Identification of Hazardous Weather for Offshore Forecasting

Forecast offices responsible for monitoring weather conditions over open water rely on antiquated forecasting tools. Geostationary (GEO) imagery and numerical weather prediction (NWP) analyses are the primary information sources for forecasters when outside of radar range. There is a lack of products specifically developed and designed for offshore operational forecasting of mesoscale and synoptic scale meteorological events. A new analysis tool is under development to identify developing and sustained weather systems that have the potential to produce navigational hazards.

This project is analogous to the ProbSevere model, developed by CIMSS, which calculates the likelihood that a radar-identified precipitation cell will produce severe weather (wind speed > 50 knots or hail > 1 inch) in the next 60 minutes. ProbSevere collocates precipitating cells with the glaciation rate and growth rate from GOES IR imagery, the maximum expected size of hail (MESH) from radar, and the effective bulk shear and convective available potential energy (CAPE) from NWP. In the future, lightning intensity will also be a predictor. A Bayesian model compares the observed conditions to previous storms and severe storm reports to calculate the probability of the storm becoming severe. This strategy will be modified for forecasting applications over the ocean, where there is no radar data available. Low earth orbit (LEO) data can supplement the GEO and NWP predictors to measure characteristics of long-lived systems.

The introduction of GOES-R will dramatically increase the amount of GEO data, and require intelligent data processing to deliver the most important information to forecasters. Rather than calculating the probability of meeting severe weather criteria, the probability of a storm producing gale-forced winds will be computed.