Access to this full-text is provided by IOP Publishing.
Content available from Environmental Research Letters
This content is subject to copyright. Terms and conditions apply.
Environ. Res. Lett. 16 (2021) 093003 https://doi.org/10.1088/1748-9326/ac1c29
OPEN ACCESS
RECEIVED
6 February 2019
REVISED
2 August 2021
ACC EPT ED FOR PUB LICATI ON
10 August 2021
PUBLISHED
2 September 2021
Original content from
this work may be used
under the terms of the
Creative Commons
Attribution 4.0 licence.
Any further distribution
of this work must
maintain attribution to
the author(s) and the title
of the work, journal
citation and DOI.
TOPICAL REVIEW
Arctic amplication of climate change: a review of underlying
mechanisms
Michael Previdi1,∗, Karen L Smith1,2and Lorenzo M Polvani1,3
1Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY 10964, United States of America
2Department of Physical and Environmental Sciences, University of Toronto Scarborough, Toronto, Ontario M1C 1A4, Canada
3Department of Applied Physics and Applied Mathematics and Department of Earth and Environmental Sciences, Columbia University,
New York, NY 10027, United States of America
∗Author to whom any correspondence should be addressed.
E-mail: mprevidi@ldeo.columbia.edu
Keywords: arctic amplification, polar amplification, climate change, climate feedbacks
Abstract
Arctic amplification (AA)—referring to the enhancement of near-surface air temperature change
over the Arctic relative to lower latitudes—is a prominent feature of climate change with important
impacts on human and natural systems. In this review, we synthesize current understanding of the
underlying physical mechanisms that can give rise to AA. These mechanisms include both local
feedbacks and changes in poleward energy transport. Temperature and sea ice-related feedbacks
are especially important for AA, since they are significantly more positive over the Arctic than at
lower latitudes. Changes in energy transport by the atmosphere and ocean can also contribute to
AA. These energy transport changes are tightly coupled with local feedbacks, and thus their
respective contributions to AA should not be considered in isolation. It is here emphasized that the
feedbacks and energy transport changes that give rise to AA are sensitively dependent on the state
of the climate system itself. This implies that changes in the climate state will lead to changes in the
strength of AA, with implications for past and future climate change.
1. Introduction
Arctic amplification (AA) of climate change is
a prominent feature of the paleoclimatic record
(Hoffert and Covey 1992, Miller et al 2010), present-
day observations (Serreze et al 2009, Bekryaev et al
2010, Wang et al 2016, Richter-Menge et al 2019),
and model projections of future climate (Intergov-
ernmental Panel on Climate Change (IPCC) 2013,
Davy and Outten 2020). AA refers to the enhance-
ment of near-surface air temperature change in
the Arctic relative to lower latitudes and the global
mean, a feature that is readily apparent in obser-
vations over recent decades (figure 1(a)). Climate
models from the fifth Coupled Model Intercompar-
ison Project (CMIP5) reproduce the observed pattern
of polar-amplified warming (figure 1(b)), although
the models on average simulate less warming in
the Arctic lower troposphere and more warming
in the tropical lower troposphere (and thus weaker
AA) compared to observations. CMIP5 models also
simulate tropically-amplified warming in the upper
troposphere that is not seen in observations (see also
Po-Chedley and Fu 2012, Mitchell et al 2013,2020,
Santer et al 2013). We note, however, that the level of
agreement between modeled and observed tropical
tropospheric temperature trends may depend on the
choice of observational dataset. For example, Santer
et al (2017) found that observed trends based on one
updated satellite dataset agreed well with the trends
simulated by CMIP5 models.
Whatever the cause(s) of model-data differences
(e.g. internal variability, model bias, observational
uncertainty), these features of recent climate change
are qualitatively consistent with the expected response
to increases in atmospheric CO2(figure 1(c)). As will
be discussed, however, AA is a robust response of the
Earth system to climate forcing4, and other forcings
besides CO2very likely played a role in determining
4See section 3.1 for definitions of climate forcing and feedback.
© 2021 The Author(s). Published by IOP Publishing Ltd
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 1. (a) Latitude-pressure plot of 1979–2018 annual mean air temperature change (in K) based on linear trends (mean
change of three atmospheric reanalyses: ERA5, JRA-55 and MERRA-2). (b) As in figure 1(a), but for the mean change of 12
CMIP5 models (bcc-csm1-1, CanESM2, CCSM4, CSIRO-Mk3-6-0, GFDL-CM3, inmcm4, IPSL-CM5A-LR, MIROC5,
MPI-ESM-LR, MPI-ESM-MR, MRI-CGCM3 and NorESM1-M). For each model, 1979–2018 time series were created by
concatenating the first ensemble member (r1i1p1) of the historical simulation with the first ensemble member of the RCP8.5
simulation. (c) Latitude-pressure plot of annual mean air temperature change (in K) based on the difference between the last
30 years of the CMIP5 abrupt 4 ×CO2and pre- industrial control simulations (mean change of the 12 models listed in figure
1(b)). Note the different color scale in figure 1(c) compared to figures 1(a) and (b).
the magnitude of AA seen in recent decades (e.g.
Najafi et al 2015, Polvani et al 2020). This polar-
amplified warming has occurred in all seasons except
boreal summer, with the strongest warming in fall
and winter (figures 2(a) and (b)). Once again, this is
in qualitative agreement with the expected response
to CO2forcing (figure 2(c)).
AA has had and will have important impacts on
a range of human and natural systems, both within
and outside the Arctic. Within the Arctic, enhanced
warming and the associated melting of snow and ice
present both challenges and opportunities to local
human populations. Many communities face risks
to food and water security as a result of changes in
access to hunting areas and the distribution of tradi-
tional food sources, contamination of drinking water,
changes in traditional food preservation methods,
and increases in food contaminants (AMAP 2017,
Osborne et al 2018). Additionally, the loss of snow
and ice can create safety concerns as a result of damage
and other alteration to infrastructure (e.g. roadways,
buildings), and by leading to potentially dangerous
conditions for hunting, recreation and other activities
(e.g. Hovelsrud et al 2011, Schneider Von Deimling
et al 2021). On the other hand, climate change offers
some benefits to local communities such as increased
opportunity for marine shipping, commercial fisher-
ies, tourism, and access to resources as a result of the
lengthening open water season in the Arctic Ocean
(e.g. Melia et al 2016).
Natural systems within the Arctic are similarly
affected by the pace and magnitude of climate warm-
ing. Changes in many animals’ feeding, mating and
migration habits are altering species distribution and
abundance, introducing non-native species to the
Arctic and in some cases threatening native spe-
cies with extinction (CAFF 2013). Vegetation changes
have also been observed in recent decades, includ-
ing an overall ‘greening’ of the Arctic tundra that
reflects an increase in plant growth and productivity
2
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 2. (a) Observed 1979–2018 zonal mean surface air temperature (SAT) change (in K) for each season and the annual mean
based on linear trends. Lines indicate the mean change of the three atmospheric reanalyses listed in figure 1(a), and shading
denotes the ±1- sigma spread about this mean change (providing an estimate of observational uncertainty). (b) As in figure 2(a),
but for the 12 CMIP5 models listed in the figure 1(b) caption (historical +RCP8.5 simulations). (c) As in figure 2(b), but for the
SAT change based on the difference between the last 30 years of the abrupt 4 ×CO2and pre-industrial control simulations. Note
the different scale on the ordinate compared to figures 2(a) and (b).
(Osborne et al 2018). Arctic greening is an expec-
ted response to higher temperatures, although other
factors such as moisture availability and the occur-
rence of extreme events (e.g. wildfires, insect out-
breaks) will continue to play an important role in tun-
dra vegetation dynamics (e.g. Wu et al 2020). Higher
temperatures and the associated loss of sea ice are fur-
ther driving increases in primary production by mar-
ine algae, as well as an expansion of harmful algal
blooms that threaten human and ecosystem health
(Osborne et al 2018).
The impacts of AA are by no means limited to
the Arctic. Sea-ice loss, permafrost thaw, and other
physical changes to the Arctic environment due to
climate warming have the potential to significantly
alter Arctic carbon cycling (e.g. McGuire et al 2009,
Parmentier et al 2013, Schuur et al 2015, Natali et al
2019, Bowen et al 2020, Ito et al 2020, Bruhwiler
et al 2021). This can impact the atmospheric con-
centrations of the well-mixed greenhouse gases CO2
and methane, and thus the global climate forcing
from these gases. Increases in global sea level due
to melting of Arctic land ice are another important
impact of Arctic warming (e.g. Shepherd et al 2012,
Box et al 2018, Bamber et al 2019). This ice melt,
along with projected increases in high-latitude pre-
cipitation and ocean warming, are expected to weaken
the ocean’s meridional overturning circulation (IPCC
2013), with implications for regional climate change
and global ocean heat and carbon uptake. AA may
also influence the atmospheric jet streams (both in
the troposphere and stratosphere) in ways that lead
to increased weather and climate extremes at mid-
latitudes (e.g. Francis and Vavrus 2012, Cohen et al
2014, Barnes and Screen 2015, Coumou et al 2018).
However, the exact physical mechanisms involved,
and the relative importance of Arctic warming com-
pared to other influences (including internal variab-
ility), remain uncertain (e.g. McCusker et al 2016,
Blackport and Screen 2020).
3
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Despite this wide array of impacts, a clear con-
sensus is still lacking within the scientific com-
munity as to which physical mechanisms are the most
important in causing AA. Previous review articles
addressing this topic were published by Miller et al
(2010), Serreze and Barry (2011), and Walsh (2014).
More recently, Goosse et al (2018) reviewed some
of the key climate feedbacks in polar regions, and
the methods for quantifying them. While AA was
briefly discussed, it was not the focus of the art-
icle per se. Thus, the goal of the current article is to
provide an up-to-date, focused review of the phys-
ical mechanisms that are likely responsible for AA:
our review draws on the wealth of new research that
has happened since the publication of earlier reviews.
This will serve to synthesize our current understand-
ing, and to identify remaining knowledge gaps and
important paths for future research.
2. Methodology
Before beginning the formal review, we briefly
describe our approach for identifying articles that
were used as evidence in the review. Article selection
began by searching for ‘Arctic amplification’ and then
‘polar amplification’ using both Web of Science and
Scopus. For both databases, a document search for
these phrases within the article title, abstract, and/or
keywords was performed. Documents identified in
this manner were then screened using the following
four-step procedure:
(a) Irrelevant documents unrelated to polar amp-
lification as it pertains to climate change were
eliminated. For example, an initial search for
‘polar amplification’ yielded multiple docu-
ments on topics as diverse as electronics and bio-
logy, which are clearly irrelevant to this review
article.
(b) Documents focused on polar amplification only
in the Antarctic were eliminated. In addition to
our review article being focused on the Arctic,
there is evidence that the magnitude of polar
amplification—as well as its governing physical
processes—are different in the two hemispheres.
(c) Opinion pieces and duplicate hits were elimin-
ated.
(d) Documents not focused on the physical pro-
cesses driving AA were mostly eliminated. Two
especially prevalent examples of such documents
are: a) articles identifying the occurrence of AA
(either in observations or climate model simu-
lations) but not explicitly addressing its physical
causes; and b) articles discussing the impacts of
AA (e.g. on midlatitude weather and climate).
While the main findings from some of these art-
icles were touched upon above in the Introduc-
tion, the primary focus of our review is the causes
of AA.
Articles that survived this elimination procedure
were used as evidence in our review. These articles
were supplemented with additional relevant articles
that were identified by the authors while writing the
review.
3. Physical mechanisms
We now commence with the formal review of the
underlying physical mechanisms of AA. We begin
with some notes on terminology (section 3.1), fol-
lowed by discussion of the roles of climate for-
cing (section 3.2), climate feedbacks (section 3.3),
and changes in poleward energy transport (PET)
(section 3.4) in AA. The coupling between indi-
vidual climate feedbacks, and between climate feed-
backs and energy transport changes, is then discussed
(section 3.5). Finally, a synthesis is provided at the
end of the section (section 3.6). The important ques-
tion of the dependence of forcing, feedbacks and
energy transport on the state of the climate itself is
considered in the next section (section 4).
3.1. Terminology
Throughout the discussion, we use the term climate
forcing to refer to a change in an external driver of
climate change, such as an increase in the atmo-
spheric CO2concentration (see section 3.2 for other
examples). The term radiative forcing is used to refer
to the radiative flux change caused by a climate for-
cing. Radiative forcing without additional qualifiers
refers to the radiative flux change at the top-of-
atmosphere (TOA; Intergovernmental Panel on Cli-
mate Change (IPCC) 2013). However, we will also
consider the surface radiative forcing and atmospheric
radiative forcing, which denote, respectively, the radi-
ative flux change at the surface, and the change
in radiative flux divergence within the atmospheric
column (i.e. the difference between the TOA and sur-
face radiative forcing). In all cases, a positive (negat-
ive) forcing is one that induces warming (cooling).
In addition to the vertical level considered, radi-
ative forcing can also be distinguished by whether or
not the effects of rapid adjustments are included. A
rapid adjustment is a response to a climate forcing
that is driven directly by the forcing, independently
of any change in the global mean surface temperature
(Intergovernmental Panel on Climate Change (IPCC)
2013). When rapid adjustments are not included, the
resulting radiative forcing is referred to as the instant-
aneous radiative forcing. In this case, the radiative
flux change is due solely to the change in the cli-
mate forcing agent (e.g. the increase in atmospheric
CO2). Other types of radiative forcing include the
radiative effects of one or more rapid adjustments.
The stratosphere-adjusted radiative forcing includes
the effect of stratospheric temperature adjustment,
whereas the effective radiative forcing additionally
includes the effects of other rapid adjustments (e.g.
4
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
to tropospheric temperature, water vapor, and cloud
cover). Additional discussion of these different types
of radiative forcing can be found in Hansen et al
(1997), Shine et al (2003), Hansen et al (2005), IPCC
(2013), and Forster et al (2016), among others.
While rapid adjustments represent the
temperature-independent part of the response to
a climate forcing (see above), climate feedbacks
are the part of the response that is dependent on
global mean surface temperature change (Inter-
governmental Panel on Climate Change (IPCC)
2013). In the discussion that follows, we will con-
sider feedbacks associated with changes in surface
and tropospheric temperatures5(section 3.3.1), sur-
face albedo (section 3.3.2), clouds and water vapor
(section 3.3.3), and other properties of the Earth sys-
tem (section 3.3.4). As for climate forcing, we will
consider the impacts of climate feedbacks on the
energy budget at both the TOA and surface. Positive
(negative) feedbacks are those that reinforce (oppose)
the effects of a climate forcing. Thus, for a positive
forcing that induces warming, a positive feedback will
lead to further warming.
