Polar amplification dominated by local forcing and feedbacks

Abstract

The surface temperature response to greenhouse gas forcing displays a characteristic pattern of polar-amplified warming 1-5 , particularly in the Northern Hemisphere. However, the causes of this polar amplification are still debated. Some studies highlight the importance of surface-albedo feedback 6-8 , while others find larger contributions from longwave feedbacks 4,9,10 , with changes in atmospheric and oceanic heat transport also thought to play a role [11][12][13][14][15][16] . Here, we determine the causes of polar amplification using climate model simulations in which CO 2 forcing is prescribed in distinct geographical regions, with the linear sum of climate responses to regional forcings replicating the response to global forcing. The degree of polar amplification depends strongly on the location of CO 2 forcing. In particular, polar amplification is found to be dominated by forcing in the polar regions, specifically through positive local lapse-rate feedback, with ice-albedo and Planck feedbacks playing subsidiary roles. Extra-polar forcing is further shown to be conducive to polar warming, but given that it induces a largely uniform warming pattern through enhanced poleward heat transport, it contributes little to polar amplification. Therefore, understanding polar amplification requires primarily a better insight into local forcing and feedbacks rather than extra-polar processes.Polar amplification-commonly defined as the ratio of polar warming to tropical warming 4,10 -is a robust feature of climate change seen in historical observations and climate model simulations [1][2][3][4][5] . Accurate predictions of polar warming are critical given the fundamental role that polar ice plays in the climate system, terrestrial and marine ecosystems, and human society.A key challenge is identifying the roles that local (that is, polar) and remote (that is, extra-polar) processes play in polar amplification within the inherently coupled climate system. Indeed, different conclusions have been reached as to which feedbacks most contribute to polar amplification and whether poleward heat transport plays a significant role 4,5,[7][8][9][10][11][12][13][14]17,18 . These differences may, in part, be due to different analysis methods. For instance: using simulations with prescribed changes in sea-ice and sea-surface temperatures (SSTs), Screen et al. 7 argue that sea-ice loss is the main contributor to Arctic

Arctic Amplification Response to Individual Climate Drivers

“…The importance of lower‐latitude impacts on the Arctic through transport of heat and moisture has been stressed by many studies (Graversen & Wang, ; Screen et al, ; Yoshimori et al, ), and a recent analysis by Yoshimori et al () found that the effect of remote forcing is predominant in the Arctic warming. Other studies report that the AA is dominated by local forcings and feedbacks (Stuecker et al, ) and point to the importance of processes such as the water vapor and cloud feedbacks (Graversen & Wang, ) or the lapse‐rate and Planck feedbacks (Pithan & Mauritsen, ), in addition to the ice albedo feedback (Laîné et al, ; Screen & Simmonds, ). Whether or not local processes dominate, the possibility of an influence from forcing at lower latitudes remains.…”

Is Arctic Amplification Dominated by Regional Radiative Forcing and Feedbacks: Perspectives From the World‐Avoided Scenario

“…While our results suggest that AA can exist even in the presence of a negative regional RF, we note that we are using the stratosphere‐adjusted definition of RF, whereas Stuecker et al () use the alternatively defined effective radiative forcing (ERF; Hansen et al, ). With this in mind, we find a significantly greater positive energy flux convergence in our WAVD integrations relative our RCP45 integrations, which acts to offset the negative RF in the WAVD scenario and contribute substantially to AA.…”

“…This phenomenon has been well documented in observational and modeling studies (Goosse et al, ; Holland & Bitz, ; Masson‐Delmotte et al, ; Winton, ; Zhang et al, ). Linking Arctic warming contributions to different feedbacks using radiative kernels (Feldl & Roe, ; Goosse et al, ; Pithan & Mauritsen, ; Stuecker et al, ) provides a useful way to quantify the importance of various feedbacks in the Arctic relative to other regions and in different seasons (Goosse et al, ; Taylor et al, ). Using this approach, it has been shown that the ice albedo feedback makes the largest contribution to warming in summer, whereas the lapse rate and Planck feedbacks dominate in winter (Dai et al, ; Goosse et al, ; Pithan & Mauritsen, ; Screen & Simmonds, ; Stuecker et al, ; Taylor et al, ; Zhang et al, ).…”

“…Linking Arctic warming contributions to different feedbacks using radiative kernels (Feldl & Roe, 2013;Goosse et al, 2018;Pithan & Mauritsen, 2014;Stuecker et al, 2018) provides a useful way to quantify the importance of various feedbacks in the Arctic relative to other regions and in different seasons (Goosse et al, 2018;Taylor et al, 2013). Using this approach, it has been shown that the ice albedo feedback makes the largest contribution to warming in summer, whereas the lapse rate and Planck feedbacks dominate in winter (Dai et al, 2019;Goosse et al, 2018;Pithan & Mauritsen, 2014;Screen & Simmonds, 2010;Stuecker et al, 2018;Taylor et al, 2013;Zhang et al, 2018). The importance of poleward energy transport in contributing to AA has been intensely studied; however, the results are somewhat mixed with some studies arguing for a large contribution (Cai, 2005;Merlis & Henry, 2018;Zelinka & Hartmann, 2012) and others arguing for a small contribution (Hwang et al, 2011;Kay et al, 2012).…”

“…Recent Geophysical Research Letters 10.1029/2019GL082320 studies suggest that AA is largely determined by the regional component of this RF and the associated regional feedbacks (Kay et al, 2012;Shindell & Faluvegi, 2009;Stuecker et al, 2018). Notably, Stuecker et al (2018), using the feedback warming contribution approach, show that when a quadrupling of CO 2 concentrations is applied only to the polar regions in a global climate model (GCM), the RF itself accounts for one of the largest contributions to the warming in the Arctic (along with the lapse rate feedback), while contributions from atmospheric and oceanic energy transport act to decrease AA. Yet, Huang et al (2017) argue that the full meridional structure of RF plays an important role in determining the poleward energy transport, and that the structure of the RF from CO 2 in particular leads to enhanced poleward energy transport (Armour et al, 2019;Zelinka & Hartmann, 2012).…”