PNNL

Abstract

East Asia dust has a significant impact on regional and global climate. In this study, we evaluate the spatial distributions and temporal variations of dust extinction and dust optical depth (DOD) by comparing model results from Community Earth System Model (CESM) with satellite retrievals from Luo et al. (2015a, 2015b), Yu et al. (2015), and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) level3 products. We find high model biases of dust extinction in the upper troposphere over the Taklamakan Desert, Gobi Desert, and Tibetan Plateau in MAM, JJA, and SON, especially in JJA by two orders of magnitude. Modeled dust extinction and DOD have its maximum in JJA due to low removal rate of dry and wet deposition or strong dust transport, while observations peak in MAM. In the Tibetan Plateau, low model biases of dust extinction in the middle troposphere in MAM and DJF are found due to the underestimation of local dust emission. We also access dust surface concentrations and 10 m wind speed by comparing their probability density functions with meteorological records from weather stations during dust events. CESM underestimates dust surface concentrations compared with observations due to the low model biases in frequencies of high dust concentrations, which can be further attributed to the low model biases in frequencies of strong wind events. We also find large uncertainties in retrieving dust extinction from CALIPSO and considerable impacts of observational uncertainties and limitations on the model evaluation.