变化环境下湿润区干旱演变规律及风险响应研究

Drought is caused by water deficit and dominated by water cycle. Various types of droughts interact with each other. Due to the rapid increase of water demand and the change of water use structure, a higher precision and resolution of spatial-temporal scale of drought should be put forward for the hazard management. At present, the passive resistance against drought disasters has been more proactive, and the emergency management has been transformed into the risk management. Global change and human activities have exerted influences on the temporal-spatial distribution and evolution of drought. In humid area of the southern of China, increasing extreme droughts and an imbalance of water demand and supply under the rapid development of regional economy has been paid more attentions by local government for disaster prevention and reduction. Therefore, in the Pearl River basin as the case area of a humid region, studies on droughts are carried out from three aspects as follows: drought properties identification and its risk assessment, evolution mechanism and risk transform among various types of drought, and their responses under climate change and human activities. The adaptability of various types of drought indices in humid area are investigated. Several indices are expected to be improved. Multivariate droughts with various spatial-temporal scales are simulated and risk assessment system are built. The evolution mechanism and risk transform of drought in humid area are determined and the impacts of main driving factors on them are quantified.

干旱是由于水分亏缺所引起，但同时受到水分循环过程控制。干旱类型主要有气象干旱、农业干旱、水文干旱和社会经济干旱，它们之间相互作用。区域水资源需求快速增加及需水结构的变化，对干旱灾害管理提出了更高的时空尺度分辨精度要求。消极被动抗旱和应急管理，也逐渐转变为积极主动抗旱与风险管理。全球变化和人类活动,已经影响了干旱时空分布及演变规律。我国南方湿润地区极端干旱事件在不断增加，地区经济快速发展产生的水资源供需矛盾，已引起地方防灾减灾管理的高度重视。因此，本项目以珠江流域为典型研究区，拟从干旱特征识别及风险评估、干旱演变驱动机制及风险传递规律、气候变化和人类活动影响下的干旱演变及风险响应三个方面开展研究。分析湿润区不同类型的干旱指标适应性及改进方法，构建湿润区多尺度干旱多维特征识别及风险评估体系,揭示湿润区干旱驱动机制及不同类型干旱风险传递规律,量化主要驱动因素对湿润区干旱演变及风险响应的影响程度。

由于不同类型的干旱特性、区域水资源供给和需求变化和干旱防御政策转变等多因素影响，湿润区干旱评估体系、演变规律和风险响应已经成为亟待解决的热点问题。本项目通过构建湿润区多尺度干旱多维特征识别及风险评估体系，辨析干旱驱动机制和不同类型干旱风险传递规律，以及量化变化环境下的湿润区干旱演变及风险响应。基于水循环过程、水量平衡关系和多元多维数据分析方法，结合大数据和具体案例研究构建了标准化水循环指数、改进的标准化旱涝急转指数、标准化供需水指数等多种新的干旱评估指数，分别可用于评估气象干旱、水文干旱和社会经济干旱等多种干旱类型的多维干旱特征。通过深度学习算法、多元数据分析方法，建立了多种干旱类型的驱动因素分析和特征联合分布模型。利用CMIP6数据集和南方湿润区气象水文及社会数据，通过全球对比、聚焦珠江流域及其东江流域，开展了多尺度、长历时和多类型的干旱特征评估，建立了新的时空耦合特征分析方法，并依托于数据挖掘和机器学习等手段对变化环境下干旱指数和干旱特征进行精准预测。整体而言，本项目建立了操作简便且计算精准的区域适应性干旱指标，实现了基于多尺度多维干旱评估，揭示了多类型干旱之间的传递规律和驱动因素影响，量化了变化环境下的干旱演变规律及其对风险的响应能力，为变化环境下南方湿润区干旱防御和水资源精细管理提供了理论和方法基础。