Finally, we will consider changes in PET by the
atmosphere (section 3.4.1) and ocean (section 3.4.2).
While it is unclear whether PET changes can be classi-
fied as feedbacks (Goosse et al 2018), they are part of
the response to climate forcing and are likely to con-
tribute to AA, as will be discussed.
3.2. Climate forcing
The single most important climate forcing dur-
ing the industrial era is increasing atmospheric
CO2(Intergovernmental Panel on Climate Change
(IPCC) 2013). AA as a response to increasing CO2
was initially predicted by Arrhenius (1896), and,
subsequently, much of the current understanding
of AA has been gleaned from CO2-forced climate
model simulations (e.g. Manabe and Wetherald
1975, Manabe and Stouffer 1980, Holland and Bitz
2003, Winton 2006, Lu and Cai 2009a, Pithan and
Mauritsen 2014, Previdi et al 2020). It seems appro-
priate then to begin our discussion of climate forcing
with CO2. AA is a robust response to CO2increases
in the studies cited above (and many others). The first
question we wish to address, therefore, is this: is AA
caused by the direct radiative forcing from CO2, or
by subsequent responses within the climate system to
this forcing?
Figure 3shows the instantaneous radiative for-
cing from doubling CO2at the TOA, the surface,
5Stratospheric temperatures may also change significantly in
response to an imposed climate forcing. For example, the strato-
sphere cools when atmospheric CO2is increased (figure 1(c)). Such
changes, however, are (largely) a direct response to the forcing (i.e.,
a rapid adjustment), rather than a feedback that is mediated by
surface temperature change. The rapid adjustment of stratospheric
temperatures is included in the stratosphere-adjusted and effective
radiative forcing, as noted above.
and within the atmospheric column. At the TOA
(figure 3(a)), CO2radiative forcing is largest at low
latitudes and decreases toward the poles6. This meri-
dional structure is due to decreases in the clima-
tological surface temperature and tropospheric ver-
tical temperature gradient with latitude (Zhang and
Huang 2014, Merlis 2015, Smith et al 2018), leading
to the strongest enhancement of the greenhouse effect
(and thus the largest radiative forcing) at low latitudes
when CO2is increased.
A qualitatively different picture emerges when
one considers the CO2radiative forcing at the surface
(figure 3(b)). In this case, the forcing is at a minimum
at low latitudes and increases toward the poles. This
is due to the overlap between CO2and water vapor
absorption bands in the 12–18 µm spectral region.
Because of this overlap, increases in the surface down-
welling longwave radiation (DLR) are limited when
CO2is increased, particularly at low latitudes where
water vapor concentrations are highest and the emis-
sion from this spectral region is already largely satur-
ated (Kiehl and Ramanathan 1982, Huang et al 2017).
The very different spatial patterns of CO2radi-
ative forcing at the TOA and at the surface result
in a strong latitudinal gradient in atmospheric radi-
ative forcing (figure 3(c)), with positive forcing at
low latitudes transitioning to negative forcing at the
poles. This pattern of atmospheric radiative forcing
has major implications for the PET, as will be dis-
cussed in section 3.4.1. To summarize, the impact of
direct CO2radiative forcing on AA depends on per-
spective. From the perspective of the TOA and atmo-
spheric column energy budgets (figures 3(a) and (c)),
CO2radiative forcing opposes AA by preferentially
heating the tropics. However, from the perspective of
the surface energy budget (figure 3(b)), CO2radiative
forcing contributes to AA by preferentially heating the
Arctic.
While the discussion thus far has focused on
CO2, other climate forcings have also been shown
to produce AA. These include changes in solar irra-
diance (Cai and Tung 2012, Stjern et al 2019), aer-
osols and aerosol precursors (Lambert et al 2013,
Yang et al 2014, Najafi et al 2015, Acosta Navarro
et al 2016, Chylek et al 2016, Sagoo and Storelvmo
2017, Conley et al 2018, Wang et al 2018, Stjern et al
2019, Kim et al 2019b), ozone-depleting substances
(Polvani et al 2020), methane (Stjern et al 2019),
and land use/land cover (van der Molen et al 2011,
Lott et al 2020). The fact that AA occurs in response
to these various forcings, with very different spa-
tial, seasonal and spectral characteristics, suggests that
it is a robust response to climate forcing that does
not depend—at least existentially—on the details of
the forcing. However, forcing details may impact the
6This strong latitudinal dependence is also present in the
stratosphere-adjusted (Hansen et al 1997) and effective radiative
forcing (Forster et al 2016) from increasing CO2.
5
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 3. Instantaneous radiative forcing (in W m−2) from doubling CO2as calculated by the rapid radiative transfer model
(RRTM). The forcing is defined as the change in the net (i.e. longwave plus shortwave) downward radiative flux, and is shown for:
(a) the top-of-atmosphere (TOA); (b) the surface (SFC); and (c) the atmospheric column (ATM, defined as TOA minus SFC).
Adapted with permission from Huang et al (2017). © 2017. American Geophysical Union. All Rights Reserved.
magnitude of AA. Modeling studies that have system-
atically varied the forcing latitude (Kang et al 2017,
Park et al 2018, Stuecker et al 2018, Semmler et al
2020) and season (Bintanja and Krikken 2016) indic-
ate that AA is likely to be strongest for high-latitude
forcings, and for forcings occurring during spring
and summer. What is clear though is that the dir-
ect effects of climate forcing alone are insufficient
to fully understand AA (see also Virgin and Smith
2019). Such an understanding requires consideration
of the feedback mechanisms operating within the cli-
mate system that act to amplify temperature changes
6
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 4. Warming contributions of individual feedback mechanisms calculated based on the difference between the last 30 years
of the CMIP5 abrupt 4 ×CO2and pre-industrial control simulations. (a) Arctic (60◦–90◦N) versus tropical (30◦S–30◦N)
warming from a TOA perspective. (b) Arctic winter (December–January–February) versus summer (June–July–August) warming.
(c) Arctic versus tropical warming from a surface perspective. For figures 4(a) and (c) feedbacks above the 1:1 line contribute to
AA, whereas feedbacks below the line oppose AA. Grey is the residual error of the decomposition. ‘Ocean’ includes the effect of
ocean transport changes and ocean heat uptake. Reprinted by permission from Springer Nature Customer Service Centre GmbH:
Springer Nature, Nature Geoscience, Arctic amplification dominated by temperature feedbacks in contemporary climate models,
Pithan and Mauritsen (2014).
at high northern latitudes. We discuss these feedbacks
next.
3.3. Climate feedbacks
3.3.1. Temperature feedbacks
We begin our discussion of climate feedbacks with
the temperature feedbacks that are associated with
changes in surface and tropospheric temperatures.
For ease of discussion, and for relevance to anthro-
pogenic climate change, we consider a positive cli-
mate forcing that induces surface and tropospheric
warming. The total temperature feedback at the TOA
in response to such a forcing can be decomposed
into contributions from vertically-uniform warming7
(Planck response) and changes in the tropospheric
lapse rate (e.g. Hansen et al 1984, Bony et al 2006,
Soden and Held 2006). The Planck response is the
most fundamental mechanism acting to stabilize the
climate, since surface and tropospheric warming will
increase the outgoing longwave radiation (OLR) and
thus oppose the effects of a positive forcing. How-
ever, because of the nonlinear dependence of OLR
on temperature (as given by the Stefan–Boltzmann
law), the Planck response is weaker (i.e. less stabiliz-
ing) over the colder Arctic relative to lower latitudes
and to the global mean. In other words, OLR increases
less over the Arctic than at lower latitudes, per degree
of local warming, and this contributes to AA (Henry
and Merlis 2019).
Globally averaged, changes in the tropospheric
lapse rate are characterized by a reduction in the
rate of temperature decrease with height, since the
upper troposphere warms more than the lower tropo-
sphere and surface. Such a change in the temperature
7The magnitude of warming is typically assumed to be equal to
that at the surface.
profile leads to a larger increase in OLR than would
occur from the Planck response alone, representing a
negative feedback on surface warming. However, the
opposite occurs over the Arctic, where the upper tro-
posphere warms less than the lower troposphere and
surface. This opposes the basic Planck response and
represents a positive feedback on surface warming.
Thus, both the Planck and lapse rate components of
the temperature feedback are expected to contribute
to AA.
Indeed, previous studies that assessed contribu-
tions to Arctic warming and AA in both models
(Langen et al 2012, Graversen et al 2014, Pithan
and Mauritsen 2014, Payne et al 2015, Cronin and
Jansen 2016, Previdi et al 2020, Zhang et al 2020) and
observations (Hwang et al 2018, Zhang et al 2018)
concluded that temperature feedbacks are import-
ant. Notably, Pithan and Mauritsen (2014) found that
these feedbacks make the single largest contribution
to AA in CMIP5 models subjected to an instantan-
eous quadrupling of atmospheric CO2(4 ×CO2).
Their analysis demonstrates a prominent role for
temperature feedbacks (and in particular the lapse
rate feedback) both in causing AA in an annual mean
sense (figure 4(a)), and also in producing the distinct
seasonality in AA, with much stronger AA in winter
relative to summer (figure 4(b); see also figure 2).
The effects of temperature feedbacks on the sur-
face energy budget come from both surface warm-
ing and atmospheric warming, each of these contrib-
uting to AA (figure 4(c); see also Laˆıné et al 2016,
Sejas and Cai 2016). The surface warming contribu-
tion arises because a larger increase in surface tem-
perature is needed over the colder Arctic (relative to
lower latitudes and to the global mean) to balance a
given increase in downward energy flux (i.e. through
enhanced surface emission of longwave radiation).
7
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
This is fundamentally the same reason why latitud-
inal variations in the Planck response at the TOA
contribute to AA. However, the effect is more pro-
nounced when considering the surface energy budget,
since the meridional temperature gradient at the sur-
face is larger than that in the troposphere (Pithan and
Mauritsen 2014). The atmospheric warming contri-
bution to AA is associated with increases in the sur-
face downwelling longwave radiation (DLR). These
DLR increases are larger in the Arctic than at lower
latitudes as a result of the much greater warming of
the Arctic near-surface atmosphere (see figure 1).
3.3.2. Surface albedo feedbacks
As sea ice and snow cover retreat in a warming cli-
mate, the surface albedo (or reflectivity) decreases
and a larger fraction of the incoming solar radiation
is absorbed by the Earth system. Such surface albedo
feedbacks therefore act to amplify surface warming,
especially at high latitudes where sea ice and snow
cover are most extensive. Changes in land ice (e.g. Hill
et al 2014, Stap et al 2014,2017,2018) and vegeta-
tion (see also section 3.3.4) are additional sources of
surface albedo change that may be important on long
(multidecadal and longer) timescales. Here, however,
we concentrate on the surface albedo (and related)
feedbacks associated with the loss of sea ice and snow
cover.
Those feedbacks have been shown to contribute to
AA in both models (e.g. Budyko 1969, Sellers 1969,
Manabe and Wetherald 1975, Manabe and Stouffer
1980, Hansen et al 1984, Holland and Bitz 2003, Hall
2004, Winton 2006, Crook et al 2011, Bintanja and
van der Linden 2013, Baidin and Meleshko 2014,
Graversen et al 2014, Pithan and Mauritsen 2014,
Song et al 2014, Lang et al 2017, Sun et al 2018, Dai
et al 2019, Duan et al 2019) and observations (e.g.
Screen and Simmonds 2010a,2010b,2012, Pistone
et al 2014, Hu et al 2017, Hwang et al 2018, Zhang
et al 2018,2019, Cai et al 2021, Yu et al 2021). For
example, in their analysis of the CMIP5 4 ×CO2
simulations, Pithan and Mauritsen (2014) found that
surface albedo feedbacks make the second largest
contribution to AA, behind temperature feedbacks
(figures 4(a) and (c)). The former were found to be
important both in causing AA in a multimodel mean
sense, and also in explaining the intermodel spread in
the magnitude of AA (see also Boeke and Taylor 2018,
Dai et al 2019, Block et al 2020, Hu et al 2020).
It is important to point out that surface albedo
feedbacks are only active when sunlight is present,
which is mainly in late spring and summer in the Arc-
tic. AA itself, however, is absent in summer and is
most pronounced in fall and winter (figure 2). How
then can surface albedo feedbacks contribute to AA?
To answer this question, it is necessary to consider the
seasonal cycle of heat storage within the Arctic Ocean
mixed layer (e.g. Chung et al 2021, Dai 2021). In a
climatological sense, this seasonal cycle is character-
ized by ocean heat uptake in the late spring and sum-
mer, followed by a loss of heat from the ocean to the
atmosphere in fall and winter. In a warming climate,
the amplitude of this seasonal cycle is expected to
increase, meaning greater ocean heat uptake in spring
and summer and greater ocean heat loss in fall and
winter (Carton et al 2015, figure 4(b)). This is a direct
result of the loss of Arctic sea ice. In summertime, the
surface albedo feedback associated with sea ice loss
causes a greater amount of incoming solar radiation
to be absorbed by the Arctic Ocean8(Pistone et al
2014). Additionally, with less sea ice present, a lar-
ger fraction of the absorbed energy goes into increas-
ing ocean temperatures rather than into melting ice
(Carton et al 2015). The additional energy accumu-
lated within the mixed layer in spring and summer is
subsequently released to the atmosphere in fall and
winter in the form of longwave radiation and latent
and sensible heat. This process is facilitated by the loss
of sea ice (area and thickness) in the latter seasons,
and thus a weakening of the sea ice insulation effect.
The enhanced ocean-to-atmosphere energy
exchange in fall and winter, which acts to warm
the near-surface atmosphere, is now recognized
to be of primary importance to AA (e.g. Screen
and Simmonds 2010a,2010b, Bintanja and van der
Linden 2013, Sejas et al 2014, Yoshimori et al 2014a,
Kim et al 2016, Franzke et al 2017, Boeke and Taylor
2018, Dai et al 2019, Chung et al 2021, Dai 2021). Its
importance to observed AA in recent decades is sug-
gested by the strong spatial correspondence between
near-surface warming and sea ice loss over the Arctic,
as shown in figure 5. This strong spatial correspond-
ence further suggests a coupling between sea ice loss
and the Arctic lapse rate feedback (e.g. Feldl et al
2020, Boeke et al 2021; see also section 3.5). The
warming and moistening of the Arctic lower atmo-
sphere increases the surface DLR, which hinders sea
ice growth and thus helps to sustain the enhanced
ocean-to-atmosphere energy flux (Burt et al 2016,
Kim et al 2016, Alexeev et al 2017, Kim and Kim
2017, Kim et al 2019a).
