Learning Architecture for Short-Term Forecasts of Cloud Characteristics Using Satellites (LAST FOCUS)

The Learning Architecture for Short-Term Forecasts of Cloud Characteristics Using Satellites (LAST FOCUS) introduces a novel approach and state-of-the-art ML methodologies to forecast cloud properties includingĀcloud top, optical depth, phases, and reflectance.Ā Using a suite of cloud-observing geostationary and Low-Earth Orbiting satellites, atmospheric profiles, and an extensible ML ecosystem for weather modeling, LAST FOCUS produces outputs for aviation operations.Ā LAST FOCUS is interoperable with and extends existing capabilities including Navy Research LaboratoryÆs (NRL) Geo-located Information Processing System (GeoIPS) and the Air ForceÆs Weather Machine Learning Platform (WxMLP). LAST FOCUS consists of four main modules.Ā The modules are hosted within the LAST FOCUS ecosystem and includes robust training and validation processes.Ā A family of microservices provides access to heterogeneous data sources and the means to curate and align them into a canonical form and format.Ā This will feed the other three modules:Āone generates cloud mask analysis and nowcasts, one that nowcasts cloud characteristics, and a third that gleans truth used to train ML algorithms.ĀThe modules and supportive microservices are orchestrated by NextGen's commercial ML platform, StratusMLÖ, the basis of the Air Force Weather (AFW) Weather ML Platform (WxMLP).Ā The resulting ML models are used to generate analyses and nowcasts of cloud characteristics.