Automating measurement of Prairie Pothole Region wetland ponding
This project developed an open-source machine learning model for estimating high spatial resolution wetland surface water area in the Prairie Pothole Region (PPR). This model relies on freely available Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 10-m surface reflectance satellite data. We developed a multi-sensor algorithm that converts these products to analysis-ready image data and then uses those data to map multi-temporal changes in wetland surface water area, even below vegetation canopies.
We are sharing this model workflow as code that can be run by end-users on Google Earth Engine, allowing them to generate their own predictions (see example images right). These wetland surface water predictions are useful for conservation planning as they are freely available, high resolution (spatial resolution: 10-m, temporal resolution: 2 weeks), applicable across the entire PPR, and applicable across the entire waterfowl breeding season (May- August). While one or two of these attributes have previously been addressed at once, addressing all together in one workflow deals with a large number of the challenges that conservation managers have faced when attempting to use remote sensing for wetlands in the PPR.
A slide deck documenting the current state of project development is found here.
How to use this project:
See our How-to Guide for code links and detailed instructions for end-users.
A link to a video with a code walk-through for the preliminary project products is here
This is an open-science project. You may use and continue development of our project code as you like, but please share your changes and source code back with the scientific and land management community. This project is under active development. We will update this page as the project changes.
This project is licensed under the GNU General Public License v3.0. Permissions of this strong copyleft license are conditioned on making available complete source code of licensed works and modifications, which include larger works using a licensed work, under the same license. Copyright and license notices must be preserved. Contributors provide an express grant of patent rights.