Volume 11, Issue 22 (11-2023)
ifej 2023, 11(22): 57-66 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Bakhtiari S, Rostani Shahraji T, Akhavan R, Ebrahimi Atani R. (2023). Spatial Distribution Modeling of Pistacia atlantica using Artificial Neural Network in Khohir National Park. ifej. 11(22), 57-66. doi:10.61186/ifej.11.22.57
URL: http://ifej.sanru.ac.ir/article-1-515-en.html
URL: http://ifej.sanru.ac.ir/article-1-515-en.html
Guilan University
Abstract: (627 Views)
Introduction and Objective: In vestigating the presence of species in forest habitats is very important in order to locate and identify areas with the ability to plant and successfully restore species and the relationship between the presence of species to environmental factors. In this research, predicting the probability of the presence and absence of Pistacia atlantica Forest species in relation to environmental variables (topography and soil science data) in a part of Khojir National Park of Tehran province with an area of 120 hectares and the modeling was done using artificial neural network and geostatistics.
Material and Methods: Slope, Aspect and Altitude maps were prepared using the Digital Elevation Model (DEM) map of the area, sampling of trees in the studied area was done in the form of regular-random sampling based on a grid of 100 x 150 meters with 61 pieces of 12 samples. Soil sampling was done in 17 sample plots according to the variety of soil conditions by taking the spatial coordinates of the sample plots. Variables of apparent specific gravity, true specific gravity, absorbable potassium, percentage of nitrogen, absorbable phosphorus, percentage of organic carbon, conductivity Electricity, acidity, soil saturation percentage, soil lime percentage, sand percentage, silt percentage and clay percentage were measured in the laboratory. The map of environmental factors of soil variables was prepared with the help of geostatistics and using GS+ software. Then, the multi-layer perceptron artificial neural network model was designed and validated between the environmental features as model inputs and the presence and absence of the Pistacia atlantica as the model output using SPSS Modeler software. Finally, based on the results of the model and the digital map of environmental factors, a prediction map of the probability of the presence and absence of the Pistacia atlantica was prepared using ArcGIS software.
Results: The results indicate that the artificial neural network had a high accuracy (91%) in predicting the probability of the presence and absence of the Pistacia atlantica. It indicate the relationship between the presence of a Pistacia atlantica and the variables of electrical conductivity, apparent specific gravity, direction Geographically, the percentage of nitrogen and height altitude with the importance coefficient is 0.43, 0.21, 0.17, 0.15 and 0.05 respectively. In addition, the agreement of the prediction map with the ground reality map was assessed as good with a Kappa coefficient of 0.651.
Conclusion: The results showed that with acceptable accuracy, it is possible to use the combination of topographic and soil data to estimate the characteristics of the presence of Pistacia atlantica species in the researched forests, and its maps can be used to identify areas prone to the restoration of the habitat of this species.
Material and Methods: Slope, Aspect and Altitude maps were prepared using the Digital Elevation Model (DEM) map of the area, sampling of trees in the studied area was done in the form of regular-random sampling based on a grid of 100 x 150 meters with 61 pieces of 12 samples. Soil sampling was done in 17 sample plots according to the variety of soil conditions by taking the spatial coordinates of the sample plots. Variables of apparent specific gravity, true specific gravity, absorbable potassium, percentage of nitrogen, absorbable phosphorus, percentage of organic carbon, conductivity Electricity, acidity, soil saturation percentage, soil lime percentage, sand percentage, silt percentage and clay percentage were measured in the laboratory. The map of environmental factors of soil variables was prepared with the help of geostatistics and using GS+ software. Then, the multi-layer perceptron artificial neural network model was designed and validated between the environmental features as model inputs and the presence and absence of the Pistacia atlantica as the model output using SPSS Modeler software. Finally, based on the results of the model and the digital map of environmental factors, a prediction map of the probability of the presence and absence of the Pistacia atlantica was prepared using ArcGIS software.
Results: The results indicate that the artificial neural network had a high accuracy (91%) in predicting the probability of the presence and absence of the Pistacia atlantica. It indicate the relationship between the presence of a Pistacia atlantica and the variables of electrical conductivity, apparent specific gravity, direction Geographically, the percentage of nitrogen and height altitude with the importance coefficient is 0.43, 0.21, 0.17, 0.15 and 0.05 respectively. In addition, the agreement of the prediction map with the ground reality map was assessed as good with a Kappa coefficient of 0.651.
Conclusion: The results showed that with acceptable accuracy, it is possible to use the combination of topographic and soil data to estimate the characteristics of the presence of Pistacia atlantica species in the researched forests, and its maps can be used to identify areas prone to the restoration of the habitat of this species.
Type of Study: Research |
Subject:
اکولوژی جنگل
Received: 2023/07/6 | Accepted: 2023/08/13 | Published: 2024/02/3
Received: 2023/07/6 | Accepted: 2023/08/13 | Published: 2024/02/3
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
