Um sistema de armazenamento e distribuição de dados georreferenciados para o Programa Bioma Stats

Abstract

With the increasing volume of environmental data and the growing demand for effective and accessible tools to work with spatial information, the development of systems that improve the access to and processing of georeferenced data is essential. In this context, Bioma Stats emerges as an open-source tool built in the R language, aiming to centralize and automate environmental analyses. Executing the Bioma Stats workflow in its first version required manually downloading its database directly from the MapBiomas platform, forcing users to download the entire collection of land use and land cover data for Brazil and overloading their machines with unnecessary storage demands for analyzing their specific area of interest. Therefore, this work presents the implementation of a georeferenced data storage and distribution system that operates through a RESTful API developed using Node.js, Express, and MySQL. The system orchestrates the selective delivery of only the data corresponding to the user's territorial fragments of interest, which are stored in the cloud, thus replacing the dependency on centralized servers. The approach of using HTTP requests for communication between systems validates the integration between the R environment and the web API. The results demonstrate a significant improvement in download time compared to the traditional method, as well as greater system resilience and optimization of computational resource usage.