We present a reconstruction of historical (1950–2014) surface mass
balance (SMB) of the Greenland ice sheet (GrIS) using
a high-resolution regional climate model (RACMO2; ∼11km) to dynamically downscale the climate of the
Community Earth System Model version 2 (CESM2; ∼111km). After further statistical downscaling to
1 km spatial resolution, evaluation using in situ SMB
measurements and remotely sensed GrIS mass change shows good
agreement. Comparison with an ensemble of previously conducted
RACMO2 simulations forced by climate reanalysis demonstrates that
the current product realistically represents the long-term average
and variability of individual SMB components and captures the
recent increase in meltwater runoff that accelerated GrIS mass
loss. This means that, for the first time, climate forcing from an
Earth system model (CESM2), which assimilates no observations, can be
used without additional corrections to reconstruct the historical
GrIS SMB and its recent decline that initiated mass loss in the
1990s. This paves the way for attribution studies of future GrIS
mass loss projections and contribution to sea level rise.
Introduction
A common approach to project the future surface mass balance (SMB)
of the Greenland ice sheet (GrIS) is to force a regional climate
model (RCM), typically running at 5 to 10 km horizontal
resolution, at the lateral and top boundaries with the outputs of
an Earth system model (ESM; ∼100km)
. However, ESMs from the
fifth phase of the Climate Model Intercomparison Project (CMIP5) do
not accurately represent the contemporary large-scale climate of
the Greenland region . The reason is that
ESMs neither assimilate nor prescribe climatic observations as
global climate reanalyses and RCMs do
. For instance,
ESMs fail at capturing the recent summertime Arctic atmospheric
circulation change , making projections of GrIS
mass loss and contribution to sea level rise highly uncertain
. Consequently, climate forcing from CMIP5 ESMs
still requires dedicated bias correction before being used to force
RCMs over the GrIS . An
alternative approach is to directly use outputs of ESMs to estimate
GrIS SMB; however, most ESMs do not have (sophisticated) snow
models that consider meltwater retention in firn, and their coarse
spatial resolution does not accurately resolve the large SMB
gradients at the GrIS margins .
Here, we use the historical climate (1950–2014) of a CMIP6 model,
the Community Earth System Model version 2.1 (CESM2; ∼111km), to force the lateral and top boundaries of the
Regional Atmospheric Climate Model version 2.3p2 (RACMO2; ∼11km). The reason for selecting CESM2 as climate forcing
for RACMO2 stems from the active involvement of the Institute for
Marine and Atmospheric research Utrecht (IMAU) in the development
and improvement of the model for studies over both the Greenland
and Antarctic ice sheets. To obtain a meaningful comparison with in
situ observations, the resulting SMB field is then statistically
downscaled to 1 km over the GrIS and peripheral glaciers
and ice caps (Fig. a) . We show that,
without additional corrections, CESM2 climate forcing yields
a realistic reconstruction of historical GrIS SMB (1950–2014),
including its recent decline in the 1990s. This is unexpected for
an ESM which exclusively prescribes greenhouse gas (CO2
and CH4) and aerosol emissions and may herald more
accurate projections of GrIS contribution to future sea level
rise. Section 2 describes CESM2 and RACMO2, including model
initialisation, forcing set-up, as well as observational and model
data sets used for evaluation. Section 3 evaluates the CESM2-forced
RACMO2 product using in situ and remotely sensed
measurements. Model comparison to previous RACMO2 simulations is
discussed in Sect. 4, as well as representation of recent trends in
SMB components and mass loss. Conclusions are drawn in Sect. 5.
MethodsThe Community Earth System Model: CESM2
CESM2.1, hereafter referred to as CESM2, is an ESM that simulates
mutual interactions between atmosphere–ocean–land systems on the
global scale. The model incorporates the Community Atmosphere
Model version 6 (CAM6) , resolving global
atmospheric dynamics and physics, the Parallel Ocean Program model
version 2.1 (POP2.1) , and the Los Alamos National
Laboratory Sea Ice Model version 5.1 (CICE5.1) ,
modelling global oceanic circulation and sea-ice evolution. These
are coupled with the Community Land Model version 5 (CLM5)
and the Community Ice Sheet Model version 2.1
(CISM2.1) simulating land–atmosphere
interactions and ice dynamics. Here, we use a full
atmosphere–ocean coupling in CESM2, i.e. including sea ice
dynamics and sea surface temperature evolution while excluding
land ice dynamics (e.g. calving). The model is run at 1∘
spatial resolution (∼111km) and only prescribes
atmospheric greenhouse gas (CO2 and CH4) and
aerosol emissions as well as land cover use
. CESM2 has been extensively tested and adapted
to realistically reproduce the contemporary climate and SMB of the
GrIS . Detailed model
description, latest updates and evaluation are provided in
.
Regional Atmospheric Climate Model: RACMO2
RACMO2 is an RCM that is specifically adapted to simulate the SMB
of polar ice sheets . The model
incorporates the dynamical core of the High Resolution Limited Area
Model (HIRLAM) and the physics package cycle
CY33r1 of the European Centre for Medium-Range Weather Forecasts
Integrated Forecast System . It includes
a multi-layer snow module that simulates melt, liquid water
percolation and retention, refreezing and runoff
and accounts for dry-snow densification
following . RACMO2 implements an albedo scheme
that calculates snow albedo based on prognostic snow grain size,
cloud optical thickness, solar zenith angle and impurity
concentration in snow . In line with in situ
measurements , impurity concentration (soot) in
RACMO2 is prescribed as a constant in time and space at
0.1 ppmv. The model simulates drifting snow
erosion and sublimation following . The latest
model version RACMO2.3p2 accurately simulates the contemporary
climate and SMB of the GrIS when it is forced by ERA-40
(1958–1978) and ERA-Interim (1979–present) climate reanalyses
and is statistically downscaled to
1 km spatial resolution (see Sect. 2.4). For detailed model
description, latest updates and evaluation, we refer to
.
