Verification of global and mesoscale cloud forecasts over the eastern Pacific

Advancements in microphysical parameterizations combined with increases in model resolution have enabled the production of increasingly detailed cloud forecasts. These forecasts are uniquely challenging to verify due to uncertainties in cloud top height estimation and limitations of conventional satellite observing systems to observe cloud vertical structure. However, preliminary CloudSat measurements indicate that observational errors are reduced for certain deep cloud types. Retrievals of multiple properties, including liquid water path, cloud type, and cloud top temperature, can be used along with CloudSat measurements to characterize uncertainty in both the observations and the forecasts. A cloud verification scheme is presented whereby statistics ranging from simple bias and threat scores to advanced spatial sampling methods are used to estimate uncertainty. Global and mesoscale model cloud forecasts over the eastern Pacific are compared to GOES and CloudSat measurements.