Applying the Sub-Daily SWAT Model to Assess Aquatic Life ...Applying the Sub-Daily SWAT Model to Assess Aquatic Life Potential under Different Development Scenarios in the Austin,

Applying the Sub-Daily SWAT Model to Assess

Aquatic Life Potential under Different Development

Scenarios in the Austin, Texas Area

Leila Gosselink, P.E.

Roger H. Glick, P.E., Ph.D.

Watershed Protection Department

City of Austin, Texas Presented at

2011 International SWAT Conference

Toledo, Spain

Differing Growth Scenarios

Purpose and Objectives

• Identify measures of flow regime changes

critical to aquatic life at different locations

that are well-modeled by SWAT

• Evaluate changes in aquatic life potential

at different locations with respect to critical

hydrologic metrics

"T

"T

"T

"T

"T

"T"T

"T"T

"T"T

"T"T

"T

"T

"T"T

"T"T

"T

"T

"T

[¡

[¡

[¡

[¡ [¡

[¡

[¡ [¡

[¡

[¡

[¡

[¡

[¡ [¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡ [¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡[¡[¡

[¡

[¡

[¡

[¡[¡

[¡[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡ [¡

[¡

[¡ [¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡[¡

[¡

[¡[¡

[¡[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡ [¡[¡

[¡[¡[¡

[¡[¡

[¡[¡

[¡

[¡

[¡

[¡

[¡

[¡[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡[¡[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

ONI5

ONI6

BAR5

ONI4a

ONI4

BER3

ONI1

WMS2

BAR6

DRE1

WLN4

LBR1

BAR2

BUL2

DRE2

GIL2

SLA2

SLA1

WLN2

GIL6

SLA3

GIL1

ONI3

LKC2

LBR2

DKR1

BAR3

SHL3

DKR2

GIL4

HRS1

LKC1

SFD2

RAT1

BAR1

LBA3

WMS3

WBL2

CTM1

WLN1

BAR4

WLN3

LKC3

LWA3

BER2

ONI2

ELM1

HRS2

PAN1

BUL3

BER1

GIL5

WMS1

WLN5

GIL3

RIN2

NFD1

ELM2

CAR1

MAR2

SHL4

SBG2

LWA2

LBA1

BUL5

RIN1

SFD1

LBA2

RIN3

BOG1

LWA1

BUL1RDR1

EAN1

CRN1

CAR2

TRK1

CMF1

DKR3

BRW1

BOG3

TAN1

CCE1

JOH1

TAN3

RAT2

TYN1

SBG1

CCW2

WBO3

BUL4

EAN2

BOG2

LBE1

WBL1

BEE2

WLR3

SHL2WLR2

BEE1

FOR2

FOR3

DRN1

EBO2

CCW1

SHL1

DRN2

WLR1

TAN2

BLU3

BEE3

LWA4

TYS1

WBO2

MAR1

BLU2

BMK1

HRP1

BMK3

BMK2

FOR1

EBO1

EBO3

FOR4

WBO1

BLU1

Environmental Integrity (EII) Sampling Sites

and Continuous Flow Gages Citywide

for Hydrologic Regression Analyses

Current Aquatic Life Scores at EII Sites and Flow Gages

[¡ EII_Sampling_Sites

"T usgs_wqm_sites

Very Bad

Bad

Poor

Marginal

Fair

Good

Very Good

Excellent

Aquatic Life Evaluation

)_(*2655.21

)90ln(*3842.4)/(*016.07539.87

RiseAvg

QAreaQpeakAQP

where,

Qpeak/Area = peak flow rate in cms/100 sq km

Q90 = 90th percentile flow rate in cms; 90% of flow is below this value

Avg_Rise = mean of positive differences between consecutive rising

flow values (rise rate in cms/sec)

Study Area - Walnut Creek

• 145.8 km2 watershed

• USGS gage data 1967 to present

• 3-m DEMs

• SURRGO Soils

• 15-minute rainfall at 18 gauges

• Lot level land use

• 298 sub-basins

• ~4500 HRUs

• Sub-daily – NSE = 0.74

– r2 = 0.78

Walnut Land Use Scenarios

Environmental Integrity Sampling Sites for

Benthic & Diatom Communities on Walnut Creek

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

WLN4

WLN2

WLN1

WLN3

LWA3

WLN5

LWA2

LWA1

LWA4

BMK1

BMK3

BMK2

• Environmental Integrity

Index (EII) – Aquatic Life

Support Assessment

• Benthic & Diatom

Community Assessment –

Changes over time:

– Every three years, 4x/yr

– Every two years, 3x/yr

– Every two years, 1x/yr

EII Reaches – Drainage Area

1

10

100

1,000

BM

K3

BM

K2

BM

K1

LWA4

LWA3

LWA2

LWA1

WLN

5

WLN

4

WLN

3

WLN

2

8158

600

WLN

1

Are

a

(sq

uare

kil

om

ete

rs)