(a) Annual mean SMB (1950–2014) as modelled by RACMO2.3p2 forced by CESM2, statistically downscaled to 1 km resolution. (b) SMB evaluation in the accumulation zone (182 sites; white dots in a) and (c) in the ablation zone of the GrIS (213 sites; yellow dots in a). Statistics including the number of observations (N), slope (b0) and intercept (b1) of the regressions, determination coefficient (R2), RMSE and bias are listed for the ERA (red) and CESM2-forced RACMO2.3p2 simulation (blue). (d) Period (2003–2014) cumulative SMB (black), glacial discharge (orange) , mass balance (MB = SMB - discharge; blue) and mass loss derived from GRACE (red) . To enable a direct comparison with GRACE in (d), SMB is integrated over the GrIS, peripheral ice caps and tundra regions of Greenland. Boxes represent 1 standard deviation around the mean (horizontal bars).
Model initialisation and set-up
Here, we conduct a CMIP6-style historical simulation (1950–2014)
using RACMO2.3p2 at 11 km horizontal resolution
to dynamically downscale the outputs of CESM2
prescribed in a 24-grid-cell-wide relaxation zone at the model
lateral boundaries. Forcing consists of atmospheric temperature,
pressure, specific humidity, wind speed and direction being
prescribed on a 6-hourly basis at the 40 model atmospheric
levels. Upper atmosphere relaxation is implemented
. Sea surface temperature and sea ice extent/cover
are prescribed from the CESM2 forcing every 6 h. RACMO2.3p2
has typically 40 to 60 active snow layers that are initialised in
January 1950 using temperature and density profiles derived from
the offline IMAU Firn Densification Model (IMAU-FDM)
. Glacier outlines and surface topography are
prescribed from a down-sampled version of the 90 m
Greenland Ice Mapping Project (GIMP) digital elevation model (DEM)
. Bare ice albedo is prescribed from the
500 m MODerate-resolution Imaging Spectroradiometer (MODIS)
16 d Albedo product (MCD43A3), as the 5 % lowest surface
albedo records for the period 2000–2015, clipped between 0.30 for
bare ice and 0.55 for bright ice covered by perennial firn in the
accumulation zone. The current study uses the climate forcing of
1 out of the 12 members of the CESM2 historical
ensemble. Forcing RACMO2 with other CESM2 members would have been
ideal, but doing so in a transient fashion and at high spatial and
temporal resolution is computationally prohibitive. Instead, we
select one member that offers the 6-hourly climate forcing required
to drive RACMO2 while being representative of other CESM2 members
(see Sect. 4.3 and Fig. a).
Statistical downscaling
Following , the historical simulation at
11 km, hereafter referred to as CESM2-forced RACMO2.3p2,
is further statistically downscaled to a 1 km ice mask and
topography derived from the 90 m GIMP DEM
. In brief, the downscaling procedure corrects
individual SMB components (except for precipitation),
i.e. primarily meltwater runoff, for elevation and ice albedo
biases on the relatively coarse model grid at 11 km
resolution. These corrections reconstruct individual SMB
components on the 1 km GrIS topography using
daily-specific gradients estimated at 11 km and minimise
the remaining runoff underestimation using a down-sampled
1 km MODIS 16 d ice albedo product averaged for
2000–2015. Precipitation, including snowfall and rainfall, is
bilinearly interpolated from the 11 km onto the
1 km grid without additional corrections
. Statistical downscaling proves essential to
resolve narrow ablation zones, outlet glaciers and ice caps at the
GrIS margins that significantly contribute to contemporary mass
loss of Greenland land ice . For instance,
applying statistical downscaling increases GrIS-wide runoff by
55 Gtyr-1 (+23%) on average for the
period 1950–2014, resulting in a SMB decrease of
56 Gtyr-1 (-13 %).
Evaluation data sets
For evaluation, we use a compilation of in situ SMB measurements
derived from 182 stakes, snow pits and airborne
radar campaign in the GrIS accumulation area
(182 records; white dots in Fig. a) and collected at
213 sites in the ablation zone (1073 records;
yellow dots in Fig. a) . In addition, combined modelled
SMB and glacial discharge estimates (1972–2018) are
compared to mass changes from GRACE over the period 2003–2014
. The CESM2-forced RACMO2.3p2 historical
simulation is also compared to SMB and individual components from
an ensemble of eight previous RACMO2 simulations
,
using different climate forcing (ERA reanalysis or the ESM
HadGEM2) at various spatial resolutions (1, 5.5 and
11 km). These simulations are listed and further compared
in Tables and for the overlapping model
period 1972–2012.
Annual mean SMB and components integrated over the GrIS (Gtyr-1) for the period 1972–1990 from an ensemble of RACMO2 simulations using various spatial resolutions and lateral forcing. SMB components include total precipitation (PR), runoff (RU), melt (ME), refreezing (RF) and rainfall (RA). The uncertainty range corresponds to 1 standard deviation around the mean. Here mass balance of the GrIS (MB) is estimated as GrIS-integrated SMB minus glacial discharge for the period 1972–1990 (458 Gtyr-1) .
Annual mean SMB and components integrated over the GrIS (Gtyr-1) for the period 1991–2012 from an ensemble of RACMO2 simulations using various spatial resolutions and lateral forcing. SMB components include total precipitation (PR), runoff (RU), melt (ME), refreezing (RF) and rainfall (RA). The uncertainty range corresponds to 1 standard deviation around the mean. Here mass balance of the GrIS (MB) is estimated as GrIS-integrated SMB minus glacial discharge for the period 1991–2012 (485 Gtyr-1) .
