Document Type : Research Paper
Authors
1 Soil and Water Research Department, َArdabil Agricultural and Natural Resources Research and Education Center, Agricultural Research, Education and Extension Organization (AREEO), Ardabil, Iran.
2 Department of Irrigation and Soil Physic, Soil and Water Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Karaj, Iran.
Abstract
Keywords
Main Subjects
Understanding the influence of climatic parameter variability on reference evapotranspiration (ET₀) is critical for irrigation planning and agricultural water resource management. Integrating trend, sensitivity, and contribution analyses provides a robust quantitative framework for evaluating the impact of climatic variables on ET₀, thereby enhancing the precision and reliability of the results. This study aims to quantify the impact of climatic parameter changes on ET₀ in Ardabil Province, Iran.
To achieve this objective, the following steps were undertaken:
Monthly, seasonal, and annual time series of ET₀ and climatic parameters were analyzed for 10 synoptic stations across Ardabil Province.
The stations included Ardabil and Namin (temperate zone); Sareyn, Khalkhal, Meshginshahr, Nir, and Firouzabad (cold zone); and Parsabad, Germi, and Bilehsavar (warm zone). These climatic zones were classified based on elevation, air pressure, and climatic variables using a hierarchical cluster analysis.
The dataset, covering the period from 1999 to 2024, was obtained from the Soil and Water Research Institute. ET₀ was calculated using the FAO Penman–Monteith method, a globally accepted standard.
The analyzed climatic parameters included maximum and minimum air temperature (Tmax and Tmin), maximum and minimum relative humidity (RHmax and RHmin), sunshine duration (SH), and wind speed (U).
The Modified Mann–Kendall test and Sen’s slope estimator were used to detect the relative change rate (RCv) in time series at confidence levels of 90%, 95%, and 99%. ET₀ sensitivity was assessed using the dimensionless relative sensitivity coefficient (Sv), calculated via the partial derivative method by applying a ±10% change to each parameter (Chu et al., 2017).
Based on the trend and sensitivity results, the contribution rate (Conv) of each climatic variable to ET₀ variations was estimated by multiplying Sv by the RCv of the respective parameter, using Equations 6 to 8 (Li et al., 2017).
Finally, trends, sensitivity coefficients, and contribution rates were systematically analyzed and compared across time scales and stations to reveal spatiotemporal patterns.
Annual ET₀ exhibited a statistically significant increasing trend at 60% of the stations across all three climatic zones, while no station showed a significant decreasing trend. The highest increases in monthly, seasonal, and annual ET₀ were observed at Firouzabad Station, with values of 4.37 mm (July), 9.59 mm (summer), and 14.48 mm annually. In contrast, the most notable decreases occurred in June and spring at Germi Station, with rates of –3.31 mm and –4.69 mm, respectively (Figure 2). Overall, the results indicated an increasing trend in ET₀ across most areas of Ardabil Province. Tmax, Tmin, and U showed increasing trends on both seasonal and annual time scales, while other parameters showed decreasing trends. These findings are consistent with earlier studies conducted in western regions of Iran.
The highest annual sensitivity coefficients were observed for Tmax (0.40), RHmin (–0.37), RHmax (–0.32), U (0.24), SH (0.14), and Tmin (0.03). In general, ET₀ was more sensitive to most climatic parameters in warm regions, followed by temperate and cold zones. During summer, ET₀ was most sensitive to Tmax and SH, whereas in autumn and winter, it was more sensitive to RHmax and RHmin. This sensitivity was most pronounced at warm-zone stations such as Parsabad, Germi, and Bilehsavar (Figure 3).
The contribution analysis showed that U and SH had the largest impacts on ET₀ variation, with contribution rates of 12.21% (positive) and –11.39% (negative), respectively. Annual average contributions of RHmax and RHmin were 2.72% and 4.53%, respectively, while Tmax and Tmin contributed 3.41% and 0.96%, respectively (Table 3). At the seasonal level, U showed the highest contribution to ET₀ increase during summer and winter, attributable to its high sensitivity coefficient and trend magnitude in these seasons. SH contributed most to ET₀ reduction during spring and summer, acting as a limiting factor against excessive increases in ET₀ during these periods.
The concurrent increasing trends and positive sensitivity coefficients of Tmax, U, and Tmin, along with the decreasing trends and negative sensitivity coefficients of RHmin and RHmax—particularly during spring and summer—indicated a likely rise in crop water requirements in warm regions. In contrast, the decreasing trend and positive sensitivity of SH in most seasons contributed to moderating ET₀ increases.
Tmax, Tmin, SH, and U were found to positively influence ET₀, while RHmax and RHmin had negative effects.
Although U and SH exhibited lower sensitivity coefficients than other variables at most stations, their contribution rates to ET₀ change were comparatively higher. This is due to their relatively steeper trend slopes. These findings highlight the importance of quantifying the impacts of climatic variables on ET₀ to support precise agricultural water management strategies. For instance, implementing windbreaks around farms and orchards could help mitigate the effect of wind speed on ET₀.
While analyzing climatic trends at broad spatial scales remains essential, this study underscores the importance of detailed assessments at basin and provincial levels. Combining trend, sensitivity, and contribution analyses offers a comprehensive approach for understanding ET₀ variability and its driving forces—crucial for the effective planning and optimization of agricultural water resources.
All authors contributed equally to the conceptualization of the article and writing of the original and subsequent drafts.
Data is available on reasonable request from the authors.
The authors would like to thank all participants of the present study.
The authors avoided data fabrication, falsification, plagiarism, and misconduct.
The author declares no conflict of interest.
)