Impact of paleoclimate on present and future evolution of the Greenland Ice Sheet.

Using transient climate forcing based on simulations from the Alfred Wegener Institute Earth System Model (AWI-ESM), we simulate the evolution of the Greenland Ice Sheet (GrIS) from the last interglacial (125 ka, kiloyear before present) to 2100 AD with the Parallel Ice Sheet Model (PISM). The impact of paleoclimate, especially Holocene climate, on the present and future evolution of the GrIS is explored. Our simulations of the past show close agreement with reconstructions with respect to the recent timing of the peaks in ice volume and the climate of Greenland. The maximum and minimum ice volume at around 18-17 ka and 6-5 ka lag the respective extremes in climate by several thousand years, implying that the ice volume response of the GrIS strongly lags climatic changes. Given that Greenland's climate was getting colder from the Holocene Thermal Maximum (i.e., 8 ka) to the Pre-Industrial era, our simulation implies that the GrIS experienced growth from the mid-Holocene to the industrial era. Due to this background trend, the GrIS still gains mass until the second half of the 20th century, even though anthropogenic warming begins around 1850 AD. This is also in agreement with observational evidence showing mass loss of the GrIS does not begin earlier than the late 20th century. Our results highlight that the present evolution of the GrIS is not only controlled by the recent climate changes, but is also affected by paleoclimate, especially the relatively warm Holocene climate. We propose that the GrIS was not in equilibrium throughout the entire Holocene and that the slow response to Holocene climate needs to be represented in ice sheet simulations in order to predict ice mass loss, and therefore sea level rise, accurately.

Introduction

Under global warming, melting of the Greenland Ice Sheet (GrIS) is expected to be one of the dominant contributors to future sea level rise [1, 2]. However, uncertainties remain regarding how much climate change will affect the speed and amount of GrIS retreat. Systematic observations cover only a few decades and cannot inform about the long-term response of the GrIS to the ongoing, intensifying warming. The past and future evolution of the GrIS, as either reconstructed or simulated, provides an important insight into the processes which become important on longer time scales.

Ice sheet reconstructions based on indicators of past ice extent and relative sea level data indicate that the GrIS was substantially larger during the onset of the last deglaciation (18–16 kiloyear before present, hereafter referred to as ka) than present [3-6]. In contrast, the GrIS reached a minimum extent at around 6–4 ka after a period of warm climate called the Holocene Thermal Maximum (8 ka) [5, 7–15]. Besides these direct geologically based reconstructions, several modelling efforts on the paleo GrIS have been conducted by using forcing constrained by paleo proxies of temperature, ice extent and relative sea level [16-19]. For instance, several studies [16, 17] simulated the GrIS since the Last Glacial Maximum (LGM). Their results suggest that the GrIS reached a maximum size around 16.5 ka, corresponding to an additional ice volume of 4.6 m to 5.1 m sea level equivalent (SLE) relative to present-day, while a minimum size was approached around 5–4 ka with ice volume being reduced by 0.16 m to 0.17 m SLE relative to present-day. Another extended GrIS simulation [20] covering the past two glacial-interglacial cycles proposed that the GrIS contributed 1.46 m and -2.59 m to global sea level during the last interglacial and LGM, respectively. Under an idealised forcing by prescribing temperature and precipitation following several Greenland ice core records, Nielsen and his colleagues simulated the evolution of the GrIS during the last ten thousand years [18]. Their simulations suggest that the Holocene Thermal Maximum potentially reduced the GrIS to a minimum size of 0.15 m to 1.2 m SLE smaller than present at around 9–7 ka. By forcing the GrIS using improved seasonal temperature reconstruction, another simulation [21] obtained a Holocene minimum in GrIS mass of 0.55 m SLE below present day.

Besides the studies on the past evolution of the GrIS, a great effort has been carried out to predict the future evolution of the GrIS [22-30]. Depending on the greenhouse gases emissions scenarios and numerical model strategies, studies have projected that the GrIS may contribute to 2 cm to 33 cm of sea level rise by the end of 21st century. By integrating the GrIS model over the next millennium, Aschwanden et al [31] illustrated that the GrIS may melt away under the strongest warming scenario.

Many studies find that different initialisation methods affect the future projections of the GrIS [23, 24, 32, 33]. Sensitivity simulations [32] indicate that the short-term GrIS evolution projected by an ice-sheet model is strongly influenced by the initial conditions set in the simulation. Therefore, it is essential for numerical simulations to provide an initial model state that is consistent with the climate forcing [24]. Previously, two major initialisation strategies were widely used, i.e., paleo-spinups and equilibrium-spinups. The paleo-spinup strategy uses the evolution of past climate to force the ice sheet model to present, while the equilibrium-spinup applies observed climate from the present (or pre-industrial) to force the model into an equilibrium state. Previous literature [23] noticed that simulations with paleo-spinup produce results closer to observations than those with equilibrium-spinup.

In this study, we simulate the evolution of the GrIS from the last interglacial to the year 2100 using forcing prescribed by a climate model. The impact of paleoclimate, especially the relatively warm Holocene climate, on present and future evolution of the GrIS is explored by comparing sensitivity simulations that do not contain information of paleoclimate change.

Experiment design

Model setup

The evolution of the GrIS from the last interglacial (125 ka) to 2100 AD is simulated by the Parallel Ice Sheet Model (PISM, version 0.7.3, [34, 35]) with a resolution of 5 km. We employ PISM with a hybrid stress balance of non-sliding shallow ice approximation and the shallow-shelf approximation (SIA+SSA). The ice velocity is determined by a linear superposition of the SIA and the SSA velocities. In areas where the basal sliding is negligible, the SIA dominates the ice velocity, while for the ice shelves and fast flowing regions, the ice flow is primarily controlled by the SSA. We use the Lingle and Clark bed deformation model [36, 37] to simulate the solid earth deformation. By doing so, the deformation of the Earth’s crust and the related regional relative sea level variations are taken into account. The Positive Degree Day (PDD) scheme is used to compute the ice ablation [38]. A constant temperature lapse rate of 5°C km−1 is adopted to account the elevation-induced changes in the near-surface temperature, which also impacts the surface mass balance. The eigen-calving and thickness calving parameterizations are used to determine the calving-front dynamics. Eigen-calving is a simple formula for first-order kinematic contribution to iceberg calving, in which volume loss through calving at the ice front is proportional to the determinant of the strain rate tensor, i.e. the product of its eigenvalues [39]. The thickness calving scheme is a simplified parameterization which removes the ice shelf when its thickness exceeds a certain thickness threshold (i.e., thickness_calving_threshold in PISM). Calving takes place when the ice front meets the threshold from either of these two calving laws.

Considering that the GrIS covered a large portion of the continental shelf of Greenland during the last glacial period [3, 4], the chosen model domain is selected to cover it entirely (as shown in Fig 1). A global sea level reconstruction [40] and basal geothermal heat flux [41] are used as external forcing. These boundary conditions and all other input data are bilinearly interpolated onto the 5 km model grid. The present-day bedrock elevation and ice thickness from the ETOPO1 (Fig 1, [42]) are used as the initial conditions at the very beginning of our simulation, i.e., the last interglacial (125 ka).

Fig 1

The model domain of the ice sheet model.

The model domain of the ice sheet model. The left subpanel shows the ice thickness used to initialise the model (shading) and the contemporary coast line (black contour line) while the right subpanel depicts the corresponding bedrock topography. These data are derived from the ETOPO1 data set [42].

Previous modelling studies of the GrIS, such as [16, 17], use geological data to tune the ice extent and thickness. In these simulations, the ice extent is set to follow these observations by adjusting the grounding line parameterizations and ice extent, but this may cause violations in the ice dynamics of the GrIS. To prevent the ramifications of this kind of tuning of the ice dynamics, we do not artificially constrain the ice extent or thickness, except to approximate present day conditions. The ice sheet develops freely, which is driven solely by the changing climate forcing. We tuned the enhancement factor (sia_e = 5) to get a reasonable ice volume (i.e., close to 7.4 m SLE) for the Pre-Industrial era (i.e., the end of the paleo GrIS simulation).

