气候变化背景下黄河流域干旱情景预估与响应机制研究

Under the background of climate changing, it is necessary to do some work on drought prediction and hydrological process research. Taking the Yellow River basin as the study area, we will study the spatio-temporal characteristics of drought and predict the trend of drought, based on multi-model integrated, model simulating, and statistics analyzing methods. The study themes include: (1) Due to the incapability of commonly used statistical downscaling methods in reflecting extremes of precipitation and temperature, an equidistant quantile matching-based statistical downscaling model (EQSM) will then be developed to effectively improve the accuracy in simulating temperature and precipitation extremes; (2) In consideration of the deficiencies of SC-PDSI in its water balance model basis and spatio-temporal comparability, this study constructs drought assessment index (VIC-PDSI) integrating VIC model and SC-PDSI, which could serve as the useful indicator for regional drought monitoring and assessment; (3) Based on the simulated result of statistical downscaling and VIC-PDSI, the Copula is used to predict the multivariate distribution of drought features under 23 climate models and three climate emissions scenarios. The study will contribute to drought monitoring and prediction in Yellow River basin and similar areas under climate change, providing scientific basis and technical support for drought management and decision-making.

本项目以黄河流域为研究对象，针对变化环境下流域水文过程演变以及干旱情势监测的需求，通过模型耦合、数值模拟、统计分析，探究干旱发生演变时空特征，预估和分析干旱的变化趋势，内容包括：①发展等距离分位数统计降尺度模型（EQSM），克服全球气候模式常用统计降尺度方法无法反映降水和气温极值的缺陷，实现全球气候模式与分布式水文模型的完整对接；②利用分布式水文模型和SC-PDSI指数的优势，构建黄河流域具有物理机制、地区适应性和时空可比性强的干旱评估模式（VIC-PDSI），提高黄河流域干旱过程模拟和评估的精度；③通过降尺度模型、VIC-PDSI干旱评估模式、多变量联合概率分布函数，预估未来不同气候模式和不同排放情景下黄河流域干旱发生概率、重现期的变化趋势，揭示全球气候变化背景下流域尺度干旱的多维响应机制。研究成果将为黄河流域及类似地区气候变化背景下的旱情评估、预测与管理决策提供科学依据和技术支撑。

近年来，黄河流域干旱频发，对生产、生活和生态造成了严重影响。借助合理的干旱指标体系，预估未来时期的干旱演变趋势，有利于指导人们有意识地防旱、抗旱，规避干旱风险，从根本上提高社会应对干旱的能力并降低面对旱灾的脆弱性。本项目首先基于所发展的等距离分位数统计降尺度模型，构建了黄河流域和中国区域的未来气候情景数据集，极大地提高了CMIP5的模拟精度和空间分辨率，为分布式水文模型提供了气候情景驱动数据。然后，基于VIC模型和自率定PDSI指数，构建了具有物理机制、时空可比性好、地区适应性强的干旱评估指数，提高了黄河干旱过程模拟和评估的精度。最后，基于高精度未来情景数据、VIC-scPDSI干旱指数和“强度-面积-时间”方法、多变量联合概率分布函数，预估未和分析了未来黄河流域和中国区域干旱的时空演变趋势。研究成果对于提高干旱事件识别的准确度和降低干旱变化趋势预估的不确定性，制定黄河流域干旱预警与应对策略等极具理论与现实意义。