantify baseflow and stormwater loads into both Lake Trafford and the Gordon River from their surrounding watersheds (FDEP 2008a and 2008b, respectively). Consequently, the HSPF representation of nitrogen uptake may not be accurate; thereby invalidating the conclusions reached for TMDL development. 4.2 TMDL Consistency with Recent Water Quality Improvements A large - scale, multi -year habitat restoration project involving sediment removal has already been initiated for Lake Trafford, and Phase I efforts have already been completed. The 10 Collier County Watershed Model Update and Plan Development objective of the restoration project is to enhance fisheries habitat by the removal of nutrient laden organic sediment from the lake bottom. These activities are also likely to reduce the resuspension of sediments and nutrients to the water column from the lake bottom. Approximately 3.5 million cubic yards of organic muck was dredged from the lake between November 2005 and May 2006. Additional dredging is planned within the Lake to remove the remaining organic material. This organic material is thought to have accumulated over time due to infestation of the lake with Hydrilla and subsequent management actions which killed off this species in- place. The HSPF model calibration period used to develop the TMDL for Lake Trafford preceded lade dredging activities and the model output was used to develop scenarios for water quality responses to various nutrient levels, which were in turn used to derive the substantial load reductions called for in the FDEP (2008a) TMDL report.re is evidence of improvements in water quality within Lake Trafford that seem to have arisen in response to sediment removal, the values and coefficients used in the model's nutrient - phytoplankton response equations may no longer be appropriate or accurate and the proposed pollutant load reductions, 77 and 60 percent for TP and TN, respectively, would not be applicable. 4.3 Dredging Effects on Water Quality in Lake Trafford An evaluation of pre and post sediment removal in Lake Trafford was completed to determine if water quality improvements have occurred in response to sediment dredging activities. It must be noted that sediment removal has been undertaken on a number of lakes in Florida (e.g., Banana Lake, Lake Hollingsworth, and Lake Maggiore) usually with the intent to improve water quality. As there is a much larger data set collected prior to restoration than after restoration, we chose to make the data sets as similar as possible by comparing pre- restoration water quality data to the same length of time, 27 months, as that which was available post- restoration. In this way, differences in the sample size of pre vs. post- restoration were minimized. Data collected between June 2003 and September 2005 were evaluated as representative of pre- dredging water quality. Data collected between September 2006 and December 2008 were considered representative of post- dredging water quality. The non - parametric Mann - Whitney W test (comparison of medians) was used due to the non - normal distribution of the data. Multiple parameters were evaluated to assess water quality (Table 3). 11 Collier County Watershed Model Update and Plan Development Table 3 - Comparison of Median concentrations of multiple parameters in Lake Trafford. Values represent sample size (N), median values, and statistical significance (Mann - Whitney W Test, significant if p <0.05). Parameter Median N Pre N Post p -value Chl -a (µg /1) 26 41.5 31 13.7 {).Ot) {); TN (mg/1) 23 2.6 31 2.1 0.000") TP (mg/1) 23 0.200 31 0.203 0.986 Turbidity (NTU) 26 8.8 31 4.6 0.00017 Secchi depth (m) 29 0.43 18 0.48 0.7 Dissolved Oxygen (mg /1) 38 9.38 18 8.36 0.076 Conductivity (µS /cm) 29 276 18 332 ().Ot nl) I TN:TP 23 11.7 31 6.5 0.077 Concentrations of Chl -a were significantly lower in the post- restoration time period than in the years prior to sediment removal (13.7 vs. 41.5 µg / liter, respectively) for a decline of 67 percent. Concurrent with the decline in Chl -a is a decline of TN from 2.6 to 2.1 mg / liter, a decline of 19 percent. In contrast, there was no evidence of a decline in levels of TP associated with sediment removal. However, the decline in TN was not sufficient in magnitude to produce a change in the ratio of TN to TP Turbidity declined from 8.8 to 4.6 NTU, a decline of 48 percent, after sediment removal. However, there was no evidence of a statistically significant change in water clarity (measured as Secchi depths). Similar to recent findings in Lake Maggiore, in Pinellas County, sediment removal seems to have been accompanied by an increase in conductivity, as values increased from 276 to 332 µS / cm, an increase of 20 percent possibly because of increased groundwater inflows. 5.0 Guidance on Developing Site - Specific Criteria and Modeling Approaches for Lake Trafford and Elsewhere For Lake Trafford, efforts to protect and restore water quality likely require the development of site - specific water quality criteria for levels of Chka, as well as for the nutrients TN and TP. FDEP's TMDL report for Lake Trafford (FDEP 2008a) is based on output from efforts to model potential responses of phytoplankton populations within the lake to various lake management activities, particularly those related to nutrient availability. In their report, FDEP (2008a) concluded that phytoplankton within Lake Trafford is most likely limited by the abundance of phosphorus. Upon examining the water quality data for the period of record for Lake Trafford, it appears that nitrogen might play a currently unrecognized role in controlling phytoplankton levels, based on the following findings: 12 Collier County Watershed Model Update and Plan Development I . The temporal pattern of TN better matches the temporal pattern of Chl -a in Lake Trafford, compared to TP 2. The value of the correlation (Pearson product- moment) between TN and Chl -a is three times the correlation value between TP and Chi -a. However, TP correlates with Chl -a quite well during the calibration period of 1998 to 2002. 