Finger Millet (Eleucine coracana) Yield Estimation: Integrating Remote Sensing and Farm Management Practices in Busia-Kenya

Abstract

Early crop yield prediction is important for planning and taking various decisions by the farmer. Conventional techniques of data collection for crop monitoring and yield estimation rely on ground-based visits and reports. These methods are subjective, very costly and time consuming. With the launching of satellites, remotely sensed data is being used for crop monitoring and yield prediction. This study applied space-borne satellite based NDVI to predict crop yield at field level. It was carried out in Kenya’s Busia County. The purpose of the study was to investigate the relationship between space-borne Satellite based NDVI and Finger millet (Eleucine coracana) yield and combining NDVI with farmer’s indigenous knowledge (IK) and practices for yield prediction at field level. A survey was carried out to investigate management and land factors in Finger millet growing area during the long rains growing season, January-June 2014. This was followed by satellite image analysis of farms during different stages of Finger millet growth. Data collected was correlated against Finger millet yield in order to analyse the relationships among Remote sensed data, indigenous practices and crop yield. The results showed that there is significant correlation between remotely sensed NDVI and field level finger millet yield (r = 0.19, p =0.040, N = 57). The most yielding farms were those with mixed variety of seed (i.e local and improved u15 variety) and broadcasting as a planting technique. The study also showed that not all the factors affecting yield also affect NDVI. This paper concludes that remote sensing can be used as a tool for monitoring crop growth and vigor based on different farm management practices. The paper further suggests that the integration of farmer IK in the management of finger millet crop and use of mixed variety of seed is likely to improve yield.