CPClimate of the PastCPClim. Past1814-9332Copernicus PublicationsGöttingen, Germany10.5194/cp-12-1619-2016Sensitivity of Pliocene climate simulations in MRI-CGCM2.3 to respective
boundary conditionsKamaeYouichikamae.yoichi.fw@u.tsukuba.ac.jphttps://orcid.org/0000-0003-0461-5718YoshidaKoheihttps://orcid.org/0000-0002-2422-5584UedaHiroakiFaculty of Life and Environmental Sciences, University of Tsukuba,
Tsukuba, 305-8572, JapanScripps Institution of Oceanography, University of California San
Diego, La Jolla, 92093-0206, USAMeteorological Research Institute, Tsukuba, 305-0052, JapanYouichi Kamae (kamae.yoichi.fw@u.tsukuba.ac.jp)8August20161281619163427April20169May201613July201619July2016This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://cp.copernicus.org/articles/12/1619/2016/cp-12-1619-2016.htmlThe full text article is available as a PDF file from https://cp.copernicus.org/articles/12/1619/2016/cp-12-1619-2016.pdf
Accumulations of global proxy data are essential steps
for improving reliability of climate model simulations for the Pliocene
warming climate. In the Pliocene Model Intercomparison Project phase 2
(PlioMIP2), a part project of the Paleoclimate Modelling Intercomparison
Project phase 4, boundary forcing data have been updated from the PlioMIP
phase 1 due to recent advances in understanding of oceanic, terrestrial and
cryospheric aspects of the Pliocene palaeoenvironment. In this study,
sensitivities of Pliocene climate simulations to the newly archived boundary
conditions are evaluated by a set of simulations using an atmosphere–ocean
coupled general circulation model, MRI-CGCM2.3. The simulated Pliocene
climate is warmer than pre-industrial conditions for 2.4 ∘C in
global mean, corresponding to 0.6 ∘C warmer than the PlioMIP1
simulation by the identical climate model. Revised orography, lakes, and
shrunk ice sheets compared with the PlioMIP1 lead to local and remote
influences including snow and sea ice albedo feedback, and poleward heat
transport due to the atmosphere and ocean that result in additional warming
over middle and high latitudes. The amplified higher-latitude warming is
supported qualitatively by the proxy evidences, but is still underestimated
quantitatively. Physical processes responsible for the global and regional
climate changes should be further addressed in future studies under
systematic intermodel and data–model comparison frameworks.
Introduction
Atmosphere-ocean coupled general circulation models (AOGCMs) have widely
been used for climate projections on decadal to centennial timescales since
the late 20th century. Recently, representations of large-scale climate
state, its variability, and parameterized processes in the AOGCMs (e.g.
cloud and convection) have been improved substantially (e.g. Reichler and
Kim, 2008; Klein et al., 2013; Bellenger et al., 2014). A large ensemble of
multiple climate models has contributed to addressing robust climate trends
in the future projections (e.g. Xie et al., 2015). Model intercomparisons of
the past climate changes and data–model comparisons are powerful frameworks
for understanding physical processes responsible for climate change and
variability and assessing reliability of the climate model projections (e.g.
Braconnot et al., 2012; Masson-Delmotte et al., 2014).
Since the 1990s, globally warmed climate during the Pliocene
(∼ 5 to 3 Ma) has attracted much attention as a potential
analogue for the ongoing global climate change (e.g. Masson-Delmotte et al.,
2014). Previous modelling and proxy-based studies revealed that the Pliocene
climate can be characterized by substantial global warming (2.7 to 4.0 ∘C) with anomalous zonal and meridional temperature gradients
(Dowsett et al., 1992, 2010; Wara et al., 2005; Fedorov et al., 2013;
Haywood et al., 2013, 2016a). Under the Pliocene Model Intercomparison
Project phase 1 (PlioMIP1; Haywood et al., 2010, 2011), a part project of
the Paleoclimate Modelling Intercomparison Project phase 3 (PMIP3), nine
AOGCMs provided results of mid-Pliocene (3.264 to 3.025 Ma) climate
simulations. A palaeoenvironmental reconstruction project named the Pliocene
Research Interpretation and Synoptic Mapping (PRISM) for PlioMIP1 (PRISM3D;
Dowsett et al., 2010) provided global boundary condition dataset that is
needed for performing the mid-Pliocene climate simulations. Through
intercomparisons of multiple model results and data–model comparisons,
large-scale climate pattern (Haywood et al., 2013), East Asian monsoon
behaviour (R. Zhang et al., 2013), Atlantic Meridional Overturning
Circulation (AMOC; Z.-S. Zhang et al., 2013), terrestrial climate (Salzmann
et al., 2013), and oceanic conditions (Dowsett et al., 2012, 2013) were
examined systematically. The PlioMIP1 was the first model intercomparison
project comparing past modelled climate prescribed with proxy-based
vegetation pattern that is distinct from the present-day condition. Numerous
independent proxy data for the Pliocene (Haywood et al., 2016a) suggested
predominant warming over the North Atlantic and the Arctic region and wetter
climate over land (Salzmann et al., 2008). The climate simulations under the
PlioMIP1 tended to underestimate the warming over the Northern Hemisphere
middle and high latitudes (Salzmann et al., 2013; Dowsett et al., 2012,
2013) although the latitudinal warming gradient was reproduced
qualitatively. Hill et al. (2014) pointed out a robust contribution of
surface albedo due to changes in vegetation and ice sheets and ice-albedo
feedbacks (sea ice and snow) to the warming amplifications over the polar
regions.
Since the PRISM3D/PlioMIP1, newly archived proxy evidences have been
integrated as PRISM4 dataset (Dowsett et al., 2016; hereafter D16) which is
planned to be used for an ongoing modelling intercomparison project, PlioMIP
phase 2 (PlioMIP2; Haywood et al., 2016b; hereafter H16b), a part project of
PMIP4 (Kageyama et al., 2016). In the PRISM4/PlioMIP2, boundary conditions
for the Pliocene climate simulations including orography and ice sheets have
been updated. In addition, global lakes and soil data were newly included in
the PRISM4 dataset. Numerous mega-lakes over land associated with the wetter
terrestrial climate during the Pliocene were suggested to be important for
simulating local and large-scale anomalous climate (Pound et al., 2014).
In the PlioMIP2, dynamical predictions of vegetation and lakes, changes in
land and ocean topography, and change in the land–sea mask are recommended
for the Pliocene climate simulations. However, the respective roles of the
revised boundary conditions in the Pliocene climate simulations have not
been addressed sufficiently. A set of sensitivity experiments with different
combinations of modern and Pliocene boundary conditions is planned to be
performed in the PlioMIP2. The respective roles of the boundary conditions
can be evaluated by comparing results of the sensitivity runs. In this
study, we conduct PlioMIP2 climate simulations by using an AOGCM,
MRI-CGCM2.3, that was also used in the PlioMIP1 (Kamae and Ueda, 2012;
hereafter KU12). The results of the PlioMIP1 run, the PlioMIP2 run, and the
PlioMIP2 sensitivity runs with swapped boundary conditions are used to
examine respective roles of the updated boundary conditions. This study
reports that the revised ice sheets in the high latitudes and global land
properties including vegetation result in an amplified high-latitude warming
via direct influence, radiative feedback, and atmospheric and oceanic heat
transports. The anomalous warming simulated in the PlioMIP2 protocol is more
consistent with the proxy evidences than the PlioMIP1 run. Section 2
describes the data and methods including proxy-based boundary conditions and
modelling strategy. Section 3 presents general characteristics of simulated
Pliocene climate and compares it with the PlioMIP1 results. Section 4
examines the roles of atmospheric and oceanic meridional heat transports in
the simulated middle and high-latitude warming in the model. Section 5
compares modelled and reconstructed sea surface temperature (SST) and
assesses SST reproducibility of the model simulation. In Sect. 6, we present a
summary and discussion of this study.
Summary of model and experimental setting.
ModelMRI-CGCM2.3PalaeogeographyStandardDynamic vegetationNoCarbon cycleNoDynamical lakeNoCH4760 ppbvN2O270 ppbvOzoneWang et al. (1995)Solar constant1365 W m-2Eccentricity0.016724Obliquity23.446∘Perihelion102.04∘Integration length500 yearsData and methodsClimate model
An AOGCM named MRI-CGCM2.3 (Yukimoto et al., 2006) was used for performing
the PlioMIP2 experiments in this study. The model is identical to that used
in the PlioMIP1 (KU12). Atmospheric model has a horizontally T42 resolution
(∼ 2.8∘) and vertically 30 layers (model top is 0.4 hPa). Oceanic component is a Bryan-Cox-type ocean general circulation model
with a horizontal resolution of 2.5∘ longitude and 2.0–0.5∘ latitude and 23 layers (the deepest layer is 5000 m).
