特邀报告人：丹尼斯·拉特玛依尔, 美国工程院院士 Speaker：Dennis Lettenmaier, Member of National Academy of Engineering, USA 题目：干旱监测与预测 Title：Progress in Drought Monitoring and Prediction 时间：2015年5月29日 17:15-17:45 Time：May 29th 2015 17:15 pm -17:45 pm 摘要 干旱作为一种普遍发生的自然灾害， 每年在全球范围内引发高达数百亿美金的 经济损失。然而目前对干旱的预测能力非 常有限，现有研究成果最多是基于对水文 初始条件的理解，缺乏对超过一个星期甚 至更长时间后的降雨量的预测能力。但是 目前对干旱监测的研究已经有了实质的进 展，通过改进的地表模型，可以预测实际 监测中无法直接获取的农业和水文干旱的 首要指标-土壤湿度。地表模型也被用来 重建过去的干旱情景，可以把现在发生的 干旱灾害放入历史背景中。本次报告将介 绍在华盛顿大学和加州大学洛杉矶分校使 用基于模型的地表水监测仪进行干旱监测 的情况，其数据会每日更新，并已被美国 国家海洋气象局气候预测中心用来监测美 国干旱情况。此外，还有一台更高分辨率 的美国西海岸监测仪，对目前加利福尼亚 干旱状况监测。同时，报告还会回顾已有 的历史干旱情景重建工作， 包括中国 1950-2006年间的干旱情况，以及目前正 在试图进行中国19世纪大饥荒以来的干旱 情景重建工作。 Abstract Drought is a pervasive natural hazard, with economic implications that run to the tens of billions of dollars globally each year. Nonetheless, the ability to predict droughts is tenuous at best, and rests mostly on understanding of hydrologic initial conditions, absent the ability to predict precipitation more than a week or so in advance. There have, however, been substantial advances in drought monitoring, made possible mostly through improved land surface models, which can predict soil moisture – the primary index to agricultural and hydrological drought – in the absence of direct measurements. Land surface models have also been used to reconstruct past droughts, which allows us to place contemporary drought in the context of the historic past. The speaker review the experience at the University of Washington and UCLA with a model-based Surface Water Monitor, which is updated daily, and is used by NOAA’s Climate Prediction Center in its contribution to the U.S. Drought Monitor, and a higher resolution U.S. West Coast monitor, including its use in the ongoing California drought. The speaker also review past drought reconstructions, including for drought over China for the period 1950-2006, and ongoing attempts to reconstruct drought over China including famines of the late 1800s. | 
