基于星地多源数据的干旱演变过程连续监测与定量分析方法研究

Remote sensing based method for drought monitoring and analysis has become one core method for drought disaster response. However, most current methods only focus on the average status or limited profiles of drought event, resulting in time gap and blind area. So there is still lack of continuous monitoring for drought propagation process. This paucity of research on the drought propagation process further limits the accuracy of drought analysis and prediction. Therefore, this project focuses on the scientific question of continuous monitoring and quantitative analysis for drought propagation process, and 1) analyzes several core drought variables in meteorology, hydrology, soil, vegetation, and human activity, then builds satellite-ground observation oriented information model for drought propagation process; 2) based on this information model, studies the synergic relationship between satellite remote sensing and ground observation, and proposes continuous monitoring method based on satellite-ground fusion for drought propagation process; 3) based on the continuous data obtained from this monitoring method, constructs quantitative analysis methods for calculating the spatio-temporal characteristic and typology features of drought propagation. In the experiment, this project will select several typical agricultural production areas for validation and comparison purpose. The innovation method proposed by this project, i.e., continuous monitoring and quantitative analysis for drought propagation process, can provide new dataset and method for high efficient drought monitoring and scientific drought analysis. The outcome of this project is also promising for constructing a new drought monitoring operational system based on drought propagation process, which will be valuable for improving the efficiency of drought monitoring and prediction.

基于遥感的干旱监测分析，已经成为应对干旱灾害的核心方法之一。现有研究主要针对干旱灾害事件的平均状态或断面信息，监测数据存在时间间断和空间盲区，因此缺乏对干旱演变过程的连续监测，进而制约了干旱分析和预测的准确性。为此，本项目针对干旱演变过程连续监测与定量分析的科学问题，1）分析气象、水文、土壤、植被和人类活动等多种干旱核心因子，建立面向星地监测的干旱演变过程监测信息模型；2）基于监测信息模型，研究卫星遥感和地面监测的互补特征，提出基于星地数据融合的干旱演变过程连续监测方法；3）基于连续监测方法获取的连续监测数据，构建干旱演变过程的时序特征、空间特征和形态特征的定量分析方法。最后以我国主要农业产区为典型研究区域，开展验证与对比实验。本研究提出的创新方法，能够为干旱遥感高效监测与科学分析提供新的数据支撑和技术方法，并有助于基于干旱演变过程规律构建新的干旱监测业务系统，提升干旱监测和预测的效率。

干旱作为一种典型的灾害，严重威胁我国粮食安全和社会稳定，成为制约社会经济发展的主要挑战之一。针对现有研究采用平均状态或断面数据，监测分析存在时间间断和空间盲区的关键问题。本项目针对干旱演变过程连续监测与定量分析的科学问题，取得如下研究成果：.1）分析了影响干旱发生发展的气象、水文、土壤、植被和人类活动等多种因子，开展了基于最优指数的长江中下游干旱监测与演变过程分析，揭示了多种单一和复合干旱指数在我国农业主产区长期干旱和骤发干旱监测分析中的适用性；.2）研究了卫星遥感和地面监测的互补特征，提出基于星地数据融合的干旱关键变量（降水、土壤水分和气温）连续监测方法，生成了长江流域时间连续、空间无缝且精度较高的融合数据，为后续干旱监测分析提供了高质量数据基础；.3）构建了干旱演变过程的定量分析方法，针对我国长江流域和华北平原，采用线性和非线性模型，明确了干旱发生发展过程中演变时长和烈度的定量规律，特别是分析了2003年前后三峡工程建设对于干旱演变过程的影响，并进一步提出了多种未来干旱格局的预测方法；.4）项目成果共发表研究论文27篇，其中包括WRR、GRL、JOH、AFM和RSE等权威期刊论文10篇；出版专著《基于多源数据融合的干旱综合监测与分析》1部；培养硕博士研究生11名，其中4人获得研究生国家奖学金；参加中国地理学会、国际农业信息化和青年地学论坛等国内外学术会议10余次。.本研究提出的创新方法，能够为干旱遥感高效监测与科学分析提供新的数据支撑和技术方法，并有助于未来基于干旱演变过程规律构建新的干旱监测业务系统，提升干旱监测和预测的效率。