(TMDL) Assessment for the Neshaminy Creek Watershed in Southeast Pennsylvania Table of Contents

Home , Ironworks Creek, Little Neshaminy Creek, Neshaminy Creek, Pennsylvania Route 309, Pennsylvania Route 611

Total Maximum Daily Load (TMDL) Assessment for the Neshaminy Creek Watershed in Southeast Pennsylvania Table of Contents

Page

A1.0 OVERVIEW………………………………………………………………………… 1

A2.0 HYDROLOGIC /WATER QUALITY MODELING………………………………. 8

A2.1 Data Compilation and Model Overview………………………………………… 8 A2.2 GIS-Based Derivation of Input Data……………………………………………. 10 A2.3 Watershed Model Calibration…………………………………………………… 10 A2.4 Relationship Between Dissolved Oxygen Levels, Nutrient Loads and Organic Enrichment…………………………………………………………….. 16

B. POINT SOURCE TMDLs FOR THE ENTIRE NESHAMINY CREEK WATERSHED (Executive Summary)……….……….………………………. 18

B1.0 INTRODUCTION.…………………………………………………………………. 19

B2.0 EVALUATION OF POINT SOURCE LOADS……………………………………. 20

B3.0 REACH BY REACH ASSESSMENT……………………………………………… 27

B3.1 Cooks Run (482A)………………………………………………………………. 27 B3.2 Little Neshaminy Creek (980629-1342-GLW)………………..………………… 27 B3.3 Mill Creek (20010417-1342-GLW)…………………………………………….. 29 B3.4 Neshaminy Creek (467)………………………………………….……………… 30 B3.5 Neshaminy Creek (980515-1347-GLW)..…………………….………….……… 32 B3.6 Neshaminy Creek (980609-1259-GLW)………………………..………………. 32 B3.7 Park Creek (980622-1146-GLW)………………………………..……………… 34 B3.8 Park Creek (980622-1147-GLW)……………………………..………………… 35 B3.9 West Branch Neshaminy Creek (492)……………………………..……………. 35 B3.10 West Branch Neshaminy Creek (980202-1043-GLW)…………..……………. 36 B3.11 West Branch Neshaminy Creek (980205-1330-GLW)………………………… 37 B3.12 West Branch Neshaminy Creek (980205-1333-GLW)………………………… 38

C. LITTLE NESHAMINY CREEK……………………………………………………… 40

D. LAKE GALENA……………………………………………………………………… 58

E. PINE RUN…………………………………………………………………………….. 78

i Table of Contents (cont.)

Page

F. SUB-BASIN #1 OF WEST BRANCH NESHAMINY CREEK……………………… 94

G. SUB-BASIN #2 OF WEST BRANCH NESHAMINY CREEK……………………… 107

H. SUB-BASIN #3 OF WEST BRANCH NESHAMINY CREEK…………………….. 121

I. SUB-BASIN #4 OF WEST BRANCH NESHAMINY CREEK……………………… 138

J. COOKS RUN………………………………………………………………………….. 155

K. NESHAMINY CREEK TRIBUTARY #1…………………………………………….. 168

L. NESHAMINY CREEK TRIBUTARY #2…………………………………………….. 182

M. NESHAMINY CREEK TRIBUTARY #3……………………………………………. 196

N. NESHAMINY CREEK TRIBUTARY #4……………………………………………. 211

O. MILL CREEK SUB-BASIN #1………………………………………………………. 213

P. MILL CREEK SUB-BASIN #2……………………………………………………….. 227

Q. CORE CREEK………………………………………………………………………… 241

R. NESHAMINY CREEK SOUTH #1…………………………………………………… 255

S. NESHAMINY CREEK SOUTH #2…………………………………………………… 269

T. NESHAMINY CREEK SOUTH #3…………………………………………………… 271

Appendix A. GWLF Users Manual Appendix B. Streambank Erosion Calculations within AVGWLF Appendix C. TMDL Information Sheet Appendix D. TMDL Methodology Used in Pennsylvania Appendix E. Proposed Comprehensive Stormwater Management Policy Appendix F. AVGWLF Modeling Output for Sub-watersheds Appendix G. EMPR Results for Selected Watersheds

ii A1.0 OVERVIEW

This comprehensive document has been prepared to address all of the impaired stream segments contained within the Neshaminy Creek watershed located in Bucks and Montgomery Counties, Pennsylvania (see Figure A1). This urbanized watershed, which comprises about 148,679 acres, is located primarily in the Piedmont physiographic province to the north of Philadelphia. In terms of land use/cover, it is approximately 24% developed, 38% agriculture, 36% wooded, and 2% other (primarily wetland and disturbed), and has approximately 418.3 miles of streams. Since 1996, 203.3 miles of these streams (about 48.6%) have been included on Pennsylvania’s 303(d) list of streams having aquatic life use impairments. A complete listing of the segments impacted, along with their respective sources and causes of impairment, is given in Table A1.

Within this document, the different sections have been organized to address impairments by geographic unit (i.e., sub-watershed) and by sources and causes of impairment. The first section addresses municipal point source impairments within the entire Neshaminy Creek watershed. Subsequent sections address the different sub-watersheds listed in Table A1 (which are also depicted in Figure A2). To be consistent with other TMDL documents prepared by the Pennsylvania Department of Environmental Protection (PaDEP), an Executive Summary has also been prepared for each section.

Figure A1. Location of Neshaminy Creek watershed.

1 Table A1.

Aquatic Life Use 303(d) List of Streams and Sources of Impairments Year Targeted Data Source Source Cause Priority Listed for TMDL ------State Water Plan:02F Named Streams: 02F - Cooks Run Segment ID: 482 Miles Assessed: 1.4 Surface Water Monitoring Program Urban Runoff/Storm Sewers Cause Unknown Low 1996 Surface Water Monitoring Program Urban Runoff/Storm Sewers Nutrients Medium 1996 Segment ID: 482A Miles Assessed: 3.2 Surface Water Monitoring Program Municipal Point Source Nutrients Medium 1996 2003 Surface Water Monitoring Program Urban Runoff/Storm Sewers Cause Unknown Low 1996 2003 Surface Water Monitoring Program Urban Runoff/Storm Sewers Nutrients Medium 1996 2003

Named Streams: 02F - Core Creek Segment ID: 980602-0954-GLW Miles Assessed: 15.8 Surface Water Assessment Program Agriculture Siltation Medium 2002

Named Streams: 02F - Little Neshaminy Creek Segment ID: 980616-1108-GLW Miles Assessed: 5.5 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 1998 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 1998 Segment ID: 980616-1316-GLW Miles Assessed: 6.2 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Segment ID: 980629-1341-GLW Miles Assessed: 10.5 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 1996 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 1996 Segment ID: 980629-1342-GLW Miles Assessed: 15.7 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 1996 2003 Surface Water Assessment Program Municipal Point Source Nutrients Medium 1996 2003 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 1996 2003

Named Streams: 02F - Mill Creek Segment ID: 20000525-1017-GLW Miles Assessed: 0.8 Surface Water Assessment Program Surface Mining Siltation Medium 2002 Surface Water Assessment Program Surface Mining Flow Alterations Low 2002 Segment ID: 20010417-1342-GLW Miles Assessed: 1.8 Surface Water Assessment Program Municipal Point Source Nutrients Medium 2002 Segment ID: 20010426-1512-GLW Miles Assessed: 2.2 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Segment ID: 980609-1425-GLW Miles Assessed: 3.9 Surface Water Assessment Program Small Residential Runoff Flow Alterations Low 2002 Surface Water Assessment Program Small Residential Runoff Siltation Medium 2002

Named Streams: 02F - Park Creek Segment ID: 20010510-1303-GLW Miles Assessed: 0.2 Surface Water Assessment Program Other Oil and Grease Medium 1996 Segment ID: 20010511-1045-GLW Miles Assessed: 1.7 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Segment ID: 980622-1146-GLW Miles Assessed: 6.2 Surface Water Assessment Program Municipal Point Source Nutrients Medium 1996 2003 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 1996 2003 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 1996 2003 Segment ID: 980622-1147-GLW Miles Assessed: 1.2 Surface Water Assessment Program Municipal Point Source Nutrients Medium 1996 2003 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 1996 2003 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 1996 2003

2 Table A1. (cont.)

Aquatic Life Use 303(d) List of Streams and Sources of Impairments Year Targeted Data Source Source Cause Priority Listed for TMDL ------Named Streams: 02F - Neshaminy Creek Segment ID: 20010426-1235-GLW Miles Assessed: 5.2 Surface Water Assessment Program Land Development Siltation Medium 2002 Surface Water Assessment Program Agriculture Siltation Medium 2002 Segment ID: 20010525-1250-GLW Miles Assessed: 7.6 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Segment ID: 20010525-1330-GLW Miles Assessed: 5.4 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Segment ID: 467 Miles Assessed: 36.5 Surface Water Assessment Program Municipal Point Source Nutrients Medium 1996 2003 Surface Water Assessment Program Other Cause Unknown Low 1996 2003 Surface Water Assessment Program Municipal Point Source Organic Enrichment/Low D.O. Medium 1996 2003 Surface Water Assessment Program Municipal Point Source pH Medium 1996 2003 Segment ID: 980202-1313-GLW Miles Assessed: 3.3 Surface Water Assessment Program Agriculture Excessive Algal Growth Medium 1996 2003 Surface Water Assessment Program Agriculture Siltation Medium 1996 2003 Surface Water Assessment Program Construction Siltation Medium 1996 2003 Segment ID: 980205-1211-GLW Miles Assessed: 1.5 Surface Water Assessment Program Land Development Water/Flow Variability Low 2002 Surface Water Assessment Program Urban Runoff/Storm Sewers Flow Alterations Low 2002 Segment ID: 980427-0945-GLW Miles Assessed: 4.6 Surface Water Assessment Program Land Development Siltation Medium 2002 Segment ID: 980514-1004-GLW Miles Assessed: 1.5 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Segment ID: 980515-1347-GLW Miles Assessed: 2 Surface Water Assessment Program Municipal Point Source Water/Flow Variability Low 1996 2003 Surface Water Assessment Program Construction Siltation Medium 1996 2003 Segment ID: 980515-1348-GLW Miles Assessed: 1.3 Surface Water Assessment Program Construction Siltation Medium 2002 Segment ID: 980609-1258-GLW Miles Assessed: 3.2 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 Segment ID: 980609-1259-GLW Miles Assessed: 3.6 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 Surface Water Assessment Program Municipal Point Source Nutrients Medium 2002 Segment ID: 980713-1351-GLW Miles Assessed: 9.9 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 2003

Named Streams: 02F - Neshaminy Creek, West Branch Neshaminy Creek Segment ID: 980202-1040-GLW Miles Assessed: 8.5 Surface Water Assessment Program Agriculture Siltation Medium 1996 2003 Surface Water Assessment Program Land Development Water/Flow Variability Low 1996 2003 Surface Water Assessment Program Agriculture Excessive Algal Growth Medium 1996 2003

Named Streams: 02F - North Branch Neshaminy Creek Segment ID: 980210-1123-GLW Miles Assessed: 3.3 Surface Water Assessment Program Upstream Impoundment Siltation Medium 2002 2003 Surface Water Assessment Program Upstream Impoundment Water/Flow Variability Low 2002 2003

Named Streams: 02F - Pine Run Segment ID: 980210-1240-GLW Miles Assessed: 2.1 Surface Water Assessment Program Land Development Siltation Medium 2002 Surface Water Assessment Program Upstream Impoundment Excessive Algal Growth Medium 2002 Segment ID: 980210-1242-GLW Miles Assessed: 1.3 Surface Water Assessment Program Land Development Siltation Medium 2002 Segment ID: 980211-1241-GLW Miles Assessed: 5 Surface Water Assessment Program Land Development Siltation Medium 2002

3 Table A1. (cont.)

Aquatic Life Use 303(d) List of Streams and Sources of Impairments Year Targeted Data Source Source Cause Priority Listed for TMDL ------Named Streams: 02F - West Branch Neshaminy Creek Segment ID: 492 Miles Assessed: 0.1 Surface Water Assessment Program Municipal Point Source Nutrients Medium 1996 2003 Segment ID: 980202-1043-GLW Miles Assessed: 7.7 Surface Water Assessment Program Municipal Point Source Excessive Algal Growth Medium 1996 2003 Surface Water Assessment Program Land Development Water/Flow Variability Low 1996 2003 Surface Water Assessment Program Agriculture Excessive Algal Growth Medium 1996 2003 Segment ID: 980202-1441-GLW Miles Assessed: 4.9 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 Surface Water Assessment Program Land Development Siltation Medium 2002 Segment ID: 980205-1330-GLW Miles Assessed: 1.8 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 2002 Surface Water Assessment Program Municipal Point Source Flow Alterations Low 2002 Surface Water Assessment Program Land Development Water/Flow Variability Low 2002 Segment ID: 980205-1333-GLW Miles Assessed: 1.6 Surface Water Assessment Program Municipal Point Source Flow Alterations Low 1996 2003 Surface Water Assessment Program Land Development Water/Flow Variability Low 1996 2003 Surface Water Assessment Program Urban Runoff/Storm Sewers Siltation Medium 1996 2003 Segment ID: 980205-1430-GLW Miles Assessed: 5.1 Surface Water Assessment Program Urban Runoff/Storm Sewers Water/Flow Variability Low 2002 Surface Water Assessment Program Agriculture Excessive Algal Growth Medium 2002 Surface Water Assessment Program Land Development Water/Flow Variability Low 2002 Surface Water Assessment Program Agriculture Siltation Medium 2002

Figure A2. Location of sub-watersheds used for TMDL assessments.

4 As reported in greater detail in later sections, nutrients from municipal sources have been listed as the cause of impairment for a number of stream segments within the Neshaminy Creek watershed. The original 303(d) listings of most of these streams (ca. 1996) were based on a 1988 SERA survey, which is done to determine if nutrient loadings are impacting specific streams. In this case, nutrients from municipal treatment facilities were documented as causing organic enrichment problems in many streams within the watershed. As a result of this and water quality modeling studies conducted by USEPA Region 3 in 1982, all of the treatment facilities had nutrient limits included in their NPDES permits. Nutrient limits are applied April 1 through October 31.

Many changes have taken place in the watershed over the past 20 years. Within the past 10 years, municipal treatment plants have been upgraded to provide tertiary treatment, and the closing of some facilities and transference of waste flows to regional treatment facilities has occurred. There has also been a tremendous amount of growth. The amount of developed land, for example, has increased by about 20 percent in the watershed over the last decade. This represents drastic changes to the landscape and has a great effect on precipitation-driven runoff patterns.

It is very difficult to ascertain if excess nutrient loading is still a problem in these waters, and being overshadowed by the changes in runoff patterns, or it is not a problem at all. Currently the largest problem in the watershed is the increase in hydraulic energy of the stream flows that are causing sedimentation problems and stream bank erosion. With the increase in development comes an increase in the amount of impervious surface (pavement) which causes the water to run off the landscape at a much faster rate with greater force. This causes the stream banks to erode and deposit sediment in the stream channel. This sediment covers the stream bottom and reduces the habitat for aquatic organisms. The other effect of the increased hydraulic energy (flashy flows) is that these conditions occur more frequently with smaller rainfall events and can have enough force to essentially wash away the aquatic community.

For these reasons most of the TMDLs developed for sub-areas of the Neshaminy Creek watershed are focused on sediment control. Nutrients from overland sources and stream bank erosion are also addressed as needed. However, at this time it is not our intent to ask municipal treatment facilities to make any further reductions to existing permitted loads. Rather, it is recommended that municipal waste load allocations be equivalent to their respective permit limits.

A2.0 HYDROLOGIC/WATER QUALITY MODELING

A2.1 Data Compilation and Model Overview

The TMDLs within the Neshaminy Creek watershed were for the most part developed using the Generalized Watershed Loading Function (GWLF) model. The GWLF model provides the ability to simulate runoff, sediment, and nutrient (N and P) loadings from watershed given variable-size source areas (e.g., agricultural, forested, and developed land). It also has algorithms for calculating septic system loads, and allows for the inclusion of point source

5 discharge data. It is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values.

GWLF is a combined distributed/lumped parameter watershed model. For surface loading, it is distributed in the sense that it allows multiple land use/cover scenarios. Each area is assumed to be homogenous in regard to various attributes considered by the model. Additionally, the model does not spatially distribute the source areas, but aggregates the loads from each area into a watershed total. In other words, there is no spatial routing. For sub-surface loading, the model acts as a lumped parameter model using a water balance approach. No distinctly separate areas are considered for sub-surface flow contributions. Daily water balances are computed for an unsaturated zone as well as a saturated sub-surface zone, where infiltration is computed as the difference between precipitation and snowmelt minus surface runoff plus evapotranspiration.

GWLF models surface runoff using the Soil Conservation Service Curve Number (SCS-CN) approach with daily weather (temperature and precipitation) inputs. Erosion and sediment yield are estimated using monthly erosion calculations based on the Universal Soil Loss Equation (USLE) algorithm (with monthly rainfall-runoff coefficients) and a monthly composite of KLSCP values for each source area (e.g., land cover/soil type combination). The KLSCP factors are variables used in the calculations to depict changes in soil loss erosion (K), the length slope factor (LS) the vegetation cover factor (C) and conservation practices factor (P). A sediment delivery ratio based on watershed size and a transport capacity based on average daily runoff are applied to the calculated erosion to determine sediment yield for each source area. Surface nutrient losses are determined by applying dissolved N and P coefficients to surface runoff and a sediment coefficient to the yield portion for each agricultural source area. Point source discharges can also contribute to dissolved losses to the stream and are specified in terms of kilograms per month. Manured areas, as well as septic systems, can also be considered. Urban nutrient inputs are all assumed to be solid-phase, and the model uses an exponential accumulation and washoff function for these loadings. Sub-surface losses are calculated using dissolved N and P coefficients for shallow groundwater contributions to stream nutrient loads, and the sub-surface sub-model only considers a single, lumped-parameter contributing area. Evapotranspiration is determined using daily weather data and a cover factor dependent upon land use/cover type. Finally, a water balance is performed daily using supplied or computed precipitation, snowmelt, initial unsaturated zone storage, maximum available zone storage, and evapotranspiration values. All of the equations used by the model can be viewed in Appendix A, GWLF Users Manual.

In addition to the original model algorithms described above, a streambank erosion routine was also implemented as part of the present study. This routine is based on an approach in which monthly streambank erosion is estimated by first calculating a watershed-specific lateral erosion rate (LER) for streams within the watershed. After a value for LER has been computed, the total sediment load generated via streambank erosion is then calculated by multiplying the above erosion rate by the total length of streams in the watershed, the average streambank height, and the average soil bulk density. More information on the specific details of this approach is provided in Appendix B.

6 For execution, the model requires three separate input files containing transport-, nutrient-, and weather-related data. The transport (TRANSPRT.DAT) file defines the necessary parameters for each source area to be considered (e.g., area size, curve number, etc.) as well as global parameters (e.g., initial storage, sediment delivery ratio, etc.) that apply to all source areas. The nutrient (NUTRIENT.DAT) file specifies the various loading parameters for the different source areas identified (e.g., number of septic systems, urban source area accumulation rates, manure concentrations, etc.). The weather (WEATHER.DAT) file contains daily average temperature and total precipitation values for each year simulated.

A2.2 GIS-Based Derivation of Input Data

The primary sources of data for this analysis were geographic information system (GIS) formatted databases. A specially designed interface was prepared by the Environmental Resources Research Institute (ERRI) of the Pennsylvania State University in ArcView (GIS software) to generate the data needed to run the GWLF model, which was developed by Cornell University. The new version of this model has been named AVGWLF (ArcView Version of the Generalized Watershed Loading Function)

In using this interface, the user is prompted to identify required GIS files and to provide other information related to “non-spatial” model parameters (e.g., beginning and end of the growing season, the months during which manure is spread on agricultural land and the names of nearby weather stations). This information is subsequently used to automatically derive values for required model input parameters which are then written to the TRANSPRT.DAT, NUTRIENT.DAT and WEATHER.DAT input files needed to execute the GWLF model (see Appendix A). For use in Pennsylvania, AVGWLF has been linked with statewide GIS data layers such as land use/cover, soils, topography, and physiography; and includes location- specific default information such as background N and P concentrations and cropping practices. Complete GWLF-formatted weather files are also included for eighty weather stations around the state. Table A2 lists the statewide GIS data sets and provides explanation of how they were used for development of the input files for the GWLF model.

A2.3 Watershed Model Calibration

In the Neshaminy watershed as a whole, nutrient and sediment loads can originate from a variety of sources including upland erosion and runoff (particularly from agricultural activities and ongoing residential development), sub-surface flow (primarily from agricultural areas and septic systems), streambank erosion, and point source discharges. To adequately assess the various contributions and their resultant impacts on specific stream segments, AVGWLF was calibrated for this watershed using available water quality sample data for the period 4/93 to 3/99. In the calibration process, various model parameters were “fine-tuned” to more accurately depict critical model parameters such as sub-surface concentrations of nitrogen and phosphorus (GWN and GWP, respectively), as well as the “cover factor” (i.e., USLE “c” factor) used in the soil loss sub-model of GWLF for the “cropland” category. Actual point source discharge data were also compiled from PaDEP monthly discharge reports to assist in the calibration process. In addition, the “a” factor used within AVGWLF to determine the lateral erosion rate (LER) for

7 streambank erosion was based on the calibrated value previously derived by Evans (2002) for this particular watershed (in this case, 1.796 x 10-4).

Table A2. Description of GIS-based model input.

GIS Layer Description

Censustr Coverage of Census data including information on individual homes septic systems. The attributeusew_sept includes data on conventional systems, and sew_other provides data on short circuiting and other systems. County The County boundaries coverage lists data on conservation practices that provides C and P values in the Universal Soil Loss Equation (USLE). Gwnback A grid of background concentrations of N in groundwater derived from water well sampling. Landuse5 Grid of the MRLC that has been reclassified into five categories. This is used primarily as a background. Majored Coverage of major roads. Used for reconnaissance of a watershed. MCD Minor civil divisions (boroughs, townships and cities). Npdespts A coverage of permitted point discharges. Provides background information and cross check for the point source coverage. Padem 100 meter digital elevation model. This used to calculate landslope and slope length. Palumrlc A satellite image derived land cover grid which is classified into 15 different landcover categories. This dataset provides landcover loading rate for the different categories in the model. Pasingle The 1:24,000 scale single line stream coverage of Pennsylvania. Provides a complete network of streams with coded stream segments. Physprov A shapefile of physiographic provinces. Attributes rain_cool and rain_warm are used to set recession coefficient Pointsrc Major point source discharges with permitted N and P loads. Refwater Shapefile of reference watersheds for which nutrient and sediment loads have been calculated. Soilphos A grid of soil phosphorous loads which has been generated from soil sample data. Used to help set phosphorus and sediment values. Smallsheds A coverage of watersheds derived at 1:24,000 scale. This coverage is used with the stream network to delineate the desired level watershed. Statsgo A shapefile of generalized soil boundaries. The attribute mu_k sets the k factor in the USLE. The attribute mu_awc is the unsaturated available capacity., and the muhsg_dom is used with landuse cover to derive curve numbers. Strm305 A coverage of stream water quality as reported in the Pennsylvania’s 305(b) report. Current status of assessed streams. Surfgeol A shapefile of the surface geology used to compare watersheds of similar qualities. T9sheds Data derived from a DEP study conducted at PSU with N and P loads. Zipcode A coverage of animal densities. Attribute aeu_acre helps estimate N & P concentrations in runoff in agricultural lands and over manured areas. Weather Files Historical weather files for stations around Pennsylvania to simulate flow.

8 Figure A3 shows that portion of the watershed used in the calibration process and the location of the WQN sampling station for which historical nutrient concentration and total suspended solids data were obtained in order to compute observed loads. Also shown in this figure are the locations of the point sources that discharge nitrogen and/or phosphorus to surface waters in the watershed. For calibration purposes, current discharge data for the period 1998-2000 were used to represent point source loads in the watershed. Figures A4 and A5 show the nutrient- and transport-related model parameters for GWLF derived as a result of the calibration effort. For illustration purposes, the calibration results for total phosphorus are graphically depicted in Figure A6. The mean annual nutrient and sediment loads predicted by GWLF for the calibration period are shown in Figure A7, and Table A3 provides a summary of the nutrient and sediment load contributions within the watershed by source.

Figure A3. Locations of sampling station and point source discharges within the Neshaminy Creek watershed.

9 As alluded to above, model calibration was performed in this watershed for the specific purposes of establishing loading rates from various sources more precisely than can be done solely using default data sets and parameter settings within AVGWLF. As with any model calibration, this can more easily be accomplished when existing water quality sampling and point source discharge data are available, as was the case for this study. After successful model calibration, the newly-derived parameter estimates can, in turn, be used to more precisely estimate model parameter values used for smaller sub-basins contained within the larger watershed as well. For example, estimates for the critical GWLF model parameters cited earlier that were generated for this basin using default AVGWLF algorithms and via calibration are shown in Table A4. Also shown are the percent changes in the default values relative to the final calibrated values. Under the assumption that these adjustments provided reasonable results during the calibration step for the larger basin, the factors shown were used to make similar parameter adjustments in sub-areas as well for the purposes of supporting the TMDL assessments described in other sections of this report.

Figure A4. Calibrated transport file (transprt.dat) for GWLF.

Figure A5. Calibrated nutrient file (nutrient.dat) for GWLF.

Observed vs. Predicted Phosphorus Loads (kg) 15000

Observed 12000 Predicted

9000

6000

3000

3 4 5 6 7 8 9 9 9 9 96 9 9 9 r-94 r-95 t- r-9 t- r- t-97 r- t-98 c c c c pr- Apr-93 Oct- Ap Oct- Ap O Ap O Ap O Ap O A

Figure A6. Comparison of observed vs. simulated total phosphorus loads for the Neshaminy Creek watershed.

Figure A7. Calculated mean annual loads based on final calibrated GWLF model run.

Table A3. Relative contribution of nutrient and sediment loads by source.

Source Percent of N Percent of P Percent of Sediment

Upland Erosion/Runoff 14.6 25.2 8.0 Streambank Erosion 10.4 27.7 92.0 Groundwater 35.3 22.4 - Point Sources 31.8 24.1 - Septic Systems 7.9 0.6 -

12 Table A4. Adjustments made to critical GWLF model parameters.

Parameter Default Calibrated Adjustment

Sediment “a” factor 1.665 x 10-4 1.796 x 10-4 1.08 GWN 2.404 1.149 0.48 GWP 0.0405 0.0565 1.4 USLE “C” factor 0.42 0.21 0.5

A2.4 Relationship Between Dissolved Oxygen Levels, Nutrient Loads and Organic Enrichment

As indicated in Table A1, various streams within the Neshaminy Creek watershed have been listed as being impaired due to problems associated with dissolved oxygen levels, nutrient loads, and organic enrichment. In stream systems, elevated nutrient loads (nitrogen and phosphorus in particular) can lead to increased productivity of plants and other organisms (Novotny and Olem, 1994). Oxygen in water is used by plants (at night) and organisms in the stream. Excessive nutrient input can lead to elevated levels of productivity, which can subsequently lead to depressed dissolved oxygen levels when an abundance of aquatic life is drawing on a limited oxygen supply. Additional problems arise when these organisms die because the microbes that decompose this organic matter also consume large amounts of oxygen. A second effect of nitrogen (specifically ammonia) occurs when bacteria convert ammonia-nitrogen to nitrate- nitrogen. This process, called nitrification, also results in lower dissolved oxygen levels in streams.

Typically in aquatic ecosystems the quantities of trace elements are plentiful; however, nitrogen and phosphorus may be in short supply. The nutrient that is in the shortest supply is called the limiting nutrient because its relative quantity affects the rate of production (growth) of aquatic biomass. If the nutrient load to a water body can be reduced, the available pool of nutrients that can be utilized by plants and other organisms will be reduced and, in general, the total biomass can subsequently be decreased as well (Novotny and Olem, 1994). In most efforts to control eutrophication processes in water bodies, emphasis is placed on the limiting nutrient. This is not always the case, however. For example, if nityrogen is the limiting nutrient, it still may be more efficient to control phosphorus loads if the nitrogen originates from difficult to control sources such as nitrates in ground water.

13 In the case of the Neshaminy Creek watershed, the N/P ratio is approximately 13, which implies that phosphorus is the limiting nutrient. Therefore, it is expected that controlling the phosphorus load to streams in this watershed will limit plant growth and result in raising the dissolved oxygen level.

14 B. EXECUTIVE SUMMARY – Point Source TMDLs Within the Entire Neshaminy Creek Watershed

Based upon a 1988 SERA survey and additional field assessments completed under the State’s Unassesed Waters Program, 81.4 miles of the 418.3 stream miles in the watershed have been listed as being impacted by municipal point sources on the State’s current 303(d) List since 1996. However, it is questionable whether problems with point source discharges still exist in many cases since many of the “impaired” segments which appeared on the 303(d) list may no longer be impaired due to municipal treatment plant upgrades undertaken in the mid-1990s. To assess the likelihood of this possibility, an evaluation of point source loads from municipal wastewater facilities was made within the entire Neshaminy Creek watershed to evaluate changes in plant discharges and in-stream nutrient concentration over the last decade. Based on this evaluation, such upgrades appear to have reduced in-stream loads of nutrients (particularly phosphorus) over the last half-dozen years. Additionally, for the most part point source dischargers appear to be complying with limitations stated in their respective NPDES permits, and where occasional compliance problems have occurred, the frequency and magnitude of such events suggest minimal compliance problems with point source discharges in the watershed. Given this evidence, it is not PaDEP’s intent to ask municipal treatment facilities to make any further reductions to existing permitted loads. Rather, it is recommended that municipal wasteload allocations be equivalent to their respective permit limits.

15 B1.0 INTRODUCTION

This section addresses stream segments listed due to municipal point source discharges within the entire Neshaminy Creek watershed. Information pertaining to these segments is summarized in Table B1, and the segments themselves are also depicted in Figure B1. Based upon field assessments completed under the State’s Unassesed Waters Program, 81.4 miles of the 418.3 stream miles in the watershed have been listed as being impaired by municipal point sources on the State’s current 303(d) List since 1996. However, it is questionable whether problems with point source discharges still exist in many cases since many of the “impaired” segments which appeared on the 1996 list may no longer be impaired due to municipal treatment plant upgrades undertaken in the mid-1990s. In a modeling study conducted by the U.S. EPA in the early 1980s (U.S. EPA, 1982), it was determined that Neshaminy Creek was experiencing dissolved oxygen and organic enrichment problems due to excessive nutrient loads from municipal wastewater treatment plants in the watershed. Two of the recommendations of this report for eliminating the above problems were: 1) that municipal treatment plants install advanced treatment to remove excess phosphorus, and 2) that DEP implement a phosphorus discharge limit of 2 mg/l during low-flow periods for municipal facilities. These recommendations have since been implemented, and as described in the next section, such upgrades and phosphorus discharge limits appear to have reduced in-stream loads of nutrients (particularly phosphorus) over the last half-dozen years. Also, as described later, most point source dischargers appear to be complying with limitations stated in their respective NPDES permits, and where occasional compliance problems have occurred, the frequency and magnitude of such events suggest minimal compliance problems with point source discharges in the watershed.

Table B1. Segments impaired by municipal point source discharges as described on the State’s current 303(d) list.

Stream Segment ID Cause Year Listed

Cooks Run 482A Nutrients 1996 Little Neshaminy Cr. 980629-1342-GLW Nutrients 1996 Mill Creek 20010417-1342-GLW Nutrients 2002 Neshaminy Creek* 467 Nutrients 1996 Neshaminy Creek* 467 Organic Enrichment/Low DO 1996 Neshaminy Creek* 467 PH 1996 Neshaminy Creek 980515-1347-GLW Water/Flow Variability 1996 Neshaminy Creek 980609-1259-GLW Nutrients 2002 Park Creek 980622-1146-GLW Nutrients 1996 Park Creek 980622-1147-GLW Nutrients 1996 W. Branch, Nesh. Cr. 492 Nutrients 1996 W. Branch, Nesh. Cr. 980202-1043-GLW Excessive Algal Growth 1996 W. Branch, Nesh. Cr. 980205-1330-GLW Flow Alterations 2002 W. Branch, Nesh. Cr. 980205-1333-GLW Flow Alterations 1996

* Note: Same segment is listed for three different impairment causes

Figure B1. Map depicting location of streams impaired (Y) or not impaired (N) by point source discharges as noted on DEP’s current 303(d) list.

B2.0 EVALUATION OF POINT SOURCE LOADS

For the purposes of this TMDL assessment, it has been assumed (at least theoretically) that potential problems related to point source discharges are adequately being addressed by the NPDES permitting process in Pennsylvania. That is to say, if a given facility is discharging nutrients at or below it’s permitted limit, then further reductions need not be specified as part of the TMDL load allocation process. However, it is possible that some dischargers may not be operating within specified limits, and that in these cases, an explicit waste load allocation may be justified.

To determine whether potential compliance problems may exist, an assessment of reported versus permitted nutrient loads for those facilities located in the Neshaminy Creek watershed (see Figure B1) was undertaken. Included in Tables B2 and B3 are permitted loads for these

17 facilities as well as loads computed using monthly discharge data from the periods “1988” (generally between 1987-1989) and “1999” (generally between 1998-2000). To estimate “permitted” loads for total N and total P, the stated NPDES permit loads (i.e., flows and concentrations) for each facility for each time period were used. For total N, this varied between 11-20 mg/l depending on the size of the facility. For total P, a permitted load of 2 mg/l was used. Although the “permitted” loads were calculated in this fashion, they in theory could be higher since the permits only address flows and concentrations during critical dry periods (in this case, April 1 through October 31). For the purposes of this exercise, however, the stated permitted loads were applied to the entire year.

As shown in Table B2, the permitted and reported loads for total nitrogen for the 1988 period were 404,837 and 329,281 kilograms per year, respectively. For this same period, the permitted and reported loads for total phosphorus were 63,291 and 67,413 kilograms per year, respectively. During the 1999 period, the permitted and reported loads for total nitrogen were 379,635 and 336,073 kilograms per year, respectively; and the permitted and reported loads for total phosphorus were 60,104 and 26,037 kilograms per year, respectively (see Table B3). (Note: The change in permitted loads for both nutrients between the two time periods is due to the closure of several facilities and the consolidation of waste flows into regional facilities).

As illustrated by the two tables, the reported nitrogen loads have been consistently below permitted loads, and have increased very slightly since the 1988 time period. During the 1988 period, reported phosphorus loads were slightly higher than permitted loads. However, since that time, phosphorus discharges appear to have been significantly reduced to the point where reported loads are less than half those allowed by NPDES permits. This reduction appears to be mirrored by drops in mean monthly total P concentrations observed at DEP’s water quality monitoring station shown (see Figure A3). As shown in Figure B2, the mean monthly total P concentration at the end of 1988 appeared to be around 0.25 mg/l. By the end of 1999, the mean monthly concentration had apparently dropped to about 0.16 mg/l for a reduction of about 36%. This seems to strongly suggest that continuing point source reductions are having an effect on in- stream phosphorus concentrations. Mean total nitrogen concentrations have also decreased slightly at this station for the same time period (see Figure B3). Similarly, dissolved oxygen concentrations have been dropping slightly as well (see Figure B4), which is not a positive trend if it continues too long into the future. However, even with this drop, recorded concentrations are still consistently above 8.0 mg/l, which suggests relatively good stream water quality in general.

Additionally, a review of recent monthly discharge reports compiled by PaDEP was completed to determine if permitted facilities were in compliance with effluent limitations stated in their respective NPDES permits. More specifically, an evaluation of all facilities within the Neshaminy Creek watershed was made to determine if any facility had exceeded effluent limitations for various pollutant parameters, including ammonia (NH3-N), nitrite plus nitrate (NO2-N + NO3-N), phosphate phosphorus (PHOS-P), and dissolved oxygen (DO) during the time period reviewed (i.e., 1998-2000). The results of this review are provided in Table B4.

Table B2. Estimated point source discharges for the period 1988.

Table B3. Estimated point source discharges for the period 1999.

20 Total P Concentration 0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

Jan-89 Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00

Figure B2. Trend in total P concentration at DEP monitoring station (in mg/l).

Total N Concentration 6

Jan-89 Jan-90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00

Figure B3. Trend in total N concentration at DEP monitoring station (in mg/l).

Dissolved Oxygen 20

an-88 an-89 an-90 an-91 an-92 an-93 an-94 an-95 an-96 an-97 an-98 an-99 J J J J J J J J J J J J

Figure B4. Trend in dissolved oxygen concentration at DEP monitoring station (in mg/l).

In this table, the values recorded indicate the number of times the permitted limit for a given parameter was exceeded based on monthly observations. During this period, effluent limitations collectively were exceeded 67 times for the four pollutant parameters evaluated. While at first glance this number seems high, this number is actually relatively low given the potential number of times these limits could have been exceeded. For example, given the number of facilities, number of parameters, and the time period reviewed (1998-2000), the number of potential observations is approximately 3744 (26 facilities x 12 months x 3 years x 4 parameters). Under this scenario, effluent limitations for the stated parameters were collectively exceeded less than 2 % of the time (67/3744 = 0.018). (Note: This percentage is likely somewhat higher since concentrations for two of the parameters [i.e, NO2-N + NO3-N and P] are only reported for “critical periods”, and hence are not reported for all 12 months of the year). In general, this implies that point source discharges may not be significant sources of water quality problems within the Neshaminy watershed as a whole. However, it is possible that localized problems related to specific discharges are occurring at isolated areas. A more comprehensive evaluation of potential problems on a reach by reach basis is presented in the following section.

Table B4. Number of times stated permit limitation has been exceeded during the period 1998-2000 for selected pollutant parameters.

Name Permit No. NH3-N NO2+NO3-N PHOS-P DO

Boro Lansdale PA0026182 Hatfield Township PA0026247 10 Chalfont/New Britain Twp PA0025917 1 2 1 Boro Doylestown-Green PA0021181 2 Boro Doylestown-Harvey PA0021172 1 Doylestown Twp.-Kings Pl PA0051250 11 3 1 Paul M. Weisser PA0054992 Continental Care Center PA0052761 2 Andrew Azzara PA0055166 Warwick W&S, Inc. PA0050148 2 1 3 Gynmount Farms PA0054798 3 3 Montgomery Sewer Co. PA0052094 1 North Penn School Dist. PA0050881 English Village PA0050059 1 2 Willow Grove Naval Air PA0022411 Warminster Twp Mun Auth PA0026166 1 2 US Dept Navy – USNADC PA0022420 WR & JM Elsig PA0054879 Jehovah’s Witnesses PA0053384 Florence Coleman PA0052493 1 4 Horsham Twp. – Park Creek PA0051985 1 Mckee Group - Buckingham PA0053279 Montgomery Twp. - Eureka PA0053180 Warwick Twp. – Country Cr PA0056421 4 3 Warrington Twp. STP PA0056758 1 Buckingham Twp. STP PA0052353

Totals 20 13 16 18

23 B3.0 REACH BY REACH ASSESSMENT

As shown earlier in Table B1, twelve (12) stream segments (reaches) spanning six (6) streams have been included on DEP’s current 303(d) list as being impaired by municipal point source discharges in one way or another. Brief reviews of conditions within each of these segments vis-à-vis point source impacts are provided below.

B3.1 Cooks Run (482A)

This particular segment is listed as being impaired by nutrients. The point source located along this stream as shown in Figure B5 (NPDES No. PA0021172) discharges approximately 7087 kg/year of N and 930 kg/year of P. Having been listed for nutrients, the primary pollutant of concern here is phosphorus. From Tables B2 and B3, it can be seen that this facility consistently discharges loads below it’s permitted limit, and did not exceed established phosphorus limits at all during the period 1998-2000. However, during the “1988” time period reflected in Table B2, this facility appeared to be discharging about 5 times more phosphorus than it did during the “1999” time period. Since this facility was upgraded in the mid-1990s for phosphorus removal, it is quite likely that nutrient impairments were still evident during the time that the stream was initially surveyed and listed (ca. 1996). It is very probable that visible indications of improved stream health related to such reductions would have taken several years to manifest (and may still not be evident at present due to drought conditions that have existed in the area for several years). Consequently, it does not appear that further reductions in nutrients from this point source are required for this segment since this particular point source is operating within the limitations of it’s current NPDES permit. However, it would be prudent to assess this stream at a later date to verify if observed nutrient reductions have resulted in improved stream health. In this case, the recommended TMDL for this point source would be the load presently specified by its existing NPDES permit.

B3.2 Little Neshaminy Creek (980629-1342-GLW)

Similar to Cooks Run, this segment is also listed as being impaired by nutrients from municipal point sources. Figure B6 shows this segment in relation to the nine point source discharges that potentially affect it. This stream has five (5) point sources located directly on it (NPDES Nos. PA0026166, PA0050881, PA0054798, PA0052094, and PA0053180) and four located on tributaries that flow into it (NPDES Nos. PA0056421, PA0050059, PA0022411, and PA0051985). The last three of the tributary point sources are located on Park Creek, which is discussed again in a later section. Having been listed for nutrients, the primary pollutant of concern here is phosphorus. (Note: Three NPDES facilities listed in Table B3 that are in this watershed [Nos. PA0051985, PA0053180 and PA0056421] did not exist in 1988, and eight facilities shown in Table B2 that were in the watershed [Nos. PA0050083, PA0052841, PA0050253, PA0050695, PA0030767, PA0051144, PA0028479, and PA0022420] are no longer operational).

Figure B5. Cooks Run point source impacts.

Figure B6. Little Neshaminy Creek point source impacts.