It should be noted that most of our parameterizations are similar to a previous study [43]. However, we do not downscale the precipitation with respect to elevation changes, because compared to [43], the climate forcing derived from the Alfred Wegener Institute Earth System Model (AWI-ESM) has a relatively higher spatial resolution that resolves a precipitation pattern that resembles observations (Section: Features of climate forcing and S1 Fig in S1 File).

Transient climate forcing strategies

The climate forcing for the GrIS comes from the AWI-ESM [44, 45]. The AWI-ESM consists of the atmosphere and land-surface model ECHAM6.3 [46] coupled with the Finite Element Sea ice-Ocean Model (FESOM 1.4, [47]). The simulations performed in this study used the AWI-ESM with T63 spectral resolution in the atmosphere (∼80 km over the GrIS). Ocean and sea ice were simulated on a mesh with resolution varying from nominal one degree in the interior of the ocean to 1/3° in the equatorial belt and ∼24 km north of 50°N. This configuration of AWI-ESM is rather costly. Hence, it is unrealistic to simulate the entire paleoclimate history to construct a continuous forcing for the simulations of the GrIS. Therefore, we perform six time-slice simulations representing different climate stages (i.e., 127 ka, 21 ka, 9 ka, 8 ka, 6 ka, Pre-Industrial) following the protocol of the Paleoclimate Modelling Intercomparison Project phase 4 (PMIP4, [48, 49]). This protocol describes the modifications of the orbital parameters, concentration of greenhouse gases, topography and ice sheet coverage to simulate past climate. Note that some of the above simulations were created as part of the fourth phase of the Paleoclimate Modelling Intercomparison Project (PMIP4). Detailed information of these simulations can also be found in the online Coupled Model Intercomparison Project Phase 6 (CMIP6) documentation [50-53]. The model intercomparison indicates that AWI-ESM produces reasonable paleoclimate patterns when compared with other PMIP4 models [54-60].

The time-slice climate conditions are interpolated temporarily to obtain a continuous transient paleoclimate forcing. As in [43], the glacial index method [61] is used to prescribe the climate forcing evolution for the simulation from the last interglacial to the Pre-Industrial era. Under such an approach, the atmospheric forcing (i.e., precipitation and near-surface air temperature) from the last interglacial (127 ka) to the LGM (21 ka) is generated as a linear combination between the two corresponding climatic conditions obtained from the AWI-ESM. Here, the NGRIP ice core record sampled from the summit of the GrIS [62] is used as an index to derive the temporal evolution of climate from the last interglacial to the LGM. The glacial index method is widely used to generate transient climate forcing for the paleo ice sheet modelling [43, 63–67].

During the last deglaciation, climate changes, especially those associated with the retreat of the GrIS, are primarily driven by variations of Northern Hemisphere summer temperature. The oxygen isotope signal in the NGRIP ice core, however, is strongly affected by the fluctuation of the Atlantic meridional overturning circulation [68], which is predominantly a winter temperature anomaly [21]. To avoid overestimating summer temperature variations from 21 ka to Pre-Industrial era, the AWI-ESM time-slice simulations of 21 ka, 9 ka, 8 ka, 6 ka, and Pre-Industrial are linearly interpolated to derive a continuous climate forcing. Such linear transient strategy also lies in the fact that we have better temporal coverage of climate simulations after 21 ka, in comparison with the climate snapshots available before 21 ka.

Prior investigations [19, 69–74] revealed that oceanic temperature is an important condition for the evolution of the GrIS, especially during the last deglaciation. To take into account the effects of the ocean, the Greenland weighted-average coastal ocean temperature (upper 100 meters) is used as an oceanic temperature forcing. Unlike the prescribed atmospheric forcing that contains spatial and seasonal variations, we apply a simplified, spatially uniform ocean temperature forcing using the glacial index method, based on the yearly mean ocean temperature between the LGM (−1.43°C) and the Pre-Industrial (−0.19°C) experiments. The ice shelf–ocean interaction scheme introduced by Beckmann and Goosse [75] is applied to the ocean component with a melt factor of 0.01. The sub-shelf ice temperature is set to the pressure melting point and the sub-shelf melt rate is assumed to be proportional to the heat flux from the ocean into the ice.

In order to prescribe the climate forcing for both present and future projections of the GrIS, a transient climate simulation covering the period of 1850–2100 is carried out by the AWI-ESM. This transient climate simulation consists of two parts (i.e., historical transient (1850–2005) and Representative Concentration Pathway 8.5 (RCP8.5) future projection (2006–2100)), following the protocol of the Coupled Model Intercomparison Project, Phase 5 [76]. The 250 years of monthly mean near-surface temperature, precipitation and yearly mean ocean temperature are then used to force our industrial era GrIS simulation between 1850 and 2100.

Ensemble simulations

In addition to the above described GrIS simulation, we perform another five simulations by changing the PISM parameters on sliding (till_effective_fraction_overburden), calving (thickness_calving_threshold), lapse rate (temp_lapse_rate), sub-shelf melting (meltfactor_pik) and ice flow (sia_e) (see Table 1). These parameters are selected because their actual ranges have large uncertainties. The parameter till_effective_fraction_overburden affects the effective pressure in the till. A lower till_effective_fraction_overburden value helps the ice to slide more easily. The parameter thickness_calving_threshold provides the thickness threshold for ice shelf calving. The temp_lapse_rate gives the lapse rate for vertical air temperature downscaling. The meltfactor_pik determines the efficiency of the heat exchange between the ocean water and marine terminating ice. Higher meltfactor_pik promotes more ocean-ice heat exchange. The flow enhancement factor for the shallow ice approximation is given by the parameter sia_e. More detailed information on these parameters can be found in PISM user’s manual [34, 35]. Owing to the gravitational effects of the Laurentide Ice Sheet [6], the magnitude of sea level change around the GrIS is likely smaller than the scaled global mean sea level based on benthic δ18O values [77]. Considering the uncertainties on regional sea level, we perform a simulation by scaling the forcing of global mean sea level [40] with a factor of 0.9.

Table 1

List of ensemble simulations in this study.

Exp Name	Parameter description	Parameter name	Sensitivity simulation	Default simulation
Sliding	basal sliding	till_effective_fraction_overburden	0.0096	0.01
Calving	thickness calving	thickness_calving_threshold	250	200
LapseR	lapse rate	temp_lapse_rate	6	5
MeltF	ocean melt factor	meltfactor_pik	0.009	0.01
SIA	shallow ice approximation	sia_e	4	5
SeaL	sea level forcing	sea level scale	0.9	1

Features of climate forcing

Figs 2 and 3 show the 100 year mean climate patterns from the time slice experiments and the final 10 years (i.e., 2091–2100) of the transient simulation as anomalies relative to Pre-Industrial conditions. Since the climate primarily influences the GrIS by the year-round accumulation of snow and ablation during the summer melt season, we present the summer (JJA) temperatures and yearly mean precipitation for comparison.

Fig 2

The AWI-ESM prescribed summer (June-July-August: JJA) 2m-air temperature in different periods (plotted as anomaly to the Pre-Industrial conditions).

The absolute 2m-air temperature at the Pre-Industrial era is presented in the bottom panel. The contemporary coast line is highlighted in the black line (see Fig 1 for details). The grey contours give the ice sheet geometry used in the climate model.

Fig 3

Same as Fig 2, but for annual mean precipitation.

