Document Type : Research Paper

Authors

• Reza Norooz Valashedi ¹
• Sedigheh Bararkhanpour-Ahmadi ²

¹ Assistant Professor, Water Engineering Department, Sari Agricultural Sciences and Natural Resources University, Sari, Iran.

² PhD student of Agricultural Meteorology, Water Engineering Department, Sari Agricultural Sciences and Natural Resources University, Sari, Iran

Abstract

Keywords

• Evapotranspiration
• Temporal - Spatial Trend
• Change Point
• Extreme Values

20.1001.1.2008479.1402.54.1.2.2

Main Subjects

• ...

EXTENDED ABSTRACT

Introduction

Reference evapotranspiration (ET₀) as an important indicator of demand for air evaporation is an important factor for climatic and hydrological studies. Due to the occurrence of climate change and the occurrence of large fluctuations in precipitation and the occurrence of mild to severe droughts, it is important to study the evapotranspiration changes. Extreme events such as floods, storms and droughts are often caused by extreme weather. Therefore, in addition to studying the average data in investigating changes in meteorological and hydrological parameters, it is necessary to examine very low or very high values of evapotranspiration and other meteorological parameters. Therefore, in this research, regression methods and statistical tests were used in order to investigate the temporal-spatial changes of ET₀ in Mazandaran province. In order to investigate the trend and change point in the ET₀ time series, the Mann-Kendall, Sen's slope and Pettit non-parametric tests were used, and to investigate the trend in extreme values of ET₀, the quantile regression method was used.

Material and Methods

In this research, the monthly ET₀ grid data (monthly sum in mm) was used, which was extracted from the TerraClimate database (with a resolution of about 5 km) and was calculated using the FAO Penman Monteith method. In the first step, checking the accuracy of grid data with the ET₀ data from the nearest synoptic station to the grid position (calculated by Torrent-White method) was performed using Root Mean Square Error, BIAS and Spearman correlation test statistical evaluation indicators. Then, a seasonal time series of grid ET₀ data was prepared for the entire Mazandaran province in a period of 40 years (1981-2020). In the next step, the trend of changes and the change point in the monthly ET₀ time series in different seasons were performed using Mann-Kendall, Pettitt statistical tests and the quantile regression method, and the results (slope and significance of the trend and the year of the change point ) was zoned for the whole province in the GIS environment. Then the results were analyzed and compared.

Results and discussion

The correlation coefficient between network evaporation-transpiration data and ground station data was estimated to be more than 0.9 in most of the stations and the average value of BIAS was 0.24. The results of Pettitt's test showed the time of sudden changes in reference evapotranspiration in spring, autumn and winter seasons in 2013, 2007-2010 and 1999, respectively, in most cases. Very high values of evapotranspiration increased in spring in the eastern half, in summer in the northern half and in winter in the southern and western strips (with a seasonal slope percentage of 45, 75 and 120 percent, respectively), but they decreased in autumn in the north of the province (with a slope of -45 percent). In general, significantly increasing slope rates for high values of evapotranspiration were higher than average.

Conclusion

The significant increase in high amounts of evapotranspiration, especially in the dry season, will reduce water resources and disrupt the agricultural sector. , which, if not managed on time, will cause irreparable damages in the agricultural sector, especially in the dry season. Therefore, scientific and practical methods for managing reference evapotranspiration in the province should be considered.