University of Twente Student Theses
A Bayesian belief network approach to map seasonal changes in ecosystem services trade-offs and synergies
Sood, Aanchal (2020) A Bayesian belief network approach to map seasonal changes in ecosystem services trade-offs and synergies.
|
PDF
9MB |
| Abstract: | To gain maximum benefits from nature, there is a need to identify the relationship among different interacting services and acknowledging that these relationships might not be constant over time. This information can help to make well-informed and efficient management decisions, that can minimise the trade-offs and maximise the benefits retrieved from nature. This explorative study used a spatial Bayesian Belief Network (BBN) to assess its ability to map seasonal variations in the trade-offs and synergies between different ecosystem services (ESS). The output maps from Bayesian Belief Network can help to identify and prioritise areas for adaptive management, keeping in mind the underlying parameter uncertainties. Though several methods successfully have identified and quantified relationships among services, yet not many studies have explored the potential of the BBN to map ESS trade-offs and synergies. Therefore, this research presents a way to identify and map the interactions (trade-offs, synergies, and no-effects) among ESS using expert knowledge and then see how these interactions vary over seasons. In this regard, a case study area was selected as Mt. Oiti National Park in Central Greece. The creation of the network was based on data collected from twelve experts in the face - to - face semi-structured interviews. The interviews also included participatory mapping for collecting spatial data. It can be seen from the results that Bayesian Belief Network can be used for generating ecosystem services interactions maps which may assist the decision-makers in making well-informed decisions. This can lead to effective and efficient utilisation of resources as it allows to identify areas of priority that need immediate attention during different seasons. It was found that there was a seasonal pattern in ESS pairs such that the services may have synergy or trade-off during one season and no-effect in the other. The outputs also help to visualise the change in the magnitude of interactions among four ecosystem services over the season. The interactions among provisioning service as timber extraction and cultural service as recreational canyoning, hiking and hunting services, show the ability of Bayesian Belief Network to combine qualitative and quantitative information captured for the four ecosystem services in a single model and capture relationships among them. Synergy was observed between recreational ecosystem services like hiking and canyoning and also provisional service timber extraction and cultural service as recreational canyoning service. Trade-offs are identified among cultural ESS like recreational hunting, and recreational hiking and also recreational canyoning. The most influential relationship is from provisional ecosystem services: timber extraction to all other services. It was found that this provisional service impacts all other services but is not influenced back by any other service. The Bayesian Belief Networks are suitable for applications like this study where there is a small dataset available with large uncertainty which impacts decision-making considerably. Altogether, the developed model was the first attempt in modelling the seasonal variations in interactions among the ecosystem services focusing on their flow component. Keywords: Bayesian belief network; ecosystem services flow; ecosystem services interactions; trade-offs; synergies; no-effect; uncertainty; expert interviews; mapping. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/90787 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page