Bison body size and climate change

Abstract

The relationship between body size and temperature of mammals is poorly resolved, especially for large keystone species such as bison (Bison bison). Bison are well represented in the fossil record across North America, which provides an opportunity to relate body size to climate within a species. We measured the length of a leg bone (calcaneal tuber, DstL) in 849 specimens from 60 localities that were dated by stratigraphy and ¹⁴C decay. We estimated body mass (M) as M = (DstL/11.49)³. Average annual temperature was estimated from δ¹⁸O values in the ice cores from Greenland. Calcaneal tuber length of Bison declined over the last 40,000 years, that is, average body mass was 37% larger (910 ± 50 kg) than today (665 ± 21 kg). Average annual temperature has warmed by 6°C since the Last Glacial Maximum (~24-18 kya) and is predicted to further increase by 4°C by the end of the 21st century. If body size continues to linearly respond to global temperature, Bison body mass will likely decline by an additional 46%, to 357 ± 54 kg, with an increase of 4°C globally. The rate of mass loss is 41 ± 10 kg per°C increase in global temperature. Changes in body size of Bison may be a result of migration, disease, or human harvest but those effects are likely to be local and short-term and not likely to persist over the long time scale of the fossil record. The strong correspondence between body size of bison and air temperature is more likely the result of persistent effects on the ability to grow and the consequences of sustaining a large body mass in a warming environment. Continuing rises in global temperature will likely depress body sizes of bison, and perhaps other large grazers, without human intervention.

Keywords: Bergmann’s rule; North America; body size change; climate change; fossil; ungulate.

Figure 1

Conceptual model of the direct and indirect effects of elevated ambient temperature on body size of Bison

Figure 2

Localities (n = 60) of fossil specimens in North America that correspond with body mass estimates of bison with calibrated age. Sites are further described in Data S2

Figure 3

Standard metrics on a typical fossil calcaneum from a Bison (a) hock (b) shown in dorsal view (c) and medial view (d). Two measures for assessing body size of bison are illustrated: GL, greatest length; DstL, distal tuber length. Additional measures of the calcaneum are described by Von Den Driesch (1976) and Hill (1996)

Figure 4

Average body size of fossil bison measured as calcaneal lengths (DstL) and body mass at 60 localities in North America from 40,000 years ago (left) to today (right)

Figure 5

Sequence of Greenland mean annual temperature (°C derived from GISP2 δ¹⁸O values (Alley & Ágústsdóttir, 2005)) and relative global temperature anomaly derived from modern Greenland temperatures (μ29.9°C mean annual temperature) from 40,000 years ago (left) to today (right)

Figure 6

Relationship between estimated body mass (kg; ± SE) and the linear effect of relative global temperature (°C derived from GISP2 δ¹⁸O values) from the mixed model regression with fixed effects of temperature and the random effect of site. Regression line (y = −40.9 kg/°C ± 10) with lines for specific regressions (intercepts for B. bison (black): 520.9 ± 36.1; B. occidentalis (dark gray): 675.6 ± 36.2; B. antiquus (light gray): 737.3 ± 44.7; p < .001, n = 849, N = 53). Regression line for the small‐sized Bison clade (red line) is −63 kg/°C (± 10; z = −6.11 p < .001) with an intercept at 648 ± 26 kg