Document Type : Research Paper
Authors
1 Phd In Water Resources Engineering, Water Engineering Depth, Qazvin, Iran
2 Assistant Professor in Water Engineering Department/ Imam Khomeini International University
Abstract
Keywords
Main Subjects
Rising temperatures and shifts in precipitation patterns due to global warming present substantial risks to the longevity of vegetation and agricultural systems in arid and semi-arid zones, such as the Karun River Basin in Iran. These alterations foster more frequent and intense droughts, affecting water availability, plant cover, farming practices, and regional ecology, with forecasts suggesting a 1.5°C rise in air temperature by 2050 if current warming persists. In these vulnerable environments, ecosystems exhibit heightened sensitivity to climatic fluctuations, leading to potential disruptions in agricultural growth, food security, and sustainable livelihoods, ultimately causing notable economic damages. Vegetation serves as a pivotal indicator for drought assessment and monitoring, intersecting climate, soil, and human-animal influences; although subject to various factors, its variations remain a reliable measure of natural environmental changes driven by climatic parameters. Recent advancements in remote sensing technologies have enabled widespread utilization of satellite data for tracking vegetation shifts and droughts, particularly through spectral indices like the Normalized Difference Vegetation Index (NDVI) and Vegetation Condition Index (VCI), which accurately reflect vegetation status based on near-infrared and red wavelength reflectance. Numerous recent studies have explored climate parameter impacts on vegetation using these indices; for instance, correlations between NDVI and temperature/precipitation across diverse plant covers have been significant, with stronger ties to precipitation in lush areas. Investigations in regions like China, India, Iraq, and North Africa have demonstrated joint influences of temperature and rainfall on vegetation growth, increasing drought trends, and NDVI sensitivity varying by climate, often higher in colder, drier locales. While spectral indices are established for historical vegetation and agricultural drought analysis, most research is confined to past periods due to reliance on remote sensing imagery, with limited evaluations of future changes under climate scenarios. Yet, projecting vegetation responses to climate change (VRCC) and pinpointing high-risk areas under diverse scenarios is crucial for devising adaptation strategies by policymakers to mitigate hazards. Accordingly, this study examines relationships between climatic factors (precipitation, temperature) and NDVI, employing a Theil-Sen nonparametric regression model to forecast NDVI variations and agricultural drought via VCI under IPCC's SSP3 and SSP5 scenarios from the sixth report. Conducted at five synoptic stations in the Karun Basin over the observational period 1991–2014, it assesses future shifts across three statistical intervals: 2020–2045, 2046–2072, and 2073–2099.
This research numerically evaluates climate change effects on vegetation and drought in the Karun Basin, utilizing NDVI for vegetation health and VCI for agricultural drought monitoring under SSP3 and SSP5 scenarios. The study area encompasses the Karun River Basin in southwest Iran, featuring varied climates due to the Zagros Mountains, diverse temperature and precipitation regimes, and geological structures supporting multiple vegetation types. Data included precipitation and temperature from synoptic stations (Boroujen, Borujerd, Abadan, Kouhrang, Yasuj) for 1991–2014, alongside outputs from CMIP6 models (MRI-ESM2, IPSL-CM6A-LR, GFDL-ESM4) for historical and future periods up to 2099. Spearman's nonparametric correlation assessed links between NDVI (from AVHRR satellite imagery at 0.05° resolution via Google Earth Engine) and climatic variables. Temperature data validation against observations used R² and RMSE metrics. The Theil-Sen nonparametric regression, robust to outliers via median slopes of paired data differences, estimated NDVI from temperature, with equations derived per station and evaluated via R², MSE, p-values, and Nash-Sutcliffe efficiency. VCI, calculated from NDVI time series as (NDVI - NDVI_min) / (NDVI_max - NDVI_min) * 100, classified drought levels: extreme (<10), severe (10–19.99), moderate (20–29.99), mild (30–39.99), and wet (>40). Future projections applied regression outputs to model temperatures, analyzing NDVI percentage changes, VCI drought frequency, and monthly VCI variations relative to baseline. All analyses ensured replicability with specified data sources and statistical methods, focusing on accuracy through trial validations of model fits.
Correlation assessments revealed inverse relationships between NDVI and precipitation (Spearman's rho -0.56 to -0.78), strengthening in humid stations, while direct positive ties with temperature (0.65–0.83) indicated temperature-limited vegetation growth, with reduced correlations in hot areas like Abadan due to excessive heat. Climate models accurately simulated baseline temperatures (R² >0.95, RMSE 1.5–2.2°C). Theil-Sen regressions yielded R² 0.49–0.77, MSE ~0.003, and p<0.05, with strong fits in cooler stations (Nash-Sutcliffe 0.78 in Kouhrang) but underestimation in dry ones (0.43 in Abadan), suggesting potential enhancements via multivariate inclusions like land surface temperature. Future NDVI projections showed upward trends, amplified under SSP5: increases up to 68.56% in cold, high-elevation Kouhrang by 2073–2099, 33.47% in Boroujen, and milder 14% in warm Abadan, reflecting greater benefits in colder zones from alleviated cold stress. VCI analyses indicated rising severe and extreme drought frequencies (e.g., extreme from 6.9% baseline to 11% in Kouhrang under SSP5), with wet conditions declining slightly, trends intensifying in SSP5 and southern/mountainous sites. Monthly VCI changes displayed winter/spring declines (up to 40% in Borujerd, 24% in Boroujen under SSP5 in late century), signaling heightened moisture stress and disrupted growth seasons, contrasted by minor summer gains possibly from extreme precipitation. Overall, colder regions exhibit NDVI gains but face amplified drought risks in long-term high-emission scenarios.
This investigation highlights climate change's dual impacts on the Karun Basin, enhancing NDVI in colder areas through warming but escalating agricultural droughts via VCI declines, particularly under SSP5 with intensified temperature and moisture stresses. The Theil-Sen model and VCI proved effective for projections, though underperformance in arid zones underscores needs for multivariate refinements incorporating soil moisture or surface temperature. Dimensionless NDVI increases reached 56–68% in elevated sites, while drought frequencies rose by up to 11%, with winter VCI drops of 24–40% threatening soil humidity and crop cycles. These outcomes pose serious risks to rainfed agriculture, food security, and ecosystems, emphasizing adaptive policies like optimized water management, drought-resistant cultivars, and land-use reforms in vulnerable southern/high-altitude areas such as Kouhrang and Abadan. Integrating remote sensing platforms like Google Earth Engine with climate models offers robust tools for drought risk mapping, land planning, and ecosystem restoration funding, including reforestation or wetland revival, to bolster resilience.
Conceptualization, Koohi S. and Azizian A.; methodology, Koohi S. and Azizian A.; software, Koohi S.; validation, Koohi S. and Azizian A.; writing—original draft preparation, Koohi S.; writing—review and editing, Koohi S. and Azizian A. All authors have read and agreed to the published version of the manuscript.
Data available on request from the authors.
The authors avoided data fabrication, falsification, plagiarism, and misconduct.
The author declares no conflict of interest.
)