GOES-R IFR probabilities and Tropical desk IFR forecasts

One of the most challenging parts of the Tropical desk is verification. Given the data sparse regions in which these forecasts are issued, it is often difficult to find a sufficient method by which to verify the various forecasts. An earlier post described this challenge with forecast winds and how ASCAT winds have been explored as a way to verify those forecasts. However, another challenging forecast to verify is for IFR and MVFR. Yesterday, we explored a comparison of the GOES-R IFR probabilities in the tropics to the Tropical desk forcast for IFR conditions.

20150820 18Zf12 Tropical desk forecast with the IFR forecast in red and MVFR in scalloped blue. The GOES-R IFR probabilities are underlaid. Note the domain of the latter ends around ~10N

Note in particular the blue scalloped area over Haiti indicating forecast MVFR, and within that, the red area indicating forecast IFR conditions. The IFR probability had high probabilities in this area at 0615 UTC. It also was devoid of IFR probabilities anywhere else in the tropical domain. In this case, the GOES-R IFR probabilities became a very useful verification tool. This concept will be further explored in the future.