Details of the atmosphere and ocean models can be found in Yukimoto et al. (2006) and KU12. Vegetation, lakes and atmospheric CO2 concentration
are prescribed in the model as boundary conditions (see Sect. 2.2) because
the model does not predict vegetation, lakes, and carbon cycle (Table 1).
Land scheme is simple biosphere model (SiB2; Sellers et al., 1986; Sato et
al., 1989), which predicts soil water and surface heat budget. Parameters
for the land scheme depend on 13 types of vegetation category (see Sect. 2.2). Lakes are treated in the land surface model as inland water in grid
boxes with a drainage basin unconnected to oceans. The model predicts water
budget for lakes, but the lake surface temperature is predicted by the heat
budget at the water surface, assuming a slab with a thickness of 50 m
(Yukimoto et al., 2006).
Although both the AOGCM and an atmosphere-only general circulation model
were used in the PlioMIP1 (Kamae et al., 2011; KU12), we perform only the
AOGCM simulations in this study according to the PlioMIP2 protocol (H16b).
In KU12, a set of AOGCM simulations with and without flux adjustments (heat,
fresh water flux and wind stress) were performed. In the present study, we
only integrated the model without any flux adjustments (similar to
AOGCM_NFA run in KU12).
Experimental designs for pre-industrial and Pliocene climate
simulations
Details of six PlioMIP2 experiments examined in this study.
PlioMIP1 Pliocene run represents AOGCM_NFA run in Kamae and
Ueda (2012). Soil and land–sea mask are identical among all the
experiments.
ExperimentsOrography, lakesVegetationIce sheetCO2(ppmv)E280 (Pre-industrial)ModernModernModern280E400ModernModernModern400E560ModernModernModern560Ei280ModernModernPliocene280Eo280PliocenePlioceneModern280Eoi400 (Pliocene)PliocenePliocenePliocene400PlioMIP1 PliocenePlioMIP1 orography,PliocenePlioMIP1405modern lake
Anomalies in global-mean surface air temperature (ΔSAT; ∘C) in the PlioMIP2 runs relative to E280. The anomalies are
calculated by averages for the last 50 years of the individual runs.
In addition to two core experiments (pre-industrial and Pliocene), a set of
sensitivity experiments (Table 2) is also proposed in the PlioMIP2 (Table 3
in H16b). Alterations of soil and land–sea mask (e.g. the Bering Strait and
the Canadian Arctic Archipelago) were recommended in the PlioMIP2. Due to
technical difficulties, these alterations were not incorporated into the
current PlioMIP2 simulations. Due to the low resolution and the simple
experimental setting (i.e. low computational cost), we can conduct all the
set of simulations proposed in the PlioMIP2 (in total 12-type 500-year long
simulations; H16b). We also plan to conduct higher resolution and/or higher
complexity PlioMIP2 simulations with a higher-resolution AOGCM and/or an
Earth system model (see Sect. 6).
Prescribed land cover (SiB2 classification) for (a) modern and
(b) Pliocene conditions.
We can compare results of this study with the PlioMIP1 directly because the
experimental setting for the PlioMIP2 pre-industrial run (Tables 1 and 2) is
identical to the PlioMIP1 (KU12). In the pre-industrial run, CO2,
CH4 and N2O concentrations were set to be 280 ppmv, 760 and
270 ppbv, respectively. Ozone concentration in each month was derived from
climatology in Wang et al. (1995). Orbital parameters were identical to the
PlioMIP1 (Table 1). Modern land orography, lakes, and ice sheets of the
MRI-CGCM2.3 (KU12; Figs. 1a, 2a) were used in the pre-industrial run.
Vegetation pattern was derived from PRISM3D modern map converted into 13
types of SiB2 classification (Table 3 in KU12; Fig. 1a).
To simulate the Pliocene climate, the PRISM4 global palaeoenvironmental
dataset (D16) including atmospheric trace gases, orography, vegetation
covers, ice sheets, and lakes are prescribed in the model. Pliocene
orography was updated from the PRISM3D (Sohl et al., 2009) by considering
mantle flow (Rowley et al., 2013) and glacial isostatic response of ice
sheet loading (Raymo et al., 2011). We added anomalous orography (Pliocene
minus modern) to the model's modern orography (Fig. 3) according to the
“anomaly method” recommended in the PlioMIP2 (H16b). Direct proxy
evidences for the Greenland and Antarctic ice sheets during the Pliocene
were not available. The PRISM4 provided Greenland ice sheet data that are
confined to high elevations in the East Greenland Mountains suggested by
results of Pliocene Land Ice Sheet Model Intercomparison Project (Dolan et
al., 2015; Koenig et al., 2015). According to proxy data for the Antarctic
ice sheet (Naish et al., 2009; Pollard and DeConto, 2009), ice-free
condition and the identical ice sheet estimate to the PRISM3D were assumed
in the West and East Antarctica, respectively (D16; Figs. 1b, 3). The
resultant prescribed ice sheets over the Greenland and Antarctica are
smaller than that in KU12. We prescribed the Pliocene ice sheets by changing
land orography (Fig. 3) and land cover (Fig. 1) over the ice sheets regions.
Because the model does not predict dynamic vegetation, we prescribe
vegetation pattern (Fig. 1b) that is identical to the PRISM3D/PlioMIP1
(Salzmann et al., 2008), according to the recommendation in H16b.
Prescribed lake area fraction over land. (a) Modern, (b) Pliocene,
and (c) Pliocene minus modern.
In addition to vegetation, global lake distribution during the Pliocene was
also suggested to be distinct from the present day. Schuster et al. (2001,
2006) and Griffin (2006) suggested the existence of the African Megalakes
existed during the Miocene and the Pliocene. Contoux et al. (2013) and Pound
et al. (2014) pointed out an important role of the Pliocene lakes in the
global climate through local atmosphere–land interaction and remote
influences. The PRISM4 provides global lake area data (Pound et al., 2014)
for the PlioMIP2 (Fig. 2). We prescribe the Pliocene lakes (Fig. 2b; Table 2) by adding anomalous areas of lakes (Fig. 2c) to model's modern lakes (the
Caspian Sea, the Aral Sea, Lake Balkhash, Lake Chad and Lake Eyre; Fig. 2a).
In this paper, the results of six PlioMIP2 experiments (Table 2) are
reported: pre-industrial run; Pliocene run; pre-industrial runs but with
CO2 concentration of 400 and 560 ppmv (hereafter E400 and E560 runs);
pre-industrial run but with Pliocene ice sheets (Ei280 run); and
pre-industrial run but with Pliocene orography, vegetation, and lakes (OVL)
except over the ice sheet regions (hereafter Eo280 run). CH4,
N2O, ozone, solar constant, orbital parameters are identical to the
pre-industrial run (Table 1). Land orography, lake area, vegetation, land
ice, atmospheric CO2 concentration are altered in the Pliocene run from
the pre-industrial run. Combinations of the boundary conditions are changed
in the sensitivity runs so that the impacts of the individual components can
be evaluated (Table 2; Sect. 2.3). A large spread remains in the assessment
of the Pliocene CO2 concentration (e.g. Raymo et al., 1996; Seki et
al., 2010; D16). Although 400 ppmv of CO2 concentration is prescribed
in the Pliocene run, other concentrations are also used in the other
sensitivity runs (H16b). In this paper, we examine the results of
simulations with CO2 concentration of 280, 400, and 560 ppmv (Table 2).
Prescribed anomaly in land orography (m) for Pliocene relative to
modern.
The results are also compared with PlioMIP1 run conducted in KU12 (Table 2).
Pliocene lakes were set to be identical to the modern. The PRISM3D-based
vegetation (Salzmann et al., 2008) was also prescribed in the PlioMIP1
Pliocene run (Fig. 1b). A prescribed CO2 concentration of 405 ppmv was
slightly higher than the PlioMIP2.
The initial condition for the PlioMIP2 runs is identical to the
pre-industrial run: 31 December of the PlioMIP1 control run (in
NFA_AOGCM) after 500-year spin up (Fig. 3b in KU12). The model
is integrated for another 500 years with swapped boundary conditions and the
last 50 years are used for analyses. Note that reconstructed deep ocean
temperature was added to the initial condition of the PlioMIP1 Pliocene run
(KU12), distinct to the PlioMIP2. Despite the difference in the initial deep
ocean temperature, ocean circulation and SST after the 500-year integrations
are generally similar between the two (see Sects. 3.3, 4, 5).