25 During the time periods reflected in Tables B2 and B3, the actual nitrogen loads from the point sources in this watershed have increased slightly from 130,960 kg/yr in 1988 to 154,378 kg/yr in 1999. Both of these loads are above the calculated “permitted” loads of 126,236 for 1988 and 113,261 for 1999, and several facilities have exceeded permitted limitations for nitrogen-related parameters for the period 1998-2000 as shown in Table B4. However, as noted above, the critical nutrient here is phosphorus. From 1988 to 1999, the actual phosphorus loads dropped drastically from 31,618 kg/yr to 9,921 kg/yr. In 1988, the actual load was about 50% higher than the calculated permitted load of 20,236 kg/yr; but in 1999, the actual load was about half the permitted load of 18,908 kg/yr. Similar to Cooks Run, it is likely that plant upgrades in the mid-1990s did not yet have any noticeable effects on stream health that would have been observed during the field work conducted for the 1996 303(d) listing of these stream segments. Therefore, it does not appear that further reductions in nutrients from point sources adjacent to this segment are required since the point sources are operating within the limitations of current NPDES permits. Additionally, phosphorus loads from point sources appear to have decreased significantly over the last few years. However, it would be prudent to assess this stream at a later date to verify if observed nutrient reductions have resulted in improved stream health. In this case, the recommended TMDL for the point sources would be the cumulative load reflected in existing NPDES permits in this watershed.

B3.3 Mill Creek (20010417-1342-GLW)

This segment is about 1.8 miles long, and is listed as being impaired by nutrients from municipal point sources (see Figure B7). The one NPDES facility located along this reach is PA0056758. This facility went on-line in the early 1990s, and discharges relatively little in terms of nitrogen and phosphorus loads in comparison to other facilities in the Neshaminy Creek watershed (i.e., 1523 kg/yr of N and 155 kg/yr of P). This facility also had only one permit infraction for NH3-N during the 1998-2000 time period as shown in Table B4. Based on the 303(d) listing, it was thought to be having an effect on stream health. However, given that this particular point source is operating within the limitations of its current NPDES permit, it does not appear that further reductions in nutrients from this point source are warranted. Although, it would be prudent to assess this stream at a later date to verify if observed nutrient reductions have resulted in improved stream health. In this case, the recommended TMDL for this point source would be the load presently specified by its existing NPDES permit.

Figure B7. Mill Creek point source impacts.

B3.4 Neshaminy Creek (467)

This segment is the single longest listed impaired reach in the entire Neshaminy Creek watershed (36.5 miles), and was added to the 303(d) list in 1996 (see Figure B8). With respect to municipal sources, this stream segment was listed as being impaired by nutrients, organic enrichment/DO, and pH. Since this segment represents the main stem of Neshaminy Creek, and because all of the tributaries within the entire watershed flow into it at some point, it is difficult to isolate specific municipal point sources as being the principal sources of impairment. Rather, it is more correct to say that whatever impairments may exist due to such point sources probably result from the cumulative effects of all of them within the entire watershed. This would include all of those sources listed in Tables B2 and/or B3, depending on the time period being considered.

As discussed in section A2.4, there is a very direct relationship between nutrients, organic enrichment, and dissolved oxygen level. Over time, it is expected that problems with excessive plant growth (organic enrichment) and low dissolved oxygen levels will decrease as smaller loads of the limiting nutrient (in this case, phosphorus) are introduced into Neshaminy Creek and its tributaries. In fact, as discussed in earlier in this section, lower observed concentrations of phosphorus at the WQN monitoring station (see Figure A3) appear to mirror reported drops in phosphorus loads discharged by municipal sources in the watershed since the mid-1990s. Again, as discussed previously, it is likely that plant upgrades in the mid-1990s did not yet have any noticeable effects on stream health that would have been observed during the field work

Figure B8. Neshaminy Creek point source impacts.

conducted for the 1996 303(d) listing of these stream segments. Therefore, it does not appear that further reductions in nutrients from point sources discharging to this segment (or its tributaries) are required since the point sources are, for the most part, operating within the limitations of current NPDES permits. However, it would be prudent to assess this stream at a later date to verify if observed nutrient reductions have resulted in improved stream health. In this case, the recommended nutrient TMDLs for the point sources would be the cumulative load reflected in existing NPDES permits in this watershed.

As described above, this particular segment was also been listed in 1996 for pH impairment. From Figure B9, it can be seen that observed pH at the WQN monitoring station on Neshaminy Creek has stayed at a fairly consistent concentration of about 7.8 over the last decade, and varied from a low of about 6.5 to a high of about 9.5 (a value of 7.0 is considered to be neutral). These values are generally thought to be tolerable for most aquatic organisms (Pennsylvania Fish and Boat Commission, 2000).

pH 10 9 8 7 6 5 4 3 2 1 0

3 4 -90 -91 -92 -9 -9 -95 -96 -97 -98 -99 n n Jan-89 Jan Jan Jan Jan Jan Ja Ja Jan Jan Jan Jan-00

Figure B9. Trend in pH at DEP monitoring station.

B3.5 Neshaminy Creek (980515-1347-GLW)

This segment is about 2 miles long, and is listed as being impaired by water/flow variability from municipal point sources (see Figure B10). This particular impairment is not addressed in this document since neither the U.S. EPA or PaDEP have criteria or quantitative measures for dealing with this impairment.

B3.6 Neshaminy Creek (980609-1259-GLW)

This particular segment was listed as being impacted by nutrients from a municipal point source. The sole point source on this reach is NPDES facility PA0021181. Although this segment was listed as being impaired by nutrients, as shown in Table B3, this facility has been discharging about one-half of its calculated permit load for nitrogen and about one-third of its calculated permit load for phosphorus. As also shown in Table B4, it only experienced two permit infractions during the 1998-2000 time period. Based on this evidence, it does not appear that additional nutrient reductions are needed at this facility. Consequently, it is recommended that the nutrient TMDL be the load reflected in existing NPDES permit for this facility.

Figure B10. Neshaminy Creek point source impacts.

Figure B11. Neshaminy Creek point source impacts.

B3.7 Park Creek (980622-1146-GLW)

Park Creek is a tributary of Little Neshaminy Creek (see section B3.2), and was listed in 1996 as being impaired by nutrients from municipal point sources. As described in section B3.2 (and illustrated in Figure B12), there are three municipal facilities adjacent to Park Creek and its tributaries (PA0050059, PA0022411, and PA0051985). One of these facilities (PA0051985) did not exist in 1988, and two facilities that did exist then (PA0050253 and PA0050695) are no longer operational. In 1988, the four operating municipal facilities (PA0050253, PA0050695, PA0050059, and PA0022411) collectively discharged about 7943 kg of nitrogen and 1028 kg of phosphorus per year (from Table B2). Both of these loads were well below the calculated permit loads of 23,490 and 3995 for nitrogen and phosphorus, respectively. In 1999, the three remaining facilities discharged about 6634 kg of nitrogen and 1498 kg of phosphorus per year (from Table B3). Again, these loads were well below the calculated permit loads of 25,305 and 4507 for nitrogen and phosphorus, respectively. Given these loads, it does not appear that additional nutrient reductions are needed at any of these facilities. Consequently, it is recommended that the nutrient TMDL be the cumulative load reflected in existing NPDES permits for these facilities.

Figure B12. Park Creek point source impacts.

B3.8 Park Creek (980622-1147-GLW)

This particular 1.2-mile segment represents the lower end of Park Creek and a small tributary that flows into it about a half mile upstream from the confluence with Little Neshaminy Creek (see Figure B13). This segment has no point source discharges directly adjacent to it; rather, it is an extension of the segment discussed above (980622-1146-GLW). For the same reasons given above, it does not appear that additional nutrient reductions are needed at any of these facilities upstream of this segment. Consequently, it is recommended that the nutrient TMDL be the cumulative load reflected in existing NPDES permits for the upstream facilities.

Figure B13. Park Creek point source impacts.

B3.9 West Branch Neshaminy Creek (492)

This tiny segment is a 0.1-mile reach that is immediately downstream of the Borough of Lansdale municipal facility (see Figure B14), and the listed cause of impairment is nutrients. As shown in Tables B2 and B3, this facility has consistently discharged both nitrogen and phosphorus at levels of less than one-quarter of calculated permit loads. As also shown in Table B4, it did not exceed established levels for any of the critical parameters during the 1998-2000 time period. Given this, it does not appear that additional nutrient reductions are needed at this

facility. Therefore, it is recommended that the nutrient TMDL be the load reflected in the existing NPDES permit for this facility.

Figure B14. West Branch Neshaminy Creek point source impacts.

B3.10 West Branch Neshaminy Creek (980202-1043-GLW)

This is a 7.7-mile segment that was listed in 1996 as being impaired by excessive algal growth (see Figure B15). Based on the discussion provided earlier, this growth (i.e., increased productivity) is due to excessive nutrient levels, with phosphorus most likely being the limiting nutrient. In 1988, there were three municipal treatment plants operating on this segment (PA0026247, PA0020231, and PA0052795), in addition to another plant (PA0021682) on segment 492 that flows into it (see above discussion). In 1999, two of these plants were taken off-line (PA0020231 and PA0052795), and are no longer operational.

During the 1988 time period, the four plants were collectively discharging about 117,934 kg/year and 10,942 kg/year of nitrogen and phosphorus, respectively. These loads are about two-thirds and one-third of the calculated permit loads for nitrogen and phosphorus (193,514 kg/year and 30,241 kg/year, respectively). In 1999, the reported nitrogen and phosphorus discharges had been reduced even further to 107,304 kg/year and 8,274 kg/year, respectively; primarily, due to the discontinued use of two of the treatment plants. As also shown in Table B4,

Figure B15. West Branch Neshaminy Creek point source impacts.

neither of the two plants still in operation exceeded established levels for nutrients during the 1998-2000 time period. Given this, it does not appear that additional nutrient reductions are warranted for either of these facilities. Therefore, it is recommended that the nutrient TMDL be the cumulative load reflected in the existing NPDES permits for the two operating facilities.

In this particular sub-watershed, it may be that the excessive algal growth is due to organic matter and nutrients emanating from agricultural and developed areas through which this segment flows. (As shown in Table A1, this segment was also listed for excessive algal growth impairments due to agriculture, which is addressed later in section I). Given the urban nature of this sub-watershed, it may be that surface water runoff from these areas (particularly the urban areas) also has a higher temperature than more natural settings, which may be speeding up in- stream growth processes.

B3.11 West Branch Neshaminy Creek (980205-1330-GLW)

This segment is about 1.8 miles long, and is actually comprised of three smaller reaches (see Figure B16). It is listed as being impaired by flow alterations from municipal point sources. This particular impairment is not addressed in this document since neither the U.S. EPA or PaDEP have criteria or quantitative measures for dealing with this impairment. In this case, it is more likely that such flow alterations are due to urban runoff. It is assumed that addressing sediment

loads through the use of urban BMPs will at the same time reduce water flow variability within the watershed (see related discussion in section F).

Figure B16. West Branch Neshaminy Creek point source impacts.

B3.12 West Branch Neshaminy Creek (980205-1333-GLW)

This segment is about 1.6 miles long, and is listed as being impaired by flow alterations from municipal point sources (see Figure B17). The segment is essentially split in half by segment 492 identified in Figure B14. This particular impairment is not addressed in this document since neither the U.S. EPA or PaDEP have criteria or quantitative measures for dealing with this type of impairment. In this case, it is more likely that such flow alterations are due to urban runoff. It is assumed that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow variability within the watershed (see related discussion in section F).

Figure B17. West Branch Neshaminy Creek point source impacts.

C. Total Maximum Daily Loads (TMDLs) Development Plan for Little Neshaminy Watershed

Table of Contents Page

Executive Summary ………………………………………………… ……… 40

C1.0 Introduction ……………………………………………………………….……. 41 C1.1 Watershed Description …………………………………………… .…. 41 C1.2 Surface Water Quality ………………………………………… ……... 41 C2.0 Approach to TMDL Development………………………………………...…….. 43 C2.1 Water/Flow Variability Resulting from Urban Runoff/Storm Sewers … 44 C2.2 Siltation Caused by Urban Runoff/Storm Sewers ……………….……. 44 C2.3 Nutrients………………………………………………………………. 44 C2.4 Watershed Assessment and Modeling…………………………………….. 45 C3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 46 C3.1 Sediment TMDL Total Load ………………………………………..….. 49 C3.2 Margin of Safety ………………………………………...……. 49 C3.3. Load Allocation …………………………………………..….. 50 C3.4. Adjusted Load Allocation …………………………………..… 50 C3.5. Load Reduction Procedures ………………………………..…. 50 C4.0 Consideration of Critical Conditions ……………………………………… … 52 C5.0 Consideration of Seasonal Variations …………………………………..…….. 52 C6.0 Reasonable Assurance of Implementation …………………………………… 52 C7.0 Public Participation …………………………………………………………. 54

List of Tables Page

C1. Physical Characteristics of Little Neshaminy ……………………………………. 42 C2. Loading Values for Little Neshaminy Watershed, Year 1992 Land Use Conditions …………………………..………………. 47 C3. Loading Values for Little Neshaminy Watershed, Year 2000 Land Use Conditions …………………………..………………. 48 C4. Header Information for Tables G2 and G4………………………………..…. 49 C5. Summary of TMDLs for Little Neshaminy Watershed ……………….. 50 C6. Load Allocation for each contributing source in Little Neshaminy Watershed 51 C7. Load Allocation of Sediments by Source for Each Sub-watershed………….……….. 53 C8. Sediment Load Allocation by Land Use/Source ………………………... 53

List of Figures Page

C1. Little Neshaminy Watershed ……………………………………………….…. 42 C2. Little Neshaminy Sub-watersheds………………………………………………. 51

EXECUTIVE SUMMARY

Little Neshaminy Creek is a tributary of Neshaminy Creek and its watershed covers an area of 43.2 square miles in Bucks and Montgomery Counties. The protected uses of the stream include water supply, recreation, and aquatic life. The designated aquatic uses for Little Neshaminy Creek, its tributary Park Creek, and several unnamed tributaries are warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 47.20 miles of the main stem of Little Neshaminy Creek (Stream Code ID# 980628-1342-GLW), its tributary Park Creek (Stream Code IDs# 980622-1146-GLW, 20010511-1045-GLW, and 980622-1147-GLW), and several unnamed tributaries. They were developed to address the impairments noted on Pennsylvania’s 1996, 1998, and 2002 Clean Water Act Section 303(d) Lists. The impairments to be addressed in this document are those caused by sediments due to continuing land development in the watershed, and nutrients from municipal point sources. Water/flow variability, also listed as a cause of impairment, was not explicitly addressed because it was believed that the implementation of BMPs in the urban land use areas (High and Low Intensity Developed) to reduce sediment would also decrease water flow and volume to the stream and therefore stabilizes stream flow.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of the Little Neshaminy watershed. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining and Year 2000 when it was found to be impaired. Siltation, the cause of impairment in Little Neshaminy Creek, resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the amounts of sediment loading that will meet the water quality objectives for Little Neshaminy were 5,657,771 pounds per year. It is assumed that the Little Neshaminy Creek will support its aquatic life uses when this value is met. The sediment TMDL for Little Neshaminy is allocated as shown in the table below.

Summary of TMDL for Little Neshaminy Creek Watershed (lbs/yr) Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 7,708,16 770,817 - 6,937,35 1,279,58 5,657,77 stream bank erosion 8 1 0 1

The TMDL for sediment is allocated to non-point source from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 47.20 miles of the main stem of Little Neshaminy Creek, its tributary Park Creek and several of its unnamed tributaries. The TMDL establishes a reduction for total sediment loading of 17% from the current annual loading of 8,369,480 pounds.

C1.0 INTRODUCTION

C1.1 Watershed Description

The following discussion provides information on the physical characteristics of Little Neshaminy and its watershed including location, land use distributions, and geology. Little Neshaminy watershed is located in the Piedmont physiographic province, and is evenly distributed between Bucks and Montgomery counties. It covers an area of approximately 43.2 square miles. Little Neshaminy drains into the main stem of Neshaminy Creek from the west. The watershed is bounded by Pennsylvania Route 309 to the west, Route 232 to the east, and US Interstate 276 (Pennsylvania Turnpike) to the south. It can be accessed in the north-south direction via Route 463 or from Doylestown via Route 611. Figure C1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the main stem of Little Neshaminy Creek and its tributaries is warm water fishes and migratory fishes.

The current land use distribution in the Little Neshaminy watershed was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land uses in the Little Neshaminy watershed are agriculture (34%), forested land (33%), and developed land (33%). It is important to note that development in the watershed changed from 11.81 to 14.14 square miles from 1992 to 2000, or a 20% increase.

The surficial geology of Little Neshaminy watershed consists primarily of sandstone of the Stockton formation (66%) and the Lockatong shale formation (33%). The Stockton formation is the best source for water supply wells in the area although yields vary considerably. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.38). Watershed characteristics are summarized in Table C1.

C1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for the Little Neshaminy Creek watershed to address the impairments noted on Pennsylvania’s 1996, 1998, and 2002 Clean Water Act Section 303(d) Lists (see Table A1 in section A1.0). It was first determined that Little Neshaminy was not meeting its designated water quality uses for protection of aquatic life based on a 1994 aquatic biological survey, which included kick screen analysis and habitat surveys. In 1997, the creek was evaluated under the Unassessed Waters Program and was found to be still impaired. In 2001, the Department again surveyed the stream and its tributaries. In addition to the main stem of Little Neshaminy and Park Creek, additional stream segments were found

impaired. As a consequence of the surveys, Pennsylvania listed Little Neshaminy Creek on the 1996, 1998, and 2002 Section 303(d) Lists of Impaired Waters.

Figure C1. Little Neshaminy Creek Watershed.

Table C1. Physical Characteristics of Little Neshaminy Creek Watershed

Physiographic Province Piedmont Area (square miles) 47.24 Predominant Land Use Agriculture (34%) Forested land (33%) Developed land (33%) Predominant Geology Sandstone and Shale Soils Dominant HSGs C Average K Factor 0.38 20-Year Average Rainfall (in) 44.2 20-Year Average Runoff (in) 11.8

The 1996 303 (d) List reported 26.2 miles of the Little Neshaminy main stem and 7.6 miles of Park Creek to be impaired to be impaired by nutrients, water/flow variability, and siltation. The 1998 303(d) List added 5.5 miles to the main stem of Little Neshaminy. The 2002 303 (d) List added 6.2 miles to the little Neshaminy and 1.7 miles to Park Creek impaired segments. This list reported 37.90 miles of Little Neshaminy creek (plus its several unnamed tributaries) and 9.30 miles of Park Creek (and its two unnamed tributaries) to be impaired by nutrients from municipal point sources, and by siltation and water/flow variability as a result of urban runoff/storm sewers.

Stream segments of Little Neshaminy Creek and its tributaries are impacted by siltation as a result of “new land development” in the watershed. New land development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in Little Neshaminy is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events. As indicated above, land development has increased by approximately 20% between 1992 and 2000.

Sediments, which are often the cause of stream impairment in urban and suburban areas, are primarily from two sources: 1) disturbed land and unprotected soils at construction sites, and 2) stream channel erosion. Transitional land uses, mainly new construction sites, are one of the main sources of sediments in streams draining newly developed areas. Sediment production and sedimentation in streams are typically important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the Little Neshaminy Creek watershed. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in the watershed. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur in the stream.

C2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDLs address impairment by sediments in Little Neshaminy stream segments as reported on the 2002 303(d) Lists. The stream water flow variability impairment caused by urban runoff/storm sewer will not be explicitly addressed by these TMDLs because it is assumed that management practices that will be used to address storm water runoff and sediment

production at new construction sites will reduce problems associated with flow variability as well. These TMDLs were derived as follows:

C2.1 Water/Flow Variability Resulting from Urban Runoff/Storm Sewers

TMDLs were not determined for water/flow variability. It was assumed that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow variability within the watershed.

C2.2 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediments caused by newly developed land in the watershed were the cause of impairment of Little Neshaminy stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. The TMDLs for Little Neshaminy watershed address sediments from construction sites or “Transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

Comparing simulated annual sediment loads for Year 1992 and Year 2000 land use conditions in the watershed. It appeared from several field visits in the watershed that most of the siltation and turbidity observed in the stream have accumulated during several years. This assumption is supported by the fact that siltation was not found as a cause of impairment during the 1994 survey and 1997 assessments. Year 1992 is considered here as the benchmark because (as indicated earlier) the analysis of classified satellite images showed that development in the watershed increased by about 20% between 1992 and 2000.

C2.3 Nutrients

As reported earlier, nutrients from municipal sources were also listed as the cause of impairment for several stream segments of the Little Neshaminy and its tributaries. Although listed as the primary cause, it was determined on the basis of an evaluation of NPDES facilities in the watershed (see section B) that point source discharges were not likely the cause of observed nutrient impairments. Although it is highly probable that point sources continued to contribute to nutrient problems throughout the 1990s (during the time when the initial stream surveys upon which the 303d listings were based), it now appears that organic enrichment problems caused by point source discharges of nutrients (i.e., phosphorus) have been mitigated as evidenced by decreasing in-stream concentrations of phosphorus over the last decade in the larger Neshaminy Creek watershed, of which the Little Neshaminy watershed is a substantial part (see section B for more details). The lack of obvious positive improvements in aquatic health indicators over the last few years is very likely due to the lack of sufficient recovery time in the stream since wastewater treatment plant upgrades have been implemented. This recovery has also likely been hindered by the below-normal levels of precipitation experienced in this particular region over the last few years.

Due to the fact that nutrient loads from point sources are now at or below NPDES-permitted loads (particularly in the case of phosphorus), it is believed that no further reductions in point source loads are needed at this time. Consequently, problems related to in-stream nutrients could be addressed by reducing actual yearly phosphorus loads from other sources in the watershed, specifically nutrient loads associated with urban development (i.e., “transitional land use”) and streambank erosion for the “pre” and “post” development periods cited above for siltation (i.e., 1992 and 2000). In this case, it is assumed that the stream would attain its beneficial use when actual phosphorus loads are reduced to or below the level of the sum of nutrient loads associated with transitional land use and stream bank erosion in 1992. However, as shown in Section C4.4, Equation 9, the increase in phosphorus resulting from these land use changes was not substantial enough to allow the computation of the TMDL for nutrients.

The objective of the TMDL process for Little Neshaminy Creek is to reduce the average loading rate of sediments in the impaired stream segment to the levels equivalent to or slightly lower than the average loading rate at around 1992. This load reduction will allow the biological community to return to the impaired stream segments. The TMDL endpoints established for this analysis are discussed in detail in the TMDL section. The listing for impairment caused by siltation is addressed through reduction of sediment loads, respectively.

C2.4 Watershed Assessment and Modeling

The AVGWLF model was run for the Little Neshaminy Creek watershed to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Prior to running the model for the two land use conditions as described, historical stream water quality data for the period 4/89 to 3/96 were first used to calibrate various key parameters within the GWLF model. Such data sets are typically not available in AVGWLF-based TMDL assessments done elsewhere in Pennsylvania. In this case, however, it was felt that model calibration would provide for better simulation of localized watershed processes and conditions. A description of the calibration procedure used can be found in section A2.3 of this document.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the Little Neshaminy watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the 1992 and 2000 land use/ cover conditions are shown Tables C2 and C3, respectively. Point source N and P loads in this table, which are the cause of impairment in the watershed, are mean annual loads that were computed using data from recent PaDEP discharge monitoring reports (DMR) for individual municipal point sources. The Unit Area Load for each pollutant in each watershed was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area

for the watershed. Table C4 presents an explanation of the header information contained in Tables C2 and C3. Modeling output for Little Neshaminy watershed for 1992 and 2000 land use conditions is presented in Appendix F.

C3.0 LOAD ALLOCATION PROCEDURE FOR NUTRIENT AND SEDIMENT TMDLs

The load allocation and reduction procedures were applied to the entire Little Neshaminy watershed and its sub-watersheds. The sub-watersheds were obtained by delineating contributing areas to each of the major impacted stream segments (Figure C2). SubWatersh_2638 consists of the area drained by stream segments of the main stem of Little Neshaminy (Segment ID# 2638) and its tributaries. SubWatersh_2661 is the drainage area that contains impaired segments of Park Creek (Stream Segment ID# 2661) and its tributaries. In addition to sub-watershed delineations, a GIS analysis was performed to determine land use distributions in each sub- watershed. This data, along with watershed area and miles impaired in each sub-watershed, are needed for load reduction analyses.

The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDLs for Little Neshaminy watershed. TMDLs and load reductions were performed separately depending of the origin of the source and type of pollutant. These computations were done for sediments emanating from upland erosion based on the unit watershed area and for those from stream bank erosion using unit stream length (in feet).

Although nutrients from point sources were reported as the cause of nutrient impairment on the 303 (d) List, data analyses of permitted and current nutrient loadings (see section B) showed that point source discharges were not likely to be the cause on nutrient impairments observed in the watershed. Therefore, it was assumed that nutrients associated with transitional land uses and stream bank erosion resulting from land use changes from 1992 to 2000, may have been the cause of stream impairment. This is in part based on the corollary assumption that the NPDES permitted discharges were established at levels to ensure that designated uses for affected streams were adequately met. However, as shown in Equation 9 below, the adjusted load allocation is lower then zero (-1,642 lbs). This indicates that P increases associated with transitional land use and stream bank erosion resulting from land use changes between 1992 and 2000 was not substantial enough to cause nutrient impairments observed in the watershed. Therefore, nutrient TMDL was not performed for Little Neshaminy watershed.

The equations defining TMDLs for sediments and nutrients are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is

Table C2. Loading Values for Little Neshaminy Watershed, Year 1992 Land Use Conditions

Unit Area P Total N Unit Area N Sediment Unit Area Land Use Category Area Total P Load (lbs/yr) Load Load Sediment Load (acres) (lbs/yr) (lbs/acre/ yr) (lb/acre/ yr) (lbs/year) (lbs/acre/yr) Hay/Pasture 2,726 565 0.21 4,857 1.78 47,108 17.28 Cropland 7,989 3,455 0.43 27,312 3.42 1,286,115 160.99 Con Forest 296 1 0.00 33 0.11 240 0.81 Mixed Forest 1,911 8 0.00 210 0.11 2,250 1.18 Decid Forest 6,918 30 0.00 771 0.11 10,504 1.52 Unpaved Roads 7 4 0.57 46 6.57 3,285 469.29 Transition 17 8 0.47 95 5.59 3,991 234.76 Low Intensity Dev 5,640 187 0.04 1,400 0.25 127,814 22.66 High Intensity Dev 1,758 168 0.09 1,419 0.81 29,731 16.91 Stream Bank 1,921 9,296 6,197,130 Groundwater 4,429 90,067 Point Source1 29,638 260,185 Septic Systems 211 21,695 Total 27,262 40,614 1.49 417,386 15.31 7,708,168 282.75

1Nutrients from point sources are permitted average loads during the months of April through October and current loads from DMR reports for the period of November through March.

Table C3. Existing Loading Values for Little Neshaminy Watershed, Year 2000 Land Use Conditions

Unit Area P Total N Unit Area N Sediment Unit Area Land Use Category Area Total P Load (lbs/yr) Load Load Sediment Load (acres) (lbs/yr) (lbs/acre/ yr) (lb/acre/ yr) (lbs/year) (lbs/acre/yr) Hay/Pasture 2,570 530 0.21 4,561 1.77 43,465 17.03 Cropland 6,926 2,949 0.43 23,424 3.38 1,053,201 152.06 Con Forest 296 1 0.00 33 0.11 243 0.82 Mixed Forest 1,908 8 0.00 210 0.11 2,252 1.18 Decid Forest 6,655 29 0.00 740 0.11 10,110 1.52 Unpaved Road 7 4 0.57 46 6.57 3,289 469.86 Transition 1,187 907 0.76 8,183 6.89 826,324 696.14 Low Intensity Dev 5,909 207 0.04 1,556 0.26 136,071 23.03 High Intensity Dev 1,804 168 0.09 1,510 0.84 30,949 Stream Bank 1,948 9,395 6,263,576 Groundwater 4,450 90,490 Point Source1 29,638 260,185 Septic Systems 211 21,695 Total 27,262 45,500 1.67 395,772 14.52 8,369,480 307.00

1Nutrients from point sources are permitted average loads during the months of April through October and current loads from DMR reports for the period of November through March.

Table C4. Header Information for Tables 4 and 5.

Land Use The land cover classification that was obtained by from the MRLC Category database Area (acres) The area of the specific land cover/land use category found in the watershed. Total P The estimated total phosphorus loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area P Load The estimated loading rate for phosphorus for a specific land cover/land use category. Loading rate is expressed in lbs/acre/year Total N The estimated total nitrogen loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area N Load The estimated loading rate for nitrogen for a specific land cover/land use category. Loading rate is expressed in lbs/acre/year Total Sediment The estimated total sediment loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area The estimated loading rate for sediment for a specific land cover/land use Sediment Load category. Loading rate is expressed in lbs/acre/year

reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Little Neshaminy Creek to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

C4.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for Little Neshaminy watershed are based on the sediment loads obtained using the 1992 land use conditions, and are equal to 7,708,168 lbs/year and 40,614 lbs/year for sediment and P, respectively (see Table C2). P loads from municipal point sources in this table are permitted and not current loads as explained earlier.

C4.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDLs were reserved as the MOS.

MOS (Sediments) 7,708,168 lbs/yr x 0.1 = 770,817 lbs/yr (3)

C4.3 Load Allocation

The Load allocation (LA), consisting of all sources in the watershed, was computed by subtracting the margin of safety. Waste load allocation (WLA), which is usually subtracted from the TMDL total load, was not in both cases because: 1) for sediments, there is no waste load and 2) for phosphorus, point sources are also sources of pollution in the watershed. Nitrogen is not addressed in this analysis because it is a compliance issue rather than a TMDL issue.

LA (Sediments) 7,708,168 lbs/yr - 770,817 lbs/yr = 6,937,351 lbs/yr (4) LA (Phosphorus) 40,614 /bs/yr – 4,061 lbs/yr = 36,553 lbs/yr (5)

C4.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table C3 except those from the transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table C3,

LNR (Sediments) = 43,465 lbs/yr + 1,053,201 lbs/yr + 243 lbs/yr +2,252 lb/yr + 10,110 lb/yr + 3,289 lbs/yr + 136,071 lbs/yr + 30,949 lbs/yr = 1,279,580 lbs/y (6) ALA (Sediments) = 6,937,351 lbs/yr – 1,279,580 lbs/yr= 5,657,771 lbs/yr (7)

LNR (Phosphorus) = 530 lbs/yr + 2,949 lbs/yr + 1 lb/yr + 8 lbs/yr + 29 lbs/yr + 4 lbs/yr – 207 lbs/yr +168 lbs/yr + 4,450 lbs/yr + 29,638 lbs/yr +211 lbs/yr = 38,195 lbs/yr (8) ALA (Phosphorus) = 36,553 lbs/yr – 38,195 lbs/yr = -1,642 lbs/yr (9)

Table C5 below presents the TMDL for the Little Neshaminy Creek watershed.

Table C5. Summary of TMDLs for Little Neshaminy Creek Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 7,708,168 770,817 - 6,937,351 1,279,580 5,657,771 stream bank erosion

The ALA computed above is the portion of the load that is available to allocate among contributing land use/sources and sub-watersheds as described in the next step. The following section shows the allocation process in detail for the entire watershed and sub-watersheds.

C4.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in Little Neshaminy was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method

(as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table C6. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 659,410 and 4,998,361 pounds, respectively for Little Neshaminy Creek to attain its specific uses.

SubWatersh_2638

Park Creek Sub-watershed (SubWatersh_2661)

Figure C2. Little Neshaminy Creek Sub-watersheds

Table C6. Load Allocation for Each Contributing Source in Little Neshaminy Creek Watershed.

Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 826,324 12 659,410 20 - Stream bank erosion 6,263,576 88 4,998,361 20 TOTAL 7,089,900 100 5,657,771 100

The next step is to allocate sediment loads for each contributing source to each of the two sub-watersheds. The sub-watershed allocation for transitional land use was performed by areal coverage of each sub-watershed. Load allocation for stream bank erosion was accomplished according to the mileage of impaired stream segments in each sub-watershed as recorded on the 2002 303 (d) List. Load allocations and corresponding reductions by sub-watershed are presented in Table C7. Table C8 provides sediment load allocation when all land uses in the Little Neshaminy watershed are taken into consideration. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

The total allowable sediment load in the Little Neshaminy Creek and its tributaries when all land use/cover sources are considered is 6,937,351 pounds per year. In order for all stream segments to attain their specific uses, the total sediment load should be reduced from 8,369,480 pounds per year. Consequently, sediment load should be reduced by 17%.

C4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing these TMDLs using average annual conditions is protective of the waterbody.

C5.0 CONSIDERATION OF SEASONAL VARIATIONS

The continuous simulation model used for this analysis considers seasonal variation through a number of mechanisms. Daily time steps are used for weather data and water balance calculations. The model requires specification of the growing season, and hours of daylight for each month. The model also considers the months of the year when manure is applied to the land. The combination of these actions by the model accounts for seasonal variability.

C6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Table C7. Load Allocation of Sediments by Source for Each Sub-watershed

Transitional Land Use Stream Bank Erosion TOTAL

Sub-watershed Area Ave Ave Red Stream Ave Ave ALA Reduction Ave Ave Red Load ALA Length1 Load Load ALA Acres Lbs/yr lbs/yr - % - Miles lbs/yr Lbs/yr - % - lbs/yr Lbs/yr - % - SubWatersh_2638 781 543,687 433,866 20 37.90 5,034,773 4,017,771 20 5,578,460 4,451,637 20 SubWatersh_2661 406 282,637 225,544 20 9.25 1,228,803 980,590 20 1,511,440 1,206,134 20 TOTAL 1,187 826,324 659,410 20 47.15 6,263,576 4,998,361 20 7,089,900 5,657,771 20

1This is the mileage (stream length) of impacted sediments in each sub-watershed

Table C8. Sediment Load Allocation by Each Land Use/Source

Land Use Category Area (acres) Unit Area Load ALA Reduction Load (lbs/year) (lbs/year) (%) (lbs/acre/yr) Hay/Past 2,726 17.02 43,465 43,465 0 Cropland 7,989 152.06 1,053,201 1,053,201 0 Coniferous Forest 296 0.82 243 243 0 Mixed Forest 1,911 1.18 2,252 2,252 0 Deciduous Forest 6,918 1.52 10,110 10,110 0 Unpaved Road 7 469.86 3,289 3,289 0 Transition 17 696.14 826,324 659,41020 Low Int. Dev 5,640 23.03 136,071 136,071 0 High Int. Dev. 1,758 30,949 30,949 0 Stream Bank 6,263,576 4,998,361 20 Groundwater Point Source Septic Systems Total 27,262 77.25 8,369,480 6,937,351 17

erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within the Little Neshaminy Creek watershed will be affected by this policy, which has been included in Appendix E.

Implementation of BMPs aimed at sediment reduction will also assist in the reduction of phosphorus originating from transitional land uses and stream bank erosion. Other possibilities for attaining the desired reductions in phosphorus and sediment include streambank stabilization and fencing. Further field work will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet the nutrient and sediment reductions outlined in this report.

C7.0 PUBLIC PARTICIPATION

Notice of the draft TMDLs will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDLs. Notice of final TMDL approval will be posted on the Department website.

D. Total Maximum Daily Loads (TMDLs) Development Plan for Lake Galena

Table of Contents Page

Executive Summary……………………………………………………………………… 57

D1.0 INTRODUCTION..………………………………………………………………. 59 D1.1 Physical Setting………………………………………………………………… 59 D1.2 Lake Eutrophication……………………………………………………………. 60 D1.3 Lake Galena Water Quality…………………………………………………….. 61 D 1.4 Water Quality Standards……………………………………………………….. 61 D 1.5 Numeric Water Quality Target………………………………………………… 62

D2.0 TMDL ASSESSMENT METHODOLOGY………………………………………. 63 D2.1 Overview……………………………………………………………………….. 63 D2.2 Watershed Modeling……………………………………………………………. 64 D2.3 Lake Water Quality Modeling………………………………………………….. 68 D2.4 Load Allocation………………………………………………………………… 70

D3.0 NORTH BRANCH NESHAMINY CREEK……………………………………… 72

D4.0 CONSIDERATION OF CRITICAL CONDITIONS……………………………… 73

D5.0 CONSIDERATION OF SEASONAL VARIATIONS……………………………. 73

D6.0 REASONABLE ASSURANCE OF IMPLEMENTATION………………………. 74

D7.0 PUBLIC PARTICIPATION………………………………………………………. 74

EXECUTIVE SUMMARY

Lake Galena is located in the upper end of the Neshaminy Creek basin. The lake is approximately 2.33 miles long by 0.25 miles wide, and encompasses an area of about 370 acres. The watershed area that drains into this lake is approximately 9798 acres in size. In addition to water generated by precipitation within the watershed, Lake Galena also receives substantial inflow of water diverted from the nearby Delaware River for use by the North Penn and North Wales Water Authorities. The watershed contains a mix of land use/cover, including primarily agriculture, woodlands, and residential developments. In particular, this watershed has experienced a significant increase in residential development over the last 5-10 years, which has been identified as an important source of sediment to the lake during this time period. The designated aquatic uses for Lake Galena and North Branch Neshaminy Creek, is Trout Stocking, Migratory Fishes (TS, MF).

The lake was identified on Pennsylvania’s current 303(d) list as being impaired by nutrients and suspended solids from various sources, including on-site wastewater, agriculture, urban runoff/storm sewers, and other. Since Pennsylvania does not currently have numeric water quality standards for nutrients or suspended solids, the overall goal of this TMDL is to improve the trophic status of Lake Galena from hyper-eutrophic to mesotrophic. Based on this goal, the water quality target to address the stated impairments has been set at 10 ug/l of chlorophyll-a.

A combined watershed/lake water modeling approach was used to estimate current nutrient and sediment loads to the lake as well as to estimate phosphorus reductions needed to achieve the chlorophyll-a target of 10 ug/l. The land use/cover data layer for the lake watershed was updated using 1999-vintage satellite imagery. The updated layer was used to represent development conditions within the area as well as to estimate the amount of land being developed on a mean annual basis. Based on the modeling, it was estimated that the current annual phosphorus load of 3909 kg/yr (1776.8 lbs/yr) would have to be reduced to 1127 kg/yr (512.3 lbs/yr) in order to achieve the chlorophyll-a target of 10 ug/l. With an additional 10% margin of safety (MOS) factor, this target phosphorus load is further reduced to 1014.3 kg/yr (461.0 lbs/yr). The final load allocations by source are as shown in the table on the following page.

Segments of the North Branch of Neshaminy Creek immediately downstream from Lake Galena (Stream Segment ID# 980210-1123-GLW) were listed also as being impaired by both siltation and water/flow variability caused by an upstream impoundment (i.e., Lake Galena). A TMDL for water/flow variability was not developed in this case because neither the U.S. Environmental Protection Agency (EPA) nor PaDEP currently have water quality criteria for this impairment. Furthermore, quantitative measures for water/flow variability as an “impairment” are not currently available. However, it is assumed for these segments that addressing sediment loads through the use of various BMPs will at the same time reduce water flow variability or alterations within the watershed. With respect to the listed impairment from siltation, it is believed that a separate TMDL for this stream segment is not needed since the reductions for phosphorus specified for the Lake Galena watershed will directly result in sediment reductions in the North Branch as well. It is estimated that the targeted phosphorus reduction will result in approximately a 45% reduction in sediment to the lake as well. It is presumed that this would result in a similar reduction in sediment being delivered to the North Branch.

SOURCE OF P CURRENT LOAD (kg/yr)1 REDUCED LOAD (kg/yr)2

Agricultural land3 2097 314.6 Transitional land (in development) 71 10.7 Streambank erosion 216 32.4 Diversion water 1022 153.3 Other sources4 503 503

TOTAL 3909 1014.0

1Based on watershed and lake water quality modeling 2Loads reduced by 85% (i.e., current load x 0.15) 3Includes 347 kg/yr from groundwater load 4Other sources not reduced due to small size of load and/or difficulties in controlling them

D1.0 INTRODUCTION

Lake Galena has been identified on Pennsylvania’s current 303(d) list as being impaired by suspended solids and nutrients from various sources (see Table D1 below). Lake Galena was initially listed in 1998 as a result of a Clean Lakes Project, and was given a medium priority for TMDL development. Based on this listing, a TMDL for suspended solids and nutrients is being developed.

Table D1. Sources and causes of impairment for Lake Galena.

SOURCE CAUSE

On site Wastewater Nutrients Agriculture Suspended Solids On site Wastewater Suspended Solids Urban Runoff/Storm Sewers Suspended Solids Other Suspended Solids Agriculture Nutrients Urban Runoff/Storm Sewers Nutrients Other Nutrients

D1.1 Physical Setting

Lake Galena is located in the upper end of the Neshaminy Creek basin (see Figure D1). The lake is approximately 2.33 miles long by 0.25 miles wide, and encompasses an area of about 370 acres. The watershed area that drains into this lake is approximately 9798 acres in size. In addition to water generated by precipitation within the watershed, Lake Galena also receives substantial inflow of water diverted from the nearby Delaware River for water supply use by the North Penn and North Wales Water Authorities. This watershed contains a mix of land use/cover, including primarily agriculture, woodlands, and residential developments. In particular, this watershed has experienced a significant increase in residential development over the last 5-10 years, which has been identified as an important source of sediment to the lake during this time period. In terms of surface geology, the lake and surrounding drainage area is primarily underlain by the Lockatong shale formation. Only one point source (PA0052493) currently exists in the watershed. As shown earlier in Table B3, however, the nutrient loads from this source are relatively insignificant. A municipal discharge (PA0031615) that had existed in 1988 is no longer operational.

