Texas has experienced several devastating TCs in the 21st century, including Harvey, Ike, and Allison, causing billions of dollars in damage. Harvey and Allison caused extensive flooding due to record rainfall, while Ike's damage stemmed from strong winds and storm surge. The varying damage characteristics highlight the importance of TC movement (intensity, size, landfall angle, translation speed, and sea level) in determining the severity of impact. Climate change influences these factors through higher sea-surface temperatures (SST), increased atmospheric moisture, and shifts in large-scale circulation. While previous research has explored changes in TC translation speed, findings remain inconclusive. This study addresses the specific question of how climate change will influence the movement, particularly the translation speed, of future landfalling Texas TCs, using large-ensemble/multi-model datasets, clustering, and downscaling techniques to minimize uncertainties related to natural variability and model biases.

Numerous studies have examined changes in TC frequency and intensity in relation to climate change. However, investigations into changes in TC translation speed at the global or basin-wide scale have yielded inconsistent results. Some studies suggest a global slowdown of TC translation speed, while others challenge these findings, emphasizing the importance of data homogeneity and regional variations. This research builds upon these previous efforts by focusing on the specific region of Texas and employing a multi-faceted approach to reduce uncertainties associated with natural climate variability and model bias.

The study employs a multi-pronged approach using three large-ensemble datasets (LENS, MPI-GE, GFDL-LE) and data from 14 Coupled Model Intercomparison Project Phase 5 (CMIP5) models. First, the researchers analyze changes in June-September averaged steering winds and geopotential height at 500 mb (Z500). Next, they utilize a self-organizing map (SOM) cluster analysis on daily steering wind vectors from LENS to identify dominant wind regimes and their changes in the future climate. Finally, downscaling experiments using the Columbia TC HAZard model (CHAZ) are conducted with six CMIP5 models to assess changes in the translation speed of synthetic TCs making landfall near Houston. The CHAZ model considers various environmental variables such as potential intensity (PI), mid-to-low level moisture, and low-level vorticity, in addition to steering winds. Bias correction methods, using Gaussian and quantile-matching techniques, are applied to the CHAZ model outputs to account for model discrepancies with historical observations. Statistical significance is assessed using a two-tailed t-test at a 95% confidence level.

Across all models, the study consistently reveals a robust increase in northward steering winds over Texas during June-September. This is associated with an intensification and westward shift of the Atlantic subtropical high and a weakening American monsoon. The cluster analysis reveals an increase in the frequency of daily wind regimes with northward steering winds and a decrease in regimes with southward winds over Texas in the future. The downscaling experiments using CHAZ demonstrate a significant shift in the translation speed distribution toward faster-moving TCs. The relative probability of TCs with speeds ≥20 km h⁻¹ increases by approximately 6%, while the probability of TCs with speeds ≤5 km h⁻¹ decreases by approximately 4%. This 10-percentage-point shift toward faster-moving storms remains consistent even after applying bias correction to the CHAZ model outputs.

The findings indicate a robust increase in the likelihood of faster-moving landfalling Texas TCs in the late 21st century, primarily driven by enhanced northward steering winds. This northward steering wind response appears linked to changes in the Atlantic subtropical high and the American monsoon. These results contrast with some previous studies that reported a global slowdown of TCs. However, the differences can be attributed to variations in the study region, time period, and methodologies. The observed changes in steering winds are regionally specific, highlighting the necessity of regional analyses to understand climate change effects on TCs. The increased probability of fast-moving TCs doesn't necessarily imply a reduced risk, as fast-moving storms like Ike can still cause significant damage through strong winds and storm surge. Future research should consider the diverse damage mechanisms of fast and slow-moving storms to better inform adaptation and mitigation strategies.

This study provides strong evidence for an increased likelihood of faster-moving landfalling Texas TCs by the late 21st century, driven by changes in atmospheric circulation patterns. The multi-model, multi-faceted approach used effectively reduced uncertainties associated with natural variability and model biases. Further research should expand on this work by investigating the mid-21st century and alternative emission scenarios (RCP4.5), and delve deeper into the relationships between changing TC dynamics and thermodynamics to further refine risk assessments.

The study primarily focuses on changes in the late 21st century under the high-emission scenario RCP8.5. While bias correction methods were applied, the downscaling model (CHAZ) still shows some discrepancies compared to historical observations. The analysis assumes that TCs approach the Texas coastline with equal probability across different wind regimes, a simplification that may not fully capture the complexities of TC genesis and movement. Further, the study mainly focusses on the speed of translation near landfall.