Catalog of stock ponds using machine learning
Description
There are many undocumented stock ponds that retain water for use by farmers and ranchers throughout the landscape in the Dakotas. The stock pond AI model could identify unknown stock ponds to water managers for use in contaminant monitoring or water budget analysis and drought mitigation.
Detailed example
The stock pond ML algorithm outputs predicted locations of stock ponds and dams in lidar images in jpg format and a csv file of locations given in latitude and longitude.
AI / analytics pattern
Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relationships.
Automation level / stage
b) Pilot – The use case has been deployed in a limited test or pilot capacity.
Expected benefit
Identifying stock ponds in North Dakota using an automated ML model rather than manually doing so increases efficiency at the state level that could free up labor for other tasks.
Controls / human review
ATO: No; PIA: Not published
Data needed
The model is trained on the publicly available lidar data disseminated by the North Dakota Department of Water Resources using the U-Net image segmentation architecture in PyTorch. The model is evaluated during training based on validation images through the use of RMSE. After the model development the output will be evaluated against known stock pond locations downloaded from the NDDWR and the U.S. Geological Survey National Hydrologic Data.