VegIndex: open-source google earth engine computational routine for spatio-temporal analysis of vegetation index.

Despite the rapid growth in the number of applications developed on Google Earth Engine (GEE) in recent years, related to various topics in precision agriculture (PA), there is still a need to develop more specific or customized applications. In this context, instigated by the increase in the development and adoption of digital tools in agriculture, as well as the trend towards cloud data processing, in this work we aimed to develop an automated, open source and scalable computer routine (code repository with all the GEE scripts) to carry out spatio-temporal analysis of the NDVI vegetation index. For this purpose, three datasets (or collections) from the Landsat 8, Landsat 9 and Sentinel-2 instruments, available in the GEE repository, were used. The routine developed, called VegIndex, has three data processing, analysis and visualization modules, structured using object-oriented programming (functions), with JavaScript language edited in the GEE Code Editor. These modules are not sequential, can works independently. The aforementioned routine was tested on a commercial area (122.70 ha) with irrigated cotton crops. The routine developed made it possible to carry out spatio-temporal analysis of NDVI in a fast, intuitive and 100% automated way, with great potential for use in PA. The routine is open source, allowing the user (researcher, consultant, technician, grower, etc.) to adapt it to different agricultural crops and forestry species.