Mathematical Sciences:Collaboration Between Statistical and Atmospheric Sciences on Modeling the Climate System

9312686卡茨气候系统的建模和分析是一个复杂的问题，对统计、数学和地球物理科学提出了挑战。加强这些科学之间的合作将导致对气候系统及其模型的数据进行更有效的分析和更准确的解释，并对伴随而来的不确定性进行更可靠的评估。有人建议在国家大气研究中心(NCAR)组建一个统计科学家团队，以帮助促进这种合作。NCAR由国家科学基金会资助，是大气基础研究的中心，与北美各地的大学有着密切的联系。随着更多的统计和数学专业知识的增加，它可以成为促进地球物理、统计和其他数学科学之间合作的一个联络点。总体目标是开发和利用新的数据分析技术，这些技术源于物理学家、数学家和统计学家的集体理解。气候系统具有非常多的自由度，并呈现出一些混沌行为。因此，合作研究的具体重点将是开发适用于气候系统高分辨率分析的推理方法。这样的分析是适当的，因为气候系统的现实模型需要处理较小规模的现象，而且评估气候的社会影响需要区域或较小规模的信息。除了对直接涉及的科学领域的直接好处外，对气候系统的更好的了解将给社会带来潜在的好处，这些好处将无限期地累积到未来。例如，它可以提前在从小时到季节的广泛时间范围内改进天气和气候预报，以及更现实的未来气候变化情景。这样的改进将对决策者和政策分析人员具有重大价值。

9312686 Katz The modeling and analysis of the climate system is a complex problem that challenges the statistical, mathematical, and geophysical sciences. Increased collaboration among these sciences would result in more efficient analyses and more accurate interpretation of data from the climate system and its models, and in a more reliable assessment of accompanying uncertainties. It is proposed that a team of statistical scientist be assembled at the National Center for Atmospheric Research (NCAR) to help foster such a collaboration. NCAR, basefunded by the National Sciences Foundation, is a focal point for basic research on the atmosphere and has strong links to universities throughout North America. With the addition of more statistical and mathematical expertise, it could serve as well as a focal point for promoting collaboration among geophysical, statistical, and other mathematical sciences. The general objective is to develop and utilize new techniques of data analysis that arise from the collective understanding of physical scientists, mathematicians, and statisticians. The climate system has a very large number of degrees of freedom and exhibits some chaotic behavior. Consequently, the specific focus of the collaborative research would be on the development of inferential methods appropriate for high-resolution analysis of the climate system. Such analyses are appropriate both because realistic models of the climate system require the treatment of smaller scale phenomena, and because assessments of societal impact of climate require information on regional or smaller scales. Besides the immediate benefits to the scientific fields directly involved, improved understanding of the climate system would result in potential benefits to society that would be accrued indefinitely into the future. For instance, it could lead to improved weather and climate forecasts over a wide range of time scales from hours to seasons in advance, as well as to more realistic scenarios of future changes in climate. Such improvements would be of substantial value to decision makers and policy analysts.