دوماهنامه
نوع مقاله : مقالات پژوهشی
نویسندگان
1 استادیار مرکز پژوهشی علوم جغرافیا و مطالعات اجتماعی دانشگاه حکیم سبزواری
2 دستیار پژوهشی، مرکز پژوهشی مطالعات جغرافیایی و علوم اجتماعی، دانشگاه حکیم سبزواری
3 3- مرکز پژوهشی مطالعات جغرافیایی و علوم اجتماعی، دانشگاه حکیم سبزواری
کلیدواژهها
عنوان مقاله English
نویسندگان English
Introduction
Dust is one of the most important environmental hazards in arid and semi-arid regions and can affect vegetation health through reducing incoming radiation, depositing particles on leaf surfaces, disrupting photosynthesis, and intensifying environmental stresses. Khuzestan Province is considered one of the major dust-affected regions in Iran due to its geographical location, proximity to domestic and transboundary dust sources, sensitive wetlands, extensive agricultural lands, recurrent droughts, and land-use changes. The reduction of wetland water levels, decline in soil moisture, improper water resource management, and human activities have also increased the vulnerability of this region. Since vegetation plays an important role in soil stabilization and the reduction of wind erosion, investigating the effects of dust events on vegetation conditions is of great importance for environmental and agricultural management. The aim of this study is to analyze the temporal effects of dust events on vegetation health in Khuzestan Province during the period 2001–2023 using remote sensing and atmospheric reanalysis data.
Materials and Methods
In this study, monthly data for the period 2001–2023 were used. To monitor dust intensity, dust aerosol optical depth at 550 nm obtained from the MERRA-2 dataset was employed. In addition, total AOD and the ratio of dust AOD to total AOD were calculated to examine the contribution of dust to the total aerosol burden over the region. Vegetation health was evaluated using two indices, NDVI and EVI, derived from the MODIS MOD13A2 product, version 6.1. NDVI was used as an indicator of vegetation greenness and density, while EVI was applied because of its lower sensitivity to atmospheric effects and vegetation saturation.
The 16-day NDVI and EVI data were converted to a monthly scale and synchronized with the AOD data. Then, monthly anomalies of AOD, NDVI, and EVI were calculated to remove the effect of the seasonal cycle. Severe dust events were identified using percentile-based statistical thresholds; the 75th percentile was calculated as 0.296, and the 90th percentile as 0.360. Months with AOD values equal to or greater than the 90th percentile were selected as severe dust events. Subsequently, the relationship between dust and vegetation indices was examined using Pearson correlation, linear regression, seasonal analysis, month-by-month analysis, lag-correlation analysis, and event-based analysis.
Results and Discussion
The results showed that dust AOD in Khuzestan Province experienced considerable temporal fluctuations during the study period. In total, 70 months were above the 75th percentile and 29 months were above the 90th percentile, indicating the occurrence of relatively severe and severe dust events. The relative synchronization between variations in dust AOD and total AOD showed that dust constitutes an important part of the aerosol burden over the region.
The analysis of NDVI and EVI anomalies indicated that vegetation health changed noticeably during the study period, and both indices showed similar behavior in many periods. In the first step, the relationship between raw AOD and vegetation index anomalies was negative but weak; the correlation coefficient between raw AOD and NDVI anomaly was -0.144, while that between raw AOD and EVI anomaly was -0.132. This weak relationship is probably due to strong seasonal cycles and the simultaneous influence of climatic and agricultural factors.
After removing seasonal effects, the strength of the relationship increased. The correlation coefficients between AOD anomaly and NDVI and EVI anomalies were -0.307 and -0.281, respectively, and both relationships were statistically significant. This result indicates that abnormal increases in dust are generally associated with a relative decline in vegetation health. Seasonal analysis also showed that this relationship was stronger in spring; the correlations between AOD anomaly and NDVI and EVI during this season were -0.388 and -0.346, respectively. This may be related to active plant growth and the greater sensitivity of agricultural crops during this period.
Lag analysis showed that the effect of dust was not limited to the month of occurrence and continued for several months afterward. The strongest negative correlation was observed in the same month as the dust event; however, negative relationships remained at one-month, two-month, and three-month lags. In addition, event-based analysis showed that after severe dust events, the mean anomalies of NDVI and EVI decreased further in the following months. This may indicate the cumulative effect of dust, reduced photosynthetic capacity, decreased surface moisture, or the coincidence of dust events with climatic stresses.
Conclusion
The results of this study showed that there is a negative, significant, and complex relationship between dust events and vegetation health in Khuzestan Province. The use of AOD anomalies provided a more accurate representation of the effect of dust on vegetation compared with raw AOD values. The findings indicate that abnormal increases in dust, particularly in spring and in some sensitive months, are associated with relative decreases in NDVI and EVI. However, vegetation health is not affected by dust alone, and factors such as precipitation, temperature, soil moisture, irrigation, crop type, planting time, and agricultural management also play important roles. Therefore, for a more accurate interpretation of this relationship, the integration of remote sensing data with climatic, hydrological, and field data is recommended.
کلیدواژهها English