1991–2012ReferenceForcingGridMBSMBPRRUMERFRARACMO2.1HadGEM211 km-240±80245±141785±84496±114749±156318±5467±17RACMO2.1ERA11 km-155±80330±116731±64361±100588±146279±6552±14RACMO2.3p1ERA11 km-179±83306±119685±64336±99543±145233±6026±8RACMO2.3p2ERA11 km-95±78389±114709±61286±92470±133208±5326±8RACMO2.3p2ERA5.5 km-94±81391±117717±62291±91469±131203±5326±8RACMO2.3p2This studyCESM211 km-127±47358±83704±70314±93495±131211±5132±14RACMO2.3p1ERA1 km-200±95285±131757±66428±109680±146283±5327±9RACMO2.3p2ERA1 km-170±85315±121705±61357±101590±136286±5238±8RACMO2.3p2ERA1 km-157±85328±121717±63353±96594±131290±5036±8RACMO2.3p2This studyCESM21 km-195±49290±85702±70380±101620±128302±4744±13Surface mass balance evaluation and uncertainty
Figure a shows annual mean SMB from CESM2-forced
RACMO2.3p2, statistically downscaled to 1 km. As is the
case with state-of-the-art reanalysis-forced simulations
, it
accurately captures the extensive inland accumulation area as well as
narrow ablation zones, outlet glaciers and ice caps fringing the
GrIS margins (Fig. a). The model shows very good
agreement with multi-year averaged SMB observations in the
accumulation zone (R2=0.89; Fig. b), with a small
bias and RMSE of -20.5 and 63.3 mm w.e. Interestingly,
these statistics are on par with the recent RACMO2.3p2 run at
11 km forced by ERA reanalysis and statistically
downscaled to 1 km, hereafter referred to
as ERA-forced RACMO2.3p2. In the ablation zone, CESM2-forced
RACMO2.3p2 agrees reasonably well with ablation measurements:
R2=0.61 vs. 0.72 (Fig. c). The
model shows larger bias and RMSE relative to ERA-forced RACMO2.3p2
(+0.06 and +0.18m w.e.). As CESM2 neither
assimilates nor prescribes climatic observations, a larger bias was
expected. Good agreement with observations can be partly
attributed to dynamical downscaling in RACMO2, which results in
realistic SMB gradients if appropriate climate forcing is
prescribed , and to statistical downscaling, as it
minimises SMB bias by enhancing runoff in marginal ablation zones
. On the regional scale, CESM2-forced and
ERA-forced RACMO2.3p2 simulations show no significant difference
in SMB and components for the period 1958–2014 (not shown),
i.e. mean difference (CESM2-forced minus ERA-forced) lower than
1 standard deviation of the 1958–2014 period.
We follow to estimate the SMB uncertainty
(σ). Mean accumulation (20.5 mm w.e.;
Fig. b) and ablation biases (180.0 mm w.e.;
Fig. c) are accumulated over the long-term (1958–2014)
accumulation and ablation zone of the GrIS, with an area of ∼1 521 400 and ∼179 400 km2
respectively.
σ=(biasabl.×areaabl.)2+(biasacc.×areaacc.)2
This yields σ=48Gtyr-1. A similar value
(43 Gtyr-1) is obtained for the downscaled ERA-forced
RACMO2.3p2 product. Integrated over the ice sheet, CESM2-forced and
ERA-forced simulations agree very well, with an average cumulative
SMB of 365±48 and
357±43Gtyr-1 for the period
1958–2014. Figure d compares modelled and remotely
sensed (GRACE) cumulative mass change for the period 2003–2014,
respectively. Modelled mass change (-3299±1240Gt;
blue box), estimated as cumulative SMB over the GrIS, peripheral
ice caps and tundra region (2970±1097Gt; black box)
minus glacial discharge (6269±143Gt; orange box),
shows excellent agreement with GRACE (-3290±1434Gt;
red box). This highlights the ability of CESM2-forced RACMO2.3p2 to
also realistically capture the recent Greenland mass loss
(2003–2014) .
Surface mass balance variability
Figure a and c show annual mean SMB components
simulated by CESM2-forced RACMO2.3p2 at 1 km (horizontal
bars) for the periods 1972–1990 and 1991–2012. Likewise,
Fig. b and d show annual mean mass balance (MB; blue),
i.e. SMB (black) minus glacial discharge (orange), simulated by
CESM2-forced RACMO2.3p2 on the original model grid (11 km)
and statistically downscaled to 1 km for the two
periods. Ice discharge and associated uncertainties (1972–2014)
are from . Prior to 1972, ice discharge and
uncertainties are assumed constant at the 1972 value. Error bars
represent the inter-annual variability in SMB components estimated
as 1 standard deviation around the mean. Boxes show the range of
modelled SMB components derived from an ensemble of seven RACMO2
simulations forced by ERA reanalysis. To highlight recent
improvements in ESM climate forcing, we use the modelled SMB
components from a previous RACMO2.1 simulation forced by HadGEM2
(dark green dots). HadGEM2 is a coupled atmosphere–ocean model
developed by the Met Office Hadley Centre using a spatial
resolution of 1.875∘×1.25∘ (atmosphere)
and 1∘×1∘ (ocean). Similar to CESM2, the
model prescribes land cover use, greenhouse gas and aerosol
emissions and simulates the dynamic evolution of sea ice and sea
surface temperature over the historical period
. Compared to RACMO2.1, the latest RACMO2.3p2
version implements (1) significant changes in the cloud physics
favouring more snowfall in the ice sheet interior; (2) lower
impurity concentration in snow (soot) and smaller snow grain size,
both reducing the previously underestimated snow albedo; (3) less-active snow drift erosion limiting the overestimated exposure of
bare ice notably in the northeast of Greenland. For additional
information about the HadGEM2-forced RACMO2.1 simulation and
settings, we refer the reader to ; key
differences between RACMO2.1, RACMO2.3p1 and p2 are discussed in
. Annual mean SMB components and
corresponding inter-annual variability for the ensemble RACMO2
simulations are listed in Tables (1972–1990) and (1991–2012).
(a) Average GrIS SMB (black) and components at 1 km, i.e. total precipitation (blue), runoff (red), melt (orange), refreezing (cyan) and rainfall (green), for the period 1972–1990. (b) Annual mean SMB (black), glacial discharge (orange) and mass balance (MB = SMB - discharge; blue) at 11 and 1 km resolution for the period 1972–1990. (c and d) same as (a and b) for the period 1991–2012. The green dots represent values from a previous HadGEM2-forced RACMO2.1 simulation. Boxes around the CESM2-forced RACMO2.3p2 mean (horizontal bars) represent the range of modelled estimates from an ensemble of RACMO2 simulations (five at 11 km, one at 5.5 km, and four at 1 km; Tables and ). The error bars represent 1 standard deviation (σ) around the CESM2-forced RACMO2.3p2 mean.
