Spatiotemporal coupling coordination between water network and planting structure in Yilong Lake Basin—A multi-scale perspective on the present and future

Understanding the spatiotemporal coordination between the water network (WN) and planting structure (PS) is necessary to achieve high utilization efficiency of water resources and agricultural sustainability in basins. In this study, a modified coupling coordination degree (CCD) model was constructed and a hybrid model termed multi-channel mutual information-non-Euclidean response surface modeling (MCMI-NE-RSM) was proposed to unveil the spatiotemporal evolution and heterogeneity in the WN-PS coupling coordination relationship in Yilong Lake Basin in 2003-2023; the cellular automata Markov (CA-Markov) model was introduced to predict the WN-PS relationship in the basin in 2033 and 2043. It was found that, first, from 2003 to 2023, the WN-PS CCD declined before rising, presenting a trend of continued improvement across the basin, with the minimum observed in 2012. Spatially, the CCD was higher in the northern part than in the southern part of the basin and declined outwards to the basin's edges. Second, the MCMI analysis identified the agricultural irrigation water utilization rate (X11) as the most critical variable (MI_sum = 5.094). Third, CCD measured by Z1-Z4 at the sub-basin scale shifted from a pattern of "regional aggregation and lower degrees along the edges" to a pattern of "improved regional balance". Fourth, CA-Markov modelling predicted that medium-level coupling coordination will see an increased area percentage from 52.9% in 2023 to 57.2% in 2043, hence dominating the basin; however, the area of high-level coupling coordination shrank in the central and southwestern regions of the basin.