The universal phenomena of global warming caused by climate change have direct linkage with different hydro-meteorological variables which in turn affect the evaporative demand of the atmosphere. Therefore, this study evaluated the seasonal and spatial changes in reference crop evapotranspiration (ETo) during last 4.7 decades and identified the forcing mechanism behind the seasonal changes in ETo using a stepwise regression equation. To remove the effect of serial correlation, the modified Mann–Kendall test was adopted together with Sen's slope and linear regression method to identify and compare the spatial and temporal trends between Thornthwaite, Hargreaves and FAO Penman–Monteith (FAO-PM) methods. Results indicate that the annual average ETo value is higher along the south and east coastal areas and lower along the northwest side of South Korea. The highest number of stations with significant increasing trends was detected in Thornthwaite and significant decreasing trend in FAO-PM method. Spatial and temporal correlation analysis showed the existence of strong correlation between Hargreaves and FAO-PM methods with the Pearson r value ranging from 0.84 to 0.98 and R² of 0.8. Wind speed is found to be the most influencing climatic variable, especially in autumn, early winter and early summer, and maximum temperature during spring and late summer.

中文翻译：

---

韩国参考蒸散量的时空趋势

由气候变化引起的全球变暖的普遍现象与不同的水文气象变量有直接联系，而水文气象变量又影响大气的蒸发需求。因此，本研究评估了过去4.7年中参考作物蒸散量（ETo）的季节性和空间变化，并使用逐步回归方程确定了ETo季节性变化背后的强迫机制。为了消除序列相关性的影响，采用改进的Mann-Kendall检验以及Sen的斜率和线性回归方法来识别和比较Thornthwaite，Hargreaves和FAO Penman-Monteith（FAO-PM）方法之间的时空趋势。结果表明，韩国南部和东部沿海地区的年平均ETo值较高，而韩国西北部的年平均ETo值较低。在Thornthwaite中发现趋势显着增加的台站数量最多，而在FAO-PM方法中则发现显着下降的趋势。时空相关分析表明，哈格里夫斯方法和粮农组织-PM方法与皮尔逊方法之间存在很强的相关性r值介于0.84至0.98之间，R ²为0.8。发现风速是影响最大的气候变量，尤其是在秋天，初冬和初夏，以及春季和夏末的最高温度。