PARENT SESSION
Oral Session # 28: Ecological Modeling II.
Presiding: R Park
Tuesday, August 5. 1:30 PM to 5:00 PM, SITCC Meeting Room 102.

Using AOGCM output to assess the performance of teleconnection patterns derived from climate proxies.

Wahl, Eugene^*,1, Ammann, Caspar¹, ¹ National Center for Atmospheric Research, Boulder, CO, USA

ABSTRACT- Long continuous runs of coupled atmosphere/ocean general circulation models (AOGCM's) are now being developed using the full suite of forcings that are particularly relevant over the last millenium: greenhouse gases, tropospheric sulfates, variations in solar output, and volcanic aerosols. Output from these models is reported as monthly averages, allowing a number of new avenues to compare the modeled values with high-resolution paleoclimatic reconstructions and to use the physically consistent model world to examine issues in paleoclimatic analyses. In pursuit of the second goal, output from a variety of long transient model runs is being studied to analyze the fidelity of proxy-based measures of the teleconnection patterns of important global climate modes. Initial work has focused on ENSO/drought teleconnections in six runs from two models at the U.S. National Center for Atmospheric Research (NCAR). In these runs, the teleconnection patterns associated with "real" (within-model) Nino3 SST variations are compared to those produced by a within-model "proxy", December-March precipitation in Ecuador/NW Peru. Both "real" and "proxy" ENSO measures are stratified into high/low 25 percentile composites, representing warm/cool events respectively. Results indicate that the models reproduce known ENSO/moisture teleconnection patterns during the period of instrumental data (c.1870-present) quite well, indicating they provide a reasonable basis for judging the performance of the simulated proxy patterns. The teleconnection patterns produced by the simulated proxy show generally good fidelity to the "real" ones at the global scale; they are generally somewhat weaker in intensity and suggest regions where proxy performance is less good in either one or both of the ENSO phases, e.g. western North America. Analysis using running-average low pass filters indicates that the simulated proxy patterns may be more sensitive to temporal smoothing than the "real" patterns, amplifying the need to establish maximal dating accuracy when developing and using proxy-derived climate reconstructions.

Key words: Teleconnection Patterns, Climate Models, ENSO, Climate Proxies