To summarize, there is considerable evidence
that surface albedo feedbacks—and other feedbacks
associated with the loss of sea ice and snow cover
in a warming climate—make significant contribu-
tions to AA. It is worth noting, however, that polar-
amplified warming has still been simulated in cli-
mate models even when surface albedo feedbacks
(Hall 2004, Graversen and Wang 2009, Lu and Cai
2010, Graversen et al 2014) and all sea ice- and/or
snow cover-related feedbacks (Schneider et al 1997,
1999, Alexeev 2003, Alexeev et al 2005, Duan et al
2019) have been eliminated (e.g. by fixing the surface
8It is worth noting that this increase in surface solar radiation
absorption would be significantly larger if not for the presence of
extensive cloud cover over the Arctic Ocean (Alkama et al 2020b).
8
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 5. Observed Arctic SAT and sea ice concentration (SIC) changes during October–November–December (OND)
1979–2018 based on linear trends. SAT changes (in K; shading) are the mean of the three reanalyses listed in figure 1(a), and SIC
changes (in %; white contours) are taken from the HadISST dataset version 1.1. For SIC changes, only negative values
(corresponding to sea ice loss) are plotted using a contour interval of 10%; positive SIC changes during this period (greater than
+10%) are limited to a small area between northeastern Greenland and Svalbard, and are omitted here for clarity.
albedo). This suggests that other processes are cap-
able of producing AA, even in the absence of any
changes in surface albedo and/or sea ice/snow cover.
The importance of temperature feedbacks for AA
was discussed above in section 3.3.1. However, other
processes may also be important. For example, Gra-
versen and Wang (2009) linked AA in their model
to a stronger greenhouse effect over the Arctic res-
ulting from increases in clouds and water vapor, and
Alexeev et al (2005) emphasized the importance of
increased atmospheric PET. We discuss cloud and
water vapor feedbacks in more detail in the next
section, and atmospheric PET changes will be con-
sidered in section 3.4.1.
3.3.3. Cloud and water vapor feedbacks
Most climate models project that the Arctic will
become cloudier in a warmer climate (Vavrus 2004,
Vavrus et al 2009,2011, Taylor et al 2013), mainly
due to increases in low clouds in fall and winter.
This result is supported by analyses of observed cloud
changes in recent decades (Wu and Lee 2012, Jun
9
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
et al 2016, Cao et al 2017, Philipp et al 2020). Some
modeling studies also indicate that Arctic middle and
high clouds may increase as well (Vavrus et al 2009,
2011). Increases in Arctic low clouds in fall and winter
are a direct response to sea ice loss, which acts to
destabilize the boundary layer and enhance mois-
ture availability and boundary-layer convection (Kay
and Gettelman 2009, Vavrus et al 2009, Eastman and
Warren 2010, Palm et al 2010, Vavrus et al 2011, Wu
and Lee 2012, Abe et al 2016, Jun et al 2016, Mor-
rison et al 2019, Philipp et al 2020). In contrast, sea
ice loss seems to have an insignificant impact on Arc-
tic cloud cover during summer (Kay and Gettelman
2009, Morrison et al 2019, Choi et al 2020).
Given that low cloud increases occur mainly dur-
ing fall and winter when insolation is at a minimum
over the polar cap, the associated radiative effects
occur primarily in the longwave portion of the spec-
trum. However, these effects differ considerably at the
TOA and surface (Pithan and Mauritsen 2014). At
the TOA, low cloud increases have a relatively small
impact on OLR since the temperature of these clouds
is not that different from the surface temperature.
In contrast, these cloud increases strongly enhance
the emissivity of the Arctic lower troposphere and
thus the surface DLR (Vavrus et al 2011, Taylor et al
2013, Abe et al 2016, Jun et al 2016, Cao et al 2017,
Morrison et al 2019, Philipp et al 2020). These DLR
increases, which may be further enhanced by cloud
phase changes (i.e. a larger liquid-to-ice ratio; Tan
and Storelvmo 2019, Huang et al 2021), contribute to
additional surface warming and sea ice melt (Vavrus
et al 2011, Cao et al 2017, Philipp et al 2020), and
may also help to suppress the formation of Arctic
air masses over high-latitude continents (Cronin and
Tziperman 2015, Cronin et al 2017).
Understanding the effect of cloud feedbacks on
AA requires consideration of the meridional struc-
ture of these feedbacks. In other words, it is import-
ant to also consider cloud feedbacks outside of the
Arctic. Globally averaged, the net (i.e. longwave plus
shortwave) cloud feedback is positive in most cur-
rent climate models, but with large intermodel spread
mainly due to differences in shortwave cloud feedback
(Zelinka et al 2020). This suggests that the effect of
cloud feedbacks on AA is likely to be model depend-
ent. Additionally, this effect depends on whether one
considers changes to the TOA energy budget, or to
the surface energy budget. From a TOA perspective,
cloud feedbacks oppose AA in the CMIP5 multimodel
mean (figure 4(a)), whereas they contribute to AA
from a surface perspective (figure 4(c)). (Note that in
both cases, however, the magnitude of the cloud con-
tribution is relatively small; see also Middlemas et al
(2020).) This difference in the inferred cloud effect
is partly due to the greater impact of Arctic cloud
changes on the longwave radiation at the surface, as
discussed above. Cloud feedbacks (and thus the cloud
contribution to AA) estimated from observations in
recent decades are similarly uncertain, these estimates
depending on the choice of observational datasets,
the time period considered, and the method used to
quantify cloud feedbacks (Zhang et al 2018).
Now we turn to water vapor feedbacks: these are
positive in the Arctic at both the TOA and surface,
with water vapor increases in the Arctic lower tropo-
sphere being an important contributor to increases
in surface DLR (Chen et al 2011, Ghatak and Miller
2013, Cao et al 2017). However, water vapor feed-
backs in the tropics (and in the global mean) tend to
be stronger than they are in the Arctic, both at the
TOA and surface, and thus the direct effect of these
feedbacks is to oppose AA (figures 4(a) and (c)). In
order to understand why water vapor feedbacks in the
tropics tend to be stronger, we decompose the differ-
ence in the tropical (30◦S–30◦N) and Arctic (60◦–
90◦N) mean water vapor feedbacks in the CMIP5
abrupt 4 ×CO2simulations. We express the water
vapor feedback dR as the product of two terms (see
Soden et al 2008): a radiative kernel K, describing the
sensitivity of the TOA or surface radiation to incre-
mental changes in specific humidity, and a climate
response dq, defined here as the difference in spe-
cific humidity between the last 30 years of the CMIP5
abrupt 4 ×CO2and pre-industrial control simula-
tions, i.e.
dR =∂R
∂qdq ≡Kdq.(1)
The difference in water vapor feedback ∆dR between
the tropics and Arctic can then be written as:
∆dR =dq∆K+K∆dq + ∆K∆dq (2)
where dq∆Kand K∆dq represent the contributions
to ∆dR from meridional differences in Kand dq,
respectively, and ∆K∆dq is the nonlinear term rep-
resenting the covariance between Kand dq.
Figure 6shows the various terms in equation
(2) for both the TOA (figure 6(a)) and surface
(figure 6(b)) water vapor feedbacks. We find that ∆dR
is 1.71 ±0.63 K (multimodel mean ±1σ) at the TOA,
and 0.67 ±0.24 K at the surface. This TOA value is
similar to the one found by Pithan and Mauritsen
(2014) (compare figures 4(a) and 6(a)), although our
kernel (Previdi 2010, Previdi and Liepert 2012) yields
a somewhat smaller value of ∆dR at the surface (com-
pare figures 4(c) and 6(b)).
The decomposition of ∆dR (equation (2)) pro-
duces qualitatively similar results at the TOA and sur-
face (figure 6). We find that in both cases, the stronger
water vapor feedback in the tropics compared to the
Arctic can be explained by the larger increase in spe-
cific humidity in the former (i.e. by the K∆dq term).
This larger increase in specific humidity is due to
the strong (and nonlinear) dependence of the sat-
uration vapor pressure on temperature—as given by
the Clausius–Clapeyron relationship—which leads to
10
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 6. Decomposition of the difference in tropical (30◦S–30◦N) and Arctic (60◦–90◦N) mean water vapor feedback at the
(a) TOA and (b) surface in the CMIP5 abrupt 4 ×CO2simulations (blue bars represent the multimodel mean of the 12 CMIP5
models listed in the figure 1(b) caption, with black error bars bracketing ±1σabout the mean). Labels along the abscissa are the
terms in equation (2) (see discussion in text). Water vapor feedbacks are calculated using radiative kernels from ECHAM5 (Previdi
2010, Previdi and Liepert 2012), and are expressed as warming contributions (or partial temperature changes) as in figure 4.
the largest increases in water vapor amounts where
the climatological temperatures are highest (i.e. in
the tropics). Note though that K∆dq is largely offset
by dq∆Kand ∆K∆dq (particularly the latter). This
offset reflects an overall lower radiative sensitivity to
water vapor perturbations (as measured by K) in the
tropics compared to the Arctic, owing to the greater
saturation of water vapor emission under moist trop-
ical conditions. The lower radiative sensitivity in the
tropics leads to significantly smaller values of TOA
and surface ∆dR than would result from meridional
differences in dq alone.
To summarize, stronger tropical than Arctic water
vapor feedbacks in the CMIP5 abrupt 4 ×CO2simu-
lations can be explained by larger increases in specific
humidity in the tropics. The greater tropical moisten-
ing outweighs the opposing effect of lower tropical
radiative sensitivity to water vapor changes. In addi-
tion to these implications for water vapor feedback,
larger specific humidity increases in the tropics com-
pared to the Arctic also lead to enhanced poleward
moisture transport with climate warming. As will be
discussed below (see section 3.4.1), this enhanced
moisture transport is now recognized to be import-
ant for AA.
3.3.4. Other feedbacks
To conclude this section, we briefly discuss a few
other feedbacks that have been suggested to contrib-
ute to AA. One of these involves the northward shift
of boreal forest and overall greening of the Arctic
that are expected to occur with climate warming (see
section 1). These vegetation changes enhance Arc-
tic surface warming directly by reducing the surface
albedo of high latitude land areas, and thus increas-
ing the surface absorption of solar radiation. Fur-
ther enhancement of Arctic warming may also occur
indirectly through interactions between vegetation
changes and changes in water vapor, clouds and sea
ice (O’ishi and Abe-Ouchi 2011, Jeong et al 2014,
Willeit et al 2014, Chae et al 2015, Cho et al 2018,
Park et al 2020). Increases in phytoplankton biomass
in the Arctic may produce an additional positive feed-
back on surface warming, by enhancing solar radi-
ation absorption within the ocean surface layer (Park
et al 2015).
These changes in the biosphere affect not only the
surface solar radiation, but also other surface fluxes
such as the turbulent fluxes of latent and sensible heat
(e.g. evapotranspiration changes due to vegetation
change). More generally, it is important to consider
11
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Figure 7. Changes in northward energy transports (in PW) in CMIP3 models (see legend) from 2001–2020 to 2081–2100 in the
A2 scenario: (a) atmospheric energy transport; (b) moisture (solid) and dry static energy (DSE; dashed) transport; and (c) oceanic
energy transport. Reproduced with permission from Hwang et al (2011). Copyright 2011 by the American Geophysical Union.
the possible impact of changes in the surface tur-
bulent energy fluxes on AA, in addition to the sur-
face radiative flux changes discussed in some detail
above. In a warming climate, the turbulent energy
(latent plus sensible heat) transfer from the surface to
the atmosphere is generally expected to increase (e.g.
Cai and Lu 2007, Taylor et al 2013). Increases in the
surface latent heat flux (or evaporation) tend to be
largest at low latitudes (particularly over oceans), a
consequence of the strong temperature dependence
of the saturation vapor pressure (i.e. the Clausius–
Clapeyron relationship). These evaporation increases
at low latitudes provide a strong negative feedback
on surface warming, acting to stabilize tropical sea
surface temperatures, and thus contributing to AA
(Newell 1979, Hartmann and Michelsen 1993, Cai
and Lu 2007, Yoshimori et al 2014b).
3.4. Changes in PET
3.4.1. Atmospheric transport
In section 3.2, we discussed the difference in the latit-
udinal structure of CO2radiative forcing at the TOA
and surface. Recall that, at the TOA, the forcing is
at a maximum in the tropics and decreases toward
the poles (figure 3(a)). At the surface, however, this
latitudinal gradient is reversed, with the maximum
forcing occurring at high latitudes (figure 3(b)). This
difference between the TOA and surface leads to an
even stronger latitudinal gradient in CO2radiative
forcing within the atmosphere, with the forcing act-
ing to heat the atmosphere in the tropics and generally
cool it in polar regions (figure 3(c)). Such a forcing
pattern contributes to an increase in the atmospheric
PET (e.g. Cai 2006, Huang et al 2017), which redis-
tributes the additional energy that is gained at low
latitudes. In accordance with this, coupled climate
models forced with increased CO2robustly simulate
increases in atmospheric PET in the midlatitudes of
both hemispheres (Held and Soden 2006, Hwang and
Frierson 2010, Wu et al 2011, Zelinka and Hartmann
2012, Huang and Zhang 2014, Armour et al 2019).
Over the Arctic, however, model responses are
not so robust. While some models simulate increases
in atmospheric PET into the Arctic, other models
simulate decreases (figure 7(a); see also figure 3
in Pithan and Mauritsen 2014). Decreases in atmo-
spheric PET into the Arctic have also been reported
in observations over recent decades (Sorokina and
Esau 2011, Fan et al 2015), although the magnitude
and sign of the PET change are sensitive to the data-
set (Liu et al 2020), time period (Graversen 2006,
Yang et al 2010), and Arctic sector (Mewes and Jacobi
2019) considered. From an energetic perspective,
such PET decreases imply that the net effect of local
feedbacks over the Arctic must be to heat the atmo-
spheric column, thus balancing the cooling effects
from reduced PET and CO2radiative forcing. This
further implies a coupling between local feedbacks
12
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
and PET changes, which will be discussed in
section 3.5.
