Probabilistic Approaches to Data Assimilation for Earth Systems (13w5139)

Arriving in Banff, Alberta Sunday, February 17 and departing Friday February 22, 2013


Mark Buehner (Environment Canada)

Alexandre Chorin (University of California, Berkeley)

(Université de Québec à Montréal)

(University of Maryland)

Robert Miller (Oregon State University)


As Planet Earth faces critical issues associated with future climate change, monitoring and prediction of the earth system is a pressing matter at a wide range of temporal and spatial scales. Data assimilation is necessary for any retrospective climate analysis, and it provides a foundation for scientific monitoring and prediction activities. The objective of this intensive workshop is to advance data assimilation for earth systems by bringing together mathematicians, particularly those working in dynamical and stochastic systems, statisticians, and domain scientists who have a vested interest in fusing data with models. In the highly successful 2008 BIRS workshop, key mathematical questions in data assimilation were identified for atmosphere, ocean, and coupled atmosphere-ocean systems. The years following the workshop have seen a blossoming of new approaches to data assimilation, much of which has proceeded along the lines discussed in depth at BIRS in 2008. The area is now ripening for fostering further innovative ideas through effective interdisciplinary collaborations. Two main thrusts of this workshop are thus 1) mathematical approaches to data assimilation methodology and 2) advancement of ESPC as scientific monitoring and prediction.

The timing of this workshop is chosen judiciously to collectively maximize the impact of this workshop at a time of strong growth in this field. As our view of the earth system has broadened to encompass coupled physical, chemical and biological systems, weaknesses of present methods with their restrictive assumptions have become apparent, and new approaches to data assimilation will be needed. Because the BIRS workshop can bring together more mathematicians and statisticians for a period of time and at a place where the intensive focus and energy can serve to define the way forward, it offers a unique and much needed opportunity to go far beyond what we were able to achieve at the previous programs. In 2013 five years will have elapsed since the previous BIRS workshop. A number of those on our list of suggested participants have expressed specific interest in the proposed 2013 workshop.

Advanced data assimilation will form a key component of any systematic description of the earth system. This workshop will showcase the essential relevance of mathematics and interdisciplinary innovation for the advancement of the problems associated with the earth system. It is therefore particularly appropriate as part of the Mathematics of Planet Earth 2013 (MPE2013). To enhance our effort to advance data assimilation as interdisciplinary science, a 2013 summer school organization is in the works at the Center for Mathematical Modeling and Scientific Computation (CSCAMM), University of Maryland. CSAMM is a MPE201 partner institution. By bringing together the leading experts at the BIRS workshop while aiming at the training of young scientists at the CSCAMM summer school, we will also combine activities between the institutions and thus broaden impact of both programs.