Prediction of dust storm using artificial neural networks in Kermanshah

Dust is a phenomenon with significant environmental impacts across various aspects of human life, including agriculture, economy, health, and more. The purpose of this study is to investigate and predict the dust phenomenon in Kermanshah. Meteorological data with a 3-hour resolution for the statisti...

Full description

Saved in:
Bibliographic Details
Main Authors: Toba Alizadeheh, Majid rezaie banafsh, Gholamreza Goodarzi, Hashem Rostamzadeh
Format: Article
Language:fas
Published: Kharazmi University 2025-09-01
Series:تحقیقات کاربردی علوم جغرافیایی
Subjects:
Online Access:http://jgs.khu.ac.ir/article-1-4007-en.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849312988010381312
author Toba Alizadeheh
Majid rezaie banafsh
Gholamreza Goodarzi
Hashem Rostamzadeh
author_facet Toba Alizadeheh
Majid rezaie banafsh
Gholamreza Goodarzi
Hashem Rostamzadeh
author_sort Toba Alizadeheh
collection DOAJ
description Dust is a phenomenon with significant environmental impacts across various aspects of human life, including agriculture, economy, health, and more. The purpose of this study is to investigate and predict the dust phenomenon in Kermanshah. Meteorological data with a 3-hour resolution for the statistical period (2000–2020) from the Kermanshah station was obtained from the Meteorological Organization. First, the dust data were normalized, and then Artificial Neural Network (ANN) models were used to predict dust concentration, while the Adaptive Neuro-Fuzzy Inference System (ANFIS) was employed to analyze and predict the time series of dust occurrence in MATLAB software. The findings revealed that the maximum predicted dust concentration, related to the minimum dew point with the highest Pearson correlation with dust, was estimated at 3451.23 µg/m³. Additionally, the results of the time series prediction using the ANFIS model showed that the linear bell membership function with grade 3, during both the training and testing stages, was the most effective input function among other membership functions. According to the forecasting models, the highest probability of maximum dust occurrence in the next 20 years in Kermanshah is 94%. Based on the aforementioned studies, sufficient information was gathered to conduct this research. The phenomenon of dust, particularly in western Iran and the city of Kermanshah, has consistently posed significant challenges for the residents of these areas. This phenomenon is influenced by specific atmospheric conditions that cause irreparable damage annually, leading to respiratory issues and deteriorating air quality. Therefore, it is essential to pay serious attention to the issue of dust.
format Article
id doaj-art-db70d05ba10c456b8eb8953d1b9283a8
institution Kabale University
issn 2228-7736
2588-5138
language fas
publishDate 2025-09-01
publisher Kharazmi University
record_format Article
series تحقیقات کاربردی علوم جغرافیایی
spelling doaj-art-db70d05ba10c456b8eb8953d1b9283a82025-08-20T03:52:52ZfasKharazmi Universityتحقیقات کاربردی علوم جغرافیایی2228-77362588-51382025-09-012578278297Prediction of dust storm using artificial neural networks in KermanshahToba Alizadeheh0Majid rezaie banafsh1Gholamreza Goodarzi2Hashem Rostamzadeh3 Dust is a phenomenon with significant environmental impacts across various aspects of human life, including agriculture, economy, health, and more. The purpose of this study is to investigate and predict the dust phenomenon in Kermanshah. Meteorological data with a 3-hour resolution for the statistical period (2000–2020) from the Kermanshah station was obtained from the Meteorological Organization. First, the dust data were normalized, and then Artificial Neural Network (ANN) models were used to predict dust concentration, while the Adaptive Neuro-Fuzzy Inference System (ANFIS) was employed to analyze and predict the time series of dust occurrence in MATLAB software. The findings revealed that the maximum predicted dust concentration, related to the minimum dew point with the highest Pearson correlation with dust, was estimated at 3451.23 µg/m³. Additionally, the results of the time series prediction using the ANFIS model showed that the linear bell membership function with grade 3, during both the training and testing stages, was the most effective input function among other membership functions. According to the forecasting models, the highest probability of maximum dust occurrence in the next 20 years in Kermanshah is 94%. Based on the aforementioned studies, sufficient information was gathered to conduct this research. The phenomenon of dust, particularly in western Iran and the city of Kermanshah, has consistently posed significant challenges for the residents of these areas. This phenomenon is influenced by specific atmospheric conditions that cause irreparable damage annually, leading to respiratory issues and deteriorating air quality. Therefore, it is essential to pay serious attention to the issue of dust.http://jgs.khu.ac.ir/article-1-4007-en.pdfdustforecastannanfiskermanshah.
spellingShingle Toba Alizadeheh
Majid rezaie banafsh
Gholamreza Goodarzi
Hashem Rostamzadeh
Prediction of dust storm using artificial neural networks in Kermanshah
تحقیقات کاربردی علوم جغرافیایی
dust
forecast
ann
anfis
kermanshah.
title Prediction of dust storm using artificial neural networks in Kermanshah
title_full Prediction of dust storm using artificial neural networks in Kermanshah
title_fullStr Prediction of dust storm using artificial neural networks in Kermanshah
title_full_unstemmed Prediction of dust storm using artificial neural networks in Kermanshah
title_short Prediction of dust storm using artificial neural networks in Kermanshah
title_sort prediction of dust storm using artificial neural networks in kermanshah
topic dust
forecast
ann
anfis
kermanshah.
url http://jgs.khu.ac.ir/article-1-4007-en.pdf
work_keys_str_mv AT tobaalizadeheh predictionofduststormusingartificialneuralnetworksinkermanshah
AT majidrezaiebanafsh predictionofduststormusingartificialneuralnetworksinkermanshah
AT gholamrezagoodarzi predictionofduststormusingartificialneuralnetworksinkermanshah
AT hashemrostamzadeh predictionofduststormusingartificialneuralnetworksinkermanshah