These maps are from a project in ENVS-420 (GIS Analysis and Modeling) at Western Washington University. The purpose of the project was to determine where watershed restoration efforts were required in order to increase water quality in WRIA 1 (Water Resource Inventory Area 1). Stream-road intersections and agricultural land percentage are used as indicators of watershed water quality in this project.
Agricultural land can negatively impact water quality through infiltration of pesticides, manure runoff, and soil erosion. Stream-road intersections can likewise negatively impact water quality through runoff.
This model utilizes inline variable substitution and thus can be applied to other watersheds if desired. More details on the analysis portion in the model shown below (Fig. 3).
Fig. 1: Shows stream-road intersection density and percentage of agricultural land in WRIA 1 by watershed. Stream-road intersections are points where a stream and road cross paths. Density is calculated by summing the total stream-road intersections for a watershed and dividing by the total stream length for that watershed. Agricultural land percentage was calculated by dividing the total agricultural land area by the area of that watershed.
Fig. 2: Shows four USGS streamgages symbolized with circles, (the bigger the circle, the higher the discharge), as well as the Nooksack River and its tributaries.
The first part of the model selects for a single WRIA, erases any area that overlaps with a marine waters polygon, clips a watersheds dataset to the selected WRIA, and clips roads/streams to the selected WRIA.
Then, the model finds which streams belong to which watersheds, calculates the stream length, and adds the length of each individual stream for each watershed to get the total stream length per watershed. It updates the original watershed dataset with this information.
After this, the model finds where streams and roads intersect, associates those intersection points with each watershed, and counts how many intersection points there are in each watershed. It then updates the original watershed dataset.
Finally, model associates agricultural land from an input dataset with each watershed, calculates the area of agricultural land in each watershed, and updates the original watershed dataset.