Design and implementation of a web-based decision support system for climate change impact assessment on forests
|Title||Design and implementation of a web-based decision support system for climate change impact assessment on forests|
|Year of Publication||2016|
|Authors||Thiele J.C., Nuske R.S.|
|Keywords||Climate change adaption, DSS implementation, Ecological modelling, forest management, Model coupling, Risk assessment, Software architecture|
Climate change has altered and further will change the environmental conditions for many sectors. Whereas annual cropping systems can be adapted yearly, decisions in forest management usually have long-lasting effects. Depending on the region and tree species rotation periods range between 80 and 180 years in Central Europe. Therefore, todays tree species selection should also take into account the impacts of climate change in the future. Although many attempts have been made to understand single aspects of climate change impacts on forests, the available knowledge has to be conflated into an integrated assessment to support decision making. A comprehensive system comprising different impact models and an economic assessment suitable for Central European forests and driven by high-resolution temporal- and spatial data is currently missing. We present a conceptional design and a reference implementation of a Decision Support System (DSS) tailored to assess climate change impacts on German forests. To provide high ease-of-use, the system was implemented as a web application and offers information for single stands on-demand as well as interactive maps and preprocessed assessments on a coarser level for entire Germany. To create such a complex, integrated system from legacy models written in different programming languages the interfaces had to be developed carefully. The key of this DSS is its building blocks: established models describing different climate change impacts. Since the DSS is a very modular system, it is easy to replace submodels and to adapt it to other study areas, forest systems or research questions if suitable parameterized models and input data are available. The presented technical solution is adaptable to other systems integrating existing models and the source code is available.