1. What is the projected mean warming for Greenland Ice Sheet in SSP5-8.5 projections?
The projected mean warming for the Greenland Ice Sheet in SSP5-8.5 projections is 5.3 K for the 21st century. This significant increase in temperature is expected to lead to a reduction in the Greenland Ice Sheet's surface mass balance (SMB), resulting in more runoff and ultimately raising global sea levels. The conservative estimate of the equivalent sea level rise amounts to more than 10 cm by the end of this century. Surface mass balance models (EBMs) are used to infer changes in SMB from basic surface climate variables, providing a low-cost alternative to computationally intensive regional climate model simulations. However, the ERA-Interim reanalysis product, which was previously used, has been replaced by the higher resolution ERA5 reanalysis product, necessitating an update to current SMB simulations and adjustments to existing EBM parameters.
read more
2. How do ERA5 and ERAI compare in terms of spatial resolution and coverage period?
ERAI covers the period from January 1979 to August 2019 with a spatial resolution of about 80 km over Greenland, while ERA5 begins in January 1959, runs until the present, and has a finer resolution of about 30 km. ERA5 provides more recent and detailed data compared to ERAI, allowing for a more comprehensive analysis of climate properties over the Greenland Ice Sheet (GrIS).
read more
3. What are the differences between ERA5 and ERAI in summer?
During the summer months, ERA5 and ERAI exhibit pronounced differences in variables considered. ERA5 shows mean summer 2m-air temperatures (T2M) that are more than 1 * C colder than ERAI over most parts of the ice sheet, except the South Eastern margins and the southern dome region. The mean bias exceeds two standard deviations of the interannual variability almost everywhere north of 66 * N. ERA5 has a stronger shortwave downward radiation at the surface (SWD) than ERAI over the main ice sheet, with a mean bias exceeding two standard deviations on the lower parts of the ice sheet. ERA5 has a slightly higher emissivity deviation from ERA-Interim in most parts of the ice sheets, except for a pronounced negative bias at the central-eastern margins. The lower parts of the ice sheet have mostly lower emissivity values in ERA5, which is consistent with the pronounced positive shortwave radiation bias. However, these features are not accompanied by a correspondingly lower cloud cover in ERA5. Comparing ERA5 and ERAI with AWS measurements shows no significant bias in ERA5 temperatures, while ERAI temperatures are significantly warmer than observations. Applying the lapse rate correction of 5 K km -1 reduces the spread of the reanalysis data around the observational data but also reinforces the warm bias in ERAI. Over the lower parts of the ice sheet (< 2000 m), differences in ERA5 and ERAI 2m-air temperature vary around a mean of -1.0 K with a standard deviation of 0.24 K. This bias is enhanced by 25% during the period between 2002-2009. The differences in SWD and T2M between ERA5 and ERAI yield a range of melt rates from -0.25 mm day -1 for a low albedo of A = 0.4 in the dark bare ice zone to -2 mm day -1 for a fresh snow albedo of A = 0.9. This results in a lower equilibrium line and stronger melt gradients between the equilibrium line and the ice sheet's margin under ERA5 forcing.
read more
4. What are the differences between ERA5 and ERA-Interim in terms of surface energy balance?
The comparison between ERA5 and ERA-Interim reveals substantial and temporally coherent differences in surface energy balance over the GrIS. ERA5 is characterized by systematically colder near-surface temperatures and more intense insolation in summer. The difference in shortwave radiation downward (SWD) is particularly pronounced along the lower parts of the ice sheets, where ERA5's higher spatial resolution better represents the steep orography. Correcting the near-surface temperatures with a lapse rate of -5 K km-1 reduces the differences between the two reanalysis products and improves the comparison with monthly observations from PROMICE weather stations for both data sets. This result is consistent with slope lapse rates diagnosed from both data sets and highlights the benefit of this simple downscaling method when dealing with coarse-resolution temperature fields. In contrast to Delhasse et al. (2020), a significant warm bias of ERAI relative to weather station data is found, but this is only fully evident when a lapse rate correction is applied, while SWD appears to be slightly overestimated in ERA5. The observed differences between ERA5 and ERAI have implications for the estimation of surface melt and ultimately the release of runoff. Replacing ERAI with ERA5 forcing in an energy balance model of the GrIS may require re-calibration to reproduce existing observations (e.g., IMBIE Team, 2020).
read more