The spatial pattern of Pre-Industrial summer temperature primarily reflects Greenland’s topography, with temperatures falling well below −20°C around the summit of the GrIS. Temperatures near or above melting point are found around the margins. Over the ice sheet, the highest temperatures prevail in the ablation zones of southwest Greenland (Fig 2) while temperature of the same elevation range are generally lower in the northeast due to the uneven distribution of solar radiation and the effects of the sea ice covered Arctic Ocean (not shown). Compared to the Pre-Industrial summer temperature, the last interglacial period exhibits about ≈4°C higher near-surface air temperature. Temperature anomalies are mostly uniform and follow the large-scale glacial-interglacial climate evolution caused by changing summer insolation [78] and greenhouse gas concentrations. Compared to the Pre-Industrial, we find that the coldest conditions in the 21 ka experiment are about 13°C to 14°C colder than the Pre-Industrial conditions. There are spatially heterogeneous temperature anomalies at 9 ka showing that western Greenland is colder than Pre-Industrial levels, while the rest of the ice sheet is warmer. The snapshot experiments of 8 ka and 6 ka are 1°C to 2°C warmer than Pre-industrial levels. Our climate model simulations well capture the relatively warm Holocene climate as indicated by multiple proxy records [21, 79–83]. From the Pre-Industrial (1850 AD) onward, Greenland summer temperatures rise again at an accelerating rate due to increasing concentrations of greenhouse gases. Under the RCP8.5 scenario, the simulated climate around 2100 over Greenland has warmed by about 6°C relative to Pre-Industrial.

In addition to the changing climate, the spatial temperature anomaly over the GrIS is also influenced by the temporal changes in the ice sheet geometry. More specifically, we have applied a different topographic boundary conditions in the simulations of the 21 ka and 9 ka experiments, while the modern topography is used in the 127 ka, 8 ka, 6 ka, Pre-Industrial and 1850–2100 transient experiments. Compared to the Pre-Industrial, the 21 ka experiment is based on a more extensive ice sheet. Accordingly, the climate of Greenland exhibits pronounced cooling around the margins and only moderate cooling near the modern summit. The 9 ka experiment shows colder conditions in western Greenland, due to the cooling effect caused by the residual Laurentide Ice Sheet (not shown) and differences in local topography. To eliminate the impact from the ice sheet geometry, the temperature anomalies at the 700 hPa tropospheric level are also given in S2 Fig in S1 File. We note that temperature changes with similar amplitude are still visible at the troposphere level above the ice sheet, implying that the large-scale climate changes are the dominant contributors for the simulated temperature anomalies.

Similar to the present-day observed precipitation pattern [84, 85], the annual mean precipitation of our Pre-Industrial experiment exhibits a pronounced maximum along the southeast portion of the GrIS and low accumulation at high altitudes (Fig 3 and S1 Fig in S1 File). Due to the cold climate of the LGM, the 21 ka experiment has the lowest accumulation rates over most parts of the ice sheet. Precipitation intensifies after the LGM. The 9 ka climate is still considerably drier than Pre-Industrial while both the 8 ka and 6 ka climate states almost reach Pre-Industrial accumulation rates. In general, we find the lowest precipitation rates in the coldest climate (i.e., LGM), and the highest precipitation rate in the warmest climate (i.e., 2100 AD in RCP8.5 experiment). During the early and mid-Holocene (i.e., 9 ka, 8 ka, 6 ka), when the climate conditions were not significantly different from the Pre-Industrial, the precipitation anomaly is relatively heterogeneous.

Past evolution of the GrIS with a focus on the Holocene

Fig 4 provides the evolution of the prescribed climate forcing and ice volume of the GrIS from the last interglacial to the Pre-Industrial era. In general, the climate gets colder from the last interglacial through the last glacial period, contributing to increasing the ice volume of the GrIS. The climate warms from the LGM to the Holocene, which causes the the GrIS to retreat. This suggests that the temperature, rather than the snowfall rate, is the primary driver for the evolution of the GrIS. Focusing on the periods of extremes in Greenland’s climate and ice volume, we notice that the recent coldest climate is around 26 ka, followed by a warming afterwards. Despite the warming trend, the GrIS still grows for another nine thousand years, and reaches its maximum volume at around 18–17 ka. The Greenland temperature peaks at around 8 ka while the minimum ice volume is obtained 2–3 thousand years later, at around 6–5 ka. Such a delayed ice volume response to climate change happened in all of the sensitivity simulations (Fig 4), independent of the changes in model parameters.

Fig 4

The evolution of Greenland’s climate and simulated ice volume of the GrIS from the last interglacial to the Pre-Industrial era.

(A): The prescribed forcing of summer (JJA) atmospheric temperature (red line), yearly mean ocean temperature (blue line) and global mean sea level (black line). (B): The simulated ice volume evolution of the GrIS. The coloured lines are the results from individual ensemble members (see Table 1). The thick black line gives the ensemble mean values. The red shadow area marks the time period when Greenland’s climate gets warmer, and the blue patch illustrates the time when Greenland’s climate colds.

Given that the simulated evolution of the GrIS before the Holocene has large uncertainties in terms of initial conditions, sea level forcing and the ocean temperature forcing, we primarily focus our analysis on the results from the Holocene Thermal Maximum (8 ka) onward. During the Holocene, the GrIS is mostly terrestrially terminating (S3 Fig in S1 File, and [86]), suggesting that the atmospheric forcing is the primary driver for the GrIS. As shown in Fig 5, both summer insolation [78] and the Greenland ice core proxy records [62] reveal a cooling trend from the Holocene Thermal Maximum (8 ka) to the Pre-Industrial era, in agreement with our model simulations. Despite the cooling trend starting from 8 ka, the GrIS continues to lose mass until 6–5 ka. The dynamic ice loss slows down and reaches a balance with the surface mass balance around 6–5 ka when the GrIS reaches its minimum size, with an ice volume of 0.20 m SLE smaller than the Pre-Industrial volume. The minimum volume of the GrIS also coincides with the smallest ice extent (S4 Fig in S1 File). After 5 ka, summer cooling and the associated increasing surface mass balance lead to a re-advance of the GrIS, causing expanding ice extent and increasing ice volume. Approaching to the Pre-Industrial era, the surface mass balance is higher than the rate of dynamic ice loss, denoting a background growth of the GrIS. Our simulations imply that the GrIS was not in an equilibrium state at any time during the entire Holocene. There is a background trend of GrIS growth when approaching the Pre-Industrial era.

Fig 5

Evolution of Greenland Ice Sheet from the Holocene Thermal Maximum (i.e., 8 ka) to the Pre-Industrial era.

A, top: Blue line: the oxygen δ 18O isotope record from the NGRIP ice core [62]. Red line: the prescribed forcing of summer (JJA) temperature. Black line: July insolation at 70°N [78]. B, center: Black line: ice volume of the GrIS. Pink line: total ice area of the GrIS. Blue line: floating ice area. C, bottom: Black line: surface mass balance of the GrIS. Blue line: total mass balance, positive indicate mass gain, and vice versa. Red line: rate of dynamic ice loss, i.e., discharge ice flux. As the sub-shelf ice flux and grounded basal mass flux are one order of magnitude smaller than that of the surface mass balance and the dynamic ice loss, the total mass balance of the GrIS is primarily controlled by the surface mass balance and dynamic ice loss. The above results are all based on the ensemble mean of the seven ensemble simulations.

Impact of Holocene climate on the present and future evolution of the GrIS

The above results imply that the ice volume response of the GrIS strongly lags climate changes. Therefore, the current evolution of the GrIS is not only controlled by the present climate change, but is also affected by the climate of the past. In order to investigate the actual impact of paleoclimate on the present and future evolution of the GrIS, we continue our paleo GrIS simulation into the year 2100 under the forcing of an industrial era climate scenario (1850–2100). As shown in Fig 6a, under the forcing of increasing greenhouse gases (black line), the mean Greenland-wide summer temperature rises steadily after the 1900s, and accelerates after the 1970s (red line). Despite the forcing of a warming climate, we still observe growth of the GrIS from the 1850s to the 1970s (Fig 6b, black line). This is primarily because the GrIS is still growing due to the cooling from the mid to late-Holocene. Such growth is reversed around the 1970s when Greenland’s surface temperature experiences a prominent warming. Our simulations show that the mass loss of the GrIS starts around the 1970s and keeps accelerating afterwards due to enhanced anthropogenic warming. Based on the RCP8.5 scenario, the simulated Greenland-wide summer surface temperature increases about 6°C by the year 2100. The warming promotes the disintegration of the GrIS which raises the sea level by 4.5 cm SLE by 2100. Given that the RCP8.5 warming (around 6°C, Fig 2) by 2100 is much stronger than the warming during the mid-Holocene (around 1 degree Celsius, Fig 2), and the 4.5 cm SLE melting of GrIS by 2100 is very likely to be a prelude to further melting. Our preliminary simulation indicates that under the steady climate forcing taken from the RCP8.5 climate state by the year 2091–2100, the entire GrIS will melt within a few millennia (not shown).

