Desert (Dec 2023)

Bayesian network application in socio-economic and ecological effects analysis in watershed management

  • A. Afzali,
  • K. Solaymani,
  • Sh. Rastgar,
  • A. Keshtkar

DOI
https://doi.org/10.22059/jdesert.2023.96217
Journal volume & issue
Vol. 28, no. 2
pp. 403 – 413

Abstract

Read online

Natural resource management using appropriate management tools, is one of the critical issues in the watershed, especially in arid and semi-arid areas, land management and its associated resources will create a balance between economic and social needs and the sustainability of biological ecosystems. The present study, with the aim of watershed management and controlling sedimentation rates, evaluated socio-economic effects in the Damghan Rud watershed. The study area with an area of 5.66314 ha is located in Semnan Province. By choosing four economic and social activity variables, along with rangeland vegetation cover, were identified management options for the Damghan Rud basin. Based on the regional conditions and study objectives, four economic and social parameters the presence of surplus livestock, reliance of watershed residents on rangeland, Literacy of watershed residents’ level, and their participation percentage in water and soil resource conservation were determined as effective options for rangelands vegetation cover in the region. The sediment production rate at the output site of the selected basin at the Astana hydrometric station was investigated in the statistical period of 1995-96 to 2015-16, then by using the frequency distribution diagram of observational sediment, determined its classes. Then, with the implementation of the model were calculated the probabilities associated with each variable, and were evaluated the possible effects of the implementation of managerial options on the middle and target variables. The results showed that the option of reducing the reliance of watershed residents on rangelands to the minimum level led to a 15.3% increase in vegetation cover, a 5.5% reduction in river flow rate, and a 3.6% decrease in sediment production rate. Furthermore, the results showed that the Bayesian network models had a high capability and ability to express various dimensions of the issue and handle uncertainties within the system, and making them suitable tools for watershed resources management.

Keywords