Document Type : Research Paper

Authors

• Hannaneh Sadat Sadat Mousavi ¹
• Ali Jahani ²
• Afshin Danehkar ³
• vahid etemad ⁴
• Farnoush Attar Sahragard ⁵

¹ Department of Environmental Science, Faculty of Natural Resources, University of Tehran, Karaj, Iran

² College of Environment

³ Chamran Ave.

⁴ Tehran university/Head of Department of forestry

⁵ M.Sc. Student of Environmental Science and Engineering, College of Natural Resources, University of Tehran

Abstract

Vegetation plays an important role in maintaining the balance of ecosystems and is considered as a significant indicator in assessing the environment of terrestrial ecosystems. The coverage of vegetation is considered as one of the important characteristics for the assessment of vegetation. Vegetation changes cause disorder in the ecosystem and act as a stimulus to change the composition of species and habitat conditions. The present research was conducted with the aim of modeling the effect of human activities on the coverage of vegetation with the help of artificial neural network and determining the most effective ecological and human variables on the coverage in the central Alborz protected area under the management of Alborz province. For this purpose, after the characterizing homogeneous ecological units, 101 plant square plots with dimensions of 2 x 2 meters and 101 soil samples were collected in these units. soil and vegetation analyzes were performed on them. Finally, modeling was done using multi-layer perceptron method and input variables including soil physical, chemical, physiographic (based on the management plan maps), and related to human factors variables (based on the Euclidean distance map of the sampling points). According to the values of the coefficient of determination in the three categories of training, validation and test data equal to 0.86, 0.79 and 0.81, the optimal structure of the model for the coverage of vegetation with a structure of 1-19-18 (18 input variables, 19 neurons in the hidden layer and one output variable) was selected. Based on the results of the sensitivity analysis, slope variables, the percentage of organic matter and height are the most effective variables on the coverage of vegetation in the study area. The model presented in this research is used as a decision support system in assessing the effects of human activities on the coverage of vegetation in protected areas and provides the possibility of predicting the extent of these effects on the coverage of vegetation in protected areas.

Keywords

• Environmental factors
• Vegetation
• Coverage
• Artificial neural network
• Multilayer perceptron