دانشگاه اصفهان جغرافیا و برنامه ریزی محیطی 2008-5362 37 1 2026 03 21 Received: 30/11/2025 Accepted: 26/04/2026 Landslide Assessment and Prediction Using Advanced Machine-Learning Algorithms (Case Study: Alvand Watershed) ارزیابی و پیش‌بینی حساسیت وقوع زمین‌لغزش با استفاده از الگوریتم‌های پیشرفته یادگیری ماشین (محدوده مورد ‌‌‍مطالعه: حوضه آبخیز الوند) 1 22 30370 10.22108/gep.2026.147111.1752 FA سجاد تبارکی دانش‌آموخته ‌کارشناسی ارشد‌ سنجش از دور وGIS، دانشکده‌ علوم جغرافیایی و برنامه‌ریزی، دانشگاه اصفهان، اصفهان، ایران رضا ذاکری نژاد استادیار گروه جغرافیای طبیعی، دانشکده علوم جغرافیایی و برنامه‌ریزی، دانشگاه اصفهان، اصفهان، ایران ایمان خسروی استادیار گروه مهندسی نقشه‌برداری، دانشکده مهندسی عمران و حمل و نقل، دانشگاه اصفهان، اصفهان، ایران Journal Article 2025 11 30 Abstract Natural hazards are among the most pressing environmental problems, causing annual fatalities, injuries, and homelessness for millions worldwide. Landslides represent one of the most destructive natural hazards, threatening human lives and inflicting substantial damage to natural resources. Unlike other natural events that occur probabilistically over time, landslides have the potential for instantaneous occurrence. In mountainous regions and steep slopes, landslides not only lead to loss of life, but also alter and degrade the environment each year. This study evaluated and predicted landslide hazard in the Alvand watershed of Kermanshah using data mining algorithms. 14 parameters were employed to model the probability of landslide hazard: elevation, slope, slope aspect, geology, land use, Topographic Wetness Index (TWI), Stream Power Index (SPI), distance from stream, drainage density, distance from fault, fault density, mean annual rainfall, distance from road, and Normalized Difference Vegetation Index (NDVI). 4 data-mining algorithms—Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Rotation Forest (RoF)—were applied for modeling. Based on the results obtained from Overall Accuracy (OA) and the Kappa coefficient, the models achieved the following performance: DT (OA = 82.91%, Kappa = 0.78), RF (OA = 96.47%, Kappa = 0.93), RoF (OA = 87.47%, Kappa = 0.82), and SVM (OA = 93.90%, Kappa = 0.90). These findings demonstrated the high accuracy of the RF and SVM methods in predicting landslide hazard in the Alvand watershed. Keywords: Landslide, Data-Mining, Alvand Watershed. Introduction Natural hazards are among the environmental problems that annually lead to the death and injury of many people and cause homelessness for millions around the world. Landslides, in particular, are one of the most destructive natural hazards. In addition to threatening human lives, they cause further damage to natural resources and exert pressure on the economic development of countries, especially developing nations. The financial losses resulting from natural hazards can lead to stagnation in economic growth and prosperity. Slope instability in steep mountainous lands is a major challenge for land managers worldwide. Furthermore, landslides result from complex and diverse processes and are considered among the most serious and destructive natural hazards in mountainous regions globally. A distinctive feature of landslides is their potential for instantaneous occurrence. The Alvand watershed in Kermanshah Province, given its topographic, climatic, and morphological conditions, is one of the areas where investigating landslide probability is essential. The aim of this research was to zone areas susceptible to landslide occurrence and compare the performance of Decision Tree (DT), Random Forest (RF), Rotation Forest (RoF), and Support Vector Machine (SVM) models for landslide prediction in the study area The Alvand River watershed is located in the southwest of Kermanshah Province. The Alvand River is a border river between Iran and Iraq and the region lies between Mesopotamia and the Iranian plateau. The study area covers 112,332.77 hectares and is situated between 34°34' to 34°15' N latitude and 45°35' to 46°09' E longitude. Elevation in the basin ranges from 329 to 2,672 meters above sea level with a mean elevation of 1,098 meters. Materials & Methods To model the probability of landslide hazard, 14 physiographic, hydrological, geological, and environmental parameters were selected based on a review of relevant literature and the specific characteristics of the study area. These parameters included elevation, aspect, slope, land use and land cover, Topographic Wetness Index (TWI), Stream Power Index (SPI), stream density, fault density, distance from faults, average annual rainfall, distance from roads, vegetation density index (NDVI), and geology. All parameter layers were converted into raster format with a consistent spatial resolution and were resampled to ensure uniform geographic extent and cell size for integration into the modeling process. In this research, 4 efficient data-mining algorithms were employed for modeling: Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Rotation Forest (RoF). These algorithms were selected due to their proven performance in landslide susceptibility studies and their ability to handle high-dimensional, non-linear environmental data. The landslide inventory map was prepared using field surveys, historical records, and interpretation of satellite imagery. A total of landslide locations were identified and randomly divided into two independent datasets: training samples (70% of the data) and test samples (30% of the data). The training samples were used to calibrate the models by learning the relationships between the landslide occurrence and the fourteen conditioning factors. The test samples, which were not involved in the model building process, were used to evaluate the predictive accuracy and generalizability of the developed models. Model performance was assessed using the Overall Accuracy (OA) and the Kappa coefficient, both of which were derived from the confusion matrix comparing predicted and observed landslide occurrences in the test dataset. Research Findings Based on the results obtained from the Overall Accuracy (OA) and Kappa coefficient, the performances of the four models were as follows: DT Model: OA = 82.91%, Kappa = 0.78 RF Model: OA = 96.47%, Kappa = 0.93 RoF