Figure D1. Location of Lake Galena.

D1.2 Lake Eutrophication

Based on recent studies in Lake Galena, it has been suggested that the lake is undergoing accelerated eutrophication due to the input of nutrients and suspended solids originating from various sources as summarized in Table D1. Lake eutrophication is both a natural and culturally- based phenomenon. Natural eutrophication is a slow, largely irreversible process associated with the gradual accumulation of organic matter and sediments in lake basins. Cultural eutrophication is an often rapid, possibly reversible process of nutrient enrichment and high biomass production stimulated by cultural activities causing nutrient transport to lakes (Novotny and Olem, 1994). Lakes are considered to undergo a process of “aging” which can be characterized by the trophic status as oligotrophic, mesotrophic, or eutrophic. Oligotrophic lakes are normally associated with deep lakes which have relatively high levels of dissolved oxygen throughout the year, bottom sediments typically contain small amounts of organic matter, chemical water quality is good, and aquatic populations are both productive and diverse. Mesotrophic lakes are characterized by intermediate levels of biological productivity and diversity, slightly reduced dissolved oxygen levels, and generally have adequate water quality to support designated uses. However, there is a recognition that these lakes are naturally or culturally moving towards a eutrophic state. Lakes which are classified as eutrophic typically exhibit high levels of organic

matter, both suspended in the water column and in the upper portions of sediments. Biological productivity is high, often indicated by seasonal algae blooms and excessive plant growth. Dissolved oxygen concentrations are low, and may reach extreme levels during critical periods. In addition, water quality is often poor resulting in violations of the designated uses. The following table illustrates typical water quality values associated with these trophic designations.

Table D2. Trophic Status of Lakes

Variable Oligotrophic Mezotrophic Eutrophic

Total P (ug/l) <10 10-20 >20 Chlorophyll-a (ug/l) <4 4-10 >10 Secchi disc depth (m) >4 2-4 <2 Hypolimnetic oxygen (% sat.) >80 10-80 <10

(Source: Thomann and Mueller, 1987)

D1.3 Lake Galena Water Quality

As a result of a past Clean Lakes project funded by the U.S. EPA (F.X. Browne, 1995), various water quality sampling was conducted in Lake Galena. Mean annual values obtained for critical parameters as a result of this sampling for the years 1989-1998 are shown in Table D3. The “TSI” values refer to “trophic state index” values calculated by DEP based on the methodology described by Carlson (1977). This index was developed for lakes that are phosphorus limited, which is the case for most lakes in Pennsylvania. Based on observations of several northern temperate lakes (Krenkel and Novotny, 1980), most oligotrophic lakes have TSI values below 40, mezotrophic lakes typically range in value between 35 and 45, and most eutrophic lakes generally have values greater than 45. Hyper-eutrophic lakes, on the other hand, can have TSI values above 60. Based on these values (as well as the ones shown in Table D2), it can be seen that Lake Galena appears to be in an advanced state of eutrophication since the TSI values in Table D3 range from 60.44 to 73.20. This situation is likely exacerbated by the nutrients, organic solids, and nutrient-enriched sediments entering the lake from the various sources summarized in Table D1.

D1.4 Water Quality Standards

Water quality standards are typically developed to control quantities of various pollutants that may enter water bodies in order to maintain healthy conditions and usually consist of three inter- related components: 1) designated and existing uses, 2) narrative and/or numerical water quality criteria necessary to support those uses, and 3) an anti-degradation statement. Furthermore, water quality standards serve the dual purposes of establishing the water quality goals for a specific waterbody and serve as the regulatory basis for the establishment of water quality-based

treatment controls and strategies beyond the technology-based levels of treatment required by section 301(b) and 306 of the Act (USEPA, 1991).

Table D3. Selected mean annual water quality values for Lake Galena.

Parameter Measured/Calculated Value

Chlorophyll-a (ug/l) 28.99 Secchi disk depth (m) 0.97 Total nitrogen (ug/l) 1388.6 Organic nitrogen (ug/l) 774.3 Total phosphorus (ug/l) 120.1 Ortho-phosphorus (ug/l) 33.5 TSI-P 73.20 TSI-Chlorophyll-a 63.63 TSI-Secchi disk 60.44

According to Pennsylvania Code, Title 25, Chapter 93, Water Quality Standards, Section 93.4, all surface waters in the state shall be protected for the following uses: warm water fishes, potable water supply, industrial water supply, livestock water supply, wildlife water supply, irrigation, boating, fishing, water contact sports, and aesthetics. Lake Galena, and the North Branch of Neshaminy Creek that flows into it, have been designated for Trout Stocking, Migratory Fishes (TS,MF).

Pennsylvania does not currently have specific numeric water quality criteria for suspended solids or nutrients to support these uses. However, Pennsylvania does have general water quality criteria that state in Section 93.6 that: a) Water may not contain substances attributable to point or nonpoint source discharges in concentration or amounts sufficient to be inimical or harmful to the water uses to be protected or to human, animal, plant or aquatic life; and b) In addition to other substances listed within or addressed by this chapter, specific substances to be controlled include, but are not limited to, floating materials, oil, grease, scum and substances which produce color, tastes, orders, turbidity or settle to form deposits. These general water quality criteria may be interpreted to identify an acceptable water quality endpoint. Pennsylvania has numeric water quality criteria for total dissolved solids, however, these criteria only apply to public water supplies.

D1.5 Numeric Water Quality Target

In order to develop a given TMDL, a water quality indicator and numeric water quality target must be specified. As mentioned, Pennsylvania does not currently have numeric water quality standards for nutrients or suspended solids. Therefore, the overall goal of this TMDL will be to

improve the trophic status of Lake Galena from hyper-eutrophic to mesotrophic. As described above, the current hyper-eutrophic conditions are due to excessive nutrient input to the lake (particularly phosphorus, since this is the limiting nutrient in this case). In this watershed, most of the phosphorus originates from nonpoint source runoff in both dissolved and particulate (i.e., sediment-attached) forms. This runoff results in increased concentration of suspended solids in the lake which is comprised of both organic organisms as well as suspended sediment. It is expected, therefore, that reductions in both dissolved and sediment-borne forms of phosphorus will result in decreased nutrient and suspended solids in Lake Galena. This, in fact, was one of the conclusions drawn from the earlier Clean Lakes study conducted for Lake Galena and its surrounding watershed (F.X. Browne, 1995).

According to the trophic state index values given in Table D2, there are 4 parameters used to relate water quality with trophic state. In this case, chlorophyll-a will be used as the numeric water quality target. Chlorophyll-a is easy to measure, is a valuable surrogate for algal biomass, and is desirable as a water quality target because alga are either the direct (nuisance algal blooms) or indirect (high/low dissolved, pH, and high turbidity) cause of most problems related to excessive enrichment (US EPA, 1999(a)). Based on the goal of improving the trophic status of Lake Galena from hyper-eutrophic to mesotrophic, the water quality target to address nutrient impairments has been set at 10 ug/l chlorophyll-a. More specifically, as described in the next section, estimates of phosphorus reductions needed to achieve this goal are made via the combined use of a watershed model (AVGWLF) and a lake water quality model (BATHTUB).

D2.0 TMDL ASSESSMENT METHODOLOGY

D2.1 Overview

A combined watershed modeling/lake water quality modeling approach was used to conduct the TMDL assessment for Lake Galena. The lake model is BATHTUB, which performs steady- state water and nutrient balance calculations in a spatially segmented hydraulic network that accounts for advective and diffusive transport, and nutrient sedimentation (Walker, 1996). BATHTUB is used to simulate the fate and transport of nutrients and water quality conditions and responses to nutrient loads into the lake. BATHTUB has been cited as an effective tool for lake and reservoir water quality assessment and management, particularly where data are limited (US EPA, 1999). In order to simulate water quality conditions, BATHTUB requires as input information on various lake characteristics such as length, width, mean depth, and nutrient loads from various sources in the surrounding watershed. Basic physical and hydrologic information was obtained via a combination of field work, reports, GIS data sets, and topographic maps. Information on nutrient loading to the lake from the surrounding area was derived using the AVGWLF watershed modeling application (see background discussions in sections A2.1 through A2.3). Subsequent to setting up the two models, calibration was performed using actual lake water quality sampling data obtained as a result of a prior EPA-funded Clean Lakes project (F.X. Browne, Inc., 1995). This calibration was needed in order to accurately estimate phosphorus reductions required to achieve the chlorophyll-a target goal of 10 ug/l.

For the purposes of this TMDL assessment, mean annual nutrient loads to the lake for the period 1988-1998 were estimated. These loads were then used within BATHTUB to evaluate

current water quality/trophic conditions. Once current conditions had been established and compared with existing lake water quality sampling measurements, the BATHTUB model was then used as a “diagnostic” tool in order to estimate the phosphorus load reductions required to achieve the chlorophyll-a target of 10 ug/l.

D2.2 Watershed Modeling

As outlined above, AVGWLF was used to derive nutrient load information for use as input to the BATHTUB lake model. Simulations were performed for the period 1988-1998 to coincide with the time period for which existing weather and lake water sampling data were available. When using AVGWLF, the “default” GIS data sets that come with this application are typically used (see section A2.2). This means that for most applications, the “satellite-derived”, Pamrlc land use/cover data set is generally utilized (see Table A2). However, in areas where land development is considered to be a possible factor with respect to water quality degradation (as is the case in this watershed), this particular data layer may not be adequate for comprehensive problem assessment since the satellite images used to create this data layer were primarily from the year 1992. Consequently, to better reflect current land use/cover conditions in the watershed, the existing Pamrlc was updated using more recent satellite data available for the area (in this instance, 1999-vintage SPOT satellite images [see section E1.1 for related discussion on this image data]). Figure D2 shows the default land use/cover map used within AVGWLF and the map that was updated for this particular analysis.

As can be seen from Figure D2, the amount of developed land in the watershed (primarily low-density and high-density residential land) has changed significantly since 1992. In fact, between 1992 and 1999, the amount of developed land had increased from approximately 75 hectares (185 acres) to approximately 788 hectares (1946 acres). This amounts to an average of about 100 hectares (247 acres) of new development per year during the 7-year time period reflected by the two land use/cover layers. For GWLF modeling purposes, the newer land use/cover layer was used to represent current development conditions in the watershed. Additionally, it was assumed that about 100 hectares (247 acres) of land per year is “under development”. Within GWLF, this amount was included in the “transition” category to reflect land that is generally without vegetative cover and subject to significant surface erosion during precipitation events. As described previously, sediment transported via erosion can be an important source of phosphorus in the lake.

Screen captures of the transport- and nutrient-related input data for the final GWLF model run are shown in Figures D3 and D4. The resulting mean annual hydrology and loading output from this final run is shown is Figures D5 and D6. The most important results presented in these figures are those given in the “Totals” columns highlighted in each case in light blue at the bottom. More specifically, these values were used to estimate mean annual inputs to Lake Galena with respect to water flow and nutrient loads as described in the next section.

Figure D2. Land use/cover ca. 1992 (top) and ca.1999 (bottom) within the Lake Galena watershed.

Figure D3. GWLF transport file for Lake Galena watershed load estimation.

Figure D4. GWLF nutrient file for Lake Galena watershed load estimation.

Figure D5. Mean annual hydrology reported by GWLF.

Figure D6. Mean annual load calculations by GWLF.

D2.3 Lake Water Quality Modeling

Using descriptive information about the lake and surrounding drainage area, as well as output from AVGWLF, a BATHTUB model was set up for Lake Galena for the purpose of simulating current water quality and trophic conditions. For the purposes of this TMDL assessment, mean annual nutrient loadings were simulated using AVGWLF output and DEP lake sampling data for the period 1988-1998. After initial model development, the sampling data were used to “fine- tune” various input parameters and sub-model selections within BATHTUB during the calibration process. Once calibrated, as described in a later section, BATHTUB could then be used to estimate nutrient (specifically phosphorus) load reductions needed in order to achieve TMDL target loads.

Given the relatively small size and simple shape of Lake Galena in comparison to lakes and reservoirs typically modeled with BATHTUB, the lake was treated as a single “segment” (i.e., pool) within the model. Similarly, water inflow and nutrient loads from the surrounding drainage area was treated as though they originated from one “tributary” (i.e., source) in BATHTUB. The model-required flow and nutrient concentration information for this particular source was derived from the AVGWLF output as described above. Also, as mentioned in section D1.1, a significant amount of water is drawn from the nearby Delaware River and diverted to Lake Galena for use local water suppliers. Based on information from DEP, this inflow amounts to an annual volume of about 3600 million gallons per year, which is a little less than half of the water entering the lake annually. Within BATHTUB, this source was identified as an additional tributary to the lake. Unfortunately, information on current nutrient concentrations was not available from DEP for this diversion. Consequently, water quality data obtained at a DEP water quality monitoring station (WQN101) located on the Delaware River downstream from the water supply intake for Lake Galena was used to estimate nitrogen and phosphorus concentrations for this inflow. Model parameter data for the lake and its surrounding drainage area are summarized in Table D4. Flow and concentration data for the two sources of input to Lake Galena is given in Table D5.

In addition to the input summarized in the tables above, executing BATHTUB also requires that decisions be made on the use of various nutrient balance, sedimentation, and eutrophication response sub-models. Such decisions are routinely made as part of the calibration process where the primary objective is to simulate processes in a way that achieves optimal matches between values for observed lake water quality parameters and estimated values based on compiled input data. For Lake Galena, optimal results were achieved using the sub-models given in Table D6.

The results from the final BATHTUB model run for key model parameters are shown in Table D7. The matches between simulated and observed mean annual water quality conditions were considered to be quite good, and based on these results, it was felt that these conditions were being simulated well enough to allow estimation of phosphorus reductions needed to achieve the chlorophyll-a TMDL target. Therefore, subsequent to calibrating the lake water quality model, additional model runs were made to quantify the phosphorus reductions to the lake needed in order to achieve TMDL objectives with respect to chlorophyll-a levels. Specifically, phosphorus loads to the lake as represented by various model sources (watershed, diversion, and atmosphere) were

Table D4. Lake and watershed data.

PARAMETER VALUE

Lake area1 370 acres (1.5 km2) Lake length1 2.33 miles (3.74 km) Lake width1 0.25 miles (0.4 km) Mean depth of lake2 15 feet (4.6 m) Atmospheric N loading to lake3 2698 kg/km2 Atmospheric P loading to lake3 44 kg/km2 Mean annual Total P concentration of lake4 120.1 ppb Mean annual Ortho-P concentration of lake4 33.5 ppb Mean annual Total N concentration of lake4 1388.6 ppb Mean annual Organic N concentration of lake4 774.3 ppb Mean annual Chlorophyll-a concentration of lake4 28.99 ppb Mean annual Secchi disk depth of lake4 0.97 m

1Topographic map, GIS data, and Clean Lakes study (DEP/F.X. Browne) 2Clean Lakes study (DEP/F.X. Browne) 3Nizeyimana et al. (1997) 4DEP sample data

Table D5. Mean annual source flows and nutrient loads.

PARAMETER WATERSHED1 DIVERSION

Flow 12.76 MGD (17.64 hm3/yr) 9.86 MGD (13.63 hm3/yr)2 Total N concentration 1.7421 mg/l (1742.1 ppb) 0.98 mg/l (980 ppb)3 Inorganic N concentration 1.5219 mg/l (15219 ppb) 0.85 mg/l (850 ppb)3 Total P concentration 0.1599 mg/l (159.9 ppb) 0.075 mg/l (75.0 ppb)3 Ortho-P concentration 0.0719 mg/l (71.9 ppb) 0.034 mg/l (34 ppb)3

1Derived via AVGWLF-based watershed modeling 2From DEP data 3Derived from Delaware River water quality monitoring station data (WQ101)

Table D6. Sub-models used within BATHTUB for Lake Galena.

MODEL OPTION SUB-MODEL USED

Phosphorus balance/sedimentation First-order settling velocity Nitrogen balance/sedimentation First-order settling velocity Mean chlorophyll-a P, Linear Secchi depth Secchi vs. Chl-a and Turbidity

Table D7. Final lake model simulation results for Lake Galena.

VARIABLE OBSERVED VALUE ESTIMATED VALUE

Total P (mg/m3) 120.10 117.14 Total N (mg/m3) 1388.60 1442.46 Chlorophyll-a (mg/m3) 28.99 32.80 Secchi depth (m) 0.97 0.89 Organic N (mg/m3) 774.30 927.86 TP – Ortho-P (mg/m3) 86.60 61.54 Turbidity (1/m) 0.31 0.31 Carlson TSI-P 73.20 72.84 Carlson TSI-Chla 63.63 64.84 Carlson TSI-Secchi 60.44 61.71

iteratively decreased until a simulated chlorophyll-a concentration of 10.0 ug/l was reached. Based on the calibration, the mean annual total P loads from these sources were 2821, 1022, and 66 kg/yr, respectively, for a total of 3909 kg/yr. Upon re-running the lake model as described above, it was found that a mean annual chlorophyll-a concentration of 10 ug/l could be achieved if the annual P load was reduced to 1127 kg/yr. This represents a reduction of about 71% from the current load.

D2.4 Load Allocation

Based on the analyses described above, the approximate mean annual load of 3909 kg of phosphorus that enters the lake comes primarily from the sources shown in Table D8. The specific pathways by which phosphorus is transported from these sources varies considerably, and because of this, opportunities for controlling the movement and retention of this nutrient will vary considerably as well. From agricultural land, phosphorus originates principally from soil erosion and application of manure and/or fertilizers. Phosphorus is lost from wooded areas in surface water

Table D8. Mean annual P source loads for Lake Galena.

SOURCE P (kg/yr)

Agricultural land 1750 Wooded land 10 Transitional/unpaved land 71 Developed land 48 Stream bank erosion 216 Groundwater 695 Septic systems 32 Water supply diversion 1022 Atmospheric deposition 66

TOTAL 3909

runoff in dissolved and particulate organic forms. From transitional and unpaved surfaces, it is primarily delivered via sediment during erosion events. In developed areas (in this case, residential areas in the watershed), phosphorus originates primarily from applied fertilizers and eroded organic matter. In the case of stream bank erosion, phosphorus found in the inherent parent material is moved as the banks are eroded due to increased volumes during storm events. As demonstrated by abundant research (e.g., Novotny and Olem, 1994; and Nelson and Booth, 2002), stream bank erosion is a significant problem in fast-developing watersheds due to increased surface water runoff resulting from reduced surface infiltration. Although the quantities are small in comparison to nitrates, some dissolved phosphorus is also transported from septic systems. With respect to groundwater, there is typically a small “background” concentration owing to various natural sources; however, it is true that this concentration can increase substantially in agricultural areas. In the Lake Galena watershed, the estimated groundwater P concentration is about 0.052 mg/l (see Figure D4), which is about twice the concentration found in more “natural” settings around Pennsylvania (Reese and Lee, 1998). Consequently, it is very likely that about half of the estimated mean annual groundwater P load (or about 347 kg/yr) is actually from agricultural sources (i.e., leached in dissolved form from the surface). Finally, as discussed earlier, about 1022 kg/yr of P is contributed by the water supply diversion, and about 66 kg/yr is deposited from the atmosphere.

Based on the magnitude of the contributed P load and potential opportunities for load reduction, the most appropriate sources to consider controlling are agricultural land, transitional land, stream bank erosion, and the diversion water. These sources combined supply about 3406 kg/yr, or about 87%, of the annual P load to the lake. As stated earlier, the lake model simulation showed that the chlorophyll-a target of 10 ug/l could be reached if the mean annual P load were reduced to about 1127 kg/yr. If this load were reduced again by a 10% margin-of-safety factor (for a new target load

of 1014.3 kg/yr), a load reduction of 85% for each of the four primary sources would be required in order to meet this target as shown in Table D9.

Table D9. Load allocations needed to meet target P load of 1014.3 kg/yr.

SOURCE OF P CURRENT LOAD (kg/yr)1 REDUCED LOAD (kg/yr)2

Agricultural land3 2097 314.6 Transitional land (in development) 71 10.7 Streambank erosion 216 32.4 Diversion water 1022 153.3 Other sources4 503 503

TOTAL 3909 1014.0

D3.0 NORTH BRANCH NESHAMINY CREEK

Segments of the North Branch of Neshaminy Creek immediately downstream from Lake Galena (Stream Segment ID# 980210-1123-GLW) were listed as being impaired by both siltation and water/flow variability caused by an upstream impoundment. This impoundment is, in fact, Lake Galena as discussed above (see Figure D7). A TMDL for water/flow variability was not developed in this case because neither the U.S. Environmental Protection Agency (EPA) nor PaDEP currently have water quality criteria for this impairment. Furthermore, quantitative measures for water/flow variability as an “impairment” are not currently available. However, it is assumed for these segments that addressing sediment loads through the use of various BMPs as discussed below in section D6.0 will at the same time reduce water flow variability or alterations within the watershed.

With respect to the listed impairment from siltation, it is believed that a separate TMDL for this stream segment is not needed since the reductions for phosphorus specified for the Lake Galena watershed will directly result in sediment reductions in the North Branch as well. As shown in Table D9, the recommended TMDL will require a 74% reduction in phosphorus (1 – (1014/3909) = 0.74) in order to meet the target water quality goal (including the 10% margin of safety). As discussed above, a significant portion of the phosphorus contributed to the lake is delivered as sediment-attached P, primarily from agricultural land, transitional land (i.e., land under construction), and streambank erosion. It can therefore be surmised that since about 61% of the phosphorus load is being contributed by these three sources ((2097+71+216)/3909 = 0.61), that a

74% reduction in P would result in approximately a 45% reduction in sediment to the lake (0.61 x 0.74 = 0.45). This presumably would result in a similar reduction in sediment being delivered to the North Branch as well.

Figure D7. Location of North Branch Neshaminy Creek.

D4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model used for mean annual load estimation is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing these TMDLs using average annual conditions is protective of the waterbody. In this case, a 10- year simulation period reflecting actual data for the period 1988-1998 were used to account for normal year-to year fluctuations in precipitation and temperature.

D5.0 CONSIDERATION OF SEASONAL VARIATIONS

The continuous watershed simulation model used for this analysis considers seasonal variation through a number of mechanisms. Daily time steps are used for weather data and water balance calculations. The model requires specification of the growing season, and hours of daylight for each month. The model also considers the months of the year when manure is applied to the land. The combination of these actions by the model accounts for seasonal variability.

D6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

In this TMDL assessment, dissolved and sediment-borne phosphorus loads have been attributed to agricultural land, transitional land (i.e., land in development), streambank erosion, and water diversion. With respect to agricultural land, various best management practices (BMPs) are available that can reduce phosphorus loads from such areas. While reductions of 85% are substantial, combinations of BMPs such as conservation tillage, crop residue management, field buffers, and nutrient management have been shown to achieve such high levels of phosphorus management if applied consistently and conscientiously (Ritter and Shirmohammadi, 2001). In transitional areas, phosphorus attached to sediment can be controlled by requiring the implementation of rigorous erosion and sediment (E&S) control plans by land developers and monitoring such plans for adherence.

Similarly, sediment losses attributed to streambank erosion can be reduced through implementation of best management practices (BMPs) in developed areas to increase infiltration and detain sediment. Substantial reductions in the amount of sediment reaching the lake can be made through the installation of drainage controls such as detention ponds, sediment ponds, constructed wetlands, infiltration pits, dikes and ditches. In fact, the implementation of such BMPs in this particular watershed will likely occur as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within the Little Neshaminy Creek watershed will be affected by this policy, which has been included in Appendix E. In this instance, implementation of BMPs aimed at sediment reduction will also assist in the reduction of phosphorus originating from transitional land uses and stream bank erosion.

Finally, opportunities exist for reducing phosphorus from water presently being diverted to Lake Galena for water supply purposes. This water is withdrawn from the Delaware River by the Point Pleasant Pumping Station and transported to the Bradshaw Reservoir to the east of Lake Galena. This water is then transported through the North Branch Transmission Main and discharged into the North Branch of Neshaminy Creek about 2.5 miles upstream of the lake. Although the costs would likely be considerable, this water could be treated using various infiltration and phosphorus removal processes prior to being released into the North Branch of Neshaminy Creek.

D6.0 PUBLIC PARTICIPATION

E. Total Maximum Daily Loads (TMDLs) Development Plan for Pine Run Watershed

Table of Contents Page

Executive Summary ……………………………………………………………… 78

E1.0 Introduction ……………………………………………………………….…. 79 E1.1 Watershed Description …………………………………………….….. 79 E1.2 Surface Water Quality …………………………………………………. 81 E2.0 Approach to TMDL Development………………………………………...…….. 81 E2.1 Excessive Algae Growth Resulting from Upstream Impoundment …... 82 E2.2 Siltation Caused by Urban Runoff/Storm Sewers ……………….……. 82 E2.3 Watershed Assessment and Modeling…………………………………. 82 E3.0 Load Allocation Procedure for Sediment TMDLs ………………..…………… 83 E3.1 Sediment TMDL Total Load ………………………………………..….. 86 E3.2 Margin of Safety ………………………………………...………..…. 86 E3.3. Load Allocation …………………………………………..………..... 86 E3.4. Adjusted Load Allocation ………………………………………....… 87 E3.5. Load Reduction Procedures …………………………………………. 87 E4.0 Consideration of Critical Conditions ………………………………………… 88 E5.0 Consideration of Seasonal Variations …………………………………..…….. 88 E6.0 Reasonable Assurance of Implementation …………………………………… 90 E7.0 Public Participation …………………………………………………………. 90

List of Tables Page

E1. Physical Characteristics of Pine Run …………………………………………. 80 E2. Loading Values for Pine Run Watershed, Year 1992 Land Use Conditions …………………………..………………. 84 E3. Loading Values for Pine Run Watershed, Year 2000 Land Use Conditions …………………………..………………. 84 E4. Header Information for Tables E2 and E3………………………………..…. .. 85 E5. Summary of TMDLs for Pine Run Watershed …………………..………...... 87 E6. Load Allocation for each contributing source in Pine Run Watershed………… 88 E7. Load Allocation of Sediments by Source for Each Sub-watershed………….… 89 E8. Sediment Load Allocation by Land Use/Source …………………………...... 89

List of Figures Page

E1. Pine Run Watershed ……………………………………………………….…. 80 E2. Pine Run Sub-watersheds…………………………………………..…………. 86

EXECUTIVE SUMMARY

The Pine Run watershed in Bucks County is approximately12.0 square miles in size. Pine Run is a tributary of North Branch Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is trout stocking and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 7.1 miles of the main steam of Pine Run (Stream Segment ID#s 980210-1240-GLW and 980211-1241-GLW) from its mouth going upstream, and 1.3 miles of an unnamed tributary (Stream Segment ID# 980210-1242-GLW) located at about 1/5 mile north of the town of New Britain. They were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water act Section 303(d) Lists. The impairments are primarily caused by sediment loads from land development in the watershed. Consequently, the TMDL focuses on control of sediments. Excessive algae growth, also listed as a cause of impairment due to upstream impoundment, was not explicitly addressed because it is believed that the implementation of BMPs in the urban land use areas (High and Low Intensity Developed) to reduce sediment in runoff would also decrease build-up of sediments resulting from stream impoundment, and subsequently a decrease in phosphorus.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of the Pine Run watershed. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining its designated use, and Year 2000 when it was found to be impaired. Siltation, the cause of impairment in Pine Run, resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the sediment loading that will meet the water quality objectives for Pine Run was 1,140,111 pounds per year. It is assumed that Pine Run will support its aquatic life uses when this value is met. The sediment TMDL for Pine Run is allocated as shown in the table below.

Summary of TMDL for Pine Run Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 2,160,265 216,026 - 1,944,239 804,1281,140,111 stream bank erosion

The TMDL for sediments is allocated to non-point source pollution from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 8.4 miles of the main stem of Pine Run and its unnamed tributary. The TMDL establishes a reduction for total sediment loading of 52% from the current annual loading of 4,089,625 pounds.

E1.0 INTRODUCTION

E1.1 Watershed Description

The following discussion provides information on the physical characteristics of Pine Run and its watershed including its location, land use distributions, and geology. Pine Run watershed is located entirely in Bucks county and is in the Piedmont physiographic province. It covers an area of approximately 12.0 square miles. Pine Run drains into the main stem of North Branch of Neshaminy Creek in the town of Chalfont. The watershed is bounded to the south by Pennsylvania Route 202, and to the east by Route 611. It can also be accessed from Norristown via Route 202 or from Doylestown via Route 611. Figure E1 shows the watershed boundary, its location, and the state of water quality of stream segments as reported from the 2002 303(d) List. The protected uses of the watershed are water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the main stem of Pine Run and its unnamed tributary is trout stocking and migratory fishes.

The current land use distribution in Pine Run watershed was developed by updating the National Land Cover Data (NLCD) (Vogelmann et al., 1998) using a more recent satellite imagery; the 10 m-colorized SPOT (System Probatoire pour l’Observation de la Terre). The NLCD development was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://www.pasda.psu.edu) at no charge. The primary land uses in the Pine Run watershed are agriculture (37%) and forested land (37%), followed by developed land (26%). It is important to note that development in the watershed more than doubled from 1992 to 2000. It increased from 340 to 768 hectares (126%) during the 8-year period.

The surficial geology of Pine Run watershed consists primarily of sandstone of the Stockton formation. This formation is the best source for water supply wells in the area although yields vary considerably. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high K factor). Watershed characteristics are summarized in Table E1.

Figure E1. Pine Run Watershed.

Table E1. Physical Characteristics of Pine Run Watershed

Physiographic Province Piedmont Area (square miles) 12 Predominant Land Use Agriculture (37%) Forest land (37%) Developed land (26%) Predominant Geology Sandstone Soils Dominant HSGs C (47%) and B (31%) K Factor 0.37 20-Year Average Rainfall (in) 40.4 20-Year Average Runoff (in) 4.5

E1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for the Pine Run Watershed to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) Lists (see Table A1 in section A1.0). It was first determined that Pine Run and its unnamed tributary was not meeting its designated water quality uses for protection of aquatic life based on a 2001 aquatic biological survey. As a consequence of this survey, Pennsylvania listed Pine Run and its unnamed tributary on the 2002 Section 303(d) Lists of Impaired Waters.

The 2002 303 (d) List reported 7.1 miles of the Pine Run main stem and 1.3 miles of an unnamed tributary to be impaired by siltation and excessive algae growth. Stream segments of Pine Run Creek and its tributariy are impacted by siltation as a result of “new land development” in the watershed. New land development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in Pine Run is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events. As indicated above, land development has increased by approximately 126% between 1992 and 2000.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the Pine Run watershed. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in Pine Run. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur in the stream.

E2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDL addresses impairment by sediments in Pine Run stream segments as reported on the 2002 303(d) List. The excessive algae growth impairment caused by upstream impoundment will not be explicitly addressed by this TMDL because it is assumed that management practices that will be used to address storm water runoff and sediment production at new construction sites will reduce problems associated with algae growth as well. The TMDL was derived as follows:

E2.1 Excessive Algae Growth Resulting from Upstream Impoundment

A TMDL was not determined for excessive algae growth. It was assumed that addressing sediment loads (and its associated phosphorus load) through the use of urban BMPs will at the same time reduce excessive algae growth problems within the watershed.

E2.2 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediments caused by newly developed land in the watershed were the cause of impairment of Pine Run stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. The TMDL for the Pine Run watershed addresses sediment from construction sites or “transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

The objective of the TMDL process for Pine Run is to reduce the average loading rate of sediments in the impaired stream segment to levels equivalent to or slightly lower than the average loading rate ca. 1992. It is assumed that this load reduction will allow the biological community to return to the impaired stream segments. The TMDL endpoints established for this analysis are discussed in detail in the TMDL section. The listing for impairment caused by siltation is addressed through reduction of sediment loads.

E2.3 Watershed Assessment and Modeling

The AVGWLF model was run for the Pine Run watershed to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the

model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the Pine Run watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the 1992 and 2000 land use/cover conditions are shown in Tables E2 and E3, respectively. The Unit Area Load for sediment in the watershed was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area for the watershed. Table E4 presents an explanation of the header information contained in Tables E2 and E3. Modeling output for Pine Run watershed for 1992 and 2000 land use conditions is presented in Appendix F.

E3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire Pine Run watershed and its sub-watersheds. The sub-watersheds were obtained by delineating contributing areas to each of the major impacted stream segments (Figure E2). Pine run_sub1240 and Pine run _sub1241 sub-watersheds consist of areas drained by stream segments of the main stem of Pine Run (Segment ID# 980210-1240-GLW 980211-1241-GLW). Pine run _sub1242 sub-watershed is the drainage area of the impaired segment of the unnamed tributary (Segment ID# 980210- 1242-GLW). In addition to sub-watershed delineations, a GIS analysis was performed to determine land use distributions in each sub-watershed. This data, along with watershed area and miles impaired in each sub-watershed, are needed for load reduction analyses.

Table E2. Loading Values for Pine Run Watershed, Year 1992 Land Use Conditions

Land Use Category Area (acres) Sediment Load Unit Area Sediment Load (lbs/year) (lbs/acre/yr) Hay/Pasture 657 25,320 38.54 Cropland 2,980 1,273,488 427.35 Coniferous Forest 86 110 1.27 Mixed Forest 612 1,722 2.81 Deciduous Forest 2,069 8,344 4.03 Unpaved Road 2 1,810 905.00 Transition 40 24,409 610.23 Low Intensity Dev 627 15,519 24.51 High Intensity Dev 212 6,452 30.43 Stream Bank 803,091 Groundwater Point Source Septic Systems Total 7,286 2,160,265 296.49

Table E3. Loading Values for Pine Run Watershed, Year 2000 Land Use Conditions

Land Use Category Area (acres) Sediment Load Unit Area Sediment Load (lbs/year) (lbs/acre/yr) Hay/Pasture 610 23,969 39.29 Cropland 2,040 746,981 366.17 Coniferous Forest 86 110 1.28 Mixed Forest 611 1,722 2.82 Deciduous Forest 1,967 7,572 3.85 Unpaved Road 2 1,854 750.99 Transition 1,106 2,441,347 2207.02 Low Intensity Developed 643 16,291 25.38 High Intensity Developed 221 5,629 25.61 Stream Bank 844,150 Groundwater Point Source Septic Systems Total 7,286 4,089,625 561.30

Table E4. Header Information for Tables E2 and E3.

The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDL for Pine Run watershed. TMDL and load reductions were performed separately depending of the origin of the source and type of pollutant. These computations were done for sediments emanating from upland erosion based on the unit watershed area and for those from stream bank erosion using unit stream length (in feet).

The equations defining TMDLs for sediments are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Pine Run to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

Pine run_sub1241

Pine run_sub1240

Pine run_sub1242

Stream segments Attained Not attained Unassessed

Figure E2. Pine Run Sub-watersheds

E3.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for Pine Run watershed are based on the sediment load obtained using the 1992 land use conditions, and are equal to 2,160,265 lbs/year (see Table E2).

E3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDLs were reserved as the MOS.

MOS (Sediments) 2,160,265 lbs/yr x 0.1 = 216,026 lbs/yr

E3.3 Load Allocation

The load allocation (LA), consisting of all sources in the watershed, was computed by subtracting the margin of safety. Waste load allocation (WLA), which is usually subtracted from the TMDL total load, was not done in this case because there is no waste load for sediment.

LA (Sediments) 2,160,265 lbs/yr – 216,026 lbs/yr = 1,944,239 lbs/yr (3)

E3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table E3 except those from the transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table E3,

LNR (Sediments) = 23,969 lbs/yr + 746,981 lbs/yr + 110 lbs/yr + 1,722 lb/yr + 7,572 lb/yr + 1,854 lbs/yr + 16,291 lbs/yr + 5,629 lbs/yr = 804,128 lbs/yr (4)

ALA (Sediments) = 1,944,239 lbs/yr –804,128 lbs/yr= 1,140,111 lbs/yr (5)

Table E5 below presents the TMDL for Pine Run watershed.

Table E5. Summary of TMDL for Pine Run Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 2,160,265 216,026 - 1,944,239 804,128 1,140,111 stream bank erosion

E4.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in Pine Run was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table E6. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 847,102 and 293,009 pounds, respectively for Pine Run to attain its specific uses.

Table E6. Load Allocation for Each Contributing Source in Pine Run Watershed.

Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 2,441,347 74 847,102 65 - Stream bank erosion 844,150 26 293,009 65 TOTAL 3,285,497 100 1,140,111 65

The next step is to allocate sediment loads for each contributing source to each of the two sub-watersheds. The sub-watershed allocation for transitional land use was performed by areal coverage of each sub-watershed. Load allocation for stream bank erosion was accomplished according to the mileage of impaired stream segments in each sub-watershed as recorded on the 2002 303 (d) List. Load allocations and corresponding reductions by sub-watershed are presented in Table E7. Table E8 provides sediment load allocation when all land uses in the Pine Run watershed are taken into consideration. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

The total allowable sediment load in Pine Run and its tributary when all land use/cover sources are considered is 1,944,239 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced to 1,944,239 pounds per year (i.e., a 52% reduction from the present load of 3,285,497 lbs/yr).

E4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing this TMDL using average annual conditions is protective of the waterbody.

E5.0 CONSIDERATION OF SEASONAL VARIATIONS

Table E7. Load Allocation of Sediments by Source for Each Sub-watershed Transitional Land Use Stream Bank Erosion TOTAL

Sub-watershed Area Ave Load Ave Red Stream Ave Ave Red Ave Ave Red ALA Length1 Load ALA Load ALA Acres Lbs/yr lbs/yr - % - Miles lbs/yr Lbs/yr - % - lbs/yr Lbs/yr - % - Pine run_sub1240 104 229,566 79,655 65 2.1 211,038 73,252 65 440,604 152,907 65 Pine run_sub1241 980 2,163,219 750,597 65 5.0 502,470 174,410 65 2,665,689 925,007 65 Pine run_sub1242 22 48,562 16,850 65 1.3 130,642 45,347 65 179,204 62,197 65 TOTAL 1,106 2,441,347 847,102 65 8.4 844,150 293,009 65 3,285,497 1,140,111 65 1This is the mileage (stream length) of impacted sediments in each sub-watershed

Table E8. Sediment Load Allocation by Each Land Use/Source

Land Use Category Area Unit Area Load Load ALA Reduction (acres) (lbs/acre/yr) (lbs/year) (lbs/year) (%) Hay/Pasture 657 39.29 23,969 23,969 0 Cropland 2,980 366.17 746,981 746,981 0 Coniferous Forest 86 1.28 110 110 0 Mixed Forest 612 2.82 1,722 1,722 0 Deciduous Forest 2,069 3.85 7,572 7,572 0 Unpaved Road 2 750.99 1,854 1,854 0 Transition 40 2207.02 2,441,347 847,102 65 Low Intensity Developed 627 25.38 16,291 16,291 0 High Intensity Developed 212 25.61 5,629 5,629 0 Stream Bank 844,150 293,009 65 Groundwater Point Source Septic Systems Total 7,286 561.30 4,089,625 1,944,239 52

E6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDL are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within the Pine Run Creek watershed will be affected by this policy, which has been included in Appendix E.

Implementation of BMPs aimed at sediment reduction will also assist in the reduction of phosphorus originating from transitional land uses and stream bank erosion. Other possibilities for attaining the desired reductions in sediment include streambank stabilization and fencing. Further field verification will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet the nutrient and sediment reductions outlined in this section.

E7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDLs. Notice of final TMDL approval will be posted on the Department website.

F. Total Maximum Daily Loads (TMDLs) Development Plan for Sub-basin #1 of West Branch Neshaminy Creek

Table of Contents Page

Executive Summary ………………………………………………………………... 94

F1.0 Introduction ……………………………………………………………….……. 95 F1.1 Watershed Description …………………………………………….….. 95 F1.2 Surface Water Quality ……………………………………………….… 96 F2.0 Approach to TMDL Development………………………………………...…….. 97 F2.1 Water/Flow Variability From Land Development ……………………... 97 F2.2. Flow Alterations From to Municipal Point Sources ……………………. 97 F2.3 Siltation Caused by Urban Runoff/Storm Sewers ……………….……. 98 F2.4 Watershed Assessment and Modeling………………………………….. 98 F3.0 Load Allocation Procedure for Sediment TMDL ……………………………… 100 F3.1 Sediment TMDL Total Load ………………………………………..….. 100 F3.2 Margin of Safety ………………………………………...…………... 101 F3.3. Load Allocation …………………………………………..………….. 101 F3.4. Adjusted Load Allocation …………………………………………… 101 F3.5. Load Reduction Procedures ………………………………..……….... 102 F4.0 Consideration of Critical Conditions ………………………………………… 103 F5.0 Consideration of Seasonal Variations …………………………………..…….. 103 F6.0 Reasonable Assurance of Implementation …………………………………… 103 F7.0 Public Participation …………………………………………………………. 103

List of Tables Page

F1. Physical Characteristics of West Branch Sub-basin #1………………………… 96 F2. Loading Values for West Branch Sub-basin #1, Year 1992 Land Use Conditions 99 F3. Loading Values for West Branch Sub-basin #1, Year 2000 Land Use Conditions 99 F4. Header Information for Tables F2 and F4………………………………..….. 100 F5. Summary of TMDL for West Branch Sub-basin #1 ……..…………..………… 101 F6. Load Allocation for each contributing source in West Branch Sub-basin #1…… 102 F7. Sediment Load Allocation by Land Use/Source ………………………... 102

List of Figures Page

F1. West Branch Sub-basin #1 ……………..……………………………………. 95

EXECUTIVE SUMMARY

West Branch Sub-basin #1 in Bucks County is approximately 2.5 square miles in size. West Branch Sub-basin #1 is a portion of the larger watershed surrounding West Branch Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 3.5 miles of this West Branch Neshaminy Creek tributary (Stream Segments ID# 980205-1330-GLW and 980205-1333-GLW). They were developed to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water act Section 303(d) Lists. The impairments are primarily caused by sediment loads and water/flow variability due to land development/urban runoff-storm sewers in the watershed, and by flow alterations as a result of municipal point sources. This TMDL focuses on control of sediments. Water/flow variability and flow alterations were not explicitly addressed because it was believed that the implementation of BMPs in the urban land use areas (High and Low Intensity Developed) to reduce sediment would also decrease peak water volumes to the stream, and therefore stabilize stream flow.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of West Branch Sub-basin #1. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining and Year 2000 when it was found to be impaired. Siltation, the cause of impairment in West Branch Sub-basin #1, resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the sediment loading that will meet the water quality objectives for West Branch Sub-basin #1 is 51,612 pounds per year. It is assumed that the West Branch Sub-basin #1 will support its aquatic life uses when this value is met. The sediment TMDL for West Branch Sub-basin #1 is allocated as shown in the table below.

