The world of the Paleocene and Eocene was very different from that of today. It was much warmer and more equable during most of that interval than at any other time during the Cenozoic (Wing and Greenwood, 1993). Temperatures varied little seasonally or latitudinally, mid-latitudes were largely frost-free, and there were no polar ice caps. Conditions were generally wet or humid. A paleotemperature curve reconstructed from deep-sea oxygen isotope records (Zachos et al., 2001) shows that early Paleocene temperatures continued as high as, or higher than, those at the end of the Cretaceous. Following a slight decline at the start of the late Paleocene (59-61 Ma), ocean temperatures increased steadily through the rest of the late Paleocene and the early Eocene (52-59 Ma) and peaked in the late early Eocene, about 50-52 million years ago (the Early Eocene Climatic Optimum, the warmest interval of the past 65 My). Thereafter, temperatures deteriorated more or less continuously to the end of the Eocene, when an abrupt, substantially cooler interval corresponded approximately with the Eocene/ Oligocene boundary (or more accurately, the earliest Oligocene). This interval also corresponds with the appearance of permanent ice sheets in Antarctica for the first time in the Cenozoic, and possibly Northern Hemisphere glaciation as well (e.g., Coxall et al., 2005). Antarctic glaciation probably resulted in part from changes in ocean circulation following the isolation of Antarctica. The only significant interruption in these overall trends was the Initial Eocene Thermal Maximum, the short-term global warming alluded to earlier, which further raised temperatures for about 100,000 years at the beginning of the Eocene (Sloan and Thomas, 1998). A few other episodes of elevated temperature during the early Eocene have been identified recently, but they are of lesser magnitude (e.g., Lourens et al., 2005). The relatively high temperatures of the Paleocene and Eocene have led to the characterization of this interval as a "greenhouse," compared to the "ice house" of the post-Eocene.
Deep ocean temperatures during the Paleocene and early Eocene, deduced from oxygen isotope ratios in benthic foraminifera, ranged from 8 to 12° C (Zachos et al., 2001). Continental temperatures have been estimated from the proportion of leaves with entire (untoothed) margins, which has been shown to be higher in warmer climates (Wolfe, 1979; Wilf, 1997), from multivariate analysis of leaf physiognomy (Wolfe, 1993, 1994), and from oxygen isotope composition analyzed from paleosols and fossil teeth (Fricke et al., 1998; Koch et al., 2003). Although estimates based on these different methods do not always agree, the overall pattern is consistent. For western North America, leaf-margin analysis (supported by oxygen isotope data from foramini-fera) documents an increase in mean annual temperature (MAT) from 10 to 15-18° C during the last 0.5 million years of the Cretaceous, followed by an abrupt drop to about 11° C just before the K/T boundary (Wilf et al., 2003). MAT remained at about 11° C through at least the first half of the Puercan, except for a brief, small increase immediately after the K/T boundary (probably of about 3° C, according to Wilf et al., 2003, rather than the 10° C increment reported by Wolfe, 1990). Nevertheless, these early Paleocene floras contain palms. Somewhat later in the early Paleocene (about early Torrejonian) temperatures rose again, and tropical rainforest was present in Colorado ( Johnson and Ellis, 2002).
Leaf-margin analyses indicate that MAT in western North America increased from about 13 to more than 15° C during the last 2 million years of the Paleocene, and from about 18° C near the beginning of the Eocene to more than 22° C during the late early Eocene (the Early Eocene Climatic Optimum), with a possible brief cooler interval (dipping to about 11° C) in the middle of the early Eocene (Hickey, 1977; Wing, 1998b; Wing et al., 1999). For comparison, present-day MAT in Wyoming is about 6° C, with a much greater annual range than during the Early Cenozoic. Oxygen isotope analyses indicate that MAT during the Initial Eocene Thermal Maximum was 3-7° C higher than just before and just after that interval (Fricke et al., 1998; Koch et al., 2003). Wolfe (1985) estimated that latest Paleocene MAT was as high as 22-23° C in the northern High Plains. He later estimated early Eocene temperatures to have been at least 27° C at paleolatitude 45° N, and 19° C at 70° N in North America (Wolfe, 1994). Even the lower temperature estimates for the late Paleocene and early Eocene are within the range for present-day subtropical and paratropical rainforests (Hickey, 1977). The annual temperature range was small in the early Eocene, but increased substantially as the climate cooled toward the end of the Eocene.
Based on his higher temperature estimates, Wolfe (1985) inferred that tropical rainforest covered broad areas of the continents to latitude 50° during the latest Paleocene and early Eocene (the warmest interval of the Cenozoic), with paratropical rainforest extending to latitude 60-65°. Broad-leaved evergreen forest and palms extended to 70°. Farther poleward (e.g., on Ellesmere Island) were low-diversity forests of deciduous broad-leaved trees and deciduous conifers, such as Glyptostrobus (bald cypress) and Metasequoia (dawn redwood), which apparently were tolerant of seasonal darkness. One effect of a relatively frost-free climate at high latitudes—or, at least, a climate without persistent frost—was that forests of these deciduous angiosperms and conifers spread between Europe and North America, and even across Beringia (Manchester, 1999; Tiffney, 2000). Undoubtedly this situation made it easier for mammals also to disperse along these routes.
