International Meeting on Statistical Climatology

PI：Petter Guttorp提案编号：DMS 9725031项目：国际统计气候学会议摘要：这次国际统计气候学会议是自1979年以来第六次举行，将于1998年5月25日至29日在不列颠哥伦比亚省举行。这一系列国际会议聚集了统计学家和气候学家，就问题和解决问题的统计方法进行了正式和非正式的讨论。每天有两场受邀的讲座，每一场都有统计学家和气候学家的讨论。这种形式促进了跨学科的互动和协作。气候学中的许多主要问题都与大气环流模型有关。令人感兴趣的主要问题是将模型运行与历史数据进行比较，历史数据具有不同的空间和时间覆盖范围、不同的工具和非常不同的精度。这些模型被用来评估各种人为影响对气候的影响，例如各种温室气体成分的变化。本次会议将讨论的重要主题是可用于开发有用模型的统计技术，如隐马尔可夫模型、小波方法和神经网络。该奖项用于资助来自美国的参赛者的费用。助学金的很大一部分将用于研究生、新研究人员和代表性不足群体的成员的费用。

PI: Petter Guttorp Proposal Number: DMS 9725031 Project: International Conference on Statistical Climatology Abstract: This International Conference on Statistical Climatology, the sixth that has been held since 1979, will be held from May 25 to 29, 1998, in British Columbia. This series of international conferences has brought together statisticians and climatologists for both formal and informal discussions of problems and statistical methods for addressing them. There are two invited lectures per day, each with discussions by both statisticians and climatologists. This format promotes cross-disciplinary interactions and collaborations. Many of the leading issues in climatology are related to general circulation models. The major problem of interest is in comparing model runs to historical data, which have varying spatial and temporal coverage, varying instrumentation, and very different precision. These models are used to assess the effect of various anthropogenic influences on the climate, such as changing composition of various greenhouse gasses. Important topics to be discussed at this conference are the statistical techniques such as hiden Markov models, wavelet methods, and neural networks that can be used in developing useful models. This award is in support of expenses of participants from the United States. A large portion of the grant will go towards expenses of graduate students, new researchers, and members of underrepresented groups.