Approximate mass balance: 1972–1990
In the period 1972–1990, the mass balance of the GrIS was close to
zero or slightly negative
. Figure a, b and Table
show that downscaled CESM2-forced RACMO2.3p2 reproduces, within 1
standard deviation, SMB and components obtained from seven previous
reanalysis-forced RACMO2 simulations at various spatial
resolutions. For instance, precipitation
(701±98Gtyr-1) and runoff
(242±40Gtyr-1) compare well with ERA-forced
RACMO2.3p2 , i.e. 712±73 and
257±53Gtyr-1, resulting in similar SMB of 428 and
423 Gtyr-1 (-1 %) (Fig. a and
Table ). This highlights the ability of the CESM2 forcing
to capture realistic Greenland SMB before mass loss started in the
1990s.
Figure b shows that SMB on the 11 km grid falls
well within ERA-forced simulations at similar resolution (black
box). Through statistical downscaling, SMB at 1 km
decreases by 13 % from 485 to 428 Gtyr-1, in
line with other simulations (Fig. b and
Table ). Combining average GrIS-integrated SMB with
glacial discharge (458 Gtyr-1; 1972–1990),
CESM2-forced RACMO2.3p2 results in slightly negative mass balance
(-31 Gtyr-1; Table ). A previous attempt
using RACMO2.1 forced by the climate of HadGEM2
did not accurately represent GrIS-integrated SMB components (dark
green dots in Fig. a, b). While precipitation was
generally well represented (685±82Gtyr-1), runoff
was overestimated by ∼50% compared to ERA-forced
RACMO2.3p2 (Table ). As a result, SMB was underestimated
by ∼40%, driving an unrealistic mass loss of
189 Gtyr-1 over the period 1972–1990.
Mass loss: 1991–2012
In the two decades following 1990 (1991–2012), the GrIS
experienced accelerated mass loss ,
primarily driven by a decrease in SMB (-138 Gtyr-1
with respect to 1972–1990; Fig. c and Table )
combined with an increase in glacial discharge (+26Gtyr-1). Figure c shows that
CESM2-forced RACMO2.3p2 similarly reproduces the recent SMB
decrease resulting from enhanced surface runoff (+138Gtyr-1 or +57%) compared to the
ERA-forced RACMO2.3p2 simulation (+100Gtyr-1 or
+38%). This pronounced runoff increase stems from
enhanced surface melt (+163Gtyr-1 or
+36%) exceeding the increase in meltwater retention and
refreezing in the firn (+36Gtyr-1 or
+14%). Precipitation does not substantially change
after 1991, in line with the ensemble ERA-forced RACMO2 simulations
(Table ). We conclude that CESM2-forced RACMO2.3p2
captures the post-1990 SMB decrease that tipped the GrIS into
a state of sustained mass loss (195 and 170 Gtyr-1 for
downscaled CESM2-forced and ERA-forced RACMO2.3p2 for 1991–2012;
Fig. d). In contrast, SMB components in HadGEM2-forced
RACMO2.1 remain largely overestimated compared to other simulations
(Table ), particularly runoff and melt
(Fig. c), resulting in overestimated mass loss for
1991–2012 (240±80Gtyr-1; Fig. d). The
reason is that, unlike CESM2 , the HadGEM2
forcing had a strong, systematic warm bias of ∼1∘C, resulting in
overestimated meltwater runoff and thus underestimated SMB
(Fig. d).
Time series and trends
Figure a and b show time series of individual
GrIS-integrated SMB components for the period 1950–2014 as
modelled by the latest, state-of-the-art ERA-forced RACMO2.3p2 run
at 5.5 km horizontal resolution and the
current CESM2-forced RACMO2.3p2 simulation, both statistically
downscaled to 1 km. It is important to note that, compared
to forcing by reanalyses that assimilate observations, the
CESM2-forced simulation produces extreme melt years (e.g. 2005 and
2011; Fig. b) that are realistic in magnitude but not
necessarily in timing (e.g. the observed 2012 melt peak;
Fig. a). For 1960–1990, the two products show similar
and insignificant trends in total precipitation
(2.4±1.6Gtyr-2; p value = 0.14) and runoff
(1.1±1.0Gtyr-2; p value = 0.27)
(Fig. b). After 1991, CESM2-forced RACMO2.3p2 reproduces
the significant (p value = 0.0001) positive runoff trend
(10.4±2.2Gtyr-2; Fig. b) similar to the
ERA-forced simulation (8.8±2.1Gtyr-2;
Fig. a). The runoff trend in CESM2-forced RACMO2.3p2 is
no coincidence. Figure a shows the atmospheric
temperature at 700 hPa (T700) from the current CESM2
simulation (red) and from 11 additional ensemble members
(grey). Compared to T700 derived from ERA-40 (1958–1978) and
ERA-Interim (1979–2014; black line in Fig. a), the
current CESM2 simulation shows a cold bias of
0.6 ∘C over 1958–2014. For the ERA-Interim period,
the bias decreases to 0.4 ∘C. All CESM2 members
show a similar warming trend after 1991, in line with the
reanalysis data (dashed black line), highlighting the ability of
CESM2 to represent the recent climate of Greenland. As in
, we find a clear correlation (r=0.67;
Fig. b) between CESM2-forced RACMO2.3p2 runoff at
1 km and T700 from the CESM2 simulation (red). This
means that the post-1990 runoff increase would have been obtained
irrespective of the selected CESM2 member. Physical drivers of the
warming trend in the CESM2 forcing are currently being investigated
and will be discussed in a forthcoming publication. Compared to the
ERA-forced run, the more pronounced runoff trend in CESM2-forced
RACMO2.3p2 results from a significant (p value = 0.016) positive
trend in rainfall (1.3±0.3Gtyr-2
vs. 0.3±0.2Gtyr-2) for a similar melt
acceleration (11.9±3.0Gtyr-2
vs. 10.9±3.0Gtyr-2). Total precipitation in the
CESM2-forced RACMO2.3p2 simulation shows a significant (p value = 0.002) positive trend (5.8±1.7Gtyr-2;
Fig. b) in contrast to a negative trend in the ERA-forced
run (-1.9±1.8Gtyr-2; Fig. a). However,
the latter trend stems from decadal variability as it becomes
insignificant for the period 1950–2014:
0.9±0.5Gtyr-2 (p value = 0.090). In addition,
the positive precipitation trend disappears when extending time
series using a CESM2-based SSP5-8.5 scenario (not shown),
demonstrating that the latter trend originates from internal
decadal variability.
