Hydroponic farming hotspot analysis using the Getis–Ord Gi* statistic and high-resolution satellite data of Majuli Island, India

Abstract

Majuli is the world’s largest Riverine Island, formed by the headward erosion and migration of the Brahmaputra River. It is a part of the floodplains, which causes loss of arable lands, crops and farmers’ livelihood. To cope and adapt, farming communality adopted hydroponics farming. In this study, high-resolution satellite data (~3 m) were deployed to detect hotspots of hydroponics growing under flood conditions. A Getis–Ord Gi* spatial statistics technique was employed along with vegetation indices. The spatial maps exhibit that at several locations of Majuli Island, hydroponics was detected during the monsoon season. In total, 14 major sites were identified under hydroponics with a total area of 5.6 ha. The accuracy of hydroponics was assessed using precision and recall with the F1 score of 0.94. Around six sites of hydroponics were also validated with the independent high-resolution Google Earth image. This study demonstrates the utility of spatial statistics technique and high-resolution data for hotspot analysis of hydroponics farming. The method can be adopted for making plans and policies to identify locations for the commercialization of hydroponics farming, which could assist in food security, employment and farmer’s livelihood and has the potential to meet the second sustainable development goal (SDG).