Variability in projected elevation dependent warming in boreal midlatitude winter in CMIP5 climate models and its potential drivers

Abstract

resolution, their global climate sensitivity, and their elevation-dependent free air temperature response. We find that d(ΔTmin)/dz has the strongest correlation with elevationdependent increases in surface water vapor, followed by elevation-dependent decreases in surface albedo, and a weak positive correlation with the GCMs' free air temperature response.

Insights into elevation-dependent warming in the Tibetan Plateau-Himalayas from CMIP5 model simulations

“…Changes of these variables between the end and the beginning of the twenty-first century (hereafter called ∆albedo, ∆rlds, ∆rsds, and ∆huss) are computed in the same way as the minimum and maximum temperature changes and are considered as possible drivers of EDW. In addition, we include three other possible drivers, namely the normalized changes (i.e., the fractional changes relative to the averaged climatology between the mean in the years 1971-2000 and the mean in the years 2071-2100) in surface downwelling longwave and shortwave radiation (∆rlds/rlds 0 , ∆rsds/rsds 0 ) and in near-surface specific humidity (∆huss/huss 0 ) since they could be more effective in determining elevation-dependent temperature signals, as highlighted in previous studies (e.g., Rangwala et al 2013;Rangwala et al 2016). In particular, the authors of those studies pointed out that the normalized change in specific humidity is a more appropriate metric than the absolute change, because of the sensitivity of the downward longwave radiation to water vapour changes.…”

“…The low R 2 values shown by the regression models during SON indicate that the three predictors are not able to explain much of the variance of ∆tasmax in this season, and that some other relevant drivers may be at work. In this regard, it has to be noted that the predictors considered here have been chosen also considering processes and variables suggested by previous studies which have mainly focused on minimum temperatures, the winter season, and slightly different areas (see for example Liu et al 2009;Rangwala et al 2016). As such, lower R 2 for prediction of the maximum temperature changes in other seasons can be expected and are not so surprising.…”

“…Then we identify as actual EDW drivers the variables which, independently of elevation, still correlate with temperature changes at each grid point. This approach has the advantage that the spatial information inherent in the climate models is to some extent "preserved" and represents an alternative or complementary way of approaching the issue of identifying the EDW mechanisms with respect to other approaches found in the literature (e.g., Rangwala et al 2016).…”

Elevation-dependent warming in global climate model simulations at high spatial resolution

“…To our knowledge, there are only few studies that have to some extent explored the impact of the model resolution on the simulation of EDW. One is the work by Rangwala et al (2016), in which the authors analysed various CMIP5 GCMs with spatial resolution varying between 1…”

“…Some studies used simulations from a single GCM (e.g., Fyfe and Flato 1999;Liu et al 2009;Rangwala et al 2010;Yan et al 2016), while others analysed the output of several models with different characteristics, such as the latest Climate Model Intercomparison Project (CMIP5) GCM ensemble (e.g., Rangwala et al 2013Rangwala et al , 2016Palazzi et al 2017). In a recent study, Yan et al (2016) performed specific experiments to test the sensitivity of the CCSM3 GCM to changes in CO 2 concentrations and found that the changes in snow depth and cloud cover in and around the Tibetan Plateau in response to CO 2 quadruplication would lead to EDW.…”