Time series of downscaled (1 km) GrIS-integrated annual SMB components, namely total precipitation (PR; blue), runoff (RU; red), melt (ME; orange), refreezing (RF; cyan) and rainfall (RA; green), as modelled by (a) ERA-forced RACMO2.3p2 (1958–2014) and (b) CESM2-forced RACMO2.3p2 (1950–2014). (c) Time series of annual SMB (CESM2-forced RACMO2.3p2 at 1 km), glacial discharge (D) and mass balance (MB = SMB -D). Mass loss from GRACE (2003–2014) is represented by red dots . Dashed lines show the 1991–2014 trends. To enable a direct comparison with GRACE in (c), SMB is integrated over the GrIS, peripheral ice caps and surrounding tundra regions of Greenland.
(a) Time series of the annual June–July–August (JJA) atmospheric temperature at 700 hPa (T700) averaged over 60–80∘ N and 20–80∘ W for the ERA-40 (1958–1978) and ERA-Interim (1979–2014) reanalyses (black), the current CESM2 simulation (red) and 11 additional reference ensemble members (dark grey). The grey belt encompasses annual minimum and maximum values of the whole ensemble. (b) Annual GrIS-integrated runoff derived from CESM2-forced RACMO2.3p2 at 1 km resolution as a function of JJA atmospheric temperature at 700 hPa from the current CESM2 simulation (red).
In line with , Fig. c shows that ∼60% of the recent mass loss acceleration in
CESM2-forced RACMO2.3p2 is caused by decreased SMB
(6.6±3.3Gtyr-2) resulting from enhanced
meltwater runoff; the remaining ∼40% is ascribed
to increased glacial discharge
(4.7±0.5Gtyr-2). As a result, Greenland mass
balance decreased by an estimated rate of
11.3±3.2Gtyr-2 (or
9.4±1.6Gtyr-2 for the GrIS only) in good
agreement with GRACE (9.4±1.2Gtyr-2 for
2003–2014; Fig. c). It is important to note that the
main drivers of the post-1990 (surface) mass loss in CESM2-forced
RACMO2.3p2 may differ from that of the reanalysis-based products
since ESMs from CMIP5 (and likely CMIP6) do not accurately
reproduce the change in summertime Arctic atmospheric circulation
that is often associated with the recent mass loss acceleration
. In brief, the current study is meaningful for
two reasons: for the first time, an ESM, assimilating no
observational climatic data except for atmospheric greenhouse gas
and aerosol emissions, can (1) reliably reproduce the historical
average and variability of SMB and its individual components and (2)
realistically represent the recent Greenland mass loss
acceleration in line with remote sensing. These results are
essential for forthcoming attribution studies investigating
post-1990 GrIS mass loss.
Conclusions
Historical outputs (1950–2014) of the Earth system model CESM2
(∼111km) are dynamically downscaled using the
regional climate model RACMO2.3p2 (∼11km) over the
GrIS. The resulting SMB components are further statistically
downscaled to 1 km spatial resolution to resolve the narrow
ablation zones and marginal outlet glaciers of Greenland. Model
evaluation using in situ and remotely sensed measurements
demonstrates the ability of CESM2-forced RACMO2.3p2 to
realistically represent SMB as well as the rapid post-1990 melt and
runoff increase. Combining modelled SMB with observed glacial
discharge, our new ESM-based SMB product reflects an ice sheet in
approximate mass balance before 1991, followed by a rapid mass loss
acceleration resulting from enhanced meltwater runoff: two key
features that, until now, exclusively showed up in reanalysis-based
estimates. It is important to note that the main drivers of runoff
increase in CESM2-forced RACMO2.3p2 may differ from that of
reanalysis-based products since ESMs from CMIP5 (and likely CMIP6)
do not accurately reproduce the recent change in summertime Arctic
atmospheric circulation that is often associated with mass loss
acceleration. For the first time, an Earth system model (CESM2),
which does not assimilate climatic observations, can be used to
force a regional climate model (RACMO2) to yield realistic
historical GrIS SMB average and variability. Furthermore, our
results suggest that CESM2 climate forcing can be used without bias
corrections to project the SMB of the GrIS under different warming
scenarios and quantify Greenland's contribution to future eustatic
sea level rise.
Data availability
Data sets presented in this study are available from the authors upon request and without conditions.
Author contributions
BN prepared the manuscript, conducted the CESM2-forced RACMO2.3p2 simulation and analysed the data. LvK and JTML provided the historical CESM2 forcing. BW processed the GRACE mass anomalies time series. MRvdB and WJvdB helped interpreting the results. All authors commented the manuscript.
Competing interests
The authors declare that they have no conflict of interest.
Acknowledgements
Brice Noël was funded by NWO VENI grant VI.Veni.192.019. Leonardus van Kampenhout, Willem Jan van de Berg and Michiel van den Broeke acknowledge funding from NWO and NESSC. Bert Wouters was funded by NWO VIDI grant 016.Vidi.171.063. The Community Earth System Model (CESM) project is supported primarily by the National Science Foundation (NSF). This material is based upon work supported by the National Center for Atmospheric Research (NCAR), which is a major facility sponsored by the NSF under cooperative agreement no. 1852977.
Financial support
This research has been supported by the Polar Programme of the Netherlands Organisation for Scientific Research (NWO) (VENI (grant no. VI.Veni.192.019)).
Review statement
This paper was edited by Xavier Fettweis and reviewed by three anonymous referees.
ReferencesBailey, D., Hunke, E., DuVivier, A., Lipscomb, B., Bitz, C., Holland, M.,
Briegleb, B., and Schramm, J.: CESM CICE5 Users Guide, Release CESM CICE5,
Documentation and Software User's Manual from Los Alamos National Laboratory,
1–41, available at: http://www.cesm.ucar.edu/models/cesm2/sea-ice/ (last access: 17 April 2020),
2018.Bales, R. C., Guo, Q., Shen, D., McConnell, J. R., Du, G., Burkhart, J. F.,
Spikes, V. B., Hanna, E., and Cappelen, J.: Annual accumulation for
Greenland updated using ice core data developed during 2000–2006 and analysis
of daily coastal meteorological data, J. Geophys. Res., 114,
D06116, 10.1029/2008JD011208, 2009.Bamber, J. L., Westaway, R. M., Marzeion, B., and Wouters, B.: The land ice
contribution to sea level during the satellite era, Environ. Res.