Model: OA = 87.47%, Kappa = 0.82 SVM Model: OA = 93.90%, Kappa = 0.90 These findings indicated the high accuracy of the RF and SVM methods in predicting landslide hazard in the Alvand watershed. Landslide susceptibility maps of the study area were generated by using all four methods. Discussion of Results & Conclusion The performances of the DT and RoF methods were comparatively weaker than those of the other two algorithms. In contrast, the best results were achieved by using the RF and SVM methods. The high accuracy exceeding 90% (97% for RF and 94% for SVM), along with their Kappa coefficients above 0.9, indicated excellent modeling performance for landslide susceptibility zoning in the study area. The superior performance of these two ensemble-based and kernel-based methods could be attributed to their robustness in handling complex, non-linear relationships among the fourteen conditioning factors, as well as their abilities to reduce overfitting compared to single decision tree models, such as DT. In conclusion, this study demonstrated the high predictive accuracy of the RF and SVM models for landslide hazard assessment in the Alvand watershed. The outstanding accuracy—above 90% for both methods—coupled with their high Kappa values (>0.9), confirmed the suitability of these two approaches for landslide hazard zoning in the region. Future research could explore the integration of deep learning models or the application of these methods to other watersheds with similar topographic and climatic conditions to further validate their generalizability. زمین‌لغزش یکی از مخاطرات طبیعی تخریب‌کننده است که علاوه بر تهدید زندگی انسان‌ها، آسیب جدید به منابع طبیعی وارد می‌کند. پدیده زمین‌لغزش دارای پتانسیل وقوع لحظه‌ای است؛ درحالی‌که سایر حوادث و مخاطرات طبیعی هرچند وقت یک‌بار احتمال وقوع آن وجود دارد. در این پژوهش با استفاده الگوریتم‌های داده‌کاوی به ارزیابی و پیش‌بینی خطر زمین‌لغزش در حوضه آبخیز الوند در استان کرمانشاه پرداخته شد. برای مدلسازی احتمال خطر زمین‌لغزش از 14 پارامتر استفاده شده است. این پارامترها عبارتند از ارتفاع، شیب، جهت شیب، زمین‌شناسی، کاربری اراضی، شاخص رطوبت توپوگرافی (TWI)، شاخص قدرت آبراهه (SPI)، فاصله از آبراهه، تراکم آبراهه، فاصله از گسل، تراکم گسل، میانگین بارش سالانه، فاصله از جاده و شاخص تراکم پوشش گیاهی (NDVI) استفاده شده است. در ادامه این تحقیق، چهار الگوریتم کارای داده‌کاوی شامل روش ماشین بردار پشتیبان (SVM)، درخت تصمیم (DT)، جنگل تصادفی (RF) و جنگل دورانی (RoF) برای مدلسازی به کار گرفته شد. با توجه به نتایج به‌دست‌آمده از شاخص‌های آماری دقت کلی و ضریب کاپا مدل‌های DT، RF، RoF و SVM دارای دقت کلی به‌ترتیب 91/82، 47/96، 47/87 و 90/93 درصد و ضریب کاپا 78/0، 93/0، 82/0، 90/0 و 90/0 هستند که این نتایج نشان از دقت زیاد روش RF و SVM در پیش‌بینی خطر زمین‌لغزش در حوضه آبخیز الوند است.

https://gep.ui.ac.ir/article_30370_817d85d0f28e7a01fa4778c93d45016f.pdf

دانشگاه اصفهان جغرافیا و برنامه ریزی محیطی 2008-5362 37 1 2026 03 21 Monitoring and Investigating Changes in Water Use Efficiency (WUE) of Different Climates in Kerman Province پایش و بررسی تغییرات کارایی مصرف آب اقلیم‌های مختلف استان کرمان 23 42 30424 10.22108/gep.2026.146361.1742 FA علی مهرابی دانشیار گروه جغرافیا و برنامه‌ریزی شهری، دانشکده ادبیات و علوم انسانی، دانشگاه شهید باهنر کرمان، کرمان، ایران Journal Article 2025 08 19 Abstract Water Use Efficiency (WUE) is a key indicator for understanding the dynamics of carbon and water cycles across different climatic regimes, which is essential for assessing how ecosystems adapt and respond to complex environmental changes. In this study, Gross Primary Production (GPP) and Evapo-Transpiration (ET) data derived from satellite imagery were used to investigate WUE across all climatic zones—humid, semi-humid, semi-arid, and arid—in Kerman Province from 2004 to 2014. To evaluate the influence of various environmental factors—including temperature, precipitation, solar radiation, vapor pressure deficit, Leaf Area Index (LAI), and soil moisture—the geographic detector method and partial correlation analysis were applied. The results showed that over the study period, GPP and ET increased at average rates of 0.0065 g C/m²/year and 2.23 kg H2 O/m²/year, respectively, while WUE decreased at an average rate of 0.01 g C/kg H₂O/m²/year. Among the environmental factors examined, LAI made the largest contribution to WUE in most climatic regions. Furthermore, precipitation and solar radiation emerged as the most important primary climatic drivers influencing the development of climates in the central humid and semi-humid regions, as well as the central and western humid and semi-humid regions of the province, respectively. Keywords: Water Use Efficiency (WUE), Climate Change, Evapo-Transpiration (ET), Gross Primary Productivity (GPP), Kerman Province. Introduction Water Use Efficiency (WUE) is an indicator that links plant carbon gain to water loss through evapotranspiration, operating within the interconnected plant–soil–atmosphere continuum. At the ecosystem level, WUE is commonly defined as the ratio of Gross Primary Production (GPP) to Evapo-Transpiration (ET), representing the amount of organic carbon sequestered per unit of water lost. This metric offers valuable insights for sustainable ecosystem management aimed at optimizing water resource use. WUE is regulated by vegetation traits—such as stomatal conductance and intercellular CO2 concentration—as well as by various environmental factors. Key environmental drivers affecting plant ecosystems include temperature, precipitation, solar radiation, and vapor pressure deficit. Each of these factors has a critical threshold in relation to WUE and these thresholds vary across ecosystems due to differing local environmental conditions. For instance, an increase in average precipitation generally enhances WUE, whereas excessive precipitation may suppress it. In this study, remotely sensed GPP and ET data were used to estimate the WUE of terrestrial ecosystems in Kerman Province. Partial correlation analysis and the geographic detector method were employed to explore the relationships between WUE and its driving factors. The primary objectives of this study were to investigate the spatiotemporal changes in WUE from 2004 to 2024 and identify the dominant driving factors, along with their spatial characteristics that