Logarithmic Mean Divisia Index (LMDI) decomposition analysis of changes in agricultural water use: a case study of the middle reaches of the Heihe River basin, China

To quantitatively analyze the main driving factors of agricultural water use in different stages in the middle reaches of the Heihe River basin, the Logarithmic Mean Divisia Index (LMDI) decomposition method was employed to calculate the contribution of each driving factor to agricultural water use. The crop-planting scale, cropping pattern, irrigation quota, and irrigation efficiency of different crops were chosen as representative factors of agricultural water use. The study revealed that (1) from 1991 to 2015, agricultural water use exhibited a fluctuating growth trend that resulted in a 0.031 billion m(3) increase in use. From 1991 to 2001, agricultural water use increased by 0.069 billion m(3), and from 2002 to 2015, it decreased by 0.038 billion m(3). (2) In each research period, the expansion of the crop-planting scale and unreasonable cropping patterns increased agricultural water use. However, decreases in irrigation quotas and improvements in irrigation efficiency decreased agricultural water use. The contributions of these changes were 1.138 billion m(3), 0.109 billion m3, -1.08 billion m(3), and -0.136 billion m(3), respectively, from 1991 to 2015. Comparing the period 1991 to 2001 with 2002 to 2015, the increase associated with the crop-planting scale and the decrease related to irrigation quotas were prominent and dramatically changed agricultural water use. (3) The effects of crops varied in different research periods. From 1991 to 2001, the contribution of cash crops' increase was 0.446 billion m(3), which was more prominent than that of food crops' decrease (-3.78 billion m(3)), and from 2002 to 2015, the agricultural water use was decreased for all crops except maize. In conclusion, the best measures to decrease agricultural water use in the middle reaches of the Heihe River basin are to control the crop-planting scale and optimize the cropping pattern. The results of this study indicate how diverse determinants affect agricultural water use and provide insight for local agricultural water savings.