Connecting geomorphology to dust emission through high-resolution mapping of global land cover and sediment supply

Abstract

A key challenge in modeling dust emissions is to represent the location and strength of dust sources. One critical aspect of dust sources that is not well understood and thus not represented in dust models is their geomorphology. In this work, we investigate the geomorphology of global dust sources by developing two high-resolution (∼500 m), seamless, global maps. First is a land surface map in which landforms are classified into different categories based on geomorphology using an image classification technique. The land surface map shows the distribution of landforms in dust source regions and is useful in defining the boundaries of different dust sources in dust models. Second is the sediment supply map developed by combining the upstream drainage area with the visible reflectance retrieved by the Moderate-resolution Imaging Spectroradiometer (MODIS). This map, due to the inclusion of surface reflectance, highlights dust sources such as playa/sabkha and sand dunes and anthropogenic dust sources such as agricultural areas, that may not be captured by the commonly used elevation-based erodibility maps. We establish the connection between geomorphology and dust emission by comparing the sediment supply map with the land surface map and dust frequency map, qualitatively and quantitatively. We show that the sediment supply is linked to the land surface type and that playa/sabkha corresponds to the greatest inferred sediment supply. The sediment supply map is largely consistent with the land surface map and correlates well with the frequency of occurrence map derived from high-resolution MODIS level-2 aerosol optical depth (AOD) data.