Nucaps in AWIPS: Rethinking Information Compression and Distribution for Fast Decision Making

NUCAPS–the NOAA Unique Combined Atmospheric Processing System for MetOp, Suomi NPP, and JPSS–was released in AWIPS-II in 2014. This marked the first time that forecasters had operational access to satellite sounding products, which are remote measurements of the 3-dimensional (3-D) atmospheric structure defined by temperature and moisture profiles. Previously forecasters were limited to interrogating single radiosondes, which are in situ measurements of vertical structure. Satellite soundings differ from radiosondes in the degree of vertical variability they measure. Satellite soundings are retrieved from top-of-atmosphere infrared and microwave radiances and measure the atmospheric state in broad layers ranging from 1-4 km depending on atmospheric pressure. Their profiles are, therefore, smoother that what is seen in radiosondes. This said, satellite soundings offer an advantage over radiosondes in their temporal and spatial frequency. Per satellite platform, NUCAPS soundings are available twice a day, every day for as long as the satellite is operational. All platforms combined, there are at present an excess of 6 NUCAPS soundings per area, per day. NUCAPS from Suomi NPP and JPSS (in future) measure the early afternoon (~1:30 pm local time) pre-convective environment, which is particularly valuable to forecasters.

After three years of evaluation at the hazardous weather testbed (HWT) the wide swaths of NUCAPS profiles in AWIPS are valued for being (i) model independent–NUCAPS has a regression first guess, (ii) promoting situational awareness and, (iii) enhancing mesoscale analysis. With both spatial and vertical information available, visualizing a single NUCAPS 100-layer sounding is not always the most useful option. Recent efforts demonstrated the information gained from visualizing gridded 2-D plan views of NUCAPS soundings on different pressure layers, e.g., maps of mid-level moisture that display complex processes as gradients over large areas. Interrogating individual profiles or pressure layers in AWIPS, however, is not always feasible because it can be time consuming to find the right information for a specific scenario–there are hundreds of profiles per satellite swath, and 100 layers per profile. As it stands, forecasters rarely have time for exploration in a real-time operational environment. This team aims to design and test different information metrics (as means of data compression) that can be delivered through direct broadcast pathways (most notably the Community Satellite Processing Package; CSPP) with the goal to significantly improve information access and efficiency. We envisage these compressed, low latency NUCAPS products to be part of a “dashboard” display that alert forecaster to areas of interest, such as rapid change, steep gradients or disagreement between observations and models. Forecasters can then save their deep-dive evaluation of the 3-D atmosphere for target areas that are most relevant to their problem-solving scenario at hand. This work is in direct response to forecaster feedback at the HWT as they expressed a need for improved, low latency decision-making tools of the observed atmospheric state.