3. The reduction of Chl -a in response to sediment dredging occurred simultaneously with a decline in levels of TN, without any evidence of a concurrent reduction in TP post- dredging. However, hyper - eutrophic lakes such as Lake Trafford (or at least Lake Trafford in its pre - dredging condition) cannot be truly nitrogen- limited, as cyanobacteria that dominate in such systems can potentially access a pool of nitrogen, di- nitrogen gas in the atmosphere that is not dependent upon stormwater loads and /or bottom resuspension. It is therefore recommended that TMDL load reductions called for in the Lake Trafford TMDL report (FDEP 2008a) should not be implemented until it is clear whether or not the desired improvements in water quality for the lake may have already been partially or wholly accomplished through the completion of Phase I of the large -scale sediment removal project. Data collected to -date suggest that phytoplankton levels have declined to levels meeting State Water Quality Standards in the years after the completion of the sediment removal project (see Table 3). In addition to the recommendation to delay implementation of TMDL load reductions until post- dredging water quality can be further assessed, it is recommended that computer models that may be used in the future to estimate phytoplankton responses to nutrient availability be calibrated based on data that reflect water quality, sediment, and biological conditions consistent with the simulations period. The models should also have the capability to simulated nutrient uptake kinetics associated with nitrogen - fixing cyanobacteria. Locally - collected data on nitrogen fixation rates and the role, if any, of phosphorus on stimulating rates of nitrogen fixation (i.e. PBS &J 2008) should be incorporated into future phytoplankton growth models. The collection of such data is integral to developing site - appropriate water quality paradigms, as conducted for both Lake Jesup, located in Seminole County, and Lake Hancock, located in Polk County (PBS &J 2006 and 2008, respectively). F%) 13 Collier County Watershed Model Update and Plan Development References Florida Department of Environmental Protection. 2008a. TMDL Report Nutrient, Un- ionized Ammonia and DO TMDLs for Lake Trafford (WBID 3259W). Florida Department of Environmental Protection, Tallahassee, FL. 88 pp. Florida Department of Environmental Protection. 2008b. Dissolved Oxygen TMDL for the Gordon River Extension, WBID 3278K (formerly 3259C). Florida Department of Environmental Protection, Tallahassee, FL. 40 pp. PBS &J. 2006. Nitrol?en- fixation in Lake Jesup. Final Report to Seminole County. 6 pp. PBS &J. 2008. Assessment of Water Quality Responses to Sediment Removal in Lake Hancock. Final Report to Polk County and the Florida Department of Environmental Protection. 84 pp. ��r� 14 Collier County Watershed Model Update and Plan Development Protecting Southwest Florida's unique natural environment and quality Of life ... now and August 23, 2010 x� W Mac Hatcher Collier County Government Fd 7 7 3050 N. Horseshoe Drive, Suite 145 Naples, FL 34104 TMf CONSERVANCY ©f Southwest Florida RE: COLLIER COUNTY WATERSHED MANAGEMENT PLAN DEVELOPMENT 1450 Merrihue Drive Naples. Florida 34102 The Conservancy of Southwest Florida has reviewed the documents recently released with regard to the development of Collier County Watershed Management Plans and offers the 239.262.0304 following comments. As you know, we have been very supportive of watershed management plan development as a tool towards more effective water resource management, but seg`ith;39'262.0672 current emphasis in disputing water quality regulations, rather than adhering to them ,F , threatening to produce the opposite effect. Therefore, we urge Collier County staff to address and rectify the following identified issues in order to ensure the resulting plans produce the intended benefit. GENERAL CONCERNS WITH HANDLING OF WATER QUALITY ASPECTS The Florida Department of Environmental Protection (FDEP) already thoroughly compiles, assesses data, and develops Total Maximum Daily Loads (TMDLs) and Basin Management Plans (BMAPs) as necessitated by state and federal water quality policies and laws. Therefore, the Scope of Work for these plans should have never included a task (Element 4 — Task 1) to revaluate Impaired Waters Rule (IWR) data. Doing so was redundant, resulting in a waste of precious limited public funds and time to complete these plans. Moreover, unfounded consultant opinions as to whether current legally binding standards have merit, or should be adhered to, defies the County's legal obligation to comply with state and federal water quality standards and regulations. The purpose of the Watershed Management Plans (WMPs) is "to protect the County's estuarine and wetland systems "', not to analyze actions which could potentially allow the County to skirt the existing regulations designed to protect estuarine and wetland systems. IMPROPER ASSESSMENT OF WATERBODIES BELIEVED TO BE IMPAIRED DUE TO NATURAL CONDITIONS The technical memorandum from David Tomasko to Mac Hatcher dated 8/24/09 raises numerous points of disagreement between the contracted consultant and state and federal water quality law. The first is that waterbodies which receive loadings from natural sources should never be listed as "impaired ". This is erroneous from both a state and federal legal perspective. The Florida Department of Environmental Protection (FDEP) determines and classifies whether impairment is believed to be the result of a natural pollutant source - specifically for potential natural dissolved oxygen (DO), iron, copper, and nutrient impairments. These waterbodies are noted as Category 4c impaired waterbodies on the state 303d lists. Many of the waterbodies that Mr. Tomasko speculated as "naturally polluted" have not been classified as Category 4c by the state and thus, are not even believed by FDEP to be impaired based on natural pollutant sources. Moreover, according to federal law and guidance, even FDEP - classified Category 4c waterbodies should be relisted as Category 5 (needing a TMDL) if there is any indication that ' Collier County Growth Management Plan. CONSERVATION AND COASTAL MANAGEMENT ELEMENT. January 25, 2007. Policy 1.3.5. GOAL 2. (VI) Objective 2.1 anthropogenic pollution is, or could be present The EPA guidance figure below illustrates that unless pollutant concentrations are solely linked to natural sources, a TMDL must be completed for that waterbody2. Similarly, a waterbody must remain on the 303(d) list even if some portion of the exceedance can be contributed to natural sources, as shown in bar "C" of the figure. 