Quantifying Dynamic Characteristics of Dune Migration in Northwestern China with Multitemporal Optical Satellite Observations

摘要

Quantifying sand dune movements might contribute to deepen the understanding of local aeolian processes and offer decision support for disasters prevention. Different with the traditional field survey, the increasingly rich satellite images may provide the possibility to investigate the spatio-temporal evolution of dune migration from the perspective of synoptic view. In this study, one new pairing strategy for optical imagery correlation is proposed. With this strategy, least-squares adjustment, and singular values decomposition (SVD) approaches, we investigate the dune displacement time series and rates near Minqin and Dunhuang oasis with the time series observations of Landsat 8 (L8) and Sentinel 2 (S2) collected from 2013 to 2018. The displacement time series and rates derived from this technique can better interpret the spatio-temporal dynamic patterns of dune migration, which is consistent with local wind conditions and other related observations. The technique provides an interesting attempt to investigate the dynamic characteristics of dune migration on dune-field scale with free-accessible optical satellite observations.