A major challenge in environmental modeling is to identify structural changes in the ecosystem across time, i.e., changes in the underlying process that generates the data.
In this paper, we analyze the Baltic Sea food web in order to 1) examine potential unobserved processes that could affect the ecosystem and 2) make predictions on some variables of interest. To do so, dynamic Bayesian networks with different setups of hidden variables (HVs) were built and validated applying two techniques: rolling-origin and rolling-window. Moreover, two statistical inference approaches were compared at regime shift detection: fully Bayesian and Maximum Likelihood Estimation.
Our results confirm that, from the predictive accuracy point of view, more data help to improve the predictions whereas the different setups of HVs did not make a critical difference in the predictions. Finally, the different HVs picked up patterns in the data, which revealed changes in different parts of the ecosystem.
Research news | 2020-11-24
While hurdles remain, the region can build on the shared history of collaboration to tackle food system challenges together, new insight papers show
Research news | 2020-11-19
Stephan Barthel, Oonsie Biggs, Örjan Bodin, Thomas Elmqvist, Carl Folke, Per Olsson, Garry Peterson and Johan Rockström on exclusive list of world’s most influential researchers
Research news | 2020-11-18
Four ways to understand the complexity of global environmental change sufficiently well to take policy action
Research news | 2020-11-16
On the island of Stora Karlsö, the steel built lab gives researchers access to observe the largest seabird colony in the Baltic Sea
Research news | 2020-11-13
The clothing industry is dominated by a clutch of powerful companies, but interest groups around them are crucial in efforts to make it more sustainable
Research news | 2020-11-12
The pandemic is hitting the seafood industry hard. How can past crises help it survive this one?