It takes a village to raise a .. digital twin
Evert, F.K. van; Been, T.H.; Berghuijs, H.N.C.; Heinen, M.; Jabloun, M.; Keizer, M. de; Kempenaar, C.; Maestrini, B.; Oort, P.A.J. van; Pronk, A.A.; Sijbrandij, F.D.; Wit, A.J.W. de
In precision agriculture, farmers need precise, real-time information about the status of crops, soils and livestock, as well as information about the likely outcome of management decisions. Several decades ago, the use of crop growth models for this purpose has been proposed but it has proven difficult to obtain the input data required to run crop growth models for commercial farms in a timely manner, if at all. In this paper we document how in the case of Farmmaps input data is obtained, and how simulation results are presented to farmers. We consider that the crop growth models used have been extensively validated during the past decades; hence this paper focuses on the application of these models in commercial farming. Farmmaps is a web-based platform developed by Wageningen University & Research, in collaboration with farmers and a software development company; it is owned by a not-for-profit foundation. Farmmaps offers various precision agriculture applications, some of which are offered for a fee. The revenue thus obtained allows purchasing the world-wide weather and satellite data needed to offer the services. Farmmaps links to other data sources, such as the national soils database, soil analysis labs, and commercially available Farm Management Information Systems (FMIS) that are used by farmers in The Netherlands. The precision agriculture applications available on Farmmaps have been developed in close collaboration with farmers, which ensures that they meet a real, practical need and that farmers have a keen appreciation of what the applications can do. One of the applications available on Farmmaps is a digital twin for arable crops called the Digital Future Farm (DFF). The DFF consists of dynamic models of arable crops and grass, as well as a method to use real-time data from sensors to keep the models synchronized. The DFF provides a comprehensive overview of the current state of crops and soil. The DFF can also be used to predict the future state (by using forecast and/or stochastic weather). Finally, the DFF can be used to investigate the expected outcome of alternative management scenarios, especially with respect to fertilization and irrigation. Simulation output of the DFF is presented in tables and graphs that have been designed together with the farmers that use them. The title of this paper is an attempt to express the salient fact that it took a community of farmers, scientists, software companies, data providers, sensor providers, and others (“the village”) to bring the digital twin (“a child”) to maturity.