A comparison of feedback processes in the Arctic during past and future warm climates

Warm paleoclimates are tempting analogs for future climates, which are expected to be dominated by the warming signal from increased greenhouse gases. One such Arctic paleoclimate occurred during the relatively warm early-mid Holocene period (6,000 to 10,000 years ago). Temperatures are believed to have been up to a few degrees warmer than present, triggered by a Milankovitch orbital forcing that produced greatly enhanced summertime insolation and a moderate increase in mean annual insolation relative to modern. Researchers investigating this time period have argued for and against the usefulness of the mid-Holocene warm period as an analog for future warming in the Arctic, but little analysis of the common physical processes has been carried out.

Here, I compare and contrast the response of the Arctic climate system to positive radiative perturbations that result from (1) orbital variations (mid-Holocene period) and (2) increased CO₂. The focus is on four processes believed to generate important climatic feedbacks in the Arctic: changes in clouds, vegetation, sea ice motion, and atmospheric energy import from lower latitudes. These processes are examined with two widely used GCMs: an atmosphere/mixed-layer ocean model (GENESIS) and an atmosphere/dynamical-ocean model (Community Climate System Model (CCSM)). The results show that sea ice dynamics act as a cooling mechanism to temper the magnitude of warming in both scenarios, due to internal ice feedbacks and to atmospheric circulation changes induced by a mobile ice pack. The flux of atmospheric moist static energy into the Arctic increases under greenhouse warming (positive feedback) but decreases in the warm paleoclimate (negative feedback). Arctic cloud cover shows virtually no change in the mid-Holocene simulations, whereas cloud fraction under higher CO₂ increases substantially in GENESIS but decreases slightly in CCSM. Vegetation changes appear to act as a positive feedback mechanism in both scenarios, due to the poleward expansion of low-albedo boreal forest replacing high-albedo snow cover.