Respective roles of boundary conditions
In the PlioMIP2, relative contributions of the boundary conditions to
climate anomalies (e.g. global mean warming) can be evaluated by comparing
the set of sensitivity experiments (H16b). In this study, we use the six
PlioMIP2 simulations (Table 2) and evaluate the respective contributions by
following Eqs. (1)–(6):
All=Eoi400-[E280],CO2=E400-[E280],OVL=Eo280-[E280],Ice Sheet=Ei280-[E280],Sum=CO2+OVL+IceSheet,Residual=All-Sum,
where [] represents the experiment name (Table 2). E280 and Eoi400 indicate
the pre-industrial and Pliocene runs, respectively. Difference of results
between Eo280 and E280 run corresponds to effect of differences in OVL. Sum
of Eqs. (2)–(4) is used as a reconstruction of Eq. (1). By Eqs. (5), (6), we
can compare the simulated climate anomaly between the Pliocene run and the
pre-industrial and its reconstruction. Residual (Eq. 6) indicates a nonlinear effect
of combination of the boundary conditions associated with nonlinearity in
forcing–response relationship (e.g. Shiogama et al., 2013). This
decomposition method is not identical to that recommended in H16b. While
decomposed relative contributions could be dependent on the choice of
quantifying methods, Residual term shown in this study is generally minor to All (see
Sect. 3), suggesting an effectiveness of the decomposing method used in this
study.
Time evolution of annual-mean global-mean surface air temperature
(SAT; ∘C) in each run.
(a) Anomaly in SAT (∘C) in Pliocene run relative to
pre-industrial run (hereafter All). (b) E400 run minus pre-industrial run
(CO2), (c) Eo280 run minus pre-industrial run (OVL), and (d) Ei280 run minus
pre-industrial run (Ice Sheet), respectively. The anomalies are calculated by
averages for the last 50 years of the individual runs. (e–h) Similar to (a–d), but for sea surface temperature
(SST; ∘C). Intervals of
grey contours are 1.5 ∘C.
ResultsGlobal mean warming in Pliocene run
First, we compare global mean surface air temperature (SAT) change between
the simulations. Figure 4 shows the time evolution of the global mean SAT
during the 500-year model integrations. Compared with the pre-industrial run,
identical to the PlioMIP1 control simulation (KU12), all the experiments
show higher global mean SAT. The Pliocene run shows a more stable long-term
trend (year 70–500) than other sensitivity runs prescribed with the OVL
(Eo280) or CO2 (E400). The doubling CO2 experiment (E560) shows a
long-term warming trend and the resultant warming for the last 50 years
accounts for 2.8 ∘C (Table 3). In the E400 run prescribed with
CO2 concentration of 400 ppmv, global mean warming accounts for 1.7 ∘C, which is the largest contributor to the Pliocene warmth among
the boundary conditions (68 %; Table 3), consistent with Willeit et al. (2013). Here Sum (2.5 ∘C) can reconstruct All (2.4 ∘C)
quantitatively. Contributions of Ice Sheet and OVL are 12 and 20 %, respectively.
Compared with the PlioMIP1 run (1.8 ∘C), the PlioMIP2 Pliocene
run shows a larger warming (+39 %, 0.7 ∘C) although the
prescribed CO2 concentration (400 ppmv) is slightly lower (405 ppmv in
PlioMIP1). In the next section, we compare spatial patterns of the results
and decompose respective contributions of the individual boundary
conditions.
Regional changes in temperature and precipitation
Figure 5 shows spatial distributions of annual mean SAT anomalies averaged
over the last 50 years. Similar to the PlioMIP1 run (KU12; Haywood et al.,
2013), the Pliocene anomaly exhibits a polar amplification of surface
warming (i.e. warming peaks over the high latitudes including Greenland and
the Antarctica). While land surface warming is generally larger than the
ocean surface warming (in response to CO2 forcing; Fig. 5b; Manabe et
al., 1991; Kamae et al., 2014), regional differences in SAT change (weak
warming compared with surrounding areas) are found over southern North
America, tropical Africa, Indian subcontinent, and eastern Siberia, similar
to the PlioMIP1 (KU12). Zonally averaged SAT change exhibits (1) minimum
warming over the Southern Hemisphere middle latitude; (2) moderate warming
over the tropics; and (3) warming peaks over the Southern and Northern
Hemisphere high latitudes (Fig. 6a). SST also exhibits the inter-hemispheric
warming asymmetry (Figs. 5e, 6b; see Sect. 3.3) and warming peak in the
Northern Hemisphere mid-to-high-latitude (particularly in the eastern North
Pacific and North Atlantic; Figs. 5e, 6b). Here Sum of zonal mean SAT, SST and
other variables (e.g. precipitation) are similar to All (Fig. 6), indicating a
limited Residual term in the zonal mean (Eq. 6). The effectiveness of reconstruction
of All by Sum suggests that characteristics of the Pliocene climate anomaly can be
decomposed into the individual contributions by Eq. (5). Note that Sum tends to
underestimate (overestimate) the Southern (Northern) Hemisphere middle and
high-latitude warming, suggesting an importance of nonlinear effects (see
Sect. 4).
Compared to the PlioMIP1, the Pliocene run exhibits a larger warming over
Antarctica and the Northern Hemisphere middle and high latitudes (Fig. 6a),
resulting in the larger increase in global mean SAT (Sect. 3.1). In addition
to the less ice sheets over Greenland and West Antarctica, other factors
also contribute to the polar amplification of surface warming (Figs. 5, 6a).
The decomposition based on the sensitivity runs (Sect. 2.3) indicates that
all the boundary conditions contribute to the polar warming (Figs. 5, 6a).
Contributions of OVL and CO2 to the zonal mean polar warming are dominant
(Figs. 5b, c, 6a). Here OVL is the largest contributor to the latitudinal difference
in the Northern Hemisphere warming and the inter-hemispheric warming
contrast (Fig. 6a). Although spatially smooth CO2 radiative forcing
also leads to the polar amplification (Fig. 6a; e.g. Serreze and Barry,
2011; Hill et al., 2014), OVL effect dominates the meridional warming contrast.
Note that CO2 is the largest contributor to the global mean Pliocene warming
(Sect. 3.1). In Sect. 5, we assess reproducibility of the SST in the
PlioMIP2 and 1 runs by comparing with proxy-based estimate.
Zonal-mean anomalies in (a) SAT (∘C), (b) SST
(∘C), (c) precipitation (mm day-1), (d) mass stream function
of mean meridional circulation at 500 hPa level (1010 kg s-1),
(e) surface albedo, (f) sea ice concentration (%), (g) northward heat
transport due to atmosphere (PW) and (h) the Atlantic (PW). Black, yellow,
green, and blue lines represent All, CO2, OVL, and Ice Sheet, respectively. Grey line
represents Sum. Dashed red lines represent results of AOGCM_NFA
run conducted in the identical model in PlioMIP1 (Kamae and Ueda, 2012).
Dotted grey line in (d) represents climatology in pre-industrial run
multiplied by 0.5.
Similar to Fig. 5, but for surface albedo.
Figures 7 and 6e show change in surface albedo and its zonal mean.
Increasing albedo over North America middle latitude and eastern Siberia due
to change in vegetation (boreal forest in modern but grassland in the
Pliocene; Fig. 1; Haywood et al., 2013; Hill et al., 2014), a part of OVL
effect, contributes to the regional difference in the SAT change (Figs. 5c,
7c). Over the Northern Hemisphere high latitude, the reduced ice sheets
(Figs. 1, 3) and a reduction of surface albedo (Figs. 6e, 7c) due to
northward shift of boreal forest (deciduous conifer, tundra and bare soil
regions over northern Canada and northeastern Eurasia; Fig. 1) result in
regional warming (Fig. 5c, d). Surface snow cover (not shown) also affects
partly the land surface albedo. The high-latitude SAT is more sensitive to
imposed albedo change due to the altered vegetation cover (Fig. 1) than low
latitude (Davin and de Noblet-Ducoudré, 2010), consistent with the
substantial Arctic warming found in the current study (Fig. 5) and previous
studies (Willeit et al., 2013; Zhang and Jiang, 2014). Decreasing albedo
over the high-latitude ocean (the Arctic and Antarctic Ocean) corresponds to
sea ice reductions (Fig. 6e, f) that are larger than the PlioMIP1 (Fig. 6f;
Howell et al., 2016). Sea ice and snow albedo feedback contributes to the
differential polar and global-mean warming between the two runs. Over
semi-arid and arid land regions, local SAT change corresponds well with
precipitation change (Fig. 8; e.g. Kamae et al., 2011). OVL leads to increased
precipitation over western North America, tropical Africa, and Indian
subcontinent (Fig. 8c), resulting in surface cooling (Fig. 5c) via changing
surface heat fluxes (not shown).
Similar to Fig. 5, but for precipitation (mm day-1).
