The role of biogeophysical feedbacks and their impacts in the arctic and boreal climate system

Sammanfattning: The physical environment in the northern high latitudes including the Arctic cryosphere has undergone dramatic changes due to anthropogenic greenhouse gas warming, which since pre-industrial times has been twice or more the rate of global mean warming. Global climate models predict that this accelerated warming will continue for at least the next few decades. Meanwhile, the arctic and subarctic vegetation have been reported to be rather sensitive to such rapid warming. Biogeophysical feedbacks associated with ecosystem responses to climate change are regarded as important contributors to the amplified warming seen over the Arctic. This motivates a study to assess firstly how vegetation dynamics and ecosystem biogeochemistry will evolve under plausible future scenarios, and further how biogeophysical feedbacks associated with vegetation change will influence the climate, carbon cycle and sea ice. In addition, a regional Earth system model (ESM), as a complementary modeling alternative to relatively well-established global ESMs, can describe relevant processes and interactions in more detail and at a finer resolution in time and space. This can lead to better understanding of feedback phenomena characteristic of the Arctic climate system, as well as providing useful information on ecosystem impacts and the associated needs for adaptation they may imply. In this thesis, I present findings from studies using an individual-based dynamic vegetation model (LPJ-GUESS) and regional Earth system models (RCA-GUESS, and RCAO-GUESS) to explore the role of biogeophysical feedbacks and their impacts on the Arctic climate system. These models demonstrate good performance in reproducing the present-day dominant vegetation distribution, carbon, water and energy exchange between the land and atmosphere, the mean state of carbon pools and climate, sea ice concentration and areal extent. Thereby they provide a robust base-line for understanding and characterizing ecosystem feedbacks to the Arctic climate. Under future projections, off-line (non-feedback) simulations using LPJ-GUESS indicate that the pole-ward shift of shrubs and trees and a reduced distribution and abundance of deciduous needle-leaved trees (larch) in favor of evergreen forest is likely to cause positive feedbacks arising from reduced albedo and increased methane emission to outweigh negative feedbacks arising from increased latent heat flux and carbon sequestration. Coupled vegetation-climate simulations using RCA-GUESS projects similar changes in vegetation, which results in a further carbon sink due to biogeophysical feedbacks, and most of this carbon sink is located in the present-day arctic tundra areas. The net biogeophysical feedback is a result of the balance between two opposing feedbacks, the albedo feedback and the evapotranspiration feedback. When evolving under different levels of CO2-induced warming, biogeophysical feedbacks to near-surface warming differ both in feedback sign and magnitude depending on spatial and temporal scale, varying by season and among sub-regions of the Arctic depending on the level of CO2-induced radiative forcing. Results are discussed in terms of the resilience of ecosystems to climate change. When coupling with an ocean sea-ice model, RCAO-GUESS reveals that biogeophysical feedbacks of vegetation change could amplify variations in summer and autumn sea ice areal extent. Increased down-ward long wave radiation aided by a mean sea level pressure anomaly is found to be the main contributing factor to a strengthened sea ice decline. A further investigation is therefore needed to disentangle the complex chain of cause and effect between the Arctic vegetation and sea ice, including the spatial and temporal variability, touched upon in this initial study.