On the influence of land cover on early Holocene climate in northern latitudes

Abstract

This study presents a factorial experimental strategy for assessing the effects of the changing land surface on early Holocene (∼11 ka) climate in Beringia. The strategy uses a coupled land‐atmosphere single column model in a series of perturbation experiments that vary the vegetation type, lake cover, and soil specification simultaneously. Two sites and eight years of model forcing are chosen to determine the importance of differing terrestrial histories, local climate, and interannual variability. Eastern Beringia is represented by Fairbanks, Alaska, and western Beringia is represented by Elikchan Lake, Siberia. Evaluations of model response are performed using metrics important for vegetation growth, such as growing degree days and moisture availability. For both sites, large‐scale interannual variability has a greater impact on the local climate than changes in the land surface specification for temperature‐based responses, but not for moisture availability. Of the three land surface parameters tested, vegetation type is most influential. Vegetation transitions from tundra to boreal forest leads to increased precipitation, winter snow depth, and low cloud, leading to a delayed snowmelt in forested regions. This is in contrast to expectations based on albedo arguments alone. In western Beringia, a positive feedback is suggested in warming due to the presence of boreal forest, consistent with some observations that Siberian deciduous forest persisted longer than Alaskan deciduous forest in the context of large‐scale cooling of the climate. The reverse was found to be true in eastern Beringia, where deciduous forests exhibited a cooling effect due to increases in albedo. The response to soil variations was relatively small, but the inclusion of inland lakes tends in general to cool the surface, although in some years a warming is evident due to the surface energy balance shifting in favor of downwelling longwave heating over evaporative cooling.