Sichuan Basin is a major grain-producing area in Sichuan Province. Agricultural water usage also accounts for 65.5% of the total provincial water consumption. However, the low efficiency of agricultural water use has limited water resource security and sustainable utilization in the context of significant seasonal and regional water shortages. It is very necessary to explore the crop water use for water use efficiency. In this study, the crop planting structure was optimized to evaluate the crop water use efficiency from the perspective of water footprint. Firstly, the water consumption of three major grain crops in the Sichuan Basin was quantified by water footprint theory. An analysis was made of the temporal and spatial variation of crop water consumption. Secondly, a multi-objective optimization model was constructed to minimize the blue and gray water footprints for the economic benefits. The optimal crop planting structure was obtained under different water-saving scenarios (5%, 10%, and 15%). Finally, the water resources efficiency was used to clarify the agricultural water use efficiency before and after optimization. The results showed that: 1) The water consumption of crop production showed a downward trend from 2001 to 2021. But the downward trend was significantly different. The blue water footprint and irrigation loss decreased the most, whereas, the green water footprint decreased the least. There was a gradual decrease from northeast to southwest, indicating the significant spatial difference. 2) Under the scenarios of water saving 5%, 10%, and 15%, the optimization of planting structure showed that the area of rice and wheat decreased, whereas, the area of corn increased. Among them, the blue water footprint decreased by 21.27%, 22.24%, and 23.71%, respectively, while the gray water footprint decreased by 12.68%, 13.39%, and 18.46%, respectively, and the agricultural output value increased by 3.47%, 2.48%, and 0.96%, respectively. 3) The crop water use efficiency first fluctuated and then slowly increased, thus reaching a peak in 2021; the water use efficiency of crops showed a non-equilibrium trend with the high in the north and south, while the low in the middle in space. 4) After the optimization of the planting structure, the total efficiency of water use increased by 1.68%, 1.49%, and 1.03%, respectively, under the scenarios of saving water by 5%, 10%, and 15%, compared with that in 2020; The increasing planting area of corn and the reducing planting area of rice and wheat was sed to more significantly improve the crop water use efficiency. There were two contributions after optimization. Firstly, the spatial difference in crop water use was clarified from the perspective of water footprint. Secondly, the optimization model of crop planting structure with strong applicability was established to realize the dynamic adjustment of crop area. The finding can provide a strong reference to managing agricultural water resources, in order to improve crop water use efficiency for food security.

To explore the evolution characteristics and driving factors of agricultural water footprint in the Sichuan Basin, the spatiotemporal evolution characteristics of blue water, green water, grey water, and white water footprint of major crops in the Sichuan Basin from 2001 to 2021 were analyzed. The LMDI model was used to identify the driving factors of crop water footprint change from the technical, environmental and social dimensions, and the differences of each driving factor were discussed from the global and local perspectives. The results showed that the crop water footprint in Sichuan Basin increased first and then decreased from 2001 to 2021. The spatial distribution characteristics of crop water footprint in each city were high in the north and low in the south. From the global perspective, the impact of different driving factors on changes in crop water footprint varies from large to small as follows: economic effect, population change effect, technological effect, water footprint intensity effect, water-saving effect, and production effect. The economic effect and population change effect were the strongest promoting and inhibiting factors, and the contribution values were 82.90×10⁸ m³ and -51.55×10⁸ m³, respectively. From the local perspective, there was a dislocation between the technical effect and the population change effect in space, and the production effect, water saving effect and water footprint intensity effect show a two-way driving effect in space.