The crane radar: Development and deployment of an operational eco-digital twin

As virtual copies of real-world phenomena, digital twins are rapidly gaining popularity in industry, academia and policy. Their main strength is informing decision makers in an operational environment by giving detailed up-to-date information about the systems represented in the digital twins, such as industrial production systems or weather forecasts. By running continuous predictions, digital twins evolve synchronous with the real world and related data, making digital twins a particularly powerful tool to study systems that are changing beyond their historic realm. Thereby, digital twins guide policy makers in addressing twenty-first century challenges. Yet, it is also a new concept to many scientists and policy makers in the environmental domain, which demands more clarity about what digital twins are and how they are built, in order to support constructive developments in digital twining. This paper therefore aims to outline the main concept and the steps involved in building digital twins for ecological decision making. By showing an example of the crane radar, a real-time digital twin of the common crane (Grus grus) migration, this paper illustrates what digital twins could look like in ecology, and which challenges modellers face towards actual deployment of their digital twin in an operational environment. The most important challenges identified here are: ex ante quality and error control of the input data; accounting for time lags between observations and data availability; harmonising the updating frequencies and temporal resolutions of models and data; developing robust model/software code to keep digital twins alive; the necessity of learning new skills for ecological modellers; and effectively communicating digital twin outputs to stakeholders. A real quantitative evaluation of the contribution of the crane radar to operational decision making is, however, still missing. •The crane radar is one of the first operational digital twins in ecology.•Lessons learned in its development can support developing other eco-digital twins.•Technical challenges include availability and automated quality control of input data.•Other challenges involve modellers' skills and better model-stakeholder communication.•Solutions require training and a multidisciplinary approach, from technical to social.