Fig 6

The evolution of Greenland’s climate and ice volume during 1850–2100.

(A): Black line: concentration of CO2 forcing in our climate model simulation. Blue line: yearly mean ocean temperature forcing. Red line: Greenland-wide summer temperature anomalies in respect to the Pre-Industrial values (100 years of climatology mean). (B): Anomaly of ice volume of the GrIS with respect to the Pre-Industrial values. The solid lines are the simulations using initial conditions obtained by the paleo-spinup, while the dashed lines are the results using initial conditions obtained by a ten thousand year Pre-Industrial equilibrium-spinup. The coloured lines are the results from individual ensemble members (see Table 1), and the thick black lines are the results of the ensemble mean.

To explore how the GrIS evolves without the impact of Holocene climate change, we conduct another simulation forced by the identical industrial climate (1850–2100) as the above simulation, but initialised with a different strategy. In this simulation, the initial condition is obtained by continuing the paleo GrIS simulation for ten thousand years under the forcing of the Pre-Industrial steady state climate. As described previously (Section: Past evolution of the GrIS with a focus on the Holocene), the GrIS was not in an equilibrium state during the entire Holocene. Approaching the Pre-Industrial era, the GrIS was still adjusting to a relatively cool Pre-Industrial climate in comparison to the previous warmer Holocene climate. Under steady Pre-Industrial climate forcing, the GrIS grows from an ice volume of ∼7.54 m SLE to a quasi-equilibrium state, holding ∼0.10 m SLE more ice after ten thousand years of equilibrium-spinup (Fig 7a). Increased ice volume is primarily found over the northeast (Fig 8a), where the precipitation rate is relatively low. Since drier areas tend to need more time to respond to a cooling climate, a long relaxation under PI (cold) conditions is critical for the northeast to gain mass. The ice thickness increase modelled at the northeast, northwest and west is additionally sustained by the creation of ice shelves that, by exerting resistance into the interior of the ice sheet, prevent ice-mass loss. This buttressing effect is well seen by the slowdown in the velocity modelled in these regions (Fig 6b).

Fig 7

The simulated evolution of the (A, top) ice volume, (B, center) ice temperature (average over entire GrIS) and (C, bottom) dynamic ice loss in the paleo-spinup (solid lines) and the equilibrium-spinup (dashed lines).

All results are given as anomalies with respect to the corresponding Pre-Industrial (i.e., 1850) values. The coloured lines are results from individual ensemble members, and the thick black lines are the ensemble mean.

Fig 8

(A, left) Ice thickness and (B, right) surface ice velocity anomaly between the initial conditions obtained from the equilibrium-spinup and paleo-spinup (equilibrium minus paleo).

Black line represents the coastline. As shown, individual glaciers have distinct responses. Overall, northeastern Greenland contributes to a larger GrIS ice volume in comparison with the that obtained from the paleo-spinup. With respect to the surface ice velocity, large portions of the GrIS show faster ice velocity in the equilibrium-spinup with respect to that obtained from paleo-spinup. Results are the ensemble mean of the seven ensemble members.

Compared to the simulation initialised from the paleo-spinup, the equilibrium-spinup shows almost immediate ice loss after the 1890s, almost one century earlier than the simulations starting from the paleo-spinup (Fig 6). Moreover, the estimated contribution of sea level rise from melting the GrIS is on the order of 6.2 cm SLE by 2100, which is larger than the results from the simulation initialised from the paleo-spinup (black line in Fig 6). Our results indicate that if the late-Holocene background trend of growth of the GrIS is not included, the onset and magnitude of mass loss (and therefore sea level rise) of the GrIS can both be overestimated by an ice sheet model.

To understand the behaviour of the GrIS initialised from different initial conditions, we evaluate the discrepancies in the ice temperature and dynamic ice loss between those two initial condition setups. As shown in Fig 7b, when approaching the Pre-Industrial era, the ice temperature of the GrIS is still increasing, owing to the relatively cold climate. Overall, the equilibrium condition has an 0.4°C higher ice temperature than the condition obtained from paleo-spinup. Moreover, at the end of the equilibrium-spinup, the surface mass balance of the GrIS reaches a quasi-balance with the dynamic ice loss, which is stronger than the paleo-spinup (Fig 7c). The higher dynamic ice loss rate is also expressed by an overall higher surface ice velocity (Fig 8b). Due to these differences in ice temperature and rate of dynamic ice loss, the GrIS simulation with the equilibrium-spinup can overestimate the onset and magnitude of the melt of the GrIS under the forcing of anthropogenic warming.

Discussion

Reconstructing the past evolution of the GrIS is challenging owing to the lack of observations both on temporal and spatial scales. By constraining the ice sheet model using field observations of relative sea level and ice extent, previous studies [5, 16, 17] propose that the GrIS reached its maximum size around 18–16 ka. Paleo proxies of temperature, ice extent and ice elevation [8, 11, 81–83, 87] combined with ice sheet model simulations [16-18] reveal that the GrIS retreated to a minimum size most likely between 7–4 ka.

In the present study, we revisit the past evolution of the GrIS using ice sheet model (i.e., PISM). Our simulations reproduce the timing of both maximum (around 18–17 ka) and minimum (around 6–5 ka) sizes of the GrIS, which are close to the previous reconstructions. We find that the times when the GrIS approached to its maximum and minimum size are several thousand years delayed compared to the times when the Greenland climate reaches its coldest (around 26 ka) and warmest (around 8 ka) conditions. Without any constraints imposed from observations, our simulations illustrate that the maximum size of the GrIS (at around 18–17 ka) is about 2.51 m SLE bigger than today, and the minimum size is around 0.20 m SLE smaller than today during the mid-Holocene. Previous GrIS simulations constrained by observations suggest that during the last deglaciation, the GrIS may hold 4.6 m to 5.1 m SLE more ice relative to present-day [16, 17]. Our simulated magnitudes show discrepancies with the previous studies, likely because the simulations have not been constrained by any observation. In addition, a simplified glacial index method could also contribute to such a discrepancy. We realize that the simulated ice volume of the GrIS varies depending on the choice of model parameters and also the prescribed climate forcing. Temperature reconstruction from boreholes [79], ice cores [21, 80, 81] and lake sediments [82, 83] indicate that the regional Greenland peak summer temperature during the Holocene Thermal Maximum was 2°C to 7°C warmer than the Pre-Industrial era. Our model simulation tends to underestimate this warming with a magnitude of around 1°C. Therefore, the simulated minimum ice volume of the GrIS during the mid-Holocene may be underestimated. With forcing prescribed by ice core proxies, two previous studies [18, 21] obtained a much larger retreat of the GrIS, which contributes to a sea level rise of about 0.15–1.2 and 0.55 m during the mid-Holocene, respectively. Furthermore, we can assume that the response of the GrIS to Holocene climate change is additionally underestimated in our experiments, as the applied PDD scheme does not account for the variations in insolation on orbital time scales. During the last interglacial the change in insolation substantially contributed to surface melt on the GrIS [88, 89]. For the mid-Holocene, estimation according to [90] suggests that the PPD schemes will underestimate the surface melt by approximately 10% due to the neglected effect of insolation. Disregarding quantitative differences in ice volumes at different periods, our simulations and previous studies consistently reveal that the ice volume response of the GrIS lags climate change by millennia, independently of the setup of the ice sheet model and climate forcing.

