Estimating Suspended Sediment Concentration Using Remote Sensing and Artificial Neural Network (Case Study: Karun River)
Z. Mollaee,
J. Zahiri,
S. Jalili,
M. R. Ansari,
A. Taghizadeh
Affiliations
Z. Mollaee
1. Department of Water Engineering, Faculty of Agriculture and Rural Engineering, Khuzestan Agricultural Sciences and Natural Resources University, Ahvaz, Iran.
J. Zahiri
1. Department of Water Engineering, Faculty of Agriculture and Rural Engineering, Khuzestan Agricultural Sciences and Natural Resources University, Ahvaz, Iran.
S. Jalili
1. Department of Water Engineering, Faculty of Agriculture and Rural Engineering, Khuzestan Agricultural Sciences and Natural Resources University, Ahvaz, Iran.
M. R. Ansari
2. Department of Soil Science, Faculty of Agriculture, Khuzestan Agricultural Sciences and Natural Resources University, Ahvaz, Iran.
A. Taghizadeh
3. Department of GIS and RS, Faculty of Earth Sciences, Chamran University, Ahvaz, Iran.
Spectral Reflectance of suspended sediment concentration (SSC) remotely sensed by satellite images is an alternative and economically efficient method to measure SSC in inland waters such as rivers and lakes, coastal waters, and oceans. This paper retrieved SSC from satellite remote sensing imagery using radial basis function networks (RBF). In-situ measurement of SSC, water flow data, as well as MODIS band 1 and band ratio of band 2 to 1 were the inputs of the RBF. A multi-regression method was also used to make a relationship between the in-situ data and the water reflectance data retrieved from MODIS bands. The results showed that RBF had the best SSC prediction error (RMSE=0.19), as compared to the multi-regression and sediment rating curve methods, with the RMSE of 0.29 and 0.21, respectively.
