Evaluating Rainfall Sensitivity over Land to Tropical Cyclone Parameters Using Idealized WRF

Tropical cyclones (TCs) are extreme weather events with devastating impacts that affect coastal communities worldwide. In the U.S. alone, TCs account for 8 of the top 10 costliest natural disasters to date, and it is expected that TC frequency and intensity will increase due to the influence of climate change. Apart from the extreme ocean responses produced by TCs, these storms are also capable of creating extreme rainfall that often exacerbates other coastal hazards such as storm surge, and the coincidence of these hazards is commonly referred to as compound flooding. For agencies such as the Federal Emergency Management Agency (FEMA) and the U.S. Army Corps of Engineers (USACE), the assessment of compound flooding hazards with respect to TCs is an important area of research for supporting various objectives such as inundation mapping and engineering design. Although TCs play a major role in producing both coastal and compound hazards, successfully evaluating the hazards induced by these storm events is difficult when relying on observations within historical datasets alone. For this reason, hazard analysis techniques, such as the Joint Probability Method (JPM), generally rely on the development and simulation of synthetic TC events as the basis for evaluating physical responses such as storm surge and rainfall.

As part of the Coastal Hazards System (CHS) (https://chs.erdc.dren.mil), the USACE has developed more than 4,300 synthetic TCs that cover the probability space from very frequent to very rare storms for TCs impacting the Gulf of Mexico and Atlantic coastlines. Based on the JPM, each storm is defined as a combination of the primary TC parameters, which include central pressure deficit (Δp), radius of maximum winds (R_max), translational speed (V_t), and heading (θ). These synthetic storms have been developed with the purpose of characterizing the physical atmospheric-forcing parameters that drive coastal storm responses (i.e., storm surge and waves) to support USACE coastal hazards studies. As part of these studies, hundreds of synthetic TCs are created for a given study region and simulated in hydrodynamic models for the purpose of developing hazard curves that describe the full probability space of coastal storm responses. Extensive research has been conducted to evaluate these synthetic TC events for coastal hazards, but a gap exists in evaluating these storms within rainfall models to support compound flooding analyses.

For this reason, the intent of this work is to evaluate the ability of synthetic TCs developed through the JPM to represent rainfall by conducting idealized TC simulations using the Advanced Research version of the Weather Research and Forecasting (WRF-ARW) model. This work applies WRF to obtain representative over-land rainfall estimates 1) to understand the dependence of rainfall with respect to varying synthetic TC parameters, 2) to evaluate the ability of these surge-focused synthetic TCs to represent the probability space of rainfall estimates needed for compound flooding hazard analysis, 3) and to assess the potential development of surrogate models that leverage the fidelity of WRF to more efficiently evaluate rainfall for larger suites of synthetic TC events. The results of this work are intended to address how available synthetic coastal storm databases can be leveraged to model rainfall and evaluate the applicability of WRF in compound flooding studies conducted by the USACE.