- Reclamation
- R&D
- Research Projects
- Seasonal/Temporary Wetland/Floodplain Delineation using Remote Sensing and Deep Learning
Seasonal/Temporary Wetland/Floodplain Delineation using Remote Sensing and Deep Learning
Project ID: 1867
Principal Investigator: Zackary Leady
Research Topic: Ecosystem Needs
Funded Fiscal Years:
2018,
2019 and
2020
Keywords: None
Research Question
Can recent advancements in machine learning, specifically convolutional neural network architecture in deep learning, provide improved seasonal/temporary wetland/floodplain delineation (mapping) when high temporal and spatial resolution remote sensing data is available? If so, then these new mappings could inform the management of protected species and provide critical information to decision-makers during scenario analysis for operations and planning.
Need and Benefit
Currently, the delineation of seasonal wetlands is very limited and often inaccurate by traditional methods used by Reclamation. The end-product of a fully automated toolkit for accurate seasonal wetland delineation would provide an invaluable tool for Reclamation in monitoring operational effects and water for environmental use and would be applicable to every region. Additionally, a pertinent literature review will be conducted and the developed methodology will likely be the first of its kind available in an open source format. Essentially the benefit-cost ratio is exponential to Reclamation as this methodology will not only delineate seasonal wetlands, but in the future it is likely that similar versions of this methodology could be used to develop low-cost crop classification that is more accurate than the crop data layer.
The development of the delineation methodology and the automated toolkit addresses several critical research areas and categories. The ability to detect and quantify the size of these seasonal wetlands and temporary floodplains could serve as a low-cost method for quantifying in-stream habitat for aquatic species as outlined under Environmental Issues for River Habitat Rehabilitation. The delineation of seasonal/temporary wetlands/floodplains would also better inform planning models and river operation models as described in Water Operations Models and Decision Support Systems. Additionally, with multi-year analysis of seasonal wetland delineation the effects of operational decisions, water management, drought, and climate variability could all be tracked and effectively quantified based on per area productivity loss or gain. Currently, Mid-Pacific has gained free-access to Planet's proprietary constellation satellite data to use as the foundation for the analysis, if this project is not funded this data source may not be available in the future, free of charge.
Contributing Partners
Contact the Principal Investigator for information about partners.
Research Products
Bureau of Reclamation Review
The following documents were reviewed by experts in fields relating to this project's study and findings. The results were determined to be achieved using valid means.
Seasonal/Temporary Wetland/Floodplain Delineation using Remote Sensing and Deep Learning (final, PDF, 6.8MB)
By Vanessa King
Report completed on March 31, 2021
