Development of Merged Cloud Forecasts from Satellite and Numerical Model Data

The Naval Research Laboratory (NRL) is developing a cloud nowcasting system consisting of merged data from the Coupled Ocean-Atmosphere Mesoscale Prediction System (COAMPS) and GOES 16. Predictors are drawn from COAMPS microphysical, thermodynamic, kinematic, and synthetic brightness temperature forecasts at 5 km grid spacing. Ground truth observations consist of GOES 16 reflectance and brightness temperature observations as well as retrievals of cloud top height (CTH), total condensed water path (CWP), cloud top phase, and cloud type.

Isolating relevant sets of predictors for the cloud forecasts is particularly challenging due to the many types of clouds. Since each type may require very different sets of predictors the cloud observations must be adequately characterized. Thus, current work has focused on the cloud classification issue. Although the existing NRL cloud classification scheme performed well, low-topped ice clouds were often misclassified as cirrus. Augmenting with the CTH and CWP retrievals enabled the sorting of thinner cirrus clouds with biased CTH values. Remaining clouds were then reclassified as low ice. Layered overlapping clouds remain a challenge. COAMPS thermodynamic and moisture fields have shown promise in determining the presence of cirrus over convective cumulus versus deeper, more stratiform clouds.