The precipitation response in the PlioMIP2 run, dominated by OVL effect, is
generally larger than the PlioMIP1 run (Figs. 6c, 8; Fig. 7g in KU12). The
anomalous middle and high-latitude warming associated with the altered
boundary conditions (e.g. darker land surface due to northward shift of
boreal forest; Fig. 1) could affect the large-scale precipitation pattern
via changing atmospheric circulations (Sect. 3.3). For example, tropical
precipitation associated with Intertropical Convergence Zones (ITCZs) is
sensitive to inter-hemispheric warming asymmetry (e.g. Braconnot et al.,
2007). In the tropical Atlantic, inter-hemispheric warming gradient (warmer
in the North Atlantic than the South Atlantic; Fig. 5g) affects the Atlantic
and African ITCZ precipitation (Fig. 8c; e.g. Zhang and Delworth, 2006). In
addition, regional alterations of land surface condition can also affect
local precipitation. The expansions of the lakes over tropical Africa and
mid-latitude western North America (Fig. 2c) reduce surface sensible heat
flux and enhance local hydrological cycle (e.g. surface evaporation and
precipitation; Pound et al., 2014). The enhanced precipitation over the
semi-arid regions is an important factor for simulating the Pliocene
vegetation pattern (Kamae and Ueda, 2011; Contoux et al., 2013; Pound et
al., 2014). The tropical precipitation changes may also be associated with
changes in regional land–sea temperature contrast and seasonal monsoon
circulations (e.g. R. Zhang et al., 2013).
Meridional overturning circulation
The surface temperature (SAT and SST) shows warming peaks over the middle
and high latitudes and the inter-hemispheric warming asymmetry (Figs. 5, 6a,
b). The warming peaks in the PlioMIP2 are larger than that of the PlioMIP1
(Fig. 6a). Both changes in precipitation (Figs. 6c, 8) and cloud amount (not
shown) show inter-hemispheric asymmetries (larger increase in the Northern
Hemisphere than the Southern Hemisphere), suggesting a change in large-scale
atmospheric circulation (e.g. Kang et al., 2009). Figure 9 shows changes in
atmospheric mean meridional circulation (MMC) determined by mass stream
function (MSF). The MMC, one of the important factors for the Pliocene
climate anomaly (Chandler et al., 1994; Brierley et al., 2009; Brierley and
Fedorov, 2010; Kamae et al., 2011; Sun et al., 2013; Li et al., 2015), shows
a larger change compared with the PlioMIP1 (Fig. 6d). The MMC change is
largely characterized as enhanced (weakened) Southern (Northern) Hemisphere
Hadley cell, northward shift of the tropical Hadley cells, and enhanced
mid-latitude cell over the Northern Hemisphere (Figs. 6d, 9a). The boundary
of the two Hadley cells and northern edge of the northern cell (determined
by signs of MSF at 500 hPa level; Fig. 6d) shift northward for
7.2 and 2.0∘, respectively, and the mid-latitude cell
is enhanced for 41 %, largely according to the OVL effect (Figs. 6d, 9c).
The change in the Hadley circulation is consistent with the PlioMIP1
qualitatively but larger quantitatively, suggesting an anomalous meridional
heat transport due to the MMC (see Sect. 4).
The simulated SST anomaly shows the remarkable meridional warming gradient,
particularly over the Atlantic (Fig. 5), resulting from a substantial
anomaly in the AMOC. Climatological AMOC in the pre-industrial run shown in
Fig. 10 (4.2 Sv) is weaker than observations (e.g. Buckley and Marchall,
2016) and other PlioMIP1 models (Z.-S. Zhang et al., 2013). Note that AMOC
simulated in MRI-CGCM2.3 shown in Z.-S. Zhang et al. (2013) is a result of
simulation with the flux adjustments (KU12). The Pliocene AMOC simulated in
the PlioMIP2 (without any flux adjustments) is quite stronger (+15 Sv)
than the pre-industrial run, suggesting an intensified northward heat
transport due to the Atlantic. The AMOC change dominates in OVL (+14.5 Sv;
Fig. 10c) and enhancement due to CO2 is moderate. Here change in sea surface
density flux over the North Atlantic (Speer and Tziperman, 1992) is one of
possible controlling factors for the OVL-induced AMOC change. In response to
the prescribed boundary conditions, changes in air and surface water
temperature, atmospheric humidity, cloud cover, and surface wind speed can
influence on sea surface heat fluxes (sensible heat flux, latent heat flux,
and longwave and shortwave radiative flux). In addition, changes in river
runoff, sea ice melt, and precipitation minus evaporation can affect sea
surface salinity. These heat and salinity fluxes possibly modulate AMOC
strength in the Pliocene climate. We plan to address physical processes
contributing to the OVL-induced stronger AMOC in a separated paper. In the next
section, we discuss possible factors contributing to the substantial higher
latitude warming found in the Pliocene run.
Anomalies in mass stream function of mean meridional circulation
(109 kg s-1). (a) All (shading). Contours represent climatological
mass stream function in pre-industrial run (±10, 40, 70 109 kg s-1). Solid and dashed contours represent positive and negative
anomalies, respectively. (b) CO2, (c) OVL, and (d) Ice Sheet, respectively.
Anomalies in Atlantic meridional overturning circulation (AMOC;
shading; Sv). Contours represent climatological overturning circulation in
pre-industrial run (±0, 2, 4, 6, 8 Sv). (a) All, (b) CO2, (c) OVL, and
(d) Ice Sheet, respectively.
Mid- and high-latitude warming and meridional heat transport
The PlioMIP2 run shows the larger middle and high-latitude warming over the
Northern Hemisphere (9 ∘C at 75∘ N; Fig. 6a) compared
with the PlioMIP1 run. Hill et al. (2014) evaluated the contributions of
factors for the polar amplification by using eight PlioMIP1 models. Despite
substantial inter-model spreads, strong warming due to reduced surface
albedo was robustly found in all the models and relative contribution of
meridional heat transport (due to the atmosphere and ocean) was minor.
Figure 6g shows anomalous northward heat transport due to the atmosphere.
Over the Southern Hemisphere high latitude and tropics to Northern
Hemisphere middle latitude, anomalous southward heat transport can be found
while the atmospheric heat transport is positive (northward) over the
Southern Hemisphere middle latitude (30–55∘ S). The
tropical southward heat transport is largely consistent with the MMC change
(i.e. the intensified and weakened Southern and Northern Hemisphere Hadley
cells; Figs. 6d, 9). The southward heat transport over the Southern
Hemisphere high latitude contributes to the Antarctic amplification in the
Pliocene run (6–7 ∘C over 70–90∘ S).
Generally, OVL effect dominates and CO2 and Ice Sheet effects are minor to the total change
in the atmospheric heat transport. Over 50–70∘ N,
OVL and Ice Sheet enhance northward heat transport, contributing to the Arctic warming
(70–80∘ N). Here changes in MMC over 50–70∘ N are limited in these experiments (Fig. 6d), implying an
important role of mid-latitude eddies. Changes in meridional temperature
gradient in the upper troposphere and near the surface (e.g. Li et al.,
2015) are possible factors for the anomalous mid-latitude eddy activity
(e.g. Ulbrich et al., 2009). OVL effect contributes to an enhanced (a reduced)
meridional temperature gradient in the upper troposphere (near the surface;
Fig. 6a), similar to results of PlioMIP1 AOGCMs (Figs. 6 and 7 in Li et al.,
2015). Such temperature changes imply a possible intensification of
mid-latitude eddy activity (e.g. Mizuta, 2012). In addition, orography
changes as parts of OVL and Ice Sheet effects (Fig. 3) can also affect mid-latitude
atmospheric circulation and associated meridional heat transport. Note that
nonlinear Residual term is remarkable in the northward heat transport. The northward
heat transport is quite limited in the PlioMIP1 run (Fig. 6g), consistent
with the difference in the high-latitude warming between the PlioMIP2 and 1
(PlioMIP2 run shows stronger warming than PlioMIP1; Fig. 6a).
The simulated AMOC is much stronger than the pre-industrial run (Sect. 3.3),
suggesting a substantial role in the North Atlantic warming during the
Pliocene. Figure 6h shows northward heat transport due to the Atlantic
Ocean. In contrast to divergent responses among the PlioMIP1 models (Z.-S. Zhang et al., 2013), northward heat transport is enhanced substantially over
the Northern Hemisphere (EQ–60∘ N). The enhanced heat transport
is dominated by the OVL effect, consistent with the stronger AMOC found in the
OVL (Fig. 10c). The stronger AMOC is also consistent with the substantial
meridional SST gradient over the Atlantic (Fig. 5e), contributed by OVL and
CO2 (Fig. 5g, f). The enhanced mid-to-high-latitude warming is supported
qualitatively by proxy-based SST reconstruction (Sect. 5).
