The impact of future climate change on TC intensity and structure: A downscaling study

In this study a downscaling approach is used to investigate changes in TC intensity and structure due to global warming, with several advancements from some similar previous studies. Projected changes in atmospheric temperature and moisture, and SST are computed using 13 atmosphere ocean general circulation models (AOGCMs) from 3 different emissions scenarios (A1B, A2, and B1), allowing for analysis of the sensitivity of the results to AOGCM and emissions scenario choice. Current climate conditions are derived from reanalysis data instead of AOGCM output, providing a more realistic depiction of the conditions within which TCs form in the present-day climate. Finally, TCs are simulated at very high resolution (6 or 2-km grid spacing), allowing for the omission of cumulus parameterization (CP) and more realistic TC structure. The high-resolution simulations allow for detailed analysis of TC structure changes in the future climate and how these changes impact TC intensity.

Specifically, thirty-nine separate large-scale future environments were constructed utilizing a sum of the current climate conditions and individual AOGCM-projected changes, which were calculated as the 2089-2099 mean minus the 1989-1999 mean (from the climate of the 20th century AOGCM simulations). Also, 3 large-scale environments were derived utilizing current climate conditions and ensemble mean AOGCM projected changes, which were averaged over the 13 separate AOGCMs from each emissions scenario. An initial TC vortex was inserted into each large-scale environment, and simulations with 6-km grid spacing were performed using the 39 large-scale environments, and with a nested 2-km domain for the 3 large-scale environments derived using ensemble-mean AOGCM projected changes. All model simulations were performed using version 2.2 of the Weather Research and Forecasting (WRF) model.

AOGCMs projections from the A2 emissions scenario indicate the largest increase in SST and tropospheric temperature and moisture, although significant model-to-model variability is present. Tropospheric warming increases with height typically up to the tropopause, as the tropical troposphere remains convectively neutral and tropospheric lapse rates decrease due to increased water vapor content. The largest model variability is found in the upper troposphere near the tropopause, where differences in vertical resolution, model physics, and ozone treatment likely impact projected temperature changes.

Simulations with 6-km grid spacing indicate an increase in maximum TC intensity in 38 of the 39 simulations relative to the control. Increases in central pressure deficit varied between 0 and 22%, indicating the sensitivity to AOGCM choice. Averaged over the 13 individual AOGCMs, increases in central pressure deficit of 10, 11, and 3% were found in the A1B, A2, and B1 emissions scenarios, respectively, comparable to previous studies. Simulations with 2-km grid spacing produce more intense TCs than simulations with 6-km grid spacing, but generally similar increases in future TC intensity. The high-resolution simulations produce increases in area-averaged rainfall of up to 27% with the largest increases found in the vicinity of the eyewall, consistent with projections of increased tropospheric water vapor and highlighting the potential for an increased freshwater flooding threat in future TCs. Structural changes are also apparent, with an increase in the height of the freezing level and the TC warm-core, and outflow extending to a higher altitude. Although the outflow extends to a higher altitude in future simulations it is generally warmer due to tropospheric stabilization, leading to less intensification than would occur due to SST increases alone.