The Green Stormwater Infrastructure (GSI) Suitability Mapper

The Rouge River Watershed is home to approximately 1.5 million people in Southeast Michigan, near Detroit. This watershed is roughly 467 square miles, much of which is highly developed, where natural land has been replaced by impervious surfaces such as concrete, asphalt, and roofs. Unfortunately, impervious surfaces prevent stormwater runoff from draining properly and soaking into the ground, which leads to serious problems for local residents and the ecosystem including flooding, pollution, and habitat degradation. Green Stormwater Infrastructure (GSI) provides a potential solution by reducing the amount of stormwater runoff. GSI acts to hold water in place while also filtering some pollutants out of the water. More information about GSI can be found on this link:  What is Green Infrastructure? 

Within the watershed, stormwater runoff always flows from higher elevations to lower elevations, as shown in Figure 1. While the likelihood of flooding will always be greater at lower elevations, GSI should be implemented throughout the watershed. This practice ensures optimal runoff control, significantly reducing stormwater runoff and its negative impacts. However, some locations may be better suited for GSI than others depending on the objectives of a particular project. Friends of the Rouge has developed the GSI Suitability Mapper as a tool for helping local residents and stakeholders find the best locations for any kind of GSI project. Read on to learn more about how the tool was created and how it can best be utilized.

This tool is a vector layer geodatabase that can be used to prioritize the planning and implementation of Green Stormwater Infrastructure (GSI) projects across the Rouge River Watershed. It can be used to select from a diverse range of parameters based on the priorities of different stakeholders or projects. It is based on three priority categories and one avoidance parameter. The priority categories include: Performance & Infiltration Potential, Environmental & Ecological Need, and Demographics & Social Need. We have also included a buffer around known environmental contamination sites as an important avoidance parameter. Below is a list of the data layers included in each category, their sources, and relevant notes about how they have been incorporated in the database:

• Hydrologic Soils - SSURGO and STATSGO2 shapefiles were extracted from the  USDA Web Soil Survey . The STATSGO2 data was used to update the null values in the SSURGO shapefile to produce a more complete dataset.
• Water Table Depth Annual Minimum - This data was obtained from the  USDA Web Soil Survey .
• Slope - This layer was created using LiDAR data extracted from NOAA’s  Data Access Viewer .

• Land Cover Classes - Land Cover data from 2010 was provided by SEMCOG. This layer was updated using SEMCOG’s  Building footprints (2020)  data. The result was then aggregated by census tract to calculate the percentage of Impervious Surface, Open Space, Trees, Urban Bare Space, and Water in each census tract census tract.
• Sewer System Type - This data was provided by EGLE. It is important to note that parts of this dataset may be out of date and no longer accurate.
• Land Surface Temperature (LST) - This layer was created using Landsat satellite images from July and August of 2013 though 2022. The data was extracted from  NASA  using Google Earth Engine (GEE), and the mean LST was calculated for each census tract.
• Land Use - This data came from SEMCOG’s  Land Use (2020)  dataset. A 1000ft buffer was added around areas classified as Industrial.
• Park Deserts - Using SEMCOG’s  Parks (2021 ) data, a 0.5 mile buffer was created around parks larger than 5 acres. Land outside this buffer is considered a park desert. 

• Social Vulnerability Index (SVI) - This open source data was provided by the  Center for Disease Control (CDC)  and the Agency for Toxic Substances and Disease Registry (ATSDR). Each Census Tract has a score in percentiles.
• Environmental Justice (EJ) - This data can be found on the  EJScreen mapping tool website . EJScreen provides two different indices: the EJ Index and the Supplemental Index. Both are included in the database and more information on the indices can be found  here . Each Census Tract has a score in percentiles.

• Environmental contamination sites (Red flags) - This data is from  EGLE . A 300ft buffer was created around each point in the dataset.

The GIS Suitability Mapper is designed so that users can select certain parameters within the data layers listed above to find the best locations for their GSI projects. First, users should identify criteria from the full list of parameters below based on the requirements of their project and stakeholders. It is important to note that this data is aggregated at the census tract level, though the database includes county, municipality, and sub-watershed classifications if the user is interested in looking at the data at different scales. Based on the parameters selected by the user, areas that fulfill those criteria will be selected from the database.

