Sensitivity of high-temperature weather to initial soil moisture: a case study using the WRF model

Abstract. Using a succession of 24 h Weather Research and Forecasting model (WRF) simulations, we investigate the sensitivity to initial soil moisture of a short-range high-temperature weather event that occurred in late July 2003 in East China. The initial soil moisture (SMOIS) in the Noah land surface scheme is adjusted (relative to the control run, CTL) for four groups of simulations: DRY25 (−25%), DRY50 (−50%), WET25 (+25%) and WET50 (+50%). Ten 24 h integrations are performed in each group.

We focus on 2 m surface air temperature (SAT) greater than 35 °C (the threshold of "high-temperature" events in China) at 06:00 UTC (roughly 14:00 LT in the study domain) to analyse the occurrence of the high-temperature event. The 10-day mean results show that the 06:00 UTC SAT (SAT06) is sensitive to the SMOIS change; specifically, SAT06 exhibits an apparent increase with the SMOIS decrease (e.g. compared with CTL, DRY25 generally results in a 1 °C SAT06 increase over the land surface of East China), areas with 35 °C or higher SAT06 are the most affected, and the simulations are more sensitive to the SMOIS decrease than to the SMOIS increase, which suggests that hot weather can be amplified under low soil moisture conditions. Regarding the mechanism underlying the extremely high SAT06, sensible heat flux has been shown to directly heat the lower atmosphere, and latent heat flux has been found to be more sensitive to the SMOIS change, resulting in an overall increase in surface net radiation due to the increased greenhouse effect (e.g. with the SMOIS increase from DRY25 to CTL, the 10-day mean net radiation increases by 5 W m⁻²). Additionally, due to the unique and dynamic nature of the western Pacific subtropical high, negative feedback occurs between the regional atmospheric circulation and the air temperature in the lower atmosphere while positive feedback occurs in the mid-troposphere.

Using a method based on an analogous temperature relationship, a detailed analysis of the physical processes shows that for the SAT change, the SMOIS change affects diabatic processes (e.g. surface fluxes) more strongly than the adiabatic process of subsidence in the western Pacific subtropical high in the five groups of simulations. Interestingly, although diabatic processes dominate subsidence during the daytime and night-time separately, they do not necessarily dominate during the 24 h periods (e.g. they are dominant in the WET and CTL simulations only). Further, as the SMOIS decreases, the SAT06 increases, which is largely due to the reduced cooling effect of the diabatic processes, rather than the warming effect of subsidence.

Unlike previous studies on heatwave events at climate timescales, this paper presents the sensitivity of simulated short-term hot weather to initial soil moisture and emphasises the importance of appropriate soil moisture initialization when simulating hot weather.