Remote Monitoring of Coffee Cultivation through Computational Processing of Satellite Images

The vegetation indexes (VI) obtained by satellite images play an important role in the diagnosis of productivity estimates, nutritional assessment, detection of pests and diseases, forecasting of the climate and evaluation of water needs in crop areas. The objective of this work was to evaluate the vegetation indexes calculated through the computational processing of images obtained by the Landsat-8 and Sentinel-2 satellites, with the purpose of monitoring the state of health and phenological development of coffee crops in Central American territory during the years 2017 and 2018. The vegetation indexes NDVI and EVI and were obtained with the reflection data of satellite images, these images were acquired on average every three days during the productive cycles of 2017 and 2018. The estimates of the leaf area index and the phenological stage of the crops was carried out during these productive periods. The results of the VIs were validated by a comparison with data on productivity, water stress and phenological status of coffee at each stage of the production cycles from January 2017 to December 2018. The results show that the vegetation indexes reach high levels of precision in the estimation of productivity, nutritional evaluation, detection of pests and diseases, monitoring of local temperature and assessment of water needs in specific sites of the crop. It was found that the NDVI and EVI values accurately reflect the water stress, nutritional problems and diseases that affect the development of the crops in the experimental farms. The methodology was satisfactory for obtaining parameters of leaf area indexes in coffee plantations and has potential use as a tool for monitoring and evaluating the health conditions and phenological status of different crops quickly and accurately. This monitoring methodology can assist in the application of more adequate strategies for decision making in large-scale crop areas.