Impact of satellite multi-sensor and in-situ data on high-resolution numerical simulation of the rapid intensification of Hurricane Dennis (2005)

Forecasting of the rapid intensification of hurricanes is a challenging problem. Previous studies show that numerical models commonly have problems in capturing the observed intensification rate of tropical cyclones, thus failing to produce accurate intensity forecasts for tropical cyclones during their evolutions. In addition to the imperfect physical parameterization schemes in the numerical models, uncertainties in the initial vortex representation and environmental conditions are also major sources of forecast errors.

In this study, we examine the impact of satellite and in-situ data on high-resolution numerical simulation of rapid intensification of Hurricane Dennis (2005). Specifically, the QuickSCAT ocean surface wind vectors, Aqua/AIRS temperature and moisture profiles, and aircraft dropsondes data, collected during NASA Tropical Cloud Systems and Processes (TCSP) field experiment, are assimilated into the Weather Research and Forecasting (WRF) model with its 3-dimensional variational data assimilation system in a rapidly updated cycle. The influences of each data type on the forecast of rapid storm intensification, structure and environmental conditions are investigated. In addition, numerical results are compared with the data collected from TCSP field program in order to further examine why data assimilation could result in the improved forecast and by which mechanism the improved initial conditions improve the high-resolution numerical simulation of Hurricane Dennis.