Terrestrial vegetation and its effects on climate during the latest Cretaceous

Vegetation cover is a factor that can strongly influence climate but has not been considered in modeling studies of Cretaceous climates. This study reconstructs vegetation for the latest Cretaceous (Maastrichtian) with a global database of paleoclimatic indicators, then evaluates the role of vegetation in warm Cretaceous climates through modeling experiments and comparison of model output with paleoclimatic indicators. Baseline simulation of Maastrichtian climate with the land surface coded as bare soil produces high-latitude temperatures that are too cold to explain the documented paleogeographic distribution of forest and woodland vegetation. In contrast, simulations that include forest vegetation at high latitudes show significantly warmer temperatures that are sufficient to explain the widespread geographic distribution of high-latitude forests. These warmer temperatures result from decreased albedo associated with forest vegetation and feedbacks between the land surface and adjacent oceans. The best agreement between model simulations and paleoclimatic indicators is when the simulation uses a realistic distribution of paleovegetation, though excessively cold winter temperatures still occur in the interior of North America. Positive feedbacks between high-latitude forests, the atmosphere, and ocean contributed significantly to high-latitude warming during the latest Cretaceous and imply that forest vegetation may have been an important source of high-latitude warmth during other periods of warm global climate.