Alemayehu A B A T E Shawul

Unavailability of optimum network of rain gauges and lack of well-organized precipitation data are one of the main challenges in hydrologic and climate-related studies. This study aims to evaluate the performance of three global precipitation data obtained from the Climate Forecast System Reanalysis (CFSR), Climate Prediction Center (CPC) and Tropical Rainfall Measuring Mission (TRMM) for hydrologic prediction at four main watersheds in the Upper Awash basin. Statistical indices and graphical efficiency measures were used to examine the CFSR-NCEP, CPC-NOAA, and TRMM data with the gauged data on mean annual, monthly, and daily basis at Upper Basin (UB) and Lower Basin (LB) area. The performance of global precipitation estimates was encouraging. Moreover, the spatial map of annual precipitation products demonstrated comparable spatial patterns with observed annual precipitation. The Soil and Water Assessment Tool (SWAT) model was used for hydrological simulation, and the hydrologic simulation efficiency of gauged data and global precipitation revealed good performance with R 2 and NS value of > 0.5 for daily, and > 0.7 for monthly scale simulations. However, lower to satisfactory model performance was obtained at the daily simulation in the Mojo and Akaki watersheds. The CPC-NOAA precipitation data was found superior in capturing the hydrologic variability. In relatively smaller watersheds, the suitability of global precipitation was lower due to the limited number of data points resulted from the coarse resolution of the data sources. Improvement in the satellite meteorology and climate reanalysis data could substantially increase the availability of alternative climate data for hydrologic studies.

Assessment of the changing environmental conditions is essential for planning the wise use of natural resources. The main objective of this paper is to analyze the historical and future modeled LULC changes using multi-temporal Landsat images in the Upper Awash basin, Ethiopia. The supervised image classification method was used to determine the historical LULC changes based on Landsat 1 MSS 1972, Landsat 5 TM 1984, Landsat 7 ETM + 2000, and Landsat 8 OLI TIRS 2014. The future LULC change was predicted using the machine-learning approaches of Land Change Modeler (LCM). The LULC change detection analysis exhibited significant increment in the areal extent of the cropland and urban areas, and decreasing trends in the pasture, forests and shrubland coverage. Mainly, the LULC change matrices indicated that larger conversion rate was observed from shrubland to cropland area. The urban area found to increase by 606.2% from the year 1972 to 2014 and cropland has also increased by 47.3%. Whereas, a decreasing trend was obtained in the forest by − 25.1%, pasture − 87.4%, shrubland − 28.8% and water − 21.0% in the same period. The modeled future LULC change scenarios of the year 2025 and 2035 have exhibited significant expansion of cropland and urban areas at the expense of forest, pasture and shrubland areas. The study has revealed the extent and the rate of LULC change at larger basin and subbasin level which can be useful for knowledge-based future land management practice in the Upper Awash basin.