Lett., 13, 063008, 10.1088/1748-9326/aac2f0, 2018.Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., Kobayashi,
S., Andrae, U., Balmaseda, M. A., Balsamo, G., Bauer, P., Bechtold, P.,
Beljaars, A. C. M., van de Berg, L., Bidlot, J., Bormann, N., Delsol, C.,
Dragani, R., Fuentes, M., Geer, A. J., Haimberger, L., Healy, S. B.,
Hersbach, H., Hòlm, E. V., Isaksen, L., Kållberg, P., Köhler, M.,
Matricardi, M., McNally, A. P., Monge-Sanz, B. M., Morcrette, J.-J., Park,
B.-K., Peubey, C., de Rosnay, P., Tavolato, C., Thépaut, J.-N., and Vitart,
F.: The ERA-Interim reanalysis: configuration and performance of the data
assimilation system, Q. J. Roy. Meteor. Soc.,
137, 553–597, 10.1002/qj.828, 2011.Delhasse, A., Fettweis, X., Kittel, C., Amory, C., and Agosta, C.: Brief communication: Impact of the recent atmospheric circulation change in summer on the future surface mass balance of the Greenland Ice Sheet, The Cryosphere, 12, 3409–3418, 10.5194/tc-12-3409-2018, 2018.Doherty, S. J., Warren, S. G., Grenfell, T. C., Clarke, A. D., and Brandt, R. E.: Light-absorbing impurities in Arctic snow, Atmos. Chem. Phys., 10, 11647–11680, 10.5194/acp-10-11647-2010, 2010.
ECMWF-IFS: Part IV: Physical Processes (CY33R1), Technical Report, IFS Documentation CY33R1, Part IV: Physical Processes, ECMWF, Reading UK,
2008.Ettema, J., van den Broeke, M. R., van Meijgaard, E., van de Berg, W. J., Box, J. E., and Steffen, K.: Climate of the Greenland ice sheet using a high-resolution climate model Part 1: Evaluation, The Cryosphere, 4, 511–527, 10.5194/tc-4-511-2010, 2010.Eyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., and Taylor, K. E.: Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization, Geosci. Model Dev., 9, 1937–1958, 10.5194/gmd-9-1937-2016, 2016.Fettweis, X., Franco, B., Tedesco, M., van Angelen, J. H., Lenaerts, J. T. M., van den Broeke, M. R., and Gallée, H.: Estimating the Greenland ice sheet surface mass balance contribution to future sea level rise using the regional atmospheric climate model MAR, The Cryosphere, 7, 469–489, 10.5194/tc-7-469-2013, 2013.Fettweis, X., Box, J. E., Agosta, C., Amory, C., Kittel, C., Lang, C., van As, D., Machguth, H., and Gall'ee, H.: Reconstructions of the 19002015 Greenland ice sheet surface mass balance using the regional climate MAR model, The Cryosphere, 11, 1015–1033, 10.5194/tc-11-1015-2017, 2017.Gettelman, A., Truesdale, J. E., Bacmeister, J. T., Caldwell, P. M., Neale,
R. B., Bogenschutz, P. A., and Simpson, I. R.: The Single Column Atmosphere
Model Version 6 (SCAM6): Not a Scam but a Tool for Model Evaluation and
Development, J. Adv. Model. Earth Sy., 11, 1381–1401, 10.1029/2018MS001578, 2019.Hanna, E., Fettweis, X., and Hall, R. J.: Brief communication: Recent changes in summer Greenland blocking captured by none of the CMIP5 models, The Cryosphere, 12, 3287–3292, 10.5194/tc-12-3287-2018, 2018.Howat, I. M., Negrete, A., and Smith, B. E.: The Greenland Ice Mapping Project (GIMP) land classification and surface elevation data sets, The Cryosphere, 8, 1509–1518, 10.5194/tc-8-1509-2014, 2014.Jones, C. D., Hughes, J. K., Bellouin, N., Hardiman, S. C., Jones, G. S., Knight, J., Liddicoat, S., O'Connor, F. M., Andres, R. J., Bell, C., Boo, K.-O., Bozzo, A., Butchart, N., Cadule, P., Corbin, K. D., Doutriaux-Boucher, M., Friedlingstein, P., Gornall, J., Gray, L., Halloran, P. R., Hurtt, G., Ingram, W. J., Lamarque, J.-F., Law, R. M., Meinshausen, M., Osprey, S., Palin, E. J., Parsons Chini, L., Raddatz, T., Sanderson, M. G., Sellar, A. A., Schurer, A., Valdes, P., Wood, N., Woodward, S., Yoshioka, M., and Zerroukat, M.: The HadGEM2-ES implementation of CMIP5 centennial simulations, Geosci. Model Dev., 4, 543–570, 10.5194/gmd-4-543-2011, 2011.Kuipers Munneke, P., van den Broeke, M. R., Lenaerts, J. T. M., Flanner,
M. G., Gardner, A. S., and van de Berg, W. J.: A new albedo parameterization
for use in climate models over the Antarctic ice sheet, J. Geophys. Res., 116, D05114, 10.1029/2010JD015113, 2011.Lawrence, D. M., Fisher, R. A., Koven, C. D., Oleson, K. W., Swenson, S. C.,
Bonan, G., Collier, N., Ghimire, B., van Kampenhout, L., Kennedy, D., Kluzek,
E., Lawrence, P. J., Li, F., Li, H., Lombardozzi, D., Riley, W. J., Sacks,
W. J., Shi, M., Vertenstein, M., Wieder, W. R., Xu, C., Ali, A. A., Badger,
A. M., Bisht, G., van den Broeke, M., Brunke, M. A., Burns, S. P., Buzan, J.,
Clark, M., Craig, A., Dahlin, K., Drewniak, B., Fisher, J. B., Flanner, M.,
Fox, A. M., Gentine, P., Hoffman, F., Keppel-Aleks, G., Knox, R., Kumar,
