Statistical and dynamical assessment of vegetation feedbacks on climate over the boreal forest

Abstract

Vegetation feedbacks over Asiatic Russia are assessed through a combined statistical and dynamical approach in a fully coupled atmosphere–ocean–land model, FOAM-LPJ. The dynamical assessment is comprised of initial value ensemble experiments in which the forest cover fraction is initially reduced over Asiatic Russia, replaced by grass cover, and then the climatic response is determined. The statistical feedback approach, adopted from previous studies of ocean–atmosphere interactions, is applied to compute the feedback of forest cover on subsequent temperature and precipitation in the control simulation. Both methodologies indicate a year-round positive feedback on temperature and precipitation, strongest in spring and moderately substantial in summer. Reduced boreal forest cover enhances the surface albedo, leading to an extended snow season, lower air temperatures, increased atmospheric stability, and enhanced low cloud cover. Changes in the hydrological cycle include diminished transpiration and moisture recycling, supporting a reduction in precipitation. The close agreement in sign and magnitude between the statistical and dynamical feedback assessments testifies to the reliability of the statistical approach. An additional statistical analysis of monthly vegetation feedbacks over Asiatic Russia reveals a robust positive feedback on air temperature of similar quantitative strength in two coupled models, FOAM-LPJ and CAM3–CLM3, and the observational record.