The anomalous mid-to-high-latitude warming in the PlioMIP2 run is forced by
the altered boundary conditions (Sect. 2.2) and is amplified and/or dampened by
climate feedbacks and anomalous heat transports. As shown above, the impacts
of CO2 forcing and the Greenland ice sheet are relatively limited
(Figs. 5d, 6a, b), suggesting the importance of OVL in the mid-to-high-latitude
warming including the North Atlantic and the eastern North Pacific (Fig. 5g). Note that Sum overestimates the sea ice reduction and surface warming over
the Northern Hemisphere high latitude simulated in All (Fig. 6a, b, f). In
response to the strong external forcing including CO2 and OVL, sea ice
concentration can be 0 % at the edge of sea ice cover in the control
climate (Howell et al., 2016). The limited sea ice concentration in the
pre-industrial run is one of the possible reasons for the nonlinear
relationship between forcing and sea ice reduction. Further analyses of
relative contributions and inter-model consistency of cloud and surface
albedo, longwave radiation, meridional heat transport due to atmosphere and
individual ocean basins by using PlioMIP2 multi-models may contribute to
improving the understanding of the physical mechanisms responsible for the
Pliocene polar amplification.
Data–model comparison of SST
The Pliocene AMOC is apparently distinct from the pre-industrial run (Fig. 10), resulting in the anomalous northward heat transport due to the Atlantic
Ocean (Fig. 6h). Here the larger North Atlantic warming (3–7 ∘C
in 30–70∘ N; Fig. 5e) in the PlioMIP2 Pliocene run
than the PlioMIP1 implies a better reproducibility of the
mid-to-high-latitude SST warming that was robustly underestimated among the
PlioMIP1 multi-models (Dowsett et al., 2013). Figure 11 shows comparison of
simulated SST and PRISM3D proxy-based SST reconstruction during the Pliocene
(Dowsett et al., 2009). Note that the PRISM4 SST reconstruction is not
updated (Dowsett et al., 2016) since PRISM3D. The SST reconstruction is
characterized as extremely high SST in the North Atlantic high latitude,
low-to-mid-latitude warming gradient (limited change in the tropics and
warming in the middle latitude, respectively), and remarkable warming in
mid-latitude coastal areas (off the west coast of North America and South
America, and off the east coast of the Eurasian Continent; Dowsett et al.,
2009, 2013).
(a) SST bias (∘C) in PlioMIP2 Pliocene run compared
with PRISM3D proxy-based SST reconstruction (Dowsett et al., 2009). Coloured
circles and triangles represent cool and warm biases, respectively. Sizes of
plots indicate confidence levels of SST estimate based on chronology,
sampling density, sampling quality and performance of quantitative method
(Dowsett et al., 2013). Thick black and blue open circles indicate proxy
sites in which estimated SST anomalies are large (between 4.0 and 8.9 ∘C) and extremely large (> 8.9 ∘C),
respectively. (b) Comparison of SST biases in Pliocene runs between PlioMIP2
and 1. Coloured plots indicate differences in absolute biases between the
two.
Scatter diagram of proxy-based SST reconstruction (∘C)
and simulated SST (∘C). Red and blue circles represent Pliocene
runs in PlioMIP2 and 1, respectively. Thick black and blue open circles are
identical to Fig. 11. Dashed grey line represents one-by-one line.
Figures 11 and 12 compare SST biases found in the PlioMIP2 and 1. Generally,
both the PlioMIP2 and 1 tend to underestimate the mid-to-high-latitude
warming suggested by proxy records (blue circles in Fig. 11a; Haywood et
al., 2013). However, the large part of underestimation of the
mid-to-high-latitude warming is reduced substantially in the PlioMIP2 run
(green circles in Fig. 11b). In contrast to the remarkable underestimation
(SST bias is larger than 4 ∘C; Fig. 12) of the mid-latitude
warming (the North Atlantic, off the west coast of North America and South
America, and off the east coast of the Eurasian Continent) in the PlioMIP1
(black circles in Figs. 11 and 12), the SST biases are reduced in the
PlioMIP2 run (Figs. 11b and 12). From a zonal-mean perspective, the larger
mid-to-high-latitude warming (Fig. 6a, b) in the PlioMIP2 run is more
consistent with the proxy evidences than the PlioMIP1 (Figs. 11b, 12). Note
that SST bias (8.9–12 ∘C in 69–81∘ N)
over the North Atlantic high latitude (open blue circles in Figs. 11 and 12)
is still not reduced in this simulation. This data–model discord was also
consistently found in PlioMIP1 multiple climate models (Dowsett et al.,
2013). Haywood et al. (2013) noted that this substantial data–model discord
is highly dependent on mean annual temperature estimate based on
geochemically based proxy data and is not derived from faunal-based
estimates of cold–warm month means. They suggested that we should not rely
on this data–model discord too much until more variety of proxy records is
available from more locations in the high-latitude North Atlantic. In
addition to the possible issue on the proxy-based estimate, modelled biases
in AMOC and/or sea ice can also contribute to the North Atlantic data–model
discord.
Summary and discussion
The PlioMIP2 simulations are conducted by using MRI-CGCM2.3 and prescribing
the updated Pliocene palaeoenvironmental dataset, called PRISM4. The Pliocene
climate simulation with the identical model but with slightly revised
boundary conditions from the PlioMIP1 results in the remarkable global-mean
warming with the anomalous mid-to-high-latitude warming. The sensitivity
experiments with swapped boundary conditions can largely reconstruct the
modelled Pliocene climate anomalies, suggesting the linear additivity of the
Pliocene climate simulation. However, linear additivity does not hold so
well for regional climate responses including sea ice reduction over the
high-latitude oceans. The anomalous Northern Hemisphere higher-latitude
warming can be understood as sum of direct response to the external forcing
and associated climate feedbacks. The prescribed external forcing including
CO2, reduced ice sheets, and shortwave absorption due to the Arctic
boreal forest contribute substantially to the higher-latitude warming. In
addition, the anomalous northward heat transport associated with the
large-scale atmospheric circulations and intensified AMOC, and snow and sea
ice albedo feedback are also essential factors. The resultant anomalous
warming over the mid-latitude ocean is more consistent with the proxy data
than the PlioMIP1 simulation. However, the extremely warm condition over the
Arctic to high-latitude North Atlantic region is not reproduced in this
model.
The relative contributions to the polar amplification diverged substantially
among multi-models except those of surface albedo and CO2 (Hill et al.,
2014). In the PlioMIP1, the respective roles of the atmospheric and oceanic
heat transports in the latitudinal warming gradient were not evaluated
sufficiently. The intensified southern Hadley cell and a northward shift of
the tropical cells can be confirmed in most of the PlioMIP1 models (Sun et
al., 2013; Li et al., 2015). Both the PlioMIP1 multi-models and the
MRI-CGCM2.3 PlioMIP2 run show the predominant latitudinal contrast of
surface warming over the Northern Hemisphere and the meridionally asymmetric
change in the Hadley cells. However, the anomalous Hadley circulation
transports heat southward, indicating that the change in the Hadley
circulation is not a factor but can be understood as a result of the change
in atmospheric meridional warming gradient (Li et al., 2015). Further
analyses of the surface processes over land and ocean and three-dimensional
atmospheric and oceanic processes (e.g. oceanic heat transport, atmospheric
heat transport due to mean circulations, stationary eddies, transient
eddies, feedback associated with ice albedo, water vapour, cloud, and lapse
rate; e.g. Serreze and Barry, 2011; Pithan and Mauritsen, 2014; Yoshimori et
al., 2014) are needed to evaluate the respective contributions to the polar
amplified climate in the Pliocene.
The simulated enhancement of the AMOC contributes to the North Atlantic
warming in the Pliocene run. Z.-S. Zhang et al. (2013) revealed that none of
the PlioMIP1 models simulated the substantial enhancement of the AMOC
implied by the proxy records (e.g. Raymo et al., 1996; Robinson, 2009). This
is inconsistent with the current study because Z.-S. Zhang et al. (2013)
introduced the results of flux-adjusted version of the MRI-CGCM2.3 model run
(KU12). The MRI-CGCM2.3 without any flux adjustments simulates the much
enhanced AMOC both in the PlioMIP1 and two settings (Fig. 6h). The current
study points out the importance of OVL effect to the enhanced AMOC, but does
not identify physical processes contributing to the drastic change. We plan
to clarify the physical mechanisms by comparing spatial and vertical
distribution of salinity, heat and fresh water budget at sea surface, and
its role in the AMOC strengths in the Pliocene run. Results of such analyses
will be presented in a separated paper. Recent studies suggested that
oceanic gateways (e.g. the Bering Strait; Hu et al., 2015) and bathymetry
potentially contribute to past warming and/or cooling climate anomalies (e.g. Motoi
et al., 2005; Robinson et al., 2011; Brierley and Fedorov, 2016).
