Towards Data-Driven Precision Agriculture using Open Data and Open Source Software

Abstract: Information and communications technology (ICT) within the agricultural sector is characterized by a widespread use of proprietary data formats, a strong lack of interoperability standards, and a tight connection to specific hardware implementations resulting from vendor lock-in. This partly explains why ICT has not yet had its full impact within the domain. By utilizing the vast amount of publicly available open data, ranging from topographic maps to multispectral satellite images, the economically and environmentally optimal farming practices can be advanced beyond state of the art. This paper addresses the potential of applying publicly available information sources to improve crop production, with emphasis on yield optimization. This potential is evaluated based on free public data for the growth season 2016 by examining winter wheat production for a selected region in Denmark. Data aggregation is performed by promoting opensource software tools as a foundation for decision support. That allows the farmer, or another domain expert, to query a certain crop type, merge this information with other data sets, and perform analysis on data ranging from sub-field analysis to statistics on national/regional scale. The registration of field polygons and sowed crop types for fields in Denmark, alongside with detailed geographic data and free satellite images, enable us to exploit publicly available data of high quality, which can be applied to perform further analysis.