0 0 Co s✓ a� U C O U N a� to M Making 303(d) Listing Decisions for Waterbodies with Naturally Occurring All anthropogenic Combination of Combination of Ail natural anthropogenic anthropogenic and natural and natural Sources of Impairment X Water Quality Criterion Legend A-Nal Breed a�l.ellaesd C- 1,).hd -yr.0 wkh a N.WW Conftmc Vromaim D- Licl.d -- t—Wl. Mc. Neural CoWitiom Vr.viaian (— la.l). If there are absolutely no anthropogenic pollutant factors that are plausible or can be identified, then that must be fully demonstrated with scientific evidence as part of an application to obtain site specific alternative criteria (SSAC) or place the waterbody on the "delist list ". Until such time that a SSAC has been granted or the waterbody has been delisted, the legal requirement is to develop a TMDL to meet applicable water quality standards. To that end, the County should not continue to question, but instead comply, with such regulatory requirements by utilizing EPA's Florida 303(d0 list Decision Documents as the appropriate list of impairments and waterbodies to address. IMPROPER ASSESSMENT OF DISSOLVED OXYGEN (DO) IMPAIRMENTS Also outlined in the August 24, 2009 Technical Memorandum from Dave Tomasko to Mac Hatcher3, 11 WBIDs in Collier County are not meeting the state water quality standards for DO and are currently on the FDEP Impaired Waters List. The memo cites previous work submitted (by PB &J assumingly) regarding reference sites used in the Gordon River Extension TMDL for DO - as evidence that DO impairment may not be caused by anthropogenic sources. The referenced previous work was not provided with the memo, and therefore cannot be commented on. However, the Gordon River Extension TMDL highlights an even larger problem concerning reference sites. The Conservancy would agree that the reference site approach utilized in the TMDL was not appropriate, however this does not provide evidence that natural sources are solely contributing to low DO in the subject impaired WBIDs. The four WBIDs mentioned, 3278G, 3278H, 32781, and 3278V are in fact, all impaired for DO. WBID 3278G, Fakahatchee 2 Information Concerning 2008 Clean Water Act Sections 303(d), 305(b), and 314 Integrated Reporting and Listing Decisions, pp. 10 -11 (Oct. 12, 2006). 3 Tomasko, Dave PBS &J. Technical Memorandum to Mac Hatcher, PM Collier County. Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318, Element 4 - Task 1 — Review of IWR data. August 24, 2009. Strand is included on the 303(d) list and the remaining WBIDs have exceeded state water quality standards for DO, however no causative pollutant was found (FDEP category 4d), meaning that total nitrogen (TN), total phosphorus (TP), nor biological oxygen demand (BOD) exceeded FDEP's "thresholds ". Again, had FDEP thought the DO to be the result of natural conditions, it would have classified these as 4c — which it did not. According to federal regulations, EPA is responsible for developing TMDLs for those impairments that FDEP has placed in a category 4d as required under federal water quality law. Moreover, county sampling in even in the most "pristine" areas, such as the Fakahatchee reference site Mr. Tomasko referred to, is primarily being conducted within canals that drain upstream urbanized areas containing stormwater pollution. No basin within the county is truly unaffected by anthropogenic activity and therefore, would qualify as for exemption of a TMDL due solely on natural conditions under federal law. That being said, a TMDL assessment assesses pre - development natural load and subtracts it from existing pollutant loads to determine the appropriate load reduction value — so the county would never be required to remove natural pollutant loads anyhow. Therefore, the County should comply with DO impairment determinations in creating watershed management plans that reduce the anthropogenic inputs of pollutants which depress DO levels. INAPPROPRIATE RECOMMENDATION FOR COUNTY -WIDE DO DEVIATIONS The memo suggests that a Site Specific Alternative Criteria (SSAC) "may be appropriate for deriving appropriate DO targets for Collier County waterbodies ". It should be made clear that: 1) SSAC are developed on a WBID by WBID basis, not county -wide, and 2) SSAC still have to maintain the existing designated use of the waterbody unless a Use Attainability Analysis (UAA) is completed that demonstrates that the current use is unattainable. Both a SSAC and a UAA are very costly, and since many would be needed in order to legally allow for deviation of DO standards county -wide, it would extremely expensive and risky for the County to invest such resources - unless there was definitive proof that DO standards are not influenced at all by any anthropogenic factor. Conducting water quality testing ourselves through -out the county and reviewing all other available water quality data, the Conservancy does not see the scientific justification required for such DO deviations to be granted. Additionally, Collier County staff indicated at the August 4, 2010 Collier EAC meeting that the County has no intentions of developing SSACs for DO. Therefore, the point of SSACs is moot and again, the focus of these plans should be on meeting current DO water quality standards instead. INAPPROPRIATE CONTESTING OF SALINITY CHARACTERIZATION It was also unnecessary to assess salinity regimes of WBIDs and apply DO criteria that do not match their designated use or current water quality standards. As you are aware, FDEP - with Collier County's participation - recently went through an extensive WBID re- delineation process just a couple years ago to determine boundaries that better reflect waterbodies and their designated use. In the June 25, 2009 Technical Memorandum from Dave Tomasko to Mac Hatcher4 PBS &J reports that "[w]hile sites within the Gordon River fail to meet both marine and freshwater DO criteria, so do the majority of reference sites used in the Gordon River TMDL report." As outlined above, simply because reference sites failed to meet criteria does not mean there are no anthropogenic factors contributing to the reference site impairment or the subject waterbodies impairment. Therefore, the County should not rely on this unsupported recommendation to Tomasko, Dave PBS &J. Technical Memorandum to Mac Hatcher, PM Collier County. Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500106318 Element 4, Task 3. Water Quality and Ecological Assessment of the Gordon River. revisit the characterization of WBIDs as fresh or marine as an avenue to arguing for more lenient water quality standards to be applied. Moreover, the analysis was a moot point either way, because the two WBIDs assessed as potentially marine in the Memorandum - failed water quality standards for DO when using either the marine or freshwater criteria. INAPPROPRIATE NEGATION OF IRON IMPAIRMENTS The August 24, 2009 Technical Memorandum from Dave Tomasko to Mac Hatchery suggests that "no detailed analyses have been conducted" for sources of iron and that "contaminant sources are not known to exist in these watersheds ". It should be noted that generally an assessment of pollutant sources is conducted through the TMDL process and therefore, is forthcoming after impairment has been verified. Therefore, is would be premature to expect that those sources would be identified at this point prior to an Iron TMDL. Regardless, unless Mr. Tomasko can provide sufficient scientific evidence that the iron exceedance is entirely natural and thus, a SSAC is granted or the waterbody is delisted by FDEP — the legal requirement is to meet state iron water quality standards. That being said, the Conservancy's cursory analysis shows at least one potential non- natural source of iron within WBIDs impaired for iron - active solid waste facilities (shown on the map below). In FDEP's "Evaluation of Potential Ground Water /Geologic Contributions of Iron" for the Caloosahatchee6 at least 6 potential sources of iron were established: 1) Solid Waste Sites in Planning Unit or Near WBID of interest, 2) Solid Waste Facilities in watershed of WBIDs being evaluated, 3) SUPERFUND Sites in Planning Unit or Near WBID of interest, 4) SUPERFUND Sites in watershed of WBIDs being evaluated, 5) NPDES Discharges in Planning Unit or into WBID of interest, and 6) Sites in COMET Database in Planning Unit or Near WBID of Interest. This supports that there are anthropogenic inputs of iron and thus, iron impairments would likely apply. eld3 6 Florida Department of Environmental Protection. Caloosahatchee Basin, East Caloosahatchee (WBID 32376) West Caloosahatchee (WBIDs 3235E, 3235F, 3235G, 3235J, 3235L) Orange River (WBID 3240J) Estuarine Caloosahatchee (3240B, 3240E, 32401, 3240M, 3240N). Evaluation of Potential Ground Water /Geologic Contributions of Iron. Prepared by James Dodson, Teayann Tinsley, Akia Laurant, and Edgar Wade Ground Water Protection Section, Bureau of Watershed Restoration. Therefore, the County should instead work towards identifying potential anthropogenic inputs of iron (as seen in the map above) and assess measures necessary to meet state iron water quality standards. NEED FOR SEPARATE BUT COMPATIBLE BMAPS AND WMPS FOR COLLIER COUNTY'S WATERSHEDS Watershed Management Plans (WMP) cannot act as reasonable assurance documents in place of Basin Management Action Plans (BMAP), since they will not adequately address water quality impairments based FDEP's impaired waters lists and Total Maximum Daily Loads (TMDLs) already completed within the county. The county's Watershed Management Plans are not designed to isolate and assign specific pollutant / wasteload allocation reductions as a BMAP would, as well as are not a compliance instrument with the same level of binding and enforceable measures necessary to fulfill the requirements of the BMAP. BMAPs "represent a comprehensive set of strategies -- permit limits on wastewater facilities, urban and agricultural best management practices, conservation programs, financial assistance and revenue generating activities, etc. -- designed to implement the pollutant reductions established by the TMDL. These broad -based plans are developed with local stakeholders- -they rely on local input and local commitment- -and they are adopted by Secretarial Order to be enforceable". Ill Additionally, BMAPs cross geopolitical boundaries and also developed with the participation of all affected stakeholder groups who would be required for their successful implementation - such as the County, City of Naples, City of Marco, etc. in the case of Collier County. Whereas, Collier County Comprehensive Plan Policy 2.1.4 outlines that "All Watershed Management Plans shall address the following concepts: a. Appropriate wetlands and uplands serving as a buffer to wetlands are conserved; b. Drainage systems do not degrade wetland and estuary ecosystems; c. Surface water that potentially could recharge ground water is not unduly drained away; d. When feasible the extent and effects of salt -water intrusion are lessened; e. The timing and flow of fresh water into the estuaries from the watershed shall, as a minimum, not degrade estuarine resource value; f. The needs of the watershed's natural resources and human populations are balanced; g. The effects on natural flood plains, stream channels, native vegetative communities and natural protective barriers which are involved in the accommodation of flood waters; h. Non - structural rather than structural methods of surface water management should be considered first in any proposed new works; i. Wetland and estuarine habitat functions are conserved and/or enhanced; and j. Wetland and estuarine ecosystems will be conserved and/or enhanced using a variety of innovative tools, including landowner incentives, public acquisition, conservation easements, and /or transferable development rights." Therefore, the Collier WMPs are planning tools for evaluating land use changes to reduce impacts to water resources with regards to wetlands loss, or changes in quantity, timing or distribution of flows - but they do not emphasize water quality nor provide binding enforceable pollutant / wasteload allocation and reduction requirements for meeting Total Maximum Daily Load pollutant limits to restore water quality to state water quality standards. The Conservancy thus supports separate BMAPs be done to address these impairments, which would then work in tandem with Collier's WMPs. AGRICULTURAL LANDS SHOULD NOT BE ASSIGNED A ZERO IN RESOURCE VALUE EVALUATIONS The Technical Memorandum erroneously assigns a score of "0" for Vegetation /Habitat and Hydrological areas where a natural system has been converted to a developed land use class (e.g., agriculture, urban development, golf course, and pasture). Agricultural and pasture lands provide more natural resource value than urban development. Even the document is internally inconsistent with regards to this in that assigns agriculture land cover types scores ranging from moderate to high value in its Landscape Suitability Index (LSI). These land uses include woodland pasture; with livestock (8.87), pasture; without livestock (8.03), low intensity pasture; with livestock (7.32), citrus (7.02), high intensity pasture; with livestock (6.96), and row crops (6.07). Hydrology, like vegetation, looks at the preexisting conditions (PDVM) and compares it to the current conditions. While a predevelopment vegetation community provides optimal functional value for native wildlife (e.g., for food, cover, and breeding) and hydrologic function based on its intact native vegetative state, conversion to agricultural uses does not eliminate hydrologic or habitat value as explained in the following sections. Natural Resource Value of Agricultural Lands Overview: As vital as agriculture is to Florida's economy, the agricultural lands themselves also provide key habitat and ecological functions to the surrounding