Peak refreezing in the Greenland firn layer under future warming scenarios

This data set includes the materials required to reproduce the figures and tables presented in the study: "Peak refreezing in the Greenland firn layer under future warming scenarios". The data consist of:

1. Time series of annual Greenland ice sheet (GrIS) integrated SMB components (Gigatons or Gt per year). These time series are available in ASCII format for all simulations presented in the manuscript.

RACMO2.3p2-ERA: 1 member

SMB-components_RACMO2.3p2-ERA_1958-2020_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the ERA-forced RACMO2.3p2 simulation at 5.5 km, statistically downscaled to 1 km spatial resolution (1958-2020).

RACMO2.3p2-CESM2: 3 members

SMB-components_RACMO2.3p2-CESM2-HIST-12_1950-2014_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the CESM2-forced RACMO2.3p2 historical reconstruction (HIST) at 11 km, statistically downscaled to 1 km spatial resolution (1950-2014). RACMO2.3p2 was forced by member 12 of the CESM2-HIST ensemble (HIST-12, see CESM2-HIST below).

SMB-components_RACMO2.3p2-CESM2-SSP126-4_2015-2099_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the CESM2-forced RACMO2.3p2 SSP1-2.6 projection (SSP126) at 11 km, statistically downscaled to 1 km spatial resolution (2015-2099). RACMO2.3p2 was forced by member 4 of the CESM2-SSP126 ensemble (SSP126-4, see CESM2-SSP126 below).

SMB-components_RACMO2.3p2-CESM2-SSP585-3_2015-2099_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the CESM2-forced RACMO2.3p2 SSP5-8.5 projection (SSP585) at 11 km, statistically downscaled to 1 km spatial resolution (2015-2099). RACMO2.3p2 was forced by member 3 of the CESM2-SSP585 ensemble (SSP585-3, see CESM2-SSP585 below).

CESM2-IND: 11 members

SMB-components_CESM2-IND-X_1850-1949_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native CESM2 pre-industrial reconstructions (IND) at ~111 km, statistically downscaled to 1 km spatial resolution (1850-1949). The term “X” in the filename above represents the CESM2 member ranging from 1 to 11.

CESM2-HIST: 12 members

SMB-components_CESM2-HIST-X_1950-2014_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native CESM2 historical reconstructions (HIST) at ~111 km, statistically downscaled to 1 km spatial resolution (1950-2014). The term “X” in the filename above represents the CESM2 member ranging from 1 to 12.

CESM2-SSP126: 6 members

SMB-components_CESM2-SSP126-X_2015-2099_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native CESM2 SSP1-2.6 projections (SSP126) at ~111 km, statistically downscaled to 1 km spatial resolution (2015-2099). The term “X” in the filename above represents the CESM2 member ranging from 1 to 5.

SMB-components_CESM2-SSP126-6_2100-2299_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native long-term CESM2 SSP1-2.6 projection (SSP126) at ~111 km, statistically downscaled to 1 km spatial resolution (2100-2299).

CESM2-SSP245: 6 members

SMB-components_CESM2-SSP245-X_2015-2099_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native CESM2 SSP2-4.5 projections (SSP245) at ~111 km, statistically downscaled to 1 km spatial resolution (2015-2099). The term “X” in the filename above represents the CESM2 member ranging from 1 to 6.

CESM2-SSP370: 5 members

SMB-components_CESM2-SSP370-X_2015-2099_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native CESM2 SSP3-7.0 projections (SSP370) at ~111 km, statistically downscaled to 1 km spatial resolution (2015-2099). The term “X” in the filename above represents the CESM2 member ranging from 1 to 5.

CESM2-SSP585: 8 members

SMB-components_CESM2-SSP585-X_2015-2099_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native CESM2 SSP5-8.5 projections (SSP585) at ~111 km, statistically downscaled to 1 km spatial resolution (2015-2099). The term “X” in the filename above represents the CESM2 member ranging from 1 to 6.

SMB-components_CESM2-SSP585-7_2100-2299_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native long-term CESM2 SSP5-8.5 projection (SSP585) at ~111 km, statistically downscaled to 1 km spatial resolution (2100-2299). This run (member 7) does not include ice dynamics.

SMB-components_CESM2-CISM2-SSP585-8_2100-2299_GrIS_1km.txt: time series of annual GrIS-integrated SMB, total precipitation (snow + rain), snowfall, runoff, total melt (ice + snow), refreezing and retention (Gt per year) from the native long-term CESM2-CISM2 SSP5-8.5 projection (SSP585) at ~111 km, statistically downscaled to 1 km spatial resolution (2100-2299). This run (member 8) includes ice dynamics.

2. Time series of annual GrIS-wide runoff line altitude in meters above sea-level (m a.s.l.). These time series are available in ASCII format for the RACMO2.3p2-ERA simulation (ERA), and as an ensemble mean for all pre-industrial (IND) and historical reconstructions (HIST), and projections (SSP) including both native CESM2 and RACMO2.3p2-CESM2, statistically downscaled to 1 km.

Runoff-line-altitude_ERA_1958-2020_GrIS_1km.txt: time series of GrIS-wide runoff line altitude (m a.s.l.) derived from the ERA-forced RACMO2.3p2 simulation at 1 km (1958-2020).

