Developing A Dust Storm Detection Method Combining Support Vector Machine and Satellite Data in Typical Dust Regions of Asia

Enhancing the dust storm detection is a key issue for the environmental protection, human healthy and economic development. The goal of this paper is to propose a new Support Vector Machine (SVM)-based method to automatically detect the dust storm using remote sensing data. Existing methods dealing with such kind of problem are usually threshold-based that are of great complexity and uncertainty. In this paper we proposed a simple and reliable method combining SVM with MODIS L1 data and explored the optimal band combinations used as the feature vectors of SVM. The developed method was evaluated by MODIS and OMI data qualitatively and quantitatively on three study sites located in the Arabian Desert, Gobi Desert and Taklimakan Desert, and it was also compared to three other traditional methods based on their accuracy, complexity, reliability and sensitivity to thresholds. The detection results demonstrated that the combination of (Band7-Band3)/(Band7+Band3) ((B7-B3)/(B7+B3)), Band20-Band31 (B20-B31) and Band31/Band32 (B31/B32) can detect the dust storms more precisely than other sole bands or their combination. The comparison among those cases indicated that the proposed automatic method exhibited the advantage of eliminating the uncertainty and complexity, which were the main limits of the threshold-based methods. The conclusions have strengthened research on dust aerosol detection and offered references to the choice of feature vectors of the machine learning algorithm.