• Sewer System: Combined Sewer Overflow (CSO), Municipal Separate Stormwater (MS4), NULL
• Hydrologic Soils Group: A, A/D, B, B/D, C, C/D, D, NULL
• Mean Annual Minimum Water Table Depth: 0-137cm, NULL
• Slope Percent: Greater than 6%, Less than 6%
• Mean Summer Land Surface Temperature: 85-116°F
• % Total Impervious: 0-100%
• % Tree Cover: 0-100%
• % Open Space: 0-100%
• % Urban Bare: 0-100%
• % Water: 0-100%
• Social Vulnerability Index Theme 1 Socioeconomic Status: 0-1
• Social Vulnerability Index Theme 2 Household Characteristics: 0-1
• Social Vulnerability Index Theme 3 Racial & Ethnic Minority Status: 0-1
• Social Vulnerability Index Theme 4 Housing Type & Transportation: 0-1
• Overall Social Vulnerability Index: 0-1
• EJScreen Demographic Index: 0-100
• EJScreen Demographic Supplemental Index: 0-100
• EJScreen EJ Index Particulate Matter 2.5: 0-100
• EJScreen EJ Index Ozone: 0-100
• EJScreen EJ Index Diesel Particulate Matter: 0-100
• EJScreen EJ Index Air Toxics Cancer Risk: 0-100
• EJScreen EJ Index Air Toxics Respiratory Hazard Index: 0-100
• EJScreen EJ Index Toxic Releases to Air: 0-100
• EJScreen EJ Index Traffic Proximity: 0-100
• EJScreen EJ Index Lead Paint: 0-100
• EJScreen EJ Index RMP Facility Proximity: 0-100
• EJScreen EJ Index Hazardous Waste Proximity: 0-100
• EJScreen EJ Index Superfund Proximity: 0-100
• EJScreen EJ Index Underground Storage Tanks: 0-100
• EJScreen EJ Index Wastewater Discharge: 0-100
• EJScreen Supplemental Index Particulate Matter 2.5: 0-100
• EJScreen Supplemental Index Ozone: 0-100
• EJScreen Supplemental Index Diesel Particulate Matter: 0-100
• EJScreen Supplemental Index Air Toxics Cancer Risk: 0-100
• EJScreen Supplemental Index Air Toxics Respiratory Hazard Index: 0-100
• EJScreen Supplemental Index Toxic Releases to Air: 0-100
• EJScreen Supplemental Index Lead Paint: 0-100
• EJScreen Supplemental Index RMP Facility Proximity: 0-100
• EJScreen Supplemental Index Hazardous Waste Proximity: 0-100
• EJScreen Supplemental Index Superfund Proximity: 0-100
• EJScreen Supplemental Index Underground Storage Tanks: 0-100
• EJScreen Supplemental Index Wastewater Discharge: 0-100
• Number of EJScreen EJ Indexes Over 80: 0-12
• Number of EJScreen Supplemental Indexes Over 80: 0-13

Next, the selected areas can be displayed using the database’s Priority Areas attribute. The Priority Areas attribute was created by analyzing the database’s Land Use, Park Deserts, and Environmental Contamination Site data layers so that users can determine the suitability of land at a finer scale within the census tracts that were identified in Step 1. For example, any land within the 300 ft buffer of the contamination sites likely needs further field work and assessment before developing GSI projects. Within the Land Use category, GSI projects may be particularly beneficial if placed within a 1000 ft buffer of industrial areas, and vacant lots may be ideal locations for implementation. Placing GSI on land classified as a Park Desert may have the additional benefit of adding much needed green space for communities that lack it. Considering these additional factors allows the user to gain further insights into the areas they have selected. The Priority Areas attribute includes a breakdown of these factors alone and in combination with each other. The complete list of categories within this attribute include:

• Environmental Contamination (EC)
• Industrial Areas (IA)
• Park deserts (PD)
• Vacant Lots (VL)
• Environmental Contamination Near Industry (ECI)
• Park desert + Industrial Areas (PDIA)
• Vacant Lot + Industrial Area (VLIA)
• Vacant Lot + Park Desert (VLPD)
• Vacant Lot + Park Desert + Industrial Area (VLPDIA)

In order to demonstrate how to use the GSI Suitability Mapper, we ran a number of tests based on different hypothetical GSI projects. We chose rain gardens and reforestation projects as two common GSI initiatives, and ran multiple tests with different hypothetical parameters for both project types.

Many factors are common to all GSI projects, including imperviousness, open space, sewer system type, and social vulnerability index (SVI) and environmental justice (EJ) factors. In general, census tracts with high imperviousness need more attention for the development of GSI, and the best locations for any GSI are open spaces in proximity to impervious surfaces. Areas with a combined sewer system (CSO) are also likely a higher priority for GSI installations. The SVI and EJ layers add social factors to the planning process to prioritize areas with a higher social need, making the GSI development inclusive within the communities of the Rouge River watershed. Priorities specific to siting rain gardens might include soils with medium to good infiltration potential, a lower water table, and land with a lower to gentle slope.

We ran multiple tests with different combinations of these parameters to show how the GSI Suitability Mapper might be used for different hypothetical rain garden projects. Click through the maps below to explore the parameters and results of each rain garden suitability test.

In addition to the priorities common to all GSI projects mentioned above, those specific to reforestation might include high Land Surface Temperature (LST), low canopy cover, and high EJScreen parameters. The following maps show different combinations of these parameters. Click through the maps below to explore the parameters and results of each reforestation suitability test.