Runoff-line-altitude_IND-EnsembleMean_1850-1949_GrIS_1km.txt: time series of ensemble mean GrIS-wide runoff line altitude (m a.s.l.) derived from all pre-industrial reconstructions at 1 km (1850-1949, 11 IND members).

Runoff-line-altitude_HIST-EnsembleMean_1950-2014_GrIS_1km.txt: time series of ensemble mean GrIS-wide runoff line altitude (m a.s.l.) derived from all historical reconstructions at 1 km (1950-2014, 13 HIST members).

Runoff-line-altitude_SSP126-EnsembleMean_2015-2299_GrIS_1km.txt: time series of ensemble mean GrIS-wide runoff line altitude (m a.s.l.) derived from all SSP1-2.6 projections at 1 km (2015-2299, 7 SSP126 members).

Runoff-line-altitude_SSP245-EnsembleMean_2015-2099_GrIS_1km.txt: time series of ensemble mean GrIS-wide runoff line altitude (m a.s.l.) derived from all SSP2-4.5 projections at 1 km (2015-2099, 6 SSP245 members).

Runoff-line-altitude_SSP370-EnsembleMean_2015-2099_GrIS_1km.txt: time series of ensemble mean GrIS-wide runoff line altitude (m a.s.l.) derived from all SSP3-7.0 projections at 1 km (2015-2099, 5 SSP245 members).

Runoff-line-altitude_SSP585-EnsembleMean_2015-2299_GrIS_1km.txt: time series of ensemble mean GrIS-wide runoff line altitude (m a.s.l.) derived from all SSP5-8.5 projections at 1 km (2015-2299, 9 SSP585 members).

3. Time series of annual 500hPa global temperature anomalies (ºC). These time series are available in ASCII format for the ERA5 reanalysis (ERA5-reanalysis), and as an ensemble mean for all pre-industrial (IND) and historical reconstructions (HIST), and projections (SSP) from the native CESM2 model at ~111 km spatial resolution.

Tglobal-500hPa-anomaly_ERA5-reanalysis_1950-2020.txt: time series of annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from the ERA5 climate reanalysis (1950-2020). Anomalies are estimated relative to the reference period 1950-1990.

Tglobal-500hPa-anomaly_IND-EnsembleMean_1850-1949.txt: time series of ensemble mean annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from all native CESM2 pre-industrial reconstructions (1850-1949, 11 IND members). Anomalies are estimated relative to the reference period 1850-1949.

Tglobal-500hPa-anomaly_HIST-EnsembleMean_1950-2014.txt: time series of ensemble mean annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from all native CESM2 historical reconstructions (1950-2014, 12 HIST members). Anomalies are estimated relative to the reference period 1850-1949.

Tglobal-500hPa-anomaly_SSP126-EnsembleMean_2015-2299.txt: time series of ensemble mean annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from all native CESM2 short/long term SSP1-2.6 projections (2015-2299, 6 SSP126 members). Anomalies are estimated relative to the reference period 1850-1949.

Tglobal-500hPa-anomaly_SSP245-EnsembleMean_2015-2099.txt: time series of ensemble mean annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from all native CESM2 SSP2-4.5 projections (2015-2099, 6 SSP245 members). Anomalies are estimated relative to the reference period 1850-1949.

Tglobal-500hPa-anomaly_SSP370-EnsembleMean_2015-2099.txt: time series of ensemble mean annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from all native CESM2 SSP3-7.0 projections (2015-2099, 5 SSP370 members). Anomalies are estimated relative to the reference period 1850-1949.

Tglobal-500hPa-anomaly_SSP585-EnsembleMean_2015-2299.txt: time series of ensemble mean annual anomalies in upper atmospheric (500hPa) global temperature (ºC) derived from all native CESM2 short/long term SSP5-8.5 projections (2015-2299, 8 SSP585 members). Anomalies are estimated relative to the reference period 1850-1949.

The gridded, daily downscaled SMB data sets from the ERA-forced RACMO2.3p2 simulation, and the CESM2-forced RACMO2.3p2 projections under a low-end SSP1-2.6 and high-end SSP5-8.5 warming scenario, as well as gridded, monthly downscaled SMB data sets from native CESM2 under pre-industrial (IND), historical (HIST), and short/long term climate projections (SSPs) are freely available from the authors upon request and without conditions (contact: [e-mail]). Besides SMB, the data sets include total precipitation (snow and rain), snowfall, total melt (snow and ice), runoff, refreezing and retention and total sublimation (surface and drifting snow) at 1 km horizontal resolution. 

Abstract:  Firn (compressed snow) covers approximately 90% of the Greenland ice sheet (GrIS) and currently retains about half of rain and meltwater through refreezing, reducing runoff and subsequent mass loss. The loss of firn could mark a tipping point for sustained GrIS mass loss, since decades to centuries of cold summers would be required to rebuild the firn buffer. Here we estimate the warming required for GrIS firn to reach peak refreezing, using 51 climate simulations statistically downscaled to 1 km resolution, that project the long-term firn layer evolution under multiple emission scenarios (1850–2300). We predict that refreezing stabilises under low warming scenarios, whereas under extreme warming, refreezing could peak and permanently decline starting in southwest Greenland by 2100, and further expanding GrIS-wide in the early 22nd century. After passing this peak, the GrIS contribution to global sea level rise would increase over twenty-fold compared to the last three decades.