Summary of TMDLs for West Branch Sub-basin #1 Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 134,267 13,427 - 120,840 69,228 51,612 stream bank erosion

The TMDL for sediments is allocated to non-point source from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 3.5 miles. The TMDL establishes a reduction for total sediment loading of 43% from the current annual loading of 210,596 pounds.

F1.0 INTRODUCTION

F1.1 Watershed Description

The following discussion provides information on the physical characteristics of West Branch Sub-basin #1, including location, land use distributions, and geology. West Branch Sub-basin #1 is located in the Piedmont physiographic province and is in Montgomery County. It covers an area of approximately 2.5 square miles. West Branch Sub-basin #1 drains into the West Branch of Neshaminy Creek from the west. The watershed is located north-east of the town of Lansdale, and is bounded by Pennsylvania Route 463 to the north and east and Route 63 to the west. Figure F1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the West Branch Sub-basin #1and is warm water fishes and migratory fishes.

The current land use distribution in West Branch Sub-basin #1 was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land use in West Branch Sub-basin #1 is developed land (81%). It is important to note that development in the watershed increased by 5% from 1992 to 2000.

Figure F1. West Branch Sub-basin #1.

West Branch Sub-basin #1 is underlain by a shale formation. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Watershed characteristics are summarized in Table F1.

F1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for West Branch Sub-basin #1 to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water Act Section 303(d) Lists (see Table A1 in section A1.0). It was first determined that West Branch Sub-basin #1 was not meeting its designated water quality uses for protection of aquatic life based on a 1994 aquatic biological survey, which included kick screen analysis and habitat surveys. In 2001, the Department again surveyed the stream again. In addition to impairments observed in 1994, one stream segment was found to be impaired by siltation and water flow variability. As a consequence of the surveys, Pennsylvania listed streams in West Branch Sub-basin #1 on the 1996 and 2002 Section 303(d) Lists of Impaired Waters.

Table F1. Physical Characteristics of West Branch Sub-basin #1

Physiographic Province Piedmont Area (square miles) 2.5 Predominant Land Use Developed land (81%) Predominant Geology Shale (100%) Soils Dominant HSGs C Average K Factor 0.37 20-Year Average Rainfall (in) 41.6 20-Year Average Runoff (in) 8.3

The 1996 303 (d) List reported 1.6 miles of streams in West Branch Sub-basin #1 to be impaired by flow alterations, water/flow variability, and siltation. The 2002 303(d). List added 1.8 miles. This list reported 3.4 miles of West Branch Sub-basin #1 streams to be impaired by flow alterations from municipal point sources, and by siltation and water/flow variability as a result of urban runoff/storm sewers and land development.

Stream segments of West Branch Sub-basin #1 are impacted by siltation as a result of “new land development” in the watershed. New land development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in West Branch Sub-basin #1 is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank

erosion during periods of intense storm events. As indicated above, land development has increased by approximately 5% between 1992 and 2000.

Sediments, which are often the cause of stream impairment in urban and suburban areas, are primarily from two sources: 1) disturbed land and unprotected soils at construction sites, and 2) stream channel erosion. Transitional land uses, mainly new construction sites, are one of the main sources of sediments in streams draining newly developed areas. Sediment production and sedimentation in streams are typically more important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of West Branch Sub-basin #1. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in West Branch Sub-basin #1. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur in the stream.

F2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDL addresses impairment by sediments in West Branch Sub-basin #1 stream segments as reported on the 2002 303(d) List. The flow alteration and water flow/variability impairments caused by municipal point sources and urban runoff/storm sewer will not be explicitly addressed by this TMDL because it is assumed that management practices that will be used to address storm water runoff and sediment production at new construction sites will reduce problems associated with flow variability as well. This TMDL was derived as follows:

F2.1 Water/Flow Variability from Land Development

F2.2 Flow Alterations from Municipal Point Sources

TMDLs were not determined for flow alterations. It was assumed that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow alterations within the watershed.

F2.3 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediments caused by newly developed land in the watershed were the cause of impairment of West Branch Sub-basin #1 stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macro-invertebrates. The TMDL for West Branch Sub-basin #1 addresses sediments from construction sites or “Transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

F2.4 Watershed Assessment and Modeling

The AVGWLF model was run for West Branch Sub-basin #1 to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for West Branch Sub-basin #1. Based on the use of 20 years of historical weather data, the mean annual loads for sediment for the 1992 and 2000 land use/ cover conditions were calculated as shown Tables F2 and F3, respectively. The Unit Area Load for sediment in the watershed was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area for the watershed. Table F4 presents an explanation of the

header information contained in Tables F2 and F3. Modeling output for West Branch Sub-basin #1 for 1992 and 2000 land use conditions is presented in Appendix F.

Table F2. Loading Values for West Branch Sub-basin #1, Year 1992 Land Use Conditions

Land Use Category Area Sediment Load Unit Area Sediment Load (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 32 728 22.75 Cropland 64 12,693 198.33 Coniferous Forest 10 0 0 Mixed Forest 101 221 2.19 Deciduous Forest 150 309 2.06 Transition 0 0 0 Low Intensity Dev 773 38,389 49.66 High Intensity Dev 430 19,316 44.92 Stream Bank 71,523 Groundwater Point Source Septic Systems Total 1,560 143,267 91.84

Table F3. Loading Values for West Branch Sub-basin #1, Year 2000 Land Use Conditions

Land Use Category Area Sediment Load Unit Area Sediment Load (acres) (lbs/year) (lbs/acre/yr) Hay/Past 32 751 23.47 Cropland 64 11,258 175.91 Coniferous Forest 10 0 0 Mixed Forest 101 221 2.19 Deciduous Forest 104 199 0.05 Transitional 69 69,051 1,000.74 Low Intensity Dev 754 37,660 49.95 High Intensity Dev 426 19,139 44.92 Stream Bank 72,407 Groundwater Point Source Septic Systems Total 1,560 210,596 134.99

Table F4. Header Information for Tables F2 and F3.

Land Use The land cover classification that was obtained by from the Category MRLC database Area (acres) The area of the specific land cover/land use category found in the watershed. Total Sediment The estimated total sediment loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area The estimated loading rate for sediment for a specific land Sediment Load cover/land use category. Loading rate is expressed in lbs/acre/year

F3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire West Branch Sub- basin #1. The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDL for West Branch Sub-basin #1.

The equations defining TMDLs for sediments are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for West Branch Sub-basin #1 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

F3.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for West Branch Sub-basin #1 are based on the sediment loads obtained using the 1992 land use conditions, and are equal to 134,267 lbs/year (see Table F2).

100

F3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

MOS (Sediments) 134,267 lbs/yr x 0.1 = 13,427 lbs/yr (3)

F3.3 Load Allocation

LA (Sediments) 134,267 lbs/yr - 13,427 lbs/yr = 120,840 lbs/yr (4)

F3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table F3 except those from the transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table F3,

LNR (Sediments) = 751 lbs/yr + 11,258 lbs/yr + 0 lbs/yr + 221 lb/yr + 199 lb/yr + 37,660 lbs/yr + 19,139 lbs/yr = 69,228 lbs/yr (5)

ALA (Sediments) = 120,8401 lbs/yr – 69,228 lbs/yr= 51,612 lbs/yr (6)

Table F5 below presents the TMDL for West Branch Sub-basin #1.

Table F5. Summary of TMDL for West Branch Sub-basin #1 (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 134,26 13,427 - 120,840 69,228 51,612 stream bank erosion 7

The ALA computed above is the portion of the load that is available to allocate among contributing land use/source areas as described in the next step. The following section shows the allocation process in detail for the entire watershed.

101

F3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in West Branch Sub- basin #1 was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table F6. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 25,194 and 26,418 pounds, respectively for West Branch Sub-basin #1 to attain its specific uses. In this instance, no sub- watersheds were delineated in this watershed because it was very small (2.5 square miles only). Therefore, separate sub-watershed load allocations were not performed.

Table F6. Load Allocation for Each Contributing Source in West Branch Sub-basin #1.

Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 69,051 49 25,194 64 - Stream bank erosion 72,407 51 26,418 64 TOTAL 141,458 100 51,612 64

Table F7. Sediment Load Allocation by Each Land Use/Source

Land Use Category Area Unit Area Load Load ALA Reduction (acres) (lbs/acre/yr) (lbs/year) (lbs/year) (%) Hay/Past 32 23.47 751 751 0 Cropland 64 175.91 11,258 11,258 0 Coniferous Forest 10 0 0 0 0 Mixed Forest 101 2.19 221 221 0 Deciduous Forest 104 0.05 199 199 64 Transitional 69 1,000.74 69,051 25,194 0 Low Int. Dev 754 49.95 37,660 37,660 0 High Int. Dev. 426 44.92 19,139 19,139 0

Stream Bank 72,407 26,418 64 Groundwater Point Source Septic Systems Total 1,560 134.99 210,596 120,840 43

The total allowable sediment load in West Branch Sub-basin #1 when all land use/cover sources are considered is 120,840 pounds per year. In order for the two stream segments to attain

102

their specific uses, total sediment load should be reduced from the current load of 210,596 pounds per year. Consequently, sediment load should be reduced by 43%.

F4.0 CONSIDERATION OF CRITICAL CONDITIONS

F5.0 CONSIDERATION OF SEASONAL VARIATIONS

F6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDL are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDLs. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. . These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within West Branch Sub-basin #1 will be affected by this policy, which has been included in Appendix E.

F7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

103

G. Total Maximum Daily Loads (TMDLs) Development Plan for West Branch Neshaminy Creek Sub-basin #2

104

Table of Contents Page

Executive Summary ………………………………………………………………... 107

G1.0 Introduction ……………………………………………………………….…. 108 G1.1 Watershed Description …………………………………………….…. 108 G1.2 Surface Water Quality ………………………………………………... 109 G2.0 Approach to TMDL Development………………………………………...…… 110 G2.1 Water/Flow Variability Due to Urban Runoff/Storm Sewers …………. 110 G2.2 Siltation Caused by Land Development ………….……………….……. 110 G2.3 Watershed Assessment and Modeling…………………………………. 111 G3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 113 G3.1 Sediment TMDL Total Load ………………………………………..… 113 G3.2 Margin of Safety ………………………………………...…………... 114 G3.3. Load Allocation …………………………………………..…………. 114 G3.4. Adjusted Load Allocation …………………………………..……….. 114 G3.5. Load Reduction Procedures ………………………………..………... 115 G4.0 Consideration of Critical Conditions ………………………………………… 116 G5.0 Consideration of Seasonal Variations …………………………………..…….. 116 G6.0 Reasonable Assurance of Implementation …………………………………… 116 G7.0 Public Participation …………………………………………………………. 117

105

List of Tables Page

G1. Physical Characteristics of West Branch Sub-basin #2………………………… 109 G2. Loading Values for West Branch Sub-basin #2, Year1992 Land Use Conditions 112 G3. Loading Values for West Branch Sub-basin #, Year 2000 Land Use Conditions 112 G4. Header Information for Tables G2 and G3………………………………..…. ... 113 G5. Summary of TMDL for West Branch Sub-basin #2………. … …………..….. 114 G6. Load Allocation for each contributing source in West Branch Sub-basin #2 .…. 115 G7. Sediment Load Allocation by Land Use/Source …………………………….. 115

List of Figures Page

G1. West Branch Sub-basin #2 ……………………………………………………. 108

106

EXECUTIVE SUMMARY

Sub-basin #2 of West Branch Neshaminy Creek is approximately 4 square miles in size. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 4.9 miles of this sub-basin (Stream Segments ID# 980202-1441-GLW). They were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) List. The impairments are primarily caused by sediment loads from land development in the watershed. The TMDL, therefore, focuses on control of sediments. This stream segment is also impacted by water/flow variability due to urban runoff/storm sewers. Water/flow variability was not explicitly addressed because it was believed that the implementation of BMPs in the urban land use areas (High and Low Intensity Developed) to reduce sediment would also decrease water flow and volume to the stream and therefore stabilize stream flow.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of West Branch Sub-basin #2. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining and Year 2000 when it was found to be impaired. Siltation, the cause of impairment in West Branch Sub-basin #2, resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the sediment loading that will meet the water quality objectives for West Branch Sub-basin #2 is 166,615 pounds per year. It is assumed that the West Branch Sub-basin #2 will support its aquatic life uses when this value is met. The sediment TMDL for West Branch Sub-basin #2 is allocated as shown in the table below.

Summary of TMDL for West Branch Sub-basin #2 (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 328,477 32,848 - 295,629 129,014 166,615 stream bank erosion

The TMDL for sediment is allocated to non-point source from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 4.9 miles. The TMDL establishes a reduction for total sediment loading of 57% from the current annual loading of 682,119 pounds.

107

G1.0 INTRODUCTION

G1.1 Watershed Description

The following discussion provides information on the physical characteristics of West Branch Sub-basin #2, including its location, land use distributions, and geology. West Branch Sub-basin #2 is located in the Piedmont physiographic province and is almost entirely located in Montgomery County. It covers an area of approximately 4 square miles. West Branch Sub-basin #2 drains into West Branch of Neshaminy Creek from the west. The watershed is located east of the town of Lansdale and is bounded by Pennsylvania Route 463 to the north and east and Route 63 to the west. Figure G1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the West Branch Sub-basin #2 is warm water fishes and migratory fishes.

The current land use distribution in West Branch Sub-basin #2 was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land uses in West Branch Sub-basin #2 are developed land (63%) and forested land (29%). It is important to note that development in the watershed changed from 1453 to 1584 acres, or a 9% increase from 1992 to 2000.

Figure G1. West Branch Sub-basin #2.

108

West Branch Sub-basin #2 is underlain by a shale formation. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Watershed characteristics are summarized in Table G1.

G1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for West Branch Sub-basin #2 to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). It was first determined that stream reaches in West Branch Sub-basin #2 were not meeting their designated water quality uses for protection of aquatic life in 2001 based on an aquatic biological survey. As a consequence, streams in West Branch Sub-basin #2 were included on the 2002 Section 303(d) List of Impaired Waters.

Table G1. Physical Characteristics of West Branch Sub-basin #2.

Physiographic Province Piedmont Area (square miles) 4 Predominant Land Use Developed land (63%) Forested land (29%) Predominant Geology Shale (100%) Soils Dominant HSGs C Average K Factor 0.37 20-Year Average Rainfall (in) 40.4 20-Year Average Runoff (in) 5.4

The 2002 303 (d) List reported 1.7 miles of streams in West Branch Sub-basin #2 (Stream Segment ID# 980202-1441-GLW) to be impaired by siltation from land development and water/flow variability as a result of urban runoff/storm sewers. This particular stream reach is impacted by siltation as a result of “new land development” in the watershed. New land development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in West Branch Sub-basin #2 is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events. As indicated above, land development has increased by approximately 9% between 1992 and 2000.

109

Sediments, which are often the cause of stream impairment in urban and suburban areas, are primarily from two sources: 1) disturbed land and unprotected soils at construction sites, and 2) stream channel erosion. Transitional land uses, mainly new construction sites, are one of the main sources of sediments in streams draining newly developed areas. Sediment production and sedimentation in streams are typically most important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of West Branch Sub-basin #2. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in West Branch Sub-basin #2. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur in the stream.

G2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDL addresses impairment by sediment in West Branch Sub-basin #2 stream segments as reported on the 2002 303(d) Lists. The stream water flow variability impairment caused by urban runoff/storm sewer will not be explicitly addressed by this TMDL because it is assumed that management practices that will be used to address storm water runoff and sediment production at new construction sites will reduce problems associated with flow variability as well. This TMDL was derived as follows:

G2.1 Water/Flow Variability Resulting from Urban Runoff/Storm Sewers

G2.2 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediments originating from newly developed land in the watershed were the cause of impairment of West Branch Sub-basin #2 stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom- dwelling macro-invertebrates. The TMDL for West Branch Sub-basin #2 addresses sediment from construction sites or “Transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a

110

method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

G2.3 Watershed Assessment and Modeling

The AVGWLF model was run for West Branch Sub-basin # 2 to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for West Branch Sub-basin #2 . Based on the use of 20 years of historical weather data, the mean annual loads for sediment for the 1992 and 2000 land use/ cover conditions were calculated as shown in Tables G2 and G3, respectively. The Unit Area Load for sediment in the watershed was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area for the watershed. Table G4 presents an explanation of the header information contained in Tables G2 and G3. Modeling output for West Branch Sub-basin #2 for 1992 and 2000 land use conditions is presented in Appendix F.

111

Table G2. Loading Values for West Branch Sub-basin #2, Year 1992 Land Use Conditions

Land Use Category Area Sediment Load Unit Area Sediment (acres) (lbs/year) Load (lbs/acre/yr) Hay/Pasture 104 3,929 37.78 Cropland 274 113,333 413.62 Coniferous Forest 57 154 2.70 Mixed Forest 163 487 2.98 Deciduous Forest 570 2,230 3.91 Transition 0 0 0 Low Intensity Dev 1,052 79,271 73.35 High Intensity Dev 301 14,481 48.11 Stream Bank 114,570 Groundwater Point Source Septic Systems Total 2,521 328,477 130.30

Table G3. Loading Values for West Branch Sub-basin #2, Year 2000 Land Use Conditions

Land Use Category Area Sediment Load Unit Area Sediment (acres) (lbs/year) Load (lbs/acre/yr) Hay/Pasture 89 3,091 34.73 Cropland 114 24,840 217.89 Coniferous Forest 54 154 2.86 Mixed Forest 163 487 2.99 Deciduous Forest 516 2,053 3.98 Transitional 187 425,717 2,276.56 Low Intensity Dev 1091 82,914 76.00 High Intensity Dev 306 15,475 50.57 Stream Bank 117,779 Groundwater Point Source Septic Systems Total 2,521 682,119 270.57

112

Table G4. Header Information for Tables G2 and G3.

G3.0 LOAD ALLOCATION PROCEDURE FOR NUTRIENT AND SEDIMENT TMDLs

The load allocation and reduction procedures were applied to the entire area within West Branch Sub-basin #2. Sub-watersheds were not delineated because of the watershed’s small size (4 square miles). The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDL for West Branch Sub-basin #2.

The equations defining TMDLs for sediments are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for West Branch Sub-basin #2 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

G3.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for West Branch Sub-basin #2 are based on the sediment loads obtained using the 1992 land use conditions, and are equal to 328,477 lbs/yr (see Table G2).

113

G3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

MOS (Sediments) 328,477 lbs/yr x 0.1 = 32,848 lbs/yr (3)

G3.3 Load Allocation

The load allocation (LA), consisting of all sources in the watershed, was computed by subtracting the margin of safety. Waste load allocation (WLA), which is usually subtracted from the TMDL total load, was not in this case because there is no waste load for sediment.

LA (Sediments) 328,477 lbs/yr - 32,848 lbs/yr = 295,629 lbs/yr (4)

G3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table G3 except those from the transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table G3,

LNR (Sediments) =3,091 lbs/yr + 24,840 lbs/yr + 154 lbs/yr +487 lb/yr + 2,053 lb/yr + 82,914 lbs/yr + 15,475 lbs/yr = 129,014 lbs/yr (5)

ALA (Sediments) = 295,629 lbs/yr – 129,014 lbs/yr= 166,615 lbs/yr (6)

Table G5 below presents the TMDL for West Branch Sub-basin #2.

Table G5. Summary of TMDL for West Branch Sub-basin #2 (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 328,477 32,848 - 295,629 129,014 166,615 stream bank erosion

The ALA computed above is the portion of the load that is available to allocate among contributing land use/sources as described in the next step. The following section shows the allocation process in detail for the entire watershed.

114

G3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in West Branch Sub- basin #2 was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table G6. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 130,508 and 36,107 pounds, respectively for West Branch Sub-basin #2 to attain its specific uses.

Table G6. Load Allocation for Each Contributing Source in West Branch Sub-basin #2.

Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 425,717 78 130,508 69 - Stream bank erosion 117,779 22 36,107 69 TOTAL 543,496 100 166,615 69

Table G7 provides sediment load allocation when all land uses in West Branch Sub-basin #2 are taken into consideration. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table G7. Sediment Load Allocation by Each Land Use/Source

Land Use Category Area Unit Area Load Load ALA Reduction (acres) (lbs/acre/yr) (lbs/year) (lbs/year) (%) Hay/Pasture 89 34.73 3,091 3,091 0 Cropland 114 217.89 24,840 24,840 0 Conifer Forest 54 2.86 154 154 0 Mixed Forest 163 2.99 487 487 0 Decid Forest 516 3.98 2,053 2,053 0 Transition 187 2,276.56 425,717 130,508 69 Low Intensity Dev 1091 76.00 82,914 82,914 0 High Intensity Dev 306 50.57 15,475 15,475 0 Stream Bank 117,779 36,107 69 Groundwater Point Source Septic Systems Total 2,521 270.57 682,119 295,629 57

115

The total allowable sediment load in West Branch Sub-basin #2 when all land use/cover sources are considered is 295,629 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from the current load of 682,119 pounds per year. Consequently, sediment load should be reduced by 57%.

G4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing the TMDL using average annual conditions is protective of the waterbody.

G5.0 CONSIDERATION OF SEASONAL VARIATIONS

G6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDLs are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDLs. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. . These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within West Branch Sub-basin #2 will be affected by this policy, which has been included in Appendix E.

116

determine the most cost-effective and environmentally protective combination of BMPs required to meet the nutrient and sediment reductions outlined in this report.

G7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

117

H. Total Maximum Daily Loads (TMDLs) Development Plan for Sub-Basin #3 of West Branch Neshaminy Creek

118

Table of Contents Page

Executive Summary ………………………………………………………………... 121

H1.0 Introduction ……………………………………………………………….……. 122 H1.1 Watershed Description …………………………………………….…. 122 H1.2 Surface Water Quality………………………………………………… 123 H2.0 Approach to TMDL Development………………………………………...…….. 124 H2.1 TMDL Endpoints ……………………………………………………… 124 H2.2 Selection of the Reference Watershed………………………………... 125 H2.3 Nutrient Loads and Organic Enrichment in Stream Systems ………….. 126 H2.4 Watershed Assessment and Modeling…………………………………….. 126 H3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 128 H3.1 Sediment TMDL Total Load ………………………………………..….. 129 H3.2 Margin of Safety ………………………………………...……. 129 H3.3. Load Allocation …………………………………………..….. 129 H3.4. Adjusted Load Allocation …………………………………..… 129 H3.5. Load Reduction Procedures ………………………………..…. 130 H4.0 Consideration of Critical Conditions ………………………………………… 131 H5.0 Consideration of Seasonal Variations …………………………………..…….. 132 H6.0 Reasonable Assurance of Implementation …………………………………… 132 H7.0 Public Participation …………………………………………………………. 132

119

List of Tables Page

H1. Physical Characteristics of West Branch Sub-basin #3……………………………. 123 H2. Loading Values for West Branch Sub-basin #3…………………………………... 127 H3. Loading Values for the Reference Watershed…………….……………….…. 127 H4. Header Information for Tables H2 and H3………………………………..……… 128 H5. TMDL Total Load Computation…………………………………………………. 129 H6. Summary of TMDLs for West Branch Sub-basin #3……………. …………....… 130 H7. Load Allocation for each contributing source in West Branch Sub-basin #3……… 133 H8. Sediment Load Allocation by Land Use/Source ……………………….………. 134

List of Figures Page

H1. West Branch Sub-basin #3…….. ………………………………………….……… 122 H2 Reference Watershed ……………………………………………………………… 125

120

EXECUTIVE SUMMARY

West Branch Sub-basin #3 in Bucks County is approximately 4 square miles in size and is a tributary of West Branch Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to 8.5 miles of the streams within this particular sub-basin (Stream Segments ID# 980202-1313-GLW and 20010426-1235-GLW). They were developed to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water act Section 303(d) List. The impairments are primarily caused by nutrient and sediment loads from land development and agriculture in the watershed. The TMDL focuses on control of nutrients and sediments. Phosphorus is generally considered to be the limiting nutrient in a water body when the nitrogen/phosphorus ratio exceeds 10 to 1. In West Branch Sub-basin #3, this ratio is 13 to 1.

Pennsylvania does not currently have water quality criteria for nutrients and sediments. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for nutrients and sediments in the impaired segments of the West Branch Sub-basin #3. Based upon comparison to a similar, non-impaired watershed, it was estimated that the amount of phosphorus loading that will meet the water quality objectives for West Branch Sub-basin #3 is 632 pounds per year. Sediment loading must be limited to 436,481 pounds per year. It is assumed that West Branch Sub-basin #3 will support its aquatic life uses when these values are met. The TMDLs for West Branch Sub-basin #3 are allocated as shown in the table below.

Summary of TMDLs for West Branch Sub-basin #3 (lbs/yr) Pollutant TMDL MOS WLA LA LNR ALA Phosphorus 1,233 123 0 1,110 498 632 Sediments 496,654 49,665 - 446,989 10,508 436,481

The TMDLs are allocated to non-point source from agricultural and land development activities, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to agricultural land use and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The TMDLs cover a total of 8.5 miles of streams in West Branch Sub-basin #3. The TMDL establishes a reduction for phosphorus loading from agricultural and land development activities of 21% from the current annual loading of 1,437 pounds, and a reduction in sediment loading of 52% from the current annual loading of 930,419 pounds.

121

H1.0 INTRODUCTION

H1.1 Watershed Description

The following discussion provides information on the physical characteristics of West Branch Sub-basin #3 including its location, land use distributions, and geology. West Branch Sub-basin #3 is located in the Piedmont physiographic province and in Bucks County. It covers an area of approximately 4 square miles. The streams in West Branch Sub-basin #3 drain into the main stem of West Branch Neshaminy Creek from the north. The sub-basin is located east of the town of New Britain and is bounded by Pennsylvania Route 309 to the west, Route 152 to the east, and Route 202 to the south. Figure H1 shows the sub-basin boundary and its location. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the streams in West Branch Sub-basin #3 is warm water fishes and migratory fishes.

The primary land uses in West Branch Sub-basin #3 are agriculture (53%) and forested land (45%), with areas adjacent to the stream primarily used for cropland and pasture. It was found also from a field survey of the watershed that cattle generally have free access to the stream. The majority of the streams had no protected riparian zone. The 1994 survey showed that nutrients from agricultural activities were causing increased algae growth. It also found that sediment deposited in large quantities on the streambed was degrading the habitat of bottom-dwelling macroinvertebrates.

Figure H1. West Branch Sub-basin #3.

122

In terms of subsurface geology, West Branch Sub-basin #3 is primarily underlain by a shale formation. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Characteristics of Sub-basin #3, as well as those of the reference watershed chosen for further analysis (see later discussion), are summarized in Table H1.

Table H1. Physical Characteristic Comparisons between West Branch Sub-basin #3 and Reference Watershed

Attribute West Branch Sub-basin #3 Reference Watershed Watershed Physiographic Province Piedmont Piedmont Area (square miles) 10 4 Predominant Land Uses -Agriculture (51%) -Agriculture (47%) -Forested land (45%) -Forested land (51%) Predominant Geology Shale (100%) Shale (100%) Soils - Dominant HSG C C - K Factor 0.37 0.37 20-Year Average Rainfall (in) 40.4 43.0 20-Year Average Runoff (in) 3.8 3.8

H1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for West Branch Sub-basin #3 to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water Act Section 303(d) Lists (see Table A1 in section A1.0). It was first determined that West Branch Sub-basin #3 was not meeting its designated water quality uses for protection of aquatic life in 1994 based on aquatic biological survey. The 2001 survey found that that the stream was still impaired. As a consequence, Pennsylvania listed West Branch Sub-basin #3 on the 1996 and 2002 Section 303(d) List of Impaired Waters.

The 1996 303 (d) List reported 3.3 miles of West Branch Sub-basin #3 (Stream Segment ID# 980202-1313-GLW) to be impaired by siltation from construction and agriculture, and excessive algae growth as a result of agricultural activities in the watershed. The 2002 303 (d) List reported an additional 5.2 miles (Stream Segment ID# 210426-1235-GLW) to be impaired by siltation from agriculture and land development.

123

sedimentation in streams are typically important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of West Branch Sub-basin #3. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in West Branch Sub-basin #3. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur in the stream.

H2.0 APPROACH TO TMDL DEVELOPMENT

H2.1 TMDL Endpoints

The TMDLs described herein address phosphorus and sediments. Phosphorus was determined to be the nutrient limiting plant growth in West Branch Sub-basin #3. Because neither Pennsylvania nor EPA has water quality criteria for phosphorus or sediments, we had to develop a method to determine water quality objectives for these parameters that would result in the impaired stream segments attaining their designated uses. The method employed for these TMDLs is termed the “reference watershed approach.”

With the reference watershed approach, a pair of watersheds is compared; with one attaining its uses and one that is impaired based on biological assessment. Both watersheds must have similar land use/cover distributions. Other features such as base geologic formation should be matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of the process is to reduce the loading rate of nutrients and sediments in the watershed containing the impaired stream segment(s) to a level equivalent to or slightly lower than the loading rate in the watershed with the non-impaired, reference stream segment(s). The underlying assumption is that this load reduction will allow biological health to return to the impaired stream segments.

The TMDL endpoints established for this analysis were determined using a sub-watershed of the North Branch Neshaminy Creek watershed as the reference watershed. These endpoints are discussed in detail in the TMDL section. The listing for impairment caused by nutrients and siltation is addressed through reduction of the phosphorus load. A detailed explanation of this process is included in the following section.

124

H2.2 Selection of the Reference Watershed

In general, three factors should be considered when selecting a suitable reference watershed. The first factor is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to attain water quality standards. The second factor is to find a watershed that closely resembles the watershed being assessed (i.e., West Branch Sub-basin #3) with respect to physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of desktop screening using several GIS coverages including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The watershed used as a reference for West Branch Sub-basin #3 was obtained by screen- digitizing a subwatershed of the North Branch Neshaminy Creek atershed. This watershed is also located in the Piedmont province in State Water Plan (SWP) Basin 2F. Table H1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure H2 shows the reference watershed boundary and its location in Bucks County.

An analysis of the MRLC land use/cover layer revealed that land cover/use distributions in both watersheds are very similar. The surficial geology of both West Branch Sub-basin #3 and the reference watershed is shale. A look at these attributes in Table H1 indicates that these watersheds also compare very well in terms of average runoff, precipitation, and soil K factor.

Figure H2. Reference Watershed Location.

125

H2.3 Nutrient Loads and Organic Enrichment in Stream Systems

As indicated earlier, West Branch Sub-basin #3 was listed as being impaired due to problems associated with nutrient loads and siltation. In stream systems, elevated nutrient loads (nitrogen and phosphorus in particular) can lead to increased productivity of plants and other organisms (Novotny and Olem, 1994).

Typically in aquatic ecosystems the quantities of trace elements are plentiful; however, nitrogen and phosphorus may be in short supply. The nutrient that is in the shortest supply is called the limiting nutrient because its relative quantity affects the rate of production (growth) of aquatic biomass. If the nutrient load to a water body can be reduced, the available pool of nutrients that can be utilized by plants and other organisms will be reduced and, in general, the total biomass can subsequently be decreased as well (Novotny and Olem, 1994). In most efforts to control eutrophication processes in water bodies, emphasis is placed on the limiting nutrient. This is not always the case, however. For example, if nitrogen is the limiting nutrient, it still may be more efficient to control phosphorus loads if the nitrogen originates from difficult to control sources such as nitrates in ground water.

In most fresh water bodies, phosphorus is the limiting nutrient for aquatic growth. In some cases, however, the determination of which nutrient is the most limiting is difficult. For this reason, the ratio of the amount of N to the amount of P is often used to make this determination (Thomann and Mueller, 1987). If the N/P ratio is less than 10, nitrogen is limiting. If the N/P ratio is greater than 10, phosphorus is the limiting nutrient. In the case of West Branch Sub- basin #3, the N/P ratio is approximately 13, which points to phosphorus as the limiting nutrient. It is therefore assumed that controlling the phosphorus loading to West Branch Sub-basin #3 will limit plant growth and result in raising the dissolved oxygen level.

H2.4 Watershed Assessment and Modeling

The AVGWLF model was run for both West Branch Sub-basin #3 and the reference watershed to establish existing loading conditions under existing land cover use conditions in each watershed. Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for West Branch Sub-basin #3. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the impaired and reference watersheds were calculated as shown Tables H2 and H3, respectively. Table H4 presents an explanation of the header information contained in Tables H2 and H3. Modeling output for West Branch Sub-basin #3 and the reference watershed is presented in Appendix F.

126

Table H2. Loading Values for West Branch Sub-basin #3

Unit Area Land Use Area Total P Unit Area P Total N Unit Area N Sediment Sediment Category (acres) (lbs/yr) Load (lbs/yr) Load Load Load (lbs/acre/ yr) (lbs/acre/yr) (lbs/year) (lbs/acre/yr) Hay/Past 200 44 0.22 367 1.84 7,108 35.54 Cropland 1,089 827 0.76 5,317 4.87 706,203 648.49 Coniferous For 27 0 0.00 3 0.11 44 1.63 Mixed For 237 1 0.00 28 0.12 795 3.35 Deciduous For 881 6 0.00 107 0.12 4,128 4.69 Transition 7 6 0.86 52 7.4 5,541 791.57 Lo Int Dev 64 0 0.00 0 0 1,302 20.34 Hi Int Dev 123 0 0.00 0 0 221 18.41 Stream Bank 68 310 205,077 Groundwater 476 9,684 Point Source 0 0 Septic Systems 9 2,329 Total 2,517 1,437 0.57 18,195 7.23 930,419 369.65

Table H3. Loading Values for the Reference Watershed

Unit Area Land Use Area Total P Unit Area P Total N Unit Area N Sediment Sediment Category (acres) (lbs/yr) Load (lbs/yr) Load Load Load (lbs/acre/ yr) (lbs/acre/ yr) (lbs/year) (lbs/acre/yr) Hay/Past 185 44 0.24 364 1.97 6,578 35.56 Cropland 968 558 0.58 4,033 4.17 325,960 336.74 Coniferous For 54 0 0.00 7 0.13 132 2.44 Mixed For 131 1 0.00 16 0.13 463 3.53 Deciduous For 1074 8 0.00 143 0.13 6,623 6.17 Lo Int Dev 40 0 0.00 0 0.00 2,008 50.20 Hi Int Dev 2 0 0.00 0 0.00 66 33.0 Stream Bank 46 213 142,384 Groundwater 520 10,574 Point Source 0 0 Septic Systems 23 2,146 Total 2,454 1,200 0.49 17,496 7.13 484,214 197.32

127

Table H4. Header Information for Tables H2 and H3.

Land Use Category The land cover classification that was obtained by from the MRLC database Area (acres) The area of the specific land cover/land use category found in the watershed. Total P The estimated total phosphorus loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area P Load The estimated loading rate for phosphorus for a specific land cover/land use category. Loading rate is expressed in lbs/acre/year Total N The estimated total nitrogen loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area N Load The estimated loading rate for nitrogen for a specific land cover/land use category. Loading rate is expressed in lbs/acre/year Total Sediment The estimated total sediment loading that reaches the outlet point of the watershed that is being modeled. Expressed in lbs./year. Unit Area Sediment The estimated loading rate for sediment for a specific land cover/land use Load category. Loading rate is expressed in lbs/acre/year

H3.0 LOAD ALLOCATION PROCEDURE FOR NUTRIENT AND SEDIMENT TMDLs

The load allocation and reduction procedures were applied to all of West Branch Sub-basin #3. The watershed was so small that we did not subdivide it into sub-watersheds. Therefore, sub-watershed load allocations were not performed.

The load reduction calculations in West Branch Sub-basin #3 are based on the current loading rates for phosphorus and sediments in the reference watershed. Based on biological assessment, it was determined that the reference watershed was attaining its designated uses. The phosphorus and sediment loading rates were computed for the reference watershed using the AVGWLF model. These loading rates were then used as the basis for establishing the TMDLs for West Branch Sub-basin #3.

The equations defining TMDLs for sediments and nutrients are as follows:

TMDL = MOS + LA + WLA (1) LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for West Branch Sub-basin #3 to meet water quality goals. Therefore, it is the load that originates from agricultural sources that

128

have contributed to water quality problems encountered in the watershed. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

H3.1 TMDL Total Loads

The TMDL loads for both pollutants of concern were computed in the same manner. The first step is to determine the TMDL total target load for West Branch Sub-basin #3, the impaired watershed. This value was obtained by multiplying each pollutant unit loading rate in the reference watershed by the total watershed area of West Branch Sub-basin #3. This information is presented in Table H5.

Table H5 TMDL Total Load Computation

Unit Area Loading Rate Total Watershed Area in in Reference Watershed West Branch Sub-basin #3 TMDL Total Load Type of Pollutant (lbs/acre/yr) (acres) (lbs/yr) Phosphorus 0.49 2,517 1,233 Sediment 197.32 2,517 496,654

H3.2. Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of each of the TMDLs was reserved as the MOS.

Phosphorus - 1,233 lbs/yr x 0.1 = 123 lbs/yr (3)

Sediment - 496,654lbs/yr x 0.1 = 49,665 lbs/yr (4)

H3.3 Load Allocation

The load allocation (LA), consisting of all nonpoint sources in the watershed, was computed by subtracting the margin of safety and the waste load allocation (WLA) from the TMDL total load. (Notice that sediments do not have a waste load allocation).

LA (Phosphorus) = 1,233 lbs/yr – 123 lbs/yr = 1,110 lbs/yr (5)

LA (Sediments) = 496,654 lbs/yr – 49,665 lbs/yr = 446,989 lbs/yr (6)

H3.4 Adjusted Load Allocation

129

those from agricultural land uses (Hay/Pasture, Row Crops), land development, and stream bank erosion. Therefore, using data in Table H2,

LNR (Phosphorus) = 1 lbs/yr + 6 lbs/yr + 6 lbs/yr + 476 lb/yr +9 lbs/yr = 498 lb/yr (7)

ALA (Phosphorus) = 1,110 lbs/yr - 566 lbs/yr= 632 lbs/yr (8)

LNR (Sediments) = 44 lbs/yr + 795 lbs/yr + 4,128 lb/yr + 5,541 lb/yr = 10,508 lbs/yr (9)

ALA (Sediments) = 446,989lbs/yr – 10,508 lbs/yr = 436,481 lbs/yr. (10)

Table H6 below presents the TMDLs for West Branch Sub-basin #3.

Table H6. Summary of TMDLs for West Branch Sub-basin #3 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Phosphorus 1,233 123 0 1,110 498 632 Sediments 496,654 49,665 - 446,989 10,508 436,481

The ALA computed above is the portion of the load that is available to allocate among contributing sources (Hay/Pasture, Cropland), land development and streambank erosion as described in the next step. Not all land use/source categories were included in the allocation because they are difficult to control, or provide an insignificant portion of the total load (e.g., transition land use). The following section shows the allocation process in detail for the entire watershed.

H3.5 Load Reduction Procedures

Sediment loads obtained in the previous step were allocated among the remaining land use/sources of the impaired watershed according to the Equal Marginal Percent Reduction (EMPR) method. EMPR is carried out using an Excel Worksheet in the following manner:

1) Each land use/source load is compared with the total allocable load to determine if any contributor would exceed the allocable load by itself. The evaluation is carried out as if each source is the only contributor to the pollutant load to the receiving waterbody. If the contributor exceeds the allocable load, that contributor would be reduced to the allocable load. This is the baseline portion of EMPR.

2) After any necessary reductions have been made in the baseline the multiple analysis is run. The multiple analysis will sum all of the baseline loads and compare them to the total allocable load. If the allocable load is exceeded, an

130

equal percent reduction will be made to all contributors’ baseline values. After any necessary reductions in the multiple analysis, the final reduction percentage for each contributor can be computed.

It is important to note that:

1) Load allocation to sub-watersheds was not performed due to the fact that West Branch Sub-basin #3 is very small and therefore could not be subdivided into meaningful subwatersheds;

2) Land development and stream bank erosion were lumped together as one sediment pollutant source, “land development”, because sediments in urban and suburban areas are from disturbed land and unprotected soils at construction sites, and from stream channel erosion. Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the West Branch Sub- basin #3.

The load allocation and EMPR procedures were performed using an Excel Worksheet and results are presented in Appendix G. Results of the load allocation by land use sources are presented in Table H7. Table H8 provides load allocation by considering all land uses in West Branch Sub-basin #3. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

The total allowable P and sediment loads in West Branch Sub-basin #3 when all land use/cover sources are considered is 1,130 pounds and 446,989 lb per year, respectively. In order for all stream segments to attain their specific uses, total P and sediment loads should be reduced from 1,437 lb and 960,419 pounds per year, respectively. Consequently, P sediment loads should be reduced by 21% and 52%, respectively.