The total change in atmospheric PET can be
decomposed into a change in dry static energy
(DSE) transport, and a change in moisture trans-
port. While the total change in PET into the
Arctic is not robust across model simulations—as
noted above—the changes in the DSE and mois-
ture transports individually are robust. DSE trans-
port into the Arctic decreases with climate warm-
ing, while moisture transport into the Arctic increases
(figure 7(b)). The sign of the total PET change
is primarily determined by the magnitude of the
decrease in DSE transport. In models with strong
decreases in DSE transport, the total PET into the
Arctic also decreases. However, in models with weak
decreases in DSE transport, the increase in moisture
transport (which has comparatively little intermodel
spread; see figure 7(b)) results in an increase in the
total PET.
Decreases in DSE transport into the Arctic are
a response to reductions in the meridional temper-
ature gradient due to polar-amplified warming (e.g.
Langen and Alexeev 2007, Hwang et al 2011, Skific
and Francis 2013, Graversen and Burtu 2016, Audette
et al 2021). The relationship between the DSE trans-
port and the meridional temperature gradient leads to
a negative correlation between AA and the transport
change in climate models: models with the largest AA
also tend to have the largest decrease in DSE (and
total) transport (Hwang et al 2011). We emphasize
that this correlation reflects the response of the atmo-
spheric PET to AA. However, PET changes are also
likely to be an important cause of AA, as will be dis-
cussed.
Increases in moisture transport into the Arctic
are a response to a stronger meridional gradient of
specific humidity (e.g. Graversen and Burtu 2016;
see also section 3.3.3). Several studies have emphas-
ized the importance of this enhanced moisture trans-
port for AA (e.g. Flannery 1984, Cai 2005, Cai and
Lu 2007, Langen and Alexeev 2007, Graversen and
Burtu 2016, Yoshimori et al 2017, Merlis and Henry
2018, Graversen and Langen 2019). Graversen and
Burtu (2016) demonstrated using reanalysis data that
a given increase in moisture transport across 70◦N
results in a greater Arctic surface warming than an
equivalent increase in the DSE transport. This occurs
because the increase in moisture transport acts to
strengthen the greenhouse effect over the Arctic—
both directly and by enhancing Arctic cloudiness—
leading to large increases in the surface DLR (see also
Rodgers et al 2003, Gong et al 2017, Kim and Kim
2017, Praetorius et al 2018, Hao et al 2019, Clark
et al 2021, Dunn-Sigouin et al 2021). Thus, the warm-
ing effect from enhanced moisture transport into the
Arctic with climate change can outweigh the cooling
effect from reduced DSE transport, leading to a net
Arctic warming from atmospheric PET changes even
when the total PET decreases (Graversen and Burtu
2016, Graversen and Langen 2019).
Changes in PET are the means by which heat-
ing anomalies outside the Arctic can influence the
Arctic climate. Previous studies have emphasized the
importance of heating anomalies in Northern Hemi-
sphere (NH) midlatitudes (Solomon 2006, Laliberté
and Kushner 2013, Fajber et al 2018), and in the
tropics (Rodgers et al 2003, Lee et al 2011, Yoo et al
2011, Lee 2014, Baggett and Lee 2015, Baggett et al
2016, Baggett and Lee 2017, Yim et al 2017, Baxter
et al 2019, McCrystall et al 2020, Dunn-Sigouin et al
2021). For instance, it has been suggested (Laliberté
and Kushner 2013, Fajber et al 2018) that near-surface
temperature and moisture anomalies at midlatitudes
are propagated by synoptic-scale eddies along moist
isentropes to the Arctic midtroposphere. This would
imply a direct influence of midlatitude surface warm-
ing on the lapse rate and water vapor feedbacks over
the Arctic, with implications for Arctic surface warm-
ing and AA. Alternatively, changes in atmospheric
PET into the Arctic may be induced by heating anom-
alies in the tropics. For instance, it has been suggested
that enhanced localized convection over the Maritime
Continent can trigger poleward-propagating Rossby
waves that may enhance PET into the Arctic and con-
tribute to AA (Lee et al 2011, Yoo et al 2011, Lee 2014,
Baggett and Lee 2015, Baggett et al 2016, Baggett and
Lee 2017). PET into the Arctic could also be enhanced
by a northward shift of the Inter-tropical Conver-
gence Zone, which many models project will occur
with climate warming (Yim et al 2017).
The relative importance of heating anomalies in
the tropics versus the midlatitudes has implications
for how we view changes in atmospheric PET. If trop-
ical heating anomalies are more important, a bet-
ter understanding of changes in tropical convection
would be critical for understanding atmospheric PET
changes. On the other hand, if midlatitude heating
anomalies are more important, one might wish to
place greater emphasis on better understanding the
dynamics of the eddy-driven jets and storm tracks. In
addition to these considerations, it is also worth not-
ing that tropical and midlatitude heating anomalies
have different preferred teleconnection pathways to
the Arctic, meaning that their relative importance is
likely to be regionally dependent (Ye and Jung 2019,
Hall et al 2021).
3.4.2. Oceanic transport
We now consider changes in PET by the ocean.
In response to climate warming, models generally
simulate decreases in oceanic PET at NH midlatit-
udes, and increases in oceanic PET into the Arctic
(figure 7(c); Holland and Bitz 2003, Bitz et al 2006,
Held and Soden 2006, Hwang et al 2011, Kay et al
2012, Graham and Vellinga 2013, Koenigk et al 2013,
Koenigk and Brodeau 2014, Koenigk and Brodeau
2017, Nummelin et al 2017, Singh et al 2017,2018,
13
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Docquier et al 2019, van der Linden et al 2019, Beer
et al 2020). Increased oceanic PET into the Arctic
in recent decades has also been inferred from ocean
temperature reconstructions based on marine sedi-
ments off Western Svalbard (Spielhagen et al 2011).
Modeling studies (Singh et al 2017,2018) examining
the seasonality of oceanic PET changes have found
that enhanced oceanic PET into the Arctic in the
annual mean is the result of PET increases during
boreal winter, with summertime PET into the Arc-
tic decreasing slightly. In addition to oceanic PET
increases, enhanced vertical ocean heat flux into the
Arctic Ocean mixed layer is also simulated by models
in response to climate warming (Graham and Vellinga
2013).
These changes in the oceanic energy transport
are understood to varying degrees. The simulated
decrease in oceanic PET at NH midlatitudes is mainly
the result of a weakening of the Atlantic meridi-
onal overturning circulation with climate warming
(e.g. Gregory et al 2005, Cunningham and Marsh
2010, Hwang et al 2011, Nummelin et al 2017).
At higher latitudes, however, the cause(s) of the
increase in oceanic PET into the Arctic is less cer-
tain. In particular, it is unclear to what extent this
increase is driven by circulation changes, versus the
advection of ocean heat anomalies by the back-
ground flow. Bitz et al (2006) examined simula-
tions from a coupled atmosphere-ocean model forced
with increasing CO2, and found that oceanic PET
into the Arctic was enhanced largely as a result of
strengthened inflow driven by increased convection
along the Siberian shelves. This increased convection,
in turn, was driven by enhanced sea ice production
and ocean-to-atmosphere heat flux in winter. How-
ever, other studies with coupled models have found
that the role of circulation changes is small, and that
oceanic PET increases are largely due to the advection
of warmer water into the Arctic by the background
flow (Koenigk and Brodeau 2014, Nummelin et al
2017). Nummelin et al (2017) linked this to reduced
heat loss from the subpolar ocean to the atmosphere,
which increases the heat content of the water masses
that are advected into the Arctic.
In addition to possible model dependence, the rel-
ative importance of circulation changes versus pass-
ive temperature advection for oceanic PET changes
is likely to vary with time, as patterns of ocean heat
uptake and other aspects of the climate response
evolve. For example, strengthened oceanic inflow to
the Arctic driven by increased convection along the
Siberian shelves is expected to eventually diminish in
a much warmer climate as wintertime sea ice pro-
duction declines (Bitz et al 2006). Sea ice declines are
also key to understanding the enhanced vertical ocean
heat flux into the Arctic Ocean mixed layer noted
above. Specifically, as sea ice melts, the wind stress
on the Arctic Ocean surface increases. This enhances
upper ocean mixing and entrains warmer water from
depth into the mixed layer (Graham and Vellinga
2013).
While some of the additional energy delivered
to the mixed layer through ocean transport changes
goes into warming the mixed layer, most of this addi-
tional energy goes into melting sea ice and warming
the near-surface atmosphere (Bitz et al 2006, Koenigk
and Brodeau 2014, Marshall et al 2015, Docquier et al
2019, Alkama et al 2020a). This suggests a positive
contribution from ocean transport changes to AA
(Holland and Bitz 2003, Hwang et al 2011, Mahlstein
and Knutti 2011, Graham and Vellinga 2013, Koenigk
and Brodeau 2014, Marshall et al 2014,2015,
Nummelin et al 2017, Singh et al 2017,2018, Beer et al
2020). It is worth emphasizing that this positive con-
tribution occurs during winter (figure 4(b)), the time
of year when oceanic PET into the Arctic is enhanced
(Singh et al 2017,2018), and when the ocean is a net
source of heat for the atmosphere (section 3.3.2). The
contribution of ocean transport changes to Arctic
warming and sea ice melt exemplifies the coupling
between PET changes and local feedbacks, a topic
that will be discussed further in the next section.
3.5. Coupling between mechanisms
Thus far, we have considered the contributions
of climate forcing (section 3.2), climate feedbacks
(section 3.3), and PET changes (section 3.4) to AA.
While these contributions have been discussed separ-
ately, it is now recognized that the physical mechan-
isms involved are closely coupled with one another.
In this section, we discuss the possible couplings
between mechanisms, and their implications for our
understanding of AA.
Some examples of coupling have already been
mentioned. For instance, in sections 3.3.2 and 3.3.3,
we noted that Arctic sea ice loss leads to enhanced
ocean-to-atmosphere heat flux in the fall and winter.
This is associated with a warming and moistening
of the boundary layer, and increases in low-level
clouds, which implies an impact of sea ice loss on
the lapse rate, water vapor, and cloud feedbacks
over the Arctic. The water vapor and cloud feed-
backs are further impacted by increases in poleward
moisture transport (section 3.4.1), and sea ice loss is
enhanced by increases in oceanic PET into the Arc-
tic (section 3.4.2). Changes in atmospheric PET are
also known to play an important role in modulating
the Arctic lapse rate feedback (Graversen et al 2008,
Screen et al 2012, Laliberté and Kushner 2013, Cronin
and Jansen 2016, Feldl et al 2017a, Fajber et al 2018,
Kim et al 2018, Park et al 2018, Kim and Kim 2019,
Feldl et al 2020).
The interactions between these physical mechan-
isms work in both directions. For example, increases
in Arctic lower tropospheric temperature, moisture
and cloud cover resulting from sea ice loss act to
strongly enhance the surface DLR, thus contributing
to even greater ice loss (sections 3.3.2 and 3.3.3).
14
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Table 1. Synthesis of physical mechanisms. For each of the mechanisms listed, Nindicates the total number of studies that assessed the
contribution of that mechanism to AA. The percentages of Nthat are positive (mechanism contributes to AA), negative (mechanism
opposes AA), and positive or negative (mechanism either contributes to or opposes AA) are also given.
Physical mechanism N%+%−%±
CO2forcing 29 34 48 18
Temperature feedbacks
Planck response
Lapse rate feedback
34
23
29
100
100
100
Surface albedo feedbacks
Sea ice albedo/insulation effects
Other surface albedo feedbacks
81
66
35
100
100
100
Cloud feedbacks 40 55 25 20
Water vapor feedbacks 37 32 62 6
Surface evaporation feedbacks 10 90 10
Biosphere feedbacks 8 88 12
Atmospheric PET changes 78 78 8 14
Oceanic PET changes 28 75 11 14
Additionally, local feedbacks over the Arctic that
are partly determined by PET changes modify the
energy balance and thus affect the PET (Hwang et al
2011, Feldl and Roe 2013, Merlis 2014, Huang et al
2017, Yoshimori et al 2017, Feldl et al 2017b). The
effect of a particular feedback on PET depends on
the meridional gradient of heating induced by that
feedback. Feedbacks that preferentially heat the Arc-
tic relative to lower latitudes—such as the surface
albedo feedback—act to decrease the PET into the
Arctic. In contrast, feedbacks that preferentially heat
lower latitudes—such as the water vapor feedback
(see section 3.3.3)—act to increase the PET. This is
the same line of reasoning used to explain the increase
in atmospheric PET induced by CO2radiative forcing
(see section 3.4.1). In some coupled models, this dir-
ect CO2effect is overwhelmed by an opposite-signed
effect from climate feedbacks, resulting in a decrease
in atmospheric PET into the Arctic (figure 7(a)).
The coupling between physical mechanisms that
is described here has important implications. For
instance, modeling studies suggest that nonlinear
interactions between local feedbacks will act to warm
the Arctic more than lower latitudes, thus contribut-
ing to AA (Feldl and Roe 2013, Södergren et al 2018).
Such nonlinear interactions—though not funda-
mental to the occurrence of AA—have been ignored
in traditional linear climate feedback analysis9. Addi-
tionally, the perceived contribution of a particular
physical mechanism to AA can change—at times
qualitatively—when its effects on other mechanisms
are taken into account. For example, when consider-
ing only direct effects on the TOA radiation, both CO2
forcing and water vapor feedback oppose AA, since
they preferentially heat the tropics (figure 4(a)). How-
ever, when their effects on the atmospheric PET are
taken into account (see above), they both contribute
9The relevance of feedback nonlinearities for global climate sens-
itivity has been discussed, for example, by Feldl and Roe (2013),
Knutti and Rugenstein (2015) and Sherwood et al (2020).
to AA (Russotto and Biasutti 2020). Such coupling
between physical mechanisms—and whether or not
it is accounted for—can partly explain why previous
studies have sometimes reached different (and even
contradictory) conclusions about the main causes of
AA.
3.6. Synthesis of physical mechanisms
To conclude section 3, we present a brief synthesis
of the studies that we have reviewed in table 1. For
each of the physical mechanisms discussed above,
we tabulate the total number of studies Nthat
assessed the contribution of that mechanism to AA.
