Title:

Landsat identifies aeolian dust emission dynamics at the landform scale

Publication Year:
2017
Abstract:

The modelling of windblown mineral dust emissions remains a challenge. This is in part due to the coarse spatial and temporal resolution of the data on which these models are based, but also because the processes and mechanisms of aeolian dust emission are not well understood. Satellite imagery has been used extensively in the study of dust from the late 1990s with important contributions being made in terms of sources, transport pathways and deposition areas. Using MODIS imagery, the Namib Desert has been identified as one of the largest sources of dust in southern Africa. The opening of the Landsat archive presents the opportunity to investigate these events at a higher spatial resolution (up to 15 × 15 m) than previously possible. Despite the low temporal resolution, we used Landsat imagery to identify 40 major dust episodes over the last 25 years that originated primarily from the ephemeral river valleys and pan complexes, providing new insight into the spatial and temporal evolution of the dust sources from dryland surfaces. Examination of the imagery enabled the identification of local-scale landform source points to direct ground based testing of the surfaces responsible for dust emission. Emissivity tests were undertaken using a PI-SWERL portable wind tunnel in three of the major dust producing river systems along the Namib coast, namely the Kuiseb, Omaruru and Huab Rivers. Preliminary observations suggest that human impact on the hydrological systems in two of the river basins, to cater for the increasing demand of water, have dramatically altered the emission patterns of dust. The source areas of greatest dust emission are found to be located on recently deposited fluvial surfaces which are not active in the contemporary environment. Keywords: MODIS, Namib Desert, ERA-Interim reanalysis, Preferential dust scheme (PDS), PI-SWERL.

Publication Title:

Remote Sensing of Environment

Issue:
198
Pages:
229-243
Item Type:
Journal Article
Language:
en