by Rob Lempert and Jan Kwakkel,
The 8th International Congress on Environmental Modeling and Software in Toulouse, France, on July 10-14, 2016, featured a track titled Advancing in Environmental Decision Making Under Deep Uncertainty: Emerging Tools and Challenges. The track was co-organized by Jan Kwakkel (Delft University of Technology), Patrick Reed (Cornell), Robert Lempert (RAND Corporation), and Marjolijn Haasnoot (Deltares). The track consisted of four sessions with four papers in each session.
- Session 1
The introductory session provided an overview of the various ingredients used by various deep uncertainty methods, including a taxonomy of DMDU methods, scenario discovery on CO2 emissions pathways, illustrations of robust decisionmaking with scenario discovery-derived signposts for adaptive strategies, and the challenges of model-based support for adaptation pathways in Bangladesh
- Session 2
The second session focused on open source methods and tools now available to the DMDU community, which included presentations on the Open MORDM toolkit in R, and the exploratory modeling workbench in Python, a new sampling approach extending the commonly used Latin Hypercube sampling, and iterative discovery
- Session 3
The third session featured recent methodology advances including multi-objective meta heuristics, a demonstration of direct policy search using the lake problem, a comparison of various algorithms for a water allocation case, a hybrid between MORDM and robust optimization for water supply in England, and the use of multi-objective algorithms for supporting urban planning
- Session 4
The fourth session featured various case studies in different domains: agro-ecosystems, socio-ecological systems, electricity system in Switzerland, and real options for flood risk in Bilbao
Overall, the four sessions highlighted the breadth of methods and applications among DMDU researchers, the increasingly capable tools available to the community, and exciting new advances. The 9th International Congress on Environmental Modeling and Software will be held at Colorado State University to host iEMSs 2018 in Fort Collins, Colorado, with hopefully even more DMDU work on display.