We then calculate the percentages of Nthat are posit-
ive (mechanism contributes to AA), negative (mech-
anism opposes AA), and positive or negative (mech-
anism either contributes to or opposes AA10). The
physical mechanisms considered in this synthesis
include the following: CO2forcing, temperature feed-
backs (including Planck response and lapse rate feed-
back), surface albedo feedbacks (including sea ice
albedo/insulation effects and other (snow cover, land
ice, and vegetation) surface albedo feedbacks), cloud
feedbacks, water vapor feedbacks, surface evapor-
ation feedbacks, biosphere feedbacks, atmospheric
PET changes, and oceanic PET changes.
Table 1suggests that these physical mech-
anisms can be grouped into three categories.
The first category includes mechanisms with rel-
atively large Nand a high level of agreement
between studies on the sign of the mechan-
isms’ contribution to AA. In this category, we
include temperature feedbacks (section 3.3.1), sur-
face albedo feedbacks (section 3.3.2), atmospheric
PET changes (section 3.4.1), and oceanic PET
changes (section 3.4.2). The studies we reviewed
largely agree (for atmospheric/oceanic PET changes)
10 The contribution of a particular mechanism to AA may be
assessed as either positive or negative in multimodel studies, for
instance (i.e., if models disagree on the sign of the contribution).
15
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
or unanimously agree (for temperature/surface
albedo feedbacks) that these mechanisms contrib-
ute positively to AA (often making a leading-order
contribution).
The second category includes mechanisms with
relatively large N, but with a lower level of agreement
between studies on the sign of the mechanisms’ con-
tribution to AA. This category includes CO2forcing
(section 3.2), and cloud and water vapor feedbacks
(section 3.3.3). Nearly half (48%) of the studies sur-
veyed found that direct CO2forcing opposes AA. The
majority (62%) of the studies surveyed found that
water vapor feedback also opposes AA, while a slightly
smaller majority (55%) found that cloud feedback
contributes to AA. For each of these mechanisms,
however, smaller but significant percentages of the
studies surveyed inferred an opposite-signed contri-
bution (e.g. about a third of the studies found a pos-
itive contribution to AA from direct CO2forcing and
water vapor feedback).
The third and final category includes mechanisms
with small Nand a high level of agreement between
studies. This category includes surface evaporation
and biosphere feedbacks (section 3.3.4). There is
strong (∼90%) agreement that these mechanisms
contribute positively to AA, but this is based on a rel-
atively small number of studies (N⩽10).
Based on these results, we can conclude that
the physical mechanisms included in the first cat-
egory (temperature/surface albedo feedbacks and
atmospheric/oceanic PET changes) likely contribute
positively to AA. For the mechanisms included in the
second category (CO2forcing and cloud/water vapor
feedbacks), the sign of the contribution to AA is more
uncertain given the lower level of agreement between
studies. Different studies may infer opposite-signed
contributions from these mechanisms depending, for
example, on the choice of climate model, whether
the TOA or surface energy budget is considered, and
whether effects on the energy transport are accounted
for (see discussion above). Finally, the physical mech-
anisms included in the third category (surface evapor-
ation and biosphere feedbacks) may contribute pos-
itively to AA, but given their small value of N(see
table 1), it is important that this positive contribu-
tion be confirmed through additional studies. This is
particularly true for biosphere feedbacks, which have
been studied far less than physical feedbacks (at least
in terms of their potential contribution to AA).
4. Climate state dependence
From the preceding discussion, it should already be
clear that several changes occurring within the Earth
system in response to climate forcing—including
both local feedbacks and PET changes—are import-
ant in producing AA. We argue here that some of the
most important changes are a direct consequence of
the climate state at the time the forcing is imposed.
The first aspect of the climate state that we
will consider is the meridional temperature gradi-
ent. The existence of a strong meridional temperat-
ure gradient in the present climate is the reason why
the Planck response contributes to polar-amplified
warming (section 3.3.1). In other words, colder
polar regions must warm more than lower latitude
regions in order to balance a given magnitude (pos-
itive) radiative forcing through enhanced OLR. Addi-
tionally, the meridional temperature gradient partly
explains why a given increase in atmospheric CO2
concentration produces the largest radiative forcing
at low latitudes (section 3.2). The associated meridi-
onal gradient in heating drives an increase in atmo-
spheric PET (section 3.4.1), which contributes to
AA. Even with spatially-uniform heating, however, an
increase in atmospheric PET is still expected as a res-
ult of enhanced moisture transport (e.g. Langen and
Alexeev 2007, Merlis and Henry 2018). The increase
in moisture transport is a response to a stronger meri-
dional gradient in specific humidity (section 3.4.1).
This is another consequence of the meridional tem-
perature gradient—as well as the nonlinear depend-
ence of the saturation vapor pressure on temper-
ature (i.e. the Clausius-Clapeyron relationship)—
which conspire to produce the largest increases in spe-
cific humidity at low latitudes as the climate warms
(section 3.3.3).
In addition to the meridional temperature gradi-
ent, the climatological vertical temperature struc-
ture of the atmosphere is also important for AA.
The climatological near-surface inversion over the
Arctic strongly suppresses vertical mixing and thus
confines surface heating anomalies to the lowermost
atmosphere, leading to a positive lapse rate feedback
(figures 4(a) and (b); see also Lauer et al 2020, Previdi
et al 2020). In contrast, the more unstable tropical
atmosphere permits deep vertical mixing by convec-
tion which efficiently transports heat to the upper tro-
posphere, leading to a negative lapse rate feedback
(figure 4(a)). At these upper tropospheric levels where
the atmosphere above is optically thin, the heat is
readily lost to space as OLR. This again is in stark
contrast to what occurs in the Arctic, where heating
anomalies trapped near the surface hardly contrib-
ute to the TOA OLR (Bintanja et al 2011). Instead,
these heating anomalies primarily enhance the sur-
face DLR, thus amplifying the Arctic surface warm-
ing. In summary, differences in the climatological
static stability between the polar and tropical atmo-
spheres lead to marked differences in the efficiency
of vertical mixing, and, therefore, in the efficiency
of longwave energy loss to space. In this regard, AA
can be viewed as a manifestation of the relative ineffi-
ciency of the Arctic climate system in ridding itself of
excess heat.
Lastly, we consider the importance of the cli-
matological Arctic sea ice. Holland and Bitz (2003)
examined simulations from several CMIP2 climate
16
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
models, and found that climatological sea ice condi-
tions in the models’ control climate were related to
the simulated Arctic warming and AA in response to
increasing CO2. In particular, the latitude of max-
imum warming was found to be correlated inversely
and significantly with the climatological sea ice
extent. Additionally, models with relatively thin cli-
matological ice cover tended to have stronger AA. The
robustness of these relationships, however, remains
unclear. In a more recent study of CMIP5 models that
included many more models, van der Linden et al
(2014) found that the climatological sea ice amount
could not explain the intermodel spread in Arctic
mean warming and AA in CO2forced simulations.
However, within individual models, a positive correl-
ation was found between the spatial patterns of the
climatological ice amount and the simulated winter
warming in response to increased CO2. Regardless of
the factors explaining intermodel spread, it should
be clear from the discussion above (see section 3.3.2)
that sea ice-related feedbacks contribute substantially
to AA, and thus AA is expected to be stronger when
sea ice is present. The lack of a relationship between
the climatological sea ice and AA across the CMIP5
models (van der Linden et al 2014) may simply imply
that intermodel differences in the former are subtle
enough to be overwhelmed by other factors.
To the extent that feedbacks and PET changes
depend on the state of the climate, changes in the
feedbacks and PET over time should be expected as
the climate state evolves with warming. For example,
an increase in atmospheric PET driven by CO2radi-
ative forcing11 is expected to diminish over time in
response to polar-amplified warming, and the associ-
ated reduction in the meridional temperature gradi-
ent and DSE transport (section 3.4.1). Addition-
ally, feedbacks associated with the Arctic near-surface
inversion (Bintanja et al 2012) and sea ice cover
(Bintanja and van der Linden 2013, Yim et al 2016,
Andry et al 2017, Haine and Martin 2017, Dai et al
2019) will change as the inversion weakens and sea ice
melts in a warming climate. These time-varying feed-
backs and PET changes are important for at least two
reasons. First, they imply that the relative importance
of different physical mechanisms for AA will depend
on the time period considered (see also Alexeev et al
2005, Previdi et al 2020). Second, they imply that AA
itself will also vary with time. These variations may
be significant on long timescales when changes in the
climate state are substantial. For instance, AA largely
disappears in future climate simulations over the next
couple of centuries as the Arctic Ocean becomes
essentially ice free throughout the year (Bintanja and
van der Linden 2013, Dai et al 2019).
11 More generally, increased PET will occur in response to any pos-
itive radiative forcing, provided that the forcing magnitude does
not increase strongly with latitude.
In summary, the climate state largely determines
the feedbacks and PET changes that give rise to AA.
The strength of AA thus depends on the climate state,
and should not be taken as a constant. In accordance
with this, models indicate that AA will weaken in the
future as the climate state changes dramatically with
warming.
5. Summary and outlook
In this review, we have discussed the underlying phys-
ical mechanisms that lead to AA of climate change.
Identifying and understanding these mechanisms is
important because of the wide-ranging impacts of
AA, both within and outside the Arctic. It is also of
interest from a purely scientific standpoint, as a clas-
sical problem in climate dynamics. Although uncer-
tainties linger, our review leads us to draw the follow-
ing conclusions:
(a) AA is a robust response to climate forcing. While
much of the discussion above has focused
on CO2forcing—given its preeminent role in
anthropogenic climate change—AA has also
been shown to occur in response to changes in
solar irradiance, aerosols and aerosol precursors,
ozone-depleting substances, methane, and land
use/land cover (section 3.2). The fact that AA
occurs in response to these very different climate
forcings suggests that the characteristics of the
forcing (e.g. spatial pattern, longwave or short-
wave) cannot be of primary importance. Instead,
AA must fundamentally owe its existence to feed-
backs and other mechanisms that—at least to
first order—are insensitive to the precise details
of the forcing.
(b) Temperature- and sea ice-related feedbacks are
especially important for AA. While other feed-
backs may contribute as well (section 3.3), it
is now understood that the feedbacks associ-
ated with tropospheric and surface warming,
and with Arctic sea ice loss, play a leading-order
role. Both the Planck and lapse rate compon-
ents of the temperature feedback contribute to
AA (figure 4(a)). The lapse rate feedback is espe-
cially important, however, since it changes sign
with latitude, being positive in the Arctic and
negative in the tropics (and in the global mean).
Sea ice loss plays a paramount role in establish-
ing the spatial pattern (figure 5) and seasonality
(figure 2) of Arctic warming and AA. Differences
in the magnitude of ice loss over time (and across
climate models) have been linked to correspond-
ing differences in the strength of AA (e.g. Dai et al
2019). While traditionally cast in terms of surface
albedo feedback, it is now recognized that sea ice
insulation effects—and the associated regulation
of ocean-atmosphere heat exchange—are critical
17
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
for understanding these impacts of ice loss on
AA.
(c) Increases in PET contribute to AA. In response
to a positive radiative forcing, the atmospheric
PET increases as a result of enhanced mois-
ture transport. The poleward dry static energy
(DSE) transport may also increase initially if the
radiative forcing is largest at low latitudes, as
is the case for CO2forcing (section 3.4.1). As
AA develops, however, the meridional temper-
ature gradient weakens and the poleward DSE
transport begins to decrease. In some cases, this
decrease can become large enough to outweigh
the increase in moisture transport, leading to
a reduction in the total atmospheric PET into
the Arctic (figures 7(a) and (b)). This under-
scores the need to examine the co-evolution of
the atmospheric PET and AA over time in order
to fully understand their relationship. Forcing-
driven PET increases are likely to be important in
initially establishing AA (e.g. Alexeev et al 2005,
Langen and Alexeev 2007), yet these increases
will invariably be muted once AA develops and
the DSE transport begins to weaken. Increases
in oceanic PET into the Arctic with climate
warming tend to compensate for any decreases
in atmospheric PET (compare figures 7(c) and
(a)). Much of the additional energy accumulated
within the Arctic Ocean mixed layer goes into
warming the near-surface atmosphere, thus con-
tributing to AA (section 3.4.2).
(d) Local feedbacks and PET changes are tightly
coupled. This coupling works in both directions:
local feedbacks (both within and outside the
Arctic) are partly determined by PET changes,
and these feedbacks—once established—affect
the meridional gradient in heating and thus the
PET (section 3.5). The coupling between local
feedbacks and PET changes has implications for
how one views the role of individual feedbacks
in AA. The perceived role of a particular feed-
back (i.e. whether that feedback contributes to
or opposes AA) can be different when only the
feedback’s direct effect on the meridional heat-
ing gradient is considered (e.g. see figure 4),
versus when effects on the PET are also taken into
account (such as in ‘feedback-locking’ experi-
ments with climate models; e.g. Vavrus 2004,
Graversen and Wang 2009, Graversen et al 2014,
Merlis 2014, Ceppi and Hartmann 2016, Dai et al
2019, Dekker et al 2019, Li and Newman 2020,
Middlemas et al 2020, Russotto and Biasutti
2020).
(e) The strength of AA depends on the climate state.
The existence of a strong meridional temperat-
ure gradient, a prevalent Arctic inversion, and
extensive sea ice cover in the present climate
helps to shape the feedbacks and PET changes
that give rise to AA (section 4). As these features
of the climate state change (e.g. with climate
warming), the feedbacks and PET—and thus
AA itself—will also change. Model simulations
indicate that AA becomes much weaker (and
even disappears) when the meridional temper-
ature gradient (Lutsko and Popp 2018, Merlis
and Henry 2018) and Arctic inversion (Bintanja
et al 2011,2012) are significantly weakened, and
when Arctic sea ice cover is dramatically reduced
(Bintanja and van der Linden 2013, Dai et al
2019, Chung et al 2021). These dependencies
are relevant for future anthropogenic climate
change, since the climate state is expected to
evolve in this manner with warming. Depend-
ence of AA strength on the climate state may
also be relevant for past climate changes during
periods when the climate state was very different
from today (e.g. Sluijs et al 2006).