influenced the adaptation and development of the region's ecosystems. Materials & Methods The climatic drivers analyzed in this study included temperature (TEM), precipitation (PRE), and solar radiation (RAD). Temperature and precipitation datasets were obtained from the WorldClim database consisting of monthly averages covering the period from 2004 to 2024. Solar radiation data were derived from the net Surface Solar Radiation (SSR) product of the European Centre for Medium-Range Weather Forecasts (ERA5). Leaf Area Index (LAI) data were obtained from the MOD17 product of the MODIS sensor, which provided 8-day interval data at a spatial resolution of 500 m, spanning from February 2000 to the present. Vapor Pressure Deficit (VPD) data were derived from the TerraClimate dataset, a comprehensive monthly climate product covering global land areas. Soil Moisture (SM) data were obtained from a global surface soil moisture dataset with a spatial resolution of 500 m, covering the period from 2000 to the present. The geographic detector method was employed to detect spatial heterogeneity across geographical areas and identify its driving factors. This method enabled the detection of interactions between two dependent variables, as well as the magnitude and direction of such interactions. In this study, the geographic detector was used to examine the spatial heterogeneity WUE and identify the underlying driving forces. Research Findings To identify the key drivers influencing GPP, ET, and WUE, a partial correlation analysis was conducted to examine the relationships between these three variables and the six environmental factors. The results indicated that PRE had a significantly stronger influence on GPP, ET, and WUE compared to the other climatic factors and exhibited distinct spatial distribution patterns. Precipitation was positively correlated with GPP, WUE, and ET in the central regions, particularly in humid and semi-humid areas. In contrast, a negative correlation between precipitation and these indices was observed in the southern semi-arid and arid regions. Solar radiation was the second most important factor affecting ET, GPP, and WUE. Spatially, positive correlations between solar radiation and both GPP and ET were mainly distributed in the humid and semi-humid regions of the central and western parts of the province. Conversely, a significant negative correlation between solar radiation and WUE was observed in the southern part of the study area. The influence of solar radiation on WUE was most pronounced in the humid and semi-humid central and eastern regions, where the average correlation exceeded 0.65. The effect of TEM on ET and WUE was relatively weak, except in the humid and semi-humid central and eastern regions, where the average correlation exceeded 0.2. Temperature showed a positive and relatively strong correlation with GPP although a negative correlation was observed in some western semi-arid regions. The responses of GPP, ET, and WUE to LAI were significantly stronger than those to other drivers, indicating an overall strong positive correlation. With the exception of scattered negative correlations in the western and southern arid and semi-arid regions, most areas exhibited significant positive correlations. The spatial patterns of soil surface moisture influence on the three indices indicated a generally low correlation. In contrast, the spatial distributions of the effects of Vapor Pressure Deficit (VPD) on GPP and ET showed a high degree of consistency, with positive correlations prevailing in the southern semi-arid regions. The influence of VPD on WUE was particularly prominent in the western semi-humid region, where a positive correlation dominated. Discussion of Results & Conclusion This study qualitatively and quantitatively assessed the effects of climatic and ecological environmental factors using partial correlation analysis and the geographic detector method. The influences of the various driving factors on Gross Primary Production (GPP), Evapo-Transpiration (ET), and Water Use Efficiency (WUE) exhibited clear spatial distribution patterns. Among all factors, Leaf Area Index (LAI) had the greatest impact on GPP, ET, and WUE. Specifically, LAI was more strongly correlated with GPP because changes in LAI directly alter photosynthetic rates, thereby effectively influencing vegetation productivity. Reduced rainfall resulting from successive droughts—particularly in the arid and semi-arid regions of the province—had a notable impact on changes in ET and GPP. Conversely, more favorable rainfall conditions in the humid and semi-humid regions had positively influenced GPP through their effects on vegetation. However, as the vegetation in these regions was predominantly grassland, the increase in GPP remained relatively limited, which in turn led to a comparatively weak response in WUE. The influence of solar radiation on GPP, ET, and WUE was more pronounced in humid and semi-humid regions, where water resources were more abundant. In these areas, increased solar radiation not only accelerated plant transpiration by providing energy for water vapor evaporation, but also effectively regulated vegetation photosynthetic rates, thereby enhancing GPP. However, because the increase in ET was substantially greater than that in GPP, the net effect on WUE was negatively correlated. In this study, we investigated the spatiotemporal changes in WUE in Kerman Province from 2004 to 2024, as well as the driving factors underlying these changes. The results showed that, across the study area, GPP and ET increased by 75% and 57.3%, respectively. In contrast, WUE decreased across 76.45% of the study area. The effects of 6 environmental drivers—temperature (TEM), precipitation (PRE), solar radiation (RAD), Leaf Area Index (LAI), Vapor Pressure Deficit (VPD), and Soil Surface Moisture (SM)—were analyzed by using partial correlation analysis and the geographic detector method. The results indicated that the increase in LAI was the primary driver of the increases in GPP, ET, and WUE. Moreover, the interaction between LAI and other factors was stronger than that among the other driving factors. The effects of climatic factors also exhibited clear spatial differentiation. Precipitation