Agri-food research is increasingly becoming the work of a globally connected research network. This requires stronger collaboration of research groups over the world, e.g. for co-designing and developing methods and models. These days, working together often still requires travelling and therefore opportunities to collaborate are scarce and expensive. Moreover, in times of climate change, many organisations think of reducing their contribution to CO2 emission through less travelling. The outbreak of COVID-19 has made us even more dependent on online collaboration, and here virtual co-working will be a valuable addition to help alleviate the adverse effects of being at a distance.
Virtual Research, where scientists share and collaborate in a cloud-based environment with integrated tools for data science, can increase the efficiency of research. An example is agricultural modelling, where global networks exist that harmonize, improve and apply crop simulation models. While the collaborating networks exist, there are no suitable working environments to co-work on the development and application of such models.
In AGINFRA+, Wageningen UR and its partners set up cloud-based Virtual Research Environments (VRE) for collaborative research by the global agro-climatic modelling community. We focus on large scale, high resolution crop growth modelling and yield forecasting. A combination of social media components and shared and interconnected components for data science and sharing and publishing of data and models supports efficient teamwork. A distributed computing cloud allows for running models and algorithms, using large amounts of data (e.g. high-resolution remote sensing), for large areas like Europe on the scale of individual agricultural parcels at acceptable performance. This reveals new opportunities for large scale yield forecasting, with results become usable in daily practice of farmers. Moreover, the easy access to data and data science tools, combined with an Open Science approach stimulates co-development and reuse and improvement of developed products.
The work connects to Dutch and international Big Data developments. Developed models and algorithms use the Dutch AgroDataCube and European Sentinel high resolution remote sensing data. AGINFRA+ fits into the European strategy on Open Science, adopting the key concepts and principles of the European Open Science Cloud (EOSC), e.g. the FAIR principles that state that research assets should be Findable, Accessible, Interoperable and Reusable.
AGINFRA+ proves that a virtual research environment works, and that a cloud-based approach for sharing knowledge and resources improves the innovation power of agri-food research. It helps scientists to safely work together with all data and tools at hand, and encourages to share and publish research data, algorithms and models. The tendency to reduce the environmental footprint of research and the urgent needs to search for efficient distant working formats caused by the COVID-19 crisis will increase the interest in virtual research and fit-for-purpose tooling for scientists. By developing and testing virtual research modelling workflows we pave the way and provide showcases for future research in many domains.