In closing, we briefly comment on potential paths
forward that will hopefully deepen our understand-
ing of AA. One such path involves the increased use
of coordinated modeling experiments to help address
lingering uncertainties. An example of such an effort
that is currently ongoing is the Polar Amplification
Model Intercomparison Project (PAMIP) contribu-
tion to phase 6 of the Coupled Model Intercompar-
ison Project (CMIP6; Smith et al 2019). A second path
would involve the continued exploration of alternat-
ive frameworks for studying the physical mechanisms
that contribute to AA. Traditional linear climate feed-
back analysis that focuses on the top-of-atmosphere
(TOA) energy budget alone is now recognized to have
limitations when applied to the study of AA. Specific-
ally, it is clear that the feedbacks and PET changes
contributing to AA are not independent from one
another (as is assumed in linear feedback analysis).
Additionally, changes in the TOA energy budget may
be a poor predictor of surface temperature changes
over the Arctic, where the TOA and surface are largely
decoupled as a result of the strong static stability
of the polar atmosphere (section 4). Future stud-
ies of AA should continue to investigate the coup-
ling between physical mechanisms, and examine the
energy budget at the surface (e.g. figure 4(c); see also
Alexeev 2003, Lesins et al 2012) and throughout the
atmospheric column (e.g. Lu and Cai 2009b, Taylor
et al 2013, Henry et al 2021), in addition to the TOA
energy budget. Such examination of the vertically-
resolved energy budget will further our understand-
ing of the many processes that help to shape the Arc-
tic lapse rate feedback (e.g. Cronin and Jansen 2016,
Feldl et al 2020, Henry and Merlis 2020, Boeke et al
2021). Finally, a third path forward would involve the
continued exploration of the climate state depend-
ence of AA. This can be accomplished both through
modeling experiments (including idealized experi-
ments), and through studies using paleoclimatic data.
Comparison of polar amplification in the Arctic and
18
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Antarctic (e.g. Singh et al 2018, Casagrande et al 2020,
Hahn et al 2020) may provide additional insight into
how this phenomenon changes with different bound-
ary conditions. It is our hope that the research dir-
ections outlined here will help to drive progress in
our understanding of AA and its underlying physical
mechanisms.
Data availability statement
The data that support the findings of this study are
available upon reasonable request from the authors.
Acknowledgments
We acknowledge the World Climate Research Pro-
gramme’s Working Group on Coupled Modelling,
which is responsible for CMIP, and we thank the
climate modeling groups for producing and making
available their model output. We are grateful to two
anonymous reviewers and an editorial board mem-
ber for their helpful comments on an earlier version of
the manuscript. This work was supported by a grant
from the National Science Foundation to Columbia
University (Award # 1603350).
ORCID iDs
Karen L Smith https://orcid.org/0000-0002-4652-
6310
Lorenzo M Polvani https://orcid.org/0000-0003-
4775-8110
References
Abe M, Nozawa T, Ogura T and Takata K 2016 Effect of retreating
sea ice on Arctic cloud cover in simulated recent global
warming Atmos. Chem. Phys. 16 14343–56
Acosta Navarro J C, Varma V, Riipinen I, Seland Ø, Kirkevåg A,
Struthers H, Iversen T, Hansson H-C and Ekman A M L
2016 Amplification of Arctic warming by past air pollution
reductions in Europe Nat. Geosci. 9277–81
Alexeev V A 2003 Sensitivity to CO2doubling of an atmospheric
GCM coupled to an oceanic mixed layer: a linear analysis
Clim. Dyn. 20 775–87
Alexeev V A, Langen P L and Bates J R 2005 Polar amplification of
surface warming on an aquaplanet in ‘ghost forcing’
experiments without sea ice feedbacks Clim. Dyn. 24 655–66
Alexeev V A, Walsh J E, Ivanov V V, Semenov V A and
Smirnov A V 2017 Warming in the Nordic Seas, North
Atlantic storms and thinning Arctic sea ice Environ. Res.
Lett. 12 084011
Alkama R, Koffi E N, Vavrus S J, Diehl T, Francis J A, Stroeve J,
Forzieri G, Vihma T and Cescatti A 2020a Wind amplifies
the polar sea ice retreat Environ. Res. Lett. 15 124022
Alkama R, Taylor P C, Garcia-San Martin L, Douville H,
Duveiller G, Forzieri G, Swingedouw D and Cescatti A
2020b Clouds damp the radiative impacts of polar sea ice
loss Cryosphere 14 2673–86
Andry O, Bintanja R and Hazeleger W 2017 Time-dependent
variations in the Arctic’s surface albedo feedback and the
link to seasonality in sea ice J. Clim. 30 393–410
Arctic Monitoring and Assessment Programme (AMAP) 2017
Snow, Water, Ice and Permafrost in the Arctic (SWIPA) 2017
(Oslo, Norway: AMAP) pp xiv +269
Armour K C, Siler N, Donohoe A and Roe G H 2019 Meridional
atmospheric heat transport constrained by energetics and
mediated by large-scale diffusion J. Clim. 32 3655–80
Arrhenius S 1896 On the influence of carbonic acid in the air
upon the temperature of the ground London, Edinburgh
Dublin Phil. Mag. J. Sci. 541 237–76
Audette A, Fajber R A, Kushner P J, Wu Y, Peings Y,
Magnusdottir G, Eade R, Sigmond M and Sun L 2021
Opposite responses of the dry and moist eddy heat transport
into the Arctic in the PAMIP experiments Geophys. Res. Lett.
48 e2020GL089990
Baggett C and Lee S 2015 Arctic warming induced by tropically
forced tapping of available potential energy and the role of
the planetary-scale waves J. Atmos. Sci. 72 1562–8
Baggett C and Lee S 2017 An identification of the mechanisms
that lead to Arctic warming during planetary-scale and
synoptic-scale wave life cycles J. Atmos. Sci. 74 1859–77
Baggett C, Lee S and Feldstein S 2016 An investigation of the
presence of atmospheric rivers over the North Pacific during
planetary-scale wave life cycles and their role in Arctic
warming J. Atmos. Sci. 73 4329–47
Baidin A V and Meleshko V P 2014 Response of the atmosphere at
high and middle latitudes to the reduction of sea ice area
and the rise of sea surface temperature Russ. Meteorol.
Hydrol. 39 361–70
Bamber J L, Oppenheimer M, Kopp R E, Aspinall W P and
Cooke R M 2019 Ice sheet contributions to future sea-level
rise from structured expert judgment Proc. Natl Acad. Sci.
USA 116 11195–200
Barnes E A and Screen J 2015 The impact of Arctic warming on
the midlatitude jetstream: can it? Has it? Will it? WIREs
Clim. Change 6277–86
Baxter I et al 2019 How tropical Pacific surface cooling
contributed to accelerated sea ice melt from 2007 to 2012 as
ice is thinned by anthropogenic forcing J. Clim.
32 8583–602
Beer E, Eisenman I and Wagner T J W 2020 Polar amplification
due to enhanced heat flux across the halocline Geophys. Res.
Lett. 47 e2019GL086706
Bekryaev R V, Polyakov I V and Alexeev V A 2010 Role of polar
amplification in long-term surface air temperature
variations and modern Arctic warming J. Clim. 23 3888–906
Bintanja R, Graversen R G and Hazeleger W 2011 Arctic winter
warming amplified by the thermal inversion and consequent
low infrared cooling to space Nat. Geosci. 4758–61
Bintanja R and Krikken F 2016 Magnitude and pattern of Arctic
warming governed by the seasonality of radiative forcing Sci.
Rep. 638287
Bintanja R and van der Linden E C 2013 The changing seasonal
climate in the Arctic Sci. Rep. 31556
Bintanja R, van der Linden E C and Hazeleger W 2012 Boundary
layer stability and Arctic climate change: a feedback study
using EC-Earth Clim. Dyn. 39 2659–73
Bitz C M, Gent P R, Woodgate R A, Holland M M and Lindsay R
2006 The influence of sea ice on ocean heat uptake in
response to increasing CO2J. Clim. 19 2437–50
Blackport R and Screen J A 2020 Insignificant effect of Arctic
amplification on the amplitude of midlatitude atmospheric
waves Sci. Adv. 6eaay2880
Block K, Schneider F A, Mülmenstädt J, Salzmann M and Quaas J
2020 Climate models disagree on the sign of total radiative
feedback in the Arctic Tellus A72 1–14
Boeke R C and Taylor P C 2018 Seasonal energy exchange in sea
ice retreat regions contributes to differences in projected
Arctic warming Nat. Commun. 95017
Boeke R C, Taylor P C and Sejas S A 2021 On the nature of the
Arctic’s positive lapse-rate feedback Geophys. Res. Lett.
48 e2020GL091109
Bony S et al 2006 How well do we understand and evaluate
climate change feedback processes? J. Clim. 19 3445–82
19
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Bowen J C, Ward C P, Kling G W and Cory R M 2020 Arctic
amplification of global warming strengthened by sunlight
oxidation of permafrost carbon to CO2Geophys. Res. Lett.
47 e2020GL087085
Box J E, Colgan W T, Wouters B, Burgess D O, O’Neel S,
Thomson L I and Mernild S H 2018 Global sea-level
contribution from Arctic land ice: 1971–2017 Environ. Res.
Lett. 13 125012
Bruhwiler L, Parmentier F-J W, Crill P, Leonard M and Palmer P I
2021 The Arctic carbon cycle and its response to changing
climate Curr. Clim. Change Rep. 714–34
Budyko M I 1969 The effect of solar radiation variations on the
climate of the Earth Tellus 21 611–9
Burt M A, Randall D A and Branson M D 2016 Dark warming
J. Clim. 29 705–19
Cai M 2005 Dynamical amplification of polar warming Geophys.
Res. Lett. 32 L22710
Cai M 2006 Dynamical greenhouse-plus feedback and polar
warming amplification. Part I: a dry radiative-transportive
climate model Clim. Dyn. 26 661–75
Cai M and Lu J 2007 Dynamical greenhouse-plus feedback and
polar warming amplification. Part II: meridional and
vertical asymmetries of the global warming Clim. Dyn.
29 375–91
Cai M and Tung K-K 2012 Robustness of dynamical feedbacks
from radiative forcing: 2% solar versus 2×CO2experiments
in an idealized GCM J. Atmos. Sci. 69 2256–71
Cai Q, Wang J, Beletsky D, Overland J, Ikeda M and Wan L 2021
Accelerated decline of summer Arctic sea ice during
1850–2017 and the amplified Arctic warming during the
recent decades Environ. Res. Lett. 16 034015
Cao Y, Liang S, Chen X, He T, Wang D and Cheng X 2017
Enhanced wintertime greenhouse effect reinforcing
Arctic amplification and initial sea-ice melting Sci. Rep.
78462
Carton J A, Ding Y and Arrigo K R 2015 The seasonal cycle of the
Arctic Ocean under climate change Geophys. Res. Lett.
42 7681–6
Casagrande F, De Souza R B, Nobre P and Marquez A L 2020 An
inter-hemispheric seasonal comparison of polar
amplification using radiative forcing of a quadrupling CO2
experiment Ann. Geophys. 38 1123–38
Ceppi P and Hartmann D L 2016 Clouds and the atmospheric
circulation response to warming J. Clim. 29 783–99
Chae Y, Kang S M, Jeong S J, Kim B and Frierson D M W 2015
Arctic greening can cause earlier seasonality of Arctic
amplification Geophys. Res. Lett. 42 536–41
Chen Y, Miller J R, Francis J A and Russell G L 2011 Projected
regime shift in Arctic cloud and water vapor feedbacks
Environ. Res. Lett. 6044007
Cho M-H, Yang A-R, Baek E-H, Kang S M, Jeong S-J, Kim J Y and
Kim B-M 2018 Vegetation-cloud feedbacks to future
vegetation changes in the Arctic regions Clim. Dyn.
50 3745–55
Choi Y-S, Hwang J, Ok J, Park D-S R, Su H, Jiang J H, Huang L
and Limpasuvan T 2020 Effect of Arctic clouds on the
ice-albedo feedback in midsummer Int. J. Climatol.
40 4707–14
Chung E-S, Ha K-J, Timmermann A, Stuecker M F, Bodai T and
Lee S-K 2021 Cold-season Arctic amplification driven by
Arctic ocean-mediated seasonal energy transfer Earth’s
Future 9e2020EF001898
Chylek P, Vogelsang T J, Klett J D, Hengartner N, Higdon D,
Lesins G and Dubey M K 2016 Indirect aerosol effect
increases CMIP5 models’ projected Arctic warming J. Clim.
29 1417–28
Clark J P, Shenoy V, Feldstein S B, Lee S and Goss M 2021 The role
of horizontal temperature advection in Arctic amplification
J. Clim. 34 2957–76
Cohen J et al 2014 Recent Arctic amplification and extreme
mid-latitude weather Nat. Geosci. 7627–37
Conley A J, Westervelt D M, Lamarque J-F, Fiore A M, Shindell D,
Correa G, Faluvegi G and Horowitz L W 2018 Multimodel
surface temperature responses to removal of U.S. sulfur
dioxide emissions J. Geophys. Res. Atmos. 123 2773–96
Conservation of Arctic Flora and Fauna (CAFF) 2013 Arctic
Biodiversity Assessment. Status and Trends in Arctic
Biodiversity ed H Meltofte (Denmark: Narayana Press) p 674
Coumou D, Di Capua G, Vavrus S, Wang L and Wang S 2018 The
influence of Arctic amplification on mid-latitude summer
circulation Nat. Commun. 92959
Cronin T W and Jansen M F 2016 Analytic radiative-advective
equilibrium as a model for high-latitude climate Geophys.
Res. Lett. 43 449–57
Cronin T W, Li H and Tziperman E 2017 Suppression of Arctic air
formation with climate warming: investigation with a
two-dimensional cloud-resolving model J. Atmos. Sci.
74 2717–36
Cronin T W and Tziperman E 2015 Low clouds suppress Arctic air
formation and amplify high-latitude continental winter
warming Proc. Natl Acad. Sci. USA 112 11490–5
Crook J A, Forster P M and Stuber N 2011 Spatial patterns of
modeled climate feedback and contributions to temperature
response and polar amplification J. Clim. 24 3575–92
Cunningham S A and Marsh R 2010 Observing and modeling
changes in the Atlantic MOC Wiley Interdiscip. Rev. Clim.
Change 1180–91
Dai A, Luo D, Song M and Liu J 2019 Arctic amplification is
caused by sea-ice loss under increasing CO2Nat. Commun.
