A Physically-based Tropical Cyclone Rainfall Model

Abstract:

Rainfall from tropical cyclones (TCs) can cause extreme flooding. Predicting and understanding TC rainfall is thus important but has received relatively less attention, compared to the wind and surge. Here we present a simple, physically-based rainfall model, where the rain rate is obtained from estimated vertical velocity and specific humidity in the lower troposphere. The involved rainfall mechanisms include: 1) vertical motion at the top of the boundary layer owing to frictional effects; 2) vertical motion in the middle troposphere resulted from the time evolution of the gradient wind; 3) vertical motion forced by topographic interaction as well as 4) baroclinic effect. The model has been applied to Texas and shown to generate rainfall statistics comparable to observations (Zhu et al, 2013). Here we further evaluate this model on an event basis; case studies include Hurricane Irene (2011) and Isabel (2003). Without any calibration, hourly rainfall estimated from this model compares well with those from full numerical weather prediction model (WRF) as well as rainfall climatology models (R-CLIPPER and PHRaM). This comparison demonstrates the model’s ability to capture main TC rainfall mechanisms, and it can be used as an effective tool to study the relative contribution of each rainfall mechanism. Ongoing work includes possibly improving the rainfall model by coupling it with a more accurate boundary layer model. Given its high computational efficiency, this rainfall model can be applied to large numbers of ensemble or synthetic simulations. This study fits into our long-term goal to quantify the risk of inland flooding associated with landfalling TCs.