Paleoclimatic response of the closure of the isthmus of Panama in a coupled ocean-atmosphere model
Date
1997
Authors
Murdock, Trevor Q.
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The paleoclimatic effects of the closure of the Isthmus of Panama are investigated using a global, coupled atmospheric energy balance-ocean general circulation model. The Isthmus of Panama was a climatic, oceanographic, and biological forcing factor throughout the time of gradual closure (~ 12.5 - 2.5 Ma). A complete reconstruction of Isthmus closure is unavailable, but a synthesis of available studies shows response of different aspects of climate and biological systems to different levels of closure. Hence, observations of (possible) effects of closure are spread throughout the Late Miocene and Pliocene.
The model Atlantic of the open Isthmus run is much fresher than in the present day run. Consistent with earlier ocean-only modelling studies, prior to closrue there is an absence of deep water formation in the model North Atlantic. Hence, there is a reduction in oceanic heat transport and the model North Atlantic is significantly colder prior to Isthmus closure. Model runs with varying sill depth exhibit unexpected changes in direction of Isthmus throughflow, with consequences for poleward heat transport. The utilization of an advective wind feedback is shown to make the thermohaline circulation less sensitive to open Isthmus. Model runs with the Green-Scotland ridge closed emphasize the prime importance of a closed Isthmus of Panama in order for the model to exhibit active North Atlantic Deep Water production. The possibility of the thermohaline circulation response to Isthmus closure being related to Northern Hemisphere glaciation is examined.
An extensive model-data intercomparison is made. This comparison serves several purposes, in that it: (1) increases confidence in the sensitivity results (2) increases confidence in the use of the model for other applications (3) assists in addressing questions left open from observations alone and (4) assists in suggesting directions for potential future model studies.
In each case of model-observation intercomparison, the coarse model resolution as well as several other caveats, are kept in mind. This includes the fact that several other paleogeographic and paleoclimatic forcing factors were present during Isthmus closure that are not represented by the model. Also, the fact that examining the sensitivity to one change alone rules out the possibility of the model simulating any results that depended on a combination of factors is kept in mind. This means that the observational results are relied on quite heavily, with the model used mainly to examine the feasibility of mechanism suggested by observational studies, and to suggest what the next research steps might be. Such model-observation incomparison examples include the Late Miocene carbonate crash, tropical Pacific coding, and Northern Hemisphere glaciation.