Environmental models as web services will increasingly become a feature of future environmental decision-making processes. EVO is developing a range of exemplar environmental modelling services.
Environmental models are important tools that enable raw data to be transformed into useful information, through synthesis, simulation and prediction. However, environmental models often consist of complex computer applications, written for a specific operating system, with very specific inputs and outputs and a specific purpose. What is needed is models that are flexible, to respond to very different user enquiries in many different places, and under many different conditions of data availabilities.
This requires a new generation of models that are modular, flexible and adaptable. Instead of building large, monolithic models, we may design components that can interact and be assembled on purpose for a particular application. For instance, a hydrological model may be coupled to a soil chemistry model, or a climate model to a biodiversity model. They should be flexible in their data requirements and process representations, as to adapt to data availability and local idiosyncrasies in the environmental processes. They should also be adaptable, to take advantage of new information once it comes available.
Of course we should show a healthy skepticism towards model results regardless of how sophisticated the model appears to be. It is an aphorism that all models are wrong, though some may be useful. Depending on the quality and quantity of available data and the complexity of the modeled system, errors will propagate through the model, and may be amplified by errors and simplifications in the model structure. Whether the model results are fit for a particular purpose is something only the user can assess. Consequently, the importance of enabling users to have as high a degree of interaction with the modelling process as they would like is critical to the success of the modelling exercise.
To this end, the EVO is providing a unique platform for interaction between the developers and end-users of models, and to providing feedback on the needs, possibilities, and future developments of environmental models. Find out more about how these challenges are being addressed by EVO.