An Exploration of Multi Decadal, Continental-Scale Evapotranspiration Inferred from Weather Data

陆面蒸散是指土壤、植物和湖泊中的水分转化为大气水汽的过程。它是区域水量平衡的重要组成部分，直接影响径流和地下水补给。ET也是大气水分平衡的重要组成部分，直接影响湿度和降雨量。由于蒸发水所需的能量，ET过程也会影响陆地和气温。尽管与水资源和气候有关，但ET是出了名的难以衡量，特别是在区域范围内。这项建议开发并应用了一种新发现的方法，根据随时可用的天气数据，根据相对湿度的每日模式来估计ET。通过这种方法，该提案解决了以下问题：*美国ET的季节和地理变化是什么？*强迫(温度、湿度、风和降水)和地表属性(土壤、植被、地形)在这些时空模式中的季节和地理作用是什么？*在过去的50年里，ET发生了什么趋势，是什么导致了这些趋势(例如，风、水分可获得性、辐射、温度)？*这些季节、地理和年代际变化(包括趋势)与最先进的气候模型相比如何；地面、低层大气和云形成之间的相互作用如何影响相对湿度？这项工作的更广泛影响包括为水资源管理者开发和演示一种估算ET的实用方法，更好地了解地理水资源的可用性，以及对气候模型的新的观测约束。此外，这项工作将涉及一名研究生和本科生，以及一名高三学生。了解地表和大气对ET的控制、地表和大气之间的反馈以及ET的十年时间尺度变化的进展，需要对ET进行可靠的估计。然而，直接、连续的测量是最近的(大约十年前的)，而且空间有限(数百个通量站，每个站大约测量一平方公里的陆地面积)。正因为如此，关于ET的水文气候学研究要么在空间和时间上受到限制，要么依赖于模型输出。拟议的工作克服了这些限制，并解决了与ET在十年时间尺度(包括趋势)上的时空变化、这些变化的驱动因素以及当前气候模式捕捉这些变化的能力有关的关键问题。通常情况下，估算蒸散量需要测量土壤水分状况和地表温度，或者对水分和能量平衡进行时间综合模拟。对于广泛的应用，前者需要遥感观测，而后者需要校准许多土壤水力和植被参数，从而导致高度依赖于模型公式的结果。这里开发的方法无需特定部位的校准即可提供ET估计。所需的强迫和地面数据的简洁性允许广泛的地理和时间(例如，在卫星时代之前)的应用，并产生对模型的依赖性大大降低的结果。拟议研究的学术价值在于解决了水文科学中的一个重要问题--在大陆尺度上以小时至十年为时间尺度确定水平衡的陆地蒸发分量。

An Exploration of Multi Decadal, Continental-Scale EvapotranspirationInferred from Weather DataLand surface evapotranspiration (ET) is the transformation of water in soils, plants and lakes into atmospheric water vapor. It is a significant part of the regional water balance with direct influence on streamflow and groundwater recharge. ET is also a significant part of the atmospheric water balance with direct influence on humidity and rainfall. Because of the energy required to vaporize water, the ET process also influences land and air temperature. Despite its relevance to water resources and climate, ET is notoriously difficult to measure, especially at regional scales. This proposal develops and applies a newly discovered method for estimating ET from readily available weather data, based on daily patterns of relative humidity. With this method, the proposal addresses these questions: * What is the seasonal and geographic variability of ET over the U.S? * What are the seasonal and geographic roles of forcing (temperature, humidity, wind and precipitation) and surface properties (soil, vegetation, topography) in these space-time patterns? * What trends have occurred in ET over the last 50 years, and what is responsible for those trends (for instance, wind, moisture availability, radiation, temperature)?* How do these seasonal, geographical, and decadal variabilities (including trends) compare with state-of-the art climate models; and * How do the interactions between the land surface, the lower atmosphere, and cloud formation influence relative humidity? Broader impacts of this work include development and demonstration of a practical means for water resource managers to estimate ET, better understanding of geographic water resource availability and new observational constraints on climate models. In addition, the work will involve a graduate and undergraduate student, as well as a high school senior.Progress in understanding surface and atmospheric controls on ET, feedbacks between the surface and atmosphere, and decadal time scale variations in ET, require reliable estimates of ET. Direct, continuous measurements, however, are recent (approximately one decade old), and spatially limited (a few hundreds of flux stations, each measuring approximately one square kilometer of land area). Because of this, hydroclimatological research on ET has been either limited in space and time or has relied on model output. The proposed work overcomes these limitations and addresses critical questions related to the space-time variations of ET over decadal time scales (including trends), the drivers of these variations, and the ability of current climate models to capture these variations. Typically estimating ET requires either measurement of soil water status and surface temperature, or temporally integrated modeling of the water and energy balance. For broad use, the former requires remotely sensed observations, and the latter requires calibration of numerous soil hydraulic and vegetation parameters, leading to highly model-formulation dependent results. The method developed here provides ET estimation without site-specific calibration. The parsimony of required forcing and surface data allows for broad geographic and temporal (e.g., prior to the satellite era) application, and yields results that are significantly less model dependent. The intellectual merit of the proposed research is in addressing a vital questions of hydrologic sciences - determination of the terrestrial evaporation component of the water balance at hourly to decadal time scales over continental scales.