Work Samples - GIS Watershed Climate Change Resilience Modeling

Table of Contents

Introduction

In 2016, I had the opportunity to participate in a project to model the climate change resilience of the North Santiam basin in the Willamette River watershed for the North Santiam Watershed Council. A colleague and I developed a reusable model that uses numerous indicator datasets in a multi-tier suitability analysis to assign a resilience score at all points within a modeled riparian buffer along the stream reaches. The model also computes a zonal mean score for each of the subwatersheds within the North Santiam. These two representations help the council to better prioritize their watershed preservation and restoration efforts. A summary of the project may be found in the poster below.

This browser does not support inline PDFs. Please download the PDF to view it: Download PDF


Method overview

The model incorporates dozens of datasets that are grouped into the following categories: Water, Aquatic, Riparian, and Terrestrial. Example datasets include:

  • Water
    • Predicted change in mean flow
    • Predicted change in temperature
    • Number of water quality limited stream parameters
  • Aquatic
    • Anadromous fish distribution for spawning and rearing
    • Large woody debris, side channels, pools
  • Riparian
    • NLCD vegetation cover
    • Riparian shade differential
    • Riparian bird distribution
  • Terrestrial
    • Conservation opportunity areas
    • Drinking water sources (groundwater, surface water)
    • Terrestrial bird distribution

At a high level, the model includes the following steps:

  1. Convert each dataset to raster (if necessary) and compute the resilience score for each dataset.
  2. Compute combined score for the datasets belonging to each factor (water, aquatic, riparian, terrestrial), applying predetermined weights to each dataset.
  3. Normalize the combined score for each factor
  4. Compute the composite score ("WART score") including all of the factors and applying predetermined group weights for each factor.
  5. Normalize the WART score. Call the resulting raster the riparian WART score raster.
  6. Compute the zonal mean for each of the HUC 12 subwatersheds in the basin.
  7. Produce maps showing the following:
    1. Basemap and context layers
    2. Riparian WART score raster
    3. Subwatershed zonal means

Sample maps

Map showing riparian WART score raster:

This browser does not support inline PDFs. Please download the PDF to view it: Download PDF


Map showing subwatershed zonal means:

This browser does not support inline PDFs. Please download the PDF to view it: Download PDF


Sample model

The following ModelBuilder diagram shows the model used to prepare the datasets and compute the combined score for the Water factors. Similar models were used for the Aquatic, Riparian, and Terrestrial factors, as well as to prepare context-only datasets. A sixth model was used to compute the riparian WART raster and the subwatershed zonal means. Maps were produced manually for this project, but may be automated for future projects.

This browser does not support inline PDFs. Please download the PDF to view it: Download PDF

Author: Adam Porr

Created: 2018-01-17 Wed 12:29

Validate