Dr. Adriana R. Lima, Goddard Space Flight Center UMBC/JCET
Monday, April 15, 2019 · 3:30 PM - 4:30 PM
Saharan Desert and the Recent Variability of Middle Eastern Dust AOT in Observations and Models
In this colloquium, I will talk about the Saharan desert and observed changes in the dust distribution over the Middle East region. Middle East has important dust sources that impact the climate and human health. Long-term observations from ground-based Aerosol Robotic Network (AERONET) of sun photometers and the space-based Moderate-Resolution Imaging Spectroradiometer (MODIS) have shown positive trends in aerosol optical depth (AOD) over the period between 2001 and 2012, followed by a negative trend after this period. In this study, we evaluate the ability of two recent GEOS model reanalyzes to capture such trends: The Modern-Era Retrospective analysis for Research and Applications, version 2 (MERRA-2) Reanalysis covers the period 1980 – present and includes prognostic aerosols and assimilation of AOD from ground-based and satellite observations. The MERRA-2 Replay covers the same period and is run with similar core version of the GEOS model version driven by the MERRA-2 meteorological fields, but significantly does not include AOD assimilation in the aerosol scheme. Our analysis show that the observed trends are better reproduced in the MERRA-2 Reanalysis (with aerosol data assimilated) and are mostly absent in the MERRA-2 Replay (no aerosol data assimilated), highlighting the impact of aerosol data assimilation on the ability of the model to reproduce observed variability and long term trends of atmospheric aerosols. In order to improve the capability of the model to represent such long-term changes in the dust processes we analyzed the rate of change over the Middle East region of important dust emission parameters, such as vegetation, soil moisture, and wind fields. Finally, as changes in the vegetation were not considered in our previous simulations, we investigate the sensitivity of the dust emissions to changes in the normalized difference vegetation index (NDVI) observed in the Middle East region over this period. Our sensitivity study shows that a change in the dust emission in the region where highest NDVI variability was observed reduces the differences between the model simulation and MODIS observations and indicated that further investigation of the observed parameters related to dust emission is required in order to understand and incorporate the dust emission variability in the model.