exerted the greatest influence in the humid and semi-humid central regions of Kerman Province, whereas solar radiation had a stronger impact in the humid and semi-humid central and western regions of the province. The findings of this study offer valuable perspectives for understanding the mechanisms driving changes in WUE and can inform future research on ecosystem development across different climatic regions. کارایی مصرف آب، یک شاخص مهم برای درک مکانیزم چرخه کربن و آب در اقلیم‌های مختلف به حساب می‌آید که برای درک چگونگی سازگاری و توسعه اقلیم‌ها با تغییرات پیچیده محیطی ضروری است. در این راستا از داده‌های تولید ناخالص اولیه و تبخیر و تعرق به‌دست‌آمده از پردازش تصاویر ماهواره‌ای، برای بررسی کارایی مصرف آب در کل اقلیم‌های مناطق مرطوب، نیمه‌مرطوب، نیمه‌خشک و خشک استان کرمان طی سال‌های 1383 تا 1403 استفاده شد. به‌منظور ارزیابی نقش عوامل مختلف محیطی شامل دما، بارش، تابش خورشیدی، کمبود فشار بخار، شاخص سطح برگ و رطوبت خاک، بر روی کارایی مصرف آب از روش آشکارساز جغرافیایی و تحلیل همبستگی جزئی استفاده شد. نتایج به‌دست‌آمده نشان داد که در طول دوره مورد مطالعه، شاخص‌های تولید ناخالص اولیه و تبخیر و تعرق به ترتیب با نرخ متوسط 0065/0 گرم کربن بر متر مربع در سال و 23/2 کیلوگرم آب بر متر مربع در سال افزایش، و کارایی مصرف آب با نرخ متوسط 01/0 گرم کربن بر کیلوگرم آب بر متر مربع در سال کاهش یافته است. از بین عوامل مختلف محیطی، شاخص سطح برگ، بیشترین سهم را در کارایی مصرف آب در غالب مناطق آب‌وهوایی دارد. همچنین عوامل بارش و تابش خورشیدی به ترتیب از مهمترین عوامل تأثیرگذار اولیه آب‌وهوایی در توسعه اقلیم‌های مناطق مرطوب و نیمه‌مرطوب مرکزی و مناطق مرطوب و نیمه‌مرطوب مرکز و غرب استان به حساب می‌آیند.

https://gep.ui.ac.ir/article_30424_c51c4694834fa2ad14c9cb33b5a236cc.pdf

دانشگاه اصفهان جغرافیا و برنامه ریزی محیطی 2008-5362 37 1 2026 03 21 Identification of Suitable Landfill Areas in Izeh County Considering Environmental Constraints شناسایی مناطق مستعد دفن زباله در شهرستان ایذه با اعمال محدودیت‌های زیست‌محیطی 43 66 30489 10.22108/gep.2026.148419.1776 FA معصومه اسدی استادیار پژوهشکده حفاظت آب و خاک و آبخیزداری، تهران، ایران حمید گنجائیان دکتری ژئومورفولوژی، دانشگاه تهران، تهران، ایران اعظم ابراهیمی دانشجوی دکتری جغرافیا و برنامه‌ریزی شهری، دانشگاه شهید چمران، اهواز، ایران Journal Article 2026 02 17 Abstract With rising population growth and increasing waste generation, sustainable waste management has emerged as a major environmental challenge. The selection of appropriate landfill sites plays a crucial role in mitigating environmental impacts and improving the quality of life for local residents. Given the importance of this issue, the present study aimed to identify suitable areas for landfill siting in Izeh County. A range of spatial datasets was employed, including the ALOS PALSAR digital elevation model (12.5 m spatial resolution), 1:100,000-scale geological maps, MODIS and Google Earth satellite imagery, and CHIRPS satellite precipitation data. Data processing and analysis were conducted using ArcGIS software and the Google Earth Engine platform, from which indicators like vegetation cover, mean annual precipitation, and surface features were extracted. A combined model integrating the Relative Weighting Rate (RWR) and fuzzy logic was applied to evaluate spatial suitability for landfill placement. The results indicated that approximately 45.7% of the county area fell within the moderate suitability class, while 45.8% was classified as low or very low suitability. Only 4.8% of the total area (approximately 210 km²) exhibited high potential for landfill siting. After applying environmental and land-use constraints, the final suitable area was reduced to about 76 km². These findings underscored the severe scarcity of suitable land and high environmental sensitivity of the region. The results of this study can serve as a scientific basis for waste management planning. It is recommended that, alongside site selection, complementary measures, such as waste reduction, recycling promotion, geotechnical and hydrogeological assessments, and local community engagement, be considered. Keywords: Landfill, Environmental Impacts, Suitable Landfill Areas, Izeh County. Introduction Solid waste management is regarded as one of the most pressing environmental challenges facing contemporary societies, a problem that has grown increasingly complex due to population growth, urban development, and rising consumption patterns. Landfilling as one of the most common and oldest waste disposal methods can effectively mitigate the negative impacts of waste on both the environment and human health—provided it is carried out properly and scientifically. Selection of an appropriate landfill site is therefore of great importance as improper siting can lead to contamination of surface and groundwater resources, soil degradation, greenhouse gas emissions, threats to ecosystems, and adverse health effects on human communities. Given the significance of this issue, the present study aimed to identify suitable landfill sites in Izeh County. Located in the eastern part of Khuzestan Province and situated within the high Zagros region, Izeh County is characterized by diverse and complex natural conditions. The hydrogeomorphological diversity, considerable elevation differences, severe topographic irregularities, and unique geological features of the area pose serious challenges for spatial planning and siting of various land uses, including landfills. Furthermore, the hydroclimatic conditions of the region—encompassing precipitation patterns, surface water flows, and sensitivity of groundwater resources—underscore the necessity of paying special attention to environmental considerations when selecting landfill locations. Population growth and concentration of human settlements in Izeh County have led to the generation of substantial volumes of municipal waste. At the same time, limited suitable land availability, prevailing lithology, and the risk of water and soil contamination reduce the feasibility of using many areas for landfill purposes. Inappropriate site selection or indiscriminate waste dumping may result in consequences, such as water resource contamination, soil erosion, destruction of natural landscapes, and health risks for local residents. Consequently, identifying suitable areas for landfill siting in