areas. They provide an important function to Florida's hydrology by sometimes acting as temporary water retention areas in addition to providing nesting and foraging habitat, habitat for base prey populations, and necessary components of the life cycle for various wildlife species. With proper management, they can help replenish aquifers and filter nutrients before they enter other systems. They often "serve as a buffer to encroaching urban development, and can restrict the spread of exotic and nuisance species to undeveloped areas' ". Agricultural lands support some of the states most imperiled species including, the crested caracara (Caracara plancus), southeastern American kestrel (Falco sparverius paulus), burrowing owl (Athene cunicularia floridana), wood stork (Mycteria americana), gopher tortoise (Gopherus Polyphemus), eastern indigo snake (Drymarchon couperr), and the Florida panther (Puma concolor coryt) as well as a vast array of other wildlife. In a developing landscape, Florida's wildlife has become more dependent on human impacted areas for survival. Where natural habitats are becoming scarcer because of land conversion to agriculture, species have adapted to their changing surroundings. Humans as well are learning to adapt and apply more natural water control and retainment methods on agricultural land, restoring the historic hydrology of Southwest Florida. Hydrologic Value of Agricultural Lands: The scores for hydrology were based on the length and duration of inundation and its functional value to native wildlife. Shifts in vegetation that represented a change in depth and duration of inundation were the result of the low scores that agricultural lands received. Not taken into account was that agricultural lands do or can have the ability to retain a large quantity of water similar to function of natural wetlands had on pre - developed land and are sometime used as a temporary flood storage basin. They also recharge groundwater levels as a result of rainfall or irrigation water absorption, reducing the run -off on the soil surface. A percentage of this soil retained water is then slowly released into canals and other water bodies$. Often, farm fields lie fallow and/or are seasonally flooded during the summer wet months when water storage is most needed. These storage and aquifer recharge functions can effectively stabilize water flow from the land and mitigate flood damage in downstream areas. In fact, there has been a recent movement to pay farmers for these water storage services. One such concept is Recyclable Water Containment Areas (RWCAs). A designated RWCA can be created on private land, such as an agricultural field, where it would persist for a temporarily agreed amount of time (i.e. five years). Studies have shown that temporary flooding of agricultural lands (such as in a RWCA) enhances water retention and by doing so, delays discharge from the watershed to local bays and estuaries. "Work by S. Shukla (UF /IFAS Agricultural and Biological Engineering Department) and colleagues on retention ponds in southwest Florida has shown that approximately 50% of the water in the pond is lost through 7 Restoring the Everglades: Challenges for Agriculture. Economic Research Service /USDA. Agricultural Outlook 1997). Retrieved from httr): / /www.ers.usda.aov/ publications /agoutlook/seDl997 /ao244d.Ddf Restoring the Everglades: Challenges for Agriculture. Economic Research Service /USDA. Agricultural Outlook (1997). Retrieved from http: / /www ers usda gov/ publications /agoutlook/sepl997 /ao244d Ddf lateral and downward movement9 ". Impounded water on agricultural lands however, increases the water table of adjacent lands leading to more water storage and reduced pumping for irrigation1'. As the water evaporates, detritus and other nutrients settle to the bottom. This is advantageous for future crop production, as well as the environment, by providing better soil for growth and less fertilizer and nutrient application needs. Therefore, the potential storage and aquifer recharge values that agricultural lands do or could provide should be reflected through a higher hydrologic score being assigned to them. Wildlife Value of Agricultural Lands: Agricultural lands also provide habitat for many imperiled species, with one such species being the Audubon's crested caracara. The caracara, a threatened bird of prey, occurs as an isolated population in the south central part of the state. Historically this region was dominated by xeric grassland or dry prairie, but native land cover has been subject to conversion to unimproved or improved pasture utilized for cattle ranching ". Morrison and Humphrey (2001) conducted a population study on the distribution and reproductive activity of caracara breeding pairs in relation to land ownership and usage. "Eighty - two percent of 73 active nest sites found were on privately owned cattle ranches12'. Additionally, the study found that "46 breeding areas with 4 years of known histories of occupancy and reproduction, pairs nesting on lands where the major land use was cattle ranching exhibited higher rates of breeding -area occupancy, attempted breeding during more years, initiated egg laying earlier, exhibited higher nesting success, and attempted a second brood after successfully fledging a first brood more often than pairs nesting on lands managed as natural areas 13". Populations of non - breeding caracaras also occupy habitats uncharacteristic of these breeding areas. "Specifically, citrus groves were occupied extensively, and row crops were used particularly during breeding seasonst4s. Non - breeding caracaras seem to prefer citrus groves because it serves as a refuge from high temperatures and breeding caracaras as they defend their territory'. The smallest falcon in the United States, the Southern American kestrel, also depends on agricultural fields for hunting. Kestrels, listed as threatened in Florida, utilize open pine habitats, woodland edges, prairies, and pastures throughout much of the statet5. They often perch themselves on telephone wires at the edge of a field or other open area. From this vantage point they hunt for their normal prey which includes: insects (favoring grass - hoppers and dragonflies), lizards, and small mammals16. The Florida burrowing owl, a state listed Species of Special Concern, occurs throughout the state "although its distribution is considered local and spotty and the presence of burrowing owls is primarily dependent upon habitat17". They often inhabit open native prairies and cleared areas that offer short groundcover including pastures, agricultural fields, golf courses, airports, and vacant lots in residential areas18 ". This species is one that has managed to thrive in areas affected by human development where land clearing has sometimes created new habitat for them. IBID 10 IBID 11 Morrison, Joan, and Stephen Humphrey. "Conservation Value of Private Lands for Crested." Conservation Biology. 