Transcript Item5_ModelingDetermineEffectivenessUrbanBMPs

WATERSHED MODELING IN
SOUTHERN CALIFORNIA
Kenneth Schiff
Southern California Coastal Water Research Project
www.sccwrp.org
SCCWRP’s Watershed Research Themes
• Loads and concentrations
• Runoff mechanisms and processes
• Interactions with the coastal ocean
• Source tracking and identification
• Effectiveness of BMPs
Modeling Advantages
• More complete characterization of pollutant
concentrations and loads
- All storms of a year, wet vs. dry years
• Source attribution
- Subwatershed, land use, municipal boundaries
• Assess management scenarios
- Effectiveness, cost efficiency
Water Quality Modeling
In Southern California
• Watershed-scale water quality models already exist
- HSPF, SWMM, EFDC, LSPC, etc.
• Virtually none were built in California
- Not necessarily designed for arid or urban environments
• Previous time scales largely insufficient for urban
applications
- Need minutes to hours for within storm applications
WATER QUALITY MODELING FRAMEWORK
Flow &
Pollutant
Inputs
Flow &
Pollutant
Inputs
Dispersion
Nitrification
Flow &
Pollutant
Inputs
Transport
Uptake
Decay
Partitioning
Denitrification
Instream
Concentration
Land use sites
< 0.5"
0.5" - 1.0"
> 1.0"
Total
1
5
Agriculture
3
1
Commercial
4
1
High density residential
4
2
1
7
Industrial
5
1
1
7
Low density residential
2
1
Open space
Recreational
1
1
1
Natural loadings (microsampling)
< 0.5"
Arroyo Seco
3
3
8
12
> 1.0"
Total
1
2
3
3
3
1
7
1
2
3
0.5" - 1.0"
1
Arroyo Sequit
2
Coyote Creek
2
2
4
Dominguez Channel
1
1
LAR u.s. Arroyo Seco
2
1
1
3
Los Angeles River
1
1
1
3
San Gabriel River
1
2
3
San Jose Creek
1
1
2
2
2
Santa Monica Canyon
Verdugo Wash
Walnut Creek
2
2
1
1
30 land use
site events
2
1
Natural loadings (pollutograph)
Ballona Creek
3
2
Transportation
Mass emission sites
5
17 natural
site events
34 mass
emission
site events
3
1
2
Through 4/30/05
0.1
Open Space
Open Space
LDR Unsewered
LDR Sewer
LDR Sewer
LDR Sewer
Ind Mixed
Ind Mixed
HDR Pet
HDR Pet
HDR Mixed
HDR Mixed
Comm
Comm
Comm
Ag Mixed
Ag Mixed
Ag Mixed
Ag Mixed
Copper (kg)
LAND USE LOAD COMPARISON
10000
Measured
Modeled
1000
100
10
1
Commercial Pollutograph
Event 1
0.0
1,600
Measured
Modeled
Rain
E. Coli (#/100mL)
1,400
1,200
0.1
0.2
0.3
0.4
1,000
0.5
800
0.6
600
0.7
400
0.8
200
0.9
0
02/19 12:00
1.0
02/19 15:36
02/19 19:12
02/19 22:48
Rain (in)
1,800
Copper Loading From Ballona Creek (1990-2000)
10,000,000
Hardness = 100 mg/L
CTR = 13.4 ug/L
Storm Load (grams)
1,000,000
100,000
10,000
1,000
100
0.00 0.01 0.01 0.03 0.04 0.09 0.13 0.20 0.30 0.48 0.62 0.74 1.05 1.61 2.65
Storm Rain (in)
Most Likely BMPs from Stakeholders
•
Cisterns
•
Detention basins
•
Load reduction
•
Impervious reduction
•
Wetland treatment
•
Instream
- Impoundment
- Diversion
Impoundment and Diversion
Assumptions
•
Rubber dam at end of Ballona Creek
- 5 ft high
•
Overflows during storms
•
Retained volume is diverted flow to WRP
- Constant diversion of 20 MGD
http://ca.water.usgs.gov/user_projects/sonoma/pics/rubberdam.jpg
Rubber
dam
To WRP
http://www.salinasvalleywatercoalition.org/images/rubberdam.jpg
Decadal Simulation
Change in Load
20 MGD Diversion
Volume
-16 %
Solids
-3 %
Copper
-6 %
Bacteria
-6 %
Sensitivity Analysis
Change in Load
20 MGD
(13 cfs)
40 MGD
(26 cfs)
50 MGD
(32 cfs)
Volume
-16 %
-18 %
-20 %
Solids
-3 %
-4 %
-4 %
Copper
-6 %
-7 %
-7 %
Bacteria
-6 %
-7 %
-7 %
Effectiveness of
Impoundment
•
5 ft dam retains 195 ac-ft
of water
Measured Decadal Storm Volumes
100%
90%
80%
59% of storms > 195 ac-ft
70%
Percentile
•
•
Storms > 195 ac-ft account
for 99% of decadal storm
volume
- 87% total decadal volume
60%
50%
40%
30%
20%
10%
0%
1
10
100
1,000
Storm Volume (ac-ft)
10,000
100,000
Most Likely BMPs from Stakeholders
•
Cisterns
•
Detention basins
•
Load reduction
•
Impervious reduction
•
Wetland treatment
•
Instream
- Impoundment
- Diversion
Cistern Application
Assumptions
•
Distribute cisterns on specific land uses throughout
watershed
- Public
- Open
- Transportation
- Commercial
•
Cisterns capture all rainfall for an average year (12”)
Cisterns store rain and pollutants up to their capacity
and then bypass to the stream
•
Cisterns emptied during the summer
•
- Used for irrigation, no net increase in dry weather runoff
Decadal Simulation
Change in Load
Cistern
Volume
-11 %
Solids
-8 %
Copper
-14 %
Bacteria
-11 %
Effectiveness of Cisterns
Base Volume Loads
Cisterns only applied
to specific land uses
Agriculture
Commercial
High Density Residential
Industrial
Low Density Residential
Mixed Urban
Open
Transportaion
Water
Public-Open
Base Volume Loads
Base Sediment Loads
Agriculture
Commercial
Agriculture
High Density Residential
Commercial
Industrial
High Density Residential
Low Density Residential
Industrial
Mixed Urban
Low Density Residential
Open
Mixed Urban
Transportaion
Open
Water
Transportaion
Public-Open
Water
Public-Open
Base Copper Loads
Base Bacteria Loads
Agriculture
Agriculture
Commercial
Commercial
High Density Residential
High Density Residential
Industrial
Industrial
Low Density Residential
Low Density Residential
Mixed Urban
Mixed Urban
Open
Open
Transportaion
Transportaion
Water
Water
Public-Open
Public-Open
Summary
• Watershed models are a powerful management tool
- Extrapolate to unmonitored times/locations
• Requires effort to ensure you believe them
- Calibration and validation
- Wet and dry weather
• Provides invaluable insight into evaluating
management action effectiveness
- Cost efficiency
Soon To Be Released
• More detailed and involved BMPs
• Design storm for water quality
- Function of hydrology
- Function of BMP cost
• Linked estuarine models
- Contaminated sediments
EXCEEDENCE RATE AT BALLONA CREEK
Percent of Storms Exceeding
Thresholds for Copper
40%
35%
30%
25%
20%
15%
10%
5%
0%
0
0.5
1
1.5
Rainfall Runoff Captured (in)
2