تحلیل مکانی–زمانی روند و همبستگی شاخص‌های اقلیمی و زیست‌محیطی مرتبط با بیابان‌زایی در شهرستان‌های خاتم و ابرکوه با استفاده از داده‌های ماهواره‌ای

ABSTRACT
Desertification in arid and semi-arid regions of Iran is accompanied by variations in climatic and environmental indicators. The aim of this study is to analyze the spatio-temporal trends and statistical relationships of climatic and environmental proxy indicators associated with desertification-related processes in Khatam and Abarkuh counties during the period 2000–2022. To this end, the Normalized Difference Vegetation Index (NDVI), as a proxy for vegetation cover conditions, was derived from MODIS data; soil moisture and actual evapotranspiration were obtained from TerraClimate; and precipitation and wind speed were extracted from the WorldClim dataset. Long-term trend analysis was performed using the non-parametric Mann–Kendall test and Sen’s slope estimator, while pixel-based Pearson correlation analysis was employed to examine statistical relationships among the indicators. The results indicated that 63% of the pixels exhibited a significant decreasing trend in NDVI (mean Sen’s slope of −0.002 per year), while precipitation showed a declining trend over 69% of the study area, with an average decrease of 1.5 mm per year. Simultaneously, soil moisture displayed a negative trend in 58% of the pixels, whereas actual evapotranspiration showed an increasing trend in 47% of low-lying and arid areas. Wind speed exhibited an increasing trend in 42% of the open eastern zones. Correlation analysis revealed significant positive relationships between NDVI and precipitation as well as soil moisture across large parts of the region, and notable negative correlations between NDVI and wind speed in dry plain areas. Overall, these findings indicate the spatio-temporal co-variation and statistical alignment of environmental indicator changes with conditions associated with desertification, rather than direct causal relationships.
Extended Abstract
Introduction
Desertification is a complex and multi-dimensional process resulting from the interaction of climatic, environmental, and human factors, particularly in arid and semi-arid regions. Iran, due to its location within the global arid belt, pronounced climatic variability, and increasing pressure on natural resources, is highly vulnerable to desertification-related processes. The southern part of Yazd Province—especially Khatam and Abarkooh counties—because of extremely low precipitation, high evapotranspiration rates, wide bare surfaces, and frequent erosive winds, represents one of the most environmentally sensitive areas in central Iran. In the scientific literature, desertification is conceptually distinguished from short-term droughts and climatic fluctuations; however, it is not measured directly in most regional-scale studies. Instead, variations in environmental and climatic conditions are commonly assessed using proxy indicators. In this regard, vegetation cover (NDVI), soil moisture, precipitation, actual evapotranspiration, and wind speed are widely applied as proxy indicators for evaluating conditions associated with desertification-related processes. Due to the limited spatial coverage of meteorological stations, remote sensing data provide an indispensable means for capturing the spatio-temporal variability of these indicators at regional scales. Accordingly, the objective of this study is to analyze the spatio-temporal trends and statistical relationships among climatic and environmental proxy indicators associated with desertification-related processes in Khatam and Abarkooh counties over a 22-year period (2000–2022), with an emphasis on identifying dominant spatial patterns and statistically significant associations rather than causal relationships.
 
Methodology
This study was conducted in Khatam and Abarkooh counties, located in the southern part of Yazd Province. The conceptual framework and required datasets were compiled using scientific literature, digital databases, and satellite-based products. NDVI data were obtained from the MODIS sensor as a proxy for vegetation cover conditions, while soil moisture and actual evapotranspiration were derived from the TerraClimate dataset. Precipitation and wind speed data were extracted from WorldClim. All datasets were processed within a pixel-based spatio-temporal framework to ensure consistency across spatial and temporal scales. Long-term trends were assessed using the non-parametric Mann–Kendall test and Sen’s slope estimator, which are robust against non-normal distributions and missing values. To investigate statistical relationships among variables, pixel-wise Pearson correlation analysis was performed, with NDVI considered the reference indicator. Google Earth Engine was employed as the primary platform for managing large datasets and conducting time-series and pixel-based analyses, while ArcGIS and R were used for visualization and supplementary processing.
 
Results and discussion
The 22-year trend analysis revealed that 63% of the pixels exhibited a statistically significant decreasing trend in NDVI, with the most pronounced declines occurring in central Khatam and the plains of Abarkooh. Precipitation displayed a significant negative trend over 69% of the study area, with an average annual decrease of approximately 1.5 mm. Declining precipitation was statistically associated with reduced soil moisture, and 58% of the pixels showed a negative soil-moisture trend. In contrast, actual evapotranspiration exhibited an increasing trend in 47% of the low-lying and arid areas, while wind speed increased in 42% of the open eastern zones, indicating conditions conducive to enhanced wind erosion.
Correlation analysis demonstrated strong and statistically significant positive relationships between NDVI and both precipitation and soil moisture across large portions of the region, highlighting the importance of moisture availability for vegetation dynamics in dry environments. Conversely, negative correlations between NDVI and wind speed predominated in eastern and central plains, suggesting an inverse association between vegetation cover and wind activity. The relationship between NDVI and actual evapotranspiration varied spatially, reflecting differences in moisture availability and surface conditions. Overall, these findings indicate a clear spatio-temporal co-variation and statistical alignment among environmental proxy indicators, which collectively reflect conditions associated with desertification-related processes, rather than direct causal mechanisms.
 
Conclusion
The results of this study demonstrate that the dryland ecosystems of southern Yazd Province are characterized by concurrent trends of declining vegetation cover, decreasing precipitation and soil moisture, increasing evapotranspiration, and rising wind speeds. The significant reduction in NDVI across more than half of the study area, accompanied by unfavorable trends in key climatic variables, indicates increasing environmental vulnerability and declining land stability. Statistical relationships suggest that precipitation and soil moisture exhibit the strongest positive associations with vegetation dynamics, while wind speed and evapotranspiration show inverse or spatially variable relationships with NDVI in arid and low-lying areas. Although these findings do not imply direct causality, they highlight the potential reinforcement of desertification-related conditions if current trends persist. Therefore, continuous monitoring of climatic and environmental proxy indicators using satellite-based data, along with targeted land and water management strategies, vegetation restoration programs, and wind erosion control measures, is essential for supporting sustainable land management and mitigating environmental degradation in one of Iran’s most sensitive dryland regions.
 
Funding
There is no funding support.
 
Authors’ Contribution
Authors contributed equally to the conceptualization and writing of the article. All of the authors approved thecontent of the manuscript and agreed on all aspects of the work declaration of competing interest none.
 
Conflict of Interest
Authors declared no conflict of interest.
 
Acknowledgments
We are grateful to all the scientific consultants of this paper.