Like vertebrates, plants suffered major extinctions across the K/T boundary (e.g., Wolfe and Upchurch, 1986). Floras from immediately above the K/T boundary in North America tend to be dominated by ferns, which are among the first plants to reappear after major environmental disruption, such as the K/T boundary bolide impact (Wing, 1998a). Thereafter, floral diversity increased slowly, and recovery of angiosperms—which were decimated by the bolide impact— took hundreds of thousands of years. Paleocene floras of western North America are typically characterized by a low diversity of deciduous broad-leaved trees, and many of the taxa had very broad ranges (Wing, 1998a; Manchester, 1999). There are more deciduous taxa than are usually present in tropical or subtropical floras. This relative abundance could be a result of terminal Cretaceous extinctions of evergreens, or it may indicate that continental interiors were somewhat cooler than has been inferred. In the late Paleocene and early Eocene, floras consisted of mixed deciduous and evergreen broad-leaved trees. During the climatic optimum of the late early Eocene, there was a higher proportion of evergreen species. Later Eocene cooling led to greater floristic zonation, which in turn may have stimulated a general dietary shift among mammals (e.g., rodents, perissodactyls) from mainly frugivory to more specialized browsing and folivory (Collinson and Hooker, 1987). Broad-leaved evergreen vegetation was mostly restricted to below latitude 50°, whereas farther poleward there was mixed conifer forest (Wolfe, 1985). Latitudinal variation in temperature was still relatively low, however, so that rainfall had a stronger influence on vegetation patterns (Wing, 1998a). Following the dramatic cool episode at the end of the Eocene, temperate deciduous and conifer forests prevailed in the mid-latitudes.
The principal constituents of North American Paleocene and Eocene floras are summarized here based on Brown (1962), Hickey (1977), Upchurch and Wolfe (1987), Wing (1998a,b, 2001), and Manchester (1999). Common elements of the Paleocene flora of the Western Interior were walnuts and hickories ( Juglandaceae), birches (Betulaceae), witch hazels (Hamamelidaceae), elms (Ulmaceae), dogwoods (Cor-naceae), ginkgos (Ginkgoaceae), oaks (Quercus), sycamores (Platanus), katsuras (Cercidiphyllum), and the genera Averrhoites (Oxalidaceae?) and Meliosma (Sabiaceae). Glyptostrobus and Metasequoia (Taxodiaceae) predominated in backswamps. Several of these, including Glyptostrobus, Metasequoia, Pla-tanus, and Palaeocarpinus (Betulaceae), were present during the Paleocene on all three northern continents (Manchester, 1999). Ground cover consisted of ferns, horsetails (Equi-setum), and other low herbaceous plants, for grasses did not dominate in open habitats until the latest Oligocene or earliest Miocene (Stromberg, 2005). Palms were essentially limited to the southern half of the continent. Early Eocene floras included many of the same taxa, but also more subtropical taxa. Poplars, ginkgos, and hazelnuts were present; relatives of laurels (Lauraceae), citrus (Rutaceae), and sumac, mango, and cashew (Anacardiaceae) helped to form the canopy. Still abundant in swamp forests were the widespread conifers Glyptostrobus and Metasequoia. Other common swamp plants during the warm early Eocene include palms, palmettos, cycads, tree ferns, ginger, magnolia, laurel, hibiscus, and the floating fern Salvinia. Many of these plants are similar to the largely tropical or subtropical flora present in the early and middle Eocene of England (Collinson and Hooker, 1987).
Wing and Tiffney (1987) proposed that the interaction between land vertebrates and angiosperms during the Late Cretaceous and Early Cenozoic had profound effects on both floras and faunas. The extinction of dinosaurs at the end of the Cretaceous altered selective pressures on the plant community by eliminating large herbivores. This change in pressure, in turn, may have led to denser vegetation, intensified competition among plants, and selection for larger seeds—floral changes that would have stimulated the radiation of arboreal frugivores, but might have stifled diversification of larger terrestrial herbivores. Although such a model is consistent with many Paleocene quarry assemblages from the northern Western Interior, it is less consistent with assemblages from the San Juan Basin, New Mexico, which are dominated by larger terrestrial herbivores. The relationship between floras and faunas is complex and not yet well understood. For example, mammalian diversity is not always correlated with floral diversity (e.g., Wilf et al., 1998), nor are major changes in the structure of mammal and plant communities necessarily closely associated (Wing and Harrington, 2001).
Was this article helpful?