Relationship between a spatial pattern of future atmospheric warming and Asian dust emission

Climate model projections show that atmospheric temperatures significantly increase in the lower troposphere at high latitudes of the Northern Hemisphere and the upper troposphere at low-middle latitudes, leading to a decrease in the meridional temperature gradients in the lower troposphere and an increase in atmospheric stability at low-middle latitudes. This is "a spatial pattern of future atmospheric warming".

Asian dust emission is frequently induced by cold air outbreaks at the high latitude areas. This indicates that the significant atmospheric warming in the lower troposphere at the high latitudes is expected to influence the dust emission. Also, the atmospheric stability is closely associated with the dust emission according to previous studies. This study, therefore, investigated potential impacts of the spatial pattern of future atmospheric warming on Asian dust emission mainly by numerical experiments with a dynamical downscaling method.

To obtain the spatial pattern of future atmospheric warming from a global-scale climate model, March and April monthly mean data from the Model for Interdisciplinary Research On Climate (MIROC) version 3.2-hires were analyzed for making the “Global Warming Monthly mean Differences (GWMDs)" in atmospheric temperatures, geopotential heights, and u-v wind components. The GWMD was defined as the difference between the 10-year average of the 21st century climate projections with the SRES-A1B scenario in the period from 2091 to 2100 (future climate) and that of the 20th century climate simulations from 1991 to 2000 (recent climate).

The Weather Research and Forecasting (WRF) modeling system version 3.1 was used to carry out numerical experiments on the regional-scale emission, advection, dispersion, and deposition of Asian dust. The PM10 soil dust in the WRF-Chem model was assumed to be Asian dust particle. Using this modeling system, we simulated a significant Asian dust event, occurred during the period from March 29 to April 2, 2007, with the 6 hourly NCEP-FNL data and the 40 km horizontal resolution in order to evaluate the spatial and quantitative reproducibility of the model for dust simulations by comparing the simulation results with observations in the Japanese Islands, which is located at far from the Asian dust sources.

After the reproducibility evaluation of the model, the "pseudo global warming experiments" were conducted as the main experiments in this study. First, we picked three significant events of Asian dust outbreaks and then simulated those events by using the WRF (control runs: CTL-runs). Second, the 50E-130E zonal mean GWMDs at latitudes between 20N and 80N were added to the 6 hourly NCEP-FNL to update the initial and boundary conditions of the CTL-runs. The WRF model simulations was thus given the spatial pattern of future atmospheric warming obtained from the MIROC. Third, pseudo global warming runs (PGW-runs) were executed with the updated initial and boundary conditions. Finally, future changes in Asian dust emission were analyzed mainly by calculating the differences between results of the PGW-runs and those of the CTL-runs.

Results of the pseudo global warming experiments showed that the future emission mass flux of the soil dust in the Taklimakan Desert and the Gobi Desert, i.e., the main Asian dust sources, decreases, compared with the recent emission flux. Results of the experiments also showed the future decreases in the column-integrated soil dust mass concentrations in the atmosphere in the main dust sources, which correspond to the emission flux reduction.

Analyses of results of the pseudo global warming experiments indicated that the main cause of the emission flux reduction is the future decreases in sea level pressure (SLP) gradients between the main Asian dust sources and the north of the dust sources. This can reduce velocities of surface winds blowing toward the dust sources, resulting in the reduction in the soil dust emission flux. The large increases in SLPs in the dust sources and the relatively small increases in SLPs in the north of the dust sources led to the decreases in SLP gradients between the two areas.

The small increases in SLPs in the north of the main dust sources appear to correspond to the large increases in surface air temperatures in the western part of Mongolia, in the vicinity of the Dzungaria Basin, and near Lake Baikal where cold air masses are accumulated and released toward the surrounding areas including the dust sources. Whereas, the large increases in SLPs tend to correspond to the small increases in surface air temperatures.

The results of this study support many previous studies that indicated a negative correlation between Asian dust outbreaks and atmospheric temperatures. The previous studies showed the frequency of Asian dust outbreaks has declined over the past 30-50 years, coinciding with the recent global warming.