TMDL Case Study

1
TMDL Case Study

• Delaware - EPA Region 3
• TMDL for Nutrients and Dissolved Oxygen under
  Low-Flow Conditions
  
• in the Christina River Basin

2
Christina River Basin

• Study area
• basin covers 565 square miles
• spans 3 states (PA, DE, MD)
• 39 stream segments on 303(d) list
• Major watersheds
• Brandywine Creek
• Red Clay Creek
• White Clay Creek
• Christina River

3
Christina River Basin

• Agencies involved
• EPA Region 3
• Delaware River Basin Commission
• Pennsylvania DEP
• Delaware DNREC
• Maryland Dept. of Environment
• New Castle County WRA
• USGS
• Brandywine Valley Association

4
Christina River Basin - Problem Statement

• 24 stream segments included on Pennsylvanias
  1998 303(d) list
  
• 15 stream segments included on Delawares 1998
  303(d) list
  
• Impaired due to nutrients and dissolved oxygen
  not meeting water quality criteria
  
• Exceedances of dissolved oxygen criteria occur
  primarily during low-flow, summer conditions
  
• Most probable sources during low-flow periods are
  point source discharges

5
Impairment Analysis

• In-stream water quality data inventory
• Impairment confirmation
• Magnitude of impairment
• Frequency
• Seasonal and daily patterns
• Background conditions
• Identification of data sets to support modeling
• Additional sampling to support modeling

6
Christina River Basin TMDL

• Step 1 - TMDL under low-flow conditions
• focus on point sources
• critical summer conditions
• TMDL was issued January 19, 2001
• Step 2 - TMDL under variable-flow conditions
• considers both point and nonpoint sources
• considers both high and low-flow conditions
• TMDL due December 2004

7
Water Quality Model

• Point sources
• 122 NPDES facilities
• 0.0005 mgd to 7.0 mgd
• Nonpoint sources
• 39 subwatersheds
• Agriculture (31)
• Forest (28)
• Urban (34)
• Open land, water, other (7)
• 28 water withdrawal facilities
• Free flowing streams
• Tidal rivers and streams

8
Christina River Basin Modeling Effort
Watershed Loading Model
Monitoring Program
Receiving Water Model
TMDL Development
9
Christina River BasinMonitoring Program

• PA, DE, DRBC, and EPA reached agreement in 1993
  to initiate a cooperative and coordinated
  monitoring program for the Christina River Basin
  TMDLs
• Beginning in 1995, PDEP and DNREC sampled monthly
  or semi-monthly at 23 stations
  
• The USGS monitored flow and/or quality at 17
  stations
  
• In August 1997, an intensive survey was conducted
  by PDEP and DNREC at 12 stations to characterize
  daily variations in DO and nutrients
  
• In 1999, an effluent monitoring study of the 14
  largest WWTPs in the basin was conducted to
  determine organic carbon and CBOD
  characteristics
• Historical data were used from STORET, PCS, DMRs,
  and state archives

10
Location of STORET monitoring stations
11
Christina River BasinWater Quality Model

• OBJECTIVES
• Provide a calibrated and validated hydraulic and
  water quality model of Christina River basin
  representative of critical low-flow conditions
  
• Provide a model capable of addressing variable
  flow conditions for high-flow phase of the TMDL
  
• Provide training on the model use
• Develop TMDL allocation scenarios
• Scientific credibility documented in technical
  report

12
Christina River Basin Water Quality Model
13
EFDC Water Quality Model

• 16 stream reaches
• Christina River and West Branch Christina River
  (tidal and nontidal)
  
• Brandywine Creek (tidal and nontidal)
• East and West Branch Brandywine Creek
• Buck Run
• White Clay Creek and East Branch White Clay Creek
  (tidal and nontidal)
  
• Red Clay Creek and West Branch Red Clay Creek
• Little Mill Creek
• Mill Creek
• Pike Creek
• Muddy Run
• Burroughs Run
• Delaware River (tidal)
• Designed to be linked to watershed runoff model
  for seasonal analysis
  

