Abstract Trends in monthly evapotranspiration (ET) rates across Nebraska, the most intensely irrigated state within the US, were calculated by the calibration-free version of the nonlinear complementary relationship of evaporation over the 1979–2015 period utilizing North American Regional Reanalysis (NARR) net radiation, 10-m wind velocity, as well as Parameter Regression Independent Slope Model (PRISM) air- and dew-point temperature data. State-averaged modeled ET rates rose by 5.5 mm decade−1 due to the presence of wide-spread large-scale irrigation projects in accordance with a 2.4 mm decade−1 increase in PRISM precipitation (P) and a simultaneous −2.8 mm decade−1 drop in United States Geological Survey’s state-averaged annual streamflow rates, raising the state-wide ET to P ratio from 0.89 to 0.91 over the modeled time-period. ET rates over irrigated crops increased by 7 mm decade−1 despite a −4.4 mm decade−1 drop in precipitation rates. A similar increase in ET rates (6 mm decade−1) required 8.1 mm decade−1 increase in precipitation rates across the non-irrigated Sand Hills of Nebraska. Published NARR ET rates are unable to pick up this unusual regional trend. Since an increase in precipitation rates should normally decrease the ET ratio, as predicted by the Budyko curve, this study yields evidence on how dramatically sustained large-scale irrigation can alter the regional hydrologic cycle not only through a) trivially depleting streamflow rates and/or lowering groundwater table levels; b) suppressing precipitation locally (while enhancing it a long distance downwind), but also; c) reversing the trajectory of the regional ET ratio under generally increasing trends of precipitation.

中文翻译：

---

区域尺度的人为水文循环干扰：美国内布拉斯加州全州范围的蒸散趋势（1979-2015）

摘要 内布拉斯加州是美国灌溉最密集的州，其月蒸散量 (ET) 的趋势是使用北美区域再分析 (NARR) 通过 1979-2015 年期间蒸发的非线性互补关系的免校准版本计算得出的。净辐射、10 米风速以及参数回归独立斜率模型 (PRISM) 空气和露点温度数据。由于 PRISM 降水 (P) 增加 2.4 毫米十年-1 和同时出现 -2.8 毫米十年-，由于广泛的大规模灌溉项目的存在，州平均模拟 ET 率增加了 5.5 毫米十年-1 1 美国地质调查局的州平均年流量下降，在模拟的时间段内，全州 ET 与 P 的比率从 0.89 提高到 0.91。尽管降水率下降了 -4.4 毫米十年-1，但灌溉作物的 ET 率增加了 7 毫米十年-1。内布拉斯加州非灌溉沙丘的 ET 速率（6 毫米十年-1）的类似增加要求降水率增加 8.1 毫米十年-1。已公布的 NARR ET 利率无法反映这种不寻常的区域趋势。由于降水率的增加通常会降低 ET 比，正如 Budyko 曲线所预测的那样，本研究提供了证据，证明持续的大规模灌溉不仅通过 a) 微不足道的消耗流量和/或降低地下水位水平；b) 局部抑制降水（同时在顺风长距离增强降水），而且；