On the eastern side of the Atlantic, molluscan faunas and (by inference) marine life in general experienced significant but less severe extinction. As noted earlier, patterns of extinction of bivalves—and geographic distributions of surviving species—point to climatic cooling as the dominant agent of extinction in the Mediterranean and North Sea basins, where only 54% of the total number of Early Pliocene species survive (Raffi et al., 1985). The incidence of extinction in both regions was reduced by the ability of species to survive in the southern parts of their ranges Most species restricted to either the North Sea or the Mediterranean during Early Pliocene time died out, whereas 60 of 64 species present in both basins survive today. Even so, the total fauna declined markedly in diversity. Today it includes only 198 polysyringian bivalve species, yet 323 Early Pliocene species are known.
In the first section of this chapter, we summarized two major regimes of late Cenozoic climatic change in and around the North Atlantic area: (1) the long-term cooling (and regional drying) that preceded Northern Hemisphere glaciation, and (2) the ice age cycles of the past 2.5 to 3 m.y. Here we provide a brief overview of some of the possible causes for these changes, with particular attention to processes affecting the North Atlantic Ocean and surrounding continents.
The most likely causes of climatic trends persisting for millions of years are tectonic changes in the configuration of the solid Earth that underpins the climate system, particularly changes in geography related to plate-tectonic processes. These include changes in plate position, sea-level, mountain elevations, and narrow "gateways" (sills and isthmus connections) that constrict ocean circulation. Most such tectonic changes are so gradual that it is difficult to demonstrate that they provide strong climatic forcing in the late Cenozoic. Two of these changes that may be especially relevant to climatic changes in and around the North Atlantic are the narrowing and final closing of the Isthmus of Panama and the relatively rapid uplift of plateaus and mountains in Asia and North America.
Final formation of the Isthmus of Panama occurred near 3 Ma, but was probably preceded by a long interval of gradually shallowing sill depth (Keigwin, 1982). Experiments with ocean general circulation models (OGCMs) indicate that closure should have led to a dramatic increase in the salinity of North Atlantic waters because the prior subsurface flow of low-salinity waters into the Atlantic would slow and then cease (Maier-Reimer et al., 1990). Modeling also simulates two other related changes: (1) increased formation of North Atlantic deep water (NADW), and (2) decreased formation of sea ice, resulting in a warming of circum-Atlantic waters at middle and high latitudes.
Geologic evidence confirms that a long-term increase in rates of NADW formation occurred over the past 10 or 15 m.y. (Woodruff and Savin, 1989), in agreement with the isthmus experiment. It is unclear, however, what ramifications increased NADW would have for global climate (via effects on the large oceanic carbon reservoir and thus potentially on CO2). On glacial-interglacial time scales, increased NADW formation correlates with increased, rather than decreased, levels of atmospheric CO2.
The simulated circum-Atlantic warming, resulting from closure of the Straits of Panama, matches neither the generally observed Northern Hemisphere trend toward cooler climates, nor the conclusion that some cooling of the North Atlantic sea surface occurred prior to glaciation (Dowsett and Poore, 1990). It does, however, match the evidence for very warm Early Pliocene ocean temperatures along the southeastern seaboard of the United States (Hazel, 1971; Stanley and Campbell, 1981; Cronin, 1988). It is also pertinent that an already-formed isthmus cannot account for the additional mid-Pleistocene cooling that led to larger glaciations over the past 1 m.y.
Geologic data summarized by Ruddiman et al. (1989) suggest major late Cenozoic uplift of the Tibetan Plateau in southern Asia and uplift across a broad region of high terrain in the American West centered on the Colorado Plateau (although the latter is contested by Molnar and England, 1990). Experiments with global circulation models (GCMs) show that uplift of rock masses on the scale of several million square kilometers can alter the basic planetary circulation of the atmosphere, by repositioning and intensifying meanders in the midlatitude surface westerlies and jet stream flow, and creating the strong monsoonal circulations of the subtropics (Kutzbach et al., 1989). Of the many large-scale changes due to uplift that are simulated by the model (Ruddiman and Kutzbach, 1989), two are particularly pertinent to the North Atlantic region: (1) strong winter cooling over east-central North America, because prevailing winds turn from westerly toward northwesterly; and (2) increased summer (and annual) evaporation over the Mediterranean and Eastern Atlantic, due to