14
Low-head Dam on Brandywine Creek
15
Flow Structures

• 32 flow structures included in the model
• 8 tidal inlets to connect peripheral streams to
  Christina River
  
• 24 overflow structures representing low-head
  dams, submerged weirs, bridge culverts, fall
  lines, stream confluences
  
• Rating curve representing a free overfall was
  used for the overfall structures

16
Water Quality Model Calibration

• MODELED PARAMETERS
• Chlorophyll-a
• Nitrogen (ammonia, nitrate, total)
• Phosphorus (orthophosphate, total)
• Organic carbon (dissolved, total)
• Total suspended solids
• Chloride concentrations
• Daily average dissolved oxygen
• Diel dissolved oxygen (daily minimum, maximum)

17
Water Quality Model Calibration

• Model calibration period May 1 to Sep 21, 1997
• Model validation period May 1 to Sep 21, 1995
• Nonpoint source loads from peripheral tributaries
  were computed using estimated low-flow rates and
  a characteristic concentration for each
  parameter
• Model-data statistics were computed and
  summarized in report
  

18
Calibration Results

• Results presented as longitudinal transect plots
  for the August period (low-flow condition)
  
• Results were also presented as time-series plots
  at a single monitoring station on the major
  tributaries
  
• Model statistics compare well with other similar
  studies

19
Low-flow TMDL Analysis

• Nonpoint source loads computed based on estimated
  low-flow rates
  
• Point sources set to their existing permit limits
  for both flow and concentration of CBOD, NH3-N,
  TP, and DO
  
• Stream conditions set to low-flow rates
• Stream temperature set to 75th percentile summer
  conditions

20
Low-flow Data Set

• Low-flows estimated for each watershed based on
  nearby stream gages
  
• Nonpoint source loads were estimated for each of
  the 39 subwatersheds and distributed to the EFDC
  grid cells within those watersheds
  
• Water withdrawals were set to either the safe
  yield or 75 of the peak withdrawal rate
  

21
TMDL Endpoints
22
Stream Use Designations
23
Baseline Conditions

• NPDES point sources set to existing permit
  limits.
  
• Tributary loads based on low-flow concentrations
  for the subwatersheds from the period 1988-1998
  (STORET)
  
• Reasonable agreement with Omernik (1977) study of
  watersheds in the eastern U.S.

24
Baseline conditions Daily Average DO
25
Baseline conditions Minimum DO
26
EMPR Allocation Strategy

• Equal Marginal Percent Removal
• Level 1
• Analyze each NPDES discharge individually to
  determine if water quality standards are met
  
• Set other tributaries and point sources not being
  considered to a baseline (background) condition
  
• If an individual point source does not meet water
  quality standards (WQS), then reduce the CBOD,
  TN, TP load until WQS are achieved
  
• Level 2
• Add other discharges to the analysis one at a
  time based on the size of the mass load of CBOD
  
• If WQS are not met, reduce CBOD, TN, TP load for
  all point sources in the analysis by equal
  percent
  

27
EMPR Level I Daily Average DO
28
EMPR Level I Minimum DO
29
EMPR Level 1 Allocations
30
EMPR Level 2 Allocations
31
Results of EMPR Level 2 TMDL Allocations, East
Branch Brandywine Creek
Daily Average DO Downingtown WWTP discharges
7.0 mgd at RM103.7 Sonoco discharges 3.0 mgd
at RM 104.1 Existing Permits DO sag 3.23 mg
/L After TMDL DO sag 5.92 mg/L
32
Results of EMPR Level 2 TMDL Allocations, East
Branch Brandywine Creek
Diel Minimum DO Downingtown WWTP discharges 7
.0 mgd at RM103.7 Sonoco discharges 3.0 mgd a
t RM 104.1 Existing Permits diel min DO 0.6
3 mg/L After TMDL diel min DO 4.02 mg/L