Izeh County is an essential and urgent necessity. Materials & Methods To achieve the objectives of this study, a range of spatial data, remote sensing data, and base maps were utilized. The primary datasets included the ALOS PALSAR Digital Elevation Model (DEM) with a spatial resolution of 12.5 meters, a 1:100,000-scale geological map of the region, MODIS satellite imagery for vegetation extraction, CHIRPS satellite rainfall data for estimating mean annual precipitation, and Google Earth imagery for identifying surface features, such as roads, water bodies, and residential areas. ArcGIS software served as the main tool for preparing, managing, and analyzing spatial data layers, as well as for producing the required maps. In addition, the Google Earth Engine platform was employed for processing remote sensing data, extracting vegetation indices, and calculating mean annual precipitation. The combined use of these datasets and tools enabled a more accurate analysis of the environmental and spatial characteristics of the study area. This study employed a hybrid model integrating the Relative Weighting Rate (RWR) and fuzzy logic. The research was carried out in 5 stages: Parameter identification and information layer preparation: Relevant factors were identified and the corresponding spatial layers were compiled. Standardization of information layers: Each layer was transformed into a comparable scale. Weighting of information layers:The relative importance of each layer was determined. Generation of the preliminary suitability map:The weighted layers were combined to produce an initial map of suitable landfill areas. Application of exclusion zones:Constraining factors were overlaid to derive the final map of suitable landfill locations. Discussion of Results and Conclusion Following the identification and weighting of the information layers, the assigned weight values were applied to each layer. The layers were then combined by using the fuzzy gamma operator to appropriately account for the interactions and overlaps among the various datasets. In this process, a hybrid model integrating RWR and fuzzy logic was employed to enhance the accuracy and reliability of the analysis. Based on these analyses, the final map of suitable areas for landfill siting in Izeh County was produced. An examination of the areal extent and percentage distribution of each suitability class revealed that only a limited portion of Izeh County was suitable for landfill development. According to the results, 1,143 km² equivalent to 45.8% of the county's total area (Table 2 and Figure 4), fell into the low and very low potential classes. These areas characterized by dense vegetation, high elevation and slope, and proximity to rivers and lakes were considered unsuitable for landfill siting due to their unfavorable natural and environmental conditions. Utilizing such areas would pose significant environmental and managerial risks. In contrast, only 210 km² accounting for 4.8% of the county's area was classified as relatively suitable for landfill establishment. These areas consisted of scattered portions of the central and western regions of Izeh County, where conditions were favorable according to most evaluated parameters. The findings of this study revealed that, of the total area of Izeh County, approximately 45.7% fell into the medium potential class, while 45.8% fell into the low and very low potential classes for landfill siting. This indicated that a large portion of the region was unsuitable based on environmental and spatial criteria. Only 4.8% of the county's area—equivalent to 210 km²—was identified as having high potential with these areas mostly dispersed across the central and western parts of the county. After applying constraining layers and exclusion zones—including buffers around water resources, human settlements, main roads, and areas of dense vegetation—the final suitable areas were reduced to approximately 76 km², representing about 3% of the total area of Izeh County. This substantial reduction underscored the region's high environmental sensitivity and the severe scarcity of suitable land for landfill purposes. Overall, the results demonstrated that the integrated use of GIS, fuzzy logic, and RWR could effectively enhance the accuracy of site selection and serve as a scientific basis for planning and decision-making in waste management within Izeh County. Given the limited availability of suitable land for landfill in Izeh County, it is recommended that municipal managers and waste management authorities not only utilize the findings of this study to guide site selection, but also actively pursue complementary strategies, such as waste reduction, source separation, recycling, and material reuse. چکیده با افزایش جمعیت و تولید پسماند، مدیریت پایدار زباله‌ها به یکی از چالش‌های مهم محیط‌زیستی تبدیل شده است. انتخاب مکان مناسب برای دفن یا دپوی پسماند، نقش اساسی در کاهش اثرات زیست‌محیطی و بهبود کیفیت زندگی شهروندان دارد. با توجه به اهمیت موضوع، در این پژوهش به شناسایی مناطق مستعد دفن زباله در شهرستان ایذه پرداخته شده است. در این تحقیق به‌منظور شناسایی مناطق مناسب دفن زباله در شهرستان ایذه، از مجموعه‌ای از داده‌های مکانی شامل مدل رقومی ارتفاعی ALOS PALSAR با قدرت تفکیک 5/12 متر، نقشه زمین‌شناسی ۱:۱۰۰۰۰۰، تصاویر ماهواره‌ای MODIS و Google Earth و داده‌های بارش ماهواره‌ای CHIRPS استفاده شد. پردازش و تحلیل داده‌ها با نرم‌افزار ArcGIS و سامانهGoogle Earth Engine انجام شد و شاخص‌های پوشش‌گیاهی، میانگین بارش سالانه و عوارض سطحی استخراج شد. همچنین مدل تلفیقی ضریب وزنی نسبی (RWR) و منطق فازی برای ارزیابی پتانسیل مکانی دفن زباله به‌کار گرفته شد. نتایج نشان داد که حدود 7/45 درصد مساحت شهرستان در طبقه پتانسیل متوسط و 8/45 درصد در طبقات کم و خیلی کم قرار دارند و تنها 8/4 درصد از مساحت شهرستان (۲۱۰ کیلومتر مربع) پتانسیل بالایی برای دفن زباله دارد. با اعمال محدودیت‌های زیست‌محیطی و کاربری اراضی، مساحت نهایی مناطق مناسب به حدود ۷۶ کیلومتر مربع کاهش یافت. این یافته‌ها نشان‌دهنده محدودیت شدید اراضی مناسب و حساسیت بالای محیط‌زیست منطقه است. نتایج پژوهش می‌تواند به‌عنوان مبنایی علمی برای برنامه‌ریزی مدیریت پسماند مورد استفاده قرار گیرد و توصیه می‌شود همراه با انتخاب مکان مناسب، اقدامات تکمیلی مانند کاهش تولید زباله، بازیافت، ارزیابی‌های ژئوتکنیکی و هیدروژئولوژیکی و مشارکت جامعه محلی نیز مدنظر قرار گیرد.

