Diagnostic evaluation of the Community Earth System Model in simulating mineral dust emission with insight into large-scale dust storm mobilization in the Middle East and North Africa (MENA) Sagar Prasad Parajuli a , Zong-Liang Yang a,⇑ , David M. Lawrence b a The University of Texas at Austin, Jackson School of Geosciences, 2225 Speedway, Stop C1160, Austin, TX 78712-1692, United States b National Center for Atmospheric Research, Boulder, CO 80307, United States article info Article history: Received 26 May 2015 Revised 9 January 2016 Accepted 11 February 2016 Available online 1 March 2016 Keywords: Soil erodibility Dust storms Dust emission modeling Middle East and North Africa Aerosol optical depth abstract Large amounts of mineral dust are injected into the atmosphere during dust storms, which are common in the Middle East and North Africa (MENA) where most of the global dust hotspots are located. In this work, we present simulations of dust emission using the Community Earth System Model Version 1.2.2 (CESM 1.2.2) and evaluate how well it captures the spatio-temporal characteristics of dust emission in the MENA region with a focus on large-scale dust storm mobilization. We explicitly focus our analysis on the model’s two major input parameters that affect the vertical mass flux of dust—surface winds and the soil erodibility factor. We analyze dust emissions in simulations with both prognostic CESM winds and with CESM winds that are nudged towards ERA-Interim reanalysis values. Simulations with three existing erodibility maps and a new observation-based erodibility map are also conducted. We compare the simulated results with MODIS satellite data, MACC reanalysis data, AERONET station data, and CALIPSO 3-d aerosol profile data. The dust emission simulated by CESM, when driven by nudged reanalysis winds, compares reasonably well with observations on daily to monthly time scales despite CESM being a global General Circulation Model. However, considerable bias exists around known high dust source locations in northwest/northeast Africa and over the Arabian Peninsula where recurring large-scale dust storms are common. The new observation-based erodibility map, which can represent anthropogenic dust sources that are not directly represented by existing erodibility maps, shows improved performance in terms of the simulated dust optical depth (DOD) and aerosol optical depth (AOD) compared to existing erodibility maps although the performance of different erodibility maps var- ies by region. Ó 2016 Elsevier B.V. All rights reserved. 1. Introduction Atmospheric mineral dust has a wide range of implications for Earth’s radiation budget (Sokolik and Toon, 1996; Miller and Tegen, 1998; Mahowald et al., 2006), biogeochemical cycles (Kellogg and Griffin, 2006; Yu et al., 2015), precipitation (Creamean et al., 2013; Jin et al., 2014), human health (WHO, 2006), and visibility (Wang et al., 2008). Mineral dust is one of the major contributors to the global aerosol budget (IPCC, 2013), modeling of which remains challenging because dust emission has high spatial and temporal variability. The direct/indirect effects of aerosols remain the largest source of uncertainty in estimating radiative forcing and the emissions of natural aerosols including mineral dust remains poorly characterized in climate models (IPCC, 2013). A recent evaluation of the global climate models used in CMIP5 (including the Community Earth System Model used in this study) by Evan et al. (2014) showed that there is a consider- able mismatch between model simulations and observations in terms of the climatology of dust emission and transport. Similar conclusions were made earlier by Cakmur et al. (2006) and Huneeus et al. (2011). Although numerous dust models exist, dust emission is param- eterized generally in terms of surface wind velocity usually at 10 m height (e.g., Ginoux et al., 2001) or friction velocity (e.g., Zender et al., 2003a) which is mainly a function of particle size, soil mois- ture, and clay content. Currently, the simulated global vertical mass flux of dust differs greatly across models and ranges from 1000 to 5000 Tg yr 1 (Shao et al., 2011) although this discrepancy can be reduced to some extent with observation-based constraints (Cakmur et al., 2006; Mahowald et al., 2006). Deficiencies of dust models can be attributed to multiple reasons including the use of http://dx.doi.org/10.1016/j.aeolia.2016.02.002 1875-9637/Ó 2016 Elsevier B.V. All rights reserved. ⇑ Corresponding author. Tel.: +1 512 471 3824; fax: +1 512 471 9425. E-mail address: liang@jsg.utexas.edu (Z.-L. Yang). Aeolian Research 21 (2016) 21–35 Contents lists available at ScienceDirect Aeolian Research journal homepage: www.elsevier.com/locate/aeolia