10 121
Dai H 2021 Roles of surface albedo, surface temperature and
carbon dioxide in the seasonal variation of Arctic
amplification Geophys. Res. Lett. 48 e2020GL090301
Davy R and Outten S 2020 The Arctic surface climate in CMIP6:
status and developments since CMIP5 J. Clim. 33 8047–68
Dekker E, Bintanja R and Severijns C 2019 Nudging the Arctic
Ocean to quantify sea ice feedbacks J. Clim. 32 2381–95
Docquier D et al 2019 Impact of model resolution on Arctic sea
ice and North Atlantic Ocean heat transport Clim. Dyn.
53 4989–5017
Duan L, Cao L and Caldeira K 2019 Estimating contributions of
sea ice and land snow to climate feedback J. Geophys. Res.
Atmos. 124 199–208
Dunn-Sigouin E, Li C and Kushner P J 2021 Limited influence of
localized tropical sea-surface temperatures on moisture
transport into the Arctic Geophys. Res. Lett.
48 e2020GL091540
Eastman R and Warren S G 2010 Interannual variations of Arctic
cloud types in relation to sea ice J. Clim. 23 4216–32
Fajber R, Kushner P J and Laliberté F 2018 Influence of
midlatitude surface thermal anomalies on the polar
midtroposphere in an idealized moist model J. Atmos. Sci.
75 1089–104
Fan S-M, Harris L M and Horowitz L W 2015 Atmospheric
energy transport to the Arctic 1979–2012 Tellus A67 25482
Feldl N, Anderson B T and Bordoni S 2017a Atmospheric eddies
mediate lapse rate feedback and Arctic amplification J. Clim.
30 9213–24
Feldl N, Bordoni S and Merlis T M 2017b Coupled high-latitude
climate feedbacks and their impact on atmospheric heat
transport J. Clim. 30 189–201
Feldl N, Po-Chedley S, Singh H K A, Hay S and Kushner P J 2020
Sea ice and atmospheric circulation shape the high-latitude
lapse rate feedback npj Clim. Atmos. Sci. 341
Feldl N and Roe G H 2013 The nonlinear and nonlocal nature of
climate feedbacks J. Clim. 26 8289–304
Flannery B P 1984 Energy balance models incorporating transport
of thermal and latent energy J. Atmos. Sci. 41 414–21
Forster P M, Richardson T, Maycock A C, Smith C J, Samset B H,
Myhre G, Andrews T, Pincus R and Schulz M 2016
Recommendations for diagnosing effective radiative forcing
from climate models for CMIP6 J. Geophys. Res. Atmos.
121 12460–75
Francis J A and Vavrus S J 2012 Evidence linking Arctic
amplification to extreme weather in mid-latitudes Geophys.
Res. Lett. 39 L06801
20
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Franzke C L E, Lee S and Feldstein S B 2017 Evaluating Arctic
warming mechanisms in CMIP5 models Clim. Dyn.
48 3247–60
Ghatak D and Miller J R 2013 Implications for Arctic
amplification of changes in the strength of the water vapor
feedback J. Geophys. Res. Atmos. 118 7569–78
Gong T, Feldstein S and Lee S 2017 The role of downward infrared
radiation in the recent Arctic winter warming trend J. Clim.
30 4937–49
Goosse H et al 2018 Quantifying climate feedbacks in polar
regions Nat. Commun. 91919
Graham T and Vellinga M 2013 Heat budget of the upper
Arctic Ocean under a warming climate Clim. Dyn.
40 143–53
Graversen R G 2006 Do changes in the midlatitude circulation
have any impact on the Arctic surface air temperature trend?
J. Clim. 19 5422–38
Graversen R G and Burtu M 2016 Arctic amplification enhanced
by latent energy transport of atmospheric planetary waves
Q. J. R. Meteorol. Soc. 142 2046–54
Graversen R G and Langen P L 2019 On the role of the
atmospheric energy transport in 2×CO2-induced polar
amplification in CESM1 J. Clim. 32 3941–56
Graversen R G, Langen P L and Mauritsen T 2014 Polar
amplification in CCSM4: contributions from the lapse rate
and surface albedo feedbacks J. Clim. 27 4433–50
Graversen R G, Mauritsen T, Tjernström M, Källén E and
Svensson G 2008 Vertical structure of recent Arctic warming
Nature 451 53–6
Graversen R G and Wang M 2009 Polar amplification in a coupled
climate model with locked albedo Clim. Dyn. 33 629–43
Gregory J M et al 2005 A model intercomparison of changes in the
Atlantic thermohaline circulation in response to increasing
atmospheric CO2concentration Geophys. Res. Lett.
32 L12703
Hahn L C, Armour K C, Battisti D S, Donohoe A, Pauling A G
and Bitz C M 2020 Antarctic elevation drives hemispheric
asymmetry in polar lapse rate climatology and feedback
Geophys. Res. Lett. 47 e2020GL088965
Haine T W N and Martin T 2017 The Arctic-Subarctic sea ice
system is entering a seasonal regime: implications for future
Arctic amplification Sci. Rep. 74618
Hall A 2004 The role of surface albedo feedback in climate J. Clim.
17 1550–68
Hall R J, Hanna E and Chen L 2021 Winter Arctic amplification at
the synoptic timescale, 1979–2018, its regional variation and
response to tropical and extratropical variability Clim. Dyn.
56 457–73
Hansen J et al 2005 Efficacy of climate forcings J. Geophys. Res.
110 D18104
Hansen J, Lacis A, Rind D, Russell G, Stone P, Fung I, Ruedy R and
Lerner J 1984 Climate sensitivity: analysis of feedback
mechanisms Climate Processes and Climate Sensitivity AGU
Geophysical Monograph 29 ed J E Hansen and T Takahashi
(Washington, DC: American Geophysical Union) pp 130–63
Hansen J, Sato M and Ruedy R 1997 Radiative forcing and climate
response J. Geophys. Res. 102 6831–64
Hao M, Luo Y, Lin Y, Zhao Z, Wang L and Huang J 2019
Contribution of atmospheric moisture transport to winter
Arctic warming Int. J. Climatol. 39 2697–710
Hartmann D L and Michelsen M L 1993 Large-scale effects on the
regulation of tropical sea surface temperature J. Clim.
62049–62
Held I M and Soden B J 2006 Robust responses of the hydrological
cycle to global warming J. Clim. 19 5686–99
Henry M and Merlis T M 2019 The role of the nonlinearity of the
Stefan-Boltzmann law on the structure of radiatively forced
temperature change J. Clim. 32 335–48
Henry M and Merlis T M 2020 Forcing dependence of
atmospheric lapse rate changes dominates residual polar
warming in solar radiation management climate scenarios
Geophys. Res. Lett. 47 e2020GL087929
Henry M, Merlis T M, Lutsko N J and Rose B E J 2021
Decomposing the drivers of polar amplification with a
single-column model J. Clim. 34 2355–65
Hill D J et al 2014 Evaluating the dominant components of
warming in Pliocene climate simulations Clim. Past
10 79–90
Hoffert M I and Covey C 1992 Deriving global climate
sensitivity from palaeoclimate reconstructions Nature
360 573–6
Holland M M and Bitz C M 2003 Polar amplification of climate
change in coupled models Clim. Dyn. 21 221–32
Hovelsrud G K, Poppel B, van Oort B and Reist J D 2011 Arctic
societies, cultures, and peoples in a changing cryosphere
AMBIO 40 100–10
Hu X, Fan H, Cai M, Sejas S A, Taylor P and Yang S 2020 A less
cloudy picture of the inter-model spread in future global
warming projections Nat. Commun. 11 4472
Hu X, Li Y, Yang S, Deng Y and Cai M 2017 Process-based
decomposition of the decadal climate difference between
2002–13 and 1984–95 J. Clim. 30 4373–93
Huang Y, Dong X, Kay J E, Xi B and McIlhattan E A 2021 The
climate response to increased cloud liquid water over the
Arctic in CESM1: a sensitivity study of
Wegener-Bergeron-Findeisen process Clim. Dyn.
56 3373–94
Huang Y, Xia Y and Tan X 2017 On the pattern of CO2radiative
forcing and poleward energy transport J. Geophys. Res.
Atmos. 122 10578–93
Huang Y and Zhang M 2014 The implication of radiative forcing
and feedback for meridional energy transport Geophys. Res.
Lett. 41 1665–72
Hwang J, Choi Y-S, Kim W, Su H and Jiang J H 2018
Observational estimation of radiative feedback to surface air
temperature over Northern high latitudes Clim. Dyn.
50 615–28
Hwang Y T and Frierson D M W 2010 Increasing atmospheric
poleward energy transport with global warming Geophys.
Res. Lett. 37 L24807
Hwang Y T, Frierson D M W and Kay J E 2011 Coupling between
Arctic feedbacks and changes in poleward energy transport
Geophys. Res. Lett. 38 L17704
Intergovernmental Panel on Climate Change (IPCC) 2013 Climate
Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change ed T F Stocker
et al (Cambridge: Cambridge University Press) p 1535
Ito A et al 2020 Pronounced and unavoidable impacts of low-end
global warming on northern high-latitude land ecosystems
Environ. Res. Lett. 15 044006
Jeong J-H, Kug J-S, Linderholm H W, Chen D, Kim B-M and
Jun S-Y 2014 Intensified Arctic warming under greenhouse
warming by vegetation-atmosphere-sea ice interaction
Environ. Res. Lett. 9094007
Jun S-Y, Ho C-H, Jeong J-H, Choi Y-S and Kim B-M 2016 Recent
changes in winter Arctic clouds and their relationships with
sea ice and atmospheric conditions Tellus A68 29130
Kang S M, Park K, Jin F-F and Stuecker M F 2017 Common
warming pattern emerges irrespective of forcing location
J. Adv. Model. Earth Syst. 92413–24
Kay J E and Gettelman A 2009 Cloud influence on and response to
seasonal Arctic sea ice loss J. Geophys. Res. 114 D18204
Kay J E, Holland M M, Bitz C M, Blanchard-Wrigglesworth E,
Gettelman A, Conley A and Bailey D 2012 The influence of
local feedbacks and northward heat transport on the
equilibrium Arctic climate response to increased greenhouse
gas forcing J. Clim. 25 5433–50
Kiehl J T and Ramanathan V 1982 Radiative heating due to
increased CO2: the role of H2O continuum absorption in
the 12–18 µm region J. Atmos. Sci. 39 2923–6
Kim D, Kang S M, Shin Y and Feldl N 2018 Sensitivity of polar
amplification to varying insolation conditions J. Clim.
31 4933–47
21
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Kim H-M and Kim B-M 2017 Relative contributions of
atmospheric energy transport and sea ice loss to the recent
warm Arctic winter J. Clim. 30 7441–50
Kim J-Y and Kim K-Y 2019 Relative role of horizontal and vertical
processes in the physical mechanism of wintertime Arctic
amplification Clim. Dyn. 52 6097–107
Kim K-Y, Hamlington B D, Na H and Kim J 2016 Mechanism of
seasonal Arctic sea ice evolution and Arctic amplification
Cryosphere 10 2191–202
Kim K-Y, Kim J-Y, Kim J, Yeo S, Na H, Hamlington B D and
Leben R R 2019a Vertical feedback mechanism of winter
Arctic amplification and sea ice loss Sci. Rep. 91184
Kim M J, Yeh S-W, Park R J, Son S-W, Moon B-K, Kim B-G,
Kim J-J and Kim S-W 2019b Regional Arctic amplification
by a fast atmospheric response to anthropogenic sulfate
aerosol forcing in China J. Clim. 32 6337–48
Knutti R and Rugenstein M A A 2015 Feedbacks, climate
sensitivity and the limits of linear models Phil. Trans. R. Soc.
A373 20150146
Koenigk T and Brodeau L 2014 Ocean heat transport into the
Arctic in the twentieth and twenty-first century in EC-Earth
Clim. Dyn. 42 3101–20
Koenigk T and Brodeau L 2017 Arctic climate and its interaction
with lower latitudes under different levels of anthropogenic
warming in a global coupled climate model Clim. Dyn.
49 471–92
Koenigk T, Brodeau L, Graversen R G, Karlsson J, Svensson G,
Tjernström M, Willén U and Wyser K 2013 Arctic climate
change in 21st century CMIP5 simulations with EC-Earth
Clim. Dyn. 40 2719–43
Laˆıné A, Yoshimori M and Abe-Ouchi A 2016 Surface Arctic
amplification factors in CMIP5 models: land and oceanic
surfaces and seasonality J. Clim. 29 3297–316
Laliberté F and Kushner P J 2013 Isentropic constraints by
midlatitude surface warming on the Arctic midtroposphere
Geophys. Res. Lett. 40 606–11
Lambert F, Kug J-S, Park R J, Mahowald N, Winckler G,
Abe-Ouchi A, O’ishi R, Takemura T and Lee J-H 2013 The
role of mineral-dust aerosols in polar temperature
amplification Nat. Clim. Change 3487–91
Lang A, Yang S and Kaas E 2017 Sea ice thickness and recent
Arctic warming Geophys. Res. Lett. 44 409–18
Langen P L and Alexeev V A 2007 Polar amplification as a
preferred response in an idealized aquaplanet GCM Clim.
Dyn. 29 305–17
Langen P L, Graversen R G and Mauritsen T 2012 Separation of
contributions from radiative feedbacks to polar
amplification on an aquaplanet J. Clim. 25 3010–24
Lauer M, Block K, Salzmann M and Quaas J 2020 CO2-forced
changes of Arctic temperature lapse rates in CMIP5 models
Meteorol. Z. 29 79–93
Lee S 2014 A theory for polar amplification from a general
circulation perspective Asia-Pac. J. Atmos. Sci. 50 31–43
Lee S, Gong T, Johnson N, Feldstein S B and Pollard D 2011 On
the possible link between tropical convection and the
Northern Hemisphere Arctic surface air temperature change
between 1958 and 2001 J. Clim. 24 4350–67
Lesins G, Duck T J and Drummond J R 2012 Surface energy
balance framework for Arctic amplification of climate
change J. Clim. 25 8277–88
Li F and Newman P 2020 Stratospheric water vapor feedback and
its climate impacts in the coupled atmosphere-ocean
Goddard Earth Observing System Chemistry-Climate
Model Clim. Dyn. 55 1585–95
Liu Y, Attema J, Moat B and Hazeleger W 2020 Synthesis and
evaluation of historical meridional heat transport from
midlatitudes towards the Arctic Earth Syst. Dyn. 11 77–96
Lott F C, Christidis N, Ciavarella A and Stott P A 2020 The effect
of human land use change in the Hadley centre attribution
system Atmos. Sci. Lett. 21 e972
Lu J and Cai M 2009a Seasonality of polar surface warming
amplification in climate simulations Geophys. Res. Lett.
