ANSWERS on GRASS: Integration of a watershed simulation with a geographic information system
Abstract
A program to simulate hydrologic and sediment processes of agricultural watersheds was integrated with a geographic information system (GIS). The tools developed were evaluated on a watershed located on a Purdue University farm and the significance of effects due to the extraction of elevation-based model inputs from a GIS database on the watershed simulation were highlighted. The study explored the difference between preparation of simulation inputs with the tools developed and traditional methods. ANSWERS (Areal Nonpoint Source Watershed Environmental Response Simulation), a distributed parameter model designed for planning and evaluation of various strategies for controlling non-point source pollution from agricultural land, was integrated with the GRASS (Geographical Resources Analysis Support System) GIS. A suite of GIS programs was developed as a "hydrologic toolbox" wherein spatial data needs of ANSWERS are served by GRASS. Central to the integration is a "project manager" program, which provides user interface to simulation scenario management, including the extraction and preparation of ANSWERS input data from GRASS, running the simulation, and post-processing outputs into GRASS maps of sediment movement. Other tools provide preparation of input maps and visualization and analysis of outputs prepared by the project manager. The ANSWERS-on-GRASS integration was evaluated by modeling the Purdue Animal Science watershed located in Tippecanoe County, Indiana. Topographic attributes such as slope steepness are important factors in predicting soil loss and chemical movement using hydrologic simulation models. The effects of four slope estimation algorithms (neighborhood, quadratic, best fit plane and maximum slope method) in deriving slope steepnesses and erosion estimates using the project manager were explored. Significant differences were found among the four slope prediction methods in topographic attributes and erosion estimates at the outlet of the watershed and within the watershed. Preparation of input data required by ANSWERS was explored by comparing traditional methods with those using GIS-assisted techniques. It was found that the cost of creating a GIS database needed for watershed simulation is initially expensive, but once established the GIS can provide significant improvements in time and expertise needed to run simulations and analyze outputs.
Degree
Ph.D.
Advisors
Engel, Purdue University.
Subject Area
Agricultural engineering|Environmental science|Hydrology
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.