区域农业水土资源复合系统恢复力时空特征分析及其约束效应研究

In the course of regional agricultural water and soil resources development and utilization, the system resilience has been ignored in the past, so that it was easy to generate the overload operation of the system. Aiming at the aforesaid defect, taking Sanjiang Plain as the research platform, agricultural water resources and soil resources will be seen as a composite system to comprehensively study its resilience through synthetically applying experimental analysis and modern data processing methods. First of all, the suitable evaluation index system of regional agricultural water and soil resources composite system resilience will be constructed, the comprehensive resilience index will be calculated, the optimal resilience diagnosis method will be screened, the spatial-temporal variation characteristics of the resilience will be analyzed, and the spatial variability zone map of the resilience will be drawn. After that, the key driving factors of the resilience will be calculated and obtained, the comprehensive driving mechanism of the resilience will be dissected, and then the scenarios simulation framework of the resilience will be constructed for analyzing its future evolution trend. At last, the agricultural planting structure optimization model will be established under the resilience constraint, the new method for solving model will be explored, and the agricultural planting structure optimization layout scheme will be confirmed combining with the efficiency evaluation of optimization adjustment. And then the constraint degree will be calculated for analyzing the resilience’s constraint effect on agricultural planting structure. The research achievements can provide technical and theoretical supports for solving problems of incomplete measurement and analysis methods of the system resilience, difficulty in solving optimization models and lack of the optimization adjustment efficiency evaluation and the constraint effect analysis, and have important theoretical and practical significance for promoting high efficient utilization of regional agricultural water and soil resources and regional sustainable development.

针对以往区域农业水土资源开发利用中忽视系统恢复力、容易造成系统超负荷运行的弊端，以三江平原为研究平台，将农业水、土资源视为一个复合系统，综合运用试验分析与现代数据处理方法，对其恢复力进行整合研究。首先构建农业水土资源复合系统恢复力适宜评价指标体系，计算综合恢复力指数，筛选恢复力最优诊断方法，分析恢复力时空变化特征，绘制恢复力空间变异分区图；然后计算获取恢复力关键驱动因子，剖析其综合驱动机制，进而搭建恢复力情景模拟框架，分析其未来演变态势；最后建立恢复力约束下的农业种植结构优化模型，探索模型求解新方法，并结合优化调整效率评价确定农业种植结构最优布局方案，进而计算约束度，分析恢复力对农业种植结构的约束效应。研究成果可为破解系统恢复力测度分析方法不完善、优化模型求解困难、优化调整效率评价与约束效应分析缺乏问题提供技术理论支撑，对于促进区域农业水土资源高效利用与可持续发展具有重要理论与现实意义。

作为农业水土工程领域的研究热点，区域农业水土资源复合系统恢复力研究是实现农业水土资源科学管理与高效利用的重要前提。本项目以粮食生产核心区的三江平原为研究平台，将恢复力理论引入到农业水土资源复合系统分析中，全面完成了预定的区域农业水土资源复合系统恢复力诊断及其时空变化特征、恢复力驱动机制解析及其未来情景模拟以及恢复力对农业种植结构的约束效应三方面的研究任务。在系统解析农业水土资源复合系统恢复力内涵的基础上，运用累计信息贡献率、R聚类和灰色关联度等方法优选了评价指标，解决了传统指标体系构建偏于片面或偏于宽泛的问题；综合运用粒子群、萤火虫及模拟退火等现代智能优化算法率定了支持向量机、投影寻踪及TOPSIS等评价模型的关键参数，实现了区域农业水土资源复合系统恢复力的准确测度，在此基础上借助ArcGIS技术深入分析了恢复力时空变化特征，解决了恢复力诊断方法薄弱问题；采用主成分分析、决策试验与评价试验室等方法识别了农业水土资源复合系统恢复力的关键驱动因子，运用熵权-神经网络组合模型模拟预测了恢复力未来情景，解决了农业水土资源复合系统恢复力驱动机制及未来演化特征不清问题；将恢复力诊断结果嵌入到约束条件中，建立了恢复力视角下的农业种植结构优化模型，运用多目标布谷鸟算法和多目标粒子群算法进行了模型求解，解决了传统优化模型约束条件不完善与求解精度不高问题。另外，本项目还进行了拓展研究，运用基于鲸鱼算法优化的随机森林模型和基于风驱动算法优化的投影寻踪模型评价了洪水灾害恢复力；运用最小二乘-灰色关联度模型评价了灌区灌溉用水效率；运用WEAP模型分析了水文干旱特征；运用蚁群算法优化配置了水资源。研究成果丰富完善了系统恢复力测度应用技术理论体系，为提高农业水土资源系统运行效益、保障区域水土资源安全、粮食安全及生态安全提供了科学依据。