Sensitivity of simulated Pliocene climate to the PRISM4-based
reconstructions of land–sea mask and bathymetry (D16) should be further
evaluated in multi-model frameworks.
Assessment of regional climate properties (e.g. the Asian monsoon; R. Zhang
et al., 2013) in the PlioMIP2 results is one of the remaining issues.
Detailed data–model comparison of the oceanic and terrestrial climate
should also be conducted to evaluate systematic biases in the PlioMIP2 model
ensemble. The model experiment presented in the current study did not
implement changes in the land–sea mask, soil, dynamic vegetation, and
dynamic lakes. Implementation of the proxy-based soil properties as a
boundary condition potentially affects the simulated Pliocene climate via
changing surface and atmospheric energy and water budget (Pound et al.,
2014). Low-resolution models are not suitable for simulating the regional
atmospheric circulation and hydrological cycle associated with land
orography (e.g. Xie et al., 2006). We plan to conduct more complex PlioMIP2
simulations that are more consistent with the proxy-based reconstructions by
incorporating all the requested boundary conditions to an Earth system model
or a high resolution AOGCM. Further complex and fine resolution modelling,
multiple model intercomparison, and data–model comparisons could advance
understanding of the factors for the Pliocene warming climate.
Data availability
All the PlioMIP2/PRISM4 boundary condition data are available from the USGS
PlioMIP2 web page: http://geology.er.usgs.gov/egpsc/prism/7.2_pliomip2_data.html.
Acknowledgements
We thank anonymous reviewers for giving constructive comments. The authors
acknowledge PRISM4 and PRISM3D project members for archiving and providing
global palaeoenvironmental datasets for the Pliocene climate model
simulations. We also thank A. M. Haywood and A. M. Dolan for coordinating
the model intercomparison project, PlioMIP2.
Edited by: A. Haywood
Reviewed by: two anonymous referees
References
Bellenger, H., Guilyardi, E., Leloup, J., Lengaigne, M., and Vialard, J.:
ENSO representation in climate models: from CMIP3 to CMIP5, Clim. Dynam., 42,
1999–2018, 2014.Braconnot, P., Otto-Bliesner, B., Harrison, S., Joussaume, S., Peterchmitt, J.-Y., Abe-Ouchi, A., Crucifix, M.,
Driesschaert, E., Fichefet, Th., Hewitt, C. D., Kageyama, M., Kitoh, A., Loutre, M.-F., Marti, O., Merkel, U.,
Ramstein, G., Valdes, P., Weber, L., Yu, Y., and Zhao, Y.: Results of PMIP2 coupled simulations of the Mid-Holocene and
Last Glacial Maximum – Part 2: feedbacks with emphasis on the location of the ITCZ and mid- and high latitudes heat budget,
Clim. Past, 3, 279–296, 10.5194/cp-3-279-2007, 2007.
Braconnot, P., Harrison, S. P., Kageyama, M., Bartlein, P. J.,
Masson-Delmotte, V., Abe-Ouchi, A., Otto-Bliesner, B., and Zhao, Y.:
Evaluation of climate models using palaeoclimatic data, Nature Climate Change,
2, 417–424, 2012.Brierley, C. M. and Fedorov, A. V.: Relative importance of meridional and
zonal sea surface temperature gradients for the onset of the ice ages and
Pliocene-Pleistocene climate evolution, Paleoceanography, 25, PA2214, 10.1029/2009PA001809, 2010.
Brierley, C. M. and Fedorov, A. V.: Comparing the impacts of
Miocene–Pliocene changes in inter-ocean gateways on climate: Central
American Seaway, Bering Strait, and Indonesia, Earth Planet. Sc. Lett.,
444, 116–130, 2016.
Brierley, C. M., Fedorov, A. V., Lui, Z., Herbert, T., Lawrence, K., and
LaRiviere, J. P.: Greatly expanded tropical warm pool and weakened Hadley
circulation in the early Pliocene, Science, 323, 1714–1718, 2009.
Buckley, M. W. and Marshall, J.: Observations, inferences, and mechanisms of
the Atlantic Meridional Overturning Circulation: A review, Rev. Geophys.,
54, 5–63, 2016.
Chandler, M. A., Rind, D., and Thompson, R. S.: Joint investigations of the
middle Pliocene climate II: GISS GCM Northern Hemisphere results, Global
Planet. Change, 9, 197–219, 1994.Contoux, C., Jost, A., Ramstein, G., Sepulchre, P., Krinner, G., and Schuster, M.: Megalake Chad impact on climate and
vegetation during the late Pliocene and the mid-Holocene, Clim. Past, 9, 1417–1430, 10.5194/cp-9-1417-2013, 2013.
Davin, E. L. and de Noblet-Ducoudré, N.: Climatic impact of global-scale
deforestation: Radiative versus nonradiative processes, J. Climate, 23,
97–112, 2010.Dolan, A. M., Hunter, S. J., Hill, D. J., Haywood, A. M., Koenig, S. J., Otto-Bliesner, B. L., Abe-Ouchi, A., Bragg, F., Chan, W.-L.,
Chandler, M. A., Contoux, C., Jost, A., Kamae, Y., Lohmann, G., Lunt, D. J., Ramstein, G., Rosenbloom, N. A., Sohl, L., Stepanek, C.,
Ueda, H., Yan, Q., and Zhang, Z.: Using results from the PlioMIP ensemble to investigate the Greenland Ice Sheet during the
mid-Pliocene Warm Period, Clim. Past, 11, 403–424, 10.5194/cp-11-403-2015, 2015.
Dowsett, H. J., Cronin, T. M., Poore, R. Z., Thompson, R. S., Whatley, R. C.,
and Wood, A. M.: Micropaleontological evidence for increased meridional heat
transport in the North Atlantic Ocean during the Pliocene, Science, 258,
1133–1135, 1992.Dowsett, H. J., Robinson, M. M., and Foley, K. M.: Pliocene three-dimensional global ocean
temperature reconstruction, Clim. Past, 5, 769–783, 10.5194/cp-5-769-2009, 2009.
Dowsett, H. J., Robinson, M. M., Haywood, A. M., Salzmann, U., Hill, D.,
Sohl, L., Chandler, M., Williams, M., Foley, K., and Stoll, D. K.: The
PRISM3D paleoenvironmental reconstruction, Stratigraphy, 7, 123–139, 2010.
Dowsett, H. J., Robinson, M. M., Haywood, A. M., Hill, D. J., Dolan, A. M.,
Stoll, D. K., Chan, W.-L., Abe-Ouchi, A., Chandler, M. A., and Rosenbloom,
N. A.: Assessing confidence in Pliocene sea surface temperatures to evaluate
predictive models, Nature Climate Change, 2, 365–371, 2012.Dowsett, H. J., Foley, K. M., Stoll, D. K., Chandler, M. A., Sohl, L. E.,
Bentsen, M., Otto-Bliesner, B. L., Bragg, F. J., Chan, W.-L., Contoux, C.,
Dolan, A. M., Haywood, A. M., Jonas, J. A., Jost, A., Kamae, Y., Lohmann,
G., Lunt, D. J., Nisancioglu, K. H., Abe-Ouchi, A., Ramstein, G.,
Riesselman, C. R., Robinson, M. M., Rosenbloom, N. A., Salzmann, U.,
Stepanek, C., Strother, S. L., Ueda, H., Yan, Q., and Zhang, Z.: Sea surface
temperature of the mid-Piacenzian Ocean: A data-model comparison, Sci. Rep.,
3, 2013, 10.1038/srep02013, 2013.Dowsett, H., Dolan, A., Rowley, D., Moucha, R., Forte, A. M., Mitrovica, J. X., Pound, M., Salzmann, U.,
Robinson, M., Chandler, M., Foley, K., and Haywood, A.: The PRISM4 (mid-Piacenzian) paleoenvironmental reconstruction,
Clim. Past, 12, 1519–1538, 10.5194/cp-12-1519-2016, 2016.
Fedorov, A. V., Brierley, C. M., Lawrence, K. T., Liu, Z., Dekens, P. S.,
and Ravelo, A. C.: Patterns and mechanisms of early Pliocene warmth, Nature,
496, 43–49, 2013.
Griffin, D. L.: The late Neogene Sahabi rivers of the Sahara and their
climatic and environmental implications for the Chad Basin, J. Geol. Soc.