Walnut @ Metric Blvd

Walnut @ I-35

Walnut @ Old Manor Rd

Walnut @ SPRR Bridge

Tributaries

Little Walnut Buttermilk

Land Use by Subbasin Over Time

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Pe

rce

nt

of

Wa

ters

he

d A

rea

Walnut Mainstem Land Use (sq km)

Com 7.14 21.27 36.59

ROW 6.89 16.90 17.16

Res 2.81 23.10 36.31

Undev 95.50 49.97 20.39

1964 2003 Future

0%

20%

40%

60%

80%

100%

Pe

rce

nt

of

Wa

ters

he

d

Are

a

Buttermilk Subbasin Land Use (sq km)

Commercial/Industrial 0.61 1.40 2.02

ROW (roads & utils.) 0.69 0.99 0.97

Residential 0.54 1.25 1.23

Undeveloped 2.46 0.48 0.07

1964 2003 Future

Hydrologic Metrics – SWAT Model EII Reach - Drainage Area

1

10

100

1,000

BM

K3

BM

K2

BM

K1

LWA4

LWA3

LWA2

LWA1

WLN5

WLN4

WLN3

WLN2

8158600

WLN1

Are

a

(sq

ua

re k

ilom

ete

rs)

Qpeak/area

0

5

10

15

20

25

Flo

w c

ms

/sq

km 1964

2003

Future

Hydrologic Metrics – SWAT Model

EII Reach - Drainage Area

1

10

100

1,000

BM

K3

BM

K2

BM

K1

LWA4

LWA3

LWA2

LWA1

WLN5

WLN4

WLN3

WLN2

8158600

WLN1

Are

a

(sq

ua

re k

ilom

ete

rs)

Average Rise

0.0

0.5

1.0

1.5

Ris

e,

cm

s

1964

2003

Future

Hydrologic Metrics – SWAT Model EII Reach - Drainage Area

1

10

100

1,000

Are

a

(sq

uar

e ki

lom

eter

s)

90th Percentile Flow Rate

0.001

0.01

0.1

1

10

BM

K3

BM

K2

BM

K1

LWA4

LWA3

LWA2

LWA1

WLN

5

WLN

4

WLN

3

WLN

2

8158

600

WLN

1

Flo

w, cm

s

1964

2003

Future

Aquatic Life Potential in Walnut Creek

based on SWAT Flows

0

10

20

30

40

50

60

70

80

90

BM

K3

BM

K2

BM

K1

LWA4

LWA3

LWA2

LWA1

WLN

5

WLN

4

WLN

3

WLN

2

8158

600

WLN

1

AQ

P

1964

2003

Future

*not

norm

aliz

ed t

o r

efe

rence s

ite

Changes in Aquatic Community Health

0

10

20

30

40

50

60

70

80

90

BMK3 BMK2 BMK1 LWA4 LWA3 LWA2 LWA1 WLN5 WLN4 WLN3 WLN2 8158600 WLN1

AQ

P

1964

2003

Future

Very Good

Good

Fair

Marginal

Poor

Predicted Response (AQP) of Aquatic

Communities to Development

AQP Decrease 2003 - Future

>18

15 - 18

12 - 15

9 - 12

6 - 9

3 - 6

.5 - 3

insignificant + or -

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

WLN4

WLN2

WLN1

WLN3

LWA3

WLN5

LWA2

LWA1

LWA4

BMK3

BMK2BMK1

AQP Decrease 1964 - 2003

>18

15 - 18

12 - 15

9 - 12

6 - 9

3 - 6

.5 - 3

insignificant + or -

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

[¡

WLN4

WLN2

WLN1

WLN3

LWA3

WLN5

LWA2

LWA1

LWA4

BMK3

BMK2BMK1

Walnut Land Use Scenarios

Conclusions

Flow regime is an important factor in the health of the aquatic

community and can be characterized by regression equations.

The sub-hourly SWAT model can simulate flow characteristics well

for many measures of urban impacts (Glick & Gosselink 2011)

Predicted flow, that is well modeled, can be used to estimate the

changes in aquatic life from development, making SWAT models

useful tools for environmental management.

With BMP capabilities, SWAT can evaluate management methods

to control aquatic impacts that are driven by flow alterations

Estimates of Aquatic Life Potential based on modeled hydrology

can assist in setting goals and focusing resources on appropriate

solutions

Questions?

Roger Glick

[email protected]

Leila Gosselink

[email protected]

Further Studies

Hydrologic Metrics/SWAT:

Use SWAT with BMPs to simulate conditions with & without existing

BMPs (currently not included in calibration model)

Evaluate methods to quantify & separate watershed size factors;

SWAT simulations holding land use, etc. the same for different watershed sizes could

provide insight

Evaluate hydrologic measures, appropriate time steps & normalization

Evaluate SWAT modeling capabilities relative to individual metrics

Aquatic Community Metrics:

Evaluate bioassessment data for representativeness vs. hydrologic data

Evaluate appropriateness of normalizing aquatic life relative to a reference site

Sensitivity analyses