https://gep.ui.ac.ir/article_30489_9ae3344c180dd25ae1f7eb0d2de7517a.pdf

دانشگاه اصفهان جغرافیا و برنامه ریزی محیطی 2008-5362 37 1 2026 03 21 Increasing Contribution of Absorbing Aerosols and Stabilization of Dust Pollution Regime in Isfahan Province Based on Remote Sensing and Markov Chain Analysis افزایش سهم آئروسل‌های جذبی و تثبیت الگوی آلودگی گردوغبار در استان اصفهان بر پایه سنجش‌ازدور و زنجیره مارکوف 67 106 30505 10.22108/gep.2026.148394.1775 FA علیرضا یوسفی کبریا دکتری هواشناسی کشاورزی، دانشکده مهندسی زراعی، دانشگاه علوم کشاورزی و منابع طبیعی، ساری، ایران مهدی نادی دانشیار هواشناسی کشاورزی، دانشکده مهندسی زراعی، دانشگاه علوم کشاورزی و منابع طبیعی ساری، ساری، ایران 0000-0003-0854-8380 بهاره شامگانی مشهدی دانشجوی دکتری هواشناسی کشاورزی، دانشکده مهندسی زراعی، دانشگاه علوم کشاورزی و منابع طبیعی، ساری، ایران Journal Article 2026 02 18 Abstract This study investigated the increasing trend, persistence, and spatiotemporal evolution of air pollution and dust aerosols in Isfahan Province over the period of 2020–2025 with particular emphasis on identifying intensification patterns and shifts in aerosol regimes. The spatial and temporal variations in air pollution were examined by using Aerosol Optical Depth (AOD) data from MODIS and the Absorbing Aerosol Index (AAI) derived from Sentinel-5 satellite observations. Analytical approaches included monthly and seasonal heatmaps, time series analysis, frequency distribution assessments, Kernel Density Estimation (KDE), and a Markov chain model to evaluate the stability and daily transition probabilities of pollution classes at the monitoring station of Isfahan. The results revealed that in 2020, AOD values ranged between 0.37 and 0.56; however, in critical areas—including the desiccated Gavkhouni Wetland, Isfahan City, and Khur County—values exceeded 0.96, indicating unhealthy to very unhealthy conditions. In 2021, pollution intensity escalated further with AOD levels reaching 0.72 to over 1.1 in these regions. Although a relative decline was observed during 2022–2025, the central and eastern parts of the province continued to experience unhealthy and very unhealthy conditions. Regarding AAI, a marked increase in absorbing aerosols was recorded in 2022 with values approaching 2.0 followed by a persistent high-pollution state. Notably, more than half of the days throughout the study period were classified as unhealthy or very unhealthy, underscoring the severity and continuity of the pollution episode. Statistically, AOD exhibited a bimodal distribution, whereas AAI displayed a broader distribution concentrated within the 0.5 to 1.5 range, reflecting the structural persistence of pollution and an increased frequency of extreme events. The annual mean AOD rose from 0.245 in 2020 to 0.305 in 2025, representing a 24% increase, while AAI escalated from near-zero values to approximately 1.4 over the same period. Markov chain analysis revealed that the "Moderate" class for AOD with a persistence probability of 66% constituted the most stable state, whereas the "Unhealthy" class for AAI with a 74% persistence probability emerged as the dominant state. The transition probability from Moderate to Unhealthy and Unhealthy to Very Unhealthy were estimated at about 25% and 10–15%, respectively. Overall, only 4–24% of days were classified as clean, while more than 75% of days fell within the Moderate to Very Unhealthy categories. These findings indicated the establishment of a persistent pollution regime in Isfahan Province characterized by a rising AOD baseline, a sharp AAI shift in 2022, and sustained elevated pollution levels. The concentration of pollution sources in the central and eastern regions—specifically the Gavkhouni Wetland, Isfahan City, and Khur County—underscored the need for air quality management strategies to transition toward structural and source-oriented control measures. Keywords: Air pollution, Pollution Persistence, AOD and AAI Indices, Time-Series Analysis, Markov Chain Modeling. Introduction Dust storms represent critical natural phenomena in arid and semi-arid regions, exerting severe impacts on the environment, climate, human health, and economic activities. Although remote sensing has been extensively employed to monitor these storms and identify their source areas, the majority of previous research has remained confined to descriptive analyses of spatial and temporal trends. The present study addressed this gap by focusing on Isfahan Province, a region particularly vulnerable to dust events due to its geographical setting and climatic conditions. The primary objective of this research was to analyze the temporal dynamics of dust events through an integrated approach that combined remote sensing data—specifically Aerosol Optical Depth (AOD) and Absorbing Aerosol Index (AAI)—with Markov chain modeling. This innovative methodological framework enabled the quantitative assessment of transition probabilities among different air pollution levels, thereby offering valuable insights into the stochastic, time-dependent behavior of dust phenomena. By identifying critical hotspots and modeling the evolution of pollution states over time, this study sought to establish a quantitative foundation for a more comprehensive understanding of dust dynamics, ultimately