Characteristics of Tropical Cyclones Emerging from Random Winds on the Periodic F-Plane in a Cloud-System-Resolving Numerical Model

This study examines select statistics of tropical cyclones (TCs) that emerge from random winds on the f-plane in a cloud-system-resolving (CSR) numerical model with warm-rain microphysics and longwave radiation. The simulations are run for 9-18 days with 3.9 km horizontal grid-spacing in a 4000×4000 km² doubly-periodic domain. Within the simulation set, the sea-surface temperature T_s varies between 26 ^oC and 32 ^oC, and the Coriolis parameter f varies between 10^-5 s^-1 and 10^-4 s^-1. The ratio of surface-exchange coefficients (for moist-entropy and momentum) is artificially fixed at unity. The domain-averaged atmospheric sounding is initially the same for all simulations, but its evolution due to radiative convective adjustment varies with T_s and f. Key parameters dependent on the sounding include V_th and χ, in which V_th is a theoretical estimate of the maximum potential wind speed of a TC, and χ is a dimensionless measure of the middle-tropospheric moist-entropy deficit with respect to saturated conditions. The number of TCs that develop and persist in a simulation is proportional to f ². At the time t=t_* of peak vortex intensity, the radius of maximum wind tends to increase with decreasing V_th (or T_s) or increasing f. For t>t_*, the maximum wind speed V_max of a simulated TC gradually decays while χ gradually grows. Late-time data from all simulations show a power-law decay of V_max/V_th with increasing χ (or t/t_*). The statistical decay of V_max/V_th with increasing χ coincides with a statistical enhancement of precipitation asymmetry within a TC. A subset of the data at a moderate value of χ further reveals a significant anticorrelation between V_max/V_th and a dimensionless measure of vortex tilt. The preceding results are preliminary. Sensitivity tests are underway using the same CSR model with different parameterizations and an independent CSR model. This work has been supported by NSF grants AGS-0750660 and AGS-1101713.