S., Lenaerts, J., Leung, L. R., Lipscomb, W. H., Lu, Y., Pandey, A.,
Pelletier, J. D., Perket, J., Randerson, J. T., Ricciuto, D. M., Sanderson,
B. M., Slater, A., Subin, Z. M., Tang, J., Thomas, R. Q., Martin, M. V., and
Zeng, X.: The Community Land Model version 5: Description of new features,
benchmarking, and impact of forcing uncertainty, J. Adv. Model. Earth Sy., 11, 4245–4287, 10.1029/2018MS001583, 2019.Lenaerts, J. T. M., van den Broeke, M. R., van Angelen, J. H., van Meijgaard, E., and D'ery, S. J.: Drifting snow climate of the Greenland ice sheet: a study with a regional climate model, The Cryosphere, 6, 891–899, 10.5194/tc-6-891-2012, 2012.Lenaerts, J. T. M., Medley, B., van den Broeke, M. R., and Wouters, B.:
Observing and Modeling Ice Sheet Surface Mass Balance, Rev. Geophys., 57, 376–420, 10.1029/2018RG000622, 2019.Ligtenberg, S. R. M., Helsen, M. M., and van den Broeke, M. R.: An improved semi-empirical model for the densification of Antarctic firn, The Cryosphere, 5, 809–819, 10.5194/tc-5-809-2011, 2011.Ligtenberg, S. R. M., Kuipers Munneke, P., Noël, B. P. Y., and van den Broeke, M. R.: Brief communication: Improved simulation of the present-day Greenland firn layer (1960–2016), The Cryosphere, 12, 1643–1649, 10.5194/tc-12-1643-2018, 2018.Lipscomb, W. H., Price, S. F., Hoffman, M. J., Leguy, G. R., Bennett, A. R., Bradley, S. L., Evans, K. J., Fyke, J. G., Kennedy, J. H., Perego, M., Ranken, D. M., Sacks, W. J., Salinger, A. G., Vargo, L. J., and Worley, P. H.: Description and evaluation of the Community Ice Sheet Model (CISM) v2.1, Geosci. Model Dev., 12, 387–424, 10.5194/gmd-12-387-2019, 2019.Machguth, H., Thomsen, H., Weidick, A., Ahlstrøm, A. P., Abermann, J.,
Andersen, M. L., Andersen, S., Bjørk, A. A., Box, J. E., Braithwaite,
R. J., Bøggild, C. E., Citterio, M., Clement, P., Colgan, W., Fausto,
R. S., Gubler, K. G. S., Hasholt, B., Hynek, B., Knudsen, N., Larsen, S.,
Mernild, S., Oerlemans, J., Oerter, H., Olesen, O., Smeets, C., Steffen, K.,
Stober, M., Sugiyama, S., van As, D., van den Broeke, M., and van de Wal,
R. S.: Greenland surface mass balance observations from the ice sheet
ablation area and local glaciers, J. Glaciol., 62, 1–27,
10.1017/jog.2016.75, 2016.Mottram, R., Boberg, F., Langen, P., Yang, S., Rodehacke, C., Christensen,
J. H., and Madsen, M. S.: Surface Mass balance of the Greenland ice Sheet in
the Regional Climate Model HIRHAM5: Present State and Future Prospects, Low
Temperature Science, 75, 105–115, 10.14943/lowtemsci.75.105,
2017.Mouginot, J., Rignot, E., Bjørk, A., van den Broeke B., M., Millan, R.,
Morlighem, M., Noël, B., Scheuchla, and Wood, M.: Forty six years of
Greenland Ice Sheet mass balance from the component method: 1972 to 2018,
P. Natl. Acad. Sci. USA, 116, 9239–9244, 10.1073/pnas.1904242116, 2019.Niwano, M., Aoki, T., Hashimoto, A., Matoba, S., Yamaguchi, S., Tanikawa, T., Fujita, K., Tsushima, A., Iizuka, Y., Shimada, R., and Hori, M.: NHMSMAP: spatially and temporally high-resolution nonhydrostatic atmospheric model coupled with detailed snow process model for Greenland Ice Sheet, The Cryosphere, 12, 635–655, 10.5194/tc-12-635-2018, 2018.Noël, B., van de Berg, W. J., van Meijgaard, E., Kuipers Munneke, P., van de Wal, R. S. W., and van den Broeke, M. R.: Evaluation of the updated regional climate model RACMO2.3: summer snowfall impact on the Greenland Ice Sheet, The Cryosphere, 9, 1831–1844, 10.5194/tc-9-1831-2015, 2015.Noël, B., van de Berg, W. J., Machguth, H., Lhermitte, S., Howat, I., Fettweis, X., and van den Broeke, M. R.: A daily, 1 km resolution data set of downscaled Greenland ice sheet surface mass balance (19582015), The Cryosphere, 10, 2361–2377, 10.5194/tc-10-2361-2016, 2016.Noël, B., van de Berg, W. J., Lhermitte, S., Wouters, B., Machguth, H.,
Howat, I., Citterio, M., Moholdt, G., Lenarets, J. T. M., and van den Broeke,
M. R.: A tipping point in refreezing accelerates mass loss of Greenland's
glaciers and ice caps, Nat. Commun., 8, 14730,
10.1038/ncomms14730, 2017.Noël, B., van de Berg, W. J., van Wessem, J. M., van Meijgaard, E., van As, D., Lenaerts, J. T. M., Lhermitte, S., Kuipers Munneke, P., Smeets, C. J. P. P., van Ulft, L. H., van de Wal, R. S. W., and van den Broeke, M. R.: Modelling the climate and surface mass balance of polar ice sheets using RACMO2 Part 1: Greenland (1958–2016), The Cryosphere, 12, 811–831, 10.5194/tc-12-811-2018, 2018.Noël, B., van de Berg, W. J., Lhermitte, S., and van den Broeke, M. R.:
Rapid ablation zone expansion amplifies north Greenland mass loss, Sci.