supporting effective environmental management and informed decision-making in Isfahan Province. Materials & Methods Isfahan Province is situated in central Iran, extending between 30°30' to 34°42' N latitude and 49°36' to 55°32' E longitude. The province exhibited considerable topographical diversity, encompassing the Central Zagros mountain ranges in the west and southwest, alongside arid lowland plains—such as Varzaneh and the margins of the Isfahan-Yazd desert—in the central and eastern portions. The region experiences an arid to semi-arid climate with mean annual temperatures ranging from 16 to 18°C and average yearly precipitation of 100–150 mm. Water resources including the Zayandeh-Rud River and groundwater aquifers have substantially diminished in recent decades due to prolonged drought and excessive extraction. Concurrently, industrial expansion and population growth compounded by reduced water availability have exacerbated air pollution and contributed to the intensification of local dust events (Arvin, 2019). This study utilized remote sensing data spanning the period from 2020 to 2025 to analyze dust dynamics across Isfahan Province. The principal datasets comprised AOD derived from the MODIS sensor and the AAI obtained from the Sentinel-5P satellite. All data were processed at daily, monthly, and annual temporal scales using Google Earth Engine (GEE), a cloud-based geospatial analysis platform. Remote Sensing Data Processing GEE was employed to process MODIS and Sentinel-5P data products. For AOD estimation, MODIS data—specifically the Optical_Depth_047 band—were utilized to map spatiotemporal variations in particulate matter concentrations. Sentinel-5P data (COPERNICUS/S5P/OFFL/AER_AI) were used to derive the AAI, which served as an indicator of the intensity and spatial distribution of absorbing aerosols. MODIS, aboard the Terra and Aqua satellites, provided multispectral imagery at spatial resolutions ranging from 250 m to 1000 m (Xiong et al., 2006). Air Pollution Indices AOD (Aerosol Optical Depth):This index quantifies the extinction of solar radiation by atmospheric aerosols. MODIS-derived AOD data were employed to generate spatial distribution maps across the study area. AAI (Absorbing Aerosol Index):This index indicates the presence of UV-absorbing aerosols, such as mineral dust. AAI data from Sentinel-5P were utilized in a comparable manner to assess aerosol characteristics. AQI (Air Quality Index):As a standardized metric ranging from 0 to 500, the AQI is used to report daily air quality levels. Lower values correspond to cleaner air, whereas higher values indicate elevated health risks (Yousefi Kebriya & Nadi, 2025). The classification criteria for AQI, AOD, and AAI are presented in Table 1. Generally, AOD values exceeding 0.3 and AAI values above 0.5 signify unhealthy or hazardous conditions, while an AQI of greater than 100 indicates air that is unhealthy for sensitive groups (Yousefi Kebriya et al., 2025). Markov Chain Model In this study, a Markov chain model was employed to simulate transitions between different pollution states. This discrete-time stochastic model operates on the first-order Markov property, whereby the future state depends solely on the current state, independent of the full historical sequence. The approach involves the calculation of a transition probability matrix (Pij), which estimates the likelihood of moving from one air quality class (i) to another (j) over a given time step (Stoner & Economou, 2020; Morton & Finkenstädt, 2005). Initially, Google Earth Engine was utilized to map the spatial distribution of AOD and AAI across Isfahan Province. Subsequently, the Markov chain model was applied to analyze temporal trends, assess state stability, and predict transitions among different pollution categories. The integration of satellite-derived observations with Markov chain modeling provided a robust analytical framework for spatiotemporal air pollution assessment, thereby offering a scientific foundation for the development of management strategies aimed at controlling particulate matter in Isfahan Province. Research Findings Large portions of Isfahan and Khur counties remained classified within the unhealthy and very unhealthy categories, underscoring the chronic persistence of air pollution in these areas. In 2025, the spatial pattern of the AOD index was largely consistent with that of 2024; however, a further reduction in the extent of polluted zones was observed. During this year, Ardestan and Kashan were predominantly free of significant pollution, with only limited areas falling within the unhealthy range for sensitive groups. Isolated parts of Nain recorded unhealthy conditions with AOD values ranging from 0.38 to 0.58. Nevertheless, the central and eastern regions of the province—particularly Khur County and Isfahan City—continued to experience unhealthy and very unhealthy conditions with AOD values predominantly ranging between 0.56 and 0.87 and exceeding 0.87 in some areas. In 2023, the AAI index exhibited a decreasing trend; nonetheless, numerous cities remained in an unhealthy state. Extensive areas across the eastern, northern, central, and southern parts of the province recorded AAI values between 0.46 and 0.68, as well as above 0.68, confirming