The sensitivity of pre-genesis forecasts of Hurricanes Danielle and Karl (2010) to pouch and environmental initial conditions

One of the sub-hypotheses of the PRE-Depression Investigation of Cloud-systems in the Tropics (PREDICT) experiment is that tropical cyclone genesis forecasts produced by numerical weather prediction models are limited by poor initial conditions, which amplify over time due to the chaotic dynamics of the atmosphere. As a consequence, it is important to understand where and how initial condition errors grow the fastest and why some cases of TC genesis are more predictable than others. Previous studies have suggested that the rate of genesis depends on the specification of the tropical wave and the environment in which the system moves through including vertical wind shear and moisture, thus one might hypothesize that initial condition errors associated with these might be associated with subsequent forecast errors.

This study evaluates the predictability of pre-genesis stage of Hurricanes Danielle and Karl (2010) using the ensemble-based sensitivity technique applied to ensemble forecasts generated with the Advanced Research version of the Weather Research and Forecasting (WRF) model. This ensemble is initialized from analyses produced by a continuously cycling ensemble Kalman filter data assimilation system, which ran in real-time during the PREDICT experiment. Preliminary results suggest that the intensity forecast of both systems is sensitive to coherent vertical variations in the initial-time circulation and moisture profiles within the pre-genesis system. Whereas the pre-Danielle forecast is also sensitive to the time-integrated vertical wind shear magnitude, the pre-Karl forecast is more sensitive to the time-integrated upshear mid-tropospheric moisture.