| dc.description.abstract | This study focuses on improving estimation of harvested area, a key term in estimation of yield, which guides most evaluation of improved practices and optimization of resource allocation. It validates the accuracy and precision of GPS applications in determination of land area then compares the farmer-reported areas and those determined by the GPS mediated mobile device applications. The study was carried out in three agro-ecological zones of Uganda namely, the Lake Victoria Crescent, western and southwestern located in the central, western and south western regions. A known area (0.1, 0.2, 0.3, 0.6, 1.2, 1.8, 2.4 and 3.6 ha) was demarcated on the ground in regular rectangular shapes using the traversing method. For each demarcation, Samsung tablet installed with GPS essentials application (app) were separately used to determine the coordinates at cardinal points of the demarcation. The coordinates were imported into a GIS software (QGIS) and used to generate a polygon whose area was computed using in-built routines. The data were used to evaluate the accuracy and precision of the app. Banana farmers were asked to the acreage of their banana fields. The fields’ areas were then measured using the GPS mobile devices as described for the app. The data were used to evaluate the accuracy and precision of farmer reported area for banana fields. Data were analyzed using the paired t-Test and linear regression. The app was found to be very accurate (Slope =1) and precise (R2=99.9%) in estimating field area. Farmer over-estimated their field areas leading to low accuracy and poor precision (R2= 39.5%). The accuracy decreased in the order Central >Western and South western regions. However, the precision of farmer-reported area was very poor. A regression function developed to predict actual field area from farmer-reported values using a GPS app performed poorly on an independent data set, showing that the errors in farmer reported could not be corrected based on the available data set. It was recommended that rigorous observations from a larger data set be used to generate reliable functions for correcting the farmer-reported values.
Keywords: Banana,Geographical Information Systems, Traversing, Uganda, Yield | en_US |
