Title : Impact of watershed management on vegetation cover and LULC dynamics using remote sensing in Wonba Micro-watershed, Omo Gibe Basin, Ethiopia.
Abstract:
The impact of watershed interventions on vegetation cover and LULC dynamics is one the responsible factor for changing the hydrological process of watershed by altering the magnitude of aquifer recharge, surface runoff and river flow. Thus, effective information regarding the LULC and vegetation cover responses of watershed management impact are important for hydrologists, water resources engineers and land use planner for sustainable water resource planning and management. Consequently, this study was aimed at evaluating the changes in LULC and vegetation cover due to watershed interventions in Wonba micro watershed for the period of 30 years from 1990 to 2020 which encompasses the period before and after watershed interventions. To address this concern, multi-temporal Landsat images of thematic mapper (TM), Enhanced thematic mapper plus (ETM+) and Operational land imager (OLI) with the aid of recent ground based data and historical trends were used for estimating the Normalized Difference Vegetation Index (NDVI) for detecting the vegetation cover change and to generate LULC map for assessing the land cover changes. Long-term daily rainfall data from 1982 was aggregated to derive average annual precipitation (AAP) into three sections to correspond with the processed image data. The rainfall data then converted into raster format using the inverse distance weight interpolation (IDW) method. Furthermore, the LULC maps were generated with the Maximum Likelihood Algorithm of Supervised classification. The accuracy of classified maps was also assessed using confusion matrix. The analysis was performed using ERDAS Imagine 16 software and results were mapped in ArcGIS 10.5 package. The result indicates that significant modification and conversion of LULC and vegetation cover classes of the watershed over the last three decades (1990-2020). The area under dense vegetation cover increased from 1.44 km2 (6.87%) to 4.52 km2 (17.54%) post intervention. The area under bare land decreased from 0.26 km2 (1.03%) to 0.01 km2 (0.05%). Even though, the Average Annual Precipitation (AAP) distribution shows a declining trend over the 30 years of study period, the vegetation cover shows an increasing trend. The LULC maps of the year 1990, 2005 and 2020 were generated. The integration of NDVI with supervised classification has improved the classification of LULC maps by 89.73%, 90.91% and 97% for the year 1990, 2005 and 2020 respectively. The change detection analysis showed that the bare land was continuously decreased from 12% in 1990 to 2.27% in 2020 in the expenses of the other classes. The expansion of forest land by 3.5% in the watershed had an impact on the decrease of bare land for the past 30 years. The study indicates that watershed management implemented in the study area is successfully achieved. Therefore, this study concludes by providing an indicator data support and theoretical bases for further research on land cover change and vegetation restoration for other regions of Ethiopia.
