Quantifying the Relative Roles of Progressive Land Use Change, Irrigation, and Remote Forcing in Southern Great Plains Precipitation Variability

这个项目试图了解美国南部大平原(SGP)降水变化的原因，该地区大致相当于内布拉斯加州、密苏里州、俄克拉何马州、阿肯色州、德克萨斯州和路易斯安那州覆盖的地区。该地区是美国最容易发生干旱的地区之一，带来了重大的经济影响，包括沙尘暴的严重后果，以及2011年和2012年数百亿美元的损失。该区域的降水异常，包括造成干旱的降雨不足，通常归因于海洋表面温度(SST)的波动引起的大尺度大气环流的变化。但当地的地表条件可以通过几种机制加剧远程强迫降水的可变性。例如，缺乏降雨导致土壤干燥和蒸发减少，从而减少了可用于降水的水分来源，并导致进一步的降雨量减少。此外，由于干旱导致的温暖干燥条件促进了对流层中层的隆起，这可能会进一步减少降水，因为降雨系统在脊部周围移动。该项目考虑了由于SGP中土地利用和土地覆盖变化的变化，这种陆面对降水的反馈强度发生了多大程度的变化。自20世纪50年代以来，该地区的许多本地植被已被农作物取代，这种取代会影响地表和上覆大气之间的水交换。本土植物的根要深得多，可以从农作物无法获得的储备中获取土壤水分，从而即使在干旱时期也能保持降水的水分来源。此外，农作物往往比原生植物更早变绿，因此植物对土壤水分的吸收和蒸腾作用在农地上开始得更早。前期水分需求的增加可能会增加生长季节后期的干旱易感性。进一步考虑的是灌溉的广泛采用，灌溉可以在多雨的年份增加蒸发和蒸腾，但在异常干旱的年份并不可靠。这项研究是通过使用天气研究和预报(WRF)模型进行的一套模拟进行的，WRF模型是一个在大平原上配置了高分辨率的区域模型。WRF与社区土地模型(CLM)相结合，在CLM中，可以指定土地覆盖来表示过去和现在的植被覆盖，并可以表示不同的灌溉水平。实验主要包括北美区域再分析(NARR，1979-2012年)期间的WRF-CLM模拟，首先将土地覆盖固定为现代土地覆盖属性(来自国家土地覆盖数据库)。然后将此模拟与其他模拟进行比较，在其他模拟中，土地覆盖被配置为表示前定居时代的条件(基于代表1850年的数据)、1920年前后的中间土地利用条件以及沙尘暴。现代和中等土地利用的模拟是在灌溉和不灌溉的情况下进行的。由于SGP干旱的经济和社会影响，这项工作具有更广泛的社会影响。这项研究的结果可能对农业部门和SGP水资源管理者做出关于该地区土地和水资源使用变化的决策有用。除了在同行评议的研究期刊上发表文章外，研究结果还将通过内布拉斯加州大学林肯分校的分部和SGP的其他分部服务分发。此外，该项目将支持和培训两名研究生，从而为这一研究领域的未来劳动力提供支持。

This project seeks to understand the causes of precipitation variability in the US Southern Great Plains (SGP), roughly equivalent to the region covered by Nebraska, Missouri, Oklahoma, Arkansas, Texas, and Louisiana. This area is among the most drought-prone in the US, with substantial economic impacts including the severe consequences of the Dust Bowl and losses in the tens of billions of dollars in 2011 and 2012. Precipitation anomalies in the region, including rainfall deficits contributing to droughts, are generally attributed to changes in the large-scale atmospheric circulation brought about by fluctuations in sea surface temperature (SST). But local land surface conditions can act to intensify remotely-forced precipitation variability through several mechanisms. For instance, lack of rainfall leads to dry soil and reduced evaporation, which reduces the moisture source available for precipitation and causes further rainfall reduction. Also, warm dry conditions due to drought promote ridging in the mid-troposphere, which can further reduce precipitation as rain-bearing systems move around the ridge. This project considers the extent to which the strength of such land surface feedbacks to precipitation has changed due to changes in land use and land cover change in the SGP. Since the 1950s much of the native vegetation in the region has been replaced by agricultural crops, and this substitution can affect the exchange of water between the surface and the overlying atmosphere. Native plants have much deeper roots and can access soil moisture from reserves unavailable to crops, thereby maintaining the source of moisture for precipitation even during dry spells. Also, crops tend to green up earlier than native plants, so that plant uptake and transpiration of soil moisture begins earlier on agricultural land. The enhanced early season moisture demand could increase drought susceptibility later in the growing season. A further consideration is the widespread adoption of irrigation, which can increase evaporation and transpiration during wet years but is not reliable during exceptional drought years.The research is conducted through a suite of simulations performed using the Weather Research and Forecasting (WRF) model, a regional model which is configured with high resolution over the great plains. WRF is coupled to the Community Land Model (CLM), in which land cover can be specified to represent past and present vegetation cover and varying levels of irrigation can be represented. Experiments consist largely of WRF-CLM simulations over the period of the North American Regional Reanalysis (NARR, 1979-2012), first with land cover held fixed to the modern land cover attributes (from the National Land Cover Database). This simulation is then compared with additional simulations in which land cover is configured to represent conditions during the pre-settlement era (based on data representing the year 1850), intermediate land use conditions around 1920, and the dust bowl. Simulations for modern and intermediate land use are performed with and without irrigation.The work has societal broader impacts due to the economic and societal impact of drought in the SGP. The findings of the research may be useful to the agricultural sector and SGP water managers as they make decisions regarding changes in land and water use in the region. In addition to publications in peer-reviewed research journals, results of the research will be distributed through the University of Nebraska-Lincoln Extension and other extension services across the SGP. In addition the project will support and train two graduate student, thereby providing for the future workforce in this research area.