Comparison of Soil Erosion Empirical Models in Limited-Access Area of Reclaimed Mine

Abstract

In mine reclamation, estimation of soil erosion is one of the important operations. It requires quantitative data to accurately identify critical areas. So, the accuracy in the estimation of soil erosion is a prior requirement to determine the effect of soil erosion especially in the reclamation aspect of mining. The present study uses soil erosion modeling integrates with remote sensing techniques to estimate soil erosion in the reclaim-zinc mine in Mae sot, Thailand. The integrated spatial approach based on the Revised Morgan - Morgan - Finney Model (RMMF) and the Revised Universal Soil Loss equation (RUSLE) have been generally applied to demonstrate the critical area according to its accessibility. Since these empirical models are subjected to the inclusiveness of the secondary data, especially the Digital Elevation Model (DEM), so that it was use as the main equation for the examination of the alternation of this parameter. DEMs with different sources and resolutions can generate varied topographic and hydrological features, affecting the accuracy and characteristic of the erosion model. DEMs from the Shuttle Radar Topography Mission (SRTM) at 30 m and 90 m resolution and Unmanned Aerial Vehicle (UAV) with 0.3 m resolution were acquired and applied as inputs for the RMMF and RUSLE model. The results were compared with the data from field measurement data to determine the accuracy of soil erosion models. The estimated annual average soil loss from RMMF in study area are 0.96, 0.84, 15.89, and 0.95 tonha-1year-1 which are received according to the DEMs, received from UAV, reclassifies UAV, SRTM 30 m and SRTM 90 m resolution respectively. While, the estimated annual average soil loss from RUSLE in study area are 717.87, 96.52, 5262.63, and 16772.38 tonha-1year-1 which are received according to the DEMs, received from UAV, reclassifies UAV, SRTM 30 m and SRTM 90 m resolution respectively. The results from elevation and erosion were analyzed using statistical analysis including root mean square deviation (RMSE) and relative standard error (%RSD). The comparison of RUSLE with RMMF model was done using statistical analysis. From comparison with field measurement are show that the RMMF model have percent difference value are 99.14%, 100%, 86.67% and 76.188% of UAV, reclassified UAV, SRTM30 m and SRTM90 m. While the RUSLE model have percent difference value are 933.657%, 100%, 46579.47% and 60277.26% of UAV, reclassified UAV, SRTM30 m and SRTM90 m. These results indicate that the RMMF model have better than RUSLE model in study area.