以美国内布拉斯加为例,按照耕地灌溉比例0%—30%, 30%—60%, 60%—100%将农业区分为雨养农业区、混合农业区与灌溉农业区,同时筛选丰水年(2008年)、平水年(2005年)、枯水年(2012年),比较相同年份雨养农业区、混合农业区与灌溉农业区的作物长势的峰值特征差异,以及相同农业区在丰水年、平水年、枯水年的长势过程线的相似性,并定量分析作物长势随灌溉百分比的变化规律与趋势.研究表明:(1)相同年份,灌溉农业区作物长势好于混合农业区,混合农业区的作物长势好于雨养农业区,耕地灌溉比例越高,作物长势越好;(2)不同年份的灌溉农业区作物长势差异最小,混合农业区次之,雨养农业区长势差异最大,即耕地灌溉比例越高,作物长势越稳定;(3)枯水年雨养农业区的作物长势过程线与降水过程线同增同减,受灌溉与降水的双重影响,灌溉农业区的作物长势过程线的峰值滞后于降水峰值;丰水年,作物水分胁迫减弱,灌溉农业区、混合农业区与雨养农业区作物长势过程线与降水过程线变化趋势基本一致;(4)作物长势增幅与灌溉百分比之间呈现显著的分段二次函数变化关系,当灌溉百分比增幅小于60%时,作物长势增长幅度逐步加快,当灌溉百分比大于60%时,作物长势增速逐步放缓,在枯水年时,长势随灌溉百分比增加而增长的幅度高于丰水年与枯水年.鉴于不同农业区作物长势差异,作物长势的定量监测需要进一步区分灌溉与雨养农业.

Crop yield is affected by crop condition. With a favorable crop condition, better crop yield can generally be expected; otherwise, crop yield will be lower. Thus, crop condition monitoring is very important. Normalized Difference Vegetation Index (NDVI) is widely used to monitor crop condition by global agricultural monitoring systems. In irrigated and rain-fed mixed agricultural zones, the average NDVI will exaggerate the water stress on irrigated cropland and ignore the water stress on rain-fed cropland, possibly misleading a policy maker. Therefore, crop condition monitoring in rain-fed cropland should be separated from that in irrigated cropland, and the crop conditions in rain-fed and irrigated croplands in the same crop zones should be compared. As a typically irrigated and rain-fed mixed crop zone, Nebraska in the United States was selected for the quantitative analysis of the difference between crop conditions in irrigated and rain-fed croplands.
First, this work selected the drought (2012), normal (2005), and rainy (2008) precipitation years by calculating the accumulative frequency of rainfall through the Tropical Rainfall Measuring Mission (TRMM) precipitation time series from 2001 to 2013. Second, the cropland of Nebraska was divided into rain-fed (IFC <30%), mixed (30%≤IFC<60%), and irrigated (IFC≥60%) lands according to the cropland irrigated fraction dataset. Third, the difference in maximum NDVIs and the similarity of NDVI time series profiles of the different groups (irrigated, mixed, and rain-fed zones) were analyzed in 2005, 2008, and 2012. Lastly, this work analyzed the NDVI change pattern with the increase of irrigation fraction in 2005, 2008, and 2012.
The results are as below: (1) The crop condition tends to be better with an increase in irrigation fraction at any year; an increase in NDVI accelerates faster when irrigation fraction is less than 60%, but becomes slower when irrigation fraction is larger than 60%. (2) The similarity of NDVI time series becomes strong with an increase in irrigation fraction at any year, indicating that variation in crop conditions eases with an increase in irrigation fraction. (3) In the drought year (2012), the development trend of the NDVI profile is similar with that of the rainfall profile, whereas maximum NDVI was lagging behind maximum rainfall because of the influence of irrigation water; in the rainy year (2008), the development of NDVI was consistent with that of rainfall owing to the alleviation of water stress. (4) The contribution of irrigation to crop condition in drought year (2012) is larger than those in normal and rainfall-abundant years.
In consideration of the crop condition differences in rain-fed, mixed, and irrigated croplands in different years, crop condition monitoring in an irrigated crop zone should be carried out separately from that in a rain-fed crop zone.