Adv., 5, eaaw0123, 10.1126/sciadv.aaw0123, 2019.Overly, T. B., Hawley, R. L., Helm, V., Morris, E. M., and Chaudhary, R. N.: Greenland annual accumulation along the EGIG line, 1959–2004, from ASIRAS airborne radar and neutron-probe density measurements, The Cryosphere, 10, 1679–1694, 10.5194/tc-10-1679-2016, 2016.Rae, J. G. L., Aõalgeirsdóttir, G., Edwards, T. L., Fettweis, X., Gregory, J. M., Hewitt, H. T., Lowe, J. A., Lucas-Picher, P., Mottram, R. H., Payne, A. J., Ridley, J. K., Shannon, S. R., van de Berg, W. J., van de Wal, R. S. W., and van den Broeke, M. R.: Greenland ice sheet surface mass balance: evaluating simulations and making projections with regional climate models, The Cryosphere, 6, 1275–1294, 10.5194/tc-6-1275-2012, 2012.
Smith, R., Jones, P., Briegleb, B., Bryan, F., Danabasoglu, G., Dennis, J.,
Dukowicz, J., Eden, C., Fox-Kemper, B., and Gent, P.: The parallel ocean
program (POP) reference manual ocean component of the community climate
system model (CCSM) and community earth system model (CESM), Rep.
LAUR-01853, 141, 1–140, 2010.The IMBIE team: Mass balance of the Greenland Ice Sheet from 1992 to 2018,
Nature, 579, 233–239, 10.1038/s41586-019-1855-2, 2019.
Undèn, P., Rontu, L., Järvinen, H., Lynch, P., Calvo, J., Cats, G.,
Cuxart, J., Eerola, K., Fortelius, C., Garcia-Moya, J. A., Jones, C.,
Lenderlink, G., Mcdonald, A., Mcgrath, R., Navascues, B., Nielsen, N. W.,
Degaard, V., Rodriguez, E., Rummukainen, M., Sattler, K., Sass, B. H.,
Savijarvi, H., Schreur, B. W., Sigg, R., and The, H.: HIRLAM-5, Scientific
Documentation, technical Report, HIRLAM-5 Project, Norrköping, Sweden,
2002.
Uppala, S. M., Kållberg, P. W., Simmons, A. J., Andrae, U., Bechtold, V.
D. C., Fiorino, M., Gibson, J. K., Haseler, J., Hernandez, A., Kelly, G. A.,
Li, X., Onogi, K., Saarinen, S., Sokka, N., Allan, R. P., Andersson, E.,
Arpe, K., Balmaseda, M. A., Beljaars, A. C. M., Berg, L. V. D., Bidlot, J.,
Bormann, N., Caires, S., Chevallier, F., Dethof, A., Dragosavac, M., Fisher,
M., Fuentes, M., Hagemann, S., Hòlm, E., Hoskins, B. J., Isaksen, L.,
Janssen, P. A. E. M., Jenne, R., Mcnally, A. P., Mahfouf, J.-F., Morcrette,
J.-J., Rayner, N. A., Saunders, R. W., Simon, P., Ster, A., Trenberth, K. E.,
Untch, A., Vasiljevic, D., Viterbo, P., and Woollen, J.: The ERA-40
re-analysis, Q. J. Roy. Meteor. Soc., 131,
2961–3012, 2005.Van Angelen, J. H., Lenaerts, J. T. M., van den Broeke, M. R., Fettweis, X.,
and van Meijgaard, E.: Rapid loss of firn pore space accelerates 21st century
Greenland mass loss, Geophys. Res. Lett., 40, 2109–2113,
10.1002/grl.50490, 2013a.Van Angelen, J. H., van den Broeke, M. R., Wouters, B., and Lenaerts, J.
T. M.: Contemporary (1969–2012) evolution of the climate and surface mass
balance of the Greenland ice sheet, Surv. Geophys., 35, 1155–1174, 10.1007/s10712-013-9261-z, 2013b.Van de Berg, W. J. and Medley, B.: Brief Communication: Upper-air relaxation in RACMO2 significantly improves modelled interannual surface mass balance variability in Antarctica, The Cryosphere, 10, 459–463, 10.5194/tc-10-459-2016, 2016.Van den Broeke, M. R., Enderlin, E. M., Howat, I. M., Kuipers Munneke, P., Noël, B. P. Y., van de Berg, W. J., van Meijgaard, E., and Wouters, B.: On the recent contribution of the Greenland ice sheet to sea level change, The Cryosphere, 10, 1933–1946, 10.5194/tc-10-1933-2016, 2016.Van Kampenhout, L., Rhoades, A. M., Herrington, A. R., Zarzycki, C. M.,
Lenaerts, J. T. M., Sacks, W. J., and van den Broeke, M. R.: Improving the
Representation of Polar Snow and Firn in the Community Earth System Model,
J. Adv. Model. Earth Sy., 9, 2583–2600,
10.1002/2017MS000988, 2017.
Van Kampenhout, L., Rhoades, A. M., Herrington, A. R., Zarzycki, C. M., Lenaerts, J. T. M., Sacks, W. J., and van den Broeke, M. R.: Regional grid refinement in an Earth system model: impacts on the simulated Greenland surface mass balance, The Cryosphere, 13, 1547–1564, 10.5194/tc-13-1547-2019, 2019.Van Kampenhout, L., Lenaerts, J. T. M., Lipscomb, W. H., Lhermitte, S.,
Noël, B., Vizcaíno, M., Sacks, W. J., and van den Broeke, M. R.:
Present-Day Greenland Ice Sheet Climate and Surface Mass Balance in
CESM2, J. Geophys. Res.-Earth, 125, e2019JF005318,
10.1029/2019JF005318, 2020.Van Wessem, J. M., van de Berg, W. J., Noël, B. P. Y., van Meijgaard, E., Amory, C., Birnbaum, G., Jakobs, C. L., Krüger, K., Lenaerts, J. T. M., Lhermitte, S., Ligtenberg, S. R. M., Medley, B., Reijmer, C. H., van Tricht, K., Trusel, L. D., van Ulft, L. H., Wouters, B., Wuite, J., and van den Broeke, M. R.: Modelling the climate and surface mass balance of polar ice sheets using RACMO2 Part 2: Antarctica (1979–2016), The Cryosphere, 12, 1479–1498, 10.5194/tc-12-1479-2018, 2018.Wouters, B., Bamber, J. L., van den Broeke, M. R., Lenaerts, J. T. M., and
Sasgen, I.: Limits in detecting acceleration of ice sheet mass loss due to
climate variability, Nat. Geosci., 6, 613–616,
10.1038/ngeo1874, 2013.