Our GrIS simulations use an offline prescribed climate to force the GrIS, in which several processes are simplified. First, the paleoclimate evolution is prescribed using an idealised glacial index method, which is unable to account for the spatial inhomogeneity of climate variability. Second, a spatially uniform ocean temperature is adopted to force the GrIS. This neglects the role of regional ocean currents in driving the sub-shelf melt. In addition, our ocean temperature forcing follows the evolution of the NGRIP ice core. In reality, ocean temperature variations may decouple from the atmospheric temperature variations as reconstructed from the Greenland ice core [91]. The coupling between the ice sheet, ocean and atmosphere involves many feedbacks, such as ice shelf-ocean interaction [92-94], ice geometry and precipitation feedback, and meltwater induced ocean circulation feedback [58]. These processes should be improved by a high performance coupled earth system model in the future.

Paleoclimate proxies [7, 11, 62, 80, 82, 95–97] indicate that the Northern Hemisphere summer temperature was warmer during the Holocene than the Pre-Industrial era, primarily due to the fact that the perihelion occurred during the boreal summer. This would logically drive the retreat of the GrIS. Indeed, reconstructions of ice sheet extent [11, 98] and surface elevations [8] indicate a substantial shrinking of the GrIS during the mid-Holocene, which occurs several millennia after the Holocene thermal maximum. Our model results imply that the GrIS was not in an equilibrium state during the entire Holocene, despite a relatively stable Holocene climate. After achieving a minimum extent around 6–5 ka, the GrIS kept growing until the industrial era in response to the progressive summer cooling during the Holocene. This is in agreement with paleo reconstructions [5, 11, 16, 18].

Our simulations hint that the late Holocene growth GrIS was reversed by anthropogenic global warming after the late 20th century. Satellite observations [99] and recent reconstructions of mass balance [100] and margin-position [101] suggest no prolonged mass loss of the GrIS before the 1980s, even though the air temperature had already risen globally [102] and in Greenland [103, 104] for more than a century. This implies that during the 20th century, the GrIS may even have counteracted sea level rise due other factors, such as thermal expansion and melting glaciers. This, consequently, might mean those factors are underestimated. The slow response of the GrIS to climate change highlights the critical role of past climate change on the present and future evolution of the GrIS. Our sensitivity GrIS simulations with different initialisation strategies indicate that the onset and magnitude of the GrIS melting can both be overestimated if the background trend due to the late-Holocene GrIS growth is not taken into account.

Previously, many activities have been carried out to evaluate the impact of initialisation on the projections of ice-sheet evolution, such as the initialisation intercomparison project (initMIP, [33, 105]). In the context of the initMIP, three major initialisation approaches are usually applied: paleo-spinup, equilibrium-spinup, and assimilation of ice sheet geometry and velocity based on observations. Due to the slow response time of the GrIS to climate change, we suggest that the GrIS was not in an equilibrium state around the Pre-Industrial era. There is very likely a background trend of growth of the GrIS approaching to the Industrial era. Therefore, the equilibrium-spinup may not be a good way to initialise the GrIS. Based on our results, we suggest that properly reproducing the ice velocity, or the rate of dynamic ice loss, is necessary to project the future evolution of the GrIS correctly. Thus, the paleo-spinup approach is proposed to be the better method to initialise the GrIS than the equilibrium-spinup. Previous sensitivity tests [23] also found that simulations with paleo-spinup produce results closer to observations than those with an equilibrium-spinup.

If the understanding of the GrIS evolution is only limited to centennial timescales, model projections indicate that the GrIS will probably contribute 5 cm to 33 cm sea level rise under the most extreme warming scenario (RCP8.5) by the end of 21th century [31, 106, 107]. This seems to indicate that the contribution of GrIS melting to sea level rise is only on the order of tens of centimetre, in the worst case. However, fossil plants have been found under the 1.4 km ice in northwestern GrIS, which were dated back to a time between the late Pliocene and early Pleistocene. At that time, the greenhouse gases level was similar or even lower than today [108, 109], hinting that the GrIS may actually be very sensitive to climate change, especially in the context of the equilibrium response. We highlight that the ice volume response of the GrIS strongly lags climate change. Our simulations show that the warming of Greenland during the last deglaciation was reversed around 8 ka, when Greenland’s summer temperature reached its peak value. However, despite cooling in the climate forcing, the GrIS still continuously lost mass for several thousand years from 8 ka to the mid-Holocene. The delayed response reminds us that significant sea level rise may occur irreversibly and last for several millennia even if the ongoing climate warming stops or reverses. Our preliminary simulation into the far future indicates that the GrIS is very likely to melt away under the RCP8.5 climate scenario, if the warming anomaly is kept constant at the 2100 levels. Long-term simulations [31] suggest that Greenland is likely to be ice free within a millennium if there is further warming after 2100 and no action to tackle the anthropogenic warming. Therefore, understanding and predicting the evolution of the GrIS must be extended for a longer period, on the timescale of millennia.

Conclusions

By focusing on the impact of paleoclimate on the present and future evolution of the GrIS, we simulate the GrIS from the last interglacial (125 ka) to 2100 AD by using climate forcing from a comprehensive, coupled climate model. Our model results reveal that the sea level response of the GrIS lagged climate changes by several millennia. This indicates that the observed evolution of the GrIS is not only a result of present climate change, but also affected by paleoclimate, especially the transition from the relatively warm Holocene climate to cooler pre-industrial conditions. We highlight that the GrIS was very likely still growing until the late 20th century due to the background summer cooling from the mid-Holocene to the Pre-Industrial era. Without including such a background trend, the onset and magnitude of the GrIS melt can both be overestimated by an ice sheet model.

(PDF)

Click here for additional data file.

3 Jun 2021

Submitted filename: PLos_ONE_Paper_Response Letter.docx

Click here for additional data file.

8 Jul 2021

PONE-D-21-18343

Impact of paleo climate on present and future evolution of Greenland Ice Sheet

PLOS ONE

Dear Dr. Yang,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The revised version of the manuscript have been seen by one of the previous reviewers and one new reviewer. As you can see below both are overall happy with the manuscript, but reviewer 1 has a major concern regarding the oceanic forcing applied in your study. I largely agree with the reviewer and would like to ask you to revise the manuscript accordingly. Moreover, please answer the other (minor) comments by both reviewers.

Please submit your revised manuscript by Aug 22 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Anna von der Heydt

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please update your submission to use the PLOS LaTeX template. The template and more information on our requirements for LaTeX submissions can be found at http://journals.plos.org/plosone/s/latex

3. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

4. Thank you for stating the following in the Acknowledgments Section of your manuscript:

[This work was supported through grant (Global sea level change since the Mid Holocene: Background trends and climate-ice sheet feedbacks) from the Deutsche Forschungsgemeinschaft (DFG) as part of the Special Priority Program (SPP)-1889 “Regional Sea Level Change and Society” (SeaLevel). C. Rodehacke has been financed through the German Federal Ministry of Education and Research (Bundesministerium f¨ur Bildung und Forschung: BMBF) project ZUWEISS

(grant agreement 01LS1612A) and through the National Centre for Climate Research (NCFK, Nationalt Center for Klimaforskning) provided by the Danish State. H.Y. and X.L are partly funded by the open fund of State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, CAS (SKLLQG1920). Development of PISM is supported by NASA grant NNX17AG65G and NSF grants PLR-1603799 and PLR-1644277.]

