A forest dryness monitoring tool to assess dynamics of forest dryness in the western US was developed on AWS by a Stanford University Ph.D. Student, Krishna Rao, using a combination of the Sentinel-1 and Landsat-8 satellite imagery from the AWS Open Data Registry and a trained four-layered neural network based on a long-short term memory (LSTM) model. The deep learning models utilized multiple GPU instances on AWS to cut down on training and modeling time. On cross-validation, Rao found that the landscape-scale forest dryness mapping effort was as accurate as other smaller scale analyses. The vast amounts of ground data available for model training, combined with the modifications performed to the LSTM model based on domain knowledge yielded this result. Further, the availability of satellite data and large computational power on the cloud accelerated model development.
Learn more about the use case on the AWS Public Sector Blog.