A carregar...
Publicação
Predicting the ecological status of rivers and streams under different climatic and socioeconomic scenarios using Bayesian Belief Networks
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.25 MB | Adobe PDF |
Orientador(es)
Resumo(s)
Freshwater systems have increasingly been subjected to a multitude of human pressures and the re-establishment
of their ecological integrity is currently a major worldwide challenge. Expected future climate and socioeconomic
changes will most probably further exacerbate such challenges. Modelling techniques may provide
useful tools to help facing these demands, but their use is still limited within ecological quality assessment of
water resources due to its technical complexity.
We developed a Bayesian Belief Network (BBN) framework for modelling the ecological quality of rivers and
streams in two European river basins located in two distinct European climatic regions: the Odense Fjord basin
(Denmark) and the Sorraia basin (Portugal). This method enabled us to integrate different data sources into a
single framework to model the effect of multiple stressors on several biological indicators of river water quality
and, subsequently, on their ecological status. The BBN provided a simple interactive user interface with which
we simulated combined climate and socioeconomic changes scenarios to assess their impacts on river ecological
status.
According to the resulting BBNs the scenarios demonstrated small impacts of climate and socioeconomic
changes on the biological quality elements analysed. This yield a final ecological status similar to the baseline in
the Odense case, and slightly worse in Sorraia. Since the present situation already depicts a high percentage of
rivers and streams with moderate or worse ecological status in both basins, this means that many of them would
not fulfil the Water Framework Directive target in the future. Results also showed that macrophytes and fish
indices were mainly responsible for a non-desirable overall ecological status in Odense and Sorraia, respectively.
The approach followed in this study is novel, since BBN modelling is used for the first time for assessing the
ecological status of rivers and streams under future scenarios, using an ensemble of biological quality elements.
An important advantage of this tool is that it may easily be updated with new knowledge on the nature of
relationships already established in the BBN or even by introducing new causal links. By encompassing two case
studies of very different characteristics, these BBN may be more easily adapted as decision-making tools for
water management of other river basins
Descrição
Palavras-chave
Bayesian Belief Network ecological status global change rivers streams scenarios
Contexto Educativo
Citação
Limnologica 80 (2020) 125742
Editora
Elsevier