An investigation of geospatial technologies in precision agriculture: a case study on a citrus orchard in the Eastern Cape

Abstract

Citrus production is an input-intensive farming practice that carries a high cost of production. A multitude of both local and global factors continue to put pressure on farmers to produce enough food for local consumption as well as international exports. Despite these challenges production and exports continue to increase, fighting to meet the growing rise in global demand for citrus (Genis, 2018). Growers are continuously in search of anything that may provide them with the 'edge' or an advantage to overcoming some of these challenges (Jupp, 2018). One way in which these issues could be addressed is the use of precision agriculture (PA). Precision agriculture, particularly that of commercial, Unmanned Aerial Vehicle (UAV) based PA, provides growers with solutions to these issues in the form of high quality, near real-time data, and provides access and benefits from technology driven agriculture to growers at all levels (Sishodia et al. 2020). The aim of this research therefore was to investigate the potential of high resolution, multi-spectral UAV, and satellite imagery to help citrus farmers manage their inputs better, save costs and increase their yields in a sustainable manner. Supervised image classification using a support vector machine (SVM) was applied to map and classify a citrus farm in the Eastern Cape. The approach aided the identification of Phytophthora spp in the section of interest and implies that remotely sensed data can be used to detect changes in citrus health. Guidelines for applying geospatial technologies at farm level were developed to provide a framework for enabling growers to enhance data driven farm management strategies.