Understanding tropical cyclone (TC) risks in both the present and future climate provides valuable information for stakeholders to inform local policy and preparedness measures.  Two popular tools currently available for studying TC climatology are high-resolution global climate models (GCMs; which directly simulate TCs that can be tracked in model output) and statistical-dynamical downscaling (SDD) models (which utilize a model’s large-scale climatology to generate synthetic storms). While there is overlap between the two, scientists typically apply a single tool, which may lead to inconsistent research findings and low confidence in certain aspects of future projections of TC climatology, such as frequency. 

Here, we employ these tools together to compare how landfalling TCs are represented. We aim to answer two key questions: 1) If GCMs contain climatological biases in TC activity, can we determine if these factors are due to insufficient resolution, parameterization biases, or errors in how these models simulate the large-scale environment? 2) Are TC statistics generated by SDD methods providing similar answers to those gleaned from direct storm tracking in GCM data? To assess this, we analyze simulation data from High-Resolution Model Intercomparison Project (HighResMIP) models. We take TCs tracked by TempestExtremes and compare them to observed landfalls using both IBTrACS and storm tracks objectively defined in reanalyses. We also leverage the SDD TC model described in Lin et al. (2023) to generate a parallel set of tracks derived from daily and monthly HighResMIP model fields. We examine several climatological metrics, storm track patterns, genesis and landfall points, and translation direction for both sets of tracks as well as seed conversion rates for each HighResMIP product. We also examine how higher temporal frequency data may help to improve TC representation in the SDD TC model.