the relative persistence of pollution in known hotspots. This declining trend continued into 2024 with unhealthy zones shrinking to more localized patches. In that year, only parts of Isfahan City and the counties of Khur, Nain, Ardestan, and Aran va Bidgol remained in the unhealthy class with values exceeding 0.55, indicating a relative reduction in both the intensity and spatial extent of pollution. In 2025, the AAI index showed a slight increase compared to the previous year; yet, the overall extent of polluted zones continued to diminish. In parts of Aran va Bidgol, Ardestan, and Nain, index values ranged between 0.45 and 0.65, while in portions of Khur and Isfahan City, values exceeding 0.65 were recorded—higher than in other areas—continuing to identify these regions as the primary aerosol hotspots. Time series analysis of the AOD index revealed an upward trend, increasing from approximately 0.245 at the beginning of 2020 to about 0.305 by the end of 2025, which was equivalent to a 24% increase in mean aerosol concentration over the 6-year period. Analysis of the daily transition probability matrix for AAI pollution classes indicated that Unhealthy days exhibited the highest stability with a 74% probability of persistence followed by Very Unhealthy and Good classes, which showed persistence probabilities of 47% and 56%, respectively. Discussion of Results & Conclusion The findings indicated that the spatiotemporal variations in air pollution across the province arose from the simultaneous interaction of an increased overall particulate matter load and changes in the physicochemical properties of aerosols. The convergence of peak values in both indices during critical years—particularly in 2022—suggested that this period witnessed not only a rise in general particulate matter concentrations, but also intensification in the contribution of absorbing aerosols, such as mineral dust and anthropogenic combustion particles. However, the observed discrepancies between the two indices implied that a portion of the AOD fluctuations might be attributed to the accumulation of non-absorbing particles, stable atmospheric conditions, and humidity effects, whereas the AAI more specifically delineated active dust sources and combustion-related emissions. Although these indices were complementary in aerosol pollution assessment, they differed considerably in terms of the informational content they provided and their spatial resolution. The Markovian stability analysis of this long-term distribution revealed that the majority of days in Isfahan were characterized by moderate to unhealthy pollution levels, with clean air being a rare occurrence. Conversely, critical days, though infrequent, exerted severe and abrupt impacts. Overall, the daily AOD pollution system in Isfahan exhibited relative stability within the moderately polluted and polluted classes with a natural tendency toward intensification of pollution. The marked decline in the probability of reverting to clean classes underscored the narrowing window of clean air episodes and the escalating chronic pressure on both public health and the environment. This analysis demonstrates that effective air quality management must prioritize the reduction of moderately polluted and unhealthy days, while also implementing measures to prevent transitions to the Very Unhealthy and Hazardous categories. چکیده این مطالعه با هدف تحلیل افزایش و پایداری آلودگی هوا و گردوغبار و تغییر الگوهای زمانی - مکانی آن در استان اصفهان طی دوره ۲۰۲۰ تا ۲۰۲۵، با تأکید بر شناسایی روندهای تشدید آلودگی و تغییر رژیم آئروسل‌ها انجام شد. تغییرات و پراکنش آلودگی با استفاده از شاخص AOD (MODIS) و AAI (Sentinel-5) تحلیل شد. تحلیل‌ها شامل نقشه‌های حرارتی ماهانه و فصلی، سری‌های زمانی، توزیع فراوانی، منحنی KDE و مدل زنجیره مارکوف برای ارزیابی پایداری و انتقال روزانه کلاس‌های آلودگی در ایستگاه اصفهان بود. نتایج نشان داد در سال ۲۰۲۰ مقادیر AOD در بازه ۰.۳۷ تا ۰.۵۶ قرار داشت و در نواحی بحرانی (تالاب گاوخونی، اصفهان، خور) مقادیر بیش از ۰.۹۶ مشاهده شد. در سال ۲۰۲۱ شدت آلودگی افزایش یافت و مقادیر در این مناطق به بازه ۰.۷۲ تا بیش از ۱.۱ رسید. در دوره ۲۰۲۲ تا ۲۰۲۵، پهنه‌های مرکزی و شرقی همچنان ناسالم باقی ماندند. در شاخص AAI، افزایش شدید ذرات جذبی در سال ۲۰۲۲ با مقادیر نزدیک به ۲ مشاهده شد و سپس وضعیت تثبیت شد؛ به‌طوری‌که بیش از نیمی از روزها در کلاس‌های ناسالم یا بسیار ناسالم قرار داشتند. از نظر آماری، AOD دارای توزیع دووجهی و AAI دارای توزیع پهن در بازه ۰.۵ تا ۱.۵ بود. میانگین سالانه AOD از ۰.۲۴۵ در ۲۰۲۰ به ۰.۳۰۵ در ۲۰۲۵ (افزایش ۲۴٪) و AAI از نزدیک صفر به حدود ۱.۴ افزایش یافت. نتایج زنجیره مارکوف نشان داد کلاس «متوسط» در AOD با احتمال ماندگاری ۶۶٪ پایدارترین وضعیت است، درحالی‌که کلاس «ناسالم» در AAI با احتمال ماندگاری ۷۴٪ به‌عنوان حالت غالب عمل می‌کند. احتمال انتقال از متوسط به ناسالم حدود ۲۵٪ و از ناسالم به بسیار ناسالم حدود ۱۰-۱۵٪ برآورد شد. درمجموع، تنها حدود ۴ تا ۲۴٪ روزها هوای پاک دارند و بیش از ۷۵٪ روزها در وضعیت متوسط تا بسیار ناسالم قرار دارند. نتایج بیانگر شکل‌گیری رژیم آلودگی پایدار در استان اصفهان همراه با افزایش خط پایه AOD، جهش AAI در ۲۰۲۲ و تداوم بالای کلاس‌های ناسالم است. تمرکز کانون‌های آلودگی در مناطق مرکزی و شرقی (تالاب گاوخونی، شهر اصفهان و خور) نشان می‌دهد مدیریت آلودگی باید به سمت کنترل ساختاری و منبع‌محور تغییر یابد.

https://gep.ui.ac.ir/article_30505_54e80d4128ab5bf444d860e6215961fd.pdf