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

[This work was supported through grant (Global sea level change since the Mid

Holocene: Background trends and climate-ice sheet feedbacks) from the Deutsche

Forschungsgemeinschaft (DFG) as part of the Special Priority Program (SPP)-1889

`Regional Sea Level Change and Society' (SeaLevel). C. Rodehacke has been

financed through the German Federal Ministry of Education and Research

(Bundesministerium fur Bildung und Forschung: BMBF) project ZUWEISS (grant agreement 01LS1612A) and through the National Centre for Climate Research (NCFK, Nationalt Center for Klimaforskning) provided by the Danish State. H.Y. and X.L are partly funded by the open fund of State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, CAS (SKLLQG1920). Development of PISM is supported by NASA grant NNX17AG65G and NSF grants PLR-1603799 and PLR-1644277.]

Additionally, because some of your funding information pertains to [commercial funding//patents], we ask you to provide an updated Competing Interests statement, declaring all sources of commercial funding.

In your Competing Interests statement, please confirm that your commercial funding does not alter your adherence to PLOS ONE Editorial policies and criteria by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” as detailed online in our guide for authors  http://journals.plos.org/plosone/s/competing-interests.  If this statement is not true and your adherence to PLOS policies on sharing data and materials is altered, please explain how.

Please include the updated Competing Interests Statement and Funding Statement in your cover letter. We will change the online submission form on your behalf.

5. We note that Figures 1, 2, 3, 8, S1, S2 and S3 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a) You may seek permission from the original copyright holder of Figures Figures 1, 2, 3, 8, S1, S2 and S3 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b) If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

6. Please ensure that you refer to Figure 7 in your text as, if accepted, production will need this reference to link the reader to the figure.

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: N/A

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this work the authors present simulations of the Greenland Ice Sheet (GrIS) evolution from the last deglaciation to 2100 performed by means of an ice-sheet – earth system model framework coupled in an offline mode. The authors suggest that, due to the time lag in the ice sheet response to the climate forcing, the 20th century retreat still shows some reminiscence of the late Holocene cooling. This also applies to future mass-loss projections, meaning that a paleoclimate spinup is required not to overestimate the future ice-sheet contribution to sea-level (SL) rise.

The experiment conducted here and the results achieved are certainly interesting. However, I believe there are still one major concern that the authors did not completely solve after the first round of review.

As other reviewers already pointed out, the oceanic forcing implemented in this work is inappropriate, as considering constant oceanic conditions since the last deglaciation is really unrealistic. In this experimental setup, the coastal retreat is only ensured by sea level changes through the flotation criterion, while any ocean-induced melting at the ice-water interface is neglected. The lack of evidence suggesting an ice shelf as large as those of Antarctica, as the author state, does not guarantee that subshelf melting played a minor role in the retreat. In fact, it’s plenty of work showing 1. the crucial role of the ocean in the past retreat since the last deglaciation (e.g. Jennings et al., 2017, Syring et al., 2020), but also 2. the importance of modelling oceanic changes to well simulate the past (e.g. Bradley et al., 2018, Tabone et al., 2018), 3. the current state (e.g. Rignot et al., 2016, Cowton et al., 2018, Morlighem et al., 2016) and 4. the future (Slater et al., 2020, Goelzer et al., 2020; Choi et al., 2017) of the Greenland ice sheet. Experiments such as ISMIP6 (Goelzer et al., 2020), yet with simple oceanic melt parametrisations, take into account oceanic changes in future projections, although the current percentage of the GrIS in contact with the ocean will likely decrease more.

Since this study focuses on past and future retreat simulations, I would strongly suggest to consider at least a time-dependent oceanic temperature To to capture the oceanic variability at regional spatial scales. This would mean repeating the experiments, but since the authors have already expressed their unwillingness to do it, this issue should be at least discussed thoroughly as it’s a critical flaw of the research. I don’t think this is properly done in the current version of the manuscript.

Another (minor) concern is related to the climate forcing strategy applied to the last 21ka. Why not using the same climatic index applied between 128 ka to 21 ka? In this way, abrupt events such as the Bolling Allerod could be perceived by the climatic forcing and reflected in the ice-sheet retreat. Are the authors sure that a linear interpolation until the HTM does not affect the climatic background that the GrIS would be still subjected to? I just found strange that such attention to the millennial variability is reserved for the spinup and then neglected for the most interesting part of the experiment.

List of minor changes (only those I picked, please read the manuscript carefully to avoid typos/english mistakes):

Title: please, change it to “Impact of paleo climate on present and future evolution of the Greenland Ice Sheet”

Abstract: line 7, “… the climate of Greenland”.

Age: I’d rather see “26 ka BP” or “26 ka ago” or “-26 ka” instead of “26 ka”. Please, check the journal policy on that.

Author summary: line 11, “of the GrIS is not only controlled...”

Line 9 (also further in the text): extent not extend

Line 36: “Considering that the available studies focus on the evolution of the GrIS either in

the past or into the future, the impact of past climate on present and future evolution

of the GrIS is not well known”. This might be true, but the authors are completely bypassing the fact that most of work of stand-alone ice sheet models cited here does proper glacial spinups to perform future projections (e.g. Applegate et al., 2012, Yan et al., 2013, Calov et al., 2018, ... and also Fuerst et al., 2015, Rueckamp et al. 2019, not cited) thus there is a certain “paleoclimatic background” in those future simulations. Please, add a small paragraph on this, especially since it’s related to your conclusions. Moreover, I found the introduction pretty short for the work you are presenting. Please, consider adding some relevant work focusing on the past (e.g. Buizert et al., 2018, Tabone et al., 2018) and the future (e.g. Fuerst et al., 2015, Rueckamp et al. 2019) evolution of the ice sheet.

Line 38: change to “we simulate the evolution...”

Line 40 “which do not contain...”

Line 59: “It should be noted…”

Section 1.3: I suggest to precisely describe the parameters perturbed in the ensemble simulation instead of putting the PISM user’s manual reference. Which is the “SIA” parameter, for instance? Which is the basal sliding parameter considered? I think this would be much interesting for a broad audience of ice-sheet modellers than simply look at the parameter name and try to guess what it means.

Line 113: “we perform”

Line 162: “the 21 ka experiment features...”

Line 268: “model simulations reveal...”

Line 311: “the GrIS has undergone...”

References

Applegate, P. J., Kirchner, N., Stone, E. J., Keller, K., & Greve, R. (2012). An assessment of key model parametric uncertainties in projections of Greenland Ice Sheet behavior. The Cryosphere, 6(3), 589-606.

Bradley, S. L., Reerink, T. J., Van De Wal, R. S., & Helsen, M. M. (2018). Simulation of the Greenland Ice Sheet over two glacial–interglacial cycles: investigating a sub-ice-shelf melt parameterization and relative sea level forcing in an ice-sheet–ice-shelf model. Climate of the Past, 14(5), 619-635.

Buizert, C., Keisling, B. A., Box, J. E., He, F., Carlson, A. E., Sinclair, G., & DeConto, R. M. (2018). Greenland‐wide seasonal temperatures during the last deglaciation. Geophysical Research Letters, 45(4), 1905-1914.

Calov, R., Beyer, S., Greve, R., Beckmann, J., Willeit, M., Kleiner, T., ... & Ganopolski, A. (2018). Simulation of the future sea level contribution of Greenland with a new glacial system model. The Cryosphere, 12(10), 3097-3121.

Choi, Y., Morlighem, M., Rignot, E., Mouginot, J., & Wood, M. (2017). Modeling the response of Nioghalvfjerdsfjorden and Zachariae Isstrøm Glaciers, Greenland, to ocean forcing over the next century. Geophysical Research Letters, 44(21), 11-071.

Cowton, T. R., Sole, A. J., Nienow, P. W., Slater, D. A., & Christoffersen, P. (2018). Linear response of east Greenland’s tidewater glaciers to ocean/atmosphere warming. Proceedings of the National Academy of Sciences, 115(31), 7907-7912.

Fürst, J. J., Goelzer, H., & Huybrechts, P. (2015). Ice-dynamic projections of the Greenland ice sheet in response to atmospheric and oceanic warming. The Cryosphere, 9(3), 1039-1062.

