研究采用 LMDI( logarithmic mean divisia index) 指数分解模型, 将“十一五”“十二五”时期西北 5 省( 自治区) 农业虚拟水变化的驱动因素分解为结构效应、强度效应、经济效应和需求效应等, 并分别探究不同效应对虚拟水增长的贡献情况及其时空差异。结果显示: 农业虚拟水含量在产品结构分布上, 畜牧产品> 三大主粮> 经济作物。从 时间角度来看, 2006- 2015 年全区域农业虚拟水含量呈先上升、再下降、再上升的趋势, 经济效应和强度效应分别 是虚拟水含量变化的主要促进因素和主要抑制因素, 总体上, 经济效应> 强度效应> 结构效应> 需求效应。“十一 五”时期, 该区域虚拟水含量变化主要受地区经济增长的影响;“十二五”时期, 农业生产结构调整对虚拟水含量增长的抑制作用逐渐增强。从空间角度来看, 2006- 2015 年 5 省( 自治区) 虚拟水变化的总效应省际差异显著, 新疆正 向效应最强, 为 194.26 亿 m3 , 其次为宁夏、青海和陕西, 负向效应中甘肃最强, 下降了 94.11 亿 m3。据此, 从改进 农业生产技术、调整生产结构及水生态补偿等方面提出建议, 以提高水资源利用率, 促进区域经济生态协调发展。

The Log arithrnic Mean Divisin Index ( LMDI) index decomposition model are used to divide the driving factors of agricultural virtual w ater content change into structural, intensity, economic and demand effects in the five northwest provinces( autonomous regions) from the 11th Five-Year plan and 12th Five-Year plan to explore their contributions and its spatial-temporal differences. The research results show that: the agricultural virtual water content in the product structure distribution, livest ock products> the three major staples> commercial crops. From the perspective of time, from 2006 to 2015, the agricultural virtual water content in the whole region showed a trend of rising-falling-rising. The economic effects and intensity effects were the main promoting and inhibitory factors of the virtual water content change. In general, Economic effect> intensity effect> structural effect> demand effect, respectively . During the 11th Five-Year Plan period, the change of virtual water content in the region was mainly affected by regional economic grow th. During the 12th Five-Year Plan period, agricultural production structure the adjustment has gradually inhibited the growth of virtual water content. From a spatial perspective, the inter-provincial difference in the total effect of the change invirtual water co ntent is sig nificant in the five provinces( autonomous regions) from 2006 to 2015. Xinjiang has the strongest positive effect at 11 9.426 × 1010 m3 , followed by Ningxia, Qinghai, and Shaanxi. Negative effects in Gansu is strongest dropped by 9.411×109 m3 . Based o n this, suggestions from the aspects of improving agricultural production technolog y, adjusting the production structure, compensating the water ecological environment to reduce the virtual water content in agriculture, improving the utilization of water resources, and promoting the coordinated development of regional economics and ecology .

国家自然科学基金( 41701613) ; 中央高校基本科研业务费项目( 2019B23014)