36 L16704
Lu J and Cai M 2009b A new framework for isolating individual
feedback processes in coupled general circulation climate
models. Part I: formulation Clim. Dyn. 32 873–85
Lu J and Cai M 2010 Quantifying contributions to polar warming
amplification in an idealized coupled general circulation
model Clim. Dyn. 34 669–87
Lutsko N J and Popp M 2018 The influence of meridional
gradients in insolation and longwave optical depth on the
climate of a gray radiation GCM J. Clim. 31 7803–22
Mahlstein I and Knutti R 2011 Ocean heat transport as a cause for
model uncertainty in projected Arctic warming J. Clim.
24 1451–60
Manabe S and Stouffer R J 1980 Sensitivity of a global climate
model to an increase of CO2concentration in the
atmosphere J. Geophys. Res. 85 5529–54
Manabe S and Wetherald R T 1975 The effects of doubling the
CO2concentration on the climate of a general circulation
model J. Atmos. Sci. 32 3–15
Marshall J, Armour K C, Scott J R, Kostov Y, Hausmann U,
Ferreira D, Shepherd T G and Bitz C M 2014 The ocean’s
role in polar climate change: asymmetric Arctic and
Antarctic responses to greenhouse gas and ozone forcing
Phil. Trans. R. Soc. A372 20130040
Marshall J, Scott J R, Armour K C, Campin J-M, Kelley M and
Romanou A 2015 The ocean’s role in the transient response
of climate to abrupt greenhouse gas forcing Clim. Dyn.
44 2287–99
McCrystall M R, Hosking J S, White I P and Maycock A C 2020
The impact of changes in tropical sea surface temperatures
over 1979–2012 on Northern Hemisphere high-latitude
climate J. Clim. 33 5103–21
McCusker K E, Fyfe J C and Sigmond M 2016 Twenty-five winters
of unexpected Eurasian cooling unlikely due to Arctic
sea-ice loss Nat. Geosci. 9838–42
McGuire A D, Anderson L G, Christensen T R, Dallimore S,
Guo L, Hayes D J, Heimann M, Lorenson T D,
Macdonald R W and Roulet N 2009 Sensitivity of the carbon
cycle in the Arctic to climate change Ecol. Monogr. 79 523–55
Melia N, Haines K and Hawkins E 2016 Sea ice decline and 21st
century trans-Arctic shipping routes Geophys. Res. Lett.
43 9720–8
Merlis T M 2014 Interacting components of the
top-of-atmosphere energy balance affect changes in regional
surface temperature Geophys. Res. Lett. 41 7291–7
Merlis T M 2015 Direct weakening of tropical circulations from
masked CO2radiative forcing Proc. Natl Acad. Sci. USA
112 13167–71
Merlis T M and Henry M 2018 Simple estimates of polar
amplification in moist diffusive energy balance models
J. Clim. 31 5811–24
Mewes D and Jacobi C 2019 Heat transport pathways into the
Arctic and their connections to surface air temperatures
Atmos. Chem. Phys. 19 3927–37
Middlemas E A, Kay J E, Medeiros B M and Maroon E A 2020
Quantifying the influence of cloud radiative feedbacks on
Arctic surface warming using cloud locking in an Earth
system model Geophys. Res. Lett. 47 e2020GL089207
Miller G H, Alley R B, Brigham-Grette J, Fitzpatrick J J, Polyak L,
Serreze M C and White J W C 2010 Arctic amplification: can
the past constrain the future? Quat. Sci. Rev. 29 1779–90
Mitchell D M, Lo Y T E, Seviour W J M, Haimberger L and
Polvani L M 2020 The vertical profile of recent tropical
temperature trends: persistent model biases in the context of
internal variability Environ. Res. Lett. 15 1040b4
Mitchell D M, Thorne P W, Stott P A and Gray L J 2013 Revisiting
the controversial issue of tropical tropospheric temperature
trends Geophys. Res. Lett. 40 2801–6
22
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Morrison A L, Kay J E, Frey W R, Chepfer H and Guzman R 2019
Cloud response to Arctic sea ice loss and implications for
future feedback in the CESM1 climate model J. Geophys. Res.
Atmos. 124 1003–20
Najafi M R, Zwiers F W and Gillett N P 2015 Attribution of Arctic
temperature change to greenhouse-gas and aerosol
influences Nat. Clim. Change 5246–9
Natali S M et al 2019 Large loss of CO2in winter observed across
the northern permafrost region Nat. Clim. Change
9852–7
Newell R E 1979 Climate and the Ocean: measurements of
changes in sea-surface temperature should permit us to
forecast certain climatic changes several months ahead
Amer. Sci. 67 405–16
Nummelin A, Li C and Hezel P J 2017 Connecting ocean heat
transport changes from the midlatitudes to the Arctic Ocean
Geophys. Res. Lett. 44 1899–908
O’ishi R and Abe-Ouchi A 2011 Polar amplification in the
mid-Holocene derived from dynamical vegetation change
with a GCM Geophys. Res. Lett. 38 L14702
Osborne E, Richter-Menge J and Jeffries M (ed) 2018 Arctic report
card 2018 (available at: www.arctic.noaa.gov/Report-Card)
Palm S P, Strey S T, Spinhirne J and Markus T 2010 Influence of
Arctic sea ice extent on polar cloud fraction and vertical
structure and implications for regional climate J. Geophys.
Res. 115 D21209
Park J-Y, Kug J-S, Bader J, Rolph R and Kwon M 2015 Amplified
Arctic warming by phytoplankton under greenhouse
warming Proc. Natl Acad. Sci. USA 112 5921–6
Park K, Kang S M and Kim D 2018 Contrasting local and remote
impacts of surface heating on polar warming and
amplification J. Clim. 31 3155–66
Park S-W, Kim J-S and Kug J-S 2020 The intensification of Arctic
warming as a result of CO2physiological forcing Nat.
Commun. 11 2098
Parmentier F-J W, Christensen T R, Sørensen L L, Rysgaard S,
McGuire A D, Miller P A and Walker D A 2013 The impact
of a lower sea-ice extent on Arctic greenhouse-gas exchange
Nat. Clim. Change 3195–202
Payne A E, Jansen M F and Cronin T W 2015 Conceptual model
analysis of the influence of temperature feedbacks on polar
amplification Geophys. Res. Lett. 42 9561–70
Philipp D, Stengel M and Ahrens B 2020 Analyzing the Arctic
feedback mechanism between sea ice and low-level clouds
using 34 years of satellite observations J. Clim. 33 7479–501
Pistone K, Eisenman I and Ramanathan V 2014 Observational
determination of albedo decrease caused by vanishing Arctic
sea ice Proc. Natl Acad. Sci. USA 111 3322–6
Pithan F and Mauritsen T 2014 Arctic amplification dominated by
temperature feedbacks in contemporary climate models
Nat. Geosci. 7181–4
Po-Chedley S and Fu Q 2012 Discrepancies in tropical upper
tropospheric warming between atmospheric circulation
models and satellites Environ. Res. Lett. 7044018
Polvani L M, Previdi M, England M R, Chiodo G and Smith K L
2020 Substantial twentieth-century Arctic warming caused
by ozone-depleting substances Nat. Clim. Change
10 130–3
Praetorius S, Rugenstein M, Persad G and Caldeira K 2018 Global
and Arctic climate sensitivity enhanced by changes in North
Pacific heat flux Nat. Commun. 93124
Previdi M 2010 Radiative feedbacks on global precipitation
Environ. Res. Lett. 5025211
Previdi M, Janoski T P, Chiodo G, Smith K L and Polvani L M
2020 Arctic amplification: a rapid response to radiative
forcing Geophys. Res. Lett. 47 e2020GL089933
Previdi M and Liepert B G 2012 The vertical distribution of
climate forcings and feedbacks from the surface to top of
atmosphere Clim. Dyn. 39 941–51
Richter-Menge J, Osborne E, Druckenmiller M and Jeffries M O
(ed) 2019 The Arctic [in ‘State of the Climate in 2018’] Bull.
Am. Meteorol. Soc. 100 S141–68
Rodgers K B, Lohmann G, Lorenz S, Schneider R and
Henderson G M 2003 A tropical mechanism for Northern
Hemisphere deglaciation Geochem. Geophys. Geosys. 41046
Russotto R D and Biasutti M 2020 Polar amplification as an
inherent response of a circulating atmosphere: results from
the TRACMIP aquaplanets Geophys. Res. Lett.
47 e2019GL086771
Sagoo N and Storelvmo T 2017 Testing the sensitivity of past
climates to the indirect effects of dust Geophys. Res. Lett.
44 5807–17
Santer B D et al 2013 Identifying human influences on
atmospheric temperature Proc. Natl Acad. Sci. USA
110 26–33
Santer B D et al 2017 Comparing tropospheric warming in
climate models and satellite data J. Clim. 30 373–92
Schneider E K, Kirtman B P and Lindzen R S 1999 Tropospheric
water vapor and climate sensitivity J. Atmos. Sci. 56 1649–58
Schneider E K, Lindzen R S and Kirtman B P 1997 A tropical
influence on global climate J. Atmos. Sci. 54 1349–58
Schneider Von Deimling T et al 2021 Consequences of permafrost
degradation for Arctic infrastructure—bridging the model
gap between regional and engineering scales Cryosphere
15 2451–71
Schuur E A G et al 2015 Climate change and the permafrost
carbon feedback Nature 520 171–9
Screen J A, Deser C and Simmonds I 2012 Local and remote
controls on observed Arctic warming Geophys. Res. Lett.
39 L10709
Screen J A and Simmonds I 2010a Increasing fall-winter energy
loss from the Arctic Ocean and its role in Arctic temperature
amplification Geophys. Res. Lett. 37 L16707
Screen J A and Simmonds I 2010b The central role of diminishing
sea ice in recent Arctic temperature amplification Nature
464 1334–7
Screen J A and Simmonds I 2012 Declining summer snowfall in
the Arctic: causes, impacts and feedbacks Clim. Dyn.
38 2243–56
Sejas S A and Cai M 2016 Isolating the temperature feedback loop
and its effects on surface temperature J. Atmos. Sci.
73 3287–303
Sejas S A, Cai M, Hu A, Meehl G A, Washington W and Taylor P C
2014 Individual feedback contributions to the seasonality of
surface warming J. Clim. 27 5653–69
Sellers W D 1969 A global climatic model based on the energy
balance of the Earth-atmosphere system J. Appl. Meteorol.
8392–400
Semmler T, Pithan F and Jung T 2020 Quantifying two-way
influences between the Arctic and mid-latitudes through
regionally increased CO2concentrations in coupled climate
simulations Clim. Dyn. 54 3307–21
Serreze M C, Barrett A P, Stroeve J C, Kindig D N and
Holland M M 2009 The emergence of surface-based Arctic
amplification Cryosphere 311–19
Serreze M C and Barry R G 2011 Processes and impacts of Arctic
amplification: a research synthesis Glob. Planet. Change
77 85–96
Shepherd A et al 2012 A reconciled estimate of ice-sheet mass
balance Science 338 1183–9
Sherwood S C et al 2020 An assessment of Earth’s climate
sensitivity using multiple lines of evidence Rev. Geophys.
58 e2019RG000678
Shine K P, Cook J, Highwood E J and Joshi M M 2003 An
alternative to radiative forcing for estimating the relative
importance of climate change mechanisms Geophys. Res.
Lett. 30 2047
Singh H A, Garuba O A and Rasch P J 2018 How asymmetries
between Arctic and Antarctic climate sensitivity are
modified by the ocean Geophys. Res. Lett. 45 13031–40
Singh H A, Rasch P J and Rose B E J 2017 Increased ocean heat
convergence into the high latitudes with CO2doubling
enhances polar-amplified warming Geophys. Res. Lett.
44 10583–91
23
Environ. Res. Lett. 16 (2021) 093003 M Previdi et al
Skific N and Francis J A 2013 Drivers of projected change in
Arctic moist static energy transport J. Geophys. Res. Atmos.
118 2748–61
Sluijs A et al 2006 Subtropical Arctic Ocean temperatures during
the Palaeocene/Eocene thermal maximum Nature 441 610–3
Smith D M et al 2019 The polar amplification model
intercomparison project (PAMIP) contribution to CMIP6:
investigating the causes and consequences of polar
amplification Geosci. Model. Dev. 12 1139–64
Smith K L, Chiodo G, Previdi M and Polvani L M 2018 No surface
cooling over Antarctica from the negative greenhouse effect
associated with instantaneous quadrupling of CO2
concentrations J. Clim. 31 317–23
Soden B J and Held I M 2006 An assessment of climate feedbacks
in coupled ocean-atmosphere models J. Clim. 19 3354–60
Soden B J, Held I M, Colman R, Shell K M, Kiehl J T and
Shields C A 2008 Quantifying climate feedbacks using
radiative kernels J. Clim. 21 3504–20
Södergren A H, McDonald A J and Bodeker G E 2018 An energy
balance model exploration of the impacts of interactions
between surface albedo, cloud cover and water vapor on
polar amplification Clim. Dyn. 51 1639–58
Solomon A 2006 Impact of latent heat release on polar climate
Geophys. Res. Lett. 33 L07716
Song X, Zhang G J and Cai M 2014 Quantifying contributions of
climate feedbacks to tropospheric warming in the NCAR
CCSM3.0 Clim. Dyn. 42 901–17
Sorokina S A and Esau I N 2011 Meridional energy flux in the
Arctic from data of the radiosonde archive IGRA Izv. Atmos.
Ocean. Phys. 47 572–83
Spielhagen R F, Werner K, Aagaard Sørensen S, Zamelczyk K,
Kandiano E, Budeus G, Husum K, Marchitto T M and
Hald M 2011 Enhanced modern heat transfer to the Arctic
by warm Atlantic water Science 331 450–3
Stap L B, van de Wal R S W, De Boer B, Bintanja R and
Lourens L J 2014 Interaction of ice sheets and climate during
the past 800,000 years Clim. Past 10 2135–52
Stap L B, van de Wal R S W, De Boer B, Bintanja R and Lourens L J
2017 The inf