Global Synthetic Weather Radar in AWS GovCloud for the U.S. Air Force

The United States Air Force operates manned and unmanned flight missions all over the world. Accurate weather information in both the pre-flight planning and execution portion of flight is very important for mission success. The US Air Force is partnering with MIT Lincoln Laboratory to develop a machine learning application that generates global radar-like mosaics and radar-forward forecasts in support of US Air Force flight operations. This capability will utilize a number of data sources, including global lightning, the Global Air-Land Weather Exploitation Model (GALWEM) numerical model, and global weather satellite images as input. The capability fuses these data sources together with a convolutional neural network that creates global synthetic weather radar mosaics, and these mosaics are used in conjunction with GALWEM to produce radar-forward forecasts out to 12 hours.

This capability is currently being developed for and transitioned to the Amazon Web Service (AWS) GovCloud for real-time testing, evaluation, and eventually operations. Execution in the AWS GovCloud will leverage the large compute and data storage resources available in the cloud and enable broad access to the output for use in Air Force decision support systems, such as the Weather Common Component (WxCC) (via machine-to-machine web services) and the Air Force Weather-Web Services (AFW-WEBS) viewer. From the US Air Force perspective, this effort represents a pathfinder to transition a relatively mature machine learning-based system to the AWS GovCloud. This project is comprised of four major work areas: 1) initial capability development, 2) instantiation and further development in AWS GovCloud, 3) training and user feedback, and 4) consideration of machine learning “best practices” to ensure future sustainability of the capability after transfer to the US Air Force. This presentation will provide results and perspectives on all four of these work areas.

DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited.

This material is based upon work supported by the Department of the Air Force under Air Force Contract No. FA8702-15-D-0001. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Department of the Air Force.