Lond., 163, 905–921, 2006.Haywood, A. M., Dowsett, H. J., Otto-Bliesner, B., Chandler, M. A., Dolan, A. M., Hill, D. J., Lunt, D. J.,
Robinson, M. M., Rosenbloom, N., Salzmann, U., and Sohl, L. E.: Pliocene Model Intercomparison Project
(PlioMIP): experimental design and boundary conditions (Experiment 1), Geosci. Model Dev., 3, 227–242, 10.5194/gmd-3-227-2010, 2010.Haywood, A. M., Dowsett, H. J., Robinson, M. M., Stoll, D. K., Dolan, A. M., Lunt, D. J., Otto-Bliesner, B., and
Chandler, M. A.: Pliocene Model Intercomparison Project (PlioMIP): experimental design and
boundary conditions (Experiment 2), Geosci. Model Dev., 4, 571–577, 10.5194/gmd-4-571-2011, 2011.Haywood, A. M., Hill, D. J., Dolan, A. M., Otto-Bliesner, B. L., Bragg, F., Chan, W.-L., Chandler, M. A., Contoux, C.,
Dowsett, H. J., Jost, A., Kamae, Y., Lohmann, G., Lunt, D. J., Abe-Ouchi, A., Pickering, S. J., Ramstein, G.,
Rosenbloom, N. A., Salzmann, U., Sohl, L., Stepanek, C., Ueda, H., Yan, Q., and Zhang, Z.: Large-scale features of
Pliocene climate: results from the Pliocene Model Intercomparison Project, Clim. Past, 9, 191–209, 10.5194/cp-9-191-2013, 2013.Haywood, A. M., Dowsett, H. J., and Dolan, A. M.: Integrating geological
archives and climate models for the mid-Pliocene warm period, Nat. Commun.,
7, 10646, 10.1038/ncomms10646, 2016a.Haywood, A. M., Dowsett, H. J., Dolan, A. M., Rowley, D., Abe-Ouchi, A., Otto-Bliesner, B., Chandler, M. A., Hunter, S. J.,
Lunt, D. J., Pound, M., and Salzmann, U.: The Pliocene Model Intercomparison Project (PlioMIP) Phase 2:
scientific objectives and experimental design, Clim. Past, 12, 663–675, 10.5194/cp-12-663-2016,
2016b.Hill, D. J., Haywood, A. M., Lunt, D. J., Hunter, S. J., Bragg, F. J., Contoux, C., Stepanek, C., Sohl, L.,
Rosenbloom, N. A., Chan, W.-L., Kamae, Y., Zhang, Z., Abe-Ouchi, A., Chandler, M. A., Jost, A., Lohmann, G.,
Otto-Bliesner, B. L., Ramstein, G., and Ueda, H.: Evaluating the dominant components of warming in Pliocene climate
simulations, Clim. Past, 10, 79–90, 10.5194/cp-10-79-2014, 2014.Howell, F. W., Haywood, A. M., Otto-Bliesner, B. L., Bragg, F., Chan, W.-L., Chandler, M. A., Contoux, C., Kamae, Y., Abe-Ouchi, A.,
Rosenbloom, N. A., Stepanek, C., and Zhang, Z.: Arctic sea ice simulation in the PlioMIP ensemble,
Clim. Past, 12, 749–767, 10.5194/cp-12-749-2016, 2016.
Hu, A., Meehl, G. A., Han, W., Otto-Bliestner, B., Abe-Ouchi, A., and
Rosenbloom, N.: Effects of the Bering Strait closure on AMOC and global
climate under different background climates, Prog. Oceanogr., 132, 174–196,
2015.Kageyama, M., Braconnot, P., Harrison, S. P., Haywood, A. M., Jungclaus, J., Otto-Bliesner, B. L., Peterschmitt, J.-Y., Abe-Ouchi, A.,
Albani, S., Bartlein, P. J., Brierley, C., Crucifix, M., Dolan, A., Fernandez-Donado, L., Fischer, H., Hopcroft, P. O.,
Ivanovic, R. F., Lambert, F., Lunt, D. J., Mahowald, N. M., Peltier, W. R., Phipps, S. J., Roche, D. M., Schmidt, G. A.,
Tarasov, L., Valdes, P. J., Zhang, Q., and Zhou, T.: PMIP4-CMIP6: the contribution of the Paleoclimate Modelling Intercomparison
Project to CMIP6, Geosci. Model Dev. Discuss., 10.5194/gmd-2016-106, in review,
2016.Kamae, Y. and Ueda, H.: Evaluation of simulated climate in lower latitude
regions during the mid-Pliocene warm period using paleovegetation data,
SOLA, 7, 177–180, 10.2151/sola.2011-045, 2011.Kamae, Y. and Ueda, H.: Mid-Pliocene global climate simulation with MRI–CGCM2.3: set-up and initial results of PlioMIP
Experiments 1 and 2, Geosci. Model Dev., 5, 793–808, 10.5194/gmd-5-793-2012, 2012.
Kamae, Y., Ueda, H., and Kitoh, A.: Hadley and Walker circulations in the
mid-Pliocene warm period simulated by an atmospheric general circulation
model, J. Meteorol. Soc. Japan, 89, 475–493, 2011.Kamae, Y., Watanabe, M., Kimoto, M., and Shiogama, H.: Summertime land-sea
thermal contrast and atmospheric circulation over East Asia in a warming
climate–Part II: Importance of CO2-induced continental warming, Clim. Dynam., 43, 2569–2583, 2014.
Kang, S. M., Frierson, D. M. W., and Held, I. M.: The tropical response to
extratropical thermal forcing in an idealized GCM: the importance of
radiative feedbacks and convective parameterization, J. Atmos. Sci., 66,
2812–2827, 2009.
Klein, S. A., Zhang, Y., Zelinka, M. D., Pincus, R., Boyle, J., and
Gleckler, P. J.: Are climate model simulations of clouds improving? An
evaluation using the ISCCP simulator, J. Geophys. Res.-Atmos., 118,
1329–1342, 2013.Koenig, S. J., Dolan, A. M., de Boer, B., Stone, E. J., Hill, D. J., DeConto, R. M., Abe-Ouchi, A., Lunt, D. J.,
Pollard, D., Quiquet, A., Saito, F., Savage, J., and van de Wal, R.: Ice sheet model dependency of the simulated
Greenland Ice Sheet in the mid-Pliocene, Clim. Past, 11, 369–381, 10.5194/cp-11-369-2015, 2015.
Li, X., Jiang, D., Zhang, Z., Zhang, R., Tian, Z., and Qing, Y.:
Mid-Pliocene westerlies from PlioMIP simulations, Adv. Atmos. Sci., 32,
909–923, 2015.Manabe, S., Stouffer, R. J., Spelman, M. J., and Bryan, K.: Transient
responses of a coupled ocean-atmosphere model to gradual changes of
atmospheric CO2, Part I: annual mean response, J. Climate, 4, 785–818,
1991.
Masson-Delmotte, V., Schulz, M., Abe-Ouchi, A., Beer, J., Ganopolski, A.,
González Rouco, J. F., Jansen, E., Lambeck, K., Luterbacher, J., Naish,
T., Osborn, T., Otto-Bliesner, B., Quinn, T., Ramesh, R., Rojas, M., Shao,
X., and Timmermann, A.: Information from Paleoclimate Archives, in: Climate
Change 2013: The Physical Science Basis, Contribution of Working Group I to
the Fifth Assessment Report of the Intergovernmental Panel on Climate Change
edited by: Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M. , Allen, S. K., Boschung, J., Nauels, A.,
Xia, Y., Bex, V., and Midgley, P. M., Cambridge University Press, Cambridge,
United Kingdom and New York, NY, USA, 383–464, 2014.Mizuta, R.: Intensification of extratropical cyclones associated with the
polar jet change in the CMIP5 global warming projections, Geophys. Res.
Lett., 39, L19707, 10.1029/2012GL053032, 2012.Motoi, T., Chan, W.-L., Minobe, S., and Sumata, H.: North Pacific halocline
and cold climate induced by Panamanian Gateway closure in a coupled
ocean-atmosphere GCM, Geophys. Res. Lett., 32, L10618,
10.1029/2005GL022844, 2005.
Naish, T., Powell, R., Levy, R.,Wilson, G., Scherer, R., Talarico, F.,
Krissek, L., Niessen, F., Pompilio, M., and Wilson, T.: Obliquity-paced
Pliocene West Antarctic ice sheet oscillations, Nature, 458, 322–328, 2009.
Pithan, F. and Mauritsen, T.: Arctic amplification dominated by temperature
feedbacks in contemporary climate models, Nat. Geosci., 7, 181–184, 2014.
