Sensitivity of global major crop yields to climate variables: A non-parametric elasticity analysis
Sensitivity of global major crop yields to climate variables: A non-parametric elasticity analysis
复制标题
全球主要作物产量对气候变量的敏感性：非参数弹性分析
DOI：
10.1016/j.scitotenv.2020.141431
复制
发表时间：
2020
影响因子：
9.8
通讯作者：
Deepak K. Ray
中科院分区：
文献类型：
--
作者：
Di Liu;Ashok K. Mishra;Deepak K. Ray
气候变异性控制着作物产量的变异性，对地方、区域和全球各级的粮食安全产生影响。本研究采用非参数弹性研究全球四大作物（小麦、水稻、玉米和大豆）产量对三个气候变量（降水量（PRE）、潜在蒸散量（PET）和平均气温（TMP））的敏感性。在研究期间（1961年至2014年），气候变量和作物产量都存在趋势和序列相关性。为了克服这一局限性，采用了无趋势预美白（TFPW）方法。作物产量在全球范围内对TMP最为敏感。但具体的灵敏度因大陆而异。最敏感的地区位于东南亚的部分地区。在西欧和北方美洲，小麦产量对TMP更敏感，而在南美洲以及东非和西非部分地区，玉米对TMP的敏感性更高。大豆在北美和南美更敏感。在大多数地区，小麦和水稻产量对TMP的弹性是负的（即TMP增加会降低产量），而玉米则是正的，大豆则是正负混合的信号。PRE对作物产量的影响较小。非参数弹性概念是一种简单而有效的方法，它补充了现有的线性模型方法，用于检测气候变化对作物产量的影响，并可用于调查气候变化对地方到全球规模的农业生产的未来后果。
Climate variability controls crop yield variability with impacts on food security at the local, regional and global levels. This study uses non-parametric elasticity to investigate the sensitivity of crop yields of the top four global crops (wheat, rice, maize, and soybean) to three climate variables (precipitation (PRE), potential evapotranspiration (PET), and mean air temperature (TMP)). Trends and serial correlations exist in both climate variables and crop yields over the study period (1961 to 2014). To overcome this limitation, the Trend Free Pre-Whitening (TFPW) method was applied. Crop yields are most sensitive to TMP globally. But the exact sensitivity varies across continents. The highest sensitivity regions are located in parts of the Southeast Asia. Wheat yields are more sensitive to TMP in Western Europe and Northern America, whereas maize has higher sensitivity to TMP for regions located in South America and parts of Eastern and Western Africa. Soybean is more sensitive in North and South America. The elasticities of wheat and rice yields to TMP are negative in most of the regions (i.e. increased TMP decreases yield), whereas maize witnessed positive and soybean witnessed mixed positive and negative signals depending on the region. PRE has lower influence on crop yields. The non-parametric elasticity concept is a simple and an efficient approach that complements the existing linear models methods used to detect climate change impacts on crop yields and can be used to investigate the future consequences of climate change on local to global scale agricultural production.