石羊河流域水-作物-生态系统耦合协调度时空演变与调控潜力研究

Spatiotemporal Evolution and Regulation Potential of Water-Crop-Ecosystem Coupling Coordination Degree in Shiyang River Basin

摘要:

针对西北干旱区农业水资源短缺与生态环境保护的矛盾，本文以石羊河流域为研究区，基于“压力-状态-响应”框架构建包含12个指标的水资源-作物生长-生态环境系统评价体系，采用熵权法确定权重并进行共线性检验。将贝叶斯网络引入耦合协调度评价，实现从状态描述到因果诊断的方法突破。利用多源遥感数据反演植被生理参数和环境指标，采用月尺度分析2023年4—10月作物生长周期系统演变规律。结果表明：水资源、作物生长子系统的发展指数均呈先升后降的单峰曲线趋势，分别在6、7月达到峰值0.64和0.89；生态环境子系统发展指数则持续缓慢上升（由0.43提高至0.62）；耦合协调度呈显著月际变化，从4月的0.46（面临失调）升至7月峰值0.80（中级协调），后降至10月的0.57（临界协调），与作物生长周期高度吻合；贝叶斯网络识别出不同生长阶段主控因素的动态转换：4月灌溉主导，5—7月转为肥力限制，8—10月需水肥盐协同调控；空间上中部地区协调度持续较高，下游民勤地区长期处于低值区。研究结果为干旱区农业水资源精准管理提供了“评价-诊断-调控”的完整技术体系。

Abstract:

Aiming to address the contradiction between agricultural water scarcity and ecological conservation in the arid regions of Northwest China, the Shiyang River Basin was taken as its research area. Based on the ‘pressure-state-response’ framework, a water resources-crop growth-ecological environment system evaluation framework comprising 12 indicators was established. The entropy weight method was employed to determine weights and conduct multicollinearity tests. Bayesian networks were introduced into the coupling coordination degree evaluation, achieving a methodological breakthrough from state description to causal diagnosis. Using multi-source remote sensing data to derive vegetation physiological parameters and environmental indicators, a monthly-scale analysis was conducted to characterize the system evolution patterns during the crop growth cycle from April to October 2023. The results indicated that the development indices of both the water resources and crop growth subsystems exhibited a single-peak curve trend with an initial rise followed by a decline, reaching peaks of 0.64 and 0.89 in June and July, respectively;the ecological environment subsystem development index continued to rise slowly (from 0.43 to 0.62);the coupling coordination degree exhibited significant monthly variations, rising from 0.46 in April (facing imbalanced) to a peak of 0.80 in July (moderately coordinated), and then decreased to 0.57 in October (critically coordinated), which was highly consistent with the crop growth cycle;the Bayesian network identified dynamic shifted in the dominant factors across different growth stages: irrigation-dominated in April, fertility-limited from May to July, and requiring coordinated regulation of water, and fertility from August to October;spatially, the central region maintained consistently high coordination levels, while the downstream Minqin area remained in a low-value zone. The research results can provide a comprehensive technical framework for precise management of agricultural water resources in arid regions, encompassing ‘evaluation-diagnosis-regulation’.