H4.0 CONSIDERATION OF CRITICAL CONDITIONS

131

H5.0 CONSIDERATION OF SEASONAL VARIATIONS

H6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDLs are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDLs. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within West Branch Sub-basin #3 will be affected by this policy, which has been included in Appendix E.

Implementation of BMPs aimed at sediment reduction will also assist in the reduction of phosphorus originating from transitional land uses and stream bank erosion. Other possibilities for attaining the desired reductions in phosphorus and sediment include streambank stabilization and fencing. Further ground verification will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet the nutrient and sediment reductions outlined in this report.

H7.0 PUBLIC PARTICIPATION

132

H7. Load Allocation by Each Contributing Source in West Branch Sub-basin #3.

Land Use/ Area Phosphorus Sediments Source Loading Average Average Reductio Loading Average Average Reduction Rate Load ALA n Rate Load ALA (Acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) - % - (lbs/ac/yr) (lbs/yr) (lbs/yr) - % - Hay/Past 200 0.22 44 37 15 35.54 7,108 4,772 33 Cropland 1,089 0.76 827 537 35 648.49 706,203 293,016 59 Land 187 0.78 68 58 15 1,104.81 206,600 138,693 33 Develop1 Sub-total 1,476 0.64 939 632 33 623.25 919,911 436,481 53

1Loads from land development include the amount from stream bank erosion.

133

Table H8. Load Allocation by Each Land Use/Source

Phosphorus Sediment Unit Area Unit Area Annual ALA Loading Annual ALA Area Loading average (annual Reduction Rate average (annual Reduction Source Rate load average) load average) (Acres) (lbs/ac) (lbs/yr) (lbs/year) - % - (lbs/ac/yr) (lbs/yr) (lbs/yr) - % -

Hay/Past 200 0.22 44 37 15 35.54 7,1084,772 Cropland 1,089 0.76 827 537 35 648.49 706,203293,016 Coniferous 27 0.00 0 0 0 1.63 44 44 0 Mixed For 237 0.00 1 1 0 3.35 795 795 0 Deciduous 881 0.00 6 6 0 4.69 4,128 4,128 0 Transition 7 0.86 6 6 0 791.57 5,5415,541 0 Lo Int Dev 64 0.00 0 0 0 20.34 1,302 874 23 Hi Int Dev 123 0.00 0 0 0 18.41 221 148 23 Stream Bank 68 58 15 205,077 137,671 23 Groundwater 476 476 0 Point Source 0 0 0 Septic Systems 9 9 0 2,517 0.57 1,437 1,130 21 369.65 930,419 446,989 52

134

I. Total Maximum Daily Loads (TMDLs) Development Plan for Sub-Basin #4 of West Branch Neshaminy Creek

135

Table of Contents Page

Executive Summary ………………………………………………………………... 138

I1.0 Introduction ……………………………………………………………….……. 139 I1.1 Watershed Description …………………………………………..….…. 139 I1.2 Surface Water Quality ……………………………………………..…... 139 I2.0 Approach to TMDL Development…………………………………….……….. 140 I2.1 TMDL Endpoints……………………………………………………….. 140 I2.2 Selection of the Reference Watershed…………………………………….. 141 I2.3 Water/Flow Variability and Flow Alterations Resulting from Urban Runoff/Storm Sewers and Land Development………………….. 143 I2.4 Siltation due to Agricultural Sources and Land Development…………… 143 I2.5 Excessive Algae Growth from Agriculture (Organic Enrichment in Stream Systems)……………………………………….. 143 I2.6 Excess Algae Growth due to Municipal Point Sources……………….. 143 I2.7 Watershed Assessment and Modeling………………………………… 144 I3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 146 I3.1 TMDL Total Load …………..……………………………………..….. 147 I3.2 Margin of Safety ………………………………………...……. 147 I3.3. Load Allocation …………………………………………..….. 147 I3.4. Adjusted Load Allocation …………………………………..… 148 I3.5. Load Reduction Procedures ………………………………..…. 148 I4.0 Consideration of Critical Conditions ………………………………………..… 150 I5.0 Consideration of Seasonal Variations …………………………………..……... 150 I6.0 Reasonable Assurance of Implementation …………………………………….. 151 I7.0 Public Participation …………………………………………………………. 151

136

List of Tables Page

I1. Physical Characteristic comparisons Between West Branch Sub-basin #4 and Reference Watershed ………………………. 140 I2. Loading Values for West Branch Sub-basin #4…………….……………… 145 I3. Loading Values for the Reference Watershed…………….………. 145 I4. Header Information for Tables I2 and I3………………………………..…. 146 I5. Summary of TMDLs for West Branch Sub-basin #4 ……………………....… 147 I6. Summary of Load Allocation for West Branch Sub-basin #4…………………. 148 I7. Load Allocation for each Contributing Source in West Branch Sub-basin#4…… 149 I8. Sediment Load Allocation by Land Use/Source ………………………... 150

List of Figures Page

I1. West Branch Sub-basin #4……..………………………………….…. 139 I2. Reference Watershed …………………………………………………… 140

137

EXECUTIVE SUMMARY

West Branch Sub-basin #4 is about 15 square miles in size, and is evenly divided between Bucks and Montgomery Counties. This particular sub-basin consists of the main stem of West Branch of Neshaminy Creek and several unnamed tributaries. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 22.8 miles of streams within this basin (Stream Segment ID#s 980205-1211-GLW, 980202-1040-GLW, 980202-1043-GLW, and 980205-1430-GLW). They were developed to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water act Section 303(d) Lists. The impairments are primarily caused by nutrient and sediment loads from land development and agriculture in the watershed. The TMDL, therefore, focuses on control of nutrients and sediments. Phosphorus is generally considered to be the limiting nutrient in a waterbody when the nitrogen/phosphorus ratio exceeds 10 to 1. In West Branch Sub-basin #4, the estimated ratio is about 17 to 1.

Pennsylvania does not currently have water quality criteria for nutrients and sediments. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for nutrients and sediments in the impaired segments of the West Branch Sub-basin #4. Based upon comparison to a similar, non-impaired watershed, it was estimated that the sediment loading that will meet the water quality objectives for West Branch Sub-basin #4 is 1,790,284 lbs per year. The TMDLs for West Branch Sub-basin #4 are allocated as shown in the table below.

Summary of TMDLs for West Branch Subbasin #4 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediments 2,001,827 200,183 - 1,801,644 11,360 1,790,284

The TMDLs for sediments are allocated to non-point source from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load Allocation (WLA) is that portion of the total load that is assigned to point sources,which was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 8.5 miles. The TMDL establishes a reduction for total sediment loading of 28% from the current annual loading of 2,497,519 pounds.

138

I1.0 INTRODUCTION

I1.1 Watershed Description

The following discussion provides information on the physical characteristics of West Branch Sub- basin #4 including its location, land use distribution, and geology. West Branch Sub-basin #4 is located in the Piedmont physiographic province and in Bucks and Montgomery Counties. It covers an area of approximately 15 square miles. The streams in West Branch Sub-basin #4 drain into the West Branch of Neshaminy Creek from the north. The watershed is located east of the town of New Britain and is bounded by Pennsylvania Route 309 to the west, Route 152 to the east, and Route 202 to the south. Figure I1 shows the watershed boundary and its location. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the West Branch Sub-basin #4 is warm water fishes and migratory fishes.

The primary land use in West Branch Sub-basin #4 is agriculture (35%) (with areas adjacent to the stream used for cropland and pasture) and land development (29%). It was found also from a field survey of the watershed that cattle generally have free access to the stream. The majority of the streams had no protected riparian zone. The 1994 survey showed that nutrients from agricultural activities were causing increased algae growth. It also found that sediment deposited in large quantities on the streambed was degrading the habitat of bottom-dwelling macroinvertebrates.

Figure I1. West Branch Sub-basin #4 Watershed.

In terms of subsurface geology, West Branch Sub-basin #4 is primarily underlain by a shale formation. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional

139

permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Characteristics of Sub-basin #4, as well as those of the reference watershed chosen for further analysis (see later discussion), are summarized in Table I1.

Table I1. Physical Characteristic Comparisons between West Branch Sub-basin #4 and Reference Watershed

Attribute West Branch Sub-basin #4 Reference Watershed

Physiographic Province Piedmont Piedmont Area (square miles) 15 15 Predominant Land Uses -Agriculture (35%) -Agriculture (49%) Predominant Geology Shale (100%) Shale (100%) Soils - Dominant HSG C C - K Factor 0.37 0.37 20-Year Average Rainfall (in) 40.6 40.6 20-Year Average Runoff (in) 4.5 3.8

I1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for West Branch Sub-basin #4 to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water Act Section 303(d) Lists (see Table A1 in section A1.0). It was first determined that West Branch Sub-basin #4 was not meeting its designated water quality uses for protection of aquatic life in 1994 based on aquatic biological survey. The 2001 survey found that that the stream was still impaired. As a consequence, Pennsylvania listed stream segments in West Branch Sub-basin #4 on the 1996 and 2002 Section 303(d) List of Impaired Waters.

The 1996 303 (d) List reported 17.7 miles of West Branch Sub-basin #4 (Stream Segment ID#s 980205-1211-GLW, 980202-1040-GLW, 980202-1043-GLW) to be impaired by siltation and water flow variability from land development and agriculture, and excessive algae growth as a result of agricultural activities in the watershed. The 2002 303 (d) List reported additional 5.1 miles (Stream Segment ID# 980205-1430-GLW) to be impaired by siltation, excessive algae growth, and water flow variability from agriculture and land development.

140

Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of West Branch Sub-basin #4. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in West Branch Sub-basin #4. Without effective storm water management practices and sediment traps, build-up of sediments will continue in the stream.

I2.0 APPROACH TO TMDL DEVELOPMENT

I2.1 TMDL Endpoints

The TMDLs described herein address phosphorus and sediments. Phosphorus was determined to be the nutrient limiting plant growth in West Branch Sub-basin #4. Because neither Pennsylvania nor EPA has water quality criteria for phosphorus or sediments, we had to develop a method to determine water quality objectives for these parameters that would result in the impaired stream segments attaining their designated uses. The method employed for these TMDLs is termed the “reference watershed approach.”

I2.2 Selection of the Reference Watershed

141

West Branch Sub-basin #4) with respect to physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS coverages including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

A watershed that would be similar to West Branch Sub-basin #4 Watershed in land use distributions could not be found due to the fact that all watersheds that have similar levels of agricultural land use and land development are also impacted. Therefore, the watershed used as a reference for West Branch Sub-basin #4 has less development. The watershed used as a reference for the West Branch Sub-basin #4 is comprised of both Lashaka and Mill Creek watersheds. Both watershed are located in the same physiographic province and State Water Plan as West Branch Sub-basin #4. Table I1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure I2 shows the reference watershed boundary and its location in Bucks County.

An analysis of the MRLC land use/cover layer revealed that land use/cover distributions in both watersheds are very similar. The surficial geology of both West Branch Sub-basin #4 and reference watershed is shale. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. A look at these attributes in Table I1 indicates that these watersheds compare very well in terms of average runoff, precipitation, and soil K factor.

Figure I2. Reference Watershed Location.

142

I2.3 Water/Flow Variability and Flow Alterations Resulting from Urban Runoff/Storm Sewers and Land Development

I2.4 Siltation due to Agricultural Sources and Land Development

The 1994 and 2001 surveys showed that siltation originating from agricultural activities and land development in the watershed was the cause of impairment of West Branch Sub-basin #4 stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. The TMDLs address sediments from agricultural activities and land development/urban runoff and storm sewers and from stream bank erosion. Because neither Pennsylvania nor EPA have water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses.

I2.5 Excessive Algae Growth from Agriculture (Organic Enrichment in Stream Systems)

As indicated earlier, West Branch Sub-basin #4 was listed as being impaired due to problems associated with excessive algal growth. In stream systems, elevated nutrient loads (nitrogen and phosphorus in particular) can lead to increased productivity of plants and other organisms (Novotny and Olem, 1994). In most fresh water bodies, phosphorus is the limiting nutrient for aquatic growth. In some cases, however, the determination of which nutrient is the most limiting is difficult. For this reason, the ratio of the amount of N to the amount of P is often used to make this determination (Thomann and Mueller, 1987). If the N/P ratio is less than 10, nitrogen is limiting. If the N/P ratio is greater than 10, phosphorus is the limiting nutrient. In the case of West Branch Sub-basin #4, the N/P ratio is approximately 17, which points to phosphorus as the limiting nutrient. It is therefore assumed that controlling the phosphorus loading to West Branch Sub-basin #4 will limit plant growth and result in raising the dissolved oxygen level.

I2.6 Excess Algae Growth due to Municipal Point Sources

Nutrients from municipal sources were also listed as the cause of impairment for several stream segments of West Branch Sub-basin #4 and its tributaries. Although listed as a primary cause, it was determined on the basis of an evaluation of NPDES facilities in the watershed (see section B) that point source discharges were not likely the cause of observed nutrient (i.e., algal growth) impairments. Although it is highly probable that point sources continued to contribute to nutrient problems throughout the 1990s (during the time when the initial stream surveys upon which the 303d listings were based), it now appears that organic enrichment problems caused by point source discharges of nutrients (i.e., phosphorus) have been mitigated as evidenced by decreasing in-stream concentrations of phosphorus over the last decade in the larger Neshaminy Creek watershed, of which the West Branch Sub-basin #4 is a substantial part (see section B for more details). The lack of obvious positive improvements in aquatic health indicators over the last few years is very likely due to the lack of sufficient recovery time in the stream since

143

wastewater treatment plant upgrades have been implemented. This recovery has also likely been hindered by the below-normal levels of precipitation experienced in this particular region over the last few years.

Due to the fact that nutrient loads from point sources are now at or below NPDES-permitted loads (particularly in the case of phosphorus), it is believed that no further reductions in point source loads are needed at this time. Consequently, problems related to in-stream nutrients could be addressed by reducing actual yearly P loads from other sources in the watershed specifically nutrient loads associated with urban development (i.e., “transitional land use”) and streambank erosion. In this case, it is assumed that the stream would attain its beneficial use when actual P loads are reduced to or below the level of the sum of nutrient loads associated with transitional land use and stream bank erosion. Details on nutrient impairments originating from municipal point sources are addressed for the entire Neshaminy Creek watershed in Sections B2.0 and 3.0.

The objective of the TMDL process for West Branch Sub-basin #4 is to reduce the average loading rate of nutrient and sediments in the impaired watershed to the levels equivalent to or slightly lower than the average loading rate in the reference watershed. This load reduction will allow the biological community to return to the impaired stream segments. The TMDL endpoints established for this analysis are discussed in detail in the TMDL section. The listing for impairment caused by siltation and nutrients is addressed through reduction of sediment and P loads, respectively.

I2.7 Watershed Assessment and Modeling

The AVGWLF model was run for both West Branch Sub-basin #4 and the reference watershed to establish existing loading conditions under existing land cover use conditions in each watershed.

Adjustments to specific GWLF-related parameters in the reference watershed:

-reset “C” factor from 0.21 to 0.18 for Cropland to account for use of more continuous cover crop.

-reset “P” factor to 0.45 from 0.52 for Cropland land use to account for use of riparian forest and grasses along streams, strip cropping, and buffer strips.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the West Branch Sub-basin #4. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the impaired and reference watersheds are shown Tables I2 and I3, respectively. Table I4 presents an explanation of the header information contained in Tables I2 and I3. Modeling output for West Branch Sub-basin #4 and the reference watershed is presented in Appendix F.

144

Table I2. Loading Values for West Branch Sub-basin #4

Unit Area Land Use Area Total P Unit Area P Total N Unit Area N Sediment Sediment Category (acres) (lbs/yr) Load (lbs/yr) Load Load Load (lbs/acre/ yr) (lbs/acre/ yr) (lbs/year) (lbs/acre/yr) Hay/Pasture 669 143 0.21 1,216 1.82 19,492 29.14 Cropland 2,583 1,384 0.54 10,225 3.96 880,508 340.89 Coniferous For 94 0 0.01 10 0.11 110 1.17 Mixed Forest 684 3 0.00 77 0.11 1,258 1.84 Deciduous For 2,511 13 0.00 293 0.12 8,278 3.30 Unpaved Roads 2 1 0.5 15 7.5 1,670 835 Lo Intensity Dev 2,059 32 0.02 242 0.11 60,419 29.34 Hi Intensity Dev 610 25 0.04 222 0.36 13,422 22.00 Stream Bank 454 2,267 1,512,362 Groundwater 1,568 31,884 Point Source 8,104 149,136 Septic Systems 51 6,349 Total 9,211 11,781 1.27 201,938 21.92 2,497,519 271.14

Table I3. Loading Values for the Reference Watershed

Unit Area Land Use Area Total P Unit Area P Total N Unit Area N Sediment Sediment Category (acres) (lbs/yr) Load (lbs/yr) Load Load Load (lbs/acre/ yr) (lbs/acre/ yr) (lbs/year) (lbs/acre/yr) Hay/Pasture 1,279 270 0.21 2,299 1.80 28,874 22.58 Cropland 3,617 1,609 0.44 12,494 3.45 624,481 172.65 Coniferous For 143 0 0.00 16 0.11 243 1.70 Mixed Forest 800 4 0.00 92 0.12 1,921 2.40 Deciduous For 3,563 21 0.00 424 0.12 14,305 4.02 Transition 2 2 1.00 15 7.50 1,192 596.00 Quarries 30 11 0.37 51 1.70 10,799 359.97 Coal mines 32 20 0.63 92 2.87 30,375 949.21 Lo Intensity Dev 165 0 0 1 0.00 2,936 17.79 Hi Intensity Dev 86 0 0 3 0.03 1,258 14.63 Stream Bank 453 2,084 1,398,404 Groundwater 1,644 33,358 Point Source 0 7 Septic Systems 43 11,236 Total 9,717 4,076 0.42 62,391 6.42 2,111,788 217.33

145

Table I4. Header Information for Tables I2 and I3.

I3.0 LOAD ALLOCATION PROCEDURE FOR NUTRIENT AND SEDIMENT TMDLs

The load allocation and reduction procedures were applied to all of West Branch Sub-basin #4. The watershed was so small that we did not subdivide it into sub-watersheds. Therefore, sub-watershed load allocations were not performed.

The load reduction calculations in West Branch Sub-basin #4 are based on the current loading rates for phosphorus and sediments in the reference watershed. Based on biological assessment, it was determined that the reference watershed was attaining its designated uses. The phosphorus and sediment loading rates were computed for the reference watershed using the AVGWLF model. These loading rates were then used as the basis for establishing the TMDLs for West Branch Sub-basin #4.

The equations defining TMDLs for sediments and nutrients are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that need to be reduced by the non-contributing sources (NLR) for West Branch Sub-basin #4

146

to meet water quality goals. Therefore, it is the load that originates from agricultural sources that have contributed to water quality problems encountered in the watershed. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

I3.1 TMDL Total Load

The TMDL loads for both pollutants of concern were computed in the same manner. The first step is to determine the TMDL total target load for West Branch Sub-basin #4, the impaired watershed. This value was obtained by multiplying each pollutant unit loading rate in the reference watershed by the total watershed area of West Branch Sub-basin #4. This information is presented in Table I5.

Table I5. TMDL Total Load Computation

Unit Area Loading Rate Total Watershed Area in in Reference Watershed West Branch Sub-basin #4 TMDL Total Type of Pollutant (lbs/acre/yr) (acres) Load (lbs/yr) Phosphorus 0.42 9,211 3,869 Sediment 217.33 9,211 2,001,827

I3.2. Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of each of the TMDLs was reserved as the MOS.

Phosphorus - 3,869lbs/yr x 0.1 = 387 lbs/yr (3) Sediment - 2,001,827 lbs/yr x 0.1 = 200,183 lbs/yr (4)

I3.3 Load Allocation

The Load allocation (LA), consisting of all nonpoint sources in the watershed, was computed by subtracting the margin of safety and the waste load allocation (WLA) from the TMDL total load. (Notice that sediments do not have waste load allocation).

LA (Phosphorus) = 3,869 lbs/yr – 387 lbs/yr- 8,104 lbs/yr= - lbs/yr (5) LA (Sediments) = 2,001,827 lbs/yr – 200,183 lbs/yr = 1,801,644 lbs/yr (6)

Notice that the LA for phosphorus was not computed. It is a negative value because most of the P loading in this watershed is from municipal point sources (8,104 lbs/yr or about 70% of the total load). Nutrient from municipal point sources are addressed in a different section of this document (Section B). Therefore, further analysis of P and load allocation will not be performed.

147

I3.4. Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will need reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table I2 except those from agricultural land uses (Hay/Pasture, Row Crops), land development, and stream bank erosion. Therefore, using data in Table I2,

LNR (Sediments) = 110 lbs/yr + 1,258 lbs/yr 8,278 lb/yr + 1,670 lb/yr = 11,360 lbs/yr (7)

ALA (Sediments) = 1,801,644 lbs/yr – 11,360 lbs/yr = 1,790,284 lbs/yr. (8)

Table I6 below presents the TMDL for West Branch Sub-basin #4.

Table I6. Summary of Load Allocation for West Branch Sub-basin #4 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediments 2,001,827 200,183 - 1,801,644 11,360 1,790,284

The ALA computed above is the portion of the load that is available to allocate among contributing sources (Hay/Pasture, Cropland) and land development (including streambank erosion) as described in the next step. Not all land use/source categories were included in the allocation because they are difficult to control, or provide an insignificant portion of the total load (e.g., transition land use). The following section shows the allocation process in detail for the entire watershed.

I3.5 Load Reduction Procedures

3) Each land use/source load is compared with the total allocable load to determine if any contributor would exceed the allocable load by itself. The evaluation is carried out as if each source is the only contributor to the pollutant load to the receiving waterbody. If the contributor exceeds the allocable load, that contributor would be reduced to the allocable load. This is the baseline portion of EMPR.

4) After any necessary reductions have been made in the baseline the multiple analysis is run. The multiple analysis will sum all of the baseline loads and compare them to the total allocable load. If the allocable load is exceeded, an

148

It is important to note that:

3) Load allocation to sub-watersheds was not performed due to the fact that West Branch Sub-basin #4 is a very small and therefore could not be subdivided into meaningful subwatersheds.

4) Land development and stream bank erosion were lumped together as one sediment pollutant source, “land development” because sediments in urban and suburban areas are from disturbed land and unprotected soils at construction sites, and from stream channel erosion. Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of West Branch Sub-basin #4.

The load allocation and EMPR procedures were performed using an Excel Worksheet and results are presented in Appendix G. Results of the load allocation by land use sources are presented in Table I7. Table I8 provides load allocation by considering all land uses in West Branch Sub-basin #4. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table I7. Load Allocation by Each Contributing Source in West Branch Sub-basin #4 .

Sediments Land Use/Source Loading Rate Average Load Average ALA Reduction Lbs/ac/yr lbs/yr Lbs/yr - % - Hay/Pasture 29.14 19,492 14,036 28 Cropland 340.89 880,508 634,043 28 1 Land Development 594.31 1,586,203 1,142,205 28 Sub-total 419.90 2,483,203 1,790,284 28

1Loads from land development include the amount from stream bank erosion.

149

Table I8. Load Allocation by Each Land Use/Source

Sediment

Unit Area Annual average ALA (annual Area Loading Rate load average) Reduction Source (acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) (%)

Hay/Pasture 669 29.14 19,492 14,036 28 Cropland 2,583 340.89 880,508 634,043 28 Coniferous For 94 1.17 110 110 0 Mixed Forest 684 1.84 1,258 1,258 0 Deciduous For 2,511 3.30 8,278 8,278 0 Transition 2 835 1,670 1,670 0 Lo Intensity Dev 2,059 29.34 60,419 43,508 28 Hi Intensity Dev 610 22.00 13,422 9,665 28 Stream Bank 1,512,362 1,089,032 28 Groundwater Point Source Septic Systems 9,211 271.14 2,497,519 1,801,600 28

The total allowable sediment load in West Branch Sub-basin #4 when all land use/cover sources are considered is 1,801,600 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from 2,497,519 pounds per year by 28%.

I4.0 CONSIDERATION OF CRITICAL CONDITIONS

I5.0 CONSIDERATION OF SEASONAL VARIATIONS

150

each month. The model also considers the months of the year when manure is applied to the land. The combination of these actions by the model accounts for seasonal variability.

I6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDLs are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDLs. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within West Branch Sub-basin #4 will be affected by this policy, which has been included in Appendix E.

I7.0 PUBLIC PARTICIPATION

151

J. Total Maximum Daily Loads (TMDLs) Development Plan for Cooks Run Watershed

152

Table of Contents Page

Executive Summary ……………………………………………………………...… 155

J1.0 Introduction ……………………………………………………………….……. 156 J1.1 Watershed Description …………………………………………….…… 156 J1.2 Surface Water Quality ……………………………………………….… 157 J2.0 Approach to TMDL Development………………………………………………. 157 J2.1. TMDL Endpoints ………………………………………………………………. 157 J2.1.1 Unknown Causes from Urban Runoff/Storm Sewers……………….. 158 J2.1.2. Nutrients from Urban Runoff/Storm Sewers………………………….. 158 J2.1.3 Nutrients From Municipal Point Sources…………………………….. 158 J2.2 Selection of the Reference Watershed………………………………………….. 159 J2.3 Watershed Assessment and Modeling………………………………………….. 160 J3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 162 J3.1 Sediment TMDL Total Load ………………………………………..….. 163 J3.2 Margin of Safety ………………………………………...…………… 163 J3.3 Load Allocation ……………………………………………………….. 163 J4.0 Consideration of Critical Conditions ………………………………………….. 163 J5.0 Consideration of Seasonal Variations …………………………………..…….. 164 J6.0 Reasonable Assurance of Implementation ……………………………………. 164 J7.0 Public Participation …………………………………………………………. 164

153

List of Tables Page

J1. Physical Characteristic Comparisons between Cooks Run and Reference Watersheds …………………………………………… 157 J2. Loading Values for Cooks Run Watershed…… 161 J3. Loading Values for the Reference Watershed…………….…………. 161 J4. Header Information for Tables J2 and J3………………………………..…. 162 J5. TMDL Load Computation……………… …………....………. 163

List of Figures Page

J1. Cooks Run Watershed ………………………………….… 156 J2 Reference Watershed …………………………………………………… 160

154

EXECUTIVE SUMMARY

The Cooks Run watershed is approximately 8 square miles in size. It is in Bucks county and drains into the main stem of Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 4.6 miles of Cooks Run (Stream Segment ID#s 482 and 482A). They were developed to address the impairments noted on Pennsylvania’s 1996 Clean Water act Section 303(d) List. The impairments are primarily caused by nutrient loads principally from urban runoff/storm sewers in the watershed. Phosphorus is generally considered to be the limiting nutrient in a waterbody when the nitrogen/phosphorus ratio exceeds 10 to 1. In Cooks, this ratio was calculated to be about 11 to 1.

Pennsylvania does not currently have water quality criteria for nutrients. For this reason, a reference watershed approach was used to identify the TMDL endpoints or water quality objectives for nutrients in the impaired segments of the Cooks Run watershed.

After conducting watershed modeling and an analysis of loads from all sources, it was found that phosphorus loads within the Cooks Run watershed were primarily being contributed by a single municipal point source. Based on an evaluation of discharge data, it does not appear that further reductions in phosphorus from this point source is required since it is operating within the limitations of it’s current NPDES permit. However, it would be prudent to assess this stream at a later date to verify if observed nutrient reductions have resulted in improved stream health. In this case, the recommended phosphorus TMDL for this point source would be the load presently specified by its existing NPDES permit. It is further recommended that the phosphorus TMDL for this watershed be set at its current loading rate of 2,164 lbs per year until such time when a future evaluation determines that a reduction in the point source discharge beyond its current NPDES phosphorus limit is needed in order to achieve a healthy stream environment.

155

J1.0 INTRODUCTION

J1.1 Watershed Description

The following discussion provides information on the physical characteristics of Cooks Run and its watershed including location, land use distributions, and geology. The Cooks Run watershed is located in the Piedmont physiographic province and is entirely within Bucks County. It covers an area of approximately 8 square miles. Cooks Run drains into the main stem of Neshaminy Creek from the east. The watershed is located south of the town of New Britain and west of Doylestown, and is bounded by Pennsylvania Route 152 to the west and Route 611 to the east. Figure J1 shows the watershed boundary and its location. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the Cooks Run is warm water fishes and migratory fishes.

The primary land use in the Cooks Run watershed is urban development (47%) followed by agriculture (24%), with areas adjacent to the stream used primarily for cropland and pasture. The majority of the streams had no protected riparian zone. The 1994 survey showed that nutrients from agricultural activities were causing increased algae growths.

Figure J1. Cooks Run Watershed.

156

The surficial geology of the Cooks Run watershed consists of sandstone. The geology of this area primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.38). Watershed characteristics are summarized in Table J1.

Table J1. Physical Characteristic Comparisons between Cooks and Reference Watershed

Attribute Cooks Run Watershed Reference Watershed Physiographic Province Piedmont Piedmont Area (square miles) 8 15 Predominant Land Uses - Development (47%) - Development (33%) - Agriculture (24%) - Agriculture (49%)

Predominant Geology Shale (100%) Shale (100%) Soils - Dominant HSG C C - K Factor 0.38 0.38 20-Year Average Rainfall (in) 40.4 41.5 20-Year Average Runoff (in) 4.2 5.4

J1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for the Cooks Run watershed to address the impairments noted on the Pennsylvania’s 1996 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). It was determined that Cooks Run was not meeting its designated water quality uses for protection of aquatic life in 1994 based on aquatic biological survey.

The 1996 303 (d) List reported 4.6 miles of the Cooks (Stream Segment ID#s 482 and 482A) to be impaired by nutrients and “cause unknown” from urban runoff/storm sewers. In addition, Stream Segment ID# 482A was found to be impaired by nutrients from municipal point sources.

J2.0 APPROACH TO TMDL DEVELOPMENT

J2.1. TMDL Endpoints

The TMDLs in this case were developed to address phosphorus. Phosphorus was determined to be the nutrient limiting plant growth in Cooks. Because neither Pennsylvania nor EPA has water quality criteria for phosphorus, we had to develop a method to determine water quality objectives for these parameters that would result in the impaired stream segments attaining their designated uses. The method employed for these TMDLs is termed the “reference watershed approach.”

157

In using the reference watershed approach, two watersheds are compared; with one attaining its uses and one that is impaired based on biological assessment. Both watersheds must have similar land use/cover distributions. Other features such as base geologic formation should be matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of this approach is to reduce the loading rate of nutrients in the impaired stream segment to a level equivalent to or slightly lower than the loading rate in the non-impaired, reference stream segment. The underlying assumption is that this load reduction will allow the biological community to return to the impaired stream segments. The TMDLs to be addressed in this document are for those impairments noted on the 1996 303(d) List, and are as follows:

J2.1.1 Unknown Causes From Urban Runoff/Storm Sewers

TMDLs were not developed for unknown causes. It was assumed that addressing nutrient loads through the use of urban BMPs will at the same time reduce the level of these unknown causes within the watershed.

J2.1.2 Nutrient from Urban Runoff/Storm Sewers

As indicated earlier, Cooks Run was listed as being impaired due to problems associated with nutrient loads. In stream systems, elevated nutrient loads (nitrogen and phosphorus in particular) can lead to increased productivity of plants and other organisms (Novotny and Olem, 1994). In most fresh water bodies, phosphorus is the limiting nutrient for aquatic growth. In some cases, however, the determination of which nutrient is the most limiting is difficult. For this reason, the ratio of the amount of N to the amount of P is often used to make this determination (Thomann and Mueller, 1987). If the N/P ratio is less than 10, nitrogen is limiting. If the N/P ratio is greater than 10, phosphorus is the limiting nutrient. In the case of Cooks Run, the N/P ratio is approximately 11, which points to phosphorus as the limiting nutrient. It is therefore assumed that controlling the phosphorus loading to Cooks Run will limit plant growth and result in raising the dissolved oxygen concentration.

J2.1.3 Nutrients Due to Municipal Point Sources

As reported earlier, nutrients from municipal sources were also listed as the cause of impairment for several stream segments of Cooks Run and its tributaries. Although listed as a primary cause, it was determined on the basis of an evaluation of NPDES facilities in the watershed (see section B) that point source discharges were not likely the cause of observed nutrient impairments. Although it is highly probable that point sources continued to contribute to nutrient problems throughout the 1990s (during the time when the initial stream surveys upon which the 303d listings were based), it now appears that organic enrichment problems caused by point source discharges of nutrients (i.e., phosphorus) have been mitigated as evidenced by decreasing in-stream concentrations of phosphorus over the last decade in the larger Neshaminy Creek watershed, of which the Cooks watershed is a part (see section B for more details). The lack of obvious positive improvements in aquatic health indicators over the last few years is very likely due to the lack of sufficient recovery time in the stream since wastewater treatment plant upgrades have been implemented. This recovery has also likely been hindered by the below- normal levels of precipitation experienced in this particular region over the last few years.

158

The objective of the TMDL process for Cooks Run is to reduce the average loading rate of nutrients in the impaired stream segments to levels equivalent to or slightly lower than the average loading rate in the reference watershed. This load reduction will allow the biological community to return to the impaired stream segments. The TMDL endpoints established for this analysis are discussed in detail in the TMDL section. The listing for impairment caused by nutrients is addressed through reduction of phosphorus loads.

J2.2 Selection of the Reference Watershed

In general, three factors should be considered when selecting a suitable reference watershed. The first factor is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to attain water quality standards. The second factor is to find a watershed that closely resembles the Cooks Run watershed in terms of physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS coverages including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The watershed used as a reference for the Cooks Run watershed is a sub-watershed of Ironworks Creek, which is located in the lower part of the Neshaminy Creek watershed. Both watersheds are located in the same physiographic province and State Water Plan watershed. They have same geology and k factors, and have similar land use distributions. Table J1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure J2 shows the reference watershed boundary and its location in Bucks County.

An analysis of the land use/cover layer (MRLC data set) revealed that land cover/use distributions in both watersheds are similar. The surficial geology of both the Cooks Run and reference watersheds is shale. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. A look at these attributes in Table J1 indicates that these watersheds compare very well in terms of average runoff, precipitation, and soil K factor.

159

Figure J2. Reference Watershed Location

J2.3 Watershed Assessment and Modeling

The AVGWLF model was run for both the Cooks Run and reference watersheds to establish existing loading conditions under existing land cover use conditions in each watershed. Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the Cooks Run watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the impaired and reference watersheds were calculated as shown in Tables J2 and J3, respectively. Table J4 presents an explanation of the header information contained in Tables J2 and J3. Modeling output for the Cooks Run (impaired watershed) and reference watersheds is presented in Appendix F.

160

Table J2. Existing Loading Values for Cooks Run Watershed

Land Use Area Total P Unit Area P Load Total N Unit Area N Load Category (acres) (lbs/yr) (lbs/acre/ yr) (lbs/yr) (lbs/acre/yr) Hay/Past 119 25 0.21 214 1.79 Cropland 373 172 0.46 1,366 3.66 Coniferous For 17 0 0 2 0.12 Mixed For 114 0 0 11 0.10 Deciduous For 422 2 0 49 0.12 Transition 47 42 0.89 377 8.02 Lo Int Dev 696 17 0.02 40 0.06 Hi Int Dev 274 11 0.04 102 0.37 Stream Bank 25 143 Groundwater 325 6,992 Point Source 1,541 11,733 Septic Systems 4 2,475 Total 2,062 2,164 1.05 23,504 11.40

Table J3. Loading Values for the Reference Watershed

Land Use Area Total P Unit Area P Load Total N Unit Area N Load Category (acres) (lbs/yr) (lbs/acre/ yr) (lbs/yr) (lbs/acre/ yr) Hay/Past 622 142 0.23 1,092 1.76 Cropland 528 278 2.42 1,938 3.67 Coniferous For 25 0 0 3 0.12 Mixed For 210 1 0 23 0.11 Deciduous For 160 1 0 17 0.11 Transition 17 17 1.00 132 7.76 Lo Int Dev 649 7 0.01 56 0.09 Hi Int Dev 106 1 0.01 15 0.14 Stream Bank 33 150 Groundwater 318 6,972 Point Source 0 0 Septic Systems 0 512 Total 2,317 798 0.34 10,935 4.72

161

Table J4. Header Information for Tables J2 and J3.

J3.0 LOAD ALLOCATION PROCEDURE FOR NUTRIENT TMDLs

The load allocation and reduction procedures were applied to the entire Cooks Run watershed. The watershed was so small that we did not subdivided it into sub-watersheds. Therefore, sub-watershed load allocations were not performed.

The load reduction calculations in the Cooks Run watershed are based on the current loading rates for phosphorus in the reference watershed. Based on biological assessment, it was determined that the reference watershed was attaining its designated uses. The phosphorus loading rate was computed for the reference watershed using the AVGWLF model. The loading rate was then used as the basis for establishing the TMDL for Cooks Run.

The equations defining TMDL for nutrients are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Cooks Run to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

162

J3.1 TMDL Total Load for Phosphorus

A TMDL for phosphorus was determined for the Cooks Run watershed. As shown in Table J5, this value was obtained by multiplying the pollutant unit loading rate in the reference watershed (0.34) by the total watershed area of Cooks Run.

Table J5. TMDL Total Load Computation

Unit Area Loading Rate in Reference Watershed Total Watershed Area in TMDL Total Load Type of Pollutant (lbs/acre/yr) Cooks Run (acres) (lbs/yr) Phosphorus 0.34 2,062 701

J3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

Phosphorus - 701 lbs/yr x 0.1 = 70 lbs/yr (3)

J3.3 Load Allocation

The Load allocation (LA), consisting of all nonpoint sources in the watershed, was computed by subtracting the margin of safety and the waste load allocation (WLA) from the TMDL total load.

LA (Phosphorus) = 701 lbs/yr – 70 lbs/yr- 1,541 lbs/yr= - lbs/yr (5)

Notice that the LA for phosphorus was not computed. It is a negative value because most of the P loading in this watershed is from municipal point sources (1,541 lbs/yr or about 71% of the total load). Nutrient from municipal point sources are addressed in a different section of this document (Section B). Therefore, further analysis of P nor load allocation will not be performed.

J4.0 CONSIDERATION OF CRITICAL CONDITIONS

163

J5.0 CONSIDERATION OF SEASONAL VARIATIONS

J6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Based on the analysis described above, a load allocation could not be done since the analysis showed that the greatest phosphorus load within the watershed (about 71%) is being contributed by municipal point source discharges. As described in Section B, the single point source located along this stream (NPDES No. PA0021172) discharges approximately 930 kg/year (or about 2051 lbs/yr) of P. (Note that when accounting for in-stream attenuation, the AVGWLF model reduces this load to about 1541 lbs/yr as shown in Table J2). As discussed in Section B3.1, this facility consistently discharges loads below it’s permitted limit, and did not exceed established phosphorus limits at all during the period 1998-2000. However, during the “1988” time period reflected in Table B2, this facility appeared to be discharging about 5 times more phosphorus than it did during the “1999” time period. Since this facility was upgraded in the mid-1990s for phosphorus removal, it is quite likely that nutrient impairments were still evident during the time that the stream was initially surveyed and listed (ca. 1996). It is very probable that visible indications of improved stream health related to such reductions would have taken several years to manifest (and may still not be evident at present due to drought conditions that have existed in the area for several years). Consequently, it does not appear that further reductions in nutrients from this point source are required for this segment since this particular point source is operating within the limitations of it’s current NPDES permit. However, it would be prudent to assess this stream at a later date to verify if observed nutrient reductions have resulted in improved stream health. In this case, the recommended phosphorus TMDL for this point source would be the load presently specified by its existing NPDES permit. It is further recommended that the phosphorus TMDL for this watershed be set at its current loading rate of 2,164 lbs per year until such time when a future evaluation determines that a reduction in the point source discharge beyond its current NPDES phosphorus limit is needed in order to achieve a healthy stream environment.

J7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

164

K. Total Maximum Daily Loads (TMDLs) Development Plan for Neshaminy Creek Tributary #1 Watershed

165

Table of Contents Page

Executive Summary …………………………………………………………………… 168

K1.0 Introduction ……………………………………………………………….……. 169 K1.1 Watershed Description …………………………………………….…. 169 K1.2 Surface Water Quality ………………………………………………... 170 K2.0 Approach to TMDL Development………………………………………...…….. 171 K2.1 Siltation Caused by Land Development ………….……………….……. 171 K2.2 Watershed Assessment and Modeling…………………………………… 172 K3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 174 K3.1 Sediment TMDL Total Load ………………………………………..….. 174 K3.2 Margin of Safety ……………………………………………………….. 175 K3.3. Load Allocation ………………………………………………………… 175 K3.4. Adjusted Load Allocation ……………………………………………… 175 K3.5. Load Reduction Procedures ………………………………..…………. 176 K4.0 Consideration of Critical Conditions ………………………………………… 177 K5.0 Consideration of Seasonal Variations …………………………………..…….. 177 K6.0 Reasonable Assurance of Implementation …………………………………… 177 K7.0 Public Participation …………………………………………………………. 178

166

List of Tables Page

K1. Physical Characteristics of Neshaminy Creek Tributary #1……………………. 170 K2. Loading Values for Neshaminy Creek Tributary #1Watershed, Year 1992 Land Use Conditions …………………………..………………. 173 K3. Loading Values for Neshaminy Creek Tributary #1 Watershed, Year 2000 Land Use Conditions …………………………..………………. 173 K4. Header Information for Tables K2 and K3………………………………..…. 174 K5. Summary of TMDLs for Neshaminy Creek Tributary #1 Watershed ………….. 175 K6. Load Allocation for each contributing source to Neshaminy Creek Tributary #1… 176 K7. Sediment Load Allocation by Land Use/Source ……………………….…... 176

List of Figures Page

K1. Neshaminy Creek Tributary #1 Watershed ………………………………… 169

167

EXECUTIVE SUMMARY

The Neshaminy Creek Tributary #1 watershed in Bucks County is about 5.2 square miles in size. This watershed contains a stream segment that is a tributary of Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 4.6 miles of stream (Stream Segment Id # 980427-0945-GLW). They were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water act Section 303(d) List. The impairments are primarily caused by sediment loads from land development in the watershed. The TMDL focuses on control of sediments. Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of this watershed. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining and Year 2000 when it was found to be impaired. Siltation, the cause of impairment in Neshaminy Creek Tributary #1, resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the amounts of sediment loading that will meet the water quality objectives for this tributary were 85,194 pounds per year. It is assumed that this tributary will support its aquatic life uses when this value is met. The sediment TMDLs for Neshaminy Creek Tributary #1 are allocated as shown in the table below.

