PIs propose to hold a workshop at the Colorado School of Mines, Golden, CO; February 27-29, 2008. The goals of the workshop will be to: (1) determine the key fundamental research areas in carbonates that require new advancements to develop numerical process models for carbonates; (2) determine the time scales on which the carbonate models will operate, and how models aimed at different time scales will be coupled; (3) map the environments and associated process sets for each time scale, and evaluate the state of knowledge for each of these; (4) decide on the best strategy for converting carbonate research results into numerical models that can contribute to the global CSDMS; (5) decide on the best strategy for making carbonate research results and computational methods available for education; and (6) define the scope and timing of following phases for the carbonate modeling project.
Participants will include from 30 to 35 international leaders in carbonate sedimentology, stratigraphy, climate-modeling, ocean-modeling, siliciclastic modeling, and numerical modeling. Topics to be addressed at the workshop can include, 1) What are the grand challenges in fundamental research on carbonate facies and platform evolution that incorporate new understanding of the influences of climate, ocean systems, and biology? 2) What are the main processes at different time-scales that build carbonate platforms? 3) How should we model diagenesis, and how might we use reactive transport modeling to accomplish this? 4) In terms of the carbonate suite of numerical models, what are critical time and process scales for modeling? 5) Should the carbonate CSDMS standard strictly specify things like object-oriented structure? 6) What is the best way to supply a central CSDMS core into which user-supplied process modules can be plugged? 7) Can the CSDMS be flexible enough to incorporate different modeling approaches (e.g. rule-based versus traditional deductive), or must we decide on this at the outset? 8) For the traditional case, is it important to make specific tactical choices early, like requiring a particular discretization technique (e.g. finite element) or mesh style (e.g. deforming-grid)? 9) What kind of program architecture will be best able to take full advantage of foreseeable advances in computer design (e.g. advances in parallel computing)?
