Regional Climate Sensitivity- and Historical-Based Projections to 2100

Reliable climate projections at the regional scale are needed in order to evaluate climate change impacts and inform policy. We develop an alternative method for projections based on the transient climate sensitivity (TCS), which relies on a linear relationship between the forced temperature response and the strongly increasing anthropogenic forcing. The TCS is evaluated at the regional scale (5 degrees by 5 degrees), and projections are made accordingly to 2100 using the high and low Representative Concentration Pathways emission scenarios. We find that there are large spatial discrepancies between the regional TCS from 5 historical data sets and 32 global climate model (GCM) historical runs and furthermore that the global mean GCM TCS is about 15% too high. Given that the GCM Representative Concentration Pathway scenario runs are mostly linear with respect to their (inadequate) TCS, we conclude that historical methods of regional projection are better suited given that they are directly calibrated on the real world (historical) climate. Plain Language Summary In this paper, we estimate the transient climate sensitivity, that is, the expected short-term increase in temperature for a doubling of carbon dioxide concentration in the atmosphere, for historical regional series of temperature. We compare our results with historical simulations made using global climate models and find that there are significant regional discrepancies between the two. We argue that historical methods can be more reliable, especially for the more policy-relevant short-term projections, given that the discrepancies of the global climate models directly bias their projections.