Collaborative Research: Development of 21st-Century Precipitation Scenarios Using Probabilistic Downscaling Techniques

在20世纪后半叶，美国大部分毗邻地区的年总降水量增加，这在很大程度上是由于大型降水事件的不成比例变化，并激发了人们对预测21世纪中叶和末期降水趋势的兴趣。大气-海洋环流耦合模式是建立这种预测的主要工具，但需要缩小尺度工具来获得与评估影响相关的空间尺度上的气候变化信息。缩小尺度方法往往低估地球物理变量的可变性，因此在试图量化极端(或大型)事件的变化时用处有限。这项研究项目的重点是应用创新的统计降尺度工具来从多个耦合的AOGCM输出，以获得整个美国毗邻地区的站点的日降水量和雨日降水量的概率分布的描述符。这一结果将使我们能够检验21世纪降水事件的频率和幅度，与20世纪新开发的长达一个世纪的降水记录相比。大量使用的AOGCM模拟也将提供一个前所未有的机会来明确考虑基于模型的不确定性，从而评估对衍生项目的信心。在直通项目下进行的研究与政府间气候变化专门委员会(IPCC)第四评估小组利用该计划产生的气候模式产品的活动密切相关。它也与北美区域气候变化评估计划(NARCCAP)有很强的互补性，NARCCAP只关注包括降水在内的多个变量的动态缩小尺度。将该项目的结果与NARCCAP的结果进行比较，将包括统计和动态缩小尺度，以全面概述各种可能的气候状态和轨迹。研究结果将通过出版物、在专业会议上的陈述和一个专门向气候影响社区提供研究结果的网站来传播。该项目的更广泛影响还包括对下一代气候变化科学家的教育和培训。

Increased annual total precipitation across much of the contiguous U.S. during the latter portion of the 20th century has largely been attributed to a disproportionate change in large magnitude precipitation events and has spurred interest in projecting precipitation trends into the middle and end of the 21st century. Coupled Atmosphere-Ocean General Circulation Models (AOGCMs) are the primary tools available for constructing such projections, but downscaling tools are needed to obtain climate change information at spatial scales relevant for assessing impacts. Downscaling methods have tended to underestimate the variability of geophysical variables and hence are of limited utility when seeking to quantify changes in extreme (or large magnitude) events. This research project focuses on application of innovative statistical downscaling tools to output from multiple coupled AOGCMs to obtain descriptors of the probability distributions of daily precipitation occurrence and wet-day precipitation amount for stations throughout the contiguous USA. The results will allow us to examine the frequency and magnitude of precipitation events in the 21st century relative to those derived from newly-developed century-long precipitation records from the 20th century. The large number of AOGCM simulations used will also allow an unprecedented opportunity to explicitly consider model-based uncertainty and thereby assess confidence in the derived projections.The research conducted under through project closely relates to the activities of the 4th Assessment team of the Intergovernmental Panel on Climate Change (IPCC) by utilizing climate model products generated within that program. It also is highly complementary with the North American Regional Climate Change Assessment Program (NARCCAP), which is focused solely on dynamical downscaling of multiple variables, including precipitation. The comparison of the results of this project with those from NARCCAP will encompass both statistical and dynamical downscaling to present a comprehensive overview of the range of possible climate states and trajectories. The research results will be disseminated through publications, presentations at professional meetings, and a web site dedicated to making the results available to the climate impacts community. The broader impacts of the project also include education and training of the next-generation of climate change scientists.