Keywords

ecosystem services, bayesian belief network, decision support

Start Date

1-7-2012 12:00 AM

Abstract

While being studied by scientists for decades, the term ecosystem services was only recently introduced to the general public. This introduction intended broad-scale recognition of ecosystems and their value for human well-being. Both quantitative and qualitative research on ecosystem services became emerging topics in scientific research. Ecosystem service prediction models were developed varying from basic qualitative models to complex mechanistic models which enable quantification of ecosystem services. The introduction of Bayesian belief networks in ecosystem service modelling has led to an intermediate approach between both methods. Major advantages of this Bayesian network modelling approach are the model transparency which enables stakeholder involvement in model development and evaluation, the possibility to incorporate both empirical data and expert knowledge, a straightforward combination with valuation studies and the inherent consideration of uncertainties in a transparent way. Our research focuses on the application of Bayesian belief networks to predict the ecosystem services delivered by the Burggravenstroom, a small river catchment located in the Port of Ghent region. This modelling approach enables identification of trade-offs or win-win scenarios between produced ecosystem services, evaluation of different management scenarios, assessment of effects of human interaction and enhanced system understanding.

Share

COinS
 
Jul 1st, 12:00 AM

Modelling ecosystem services using Bayesian belief networks: Burggravenstroom case study

While being studied by scientists for decades, the term ecosystem services was only recently introduced to the general public. This introduction intended broad-scale recognition of ecosystems and their value for human well-being. Both quantitative and qualitative research on ecosystem services became emerging topics in scientific research. Ecosystem service prediction models were developed varying from basic qualitative models to complex mechanistic models which enable quantification of ecosystem services. The introduction of Bayesian belief networks in ecosystem service modelling has led to an intermediate approach between both methods. Major advantages of this Bayesian network modelling approach are the model transparency which enables stakeholder involvement in model development and evaluation, the possibility to incorporate both empirical data and expert knowledge, a straightforward combination with valuation studies and the inherent consideration of uncertainties in a transparent way. Our research focuses on the application of Bayesian belief networks to predict the ecosystem services delivered by the Burggravenstroom, a small river catchment located in the Port of Ghent region. This modelling approach enables identification of trade-offs or win-win scenarios between produced ecosystem services, evaluation of different management scenarios, assessment of effects of human interaction and enhanced system understanding.