Summary of TMDLs for Neshaminy Creek Tributary #1 Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 232,83 23,283 - 209,54 124,34 85,194 stream bank erosion 0 7 9

The TMDLs for sediments are allocated to non-point source loads from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The Wasteload Allocation (WLA) is that portion of the total load that is assigned to point sources, but was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDLs cover a total of 4.6 miles. The TMDL establishes a reduction for total sediment loading of 71% from the current annual loading of 721,215 pounds.

168

K1.0 INTRODUCTION

K1.1 Watershed Description

The following discussion provides information on the physical characteristics of Neshaminy Creek Tributary #1 and its watershed including location, land use distributions, and geology. This particular watershed is located in the Piedmont Physiographic Province, and is entirely situated in Bucks County. It covers an area of approximately 5.2 square miles. This tributary drains into the main stem of Neshaminy Creek from the West. The watershed is located south of the town of St Leonard and north of Holland. It can be reached via Pennsylvania Route 522 from the east. Figure K1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for Neshaminy Creek Tributary #1 is warm water fishes and migratory fishes.

The current land use distribution in the watershed was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land uses in the Neshaminy Creek Tributary #1 watershed are developed land (43%) followed by agriculture (36%). It is important to note that development in the watershed changed from 346 to 567 acres, or a 64% increase, from 1992 to 2000.

Figure K1. Neshaminy Creek Tributary #1 Watershed.

169

The surficial geology of Neshaminy Trib#1 watershed consists of sandstone. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Watershed characteristics are summarized in Table K1.

K1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for stream segments in this watershed to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). It was first determined that Neshaminy Creek Tributary #1 was not meeting its designated water quality uses for protection of aquatic life in 2001 based on aquatic biological survey. As a consequence, Pennsylvania listed this tributary on the 2002 Section 303(d) List of Impaired Waters.

Table K1. Physical Characteristics of Neshaminy Creek Tributary #1 Watershed

Physiographic Province Piedmont Area (square miles) 5.3 Predominant Land Use - Developed land (43%) - Agriculture (36%) Predominant Geology Sandstone (100%) Soils Dominant HSGs C Average K Factor 0.37 20-Year Average Rainfall 41.5 (in) 20-Year Average Runoff (in) 4.6

The 2002 303 (d) List reported 4.6 miles of stream (Stream Segment Ids # 20010426-1512- GLW and 980609-1425-GLW) to be impaired by siltation from land development and flow alterations as a result of urban runoff/storm sewers. These stream segments are impacted by siltation as a result of “New Land Development” in the watershed. New Land Development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in this watershed is the result of years of build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events. As indicated above, land development has increased by approximately 64% between 1992 and 2000.

Sediments, which are often the cause of stream impairment in urban and suburban areas, are primarily from two sources: disturbed land and unprotected soils at construction sites, and stream

170

channel erosion. Transitional land uses, mainly new construction sites, are one of the main sources of sediments in streams draining newly developed areas. Sediment production and sedimentation in streams are typically important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1994). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the Neshaminy Creek Tributary #1 watershed. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in this watershed. Without effective storm water management practices and sediment traps, build-up of sediments will continue in the stream.

K2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDLs address impairment by sediments in Neshaminy Creek Tributary #1 stream segments as reported on the 2002 303(d) Lists. The stream water flow variability impairment caused by urban runoff/storm sewer will not be explicitly addressed by these TMDLs because it is assumed that management practices used to address storm water runoff and sediment production at new construction sites will reduce problems associated with flow variability as well. These TMDLs were derived as follows:

K2.1 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediment produced by newly developed land in the watershed were the cause of impairment of stream segments in this watershed. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. The TMDLs for the Neshaminy Creek Tributary #1 watershed address sediments from construction sites or “transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA have water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

171

considered here as the benchmark because (as indicated earlier) the analysis of classified satellite images showed that development in the watershed increased by about 64% between 1992 and 2000.

K2.2 Watershed Assessment and Modeling

The AVGWLF model was run for this watershed to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the Neshaminy Creek Tributary #1 watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the 1992 and 2000 land use/cover conditions are shown in Tables K2 and K3, respectively. The Unit Area Load for sediment was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area for the watershed. Table G4 presents an explanation of the header information contained in Tables K2 and K3. Modeling output for this watershed for 1992 and 2000 land use conditions are presented in Appendix F.

172

Table K2. Loading Values for Neshaminy Creek Tributary #1 Watershed, Year 1992 Land Use Conditions

Area Sediment Load Unit Area Sediment Load Land Use Category (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 318 19,028 59.84 Cropland 353 128,262 363.35 Coniferous Forest 5 0 0 Mixed Forest 94 221 2.35 Deciduous Forest 215 750 3.49 Transitional 0 0 0 Low Intensity Developed 304 17,571 57.80 High Intensity Developed 42 1,214 28.90 Stream Bank 65,784 Groundwater Point Source Septic Systems Total 1,331 232,830 174.93

Table K3. Loading Values for Neshaminy Creek Tributary #1 Watershed, Year 2000 Land Use Conditions

Area Sediment Load Unit Area Sediment Load Land Use Category (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 199 10,331 51.91 Cropland 286 96,424 337.15 Coniferous Forest 5 0 0 Mixed Forest 86 199 2.31 Deciduous Forest 188 662 3.52 Transition 261 524,901 2011.11 Low Intensity Developed 264 15,497 58.70 High Intensity Developed 42 1,236 29.43 Stream Bank 71,965 Groundwater Point Source Septic Systems Total 1,331 721,215 541.86

173

Table K4. Header Information for Tables K2 and K3.

K3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire Neshaminy Creek Tributary #1 watershed. Sub-watersheds were not delineated due to the small size of the watershed (5.2 square miles). The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDLs for this watershed.

The equations defining TMDLs for sediments are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) in this watershed to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

K3.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for the Neshaminy Creek Tributary #1 watershed are based on the sediment loads obtained using the 1992 land use conditions, and are equal to 232,830 lbs/year (see Table K2).

174

K3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDLs were reserved as the MOS.

MOS (Sediments) 232,830 lbs/yr x 0.1 = 23,283 lbs/yr (1)

K3.3 Load Allocation

The Load allocation (LA), consisting of all sources in the watershed, was computed by subtracting the margin of safety. Waste load allocation (WLA), which is usually subtracted from the TMDL total load, was not used because there is not a waste load for sediments. . LA (Sediments) 232,830 lbs/yr – 23,283 lbs/yr = 209,547 lbs/yr (2)

K3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table K3 except those from the transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table K3,

LNR (Sediments) =10,331 lbs/yr + 96,424 lbs/yr + 0 lbs/yr + 199 lb/yr + 662 lb/yr + 15,497 lbs/yr + 1,236 lbs/yr = 124,349 lbs/yr (3)

ALA (Sediments) = 209,547 lbs/yr – 124,349 lbs/yr= 85,194 lbs/yr (4)

Table K5 below presents the TMDL for the Neshaminy Creek Tributary #1 watershed.

Table K5. Summary of TMDL for Neshaminy Creek Tributary #1 Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 232,83 23,283 - 209,54 124,34 85,194 stream bank erosion 0 7 9

175

K3.5 Load Reduction Procedures

The allocation of the sediment load among contributing land use/cover sources in this watershed was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table K6. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 74,922 and 10,272 pounds, respectively for stream segments in this watershed to attain their specific uses.

Table K6. Load Allocation for Each Contributing Source in the Neshaminy Creek Tributary #1 Watershed. Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 524,901 88 74,922 86 - Stream bank erosion 71,965 12 10,272 86 TOTAL 596,866 100 85,194 86

Table K7 provides sediment load allocation when all land uses in the watershed are taken into consideration. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table K7. Sediment Load Allocation by Each Land Use/Source Land Use Category Area Unit Area Load Load ALA Reduction (acres) (lbs/acre/yr) (lbs/year) (lbs/year) (%) Hay/Pasture 199 51.91 10,331 10,331 0 Cropland 286 337.1596,424 96,424 0 Conifer Forest 5 0 0 0 0 Mixed Forest 86 2.31 199 199 0 Decid Forest 188 3.52 662 662 0 Transition 261 2011.11524,901 74,922 86 Low Intensity Dev 264 58.70 15,497 15,497 0 High Intensity Dev. 42 29.43 1,236 1,236 0 Stream Bank 71,965 10,272 86 Groundwater Point Source Septic Systems Total 1,331 51.91 721,215 209,543 71

176

The total allowable sediment load in this watershed when all land use/cover sources are considered is 209,543 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from 721,215 pounds per year by a factor of 71%.

K4.0 CONSIDERATION OF CRITICAL CONDITIONS

K5.0 CONSIDERATION OF SEASONAL VARIATIONS

K6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Proposed sediment reductions are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDLs. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within the Neshaminy Creek Tributary #1 watershed will be affected by this policy, which has been included in Appendix E.

177

determine the most cost-effective and environmentally protective combination of BMPs required to meet the sediment reductions outlined in this report.

K7.0 PUBLIC PARTICIPATION

178

L. Total Maximum Daily Loads (TMDLs) Development Plan for Neshaminy Creek Tributary #2 Watershed

179

Table of Contents Page

Executive Summary ………………………………………………………………... 182

L1.0 Introduction ……………………………………………………………….……. 183 L1.1 Watershed Description …………………………………………….….. 183 L1.2 Surface Water Quality ……………………………………………….... 184 L2.0 Approach to TMDL Development………………………………………...…….. 185 L2.1 Water/Flow Variability Due to Urban Runoff/Storm Sewers …………... 185 L2.2 Siltation Due to Urban Runoff/Storm Sewers ………….……………….… 185 L2.3 Watershed Assessment and Modeling…………………………………….. 186 L3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 188 L3.1 Sediment TMDL Total Load ………………………………………..….. 188 L3.2 Margin of Safety ………………………………………………………… 189 L3.3. Load Allocation …………………………………………..……………… 189 L3.4. Adjusted Load Allocation ……………………………………………….. 189 L3.5. Load Reduction Procedures ………………………………..…………… 189 L4.0 Consideration of Critical Conditions ……………………………………….… 191 L5.0 Consideration of Seasonal Variations …………………………………..…….. 191 L6.0 Reasonable Assurance of Implementation …………………………………… 191 L7.0 Public Participation …………………………………………………………. 192

180

List of Tables Page

L1. Physical Characteristics of Neshaminy Creek Tributary#2……………………… 184 L2. Loading Values for Neshaminy Creek Tributary #2Watershed, Year 1992 Land Use Conditions …………………………..………………. 187 L3. Loading Values for Neshaminy Creek Tributary #2 Watershed, Year 2000 Land Use Conditions …………………………..………………. 187 L4. Header Information for Tables L2 and L3………………………………..…. 188 L5. Summary of TMDLs for Neshaminy Creek Tributary #2 Watershed ………….. 189 L6. Load Allocation for each contributing source to Neshaminy Creek Tributary #2 … 190 L7. Sediment Load Allocation by Land Use/Source ………………………... 190

List of Figures Page

L1. Neshaminy Creek Tributary #2 Watershed ………………………………… 183

181

EXECUTIVE SUMMARY

The Neshaminy Creek Tributary #2 watershed in Bucks County is about 1.8 square miles in size. This watershed contains a stream that is a tributary of the main stem of Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 1.5 stream miles in the watershed (Stream Segment Id # 980514-1004-GLW). They were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water act Section 303(d) List. The impairments are primarily caused by sediment loads and water/flow variability due to land development in the watershed. The TMDL focuses on control of sediments. Water/flow variability was not explicitly addressed because it was believed that the implementation of BMPs in the urban land use areas (High and Low Intensity Developed) to reduce sediment would also decrease water flow and volume to the stream and therefore stabilize stream flow.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of the this watershed. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining and Year 2000 when it was found to be impaired. Siltation, the cause of impairment to this tributary, resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the amount of sediment loading that will meet the water quality objectives for the Neshaminy Creek Tributary #2 watershed were 26,808 pounds per year. It is assumed that affected stream will support its aquatic life uses when this value is met. The sediment TMDLs for this watershed are allocated as shown in the table below.

Summary of TMDL for Neshaminy Creek Tributary #2 Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 62,382 6,238 - 56,144 29,336 26,808 stream bank erosion

The TMDLs for sediment are allocated to non-point source loads from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The Waste Load Allocation (WLA) is that portion of the total load that is assigned to point sources, but was zero for sediment in this case. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDLs cover a total of 6.1 miles. The TMDL establishes a reduction for total sediment loading of 66% from the current annual loading of 165,561 pounds.

182

L1.0 INTRODUCTION

L1.1 Watershed Description

The following discussion provides information on the physical characteristics of the Neshaminy Creek Tributary #2 watershed including location, land use distributions, and geology. This watershed is located in the Piedmont Physiographic Province and is situated in Bucks County. It covers an area of approximately 1.8 square miles. Tributary #2 drains into the main stem of Neshaminy Creek from the east. The watershed is located north of the town of Langhorne in eastern Pennsylvania. It can be reached via Pennsylvania Route 432 from the east and Route 413 from the west. Figure L1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the Neshaminy Creek Tributary #2 is warm water fishes and migratory fishes.

The current land use distribution in the watershed was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land uses in the Neshaminy Creek Tributary #2 watershed are developed land (47%). It is important to note that development in the watershed changed from 168 to 216 acres, or a 29% increase from 1992 to 2000.

Figure L1. Neshaminy Creek Tributary #2 Watershed.

183

The surficial geology of the watershed consists of metamorphic/gneiss formations. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.35). Watershed characteristics are summarized in Table L1.

Table L1. Physical Characteristics of Neshaminy Creek Tributary #2 Watershed

Physiographic Province Piedmont Area (square miles) 1.8 Predominant Land Use - Developed land (47%) - Predominant Geology Metamorphic/Gneiss (100%) Soils Dominant HSGs C Average K Factor 0.35 20-Year Average Rainfall 41.5 (in) 20-Year Average Runoff (in) 3.4

L1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for Neshaminy Creek Tributary #2 to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). It was first determined that this stream was not meeting its designated water quality uses for protection of aquatic life in 2001 based on aquatic biological survey. As a consequence, Pennsylvania listed the stream on the 2002 Section 303(d) List of Impaired Waters.

The 2002 303 (d) List reported 1.5 miles of this stream (Stream Segment Id # 980514-1004- GLW) to be impaired by siltation and water/flow variability from land development. The stream segments is impacted by siltation as a result of “New Land Development” in the watershed. New Land Development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in this stream is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events. As indicated above, land development has increased by approximately 29% between 1992 and 2000.

184

sedimentation in streams are typically important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and causes soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1994). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the Neshaminy Creek Tributary #2 watershed. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in this stream. Without effective storm water management practices and sediment traps, build-up of sediments will likely continue to occur.

L2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDLs address impairment by sediments in the stream as reported on the 2002 303(d) Lists. The stream water flow variability impairment caused by urban runoff/storm sewer will not be explicitly addressed by these TMDLs because it is assumed that management practices that will be used to address storm water runoff and sediment production at new construction sites will reduce problems associated with flow variability as well. These TMDLs were derived as follows:

L2.1 Water/Flow Variability Due to Urban Runoff/Storm Sewers

TMDLs were not determined for flow alterations. It was assumed that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow alterations within the watershed.

L2.2 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediments generated by newly developed land in the watershed were the cause of impairment of stream segments in this watershed. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. The TMDLs for this watershed address sediments from construction sites or “Transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA have water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

185

L2.2 Watershed Assessment and Modeling

The AVGWLF model was run for the Neshaminy Creek Tributary #2 watershed to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the 1992 and 2000 land use/ cover conditions were simulated and are shown in Tables L2 and L3, respectively. The Unit Area Load for each pollutant in the watershed was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area for the watershed. Table L4 presents an explanation of the header information contained in Tables L2 and L3. Modeling output for this watershed for 1992 and 2000 land use conditions is presented in Appendix F.

186

Table L2. Loading Values for Neshaminy Creek Tributary #2 Watershed, Year 1992 Land Use Conditions

Sediment Load Unit Area Sed Load Land Use Category Area (acres) (lbs/year) (lbs/acre/yr) Hay/Past 12 198 16.50 Cropland 92 37,439 406.95 Coniferous Forest 20 88 4.40 Mixed Forest 40 154 3.85 Deciduous Forest 124 1,236 9.97 Transitional 0 0 0 Low Int Dev 151 10,795 71.49 High Int Developed 17 552 32.47 Stream Bank 11,920 Groundwater Point Source Septic Systems Total 456 62,382 136.80

Table L3. Loading Values for Neshaminy Creek Tributary #2 Watershed, Year 2000 Land Use Conditions

Sediment Load Unit Area Sed Load Land Use Category Area (acres) (lbs/year) (lbs/acre/yr) Hay/Past 7 110 15.71 Cropland 52 16,600 319.23 Coniferous Forest 20 88 4.40 Mixed Forest 40 154 3.85 Deciduous Forest 121 1,170 9.67 Transition 54 123,863 2,293.76 Low Int Dev 145 10,662 73.53 High Int Dev 17 552 32.47 Stream Bank 12,362 Groundwater Point Source Septic Systems Total 456 165,561 363.02

187

Table L4. Header Information for Tables L2 and L3.

L3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDLs

The load allocation and reduction procedures were applied to the entire Neshaminy Creek Tributary #2 watershed. Sub-watersheds were not delineated due to the watershed’s small size (4 square miles). The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDLs for the watershed.

The equations defining TMDLs for sediment are as follows:

TMDL = MOS + LA + WLA (1) LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) in order for this tributary to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

L3.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for this watershed are based on the sediment loads obtained using the 1992 land use conditions, and are equal to 62,382 lbs/year (see Table L2).

188

L3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDLs were reserved as the MOS.

MOS (Sediments) 62,382 lbs/yr x 0.1 = 6,238 lbs/yr (3)

L3.3 Load Allocation

LA (Sediments) 62,382 lbs/yr – 6,238 lbs/yr = 56,144 lbs/yr (4)

L3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table L3 except for those from transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table L3,

LNR (Sediments) =110 lbs/yr + 16,600 lbs/yr + 88 lbs/yr + 154 lb/yr + 1,170 lb/yr + 10,662 lbs/yr + 552 lbs/yr = 29,336 lbs/yr (5)

ALA (Sediments) = 56,144 lbs/yr – 29,336 lbs/yr= 26,808 lbs/yr (6)

Table L5 below presents TMDLs for this watershed.

Table L5. Summary of TMDL for Neshaminy Creek Tributary #2 Watershed (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 62,382 6,238 - 56,144 29,336 26,808 stream bank erosion

The ALA computed above is that portion of the load that is available to allocate among contributing land use/sources as described in the next step. The following section shows the allocation process in detail for the entire watershed.

L3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in the watershed was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For

189

example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table L6. This essentially means that sediment loads from transitional land uses and stream bank erosion should be reduced to 24,375 and 2,433 pounds, respectively in order for this watershed to attain its specific uses.

Table L6. Load Allocation for Each Contributing Source in the Neshaminy Creek Tributary #2 Watershed.

Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 123,863 91 24,375 80 - Stream bank erosion 12,362 9 2,433 80 TOTAL 136,225 100 26,808 80

Table L7 provides the sediment load allocation when all land uses in the watershed are taken into consideration. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table L7. Sediment Load Allocation by Each Land Use/Source

Land Use Area Unit Area Load Load Reduction Category (acres) (lbs/acre/yr) (lbs/year) ALA (lbs/year) (%) Hay/Past 7 15.71 110 110 0 Cropland 52 319.23 16,600 16,600 0 Conifer Forest 20 4.40 88 88 0 Mixed Forest 40 3.85 154 154 0 Decid Forest 121 9.67 1,170 1,170 0 Transition 54 2,293.76 123,863 24,375 80 Low Int. Dev 145 73.53 10,662 10,662 0 High Int. Dev. 17 32.47 552 552 0 Stream Bank 12,362 2,433 80 Groundwater Point Source Septic Systems Total 456 363.02 165,561 56,144 66

190

The total allowable sediment load in the watershed when all land use/cover sources are considered is 56,144 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from 165,561 pounds per year. Consequently, sediment load should be reduced by 66%.

L4.0 CONSIDERATION OF CRITICAL CONDITIONS

L5.0 CONSIDERATION OF SEASONAL VARIATIONS

L6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDL are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. . These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within the Neshaminy Creek Tributary #2 Creek watershed will be affected by this policy, which has been included in Appendix E.

191

determine the most cost-effective and environmentally protective combination of BMPs required to meet the nutrient and sediment reductions outlined in this report.

L7.0 PUBLIC PARTICIPATION

192

M. Total Maximum Daily Loads (TMDLs) Development Plan for Neshaminy Creek Tributary #3 Watershed

193

Table of Contents Page

Executive Summary ………………………………………………………………… 196

M1.0 Introduction ……………………………………………………………….……. 197 M1.1 Watershed Description …………………………………………….…… 197 M1.2 Surface Water Quality ……………………………………………….…. 198 M2.0 Approach to TMDL Development………………………………………...…….. 199 M2.1. TMDL Endpoints ……………………………………………………… 199 M2.2 Selection of the Reference Watershed…………………………………. 200 M2.3 Water/Flow Variability Municipal Point Sources………………………... 201 M2.4. Siltation due to Construction ……………………………………………… 201 M2.5 Watershed Assessment and Modeling……………………………………. 201 M3.0 Load Allocation Procedure for Sediment TMDL …………..………………… 203 M3.1 Sediment TMDL Total Load ………………………………………..….. 203 M3.2 Margin of Safety ………………………………………...………….. 204 M3.3. Load Allocation …………………………………………..………… 204 M3.4. Adjusted Load Allocation …………………………………………... 204 M3.5. Load Reduction Procedures ………………………………..……….. 205 M4.0 Consideration of Critical Conditions ………………………………………… 206 M5.0 Consideration of Seasonal Variations …………………………………..…….. 206 M6.0 Reasonable Assurance of Implementation …………………………………… 207 M7.0 Public Participation …………………………………………………………. 207

194

List of Tables Page

M1. Physical Characteristics of Neshaminy Creek Tributary #3……………………. 198 M2. Loading Values for Neshaminy Creek Tributary #3 Watershed ………………. 202 M3. Loading Values for the Reference Watershed…………………..……….…… 202 M4. Header Information for Tables M2 and M3………………………………..…. 203 M5. TMDL Total Load Computation………………………………………………. 204 M5. Summary of TMDLs for Neshaminy Creek Tributary #3 Watershed ………… 205 M6. Load Allocation for each Contributing Source in Neshaminy Creek Tributary #3 Watershed……………………………………. 205 M7. Sediment Load Allocation by Land Use/Source …………………………… 206

List of Figures Page

M1. Neshaminy Creek Tributary #3 Watershed ………………………………….…. 197 M2 Reference Watershed ……………………………………………………………… 200

195

EXECUTIVE SUMMARY

The Neshaminy Creek Tributary #3 watershed in Bucks County is about 7 square miles in size. The stream in the watershed are tributaries of Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to 3.3 miles of streams in the watershed (Stream Segment ID#s 980515-1347-GLW and 980515-1348-GLW). They were developed to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water act Section 303(d) List. The impairments are primarily caused by sediment loads from construction sites and water/flow variability from municipal point sources in the watershed. The water/flow variability is not addressed in this TMDL because it was assumed that BMPs that will be implemented to control siltation from urbanized areas will also decrease this flow variability. Therefore, this TMDL focuses on control of sediments.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for sediment in the impaired segments of the Neshaminy Creek Tributary #3 watershed. Based upon comparison to a similar, non-impaired watershed, it was estimated that the sediment loading must be limited to 149,029 pounds per year. It is assumed that streams in the watershed will support their aquatic life uses when this value is met. The TMDL for Neshaminy Creek Tributary #3 is allocated as shown in the table below.

Summary of TMDL for Neshaminy Creek Tributary #3 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediment 292,667 29,267 - 263,400 114,371 149,029

The TMDL is allocated to non-point source loads from construction sites (transitional land use) and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to construction (transitional land use) and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The TMDL covers a total of 3.3 miles of the streams in this watershed. The TMDL establishes a reduction for sediment loading from land construction (transitional land use) and stream bank erosion of 75% from the current annual loading of 1,054,746 pounds.

196

M1.0 INTRODUCTION

M1.1 Watershed Description

The Neshaminy Creek Tributary #3 watershed is located in the Piedmont physiographic province and is in Bucks County. It covers an area of approximately 7 square miles. The streams in the watershed drain directly into the main stem of Neshaminy Creek. The watershed is located north of the town of Jamisson and can be reached via Pennsylvania Route 263 from the east. Figure M1 shows the watershed boundary and its location. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for Neshaminy Creek Tributary #3 is warm water fishes and migratory fishes.

The current land use distribution in the Neshaminy Creek Tributary #3 watershed was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land uses in the watershed include developed land (38%), which includes low and high intensely developed and transitional land use, woodland (33%), and agricultural land (29%). It is important to note that development in the watershed increased from 10% to 38% from 1992 to 2000. The 1994 and 2001 surveys showed that sediment from construction sites deposited in large quantities on the streambed and was degrading the habitat of bottom-dwelling macroinvertebrates. Therefore, for the purposes of this watershed assessment, the amount of land in development during this time period (about 560 acres) was considered to be “transitional” land when modeling current conditions (see additional explanation below).

Figure M1. Neshaminy Creek Tributary #3 watershed.

197

The surficial geology of the Neshaminy Creek Tributary #3 watershed consists of a shale formation. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Watershed characteristics are summarized in Table M1.

Table M1. Physical Characteristic Comparisons between the Neshaminy Creek Tributary #3 and Reference Watersheds

Attribute Neshaminy Creek Tributary #3 Reference Watershed

Physiographic Province Piedmont Piedmont Area (square miles) 10 4 Predominant Land Uses - Developed land (38%) - Agriculture (49%) -Agriculture (29%) -Developed land (33%) Predominant Geology Sandstone (60%) Shale (100%) Shale (40%) Soils - Dominant HSG C C - K Factor 0.37 0.38 20-Year Average Rainfall (in) 40.4 41.4 20-Year Average Runoff (in) 4.1 4.1

M1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for the Neshaminy Creek Tributary #3 watershed to address the impairments noted on Pennsylvania’s 1996 and 2002 Clean Water Act Section 303(d) Lists (see Table A1 in section A1.0). It was first determined that Neshaminy Creek Tributary #3 was not meeting its designated water quality uses for protection of aquatic life in 1994 based on an aquatic biological survey. The 2001 survey found that this stream segment was still impaired. As a consequence, Pennsylvania listed the stream segments in this watershed on the 1996 and 2002 Section 303(d) Lists of Impaired Waters.

The 1996 303 (d) List reported 2 miles of streams in the watershed (Stream Segment ID# 980515-1347-GLW) to be impaired by siltation from construction and water/flow variability due to municipal point sources. The 2002 303 (d) List added 1.3 miles (Stream Segment ID# 980515- 1348-GLW) to be impaired by siltation from construction.

198

construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1990). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the Neshaminy Creek Tributary #3 watershed. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in this watershed. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur in these stream segments.

M2.0 APPROACH TO TMDL DEVELOPMENT

M2.1 TMDL Endpoints

This particular TMDL addresses sediment. Because neither Pennsylvania nor EPA has water quality criteria for sediment, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The method employed for this TMDL is termed the “reference watershed approach.”

With the reference watershed approach, two watersheds are compared, with one attaining its uses and one that is impaired based on biological assessment. Both watersheds must have similar land use/cover distributions. Other features such as base geologic formation should be matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of the process is to reduce the loading rate of nutrients and sediments in the impaired stream segment to a level equivalent to or slightly lower than the loading rate in the non-impaired, reference stream segment. It is assumed that this load reduction will allow the biological community to return to the impaired stream segments.

The TMDL endpoints established for this analysis were determined using a subarea of the Ironworks Creek watershed at the lower end of Neshaminy Creek as the reference watershed. The listing for impairment caused by siltation is addressed through reduction to the sediment load. A detailed explanation of this process is included in the following sections.

Stream segments of the Neshaminy Creek Tributary #3 watershed were found to be impaired by siltation as a result of construction in 1994 and 2000. This TMDL considers that developed land (10% of the 1996 land use distribution) to be primarily construction sites/newly developed land and therefore an important source of sediments in the creek. The additional development between 1992 and 2000 was considered construction sites/newly developed land as well.

199

Therefore, the cause of impairment, construction sites, consisted of “transition” land uses. It appeared from our reconnaissance surveys in the watershed that siltation presently observed in Neshaminy Creek Tributary #3 is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events.

M2.2 Selection of the Reference Watershed

In general, three factors should be considered when selecting a suitable reference watershed. The first factor is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to attaining designated water uses. The second factor is to find a watershed that closely resembles the Neshaminy Creek Tributary #3 watershed in terms of physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS coverages including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The watershed used as a reference for the Neshaminy Creek Tributary #3 watershed was obtained by screen-digitizing a sub-watershed of the Ironworks Creek watershed at the lower end of Neshaminy Creek. This watershed is also located in the Piedmont physiographic province and in State Water Plan (SWP) Basin 2F. Table M1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure M2 shows the reference watershed boundary and its location in Bucks County.

Figure M2. Reference watershed location.

200

An analysis of the MRLC land use/cover grid for the area revealed that land use distributions in both watersheds are somewhat similar. The surficial geology of the Neshaminy Creek Tributary #3 watershed is shale and sandstone, and it is primarily sandstone in the reference watershed. A look at the attributes in Table M1 indicates that these watersheds also compare very well in terms of average runoff, precipitation, and soil K factor.

M2.3 Water/Flow Variability due to Municipal Point Sources

A TMDL was not determined for water/flow variability. It was assumed that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow variability within the watershed.

M2.4 Siltation from Construction Sites

The TMDL for the Neshaminy Creek Tributary #3 watershed addresses sediment from construction sites and from stream bank erosion. As indicated above, existing developed areas in the watershed were considered to be “transitional” land use for modeling purposes since sediment impairment from constructions sites were observed in 1994 and 2001. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses.

M2.5 Watershed Assessment and Modeling

The AVGWLF model was run for both the Neshaminy Creek Tributary #3 and reference watersheds to establish loading conditions under existing land cover use conditions in each watershed using the refined parameter estimates based on the calibration results. Based on the use of 20 years of historical weather data, the mean annual loads for sediment for the impaired and reference watersheds were calculated as shown in Tables M2 and M3, respectively. Table M4 presents an explanation of the header information contained in Tables M2 and M3. Modeling output for both watersheds is presented in Appendix F.

201

Table M2. Loading Values for the Neshaminy Creek Tributary #3 Watershed

Land Use Category Area Sediment Load Unit Area (acres) (lbs/year) Sediment Load (lbs/acre/yr) Hay/Pasture 136 4,569 33.60 Cropland 395 102,539 259.59 Coniferous Forest 7 0 0.00 Mixed Forest 131 352 3.882.70 Deciduous Forest 467 1,832 3.92 Transition 560 870,839 1,555.07 Low Intensity Developed 89 3,642 40.92 High Intensity Developed 42 1,302 31.00 Stream Bank 69,536 Groundwater Point Source Septic Systems Total 1,827 1,054,746 557.31

Table M3. Loading Values for the Reference Watershed

Land Use Category Area Sediment Load Unit Area (acres) (lbs/year) Sediment Load (lbs/acre/yr) Hay/Pasture 622 36,490 58.67 Cropland 528 163,996 310.60 Coniferous Forest 25 44 1.76 Mixed Forest 210 662 3.15 Deciduous Forest 160 375 2.33 Transition 17 17,461 29.90 Low Intensity Developed 646 46,843 72.12 High Intensity Developed 106 4,636 43.74 Stream Bank 100,662 Groundwater Point Source Septic Systems Total 2,317 371,169 160.19

202

Table M4. Header Information for Tables M2 and M3.

M3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire Neshaminy Creek Tributary #3 watershed. The watershed was so small that we did not subdivided it into sub- watersheds. Therefore, sub-watershed load allocations were not performed.

The load reduction calculations in the watershed are based on the current loading rates for sediment in the reference watershed. Based on biological assessment, it was determined that streams in the reference watershed were attaining their designated uses. The sediment loading rates were computed for the reference watershed using the AVGWLF model. These loading rates were then used as the basis for establishing the TMDL for the Neshaminy Creek Tributary #3 watershed.

The equations defining TMDLs for sediments are as follows:

TMDL = MOS + LA + WLA (1) LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Neshaminy Creek Tributary #3 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

M3.1 TMDL Total Load

The first step is to determine the TMDL total target load for Neshaminy Creek Tributary #3, the impaired watershed. This value was obtained by multiplying the sediment unit loading rate in reference watershed by the total watershed area of Neshaminy Creek Tributary #3. This information is presented in M5.

203

Table M5. TMDL Total Load Computation

Unit Area Loading Rate Watershed Area of in Reference Watershed Neshaminy Tributary #3 TMDL Total Load Type of Pollutant (lbs/acre/yr) (acres) (lbs/yr) Sediment 160.19 1,827 292,667

M3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of each of the TMDLs was reserved as the MOS.

Sediment = 292,667 lbs/yr x 0.1 = 29,267 lbs/yr (4)

M3.3 Load Allocation

LA (Sediments) = 292,667 lbs/yr – 29,267lbs/yr = 263,400 lbs/yr (5)

M3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will need reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table M2 except those from agricultural land uses (Hay/Pasture, Row Crops), land development, and stream bank erosion. Therefore, using data in Table M2,

LNR (Sediments) = 4,569 lbs/yr +102,539 lbs/yr + 0 lb/yr + 352 lb/yr+ 1,832 lb/yr + 3,642 lb/yr + 1,302 lb/yr = 114,371 lbs/yr (6)

ALA (Sediments) = 263,400 lbs/yr – 114,371 lbs/yr = 149,029 lbs/yr. (7)

Table M6 below presents the TMDL for the Neshaminy Creek Tributary #3 watershed.

204

Table M6. Summary of TMDL for Neshaminy Creek Tributary #3 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediment 292,667 29,267 - 263,400 114,371 149,029

The ALA computed above is the portion of the load that is available to allocate among contributing sources (transitional land use and stream bank erosion) as described in the next step. The following section shows the allocation process in detail for the entire watershed.

M3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in Neshaminy Creek Tributary #3 was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table M7. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 138,009 and 11,020 pounds, respectively in order for the streams in this watershed to attain their specific uses.

Table M7. Load Allocation for each Contributing Source in the Neshaminy Tributary #3 Watershed.

Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 870,839 93 138,009 84 - Stream bank erosion 69,536 7 11,020 84 TOTAL 940,375 100 149,029 84

It is important to note that load allocation to sub-watersheds was not performed due to the fact that Neshaminy Creek Tributary #3 is small and therefore could not be subdivided into meaningful subwatersheds. Table G8 provides load allocation by considering all land uses in the watershed. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

205

Table M8. Load Allocation by each Land Use/Source

Sediment Unit Area Annual ALA (annual Reduction Source Area Loading Rate average load average) (acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) ( % )

Hay/Pasture 136 33.60 4,569 4,569 0 Cropland 395 259.59 102,539 102,539 0 Coniferous Forest 7 0.00 0 0 0 Mixed Forest 131 3.882.70 352 352 0 Deciduous Forest 467 3.92 1,832 1,832 0 Transition 560 1,555.07870,839 138,009 84 Lo Intensity Dev 89 40.92 3,642 3,642 0 Hi Intensity Dev 42 31.00 1,302 1,302 0 Stream Bank 69,536 11,020 84 Groundwater Point Source Septic Systems Total 1,827 557.31 1,054,746 263,265 75

The total allowable sediment load in Neshaminy Creek Tributary #3 when all land use/cover sources are considered is 263,265 lb per year. In order for all stream segments to attain their specific uses, total sediment loads should be reduced from 1,054,746 pounds per year (i.e., sediment loads should be reduced by 75%).

M4.0 CONSIDERATION OF CRITICAL CONDITIONS

M5.0 CONSIDERATION OF SEASONAL VARIATIONS

206

M6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDL are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. . These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within the Neshaminy Creek Tributary #3 Creek watershed will be affected by this policy, which has been included in Appendix E.

Implementation of BMPs aimed at sediment reduction will also assist in the reduction of phosphorus originating from transitional land uses and stream bank erosion. Other possibilities for attaining the desired reductions in sediment include streambank stabilization and fencing. Further field work will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet sediment reduction outlined in this report.

M7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

207

N. Total Maximum Daily Loads (TMDLs) Development Plan for Neshaminy Creek Tributary #4 Watershed

208

Segments in the Neshaminy Creek Tributary #4 watershed (Stream Segment ID# 980609-1258- GLW) were listed as being impaired by water/flow variability caused by urban runoff/storm sewers (see Figure N1).. A TMDL for this impairment was not developed because neither the U.S. Environmental Protection Agency (EPA) nor PaDEP currently have water quality criteria for this impairment. Furthermore, quantitative measures for water flow variability or alterations as “impairments” are not currently available. However, it is assumed for these segments that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow variability or alterations within the watershed. As discussed previously, all municipalities within the Neshaminy Creek watershed will be affected by PaDEP’s new stormwater management policy (MS4), a copy of which has been included in Appendix E.

Figure N1. Neshaminy Creek Tributary #4 watershed.

209

O. Total Maximum Daily Loads (TMDLs) Development Plan for Mill Creek Sub-basin #1

210

Table of Contents Page

Executive Summary ………………………………………………………………….. 213

O1.0 Introduction ……………………………………………………………….……. 214 O1.1 Watershed Description …………………………………………….…. 214 O1.2 Surface Water Quality ………………………………………………... 215 O2.0 Approach to TMDL Development………………………………………...…….. 215 O2.1 TMDL Endpoints…………………………………………………………. 215 O2.2 Selection of the Reference Watershed……………………………………. 216 O2.3 Flow Alterations Caused by Mining Activities …………………………. 216 O2.4 Siltation Caused by Mining Activities ……………….………….……. 217 O2.5 Watershed Assessment and Modeling…………………………………….. 217 O3.0 Load Allocation Procedure for Nutrients and Sediment TMDLs ……………… 219 O3.1 Sediment TMDL Total Load ………………………………………..….. 219 O3.2 Margin of Safety ………………………………………...……………... 220 O3.3 Load Allocation …………………………………………..…………….. 220 O3.4 Adjusted Load Allocation …………………………………………….… 220 O3.5 Load Reduction Procedures ………………………………..…………... 221 O4.0 Consideration of Critical Conditions …………………………………………… 222 O5.0 Consideration of Seasonal Variations …………………………………..………. 222 O6.0 Reasonable Assurance of Implementation ……………………………………… 223 O7.0 Public Participation …………………………………………………………… 223

211

List of Tables Page

O1. Physical Characteristics of Mill Creek Sub-basin #1………………………… 215 O2. Loading Values for Mill Creek Sub-basin #1………………………………... 218 O3. Loading Values for the Reference Watershed…………….…………………. 218 O4. Header Information for Tables O2 and O3………………………………..…. 219 O5. TMDL Total Load Computation……………………………………………… 220 O6. Summary of TMDLs for Mill Creek Sub-basin #1………….…………....… 220 O7. Load Allocation for each contributing source in Mill Creek Sub-basin #1……… 221 O8. Sediment Load Allocation by Land Use/Source ……………………….……… 222

List of Figures Page

O1. Mill Creek Sub-basin #1 ………………………………………………….…. 214 O2 Reference Watershed ………………………………………………………… 217

212

EXECUTIVE SUMMARY

Mill Creek Sub-basin #1is located in Bucks County and is about 2.3 square miles in size. This particular sub-basin contains small streams that are tributaries of Mill Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is cold water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to 0.8 miles of streams in this sub-basin (Stream Segment ID# 20000525-1017-GLW). They were developed to address the impairments noted on the Pennsylvania’s 2002 Clean Water act Section 303(d) List. The impairments are primarily caused by sediment loads and flow alterations from surface mining in the watershed. This TMDL focuses on control of sediments. TMDLs for flow alterations were not addressed because neither the U.S. Environmental Protection Agency (EPA) or PaDEP currently have water quality criteria for this impairment. Furthermore, quantitative measures for water flow variability or alterations as “impairments” are not currently available. However, it was assumed for these segments that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow variability or alterations within the watershed.

Pennsylvania does not currently have water quality criteria for sediments. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for sediment in the impaired segments of the Mill Creek Sub-basin #1. Based upon comparison to a similar, non-impaired watershed, it was estimated that the amount of sediment loading that will meet the water quality objectives for Mill Creek Sub-basin #1 is 48,669 pounds per year. It is assumed that the streams in this sub-basin will support designated aquatic life uses when this value is met. The TMDL for Mill Creek Sub-basin #1 is allocated as shown in the table below.

