Ting-Chi Wu, Colorado State University
Practical details of assimilating GPM/TRMM hydrometeor retrievals with WRF model
Coauthors
Milija ZupanskiAbstract:
In this poster we describe several important details regarding the assimilation of NASA Goddard PROFiling (GPROF) hydrometeor retrievals from the Tropical Rainfall Measuring Mission (TRMM) and Global Precipitation Measurement (GPM), customized for tropical cyclone data assimilation. This newly developed variation of the GPROF algorithm that relies upon a hurricane-specific a-priori database, including location and spatial extent of hurricanes/tropical cyclones, is called Hurricane GPROF. To assimilate the Hurricane GPROF retrieved integrated solid- and liquid-water contents, new observation operators are developed and implemented in data assimilation.
We will describe in detail the observation operators that are used to assimilate Hurricane GPROF integrated water contents, and motivation for their development. The development of such observation operator is closely related to capability of data assimilation and prediction model systems and the related choice of control variables in data assimilation. For example, the availability/complexity of cloud microphysics in the prediction model significantly influences the potential choice of control variable such as cloud hydrometeors. The prediction model employed in this study is the Weather Research and Forecast (WRF) model. On technical side, the inclusion of such new control variables may be demanding for some data assimilation algorithms. On scientific side, the change of hydrometeor initial conditions has to be supported by an adequate change of variables such as temperature, pressure, moisture and winds. This, however, strongly depends on complexity of the forecast error covariance and ultimately on data assimilation methodology. We will address the choice of control variable in assimilation of Hurricane GPROF integrated water contents observations with respect to variational, ensemble, and hybrid variational-ensemble data assimilation methodologies.