Soil moisture prediction in an agricultural field of the Gumara-Maksegnit watershed, North Gondar, Ethiopia uri icon


  • The master study on Soil Moisture Prediction in an Agricultural Field of Gumara- Maksegnit Watershed was conducted within Gumara-Maksegnit Watershed found in the Ethiopian highlands, North Gondar. This was aimed to predict moisture contained in 1meter soil profile spatially and temporally in the drying process of the 2013 wet period in small rainfed agricultural field mainly covered with cereal and legume crops. A Weekly soil moisture content (v/v) with PR2/6 Profile Probe which is product of Delta T Devices, topogrpaphic information with a spatial resolution of 5 by 5 meter measured using GPS and water level, soil texture in two soil depth ranges, crop information and representative soil bulk density were measured in the study period i.e. 30July 2013 to 9November 2013. Soil moisture observation with a PR2 Profile probe, calibrated with gravimetric measured value, were characterized and analyzed using basic statistics for data quality and consistency along time and space. Reasonable but high temporal soil moisture variability was observed in the top 10 to 20 cm soil layer compared to the rest of layers down to 1meter. Physically based hydrological model called Soil-Plant-Atmospher-Water (SPAW), developed by E. Saxon, 2006; was used to predict the temporal change of soil moisture considering the basic soil properties and agroclimatologic information together with crop data. Geostatistical analyst, Co-kriging in ArcGIS 10.2 (ESRI), was applied to predict the spatial distribution of soil moisture considering slope as a secondary variable and the temporally predicted soil moisture as the main variable. Results from both models were checked for their consistency using coefficient of determination (R2), root mean squared error and averaged standard error. R2 value of observed versus SPAW simulated soil moisture in the period of drying process shows 0.72. Spatially distributed soil water storage (SWS) with Co-kriging produced average standard error of 12.8 with maximum and minimum SWS of 325mm/m and 246mm/m respectively

publication date

  • 2014