Summary of TMDL for Mill Creek Sub-basin #1 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediment 78,261 7,826 - 70,435 21,766 48,669

The TMDL has been allocated to non-point source loading from mining activities, with 10% of the TMDL total load reserved as a margin of safety (MOS). The Waste load allocation (WLA) is that portion of the total load that is assigned to point sources. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to mining and agricultural activities, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The TMDL covers a total of 0.8 miles of streams within Mill Creek Sub-basin #1. The TMDL establishes a reduction for sediment loading from agricultural and mining activities of 52% from the current annual loading of 141,366 pounds.

213

O1.0 INTRODUCTION

O1.1 Watershed Description

Mill Creek Sub-basin #1 is located in the Piedmont physiographic province, is situated in Bucks County, and covers an area of approximately 2.3 square miles. Mill Creek Sub-basin #1 is part of the larger Mill Creek watershed which drains into the main stem of Neshaminy Creek from the west. The watershed is located south of the town of Tradesville and north of Warrington in eastern Pennsylvania. It is bounded by Pennsylvania Route 611 to the south and Route 152 to the west. Figure O1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for Mill Creek Sub-basin #1 is cold water fishes and migratory fishes.

The primary land uses in the sub-basin are agriculture (53%) and forested land (35%). It was also found from a field survey of the watershed that sediment was being deposited in large quantities on the streambed and was degrading the habitat of bottom-dwelling macroinvertebrates.

Figure O1. Mill Creek Sub-basin #1.

214

The surficial geology of Mill Creek Sub-basin #1 consists of a shale formation. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Watershed characteristics are summarized in Table O1.

Table O1. Physical Characteristic Comparisons between Mill Creek Sub-basin #1 and Reference Watershed

Attribute Mill Creek Subbasin #1 Reference Watershed Watershed Physiographic Province Piedmont Piedmont Area (square miles) 2.3 2.6 Predominant Land Uses -Agriculture (53%) -Agriculture (61%) -Forested land (35%) -Forested land (38%) Predominant Geology Shale (100%) Shale (100%) Soils - Dominant HSG C C - K Factor 0.37 0.37 20-Year Average Rainfall (in) 40.4 40.4 20-Year Average Runoff (in) 3.7 4.5

O1.2 Surface Water Quality

Total Maximum Daily Load or TMDL was developed for Mill Creek Sub-basin #1 to address the impairments noted on the Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). It was previously determined that Mill Creek Sub-basin #1 was not meeting its designated water quality uses for protection of aquatic life in 2001. As a consequence, Pennsylvania listed 0.8 miles of Mill Creek Sub-basin #1 (Stream Segment ID# 20000525-1017-GLW) on the 2002 Section 303(d) List of Impaired Waters as being impaired by siltation and flow alteration from mining operations.

O2.0 APPROACH TO TMDL DEVELOPMENT

O2.1 TMDL Endpoints

The TMDL discussed herein address sediments. Because neither Pennsylvania nor EPA have water quality criteria for sediment, we had to develop a method to determine water quality objectives for these parameters that would result in the impaired stream segments attaining their designated uses. The method employed for these TMDLs is termed the “reference watershed approach.”

With the reference watershed approach, two watersheds are compared, with one attaining its uses and one that is impaired based on biological assessment. Both watersheds must have

215

similar land use/cover distributions. Other features such as base geologic formation should be matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of the process is to reduce the loading rate of nutrients and sediments in the impaired stream segment to a level equivalent to or slightly lower than the loading rate in the non-impaired, reference stream segment. The underlying assumption is that this load reduction will allow the biological community to return to the impaired stream segments.

The TMDL endpoints established for this analysis were determined using a portion of Robin Creek as the reference watershed. These endpoints are discussed in detail in the TMDL section. The listing for impairment caused by siltation is addressed through reduction of the sediment load. A detailed explanation of this process is included in the following section.

O2.2 Selection of the Reference Watershed

In general, three factors should be considered when selecting a suitable reference watershed. The first factor is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to be attaining sufficient water quality to satisfy designated uses. The second factor is to find a watershed that closely resembles Mill Creek Sub-basin #1 in terms of physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS coverages including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The watershed used as a reference for Mill Creek Sub-basin #1 was obtained by screen- digitizing a portion of Robin Creek watershed. This watershed is also located in the Piedmont physiographic province in State Water Plan (SWP) Basin 2F. An analysis of the MRLC land use/cover grid revealed that land cover/use distributions in both watersheds are similar. Characteristics of both watersheds are summarized in Table O1, and appear to compare favorably in terms of average runoff, precipitation, and soil K factor. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure O2 shows the reference watershed boundary and its location in Bucks County.

O2.3 Flow Alterations Due to Mining Activities

A TMDL was not determined for water/flow variability. It was assumed that addressing sediment loads through the use of various BMPs will at the same time reduce water flow variability within the watershed.

216

Figure O2. Reference watershed location.

O2.4 Siltation Due to Mining Activities

The 2001 survey showed that siltation originating from mining activities in the watershed was the cause of impairment of Mill Creek Sub-basin #1 stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses.

The objective of the TMDL process for Mill Creek Sub-basin #1 is to reduce the average loading rate of sediment to the impaired stream segment to levels equivalent to or slightly lower than the average loading rate in the reference watershed. It is assumed that this load reduction will allow the biological community to return to the impaired stream segments. The TMDL endpoints established for this analysis are discussed in detail in the TMDL section. The listing for impairment caused by siltation is addressed through reduction of sediment loads, respectively.

O2.5 Watershed Assessment and Modeling

The AVGWLF model was run for both Mill Creek Sub-basin #1 and the reference watershed to establish loading conditions under existing land use/cover conditions in each watershed using the refined parameter estimates based on the calibration results. Based on the use of 20 years of historical weather data, the mean annual sediment loads for the impaired and reference watersheds were calculated as shown in Tables O2 and O3, respectively. Table O4 presents an

217

explanation of the header information contained in Tables O2 and O3. Modeling output for Mill Creek Sub-basin #1 and the reference watershed is presented in Appendix F.

Table O2. Existing Loading Values for Mill Creek Sub-basin #1.

Land Use Category Area (acres) Sediment Load Unit Area Sediment Load (lbs/year) (lbs/acre/yr) Hay/Past 20 552 27.60 Cropland 288 83,642 290.42 Coniferous Forest 2 0 0.00 Mixed Forest 22 44 2.00 Deciduous Forest 178 662 3.72 Unpaved Road 2 2,097 1,048.50 Quarries 22 14,658 666.27 Coal Mines1 27 18,653 690.85 High Intensity Developed 20 309 15.45 Stream Bank 20,751 Groundwater Point Source Septic Systems Total 581 141,366 243.32

1Although identified as coal mines, these areas are actually quarried land

Table O3. Loading Values for the Reference Watershed

Land Use Category Area (acres) Sediment Load Unit Area Sediment Load (lbs/year) (lbs/acre/yr) Hay/Past 94 2,914 31.00 Cropland 311 62,384 200.59 Coniferous Forest 20 44 2.20 Mixed Forest 69 110 1.59 Deciduous Forest 165 353 2.14 Unpaved Roads 2 0 0.00 High Intensity Developed 5 66 13.20 Stream Bank 23,841 Groundwater Point Source Septic Systems Total 666 89,712 134.70

218

Table O4. Header Information for Tables O2 and O3.

O3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire area of Mill Creek Sub-basin #1. The watershed was so small that we did not subdivide it into sub-watersheds. Therefore, sub-watershed load allocations were not performed.

The load reduction calculations in Mill Creek Sub-basin #1 are based on the current loading rates for sediments in the reference watershed. Based on biological assessment, it was determined that the reference watershed was attaining its designated uses. Sediment loading rates were computed for the reference watershed using the AVGWLF model. These loading rates were then used as the basis for establishing the TMDL for Mill Creek Sub-basin #1.

The equations defining TMDL for sediment are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) in order for Mill Creek Sub- basin #1 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

O3.1 TMDL Total Load

The first step was to determine the TMDL total target load for Mill Creek Sub-basin #1, the impaired watershed. This value was obtained by multiplying the sediment unit area loading rate in the reference watershed by the total watershed area of Mill Creek Sub-basin #1. This information is presented in Table O5.

219

Table O5. TMDL Total Load Computation

Unit Area Loading Rate Total Watershed Area in in Reference Watershed Mill Creek Sub-basin #1 TMDL Total Load Type of Pollutant (lbs/acre/yr) (acres) (lbs/yr) Sediment 134.70 581 78,261

O3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

Sediment - 78,261 lbs/yr x 0.1 = 7,826 lbs/yr (4)

O3.3 Load Allocation

The Load allocation (LA), consisting of all nonpoint source loads in the watershed, was computed by subtracting the margin of safety and the waste load allocation (WLA) from the TMDL total load. In this case, sediments did not have a waste load allocation.

LA (Sediments) = 78.261 lbs/yr – 7,826 lbs/yr = 70,435 lbs/yr (5)

O3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will need reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table O2 except those from quarries and agricultural land uses (Hay/Past, Row Crops). It is assumed that agricultural land use, which occupies a large portion of the watershed, contributed also to the deposition of sediment in the impaired stream segments. Therefore, using data in Table O2,

LNR (Sediments) = 0 lb/yr + 44 lbs/yr + 662 lb/yr + 2097 lbs/yr + lbs/yr + 309 lbs/yr + 20,751= 23,863 lbs/yr (6)

ALA (Sediments) = 70,435 lbs/yr – 23,863 lbs/yr = 48,669 lbs/yr. (7)

Table O6 below summarizes the sediment TMDL for Mill Creek Sub-basin #1.

Table O6. Summary of TMDL for Mill Creek Sub-basin #1 (lbs/yr) Pollutant TMDL MOS WLA LA LNR ALA Sediment 78,261 7,826 - 70,435 21,766 48,669

220

The ALA computed above is the portion of the load that is available to allocate among contributing sources (Hay/Past, Cropland) and quarries as described in the next section. The following section shows the allocation process in detail for the entire watershed.

O3.5 Load Reduction Procedures

Sediment loads obtained in the previous step were allocated among the remaining land use/cover sources of the impaired watershed according to the Equal Marginal Percent Reduction (EMPR) method. EMPR is carried out using an Excel Worksheet in the following manner:

5) Each land use/source load is compared with the total allocable load to determine if any contributor would exceed the allocable load by itself. The evaluation is carried out as if each source is the only contributor to the pollutant load to the receiving waterbody. If the contributor exceeds the allocable load, that contributor would be reduced to the allocable load. This is the baseline portion of EMPR.

6) After any necessary reductions have been made in the baseline the multiple analysis is run. The multiple analysis will sum all of the baseline loads and compare them to the total allocable load. If the allocable load is exceeded, an equal percent reduction will be made to all contributors’ baseline values. After any necessary reductions in the multiple analysis, the final reduction percentage for each contributor can be computed.

It should be noted that load allocation to sub-watersheds was not performed due to the fact that Mill Creek Sub-basin #1 is a small watershed and therefore could not be subdivided into meaningful sub-watersheds, and also consists of only one stream segment. The load allocation and EMPR procedures were performed using an Excel Worksheet and results are presented in Appendix G. Results of the load allocation by land use sources are presented in Table O7. Table O8 provides load allocation by considering all land uses in Mill Creek Sub-basin #1. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table O7. Load Allocation by Each Contributing Source in Mill Creek Sub-basin #1.

Land Use/ Area Sediment Source Loading Average Average ALA Reduction Rate Load (acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) ( % ) Hay/Pasture 20 27.60 522 308 41 Cropland 288 290.4283,642 28,710 66 Quarries 22 666.2714,658 3,546 76 Coal Mines1 27 690.85 18,653 11,004 41 Sub-total 357 329.06 117,475 43,568 63

1Although identified as coal mines, these areas are actually quarried land

221

Table O8. Load Allocation by Each Land Use/Source

Sediment

Source Area Unit Area Loading Annual average ALA (annual Reduction Rate load average) Acres lbs/ac/yr lbs/yr lbs/yr - % -

Hay/Past 20 27.60 552 308 41 Cropland 288 290.42 83,642 28,710 66 Coniferous 2 0.00 0 0 0 Mixed For 22 2.00 44 44 0 Deciduous 178 3.72 662 662 0 Unpaved Roads 2 1,048.50 2,097 2,097 0 Quarries 22 666.27 14,658 3,546 76 Coal Mines1 27 690.85 18,653 11,004 41 High Intensity Dev 20 15.45 309 309 0 Stream Bank 20,751 20,751 0 Groundwater Point Source Septic Systems Total 581 243.32 141,366 67,431 52

1Although identified as coal mines, these areas are actually quarried land.

The total allowable sediment loads to Mill Creek Sub-basin #1 when all land use/cover sources are considered is 67,431 pounds per year. In order for the stream to attain its specific uses, total sediment loads should be reduced from 141,366 pounds per year or by 52%.

O4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment to a waterbody and the resulting impact on beneficial uses, establishing this TMDL using average annual conditions is protective of the waterbody.

O5.0 CONSIDERATION OF SEASONAL VARIATIONS

222

each month. The model also considers the months of the year when manure is applied to the land. The combination of these actions by the model accounts for seasonal variability.

O6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDL are allocated to agricultural land uses and quarried land in the watershed. Implementation of best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches in the mining areas. These BMPs range in efficiency from 20% to 70% for sediment reduction.

Other possibilities for attaining the desired reductions in sediment include streambank stabilization and fencing. Further field verification will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet the sediment reductions outlined in this section.

O7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

223

P. Total Maximum Daily Loads (TMDLs) Development Plan for Mill Creek Sub-basin #2

224

Table of Contents Page

Executive Summary …………………………………………………………………… 227

P1.0 Introduction ……………………………………………………………….……. 228 P1.1 Watershed Description ….………………………………………….…. 228 P1.2 Surface Water Quality ….……………………………………………... 229 P2.0 Approach to TMDL Development………………………………………...…….. 230 P2.1 Flow Alterations Due to Urban Runoff/Storm Sewers…………………. 230 P2.2 Siltation Caused by Urban Runoff/Storm Sewers….……………….……. 230 P2.3 Watershed Assessment and Modeling…………………………………….. 231 P3.0 Load Allocation Procedure for Sediment TMDL………………. . ……………… 233 P3.1 Sediment TMDL Total Load ………………………………………..….. 233 P3.2 Margin of Safety ………………………………………...………………. 234 P3.3. Load Allocation ………………………………………………………… 234 P3.4. Adjusted Load Allocation ………………………………………………. 234 P3.5. Load Reduction Procedures ………………………………..…………… 235 P4.0 Consideration of Critical Conditions …………………………………………. 236 P5.0 Consideration of Seasonal Variations …………………………………..…….. 236 P6.0 Reasonable Assurance of Implementation …………………………………… ….. 236 P7.0 Public Participation …………………………………………………………. 237

225

List of Tables Page

P1. Physical Characteristics of Mill Creek Sub-basin #2………………………………. 229 P2. Loading Values for Mill Creek Sub-basin #2, Year 1992 Land Use Conditions …… 232 P3. Loading Values for Mill Creek Sub-basin #2, Year 2000 Land Use Conditions …… 232 P4. Header Information for Tables G2 and G3………………………………..………… 233 P5. Summary of TMDLs for Mill Creek Sub-basin #2 …………………………..…… 234 P6. Load Allocation for each contributing source in Mill Creek Sub-basin #2………… 235 P7. Sediment Load Allocation by Land Use/Source ………………………….... …… 235

List of Figures Page

P1. Mill Creek Sub-basin #2……………….…………………………………………. 228

226

EXECUTIVE SUMMARY

Mill Creek Sub-basin #2 is situated in Bucks County and is about 5.3 square miles in size. It is a part of the larger Neshaminy Creek watershed, and its protected uses of the watershed are water supply, recreation, and aquatic life. Its designated aquatic use is cold water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to about 6.1 miles of streams in the watershed (Stream Segment Ids # 20010426-1512-GLW and 980609-1425-GLW). They were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water act Section 303(d) List. The impairments are primarily caused by sediment loads from land development in the watershed. The TMDL focuses on control of sediments. Stream Segment ID # 980609-1425-GLW is also impacted by water/flow variability due to urban runoff/storm sewers. Water/flow variability was not explicitly addressed because it was believed that the implementation of BMPs in the developed areas to reduce sediment would also decrease water flow and volume to the stream and therefore stabilize stream flow.

Pennsylvania does not currently have water quality criteria for sediment. For this reason, a modeling approach was developed to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of Mill Creek Sub-basin #2. The approach is based on the comparison of simulated sediment loads at two time periods: Year 1992 when the stream was still attaining and Year 2000 when it was found to be impaired. Siltation, the cause of impairment in Mill Creek Sub-basin #2, has resulted from the accumulation of sediments originating from construction and newly developed land over several years. It was estimated that the amount of sediment loading that will meet the water quality objectives for Mill Creek Sub- basin #2 is 206,370 pounds per year. It is assumed that Mill Creek Sub-basin #2 will support its aquatic life uses when this value is met. The sediment TMDLs for this watershed are allocated as shown in the table below.

Summary of TMDLs for Mill Creek Sub-basin #2 (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 336,092 33,609 - 302,483 96,113 206,370 stream bank erosion

The TMDLs for sediments are allocated to non-point source from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The Waste load allocation (WLA) is that portion of the total load that is assigned to point sources but was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDLs cover a total of 6.1 miles. The TMDL establishes a reduction for total sediment loading of 10% from the current annual loading of 336,090 pounds.

227

P1.0 INTRODUCTION

P1.1 Watershed Description

The following discussion provides information on the physical characteristics of Mill Creek Sub-basin #2 including its location, land use distributions, and geology. Mill Creek Sub-basin #2 is located in the Piedmont Physiographic Province and is entirely located in Bucks County. It covers an area of approximately 5.3 square miles. Mill Creek Sub-basin #2 is one of two sub- basins that comprises the larger Mill Creek basin, which drains into the main stem of Neshaminy Creek from the west. The watershed is located south of the town of Tradesville and north of Warrington in eastern Pennsylvania. It is bounded by Pennsylvania Route 611 to the south and Route 152 to the west. Figure N1 shows the watershed boundary, its location, and water quality status of stream segments as reported on the 2002 303(d) List. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the Mill Creek Sub-basin #2 is cold water fishes and migratory fishes.

The current land use distribution in the Mill Creek Sub-basin #2 was developed by updating the National Land Cover Data (NLCD) layer described by Vogelmann et al. (1998) using a recent 10-m colorized panchromatic SPOT (System Probatoire pour l’Observation de la Terre) satellite image. The NLCD layer was based primarily on 1992 Landsat Thematic Mapper (TM). SPOT imagery was acquired in 2000 and is available for the entire Commonwealth of Pennsylvania at the Pennsylvania Spatial Data Access (PASDA) site (http://spot.pasda.psu.edu) at no charge. The primary land uses in the watershed are agriculture (37%) and forested land (37%), followed by developed land (16%). It is important to note that development in the watershed changed from 208 to 346 acres, or a 66% increase from 1992 to 2000.

Figure P1. Mill Creek Sub-basin #2.

228

The surficial geology of Mill Creek Sub-basin #2 consists of a shale formation. The bedrock geology affects primarily surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.37). Watershed characteristics are summarized in Table P1.

P1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for the Mill Creek Sub-basin #2 to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). It was first determined that the watershed was not meeting its designated water quality uses for protection of aquatic life in 2001 based on aquatic biological survey. As a consequence, Pennsylvania listed the streams in this watershed on the 2002 Section 303(d) List of Impaired Waters.

Table P1. Physical Characteristics of Mill Creek Sub-basin #2

Physiographic Province Piedmont Area (square miles) 5.3 Predominant Land Use - Agriculture (37%) - Forested land (37%) - Developed land (16%) Predominant Geology Shale (100%) Soils Dominant HSGs C Average K Factor 0.37 20-Year Average Rainfall 40.4 (in) 20-Year Average Runoff (in) 4.4

The 2002 303 (d) List reported 6.1 miles of stream in this watershed (Stream Segment Ids # 20010426-1512-GLW and 980609-1425-GLW) to be impaired by siltation from land development and flow alterations as a result of urban runoff/storm sewers. These stream segments are impacted by siltation as a result of “New Land Development” in the watershed. New Land Development is defined here as disturbed land at construction sites/new development. It appeared from our reconnaissance surveys and contacts in the watershed that siltation presently observed in the sub-basin is the result of years of a build-up of sediments in the channel bottom that started in the early 1990’s. These sediments originated from disturbed and unprotected soils at construction sites and increased channel bank erosion during periods of intense storm events. As indicated above, land development has increased by approximately 66% between 1992 and 2000.

229

Sediments, which are often the cause of stream impairment in urban and suburban areas, are primarily from two sources: disturbed land and unprotected soils at construction sites, and stream channel erosion. Transitional land uses, mainly new construction sites, are one of the main sources of sediments in streams draining newly developed areas. Sediment production and sedimentation in streams are typically important during the construction phase because soils are disturbed and exposed to detachment by raindrops and transported during storm events. Construction also renders landscapes unstable and cause soil to move in “sheets” and localized landslides during storm events.

Channel erosion and scour that occur in waterways and receiving waters located in urban and suburban areas may also be an important source of sediments. Channel erosion is primarily the result of elevated storm water runoff during storm events caused by increased impervious surfaces from residential, commercial and industrial areas; construction sites; roads; highways; and bridges in the watershed (Horner, 1994). Basically, impervious areas and disturbed land restrict water infiltration thus converting more rainfall into runoff during storm events. The visible impact of elevated storm runoff includes fallen trees, eroded and exposed stream banks, siltation, floating litter and debris, and turbid conditions in streams. All these events were observed during a reconnaissance survey of the watershed. In conclusion, addressing storm water runoff and sediment production at new construction sites through the use of management practices will assure that aquatic life use is achieved and maintained in the watershed. Without effective storm water management practices and sediment traps, build-up of sediments will continue to occur.

P2.0 APPROACH TO TMDL DEVELOPMENT

The present TMDLs address impairment by sediments in Mill Creek Sub-basin #2 stream segments as reported on the 2002 303(d) Lists. The stream water flow variability impairment caused by urban runoff/storm sewer will not be explicitly addressed by these TMDLs because it is assumed that management practices that will be used to address storm water runoff and sediment production at new construction sites will reduce problems associated with flow variability as well. These TMDLs were derived as follows:

P2.1 Flow Alterations Resulting from Residential Runoff

TMDLs were not determined for flow alterations. It was assumed that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow alterations within the watershed.

P2.2 Siltation Caused by Urban Runoff/Storm Sewers

The 2001 survey showed that sediments caused by newly developed land in the watershed were the cause of impairment of Mill Creek Sub-basin #2 stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. The TMDLs for this watershed address sediments from construction sites or “Transitional” land uses, and from stream bank erosion. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality

230

objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The approach consists of:

P2.3 Watershed Assessment and Modeling

The AVGWLF model was run for Mill Creek Sub-basin #2 to establish sediment loadings under differing land use/cover conditions (see section A for model-specific details). First, the model was run using the 1992 land use distributions provided by the National Land Cover Data (NLCD) set. As indicated earlier, NLCD land uses were developed by the MRLC Consortium using primarily a 1992 Landsat TM imagery. Second, the model was performed for the Year 2000 land use conditions using an updated version of this earlier land use data set. SPOT imagery that was acquired in the summer of 2000 was used for the land use update. In this model, land in transition (transitional land use) was considered to be new development (built after 1992) or construction sites.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for Mill Creek Sub-basin #2. Based on the use of 20 years of historical weather data, the mean annual loads for sediments, N and P for the 1992 and 2000 land use/cover conditions were calculated and are shown Tables P2 and P3, respectively. The Unit Area Load for sediment in the watershed was estimated by dividing the mean annual loading (lbs/yr) by the total area (acres) resulting in an approximate loading per unit area for the watershed. Table P4 presents an explanation of the header information contained in Tables P2 and P3. Modeling output for Mill Creek Sub-basin #2 for 1992 and 2000 land use conditions is presented in Appendix F.

231

Table P2. Loading Values for Mill Creek Sub-basin #2, Year 1992 Land Use Conditions

Land Use Category Area (acres) Sediment Load Unit Area Sed Load (lbs/year) (lbs/acre/yr) Hay/Past 118 3,267 27.69 Cropland 456 99,073 217.27 Coniferous Forest 18 22 1.22 Mixed Forest 153 243 1.59 Deciduous Forest 388 173 0.45 Transitional 0 0 0 Low Int Dev 168 4,503 26.80 High Int Developed 40 1,015 25.37

Stream Bank 95,143 Groundwater Point Source Septic Systems Total 1,341 204,061 152.17

Table P3. Existing Loading Values for Mill Creek Sub-basin #2, Year 2000 Land Use Conditions

Land Use Category Area (acres) Sediment Load Unit Area Sed Load (lbs/year) (lbs/acre/yr) Hay/Past 92 1,876 20.39 Cropland 402 87,792 218.39 Coniferous Forest 18 22 1.22 Mixed Forest 155 243 1.57 Deciduous Forest 328 662 2.02 Transition 138 141,964 1,028.72 Low Int Dev 168 4,503 26.81 High Int Dev 40 1,015 25.38 Stream Bank 98,013 Groundwater Point Source Septic Systems Total 1,341 336,090 250.63

232

Table P4. Header Information for Tables P2 and P3.

P3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire area of Mill Creek Sub-basin #2. Sub-watersheds were not delineated due to the watershed’s small size (4 square miles). The load reduction calculations are based on sediment loads that were obtained using 1992 land use conditions. This assumes that the watershed was attaining its designated uses prior to 1992. As indicated earlier, land development, which is the source of stream impairment in the watershed, has increased considerably since 1992. These loads were then used as the basis for establishing the TMDL for Mill Creek Sub-basin #2.

The equations defining TMDLs for sediment are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) in Mill Creek Sub-basin #2 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

P3.1 Sediment TMDL Total Load

As noted earlier, the TMDL total target loads for Mill Creek Sub-basin #2 are based on the sediment loads obtained using the 1992 land use conditions, and are equal to 336,092 lbs/year (see Table P2).

233

P3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDLs were reserved as the MOS.

MOS (Sediments) 336,092 lbs/yr x 0.1 = 33,609 lbs/yr (3)

P3.3 Load Allocation

LA (Sediments) 336,092 lbs/yr - 33,609 lbs/yr = 302,483 lbs/yr (4)

P3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will require reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table P3 except those from the transitional land use and stream bank erosion. Notice that loads from stream bank erosion were not adjusted. Therefore, using data in Table P3,

LNR (Sediments) =1,876 lbs/yr + 87,792 lbs/yr + 22 lbs/yr + 243 lb/yr + 662 lb/yr + 4,503 lbs/yr + 1,015 lbs/yr = 96,113 lbs/yr (5)

ALA (Sediments) = 302,483 lbs/yr – 96,113 lbs/yr= 206,370 lbs/yr (6)

Table P5 below presents the TMDL for Mill Creek Sub-basin #2.

Table P5. Summary of TMDL for Mill Creek Sub-basin #2 (lbs/yr)

Pollutant Source TMDL MOS WLA LA LNR ALA Sediment Transitional land and 336,092 33,609 - 302,483 96,113 206,370 stream bank erosion

234

P3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in Mill Creek Sub- basin #2 was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table P6. This means that sediment loads from transitional land uses and stream bank erosion should be reduced to 122,083 and 84,287 pounds, respectively for Mill Creek Sub-basin #2 to attain its specific uses.

Table P6. Load Allocation for Each Contributing Source in Mill Creek Sub-basin #2. Pollutant Source Current Load ALA Reduction Lbs/year % Lbs/year -%- Sediment - Transitional land use 141,964 59 122,083 14 - Stream bank erosion 98,013 22 84,287 14 TOTAL 239,977 100 206,370 14

Table P7 provides the sediment load allocation when all land uses in the Mill Creek Sub-basin #2 are taken into consideration. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table P7. Sediment Load Allocation by Each Land Use/Source Land Use Category Area Unit Area Load Load ALA Reduction (acres) (lbs/acre/yr) (lbs/year) (lbs/year) (%) Hay/Pasture 92 20.39 1,876 1,876 0 Cropland 402 218.39 87,792 87,792 0 Conifer Forest 18 1.22 22 22 0 Mixed Forest 155 1.57 243 243 0 Decid Forest 328 2.02 662 662 0 Transition 138 1,028.72141,964 122,083 14 Low Intensity Dev 168 26.81 4,503 4,503 0 High Intensity Dev 40 25.38 1,015 1,015 0 Stream Bank 98,013 84,287 14 Groundwater Point Source Septic Systems Total 1,341 250.63 336,090 302,483 10

235

The total allowable sediment load in Mill Creek Sub-basin #2 when all land use/cover sources are considered is 302,483 pounds per year. In order for all stream segments to attain their specific uses, the current total sediment load of 336,090 pounds per year should be reduced by 10%.

P4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing this TMDL using average annual conditions is protective of the waterbody.

P5.0 CONSIDERATION OF SEASONAL VARIATIONS

P6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

Sediment reductions in the TMDL are allocated to transitional land uses and stream bank erosion in the watershed. Implementation of best urban best management practices (BMPs) in the affected areas to increase infiltration and sediment control measures should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the installation of drainage controls such as detention ponds, sediment ponds, infiltration pits, dikes and ditches. These BMPs range in efficiency from 20% to 70% for sediment reduction. The implementation of such BMPs will likely occur in the watershed as a result of PaDEP’s Proposed Comprehensive Stormwater Management Policy. When approved, this new policy will require affected communities to implement BMPs to address stormwater control that will “reduce pollutant loadings to streams, recharge groundwater tables, enhance stream base flow during times of drought and reduce the threat of flooding and stream bank erosion resulting from storm events.” Over the next year and one-half, PaDEP will be developing a “Phase II” program for NPDES discharges from small construction sites, additional industrial activities, and for the 700 municipalities subject to the requirements for separate storm sewer systems (MS4). All of the municipalities located within Mill Creek Sub-basin #2 Creek will be affected by this policy, which has been included in Appendix E.

236

determine the most cost-effective and environmentally protective combination of BMPs required to meet the sediment reductions outlined in this report.

P7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60-day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

237

Q. Total Maximum Daily Load (TMDL) Development Plan for Core Creek

238

Table of Contents Page

Executive Summary ………………………………………………………………... 241

Q1.0 Introduction ……………………………………………………………….……. 242 Q1.1 Watershed Description …………………………………………….…. 242 Q1.2 Surface Water Quality ………………………………………………... 243 Q2.0 Approach to TMDL Development…………………………………………….. 243 Q2.1 TMDL Endpoints………………………………………………………………. 243 Q2.2 Selection of the Reference Watershed…………………………………………… 244 Q2.3 Siltation from Agricultural Activities………………………………………….. 244 Q2.4 Watershed Assessment and Modeling…………………………………………. 245 Q3.0 Load Allocation Procedure for Sediment TMDLs ……………..……………… 247 Q3.1 Sediment TMDL Total Load ………………………………………..….. 247 Q3.2 Margin of Safety ………………………………………...……………… 248 Q3.3 Load Allocation…………………………………………………………. 248 Q3.4 Adjusted Load Allocation………………………………………………. 248 Q3.5 Load Reduction Procedures…………………………………………….. 249 Q4.0 Consideration of Critical Conditions ………………………………………… 250 Q5.0 Consideration of Seasonal Variations …………………………………..…….. 251 Q6.0 Reasonable Assurance of Implementation …………………………………… 251 Q7.0 Public Participation …………………………………………………………. 251

239

List of Tables Page

Q1. Physical Characteristic Comparisons between Core Creek and Reference Watershed ………………………….……………………………. 243 Q2. Loading Values for Core Creek Watershed…………………………………. 246 Q3. Loading Values for the Reference Watershed…………….………………… 246 Q4. Header Information for Tables Q2 and Q3………………………………..…. 247 Q5. Summary of TMDL for Core Creek Watershed ……………..…………....… 248

List of Figures Page

Q1. Core Creek Watershed ……………….………………………………….…. 242 Q2 Reference Watershed ……..………………………………………………… 245

240

EXECUTIVE SUMMARY

The Core Creek watershed is about 23 square miles in size, is located in Bucks County, and drains into the main stem of Neshaminy Creek. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is cold water fishes in the upper part of the stream, warm water fishes in the lower part and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to 15.8 miles of Core Creek (Stream Segment ID# 980602-0954-GLW). They were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water Act Section 303(d) List. Pennsylvania does not currently have water quality criteria for sediments. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for sediment in the impaired segments of the Core Creek watershed. Based upon comparison to a similar, non- impaired watershed, it was estimated that the sediment loading that will meet the water quality objectives for Core Creek is 724,139 pounds per year. The TMDL for Core Creek is allocated as shown in the table below.

Summary of TMDL for Core Creek (lbs/yr) Pollutant TMDL MOS WLA LA LNR ALA Sediment 1,474,723 147,472 - 1,327,251 603,112 724,139

The TMDL for sediment is allocated to non-point source loads from transitional (i.e., “developing”) land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediment. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 15.8 miles. The TMDL establishes a reduction for total sediment loading of 26% from the current annual loading of 1,775,981 pounds.

241

Q1.0 INTRODUCTION

Q1.1 Watershed Description

The Core Creek watershed is located in the Piedmont Physiographic Province, and is located in Bucks County. It covers an area of approximately 23 square miles. Core Creek drains into the main stem of Neshaminy Creek from the east (see Figure Q1). The watershed is located east of the town of Newtown and north of Langhorne. It is bounded by Pennsylvania Route 632 to the north, and Route 432 to the east. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for Core Creek is warm water fishes in the upper part of the stream, warm water fishes in the lower part of the stream, and migratory fishes.

The primary land use in the Core Creek watershed is agriculture (57%), with areas adjacent to the stream used for cropland and pasture. The majority of the streams have no protected riparian zone. Based on the aquatic survey conducted in the area, it was found that sediment being deposited in large quantities on the streambed was degrading the habitat of bottom-dwelling macroinvertebrates.

Figure Q1. Core Creek watershed.

242

The surficial geology of Core Creek watershed consists of a sandstone formation. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.38). Watershed characteristics are summarized in Table Q1.

Table Q1. Physical Characteristic Comparisons between the Core Creek and Reference Watersheds

Attribute Core Creek Watershed Reference Watershed Physiographic Province Piedmont Piedmont Area (square miles) 23 21 Predominant Land Uses Agriculture (57%) Agriculture (50%) Predominant Geology Sandstone (90%) Shale (90%) Shale (10%) Carbonate (10%) Soils - Dominant HSG C C - K Factor 0.38 0.37 20-Year Average Rainfall (in) 40.5 40.6 20-Year Average Runoff (in) 4.1 4.3

Q1.2 Surface Water Quality

A Total Maximum Daily Load or TMDL was developed for the Core Creek watershed to address the impairments noted on the Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). The 2001 survey found that that the stream was impaired. As a consequence, Pennsylvania listed Core Creek on the 2002 Section 303(d) List of Impaired Waters. The 2002 303 (d) List reported 15.8 miles of Core Creek (Stream Segment ID# 980602- 0954-GLW) to be impaired by siltation from agricultural activities in the watershed.

Q2.0 APPROACH TO TMDL DEVELOPMENT

Q2.1 TMDL Endpoints

The TMDL described herein addresses sediment impairments. Because neither Pennsylvania nor EPA has water quality criteria for sediments, we had to develop a method to determine water quality objectives for this parameter that would result in the impaired stream segments attaining their designated uses. The method employed for this TMDL is termed the “reference watershed approach.”

243

matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of the process is to reduce the loading rate of sediments in the impaired stream segment to a level equivalent to or slightly lower than the loading rate in the non-impaired, reference stream segment. The underlying assumption is that this load reduction will allow the biological community to return to the impaired stream segments.

Q2.2 Selection of the Reference Watershed

In general, three factors should be considered when selecting a suitable reference watershed. The first factor is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to attain water quality standards. The second factor is to find a watershed that closely resembles the Core Creek watershed in physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS data layers including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The reference used for the Core Creek watershed is the Mill Creek watershed located in the north-central part of Neshaminy Creek (not to be confused with the other Mill Creek located farther upstream as discussed in sections O and P). Both watersheds are located in the same physiographic Province and State Water Plan, and are tributaries of Neshaminy Creek. Table Q1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure Q2 shows the reference watershed boundary and its location in Bucks County. An analysis of the landuse/cover layer revealed that land cover/use distributions in both watersheds are similar. The surficial geology of the Core Creek and reference watersheds is somewhat different, but it is not expected that this would significantly affect sediment loads in either case.

Q2.3 Siltation Due to Agricultural Sources

The 2001 survey showed that siltation originating from agricultural activities in the watershed was the cause of impairment of Core Creek stream segments. Sediments deposited in large quantities on the streambed were degrading the habitat of bottom-dwelling macroinvertebrates. This TMDL addresses sediments from agricultural activities and from stream bank erosion. Stream bank erosion occurred as a result of excess flow in the watershed after intense storm events.

244

Figure Q2. Reference watershed location.

Q2.4 Watershed Assessment and Modeling

The AVGWLF model was run for both the Core Creek and reference watersheds to establish existing loading conditions under existing land cover use conditions in each watershed.

Adjustments to Specific GWLF-related parameters in the Reference Watershed:

-reset “C” factor to 0.18 from 0.21 for Cropland to account for use of more continuous cover crop.

-reset “P” factor to 0.35 from 0.52 for Cropland land use to account for use of riparian forest and grasses along streams, strip cropping, and buffer strips.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the Core Creek watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediment for the impaired and reference watersheds were calculated as shown in Tables Q2 and Q3, respectively. Table Q4 presents an explanation of the header information contained in Tables Q2 and Q3. Modeling output for the Core Creek and reference watersheds is presented in Appendix F.

245

Table Q2. Loading Values for Core Creek Watershed

Sediment Load Unit Area Sediment Load Land Use Category Area (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 956 28,079 29.37 Cropland 2,415 1,144,790 474.03 Coniferous Forest 30 22 0.73 Mixed Forest 437 728 1.67 Deciduous Forest 1,022 2,230 2.18 Lo Intensity Developed 825 24,106 29.22 Hi Intensity Developed 205 4,503 21.97 Stream Bank 571,523 Groundwater Point Source Septic Systems Total 5,889 1,775,981 301.58

Table Q3. Loading Values for the Reference Watershed

Sediment Load Unit Area Sediment Load Land Use Category Area (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 464 18,168 39.16 Cropland 2,202 729,669 331.37 Coniferous Forest 12 110 9.17 Mixed Forest 336 1,015 3.02 Deciduous Forest 2,057 10,574 5.14 Transition 2 1,766 883 Lo Intensity Developed 136 4,790 35.22 Hi Intensity Developed 18 287 15.94 Stream Bank 557,616 Groundwater Point Source Septic Systems Total 5,287 1,323,995 250.42

246

Table Q4. Header Information for Tables Q2 and Q3.

Q3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire Core Creek watershed. The watershed was too small to subdivide it into meaningful sub-watersheds. Therefore, sub-watershed load allocations were not performed.

The load reduction calculations for the Core Creek watershed are based on the current loading rates for sediment in the reference watershed. Based on the biological assessment, it was determined that the reference watershed was attaining its designated uses. The sediment loading rate was computed for the reference watershed using the AVGWLF model. This loading rate was then used as the basis for establishing the TMDL for the Core Creek watershed.

The equations defining TMDLs for sediments are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Core Creek to meet water quality goals. Therefore, it is the load that originates from agricultural sources that have contributed to water quality problems encountered in the watershed. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

Q3.1 TMDL Total Load

The first step was to determine the TMDL total target load for Core Creek, the impaired watershed. This value was obtained by multiplying the sediment unit area loading rate in the

247

reference watershed by the total watershed area of the Core Creek watershed. This information is presented in Table Q5.

Table Q5. TMDL Total Load Computation

Unit Area Loading Rate in Reference Watershed Total Watershed Area for TMDL Total Load Type of Pollutant (lbs/acre/yr) Core Creek (acres) (lbs/yr) Sediment 250.42 5,889 1,474,723

Q3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

Sediment - 1,474,723 lbs/yr x 0.1 = 147,472 lbs/yr (4)

Q3.3 Load Allocation

The load allocation (LA), consisting of all nonpoint source loads in the watershed, was computed by subtracting the margin of safety and the waste load allocation (WLA) from the TMDL total load. (Notice that sediments do not have a waste load allocation in this case).

LA (Sediments) 1,474,723 lbs/yr – 147,472 lbs/yr = 1,327,251 lbs/yr (5)

Q3.4 Adjusted Load Allocation

The adjusted load allocation (ALA) is the actual load allocation for sources that will need reductions. It is computed by subtracting loads from non-point sources that are not considered in the reduction scenario (LNR). These are loads from all non-point sources in Table Q2 except those from agricultural land uses (Hay/Pasture, Row Crops), land development, and stream bank erosion. Therefore, using data in Table Q2,

LNR (Sediments) = 22 lbs/yr + 728 lbs/yr 2230 lb/yr + 24,106 lb/yr + 4,503lbs/yr + 571,523 lbs/yr= 603,112 lbs/yr (6)

ALA (Sediments) = = 1,327,251 lbs/yr – 603,112 lbs/yr = 724,139 lbs/yr (7)

Table Q6 below presents TMDL results for the Core Creek watershed.

248

Table Q6. Summary of TMDL for Core Creek (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediment 1,474,723 147,472 - 1,327,251 603,112 724,139

The ALA computed above is the portion of the load that is available to allocate among contributing sources (Hay/Pasture, Cropland) as described in the next step. Not all land use/source categories were included in the allocation because they are difficult to control, or provide an insignificant portion of the total load (e.g., transitional land use). The following section shows the allocation process in detail for the entire watershed.