Pollard, D. and DeConto, R. M.: Modelling West Antarctic ice sheet growth
and collapse through the past five million years, Nature, 458, 329–332,
2009.Pound, M. J., Tindall, J., Pickering, S. J., Haywood, A. M., Dowsett, H. J., and Salzmann, U.: Late Pliocene lakes
and soils: a global data set for the analysis of climate feedbacks in a warmer world, Clim. Past, 10, 167–180, 10.5194/cp-10-167-2014, 2014.
Raymo, M. E., Grant, B., Horowitz, M., and Rau, G. H.: Mid-Pliocene warmth:
Stronger greenhouse and stronger conveyor, Mar. Micropaleontol., 27,
313–326, 1996.
Raymo, M. E., Mitrovica, J. X., O'Leary, M. J., DeConto, R. M., and Hearty,
P. J.: Departures from eustasy in Pliocene sea level records, Nat.
Geosci., 4, 328–332, 2011.
Reichler, T. and Kim, J.: How well do coupled models simulate today's
climate?, B. Am. Meteorol. Soc., 89, 303–311, 2008.
Robinson, M. M.: New quantitative evidence of extreme warmth in the Pliocene
Arctic, Stratigraphy, 6, 265–275, 2009.
Robinson, M. M., Valdes, P. J., Haywood, A. M., Dowsett, H. J., Hill, D. J.,
and Jones, S. M.: Bathymetric controls on Pliocene North Atlantic and Arctic
sea surface temperature and deepwater production, Palaeogeogr. Palaeocl.,
309, 92–97, 2011.
Rowley, D. B., Forte, A. M., Moucha, R., Mitrovica, J. X., Simmons, N. A.,
and Grand, S. P.: Dynamic topography change of the eastern United States
since 3 million years ago, Science, 340, 1560–1563, 2013.
Salzmann, U., Haywood, A. M., Lunt, D., Valdes, P., and Hill, D.: A new
global biome reconstruction and data-model comparison for the middle
Pliocene, Glob. Ecol. Biogeogr., 17, 432–447, 2008.
Salzmann, U., Dolan, A. M., Haywood, A. M., Chan, W.-L., Voss, J., Hill, D.
J., Abe-Ouchi, A., Otto-Bliesner, B., Bragg, F. J., Chandler, M. A.,
Contoux, C., Dowsett, H. J., Jost, A., Kamae, Y., Lohmann, G., Lunt, D. J.,
Pickering, S. J., Pound, M. J., Ramstein, G., Rosenbloom, N. A., Sohl, L.,
Stepanek, C., Ueda, H., and Zhang, Z.: Challenges in quantifying Pliocene
terrestrial warming revealed by data-model discord, Nature Climate Change, 3,
969–974, 2013.
Sato, N., Sellers, P. J., Randall, D. A., Schneider, E. K., Shukla, J.,
Kinter, J. L., Hou, Y.-Y., and Albertazzi, E.: Effects of implementing the
simple biosphere model in a general circulation model, J. Atmos. Sci., 46,
2757–2782, 1989.
Schuster, M., Duringer, P., Ghienne, J.-F., Bernard, A., Brunet, M.,
Vignaud, P., and Mackaye, H. T.: The Holocene Lake MegaChad: extension,
dynamic and palaeoenvironmental implications since upper Miocene, 11th
European Union of Geoscience Meeting, Strasbourg, 2001.
Schuster, M., Duringer, P., Ghienne, J.-F., Vignaud, P., Mackaye, H. T.,
Likius, A., and Brunet, M.: The age of the Sahara desert, Science, 311,
821–822, 2006.Seki, O., Foster, G. L., Schmidt, D. N., Mackensen, A., Kawamura, K., and
Pancost, R. D.: Alkenone and boron-based Pliocene pCO2 records, Earth Planet. Sc. Lett., 292, 201–211, 2010.
Sellers, P. J., Mintz, Y., Sud, Y. C., and Dalcher, A.: A simple biosphere
model (SiB) for use within general circulation models, J. Atmos. Sci., 43,
505–531, 1986.
Serreze, M. C. and Barry, R. G.: Processes and impacts of Arctic
amplification: A research synthesis, Global Planet. Change, 77, 85–96,
2011.
Shiogama, H., Stone, D. A., Nagashima, T., Nozawa, T., and Emori, S.: On the
linear additivity of climate forcing-response relationships at global and
continental scales, Int. J. Climatol., 33, 2542–2550, 2013.
Sohl, L. E., Chandler, M. A., Schmunk, R. B., Mankoff, K., Jonas, J. A.,
Foley, K. M., and Dowsett, H. J.: PRISM3/GISS topographic reconstruction: US Geological Survey Data, Series 419, 6 pp., 2009.
Speer, K. and Tziperman, E.: Rates of water mass formation in the North
Atlantic Ocean, J. Phys. Oceanogr., 22, 93–104, 1992.Sun, Y., Ramstein, G., Contoux, C., and Zhou, T.: A comparative study of large-scale atmospheric circulation in the context of
a future scenario (RCP4.5) and past warmth (mid-Pliocene), Clim. Past, 9, 1613–1627, 10.5194/cp-9-1613-2013, 2013.
Ulbrich, U., Leckebusch, G. C., and Pinto, J. G.: Extra-tropical cyclones in
the present and future climate: A review, Theor. Appl. Climatol., 96,
117–131, 2009.
Wang,W.-C., Ling, X.-Z., Dudek, M. P., Pollard D., and Thompson, S. L.:
Atmospheric ozone as a climate gas, Atmos. Res., 37, 247–256, 1995.
Wara, M., Ravelo, A. C., and Delaney, M. L.: Permanent El Niño
conditions during the Pliocene warm period, Science, 309, 758–761, 2005.Willeit, M., Ganopolski, A., and Feulner, G.: On the effect of orbital forcing on mid-Pliocene climate, vegetation
and ice sheets, Clim. Past, 9, 1749–1759, 10.5194/cp-9-1749-2013, 2013.
Xie, S.-P., Xu, H., Saji, N. H., Wang, Y., and Liu, W. T.: Role of narrow
mountains in large-scale organization of Asian monsoon convection, J. Climate,
19, 3420–3429, 2006.
Xie, S.-P., Deser, C., Vecchi, G. A., Collins, M., Delworth, T. L., Hall,
A., Hawkins, E., Johnson, N. C., Cassou, C., Giannini, A., and Watanabe, M.:
Towards predictive understanding of regional climate change, Nature Climate
Change, 5, 921–930, 2015.
Yoshimori, M., Watanabe, M., Abe-Ouchi, A., Shiogama, H., and Ogura, T.:
Relative contribution of feedback processes to Arctic amplification of
temperature change in MIROC GCM, Clim. Dynam., 42, 1613–1630, 2014.
Yukimoto, S., Noda, A., Kitoh, A., Hosaka, M., Yoshimura, H., Uchiyama, T.,
Shibata, K., Arakawa, O., and Kusunoki, S.: Present-day and climate
sensitivity in the meteorological research institute coupled GCM version 2.3
(MRI-CGCM2.3), J. Meteorol. Soc. Jpn., 84, 333–363, 2006.Zhang, R. and Delworth, T. L.: Impact of Atlantic multidecadal oscillations
on India/Sahel rainfall and Atlantic hurricanes, Geophys. Res. Lett., 33,
L17712, 10.1029/2006GL026267, 2006.Zhang, R. and Jiang, D.: Impact of vegetation feedback on the mid-Pliocene
warm climate, Adv. Atmos. Sci., 31, 1407–1416, 2014.
Zhang, R., Yan, Q., Zhang, Z. S., Jiang, D., Otto-Bliesner, B. L., Haywood, A. M., Hill, D. J., Dolan, A. M.,
Stepanek, C., Lohmann, G., Contoux, C., Bragg, F., Chan, W.-L., Chandler, M. A., Jost, A., Kamae, Y., Abe-Ouchi, A.,
Ramstein, G., Rosenbloom, N. A., Sohl, L., and Ueda, H.: Mid-Pliocene East Asian monsoon climate simulated in the
PlioMIP, Clim. Past, 9, 2085–2099, 10.5194/cp-9-2085-2013, 2013.
Zhang, Z.-S., Nisancioglu, K. H., Chandler, M. A., Haywood, A. M., Otto-Bliesner, B. L., Ramstein, G.,
Stepanek, C., Abe-Ouchi, A., Chan, W.-L., Bragg, F. J., Contoux, C., Dolan, A. M., Hill, D. J., Jost, A.,
Kamae, Y., Lohmann, G., Lunt, D. J., Rosenbloom, N. A., Sohl, L. E., and Ueda, H.: Mid-pliocene Atlantic Meridional
Overturning Circulation not unlike modern, Clim. Past, 9, 1495-1504, 10.5194/cp-9-1495-2013, 2013.