• Home
• Search:
• Members only

• About
• Science
  & Data
• News
• Events
• Antarctic
  Treaty
• Antarctic
  Information
• International
  Polar Year
• Publications
• Communication
  and Education
• Contacts

You are in:  Home » Publications » Reports » Report 16 » Appendix 7

SCAR Report No 16,

Appendix 7

PALEOCLIMATE MODELING OF ANTARCTICA: WHAT IT HAS DONE ? ; WHAT IT CAN DO?
Robert Oglesby, Department of Earth and Atmospheric Sciences,
Purdue University, West Lafayette, Indiana, USA

Summary

Previous GCM modeling of past Antarctic climates helped consolidate our understanding of key processes (especially those responsible for the presence or absence of Antarctic glaciation), but was severely restricted due to model (and data) l!imitations. Over the past few years tremendous advances have taken place in our ability to simulate earth climate in general, and of more importance for this community, the climate at high southern latitudes. For the first time, polar meteorologists are saying that current GCMs can adequately (though not perfectly) simulate high latitude climates. The new generation GCMs can also be run at higher resolution, which both improves the overall simulated climate and also allows a better discrimination at the regional and local level. For paleoclimate studies this higher resolution is much more demanding of geology because locally-detailed reconstructions over a wide geographic area must be made both to provide model boundary conditions such as topography, surface/vegetation type, etc. and to validate the model results. This also means that strict geologic time controls are needed so that boundary conditions and forcings match up properly.

What Can Paleoclimate Modeling Of Antarctica Accomplish?

The major goal of any (physically-based) modeling effort is to serve as a test of our understanding of how some observed or reconstructed phenomena came to be. A secondary, but crucial, goal is to use the model to make a prediction (either a forecast or a hindcast). These goals inevitably lead to a complex iterative procedure between climate modelers and geologists who work on geologic reconstructions using data collection and interpretation. The climate modelers want the geologic data in order to provide boundary conditions for, and validate their models, while the geologists want the model results to guide and then validate their reconstructions. More specifically, paleoclimate modeling can be used to help examine and ‘test’ ideas and assumptions about geologically-relevant forcings, feedbacks, and reconstructions.

Previous paleoclimate !modeling has focused on mechanisms responsible for the presence (or absence) of Antarctic glaciation. Key results obtained were that to first order the topography and polar position of the continent are most important in explaining the presence or absence of large ice sheets over Antarctica, while SST play a more secondary role, acting to modify the extent of glaciation but not its occurrence. Continentality (that is, the size of the continent without regard to geographic, or polar, position) played a tertiary role at best, meaning for example that the attachment of Australia to Antarctica prior to about 40-50 Ma has little effect. Atmospheric CO2 concentration has little impact on the presence or size of the Antarctic ice sheet. Most studies have had difficulty obtaining or inferring ice-free conditions, such as might have occurred at times in the Cenozoic and much of the Cretaceous. These studies, while pioneering, need to be remade with the better models now available. It !should also be possible to better refine the questions that are addressed.

Major Issues

• Previous modeling efforts have effectively been reconnaissance-level because of model limitations and also because of uncertain or ambiguous geologic evidence and reconstructions. The situation has matured considerably over the past five years, because our models now run at higher spatial resolution, represent important physical processes more accurately, and in particular perform much better at high southern latitudes, and because we understand the geologic issues better. Thus we are poised, I think, to make tremendous advances over the next 10 years in our ability to model (and hence to understand) the past climatic and glacial history of Antarctica.
• Working with models and geology (data) is an iterative process. Geologists need understanding and guidance from mo!dels; modelers need data to evaluate their results (and pose the problems in the first place). There is a significant distinction between the way I as a modeler (and atmospheric scientist) might view things compared to a geologist or paleoceanographer. My primary interest is in explaining how the climate system can accommodate large fluctuations of Antarctic glaciation, that is, what combination of boundary conditions (e.g., paleotopography), forcings (e.g., solar or CO2) or climatic variability (e.g., ocean-atmosphere feedbacks) are necessary to account for a particular glacial state and how these interactions change through time.
• A full understanding of late Mesozoic and Cenozoic earth history demands a full understanding of Antarctic glaciation, which in turn requires a combined modeling and data approach. Understanding Antarctic glaciation is therefore also a key factor in understanding all of gl!obal climate, including how it has changed in the past and how it might change in the future.

Problems, Goals, and Priorities

From my perspective the key outstanding problem (and goal) is to model successfully the climate history of Antarctica for the past 100 million years. Obviously one cannot just let a sophisticated climate model go at 100 mybp and let it fly to the present. Instead key intervals must be chosen, presumably representing periods of maximum and minimum ice sheet extent, or times when rapid changes were taking place. A key objective of the ANTOSTRAT Workshop should be to help define the specific modeling priorities. Some possible (and general) examples of these priorities might include ‘hothouse’ and ‘icebox’ climates in high southern latitudes (and the related ‘snowgun’ hypothesis), times of extreme change versus more normal patterns of climate seasonality and variability,! linkages between the cryosphere and ocean and atmosphere circulation, and the relationships between topography, crustal deformations, and ice sheet volume.

Impact Within And Beyond Antarctica

Paleoclimate modeling is necessary to help explain and understand the history of Antarctic climate (especially concerning glaciation) over the past 100 Myr. Geologic proxy data, reconstructions, and modeling form a tightly coupled system. Only when we have sufficient data interpreted properly and explained from modeling results will we fully understand the past history of Antarctica.

These results also have major implications for climate on all scales, including global, because the climate of Antarctica plays a major role in modulating climates over the entire earth, primarily because of its ability to act as a refrigerator. The effects of this refrigeration may propagate directly through the atmosphere (as a consequence of interactions with the polar vortex) but are probably more important by affecting ocean circulation and sea ice.

Thus understanding the past climatic history of Antarctica is essential to a) an understanding of past climates for the entire earth and b) how changes to Antarctic climate might affect future global climatic change (e.g., will the West Antarctic ice sheet collapse).

Major Knowledge And Data Deficiencies

The major deficiencies that remain outstanding in paleoclimate modeling of Antarctica fall into two major areas: 1. model inadequacies which still exist even though the models themselves have gotten much better. 2. insufficient geologic data (or data too poorly-understood) for model boundary conditions and evaluation. Models are constantly being improved and indeed results from paleocli!mate modeling frequently serve as a guide to how the models need to be and can be improved (cf. the Paleoclimate Modeling Intercomparison Project PMIP). Improving the geologic data is a major topic of this workshop and a coordinated plan must be developed so that model and geologic data improvements are synthesized in the best manner.

Impact On Progress If Suggestions Not Heeded

If the suggestions presented here are not heeded we will lack crucial understanding of the past history of Antarctica. We may waste considerable time and resources collecting a variety of data from a variety of locations and yet have no way of synthesizing the individual results into a coherent, meaningful whole that describes and interprets the past history of Antarctica.