Abstract:The Zeribar wetland in Iran's western city of Marivan provides a valuable foundation for maintaining various aspects of social and cultural life, from water supply to recreational values, as well as for natural habitats. However, its different services have been put under strong anthropogenic pressure and urbanization processes. The present study is a part of ongoing efforts to improve understanding of the anthropogenic factors potential impact on management status of this vulnerable freshwater ecosystem, aiming to provide integrated management strategies using a Bayesian belief network (BBN) approach. A conceptual framework supporting the BBN was designed according to the main determining factors of the wetland's management including "human-induced factors", "economic factors and human capital", "policy context" and "social and cultural issues". The BBN model was parameterized using different data sources comprising literature review, face-to-face interviews, filed observation, model simulations, and experts' knowledge. Stakeholders' feedback on model accuracy and management strategies application was assessed in different workshops with representatives from locals, executives, and scientific communities. The results showed that the two subsystems of "policy context" and "economic factors and human capital" are the most important key factors influencing the target node, wetland management status. Then, fourteen alternative management scenarios (S2- S15) were generated using the improvement in prior probabilities of influencing factors, followed by evaluation and statistical comparison with the output of the current condition as the base-case scenario (S1). Under the S1 scenario, the probability of occurrence of the "Strong" state, in which the wetland is properly protected from anthropogenic disturbances, for the target node of the network was 25.8 %, while the most significant improvement in the wetland management status (66.4%) was observed after an improvement in all influencing variables was adopted (S15), suggesting a significant potential for improvement in the wetland management. According to the stakeholders' feedback, the constructed BBN model, and the components included in the management scenarios are useful to fill a critical gap in the current management framework. The results of this work can assist decision-makers in managing the wetland more effectively by considering all socio-economic and anthropogenic dynamics.

Human Activities Impact Prediction in Vegetation Diversity of Lar National Park in Iran Using Artificial Neural Network Model

Environmental modeling of landscape aesthetic value in natural urban parks using artificial neural network technique

Land use change modeling through an integrated Multi-Layer Perceptron Neural Network and Markov Chain analysis (case study: Arasbaran region, Iran)

Soil texture and plant degradation predictive model (STPDPM) in national parks using artificial neural network (ANN)

Prediction of hypericin content in Hypericum perforatum L. in different ecological habitat using artificial neural networks

Landscape aesthetic quality assessment of forest lands: an application of machine learning approach

Aesthetic preference and mental restoration prediction in urban parks: An application of environmental modeling approach

Monitoring of Plant Ecological Units Cover Dynamics in a Semiarid Landscape from Past to Future Using Multi-Layer Perceptron and Markov Chain Model

Modelling Temperature Variation of Mushroom Growing Hall Using Artificial Neural Networks

Noise pollution prediction and seasonal comparison in urban parks using a coupled GIS- artificial neural network model

Deforestation Prediction Using Neural Networks and Satellite Imagery in a Spatial Information System

Prediction of landslides by machine learning algorithms and statistical methods in Iran

Assessing the performance of machine learning algorithms for analyzing land use changes in the Hyrcanian forests of Iran

Spatial predication of flood zonation mapping in Kan River Basin, Iran, using artificial neural network algorithm

Evaluating the effects of vegetation and land management on runoff control using field plots and machine learning models

Aesthetic quality modeling of the form of natural elements in the environment of urban parks

Prediction of Food Production Using Machine Learning Algorithms of Multilayer Perceptron and ANFIS

Forest landscape visual quality evaluation using artificial intelligence techniques as a decision support system

Vegetation Dynamics Assessment: Remote Sensing and Statistical Approaches to Determine the Contributions of Driving Factors

Predicting the impacts of anthropogenic drivers on management scenarios using Bayesian belief network in the Zeribar freshwater wetland, Iran

Machine Learning Models of Exergoenvironmental Damages and Emissions Social Cost for Mushroom Production