延河流域实际蒸散发时空特征及影响因素分析

Analysis of temporal and spatial characteristics of actual evapotranspiration and its influence factors in Yanhe River Basin

研究气候和植被恢复综合影响下流域实际蒸散发(E_(a))的变化规律,可为水资源管理和规划提供科学参考。基于延河流域气象数据、径流资料和植被覆盖数据(NDVI),采用率定后的平流干旱模型计算E_(a),通过Mann-Kendall检验法和反距离加权法分析E_(a)时空分布特征,并利用PE_(a)rson相关分析对E_(a)成因进行探究。结果表明:AA模型最优参数α为0.85,率定期和验证期相对误差均<5%,验证期纳什系数为0.998,模型满足精度要求。延河流域多年月平均E_(a)为单峰分布,高值月份出现于6—8月;多年平均E_(a)为472.92 mm且呈现以0.23 mm/a的趋势增加;空间上,E_(a)整体呈东南高、西北低的分布特征,高值区域随着时间逐渐向南偏移。E_(a)与降水量、相对湿度、气压、2 m风速和NDVI呈现负相关,与日照时间、日最高温度、日最低温度、平均温度呈正相关。E_(a)增加受温度和日照时间的上升、气压的下降综合影响;植被覆盖度空间异质性决定着实际蒸散发的空间差异,在水分补给影响下,植被覆盖度对E_(a)作用差异显著。

[Background]This paper is to study the variation law of actual evapotranspiration(E_(a) )under the change of climate and vegetation restoration in the Yanhe River Basin,which may provide a scientific reference for the research of management and planning of water resources under the change of climate and vegetation cover.[Methods]Based on the meteorological data,runoff data,and normalized difference vegetation index data(NDVI) in Yanhe River Basin from 1978 to 2017,We firstly calibrated the parameter(α) of the advection-aridity model and used the parameter after calibration to calculate E_(a) .Then we used the Mann-Kendall test and inverse distance weighting method to analyze the changes in trends and the characteristics in spatial and temporal distribution of E_(a) .In addition,we used Pearson correlation analysis method to explore the reasons of the change in E_(a) .[Results]The optimal parameter of the AA model is 0.85 in the Yanhe River Basin.Under this parameter,both the relative error of the calculated results in the calibration period and the verification period is 5%.Moreover,the Nash Efficiency Coefficient(N) is as high as 0.998 in the verification period.The calculated results of the calibrated AA model are reliable and have high accuracy in our study in the Yanhe River Basin so that the model meets the accuracy requirements.From the view of time point,E_(a) has only one maximum value in a year,and the month at the high-value of evapotranspiration appears from June to August in a year,which is a unimodal distribution.As a whole,the average annual E_(a) for many years is 472.92 mm and shows a trend of increasing by 0.23 mm/a.From the view of spatial point,the overall spatial appearance of E_(a) is high in the southeast and low in the northwest.The distribution characteristics of the high-value area of E_(a) gradually drift southward with time.In Yanhe River Basin,E_(a) is negatively correlated with annual precipitation,relative humidity,atmospheric pressure,wind speed at 2 m height and NDVI,but it is not significant,and negative correlation coefficients are all below 0.2.E_(a) is positively correlated with actual duration of sunshine,maximum daily temperature,minimum daily temperature,and average temperature,it passes the 0.05 significance level test,and the correlation coefficient between actual duration of sunshine and E_(a) reaches 0.74.The correlation coefficient between E_(a) and NDVI is-0.063,which shows a weak negative correlation, but their relationship varies in different regions.[Conclusions]On the annual scale,the increase in E_(a) is caused by the combination of rising in temperature and actual duration of sunshine, and falling in atmospheric pressure.The spatial heterogeneity of vegetation coverage determines the actual spatial difference of actual evapotranspiration.But E_(a) is intensively affected by water replenishment.Therefore, under the influence of water replenishment,the effect of vegetation coverage on E_(a) is significantly different.