Goelzer, H., Nowicki, S., Payne, A., Larour, E., Seroussi, H., Lipscomb, W. H., ... & van den Broeke, M. (2020). The future sea-level contribution of the Greenland ice sheet: a multi-model ensemble study of ISMIP6. The Cryosphere, 14(9), 3071-3096.

Jennings, A. E., Andrews, J. T., Cofaigh, C. Ó., Onge, G. S., Sheldon, C., Belt, S. T., ... & Hillaire-Marcel, C. (2017). Ocean forcing of Ice Sheet retreat in central west Greenland from LGM to the early Holocene. Earth and Planetary Science Letters, 472, 1-13.

Morlighem, M., Bondzio, J., Seroussi, H., Rignot, E., Larour, E., Humbert, A., & Rebuffi, S. (2016). Modeling of Store Gletscher's calving dynamics, West Greenland, in response to ocean thermal forcing. Geophysical Research Letters, 43(6), 2659-2666.

Rignot, E., Xu, Y., Menemenlis, D., Mouginot, J., Scheuchl, B., Li, X., ... & Fleurian, B. D. (2016). Modeling of ocean‐induced ice melt rates of five west Greenland glaciers over the past two decades. Geophysical Research Letters, 43(12), 6374-6382.

Rückamp, M., Greve, R., & Humbert, A. (2019). Comparative simulations of the evolution of the Greenland ice sheet under simplified Paris Agreement scenarios with the models SICOPOLIS and ISSM. Polar Science, 21, 14-25.

Syring, N., Lloyd, J. M., Stein, R., Fahl, K., Roberts, D. H., Callard, L., & O'Cofaigh, C. (2020). Holocene Interactions Between Glacier Retreat, Sea Ice Formation, and Atlantic Water Advection at the Inner Northeast Greenland Continental Shelf. Paleoceanography and Paleoclimatology, 35(11), e2020PA004019.

Slater, D. A., Felikson, D., Straneo, F., Goelzer, H., Little, C. M., Morlighem, M., ... & Nowicki, S. (2020). Twenty-first century ocean forcing of the Greenland ice sheet for modelling of sea level contribution. The Cryosphere, 14(3), 985-1008.

Tabone, I., Blasco, J., Robinson, A., Alvarez-Solas, J., & Montoya, M. (2018). The sensitivity of the Greenland Ice Sheet to glacial–interglacial oceanic forcing. Climate of the Past, 14(4), 455-472.

Yan, Q., Zhang, Z., Gao, Y., Wang, H., & Johannessen, O. M. (2013). Sensitivity of the modeled present‐day Greenland Ice Sheet to climatic forcing and spin‐up methods and its influence on future sea level projections. Journal of Geophysical Research: Earth Surface, 118(4), 2174-2189.

Reviewer #2: Review of “Impact of paleo climate on present and future evolution of Greenland Ice Sheet” by Tijn Berends

General comments

The authors present a revised version of a manuscript describing simulations of the Greenland ice sheet throughout the last glacial cycle and into the near future, using the ice-sheet model PISM forced with output of the AWI-ESM climate model. They find that, in agreement with the general consensus, changes in ice volume lag changes in climate by up to several millennia. Given that the Earth’s climate, particularly in Greenland, is thought to have been relatively warm during the early Holocene, and to have cooled to pre-industrial levels during the last few thousand years, this means that the Greenland ice sheet was likely not in equilibrium, but rather was still advancing when anthropogenic climate change commenced. The authors show that this likely delayed the onset of ice-sheet retreat until several decades after anthropogenic warming started, and is even now leading to slower ice-sheet retreat than would have been the case if the Holocene climate had been completely stable. These are important findings, as they highlight the need to account for the paleoclimatic history of the ice-sheet when projecting its future evolution.

In response to comments raised in the first review round, the authors significantly altered the manuscript. They included results from several additional experiments intended to investigate the sensitivity of their results to different model parameters, which support their conclusion that the observed lag in ice volume change with respect to climate is robust. They also altered the general story of the manuscript to focus more on the impact of (recent) paleoclimate change on near-future ice-sheet evolution, which has improved the readability of the manuscript. Lastly, they adequately addressed my major concern from the first review, which was unfortunately based on a misunderstanding on my part. I think the manuscript is now fit to be published after some minor revisions.

Specific comments

Line 35: “Considering that the available studies focus on the evolution of the GrIS either in the past or into the future… ” This is phrased too strongly. Particularly the recent initMIP-GRL paper by Goelzer et al. (2018) presents a detailed study of the effect of model initialization (including paleo spin-up) on future projections. Mention this.

Line 43 (what happened to the line numbers here): “we select the Lingle and Clark bed deformation model…” I now understand the difference between your prescribed sea-level forcing and the dynamically calculated bed deformation. My previous criticism was based on a misunderstanding of this on my part, for which I apologise. I agree that the eustatic signal plus local bedrock deformation is probably sufficient for Greenland.

Line 43 (idem): “A constant temperature lapse rate of 5 ◦C km−1 is adopted…” What is this based on? It seems a little low to me (between 6 – 8 K/km is common I think).

Line 43 (idem): “The eigencalving and thickness calving parametrizations are used to determine the calving-front dynamics…” These are different approaches. Do you mean you use both of them together?

Line 57: “We tune the enhancement factor…” This requires more explanation. Do you do this in a steady-state set-up? Or do you tune it to give good results at the end of your LGC run? If so, do you also tune for LGM volume/extent in any way?

Line 60: “…we do not use the elevation induced precipitation downscaling, because…[] …our climate model has a higher spatial resolution…” Is this spatial resolution smaller than the change in ice margin position? Particularly in the south-west, where the ice margin migrates onto the continental shelf during the LGM (presumably, actually a figure showing a few time-slices of your modelled ice-sheet geometry would be very nice!), you’d expect the high-precipitation area to migrate along with it. If you don’t correct for this in some way, this might introduce a bias in your mass balance.

Line 76: “…following the protocol of the Paleoclimate Modelling 76 Intercomparison Project phase 4…” Did AWI-ESM participate in PMIP4? How do your paleoclimate time slices compare?

Line 94: “From 21ka to Pre-Industrial era, the AWI-ESM 94 time-slice simulations … are linearly 95 interpolated to derive a continuous climate forcing” Looking at Fig. 4A, this seems to introduce a warm bias during the deglaciation. Do you think this might affect your results?

Table 1: I think you changed the SIA flow enhancement factor? Why not also the SSA factor?

Section 2 in general: maybe consider summarizing this. Your article seems to be aimed mostly at ice-sheet modellers such as myself. I don’t really know enough about the details of paleoclimate at specific points in time to know what to make of this paragraph. Really all I need to know is that your paleoclimate is good. If AWI-ESM participated in PMIP4/PMIP5 (while you refer to the PMIP protocol earlier, you don’t mention if AWI-ESM participated or not) then just mentioning that would already be enough.

Line 191: “…suggesting that the surface mass balance is the primary driver for the GrIS.” Given that at present, the GrIS loses most of its mass through glacier discharge (e.g. King et a., 2018; Fettweis et al., 2020) rather than runoff, I’d phrase this differently.

Line 197: “…with an ice volume of 0.24 m SLE smaller than the Pre-Industrial volume.” I’d love to see a figure of this minimum extent.

Line 232: “…In this simulation, the initial condition is obtained by continuing the paleo GrIS simulation for ten thousand years.” In my opinion, the difference between this simulation and the “default” run is your most important result; this illustrates very clearly why including a paleo-spin-up in future projections is so important. I’d mention this one already in the introduction, and consider rearranging the text of your results to present it accordingly (as in, you’ve done your default run and this one, plus a small ensemble of model parameter sensitivity runs for both experiments, so group them like that).

Line 268: “…reveals that the 268 GrIS may retreat…” Mind your grammar.

Line 281: “Our simulated magnitudes show discrepancies with the previous studies, likely because the simulations have not been constrained by any observation.” Elaborate on this. Which previous studies are these? What ice volumes did they find?

