BRAT Vegetation Suitability Parameterization

This page provides instructions on using a database that contains LANDFIRE vegetation class frequency within the valley bottom across the US to check their BRAT suitability classification as a first step for parameterization prior to running BRAT for a given region. The database is available in the LANDFIRE EVT Class Frequency Database.

Step 1: Download the LANDFIRE EVT Class Frequency Database

Download the database from the LANDFIRE EVT Class Frequency Database and save it to your local machine.

Step 2: Use the database to find the dominant vegetation classes within the valley bottoms of a given region (e.g., a HUC4 watershed)

Open the database and connect to it in SQLiteStudio. Open an SQL editor page, and run the following query to find the dominant vegetation classes within the valley bottoms of a given region. In this example, we are using the HUC4 watershed code 1006. The 'hucs' table in the database has all of the HUC10s, in the US, so to look at data for a HUC4, we use the WHERE SUBSTR(huc10, 0, 5) = '1006' line in the query.:

SELECT VegetationID,
       area / total_area proportion
  FROM (
           SELECT VegetationID,
                  area,
                  (
                      SELECT SUM(area) 
                        FROM (
                                 SELECT VegetationID,
                                        sum(VegArea) / 1000000 area
                                   FROM veg_summary
                                  WHERE substr(huc10, 0, 5) = '1006'
                                  GROUP BY VegetationID
                             )
                  )
                  AS total_area
             FROM (
                      SELECT VegetationID,
                             sum(VegArea) / 1000000 area
                        FROM veg_summary
                       WHERE substr(huc10, 0, 5) = '1006'
                       GROUP BY VegetationID
                  )
       ) order by proportion desc;

Running this query returns the dominant vegetation classes within the valley bottoms of the HUC4 watershed 1006 and their proportion of the total area. The VegetationID corresponds to the vegetation class ID in the LANDFIRE EVT class table. The proportion column shows the proportion of the total area that each vegetation class occupies. This information can be used to check the BRAT suitability classifications for the vegetation classes in the riverscapes-tools repository. The query above returns the following results (only the first few rows shown):

VegetationID	proportion
1284	0.3563
7141	0.1539
7968	0.1143
1287	0.0790
9516	0.0423

Step 3: Check the BRAT suitability classification for the vegetation classes

Within the riverscapes-tools repository, open the BRAT Workspace. Make sure you are working on a branch to make changes (e.g., git checkout -b brat_1006_veg). Navigate to the data directory, and open the VegetationTypes.csv file. This file contains the BRAT suitability classification for each vegetation class. Check the BRAT suitability classification for the vegetation classes identified in the previous step that have a proportion above a selected threshold (e.g., 1%). In this example, this results in 12 vegetation classes to check (you can use Ctrl + F to search for the VegetationID in the VegetationTypes.csv file). Based on the name of the class, ensure that the BRAT suitability classification is appropriate for the vegetation class. If the classification is not appropriate, assign a new classification based on the BRAT suitability classification guidelines. LANDFIRE hosts tables here for 'EVT' (existing vegetation cover) and here for BpS (modeled historic vegetation cover) containing all of the attributes for each vegetation class that can be used to help determine the suitability classification. Commit any changes to the branch, push them, and create a pull request to merge the changes into the main branch.

The video below demonstrates the steps outlined above: