Updated 09/15/98
The MCR Technique
Most paleoclimatic reconstructions based on proxy data from Beringia have thus far yielded only qualitative temperature estimates (i.e., a given time interval is described as either warmer or colder than another interval, without attempting to quantify air temperatures). While this qualitative approach has most frequently been applied to fossil pollen data, it has been the dominant form of fossil beetle interpretation for this region, until now. Atkinson et al., (1987) developed a quantitative method of estimating paleoclimatic parameters based on Quaternary fossil insect assemblages from Britain. The basic principle of the method lies in establishing the climatic range, or "climate envelope" of beetle species found in fossil assemblages, then estimating past climatic conditions based on the overlap of the climate envelopes of these species. This method uses presence/absence of species rather than relative abundance (which may vary considerably, depending on depositional environments).
The development of a climate envelope for a species begins with the gathering of information on its modern distribution. The climate envelope represents the boundaries of a species’ climatic tolerances for given parameters, such as mean temperature of the warmest (TMAX) and coldest (TMIN) months, or mean annual precipitation (Elias, 1998). All that is needed for the construction of an adequate climatic envelope is a determination of the climatic tolerances of a species, based on the climatic parameters within its known range. This approach has great advantages over the geographical overlap method, previously employed to reconstruct paleoenvironments from beetle assemblages. If the geographic overlap method is to succeed, the species in a given fossil assemblage must have modern ranges that overlap.
MCR studies focus on predators and scavengers (including dung beetles). These groups show the most rapid response to climate change, because their ranges are not tied to those of plant species, so they may become established in new regions regardless of vegetation response to climate. Plant-feeding groups are not considered, because these species cannot become established in new regions until their host plants arrive. Predators and scavengers have been shown to be able to shift distributions on a continental scale in a few tens or hundreds of years (Coope, 1977; Elias, 1994). The Beringian study includes 176 species in the beetle families Carabidae, Dytiscidae, Hydrophilidae, Hydraenidae, Staphylinidae, Micropeplidae, Pselaphidae, Silphidae, and Scarabaeidae.
The MCR method provides estimates of both mean summer and winter temperatures, and it standardizes the method of paleoclimatic interpretation. This is particularly important for the fossil beetle assemblages considered here, because they span a latitudinal transect of more than 1500 km. To determine the climatic tolerances of the beetles in the fossil assemblages, climate envelopes were developed for each species, based on TMAX and TMIN values of all the North American locations where the species presently occur. The climatic parameters of each of these localities were plotted on a diagram of TMAX vs. the difference between the mean July and mean January temperature (TRANGE), based on a 25 km-grid North American climate database (Bartlein et al., 1994). This database was used to pair climate parameters with the modern beetle collection sites, using the geographically nearest grid location to each collecting site. Principal components analysis of the mean monthly temperatures from stations in Eurasia showed that over 96% of the variance in temperature of the Palearctic region is described by the mean July temperature and the difference between mean July and mean January temperatures (Atkinson et al., 1987).
Elias et al. (1996) tested the MCR accuracy for North American fossil beetle assemblages, predicting modern TMAX and TMIN temperatures of 35 sites in North America, based on overlap of climate envelopes of beetle species that live at these sites. The climate envelopes were developed for species found in Quaternary fossil assemblages, all of which are extant. A linear regression of actual versus predicted TMAX values yielded an r2 value of 0.94 (the same value obtained in a modern test of European MCR estimates), suggesting that summer temperatures (as indicated by TMAX) are a very important factor determining the distribution of beetles in the temperate, boreal, and arctic regions of North America. A regression of observed versus predicted TMIN values yielded an r2 value of 0.82. This suggests that winter temperatures (as expressed by TMIN) play a significant, if somewhat lesser role, in governing beetle species’ distribution. The standard errors of the regressions were ±0.7EC for TMAX and ± 10EC for TMIN. The modern MCR test from North America yielded estimates that deviate in a systematic way from the observed TMAX and TMIN values. MCR estimates for cold climate sites were higher than expected, and MCR estimates for warm climate sites were colder than expected. These trends hold for both TMAX and TMIN values.