Line 334: “Based on our results, we suggest that properly reproducing the ice velocity, or the rate of dynamic ice loss, is necessary to project the future evolution of the GrIS correctly. Thus, the paleo-spinup and assimilation approaches are proposed to be better methods to initialise the GrIS.” You’ve not done anything with data assimilation in your study, so this statement does not belong here. Whether or not a data-assimilated model will implicitly contain the information from the paleo-history of the ice-sheet is a question far beyond the scope of your study.

Line 345: “At that time the greenhouse gases level was similar or even lower than today…” CO2 proxies that far back have uncertainties of over 50 ppmv in either direction, and data points for the Late Pliocene / Early Pleistocene cover pretty much everything from 250 to 450 ppmv (see e.g. Berends et al. 2021, Clim. Past.). Also, you’re now comparing centennial-scale sea-level rise to equilibrated ice volumes – which, as is pretty much the entire point of your study, are not at all the same.

Throughout the manuscript: mind the difference in spelling between the verb “to extend” and the noun “extent”.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Tijn Berends

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Aug 2021

All the response can be found in the attached response letter

Submitted filename: Response_Letter.docx

Click here for additional data file.

17 Sep 2021

PONE-D-21-18343R1Impact of paleoclimate on present and future evolution of the Greenland Ice SheetPLOS ONE

Dear Dr. Yang,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

One of the reviewers has seen the revision again and finds the new version adequately addresses the raised issues. The reviewer has only minor suggestions left. Please address those in a new version of the manuscript.

Please submit your revised manuscript by Nov 01 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Anna von der Heydt

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

One of the reviewers has seen the revision again and finds the new version adequately addresses the raised issues. The reviewer has only minor suggestions left. Please address those in a new version of the manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Yang et al. investigate the influence of the past climate of the Greenland Ice Sheet (GrIS) on its future projections of mass loss. By carrying out ice-sheet-earth-system models simulations from the last glacial period to the near future (2100 AD), they suggest that a classic long paleoclimatic spinup is needed to prevent an overestimation of predictions of Greenland ice loss.

I am glad the authors decided to redo the simulations to include the ocean as an additional forcing to the ice sheet model. I appreciate the authors' effort in this sense as I understand how this might be a considerable investment of resources, especially when dealing with paleoclimatic runs.

As the authors show in the revised manuscript, the inclusion of a spatially-uniform transient oceanic temperature forcing does not have a significant effect on the simulated GrIS evolution. I agree that this might be related to the simplistic approach adopted here. In fact, subshelf temperatures at depths lower than 100 m are likely more suitable to be considered, as closer to the ice shelf base. At those depths, the ocean water is decoupled from the surface air temperature, thus a transient oceanic temperature that follows the NGRIP time series is probably not a good representation of the paleoclimatic water temperature trend. An example can be found in southeast Greenland where warmer subsurface currents during the Younger Dryas, in clear contrast with the simultaneous atmospheric cooling, led to a significant coastal mass loss (Rainsley et al., 2018). Moreover, ocean temperatures around Greenland are strongly affected by local currents which, of course, cannot be captured if their spatial average is taken. However, since increased atmospheric temperature driven by insolation changes is likely the main responsible for the GrIS retreat during the Holocene (Vasskog et al., 2015), I would not expect that a more sophisticated approach regarding the ocean would drastically change the main results of this work, especially regarding the latest part of the Holocene. I therefore thank the authors for their effort made in satisfying my requests and I suggest the publication of the manuscript after minor revisions.

Specific comments

Abstract, line 8: please change to "maximum and minimum ice volume".

Author summary

- line 3: please change to "it remains uncertain on how fast and how much the GrIS will contribute to it".

- line 8: I don't understand this sentence very well: "These results are consistent with evidence of the times of both past extremes in climates...". Please rephrase.

- line 14: please change "growing" to "growth".

Main text:

Line 39: The short term response of ice sheet models to what? Please explain.

Line 64: This is imprecise as the PDD is an ablation scheme only. Please, correct.

Line 71: Please change to "at the ice front".

Line 75: Please change to "continental shelf of Greenland". This error repeats several times. Choose either "of Greenland" or "of the Greenland Ice Sheet".

Line 101: please change to "coupled with the".

Line 104: I think it is sufficient to write the acronym.

Line 126: please change to "the index to simulate".

Line 139: Which "situation"? Please consider changing it to "climate snapshots available before 21 ka" or something alike.

Line 143-145. I would be careful with this sentence. The parametrisation you are using (Beckmann & Goose, 2008) is built for the ice shelf base as no lateral fracturing is considered. Grounding lines in Greenland are usually deeper in the water column (Wilson et al., 2017), which contrasts your choice of considering an oceanic temperature at 100 m. I would add a brief sentence in the discussion regarding this and the possible influence that this might have on your results (see my general comments).

Line 149: I guess you wanted to put "PISM" instead of "PIK".

Line 162: I would not call them "ensemble simulations" as they are rather sensitivity runs. This is misleading as it may be understood as five sets of simulations, while here we have only five runs. I would prefer to read it as "five simulations" or "five sensitivity tests" but I leave it to the authors' discretion.

Line 176: "one more ensemble simulation" as above.

Line 195: what do you mean with "there are spatially heterogeneous anomalies at 9 ka"? Do you want to point out that the west Greenland is cooler than the Pre-industrial level, while the rest of the ice sheet is warmer? Please explain.

Line 197: "This implies a long term cooling trend ...". The sentence before does not allow to get to this conclusion. Please, rephrase.

Line 214: Figure S1 maybe?

Line 219: Again, Fig. 6 is likely wrong.

Line 220: Was the acronym LGM presented before in the text?

Line 228: "is relatively heterogeneous" maybe?

Line 241: Please change to "the minimum ice volume".

Line 249: Please check the figure numbers.

Line 282-284: Figure 2?

Line 298-300: I don't understand why the northeast has a higher volume if the precipitation rate is lower. Could you explain it better?

Line 310: is this ice surface temperature or is it averaged over the ice column?

Line 312: Would you mean "the ice temperature of the GrIS is still increasing ..."?

Line 406-415: This paragraph needs some rephrasing in my view. Contributing to 33mm to sea level rise by 2100 does not imply that the GrIS is rather little sensitive to climate change. This is a subjective point of view.

References:

Beckmann, A., & Goosse, H. (2003). A parameterization of ice shelf–ocean interaction for climate models. Ocean modelling, 5(2), 157-170.

Rainsley, E., Menviel, L., Fogwill, C. J., Turney, C. S., Hughes, A. L., & Rood, D. H. (2018). Greenland ice mass loss during the Younger Dryas driven by Atlantic Meridional Overturning Circulation feedbacks. Scientific reports, 8(1), 1-9.

Vasskog, K., Langebroek, P. M., Andrews, J. T., Nilsen, J. E. Ø., & Nesje, A. (2015). The Greenland Ice Sheet during the last glacial cycle: Current ice loss and contribution to sea-level rise from a palaeoclimatic perspective. Earth-Science Reviews, 150, 45-67.

Wilson, N., Straneo, F., & Heimbach, P. (2017). Satellite-derived submarine melt rates and mass balance (2011–2015) for Greenland's largest remaining ice tongues. The Cryosphere, 11(6), 2773-2782.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

8 Oct 2021

All response are included in the response letter.

Submitted filename: PONE_response_letter.docx

Click here for additional data file.

27 Oct 2021

Impact of paleoclimate on present and future evolution of the Greenland Ice Sheet

PONE-D-21-18343R2

Dear Dr. Yang,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Anna von der Heydt

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for submitting a revised version. I can see that you have addressed all remaining issues of the reviewer and I am happy to accept the manuscript for publication in PLOS One

Reviewers' comments:

29 Oct 2021

PONE-D-21-18343R2

Impact of paleoclimate on present and future evolution of the Greenland Ice Sheet

Dear Dr. Yang:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Anna von der Heydt

Academic Editor

PLOS ONE