An application of a spatially distributed hydrologic model WetSpa (Water and Energy Transfer between Soil, Plant and Atmosphere) in Gumara watershed, Ethiopia is presented in this research. The main objective of the study is to estimate spatio-temporal water balance that is less researched in Lake Tana and Blue Nile basin watersheds. The precipitation data from five rainfall stations and potential evapotranspiration estimated using the Hargraves equation are the input data used in addition to spatial elevation, soil and land use data. The observed discharge data of Gumara river is used to calibrate and validate the model to simulate the spatio-temporal water balance of the watershed at daily timestep. The model performance was evaluated both qualitatively (visual comparison of simulated and observed hydrographs) and quantitatively using four statistical criteria. These evaluation results indicate that the model simulates a very good match with the observe discharge with Nash-Sutcliffe efficiency of 82% and 75% respectively for the calibration and validation. However, the model underestimates peak flows during validation period, while it simulates the low flow hydrographs in very good agreement with the observed flow. The water balance terms show very strong spatial and temporal variability, about 4.6% of the total precipitation is intercepted by the plant canopy; 78.4% infiltrates into the soil; 73% is groundwater recharge and 91% is total runoff. The spatial soil texture variation influences most of the water balance components than land use for spatio-temporal variation of water balance in the study area. Hence, the study recommends site specific watershed management to improve the soil texture and land use classes that influence water balance components especially in Western and Northeastern part of the study area.