Identification of Dust Source Regions at High-Resolution and Dynamics of Dust Source Mask over Southwest United States Using Remote Sensing Data

Abstract:

Identification and evaluation of sources of aeolian mineral dust is a critical task in the simulation of dust. Recently, time series of space based multi-sensor satellite images have been used to identify and monitor changes in the land surface characteristics. Modeling of windblown dust requires precise delineation of mineral dust source and its strength that varies over a region as well as seasonal and inter-annual variability due to changes in land use and land cover. Southwest USA is one of the major dust emission prone zone in North American continent where dust is generated from low lying dried-up areas with bare ground surface and they may be scattered or appear as point sources on high resolution satellite images. In the current research, various satellite derived variables have been integrated to produce a high-resolution dust source mask, at grid size of 250 m, using data such as digital elevation model, surface reflectance, vegetation cover, land cover class, and surface wetness. Previous dust source models have been adopted to produce a multi-parameter dust source mask using data from satellites such as Terra (Moderate Resolution Imaging Spectroradiometer - MODIS), and Landsat. The dust source mask model captures the topographically low regions with bare soil surface, dried-up river plains, and lakes which form important source of dust in southwest USA. The study region is also one of the hottest regions of USA where surface dryness, land use (agricultural use), and vegetation cover changes significantly leading to major changes in the areal coverage of potential dust source regions. A dynamic high resolution dust source mask have been produced to address intra-annual change in the aerial extent of bare dry surfaces. Time series of satellite derived data have been used to create dynamic dust source masks. A new dust source mask at 16 day interval allows enhanced detection of potential dust source regions that can be employed in the dust emission and transport pathways models for better estimation of emission of dust during dust storms, particulate air pollution, public health risk assessment tools and decision support systems.