本文为美国科罗拉多州大学（作者：Jeffrey Alan Siegfried）的硕士论文，共134页。

农业是全球最大的水消费领域。随着可用水资源压力的增加，用更少的水生产更多的粮食变得至关重要。精确供水方法（如变量灌溉）所需的硬件技术在商业上是可以买到的。尽管如此，为这些系统制定及时、准确的处理技术仍然是不够的。

光谱植被指数，特别是归一化差异植被指数，常用于衡量作物活力及相关参数（如叶片含氮量和产量）。然而，迄今为止的研究很少涉及土壤湿度对指数的影响。用廉价红外温度计测量的冠层温度也可以作为衡量水分的指标，但是目前利用这些数据的方法可能会很麻烦。因此，本研究的目的是确定

1）从多光谱卫星图像获得的植被指数是否有助于量化灌溉玉米生产系统中的土壤水分变化特性

2）单个图像代表土壤水分状况的时间段

3）确定同步测量作物冠层温度和田间土壤水分张力；

4）了解任意作物冠层温度应力阈值对土壤水分与作物冠层温度关系的影响。

采用变量灌溉支点形成六个水处理区。每个区域都配备了一套张力计，安装在20、45和75厘米深的地块中心，此外，还有一个指向作物冠层的红外温度计用于监测水处理区的状况。在每个处理区中，以估计蒸散量（ET）需求量的百分比进行供水：即40%、60%、80%、100%、120%和140%。从张力计收集的数据与对应于张力计地面位置的图像像素以及同步的冠层温度数据配对，分别进行了统计分析，以评估植被指数和冠层温度是否代表多种作物生长阶段的土壤水分。研究结果表明，归一化差异植被指数可以量化作物生长期V6（6叶）（r2=0.850，p=0.009）和V9（9叶）（r2=0.913，p=0.003）作物生长期土壤水分张力的变化特性。结果表明，卫星植被指数可能有助于建立大尺度土壤水分变化的时间敏感性特征。当与阈值相结合时，同步冠层温度能够量化生殖作物生长阶段的土壤水分张力。进一步的研究是必要的，以调查额外的作物生长阶段、更多的作物和其他来源的多光谱图像。未来的研究还需要评估变量灌溉管理的田间规模产量影响。

Agriculture is the largest consumer ofwater globally. As pressure on available water resources increases, the need toexploit technology in order to produce more food with less water becomescrucial. The technological hardware requisite for precise water deliverymethods such as variable rate irrigation is commercially available. Despitethat, techniques to formulate a timely, accurate prescription for those systemsare inadequate. Spectral vegetation indices, especially Normalized DifferenceVegetation Index, are often used to gauge crop vigor and related parameters(e.g. leaf nitrogen content and grain yield). However, research heretoforerarely addresses the influence of soil moisture on the indices. Canopytemperature measured using inexpensive infrared thermometers could also serveas an indicator of water stress, but current methods which exploit the data canbe cumbersome. Therefore, the objectives of this study were to determine 1) ifvegetation indices derived from multispectral satellite imagery could assist inquantifying soil moisture variability in an irrigated maize production system2) the period of time which a single image is representative of soil moistureconditions 3) to determine the relationship between synchronous measurements ofcrop canopy temperature and in-field soil moisture tension, and 4) tounderstand the influence of discretionary crop canopy temperature stressthresholds on the relationship between soil moisture tension and crop canopytemperature. A variable rate irrigation pivot was used to form six watertreatment zones. Each zone was equipped with both a set of tensiometersinstalled in the center of the plots at 20, 45, and 75cm depths and an infraredthermometer pointed into the crop canopy to individually monitor conditions inthe water treatment zones. Water was applied for each treatment as a percentageof the estimated evapotranspiration (ET) requirement: i.e., 40, 60, 80, 100,120, and 140 percent of the ET. Data collected from tensiometers was pairedwith the image pixels corresponding to the ground location of the tensiometersand with the synchronous canopy temperature data. Statistical analysis wasperformed separately to assess whether vegetation indices and canopytemperature are representative of soil moisture at several crop growth stages.Findings from this study indicate that Red Edge Normalized DifferenceVegetation Index could quantify variability of soil moisture tension at V6 (sixleaf) (r2 = 0.850, p = 0.009) and V9 (nine leaf) (r2 = 0.913, p = 0.003) cropgrowth stages. Results suggest that satellite-derived vegetation indices may beuseful for creating time-sensitive characterizations of soil moisturevariability at large field-scales. When integrated with a stress threshold,synchronous canopy temperature was able to quantify soil moisture tension withsome success during the reproductive crop growth stages. Further study isnecessary to investigate additional crop growth stages, more crops, and othersources of multispectral imagery. Future studies are also needed to evaluatefield-scale yield implications of variable rate irrigation management.

1. 用多光谱卫星图像定量分析田间土壤水分变化特性
2. 灌溉玉米田间土壤水分变化的红外测温研究

更多精彩文章请关注公众号：