Bayesian Network Modelling Provides Spatial and Temporal Understanding of Ecosystem Dynamics within Shallow Shelf Seas

Date: Wednesday 17 November 2021, 3.00PM
Location: Online
Online - joining instructions will be sent shortly before the event
Local Group Meeting


Share this event

Bayesian Network Modelling Provides Spatial and Temporal Understanding of Ecosystem Dynamics within Shallow Shelf Seas

Neda Trifonova (Aberdeen)

Abstract: There is about to be an abrupt step-change in the use of our coastal seas, specifically by the addition of large-scale offshore renewable energy developments to combat climate change. Many trade-offs will need to be weighed up for the future sustainable management of marine ecosystems between renewables and other uses (e.g., fisheries, marine protected areas). Therefore, we need a much greater understanding of how different marine habitats and ecosystems are likely to change with both natural and anthropogenic transformations. Ecosystems consist of complex dynamic interactions among species and the environment, the understanding of which has implications for predicting the environmental response to changes in climate and biodiversity. However, with the recent adoption of more explorative tools, like Bayesian networks, in predictive ecology, few assumptions can be made about the data and complex, spatially varying interactions can be recovered from collected field data. In this talk, Bayesian techniques will be presented to find the data-driven estimates of interactions among a set of physical and biological variables and a human pressure within the last 30 years in a well-studied shallow sea (North Sea, UK). A hidden variable is incorporated to model functional ecosystem change, where the underlying interactions dramatically change, following natural or anthropogenic disturbance. Then, the learned data-driven interactions will be used to build a dynamic Bayesian network model to examine the response of species to changes in their environment. Other examples will also be illustrated to show the applications of Bayesian network techniques in predictive ecology.

 
Neda Trifonova (Aberdeen)
 
Yinghui Wei (yinghui.wei@plymouth.ac.uk) and Matthew Craven for RSS South West Local Group