15.3 (2001): 675 -684. Print. t2 IBID 13 IBID 14 Dwyer, J, F. (2010) Ecology of Non- breeding and Breeding Crested Caracaras (Caracara Cheriway) in Florida. Retrieved from http: / /scholar.lib.vt.edu/ theses /available /etd- 05092010 - 132909 /unrestricted /Dwyer JF D 2010.1)df 15 Field Guide to Rare Animals of Florida, Florida National Areas inventory (2001). Retrieved from http : /twww.fnal.org /FieldGuide /pdf /Falco sparverius paulusmdf American Kestrel, Florida Fish and Wildlife Conservation Commission. Accessed by http: // mvfwc. comNVILDLIFEHABITATs /BirdSpecies American Kestrel.htm Burrowing Owl, Florida Fish and Wildlife Conservation Commission. Accessed by ht� /mvfwc.comNVILDLIFEHABITATS /BirdSpecies BurrowinoOwl.htm IBID Agricultural lands have also become vital to the wetland species, such as the endangered wood stork - which has been observed using man -made wetlands such as storm water treatment areas and ponds, golf course ponds, borrow pits, reservoirs, roadside ditches, agricultural ditches, drainages, flow -ways, mining and mine reclamation areas, and dredge spoil sites for foraging and breeding purposes19. Other protected wading bird species, such as egrets, herons, ibises, and roseate spoonbills also make use of the shallow waters that collect on agriculture fields and nearby ditches for feeding. With rapid conversion of short hydro - period wetlands into development in recent years, water retention on agricultural lands are playing a larger role in the foraging habitat for these species. A study conducted by Main and Vavrina (2009) demonstrated the usage of wading bird species on such agricultural lands. Surveys were taken in and around 12 miles of canals serving agricultural operations on a 1,000 acre potato farm for 18 weeks starting in October until March, coinciding with the nesting season of many wading birds in southwest Florida 20. The results from these surveys documented over 1,619 individuals representing 11 species of wading birds21. Additionally, the "greatest concentrations of birds were observed clustered around ditch cleaning operations during October through December 22'. Another factor influencing the population of wading birds in the canals was when the water levels were lowered during February in preparation for harvest23. Figure 3. Number of Wading Birds Observed during Agricultural Area Surveys24 1E M 250 200 p 150 L 100 z 50 X, u 10/30 11/30 12/30 1/30 2/28 3/30 Month 19 Wood stork (Mycteria americana) Five Year Review: Summary and Evaluation, U.S. Fish and Wildlife Service. Accessed by http:// www. fws. gov /northflorida/WoodStorks/2007- Review/ 2007 -Wood- stork- 5- vr- Review.pdf 20 Main, Martin, and Vavrina, Charles. "Wading birds and agriculture in Southwest Florida." University of Florida IFAS Extension. 2009. Web. 16 Aug 2010. < http: / /edis.ifas. ufl.edu /pdffles/UW /UW13900.pdf >. 21 IBID 221BID 231BID 24 [BID Table 1. Wading birds observed during surveys of agricultural canals listed by species, number and % of total birds observed, and listed status by state and federal agencies 5. Species Count i °/a Listed Status Agency I Cattle Egret 410 i 25 I Great Egret 392 j 24 Snowy Egret 238 15 Species of Special Concern State j Wood Stork 3 193 � 12 Endangered State, Fed. White Ibis ' 172 I 11 Species of Special Concern State Little Blue Heron 114 7 Species of Special Concern i State Great Blue Heron 41 3 Roseate Spoonbill 19 1 Species of Special Concern State, Fed. Under Review Tri- colored Heron 19 1 Sandhill Crane 13 1 Threatened State I Glossy Ibis 8 0 Green - backed Heron 4 0 Total 1619 100 Many species of reptiles also utilize habitat on agricultural fields including the threatened gopher tortoise which, along with its burrows, are protected by state law. Gopher tortoises live in well - drained sandy areas with a sparse tree canopy and abundant low growing vegetation. They are commonly found in habitats such as sandhill, pine flatwoods, scrub, scrubby flatwoods, dry prairies, xeric hammock, pine -mixed hardwoods, and coastal dunes which have historically been maintained by periodic wild fires however, managed agricultural lands can also provide preferred habitat 28. In areas with no dominant tree cover such as improved pasture, abandoned pasture, cropland (row and field), abandoned citrus groves, fallow crop land, and disturbed habitat like farmland there is a high potential for gopher tortoise habitat27. "Mechanical clearing and grazing by cattle can also be used to maintain open canopy and encourage forage species of plants on which the gopher tortoise feedS28". As a keystone species, the gopher tortoises' burrows also provide shelter for "more than 360 species of animals, including indigo snakes, gopher frogs, Florida mice, skunks, opossums, rabbits , quail, armadillos , burrowing owls, snakes, lizards, frogs, toads, and many invertebrates. Many of these "commensals" use tortoise burrows to escape predators, adverse weather conditions, and fire 29". Some of these species are completely dependent on these burrows and cannot exist without them30. The presence of gopher tortoises and their burrows effectively create a unique ecology, in which a vast assortment of biodiversity is dependent. 25 IBID 16 Gopher Tortoise Habitat, Florida Fish and Wildlife Conservation Commission. Accessed by ht� / /myfwc.conyWILDLIFEHASITATS /GopherTortoise Habitat.htm Ashton, Ray, and Patricia Ashton. The Natural History and Management of the Gopher Tortoise. 1st edition. Malabar, FL: Krieger Publishing Company, 2008. 65 -93. Print. 29 IBID 29 Puckett, Catherine, and Franz, Richard. "Gopher Tortoise: A Species in Decline." University of Florida IFAS Extension. Gopher Tortoise Council, 2001. Web. 11 Aug 2010. < http: / /edis.ifas.ufl.edu /uw048 >. 