| INVESTIGATION OF POTENTIAL DUST SOURCES USING SENTINEL-1 AND NEURAL NETWORK: A CASE STUDY FROM BANDAR-E EMAM-OMIDIYE |
| Paper ID : 1120-SMPR |
| Authors: |
|
Sahar khoshnoud1, SeyedMohammad MirMazloumi *2, Meisam Amani3, Hossein Mohamad Asgari4 1Environmental Geology ،Envionment group , Faculty of Marine Natural Resources , Khorramshahr University of Marine Science and Technology, Khorramshahr, Iran 2Remote Sensing Research Center, Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran 3Wood Environment & Infrastructure Solutions, St. John's, NL, Canada, A1B 1H3 4Envionment group , Faculty of Marine Natural Resources, Khorramshahr University of Marine Science and Technology, Khorramshahr, Iran |
| Abstract: |
| Aeolian erosion is a serious environmental threat that damages soils. Dust storms are one example of the consequences of aeolian erosion in dry and semi-arid areas across the world. In this regard, soil surface roughness is an important parameter for monitoring climate changes on the Earth and modelling aeolian erosion. Synthetic Aperture Radar (SAR) systems are valuable resources for estimating soil surface roughness. In arid soils, SAR backscatter is sensitive to the soil surface roughness at higher frequencies and higher incident angles. Based on these facts, an Artificial Neural Network (ANN) along with Sentinel-1 images in two polarizations (VV and VH) were used to estimate surface roughness in Bandar-e Emam-Omidiye, Khuzestan, Iran. The parameters used to train the ANN included the radar backscatter coefficient, incident angle, and in-situ roughness. These data were subsequently used to identify areas prone to dust. The results obtained from the investigation of 25 stations located in areas with five different land covers indicated that locals on clay flats are the most prone to aeolian erosion in the form of dust. |
| Keywords: |
| Dust, Sentinel-1, SAR, Neural Network, Surface Roughness |
| Status : Conditional Accept (Poster) |
