A Next Step: Soil and Water Assessment Tool SWAT Modeling for the St' Croix River Basin

1
A Next StepSoil and Water Assessment Tool
(SWAT) Modeling for the St. Croix River Basin
James E. Almendinger St. Croix Watershed Research
Station Science Museum of Minnesota
2
TAPwaters project

• Objectives
• Establish TAPwaters modeling center at the SCWRS
• Several computers, plus expertise in SWAT
  modeling
• Create watershed models of targeted sub-basins
• Willow River in WI Sunrise River in MN
• Create watershed model for entire St. Croix Basin
• A work in progress, beginning with low resolution
• Funding and Timing
• 80 LCMR, 2003-06
• 20 WDNR, 2003-05
• Pending proposals to NPS and EPA
• Personnel
• J. Almendinger (SCWRS)
• M. Murphy (Univ. of Minn. grad student)
• M. Young (NPS)
• More grad students/post-docs pending further
  funding

3
TAPwaters facilities
4
Outline TAPwaters and modeling in the St. Croix
Basin

• I. Basin Planning and Use of Watershed Models
• Some general thoughts about models and their role
  in basin planning
• II. SWAT Soil and Water Assessment Tool
• Modeling program developed by the USDA/ARS
• Plans for use in the St. Croix Basin

5
I. Basin Planning and Use of Watershed Models
6
Basin Planning

• Federally mandated in 1965
• Water Resources Planning Act authorized
  regional...
• River Basin Commissions, interagency teams with
  federal and state representatives, including
  the...
• Upper Mississippi River Basin Commission
• Funding cut in 1980 -- but law is still on the
  books
• States also bear responsibility
• Minnesota MPCA will produce basin plans for
  major river basins, including the St. Croix (1-2
  personnel)
• Wisconsin WDNR (1-2 personnel)
• St. Croix has interagency Basin Team -- but with
  no formal authority

7
Basin Planning and Computer Models

• BUT -- basin plans are difficult because large
  watersheds are complex!
• Managers need all the tools possible
• Computer modeling of watersheds is one such tool
  that is valuable because
• Models can integrate available data, help clarify
  and help quantify primary watershed processes,
  and guide further data collection
• Models can be used to help predict the effect of
  changing management practices by simulating the
  essential watershed processes

8
What is a model?

• Model simplification of reality
• Purpose To reduce a complex system to its
  essential processes, so that system behavior may
  be simulated under different conditions
• Types of models
• Qualitative/conceptual
• Quantitative
• Physical scale models
• Computational models (equations)
• Model results hypotheses
• Best interpreted in terms of relative changes,
  not absolute amounts

9
What is a model?A simplification of reality, but
with essential features -- an example
10
Whats the difference betweena modeling program,
and a model?

• Modeling program software
• Software with an array of equations that simulate
  watershed processes, e.g., SWAT, HSPF, or PRMS
• Model software data
• A modeling program data specific to that
  watershed
• Climate, topography, hydrography, soils, and
  land use

11
What goes into a watershed model?And what comes
out?
12
How are model results interpreted?

• Often harshly -- Models dont work!!
• I.e., The model results didnt match my
  expectations...
• Model results must be reasonable
• If theyre not reasonable then either
• The model needs improvement (usually), or
• Our understanding needs improvement (often)

• Expectations must be reasonable
• Model will never simulate absolute quantities
  exactly
• Models better suited to estimating relative
  changes by simulating essential watershed
  processes

13
Modeling is just one management tool...
Data interpretation hypothesis generation
Adapative Ecosystem Management
Data collection
Experimentation
14
How does modeling fit in with data collection?
1 25-yr data collation
2 1999 loads
4 Point-sources
6 Watershed modeling (SWAT)
3 Paleo-reconstructions of sed and P
5 Lake St. Croix modeling (BATHTUB)
15
How does modeling fit in with goal setting?
16
II. SWAT Soil and Water Assessment Tool

• (text quotes and some figures in the following
  presentation were taken from SWAT user manuals)

17
What is, and why, SWAT?

• SWAT (USDA/ARS)
• Objective ...to predict the impact of land
  management practices on water, sediment and
  agricultural chemical yields in large complex
  watersheds with varying soils, land use and
  management conditions over long periods of time.
• Strengths
• Best at NPS-pollution loads from agricultural
  practices improving routines for urban loads
• Lots of local SWAT expertise, esp. within WDNR
• Good interface with GIS software (ArcView)
• Other modeling software options
• HSPF (USGS/EPA)
• PRMS (USGS)

18
Main features of SWAT

• Physically based
• Needs basin-specific data on ...weather, soil
  properties, topography, vegetation, and land
  management practices...
• Benefits
• Ungaged watersheds can be modeled
• Sensitivity to different input data can be
  quantified
• Uses readily available inputs
• GIS sets often already compiled by government
  agencies
• Computationally efficient
• Even for very large basins
• Determines long-term impacts
• SWAT is a continuous time model -- ...not
  designed to simulate detailed, single-event flood
  routing.

19
Watershed Partitioning
20
Phases of hydrologic cycle simulated by SWAT
21
SWAT land phase of hydrologic cycle
22
SWAT phosphorus cycle
23
SWAT nitrogen cycle
24
SWAT pesticide fate transport
25
Willow River Watershed location map
26
Willow River Watershed
27
Issues in the Willow River Watershed
28
Summary and Conclusions
29
Phases of hydrologic cycle simulated by SWAT
Water phase
Land phase