30 IBID Eastern indigo snakes are also protected as a threatened species and utilize agricultural lands. In areas where there are populations of gopher tortoises, indigo snakes can be found sheltered in tortoise burrows where they take refuge from cold winters and desiccation 31. Studies of radio - marked eastern indigo snakes on the central ridge of South Florida indicate that they use a wide variety of natural, disturbed, and nonnatural habitat types'. "On the ridge itself, eastern indigos favor mature oak scrub, turkey oak sandhill, and abandoned citrus grove habitats, whereas snakes found off of the sandy ridges use flatwoods, seasonal ponds, improved pasture, and active and inactive agricultural lands32 ". In extreme South Florida habitats such as the Everglades and Florida Keys, eastern indigo snakes are found in tropical hardwood hammocks, pine rocklands, freshwater marshes, abandoned agricultural land, coastal prairie, mangrove swamps, and human - altered habitats33 The critically endangered Florida panther also utilizes these agricultural areas. Kautz et al. (2006) denotes three priority zones for panther conservation: primary, secondary, and dispersal zones. Primary zone, which is land necessary for the long -term viability and persistence of the panther in the wild, is 3,548 mil (9,189 km2) in size and 7.6% of it is agricultural lands 34. Secondary zone is 1,269 mil (3,287 km2) and 36% is agriculture and dispersal zone is 44 mil (113 km2) in size and is comprised of 49% agriculture 35. Panther home ranges often include contain a mosaic of natural habitats and man -made habitat such as agricultural lands, of which panthers utilize. Agricultural lands interspersed with native habitat can benefit and provide habitat for the panther's primary prey, which include deer and feral hogs. Panther telemetry data collected by Land et al. indicates that panthers use agricultural fields (primarily croplands and citrus groves) both during the day and even more so at night, albeit not as high as some other areas37. Therefore, the Conservancy urges the Hydrological and Vegetation /Habitat Scores to be revised to scores commensurate with those reflected in the Landscape Suitability Index (LSI) and a revised assessment be done. Conclusions In conclusion, Mr. Tomasko himself said that "[t]here were no discrepancies in the mathematical calculation of impairment for the previously identified impaired water bodies by FDEP in Collier County "38. Therefore, there is no scientific nor legal basis for the County to dispute state and federal water quality regulations in the development of these plans. The Conservancy urges that these watershed plans be designed to adhere and comply with existing state and federal water quality policies and laws, as well as reflect the hydrologic and habitat values provided by agricultural lands — in order to ensure that the most accurate and effective plans are produced. Thank you for your time and consideration of our comments and recommendations. JJcejrre i`y nife Hec er ector of Natural Resource Policy 31 Multi- Species Recovery Plan for South Florida: Eastern Indigo Snake, North American Wild Turkey Management Plan, Accessed by htta:/ lwww. nwtf. org /NAWTMP /downloads/Literature /Eastern Indiao Snake .pd 31 IBID 33IBID 34 Kautz, et al. "How much is enough? Landscape -scale conservation for the Florida panther." Biological Conservation. 130. (2006): 118 -133. Print. " IBID 36 Logan , et al. "Florida Panther Habitat Preservation Plan: South Florida' Population." Florida Panther Interagency Committiee. (1993) 37 Land et al. Florida Panther Habitat Selection Analysis of Concurrent GPS and VHF Telemetry Data 38 Id 3. 09/10/2010 00:27 18774686770 PROCESSOR PAGE 01/04 CONSERVANCY OF SOUTHWEST FLORIDA CONCERNS WITH COLLIER COUNTY WATERSHED MANAGEMENT PLAN DEVELOPMENT IMPROPER ASSESSMENT OF WATERBODIES BELIEVED TO BE IMPAIRED DUE TO NATURAL CONDITIONS Waterbodies which receive loadings from natural sources should still be listed as "impaired" if they are not meeting water quality standards — unless the County demonstrates that there are no plausible or identified human pollutant sources and is granted a deviation (site specific alternative criteria or use attainability analysis) with the waterbody delisted as a result. EPA Regulatory guidance graphic below: c a 0 m h a O Making 303(d) Listing Decisions for Waterbodies with Naturally Occurring Pollutants loam. aaawv Cdwr+on w•.r.e w NaWd 9KU- gw.anwlM • "w�l Cw�•s� PmrlMan P+llP�tl�M V w -IM Y � a NW nl OMtlY"1" P,wl,lw, I,rw�l. Until such time that a deviation has been granted or the waterbody has been delisted, the legal requirement is to develop a TMDL to meet applicable water quality standards. The consultant for the Watershed Management Plan said that "[t]here were no discrepancies in the mathematical calculation of impairment for the previously identified impaired water bodies by FDEP in Collier County"'. The Watershed Management Plans should therefore comply with such regulatory requirements by utilizing EPA's Florida 303(d0 list Decision Documents as the appropriate list of impairments and waterbodies to address. 2. IMPROPER ASSESSMENT OF DISSOLVED OXYGEN (DO) IMPAIRMENTS • Watershed Plan Supporting Documents dispute FDEP verified Dissolved Oxygen (DO) impairments on the basis that a "natural" site within Collier County also exhibits levels of DO below state standards. • The reference site (WBID 3278G, Fakahatchee Strand) is listed by FDEP as impaired for DO as well. Though a relatively natural area, sampling is occurring in canals traversing through it - carrying pollution collected and conveyed from outside its borders. • Despite no causative pollutant being found for reference site, FDEP has not indicated that the impairment is thought to be from natural pollutant sources and it most likely is not since it carries runoff from upstream residential and agricultural areas. • "Natural pollution" is accounted for in setting standards which allow for some pollution, and in also subtracted them from existing pollutant loads in determining the appropriate Total Maximum Daily Load pollution reduction value for compliance -- so the county would never be required to remove natural pollutant loads anyhow. ' Tomasko, Dave PBS &J, Technical Memorandum to Mac Hatcher, PM Collier County. Re: Watershed Model Update and Plan Development Contract 08 -5122, PO 4500108318, Element 4 - Task 1 — Review of IWR data. August 24, 2009. 09/10/2010 00:27 18774686770 PROCESSOR C•Jnse,vancy Concaom ,with Watershed Managemert plan Development PAGE 02/04 Fags 2 o! 4. • Watershed Plan documents suggests seeking county -wide DD deviations. Again, these would only be granted, on a waterbody by waterbody basis, if proven that no anthropogenic pollutant sources are present and levels still exceeded current criteria. Doing so would be extremely expensive and risky. • Conducting water quality testing ourselves through -out the county and reviewing all other available water quality data, the Conservancy does not see the scientific justification required for such DO deviations to be granted throughout the County. ■ Additionally, Collier County staff indicated at the dAv August 4, SSACs Collier r DOC meeting that the County has no intentions of developing } ■ The Watershed Management Plans should therefore comply with current DO impairment determinations, in creating watershed management plans that reduce the anthropogenic inputs of pollutants which depress DO levels. 