Q3.5 Load Reduction Procedures

7) Each land use/source load is compared with the total allocable load to determine if any contributor would exceed the allocable load by itself. The evaluation is carried out as if each source is the only contributor to the pollutant load to the receiving waterbody. If the contributor exceeds the allocable load, that contributor would be reduced to the allocable load. This is the baseline portion of EMPR.

8) After any necessary reductions have been made in the baseline the multiple analysis is run. The multiple analysis will sum all of the baseline loads and compare them to the total allocable load. If the allocable load is exceeded, an equal percent reduction will be made to all contributors’ baseline values. After any necessary reductions in the multiple analysis, the final reduction percentage for each contributor can be computed.

The load allocation and EMPR procedures were performed using an Excel Worksheet, and results are presented in Appendix G. Results of the load allocation by contributing sources are presented in Table Q7. Table Q8 provides load allocation by considering all land uses in the Core Creek watershed. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values. (Again, it is important to note that Core Creek only has one stream segment. Therefore, sub-watersheds could not be delineated. As a result, load allocation by sub-watersheds was not performed.)

249

Table Q7. Load Allocation by Each Contributing Source in the Core Creek Watershed.

Sediments Land Use/Source Loading Rate Average Load Average ALA Reduction Lbs/ac/yr Lbs/yr lbs/yr - % - Hay/Past 28,079 29.37 27,031 4 Cropland 1,144,790 474.03 697,108 39 Sub-total 1,172,869 347.93 724,139 38

Table Q8. Load Allocation by Each Land Use/Source

Sediment Unit Area Annual ALA (annual Source Area Loading Rate average load average) Reduction (acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) (%) Hay/Past 956 28,079 29.37 27,031 4 Cropland 2,415 1,144,790 474.03 697,108 39 Coniferous 30 22 0.73 22 0 Mixed For 437 728 1.67 728 0 Transition 1,022 2,230 2.18 2,230 0 Lo Int Dev 825 24,106 29.22 24,106 28 Hi Int Dev 205 4,503 21.97 4,503 28 Stream Bank 571,523 571,523 28 Groundwater Point Source Septic Systems Total 5,889 1,775,981 301.58 1,327,251 26

The total allowable sediment load to Core Creek when all land use/cover sources are considered is 1,801,600 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from 1,775,981 pounds per year by 26%.

Q4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment to a waterbody and the resulting impact on beneficial uses, establishing this TMDL using average annual conditions is protective of the waterbody.

250

Q5.0 CONSIDERATION OF SEASONAL VARIATIONS

Q6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

The pollutant reductions in the TMDL are allocated entirely to agricultural activities in the watershed. Implementation of best management practices (BMPs) in the affected areas should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the planting of riparian buffer zones, contour strips, and cover crops. These BMPs range in efficiency from 20% to 70% for sediment reduction. Implementation of BMPs aimed at sediment reduction will also assist in the reduction of phosphorus. Other possibilities for attaining the desired reductions in sediment include streambank stabilization and fencing. Further field verification will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet the sediment reductions outlined in this section.

Q7.0 PUBLIC PARTICIPATION

Notice of the draft TMDL will be published in the PA Bulletin and local newspapers with a 60 day comment period provided. A public meeting with watershed residents will be held to discuss the TMDL. Notice of final TMDL approval will be posted on the Department website.

251

R. Total Maximum Daily Loads (TMDLs) Development Plan for Neshaminy Creek South #1 Watershed

252

Table of Contents Page

Executive Summary …………………………………………………………………… 255

R1.0 Introduction ……………………….……………………………………….……. 256 R1.1 Watershed Description …………………………………………….….. 257 R1.2 Surface Water Quality ………………………………………………... 257 R2.0 Approach to TMDL development ……………………………………………… 257 R2.1 TMDL Endpoints…………………………………………………………………. 257 R2.3 Selection of the Reference Watershed…………………………………………….. 258 R2.4 Siltation Due to Development……………………………………………………… 258 R2.3 Watershed Assessment and Modeling…………………………………………….. 258 R3.0 Load Allocation Procedure for Sediment TMDL ………………..………………… 261 R3.1 Sediment TMDL Total Load ………………………………………..….. 262 R3.2 Margin of Safety ……………………………………………………….. 262 R3.3. Load Allocation …………………………………………………………. 262 R3.4. Adjusted Load Allocation ……………………………………………….. 262 R3.5. Load Reduction Procedures ……………………………………………… 263 R4.0 Consideration of Critical Conditions …………………………………………. 264 R5.0 Consideration of Seasonal Variations …………………………………..…….. 265 R6.0 Reasonable Assurance of Implementation …………………………………… 265 R7.0 Public Participation ………………………………………………….………. 265

253

List of Tables Page

R1. Physical Characteristic Comparisons Between Neshaminy Creek South #1 and Reference Watersheds…...……………………………. 256 R2. Loading Values for Neshaminy Creek South #1 Watershed….…….………... 260 R3. Loading Values for the Reference Watershed…………….…………. ………. 260 R4. Header Information for Tables R2 and R3………………………………..….. 261 R5. TMDL Total Load Computation…………………………………………… 262 R6. Summary of TMDLs for Neshaminy Creek South #1 Watershed…………….. 263 R7. Load Allocation for each Contributing Source in Neshaminy Creek #1……… 263 R8. Load Allocation for each Land Use/Source………………………………….. 264

List of Figures Page

R1. Neshaminy Creek South #1 Watershed …………………………………….… 259 R2 Reference Watershed ………………………………………………………… 262

254

EXECUTIVE SUMMARY

The Neshaminy Creek South #1 watershed is approximately 19 square miles in size. It is in Bucks County, and includes a portion of the main stem of Neshaminy Creek in the lower part of the Nsehaminy Creek watershed, as well as several smaller tributaries that flow into it. The protected uses of streams in the watershed are water supply, recreation, and aquatic life. Their aquatic uses include warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to 7.6 miles of streams in this watershed (Stream Segment ID#s 20010525-1250-GLW). TMDLs were developed to address the impairments noted on Pennsylvania’s 2002 Clean Water act Section 303(d) List. Pennsylvania does not currently have water quality criteria for sediment. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of the watershed. Based upon comparison to a similar, non- impaired watershed, it was estimated that the amount of sediment loading to the watershed that will meet the water quality objectives for streams therein is 253,654 pounds per year. The TMDL for the Neshaminy Creek South #1 watershed are allocated as shown in the table below.

Summary of TMDLs for Neshaminy Creek South #1 (lbs/yr) Pollutant TMDL MOS WLA LA LNR ALA Sediments 326,505 32,650 - 293,855 40,201 253,654

The TMDL for sediment is allocated to non-point source loads from developed land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The Waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediment in this case. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 7.6 miles. The TMDL establishes a reduction for total sediment loading of 44% from the current annual loading of 526,609 pounds.

255

R1.0 INTRODUCTION

R1.1 Watershed Description

The watershed contains part of the main stem of Neshaminy Creek (Stream Segment ID# 467), as well as several small tributaries that flow into. The watershed is located in the Piedmont Physiographic Province, is situated in Bucks County, and is approximately 19 square miles in size. The watershed covers the towns of South Hampton, Langhorne, and Middletown. It is bounded by Pennsylvania Route 213 to the north and Route 1 to the south. Figure R1 shows the watershed boundary and its location. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the Neshaminy Creek Tributary #3 watershed is warm water fishes and migratory fishes. The primary land use in the watershed is development (58%). As a result of aquatic surveys completed in the area, it was found that sediment deposited in large quantities on the streambed was degrading the habitat of bottom-dwelling macroinvertebrates.

Figure R1. Neshaminy Creek South #1 watershed.

The surficial geology of the watershed consists of sandstone and metamorphic/gneiss formations. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a high average K factor (0.32). Watershed characteristics are summarized in Table R1.

256

Table R1. Physical Characteristic Comparisons Between Neshaminy Creek South #1 and Reference Watersheds

Attribute Neshaminy Creek South #1 Reference Watershed

Physiographic Province Piedmont Piedmont Area (square miles) 23 21 Predominant Land Uses Development (57%) Development (50%) Predominant Geology Sandstone (50%) Metamorphic/Gneiss(95%) Metamorphic/Gneiss (45%) Soils - Dominant HSG C C - K Factor 0.32 0.34 20-Year Average Rainfall (in) 41.5 41.5 20-Year Average Runoff (in) 3.3 3.5

R1.2 Surface Water Quality

Total Maximum Daily Loads or TMDLs were developed for the Neshaminy Creek South #1 watershed to address the impairments noted on the Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). The 2001 survey found that that the stream was impaired. As a consequence, Pennsylvania listed stream segments in this watershed on the 2002 Section 303(d) List of Impaired Waters. In particular, the 2002 303 (d) List reported 7.6 stream miles (Stream Segment ID# 20010525-1250-GLW) to be impaired by siltation from development in the watershed.

R2.0 APPROACH TO TMDL DEVELOPMENT

R2.1 TMDL Endpoints

The TMDL discussed herein addresses sediment loads in the watershed. Because neither Pennsylvania nor EPA have water quality criteria for sediment, we had to develop a method to determine water quality objectives for these parameters that would result in the impaired stream segments attaining their designated uses. The method employed for this TMDL is termed the “reference watershed approach.”

With the reference watershed approach, two watersheds are compared, with one attaining its uses and one that is impaired based on biological assessment. Both watersheds must have similar land use/cover distributions. Other features such as base geologic formation should be matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of the process is to reduce the loading rate of sediments in the impaired stream segment to a level equivalent to or slightly lower than the loading rate in the non-impaired, reference stream segment. The underlying assumption is that this load reduction will allow the biological community to return to the impaired stream segments.

257

R2.2 Selection of the Reference Watershed

In general, three criteria should be considered when selecting a suitable reference watershed. The first criterion is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to attain water quality standards. The second criterion is to find a watershed that closely resembles impaired watershed in physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS coverages including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The watershed used as a reference for the Neshaminy Creek South #1 watershed is the Mill Creek watershed located in the lower part of the larger Neshaminy Creek watershed (not to be confused with the Mill Creek watershed located farther upstream discussed earlier in sections O and P) . Both watersheds are located in the same physiographic province and State Water Plan, and are tributaries of Neshaminy Creek. Table R1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure R2 shows the reference watershed boundary and its location in Bucks County. An analysis of the MRLC land use/cover GIS layer revealed that land cover/use distributions in both watersheds are similar. The major surficial geology in the two watersheds is somewhat different, but it is believed that this likely does not significantly affect sediment loads in either case.

R2.3 Siltation Due to Development

The 2001 survey showed that siltation originating from development was the cause of impairment of stream segments in this watershed. During the stream assessment, it was found that sediment deposited in large quantities on the streambed was degrading the habitat of bottom- dwelling macroinvertebrates. Stream bank erosion also occurred as a result of excess flow in the watershed due to a substantially high percentage of impervious areas (the watershed is 57% developed). Consequently, the TMDL discussed herein addresses sediment loads from development and from stream bank erosion.

R2.4 Watershed Assessment and Modeling

The AVGWLF model was run for both the Neshaminy Creek South #1 watershed and the reference watershed to establish existing loading conditions under existing land cover use conditions in each case. Prior to running the model, historical stream water quality data for the period 4/89 to 3/96 were first used to calibrate various key parameters within the GWLF model. Such data sets are typically not available in AVGWLF-based TMDL assessments done elsewhere in Pennsylvania. In this case, however, it was felt that model calibration would provide for better simulation of localized watershed processes and conditions. A description of the calibration procedure used can be found in section A2.3 of this document.

258

Figure R2. Reference watershed location.

Using the refined parameter estimates based on the calibration results, AVGWLF was re-run for the Neshaminy Creek South #1 watershed. Based on the use of 20 years of historical weather data, the mean annual loads for sediment, N and P for the impaired and reference watersheds were calculated as shown in Tables R2 and R3, respectively. Table R4 presents an explanation of the header information contained in Tables R2 and R3. Modeling output for Neshaminy Creek South #1 watershed and the reference watershed is presented in Appendix F.

259

Table R2. Loading Values for Neshaminy Creek South #1 Watershed

Land Use Category Area Sediment Load Unit Area Sediment Load (lbs/acre/yr) (acres) (lbs/year) Hay/Pasture 94 2,265 25.00 Cropland 180 28,342 157.46 Coniferous Forest 81 221 2.73 Mixed Forest 252 534 2.12 Deciduous Forest 1,425 5009 3.51 Transition 10 3,830 38.30 Lo Intensity Develop 2,240 112,320 50.14 Hi Intensity Develop 533 14,235 26.71 Stream Bank 359,853 Groundwater Point Source Septic Systems Total 4,815 526,609 109.37

Table R3. Loading Values for the Reference Watershed

Land Use Category Area Sediment Load Unit Area Sediment Load (lbs/acre/yr) (acres) (lbs/year) Hay/Pasture 254 7,698 30.31 Cropland 272 24,696 90.79 Coniferous Forest 153 276 1.81 Mixed Forest 516 1,344 2.61 Deciduous Forest 1,489 5,543 3.72 Unpaved Roads 2 1,897 948.43 Lo Intensity Dev 2,319 68,545 29.56 Hi Intensity Dev 284 3,554 12.51 Stream Bank 245,120 Groundwater Point Source Septic Systems Total 5,289 358,673 67.81

260

Table R4. Header Information for Tables M2 and M3.

R3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

The load allocation and reduction procedures were applied to the entire Neshaminy Creek South #1 watershed. The load reduction calculations in the watershed are based on the current loading rates for sediments in the reference watershed. As discussed earlier, the Neshaminy Creek South #1 watershed also contains a portion of the main stem of Neshaminy Creek that is listed as being impaired due to municipal point sources (Stream Segment ID #467). This particular impairment is not addressed here, however, since it has been discussed in an earlier section of this document (see section B). Based on biological assessment, it was determined that the reference watershed was attaining its designated uses. Sediment loading rates were computed for the reference watershed using the AVGWLF model. These loading rates were then used as the basis for establishing the TMDL for the Neshaminy Creek South #1 watershed.

The equations defining TMDL for sediment are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is that portion of the loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Neshaminy Creek South #1 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

261

R3.1 TMDL Total Load

The first step is to determine the TMDL total target load for the Neshaminy Creek South #1 watershed. This value was obtained by multiplying the pollutant unit loading rate in the reference watershed by the total watershed area of the Neshaminy Creek South #1 watershed. This information is presented in Table R5.

Table R5. TMDL Total Load Computation

Unit Area Loading Rate Total Watershed Area of in Reference Watershed Neshaminy South #1 TMDL Total Load Type of Pollutant (lbs/acre/yr) (acres) (lbs/yr) Sediment 67.81 4,815 326,505

R3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

Sediment - 326,505 lbs/yr x 0.1 = 32,650 lbs/yr (4)

R3.3 Load Allocation

LA (Sediments) 326,505 lbs/yr – 32,650 lbs/yr = 293,855 lbs/yr (5)

R3.4 Adjusted Load Allocation

LNR (Sediments) = 2,265 lbs/yr + 28,342 lbs/yr 221 lb/yr + 534 lb/yr + 5,009 lbs/yr + 3,830 lbs/yr= 40,201 lbs/yr (6)

ALA (Sediments) = = 293,855 lbs/yr – 40,201 lbs/yr = 253,654 lbs/yr. (7)

262

Table R6 below presents the sediment TMDL for the Neshaminy Creek South #1 watershed.

Table R6. Summary of TMDL for Neshaminy Creek South #1 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediment 326,505 32,650 - 293,855 40,201 253,654

The ALA computed in this case is that portion of the load that is available to allocate among contributing sources as described in the next step. The following section shows the allocation process in detail for the entire watershed.

R3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in Neshaminy Creek South #1 was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table R7. As shown in this table, sediment loads from the Low Intensity Developed and High Intensity Developed areas should be reduced to 7,606 and 58,340 pounds, respectively, in order for this watershed to attain its specific uses. Similarly, stream bank erosion should be reduced to 187,708 pounds per year. Again, it should be noted that the Neshaminy Creek South #1 watershed only has one stream segment. Therefore, subwatersheds could not be delineated. As a result load allocation by subwaterheds was not performed.

Table R7. Load Allocation for Each Contributing Source in the Neshaminy Creek South #1 Watershed.

Pollutant Source Current Load ALA Reductio n Lbs/year % Lbs/year -%- Sediment - Low Intensity Developed 112,320 23 58,340 48 - High Intensity Developed 14,235 3 7,606 ..47 - Stream bank erosion 359,853 74 187,708 48 TOTAL 486,408 100 253,654 48

Results of the load allocation by contributing land use sources are presented in Table R7. Table R8 provides load allocation by considering all land uses in the Neshaminy Creek South #1

263

watershed. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table R8. Load Allocation by Each Land Use/Source

Sediment Unit Area Annual average ALA (annual Reduction Source Area Loading Rate load average)

(acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) ( % ) Hay/Pasture 94 25.00 2,265 2,265 0 Cropland 180 157.4628,342 28,342 0 Coniferous Forest 81 2.73 221 221 0 Mixed Forest 252 2.12 534 534 0 Deciduous Forest 1,425 3.51 5009 5009 Transition 10 38.303,830 3,830 0 Low Intensity Dev 2,240 50.14 112,320 58,340 48 High Intensity Dev 533 26.71 14,235 7,606 47 Stream Bank 359,853 187,708 48 Groundwater Point Source Septic Systems Total 4,815 109.37 526,609 293,855 44

The total allowable sediment load in Neshaminy South #1 when all land use/cover sources are considered is 293,855 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from 526,609 pounds per year by 44%.

R4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing these TMDLs using average annual conditions is protective of the waterbody.

264

R5.0 CONSIDERATION OF SEASONAL VARIATIONS

R6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

The pollutant reductions in the TMDL are allocated entirely to developed land and stream bank erosion in the watershed. Implementation of best management practices (BMPs) in the affected areas should achieve the loading reduction goals established in the TMDL. Substantial reductions in the amount of sediment reaching the streams can be made through the planting of riparian buffer zones, contour strips, and cover crops. These BMPs range in efficiency from 20% to 70% for sediment reduction. Other possibilities for attaining the desired reductions in sediment include streambank stabilization and fencing. Further field verification will be performed in order to assess both the extent of existing BMPs, and to determine the most cost-effective and environmentally protective combination of BMPs required to meet the sediment reductions outlined in this report.

R7.0 PUBLIC PARTICIPATION

265

S. Total Maximum Daily Loads (TMDL) Development Plan for Neshaminy Creek South #2 Watershed

266

Segments in the Neshaminy Creek South #2 watershed (Stream Segment ID# 980713-1351- GLW) were listed as being impaired by water/flow variability caused by urban runoff/storm sewers. A TMDL for this impairment was not developed because neither the U.S. Environmental Protection Agency (EPA) nor PaDEP currently have water quality criteria for this impairment. Furthermore, quantitative measures for water flow variability or alterations as “impairments” are not currently available. However, it is assumed for these segments that addressing sediment loads through the use of urban BMPs will at the same time reduce water flow variability or alterations within the watershed. As discussed previously, all municipalities within the Neshaminy Creek watershed will be affected by PaDEP’s new stormwater management policy (MS4), a copy of which has been included in Appendix E.

Figure S1. Neshaminy Creek South #2 watershed.

267

T. Total Maximum Daily Load (TMDL) Development Plan for Neshaminy Creek South #3

268

Table of Contents Page

Executive Summary …………………………………………...…………………… 271

T1.0 Introduction ……………………….……………………………………….……. 272 T1.1 Watershed Description ………………………………….………….…. 272 T1.2 Surface Water Quality …………………………………….…………... 273 T2.0 Approach to TMDL Development ………………………………...……….… 273 T2.1 TMDL Endpoints………………………………………………………………… 273 T2.2 Selection of the Reference Watershed……………………………………………. 274 T2.3 Siltation Due to Development ……………………………………………………. 275 T2.4 Watershed Assessment and Modeling …………………………………………… 275 T3.0 Load Allocation Procedure for Sediment TMDLs ………………..…………… 276 T3.1 Sediment TMDL Total Load ………………………………………..….. 277 T3.2 Margin of Safety ………………………………………...………..…. 277 T3.3. Load Allocation …………………………………………..…………. 277 T3.4. Adjusted Load Allocation …………………………………..……….. 278 T3.5. Load Reduction Procedures …………………………………..…..…. 278 T4.0 Consideration of Critical Conditions ………………………….……………… 280 T5.0 Consideration of Seasonal Variations …………………………………..…….. 280 T6.0 Reasonable Assurance of Implementation …………………………………… 280 T7.0 Public Participation …………………………………………………………. 280

269

List of Tables Page

T1. Physical Characteristic Comparisons Between Neshaminy Creek South #3 and Reference Watershed ………………………………… 273 T2. Loading Values for Neshaminy Creek South #3……………………………. 275 T3. Loading Values for the Reference Watershed…………….…………. ……… 276 T4. Header Information for Tables T2 and T3………………………………..…. 276 T5. TMDL Total Load Computation…………………………………………….. 277 T6. Summary of TMDLs for Neshaminy Creek South #3 …………………....… 278 T7. Load Allocation for each Contributing Source in Neshaminy Creek South #3 279 T8. Load Allocation by each Land Use/Source…………………………………. 279

List of Figures Page

T1. Neshaminy Creek South #3 Watershed ………………………………….…. 272 T2 Reference Watershed …………………………………………………… 274

270

EXECUTIVE SUMMARY

The Neshaminy Creek South #3 watershed is approximately 10 square miles in size. It is situated in Bucks County and is comprised of a portion of the lower part of the main stem of Neshaminy Creek and several small tributaries that flow into it. The protected uses of the watershed are water supply, recreation, and aquatic life. Its aquatic use is warm water fishes and migratory fishes.

Total Maximum Daily Loads (TMDLs) apply to 5.4 miles of stream (Stream Segment ID# 20010525-1330-GLW). This stream segment is actually comprised of several smaller segments (i.e., tributaries) located in the lower Neshaminy Creek watershed. A TMDL was developed to address the impairments noted on Pennsylvania’s 2002 Clean Water act Section 303(d) List. Pennsylvania does not currently have water quality criteria for sediments. For this reason, we developed a reference watershed approach to identify the TMDL endpoints or water quality objectives for sediments in the impaired segments of the Neshaminy Creek South #3 watershed. Based upon comparison to a similar, non-impaired watershed, it was estimated that the sediment loading that will meet the water quality objectives for this watershed is 253,654 pounds per year. The TMDL for Neshaminy Creek South #3 is allocated as shown in the table below.

Summary of TMDL for Neshaminy Creek South #3 (lbs/yr) Pollutant TMDL MOS WLA LA LNR ALA Sediments 326,505 32,650 - 293,855 40,201 253,654

The TMDL for sediment is allocated to non-point source loads from developed land and stream bank erosion, with 10% of the TMDL total load reserved as a margin of safety (MOS). The waste load allocation (WLA) is that portion of the total load that is assigned to point sources, which was zero for sediments. The allowable loading, or adjusted loading allocation (ALA), is that load attributed to transitional land use and stream bank erosion, and is computed by subtracting loads that do not need to be reduced (LNR) from the TMDL total values. The sediment TMDL covers a total of 5.4 miles. The TMDL establishes a reduction for total sediment loading of 58% from the current annual loading of 218,488 pounds.

271

T1.0 INTRODUCTION

T1.1 Watershed Description

The watershed contains part of the lower main stem of Neshaminy Creek (Stream Segment ID# 467) as well as several small tributaries that flow into it (Stream Segment ID 20010525- 1330-GLW). The watershed is located in the Piedmont Physiographic Province and is situated in Bucks County. It covers an area of approximately 19 square miles. The watershed is south of the towns of South Hampton, Langhorne, and Midletown. It is bounded by Pennsylvania Route 213 to the west and Route 1 to the east. Figure T1 shows the watershed boundary and its location. The designated uses of the watershed include water supply, recreation and aquatic life. As listed in the Title 25 PA Code Department of Environmental Protection Chapter 93, Section 93.o (Commonwealth of PA, 1999), the designated aquatic life use for the Neshaminy Creek South #3 watershed is warm water fishes and migratory fishes. The primary land use in the is high and low intensity development (63%). During the stream assessment, it was found that sediment deposited in large quantities on the streambed was degrading the habitat of bottom-dwelling macroinvertebrates.

Figure T1. Neshaminy Creek South #3 watershed.

The surficial geology of the watershed consists of unconsolidated and metamorphic/gneiss formations. The bedrock geology primarily affects surface runoff and background nutrient loads through its influences on soils and landscape as well as fracture density and directional permeability. Soils are mostly sandy and very erodible, as indicated by a predominantly high average K factor (0.32). Watershed characteristics are summarized in Table T1.

272

Table T1. Physical Characteristic Comparisons Between Neshaminy Creek South #3 and Reference Watershed

Neshaminy Creek Attribute South #3 Reference Watershed Physiographic Province Piedmont Piedmont Area (square miles) 10 11 Predominant Land Uses Development (63%) Development (57%) Predominant Geology Unconsolidated (50%) Sandstone (95%) Metamorphic/Gneiss (50%) Soils - Dominant HSG C C - Dominant K Factor 0.32 0.32 20-Year Average Rainfall (in) 41.4 41.4 20-Year Average Runoff (in) 4.4 4.3

T1.2 Surface Water Quality

A Total Maximum Daily Load or TMDL was developed for the Neshaminy Creek South #3 watershed to address the impairments noted on the Pennsylvania’s 2002 Clean Water Act Section 303(d) List (see Table A1 in section A1.0). The 2001 survey found that that the tributaries were impaired by siltation from development in the watershed. As a consequence, Pennsylvania listed these stream segments on the 2002 Section 303(d) List of Impaired Waters. The List reported 5.4 miles of stream in the Neshaminy Creek South #5 watershed (Stream Segment ID# 20010525-1330-GLW) to be impaired

T2.0 APPROACH TO TMDL DEVELOPMENT

T2.1 TMDL Endpoints

The TMDL discussed herein addresses sediment. Because neither Pennsylvania nor EPA has water quality criteria for sediment, we had to develop a method to determine water quality objectives for these parameters that would result in the impaired stream segments attaining their designated uses. The method employed for this TMDL is termed the “reference watershed approach.”

With the reference watershed approach, two watersheds are compared, with one attaining its uses and one that is impaired based on biological assessment. Both watersheds must have similar land use/cover distributions. Other features such as base geologic formation should be matched to the greatest extent possible; however, most variations can be adjusted in the model. The objective of the process is to reduce the loading rate for sediment in the impaired stream segment to a level equivalent to or slightly lower than the loading rate in the non-impaired, reference stream segment. The underlying assumption is that this load reduction will allow the biological community to return to the impaired stream segments.

273

T2.2 Selection of the Reference Watershed

In general, three factors should be considered when selecting a suitable reference watershed. The first factor is to use a watershed that has been assessed by the Department using the Unassessed Waters Protocol and has been determined to be attaining designated uses. The second factor is to find a watershed that closely resembles the Neshaminy Creek South #3 watershed in terms of physical properties such as land cover/land use, physiographic province, and geology. Finally, the size of the reference watershed should be within 20-30% of the impaired watershed area. The search for a reference watershed that would satisfy the above characteristics was done by means of a desktop screening using several GIS data layers including the Multi-Resolution Land Characteristics (MRLC) Landsat-derived land cover/use grid, the Pennsylvania’s 305(b) assessed streams database, and geologic rock types.

The reference watershed used in this case is a part of the Mill Creek watershed located at the lower end of Neshaminy Creek. Both watersheds are located in the same physiographic province and State Water Plan, and include tributaries of Neshaminy Creek. Table T1 compares the two watersheds in terms of their size, location, and other physical characteristics. All reference watershed stream segments have been assessed and were found to be unimpaired. Figure T2 shows the reference watershed and its location within the larger Neshaminy Creek watershed.

An analysis of the land use/cover grid revealed that land cover/use distributions in both watersheds are similar. The surficial geology in the two watersheds is different, but it is not expected that this would significantly affect the sediment loads in either case.

Figure T2. Reference watershed location.

274

T2.3 Siltation Due to Development

The 2001 survey showed that siltation originating from development in the watershed was the cause of impairment to Neshaminy Creek South #3 stream segments. Sediments deposited in large quantities on the streambed were found to be degrading the habitat of bottom-dwelling macroinvertebrates. This TMDL addresses sediment from development and from stream bank erosion. Stream bank erosion occurred as a result of excess flow in the watershed due to substantially high impervious areas. The watershed is 63% developed.

T2.4 Watershed Assessment and Modeling

The AVGWLF model was run for both the Neshaminy Creek South #3 and reference watersheds to establish existing loading conditions under existing land cover use conditions. Prior to running the model, historical stream water quality data for the period 4/89 to 3/96 were first used to calibrate various key parameters within the GWLF model. Such data sets are typically not available in AVGWLF-based TMDL assessments done elsewhere in Pennsylvania. In this case, however, it was felt that model calibration would provide for better simulation of localized watershed processes and conditions. A description of the calibration procedure used can be found in section A2.3 of this document.

Using the refined parameter estimates based on the calibration results, AVGWLF was run for both watersheds. Based on the use of 20 years of historical weather data, the mean annual loads for sediments for the impaired and reference watersheds were calculated as shown Tables T2 and T3, respectively. Table T4 presents an explanation of the header information contained in Tables T2 and T3. Modeling output for both watersheds is presented in Appendix F.

Table T2. Existing Loading Values for the Neshaminy Creek South #3 Watershed

Land Use Category Area Sediment Load Unit Area Sediment Load (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 62 368 5.95 Cropland 173 9,797 56.67 Coniferous Forest 15 0 0 Mixed Forest 126 92 0.73 Deciduous Forest 565 534 0.94 Transition 42 8,158 194.23 Low Intensity Developed 1,240 20,626 16.63 High Intensity Developed 420 5,433 12.94 Stream Bank 173,480 Groundwater Point Source Septic Systems Total 2.643 218,488 82.66

275

Table T3. Loading Values for the Reference Watershed

Land Use Category Area Sediment Load Unit Area Sediment Load (acres) (lbs/year) (lbs/acre/yr) Hay/Pasture 153 5,230 34.18 Cropland 128 10,589 82.73 Coniferous Forest 52 92 1.77 Mixed Forest 269 681 2.53 Deciduous Forest 563 2,007 3.56 Low Intensity Developed 1,440 32,726 22.73 High Intensity Developed 119 1,234 10.37 Stream Bank 53,775 Groundwater Point Source Septic Systems Total 2,724 106,334 39.04

Table T4. Header Information for Tables T2 and T3.

T3.0 LOAD ALLOCATION PROCEDURE FOR SEDIMENT TMDL

Load allocation and reduction procedures were applied to the entire Neshaminy Creek South #3 watershed. The load reduction calculations are based on the current loading rates for sediment in the reference watershed. (Note: This watershed also contains Stream Segment ID# 467, a portion of the main stem of Neshaminy Creek that is listed as being impaired due to municipal point sources. However, since municipal point sources are addressed in a different section of this document (see section B), they are not discussed here). Based on biological assessment, it was determined that the reference watershed was attaining its designated uses. Sediment loading rates were computed for the reference watershed using the AVGWLF model. These loading rates were then used as the basis for establishing the TMDL for the Neshaminy Creek South #3 watershed.

276

The equations defining TMDLs for sediment are as follows:

TMDL = MOS + LA + WLA (1)

LA = ALA - LNR (2)

TMDL is the TMDL total load. The LA (load allocation) is the portion of Equation (1) that is assigned to non-point sources. The MOS (margin of safety) is the portion of loading that is reserved to account for any uncertainty in the data and computational methodology used for the analysis. The WLA (Waste Load Allocation) is the portion of this equation that is assigned to point sources. The adjusted load allocation (ALA) is the load originating from sources (Equation 2) that needs to be reduced by the non-contributing sources (NLR) for Neshaminy Creek South #3 to meet water quality goals. Details of TMDL, MOS, LA, LNR, and ALA computations are presented below.

T3.1 TMDL Total Load

The first step is to determine the TMDL total target load for Neshaminy Creek South #3, the impaired watershed. This value was obtained by multiplying the sediment unit area loading rate in the reference watershed by the total watershed area of Neshaminy Creek South #3. This information is presented in Table T5.

Table T5. TMDL Total Load Computation

Unit Area Loading Rate Total Watershed Area of TMDL Total Load in Reference Watershed Neshaminy South #3 (lbs/yr) Type of Pollutant (lbs/acre/yr) (acres) Sediment 39.04 2,643 103,183

T3.2 Margin of Safety

The Margin of Safety (MOS) for this analysis is explicit. Ten percent of the TMDL was reserved as the MOS.

Sediment - 103,183 lbs/yr x 0.1 = 10,318 lbs/yr (3)

T3.3 Load Allocation

The load allocation (LA), consisting of all nonpoint source loads in the watershed, was computed by subtracting the margin of safety and the waste load allocation (WLA) from the TMDL total load. (In this case, there was no waste load allocation for sediment).

LA (Sediments) 103,183 lbs/yr – 10,318 lbs/yr = 92,865 lbs/yr (4)

277

T3.4 Adjusted Load Allocation

LNR (Sediments) = 368 lbs/yr + 9,797 lbs/yr 0 lb/yr + 92 lb/yr + 534 lbs/yr + 8,158 lbs/yr= 18,909 lbs/yr (5)

ALA (Sediments) = = 92,865 lbs/yr – 18,909 lbs/yr = 73,956 lbs/yr (6)

Table T6 below presents the TMDL for the Neshaminy Creek South #3 watershed.

Table T6. Summary of TMDL for Neshaminy Creek South #3 (lbs/yr)

Pollutant TMDL MOS WLA LA LNR ALA Sediment 103,183 10,318 - 92,865 18,909 73,956

The ALA computed above is the portion of the load that is available to allocate among contributing sources as described in the next step. The following section shows the allocation process in detail for the entire watershed.

T3.5 Load Reduction Procedures

The allocation of sediment among contributing land use/cover sources in Neshaminy Creek South #3 was not performed according to the to the Equal Marginal Percent Reduction (EMPR) method (as commonly used) because of differences existing between the types of pollutant sources. For example, sediment detachment and transport occurs across an area of land and therefore should be considered on an areal basis. Those from channel erosion are dealt on the basis of length of stream bank eroded (source) rather than per unit area. Consequently, the allocation to contributing sources was performed using the relative contribution of each land use to the total combined current load as indicated in Table T7. This means that sediment loads from low intensity developed land and high intensity developed land should be reduced to 7,396 and 2,218 pounds, respectively for Neshaminy Creek South #3 to attain its specific uses. Stream bank erosion should be reduced to 64,342 pounds per year. Again, in this case it is important to note that this watershed only has one stream segment. Therefore, sub-watersheds could not be delineated. As a result load allocation by sub-waterheds was not performed.

278

Table T7. Load Allocation for Each Contributing Source in Neshaminy Creek South #3

Pollutant Source Current Load ALA Reductio n lbs/year % lbs/year -%- Sediment - Low intensity developed 20,626 10 7,396 65 - High intensity developed 5,433 3 2,218 ..59 - Stream bank erosion 173,480 87 64,342 63 TOTAL 199,539 100 73,956 63

As discussed, results of the load allocation by contributing sources are presented in Table T7. Table T8, on the other hand, provides load allocation by considering all land uses in the Neshaminy Creek South #3 watershed. In this case, land uses/sources that were not part of the allocation are carried through at their existing loading values.

Table T8. Load Allocation by Each Land Use/Source

Sediment

Unit Area Annual average ALA (annual Source Area Loading Rate load average) Reduction (acres) (lbs/ac/yr) (lbs/yr) (lbs/yr) ( % ) Hay/Pasture 62 5.95 368 368 0 Cropland 173 56.67 9,797 9,797 0 Coniferous 15 0 0 0 0 Mixed Forest 126 0.73 92 92 0 Deciduous Forest 565 0.94 534 534 Transition 42 194.23 8,158 8,158 0 Low Intensity Dev 1,240 16.63 20,626 7,396 65 High Intensity Dev 420 12.94 5,433 2,218 59 Stream Bank 173,480 64,342 63 Groundwater Point Source Septic Systems Total 2.643 82.66 218,488 92,905 58

The total allowable sediment load in Neshaminy Creek South #3 when all land use/cover sources are considered is 92,905 pounds per year. In order for all stream segments to attain their specific uses, total sediment load should be reduced from 218,488 pounds per year by 58%.

279

T4.0 CONSIDERATION OF CRITICAL CONDITIONS

The AVGWLF model is a continuous simulation model, which uses daily time steps for weather data and water balance calculations. Monthly calculations are made for sediment and nutrient loads, based on the daily water balance accumulated to monthly values. Therefore, all flow conditions are taken into account for loading calculations. Because there is generally a significant lag time between the introduction of sediment and nutrients to a waterbody and the resulting impact on beneficial uses, establishing this TMDL using average annual conditions is protective of the waterbody.

T5.0 CONSIDERATION OF SEASONAL VARIATIONS

T6.0 REASONABLE ASSURANCE OF IMPLEMENTATION

T7.0 PUBLIC PARTICIPATION

280

LITERATURE CITED

Bagnold, R.A., 1980. An empirical correlation of bedload transport rates in flumes and natural rivers. Proc. R. Soc. Lond., Vol 372, pp. 453-473.

Carlson, R.E., 1977. A trophic state index for lakes. Limnol. Oceanogr, 22:361-369.

Commonwealth of Pennsylvania. 1999. Pennsylvania Code. Title 25 Environmental Protection. Department of Environmental Protection. Chapter 93. Water Quality Standards. Harrisburg, PA.

Dietrich, C.R., T.R. Green, and A.J. Jakeman, 1999. An analytical model for stream sediment transport: application to Murray and Murrumbidgee river reaches, Australia. Hydrological Processes, Vol. 13, pp. 763-776.

Evans, B.M., 2002. Development of an Automated GIS-Based Modeling Approach to Support Regional Watershed Assessments. Ph.D. Dissertation in Soil Science, The Penn State Univ, University Park, PA.

F.X. Bowne, Inc., 1995. Diagnostic Feasibility Study of Lake Galena. FXB Project No. PA1200-04, 204 pp.

Green, T.R., S.G. Beavis, C.R. Dietrich, and A.J. Jakeman, 1999. Relating stream-bank erosion to in-stream transport of suspended sediment. Hydrological Processes, Vol. 13, pp.777-787.

Haith, D.A. and L.L. Shoemaker, 1987. Generalized Watershed Loading Functions for Stream Flow Nutrients. Water Resources Bulletin, 23(3), pp. 471-478.

Horner, R.R., J. Guedry and M.H. Kortenhof, 1990. Improving the Cost Effectiveness of Highway Construction Site Erosion and Pollution Control. Washington State Transportation Center. Federal Highway Administration. Seattle, WA.

Krenkel, P.A. and V. Novotny, 1980. Water Quality Management. Academic Press, New York.

Langland, M.G., P.L.Lietman, and S. Hoffman. 1995. Synthesis of nutrient and sediment data for watersheds within the Chesapeake bay Drainage Basin. U.S. Geological Survey. Water- Resources Investig. Rept. 95-4233. Lemoyne, PA.

Nelson, E.J. and D.B. Booth, 2002. Sediment sources in an urbanizing, mixed land-use watershed. Journal of Hydology, 264, pp. 51-68.

Nizeyimana, E., B.M. Evans, M.C. Anderson, G.W. Petersen, D.R. DeWalle, W.E. Sharpe, J.M. Hamlett, and B.R. Swistock, 1997. Quantification of NPS Pollution Loads within Pennsylvania Watersheds. Environ. Resources Research Inst., Penn State Univ., ER9708, 66 pp.

281

Novotny, V. and H. Olem, 1994. Water Quality: Prevention, Identification, and Management of Diffuse Pollution. Van Nostrand Reinhold, New York.

Pennsylvania Fish and Boat Commission, 2000. The Basics of Water Pollution in Pennsylvania. 8pp.

Prosser, I., P. Rustomji, B. Young, C.Moran, and A. Hughes, 2001. Constructing River Basin Sediment Budgets for the National Land and Water resources Audit. CSIRO Technical Report No. 15, 34 pp.

Reese, S. and J. Lee, 1998. Summary of Groundwater Quality Monitoring Data (1985-1997) from Pennsylvania’s Ambient and Fixed Station Network (FSN) Monitoring Program. PA Dept. of Environmental Protection, 92 pp.

Ritter, W.F. and A. Shirmohammadi, 2001. Agricultural Nonpoint Source Pollution: Watershed Management and Hydrology. Lewis Publishers, 342 pp.

Samuels, W., 1998. Case Studies: Solving Watershed-Based Problems Through the Use of GIS, Internet and EPA National Data Bases. In: Watershed Management: Moving from Theory to Implementation, Water Environment Federation, Denver, CO, pp. 1175-1182.

Thomann, R.V. and J.A. Mueller, 1987. Principles of Surface Water Quality Modeling and Control. Harper & Row, New York.

U.S. EPA, 1982. Justification Report for Advanced Treatment Facilities Proposed for Chalfont- New Britain and Doylestown, PA. EPA Region III.

VanSickle, J. and R.L. Beschta, 1983. Supply-Based Models of Suspended Sediment Transport in Streams. Water Resources Research, Vol. 19, No. 3, pp. 768-778.

Vogelmann, J.E., T. Sohl, and S.M. Howard. 1998. Regional characterization of land cover using multiple sources of data. Photogrammetric Engineering and Remote Sensing 64: 45-57.

Walker, W.W., 1996. Simplified Procedures for Eutrophication Assessment and Prediction: User Manual. Instruction Report W-96-2, US Army Corps of Engineers, Wash. DC.

282