3. INAPPROPRIATE CONTESTING OF SALINITY CHARACTERIZATION ■ Watershed plan documents also dispute boundaries between fresh and marine waterbodies — which have different water quality criteria. ■ FDEP - with Collier County's participation - recently went through an extensive re- delineation process just a couple years ago to determine boundaries that better reflect waterbodies and their designated use. Two WBIDs assessed as potentially marine in the Memorandum, failed water quality standards for DO when using either the marine or freshwater criteria — so would not change whether they are out of compliance or not. • The Watershed Management Plans should therefore utilize FDEP basin boundaries, including for division between fresh and marine waterbodies — to ensure consistency and compatibility. Solid Waste Facilities in L:ouier N \Lp �i 0 b 0 ........... - ' o 2T GS 0 0;' oo. o /y bT ............. ..... - -.. • 327RR ...� t0 - C - Legend i .ndv. "id waste P0.6 -4a 4. INAPPROPRIATE NEGATION OF IRON IMPAIRMENTS ■ Watershed Management Plan document suggests iron impairments are natural. ■ Conservancy's cursory analysis shows at least one potential non - natural source of iron within WBIDs impaired for iron - active solid waste facilities (shown on the map to left). Unless there is sufficient scientific evidence that the iron exceedance is entirely natural and a SSAC is granted with the waterbody is delisted by FDEP — the legal requirement is to meet current state iron water quality standards. • The Watershed Management flans should instead work towards assessing measures necessary to meet state iron water quality standards. 09/10/2010 00:27 18774686770 PROCESSOR PAGE 03/04 Conservancy Conceits with Watershed Management Plan DevelopmOnt Page 3 of 4 5. NEED FOR SEPARATE eUT COMPATIBLE SMAPS AND WMPS FOR COLLIER COUNTY'S WATERSHEDS • Watershed Management Plans (WMP) cannot substitute Basin Management Action Plans (BMAP) As WMPs will not adequately address water quality impairments based FDEP's impaired waters lists and Total Maximum Daily Loads (TMDLs) already completed within the county. As WMPs are not designed to isolate and assign specific pollutant 1 wasteload allocation reductions as a BMAP As WMPs are not being developed across geopolitical boundaries with the participation of all affected stakeholder groups who would be required for their successful implementation - such as the County, City of Naples, City of Marco, etc. in the case of Collier County. As WMPs do not have a compliance instrument with the same level of binding and enforceable measures necessary to fulfill the requirements he BMAP, which are adopted by Secretarial Order to be enforceable. * The Watershed Management Plans should be focused as on evaluating land use changes to reduce impacts to water resources with regards to wetlands loss, or changes in quantity, timing or distribution of flows — as well as be complimentary to existing and future TMDLs and BMAPs. 6. AGRICULTURAL LANDS SHOULD NOT BE ASSIGNED A ZERO IN RESOURCE VALUE EVALUATIONS. ■ The Watershed Management Plan documents assign a score of "0" for Vegetation /Habitat and Hydrological areas where a natural system has been converted to a developed land use class (e.g., agriculture, urban development, golf course, and pasture). • Conversion to agricultural uses does not eliminate hydrologic or habitat value, and agricultural /pasture lands provide more natural resource value than urban development. • The document is even internally inconsistent with regards to this in that assigns agriculture land cover types scores ranging from moderate to high value in its Landscape Suitability Index (LSI)_ • Agricultural lands provide important functions to Florida's hydrology by sometimes acting as temporary water retention areas in addition to providing nesting and foraging habitat, habitat for base prey populations, and necessary components of the life cycle for various wildlife species. With proper management, they can help replenish aquifers and filter nutrients before they enter other systems. • Agricultural lands support some of the states most imperiled species including, the crested caracara (Cascara plancus), southeastern American kestrel (Falco sparverius paulus), burrowing owl (Athene cunicularla floridana), wood stork (Mycteria americana), gopher tortoise (Gopherus polyphemus), eastern indigo snake (Drymarchon coupen), and the Florida panther (Puma concolor cotyl) as well as a vast array of other wildlife . 2 2 Refer to Conservancy's extensive comment letter to Mac Hatcher of Collier County for more supporting documentation on the use of agricultural lands by wildlife. 09/10/2010 00:27 18774686770 PROCESSOR ConSarvency Concen's w r"t VVefershed M8negBrnent Plan De!lelopmeni Figure: Number of Wading Birds Observed during Agricultural Area Surveys, m W, U rU - '11 _ f+i u m V% r M 9 Birds 4D k3p rerve~d during SurNrt -- zoo 250 200 100 50 - O "I0 /3O l - Y 11/30 1 2 /3Q 1/30 Man U-1 2J28 3/30 PAGE 04/04 Pege 4 of 4 Table 1. Wading birds observed during surveys of agricultural canals listed by species, number and % of intal hirric ohserved_ and listed status by state and federal agencies4. Species Count - Listed Status Agency I Cattle Egret Great Egret 410 25 392 Snowy Egret _ 238 1s Species of Special Concern i State Stork 193 12 Endangered ^�� State, Fed. White Ibis 172 11 of Special Concern Species _•• State Little Blue Heron 114 S p ecies of Specia l al onCe r n C-,� - — _ ...._ - - - t—ate i Great Blue Heron 41 1 3 i ( j 1 Species of Special Concern —~ -- Roseate Spoonbill 19 ( State, Fed. Under Review I Tri- colored Heron 19 ! 1 I Sandhill Crane 13 1 Threatened State Mossy ityls 8 0 Green- backed Heron 4 0 Total ; i 1619 100 J •, ■ The Watershed Management Plans should include Hydrological and Vegetation /Habitat Scores for agricultural lands that are commensurate with those reflected in the Landscape Suitability Index (LSI). CONCLUSION ■ The Conservancy urges that these watershed plans be designed to adhere and comply with existing state and federal water quality policies and laws, as well as reflect the hydrologic and habitat values provided by agricultural lands in order to ensure that the most accurate and effective plans are produced. 3 Main, Martin, and Vavrina, Charles. "Wading birds and agriculture in Southwest Florida" University of Florida IFAS Extension. 2009. Web, 16 Aug 2010. s http:// edis. ifas. ufl.edu/pdffiles/UWIUW13900.pdf�-. 4 IBID