Atmospheric blocking occurs when near-stationary high-pressure systems divert westerly flow for a week or more. It can cause extreme regional weather such as heatwaves in summer and cold spells in winter. In this article we explore the connection between blocking over Greenland and enhanced surface melting of the ice sheet in UKESM1.ice N96 ORCA1 present-day run. UKESM1.ice is based on a modified version of GC3.1 (Kuhlbrodt et al., 2018) that allows a two-way coupling to the ice sheet model BISICLES (Cornford et al., 2013). In particular, the modified land-surface model JULES (Best et al. 2011) uses a more sophisticated representation of the ice sheet surface on multiple elevation bands to downscale atmospheric forcing and an improved albedo scheme (Shannon et al., 2019). The run, which forms part of the UKESM contribution to Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) (Nowicki et al., 2016), starts from 1970 and finishes at the end of 2014.

We measure blocking over Greenland using the Greenland blocking index (GBI) which is defined as the 500 hPa geopotential height averaged over the region 60-80°N, 20-80°W (Hanna et al., 2013, 2016). The advantages of using GBI are that it is simple to calculate and daily NCAR/NCEP reanalysis data is available at NOAA from 1948 to present-day. We compare monthly means of GBI rather than daily means because global climate models, particularly low resolution models such as ours, underestimate the frequency and persistence of synoptic blocking events (Woollings et al., 2018 and Schiemann et al., 2020). Figure 1a suggests that the model run can capture Greenland blocking from 1980 to 2014 reasonably well on averaged over a month, where the mean GBI is 5317 m in the model and 5301 m for the reanalysis and the root-mean-square-error between the two is 83 m.

Figure 1: Comparison of Greenland blocking index (GBI) between present-day model (blue) and NCAR/NCEP reanalysis data (red) for monthly means (a) and summer means (b).

Summer GBI is of interest because anomalously high pressure over Greenland has been linked to accelerated surface melting on the ice sheet in recent summers (Hanna et al., 2013, 2016). Theory suggests that blocking can advect relatively warm air masses from the subtropics, disperse cloud cover and bring drier weather. Comparing the modelled summer GBI, where the index is averaged between June and August (JJA), with the reanalysis (see figure 1b) the mean of timeseries match where values are 5502 m and 5501 m, respectively, but the model has less variability than the reanalysis, where the standard deviations are 21.6 m and 28.2 m, respectively. The reanalysis time series has a positive trend from the mid-1990s, whereas the modelled times series has no significant trend. The model does show warming over the ice sheet, where the annual air temperature 1.5 m above the ice surface is increasing at 0.12°C/y since 1995. CMIP5 models also fail to capture the positive blocking trend (Hanna et al., 2018), although Woollings et al. (2018) argues that the trend is not distinguishable from natural internal variability.

The top three summer GBI values in the model run occur in years 2004, 1991 and 2011, which correspond to maxima in the mean surface meltrate over the ice sheet (see figure 2). Positive anomalies in GBI also correspond to positive anomalies in meltrate in the 2000s and there is a linear correlation for the whole timeseries with a coefficient of 0.76 and p-value < 1.0e-6. How is blocking increasing ice sheet melt?

Figure 2: Timeseries of summer means of Greenland blocking index and surface meltrate, albedo, cloud fraction and air temperature at 1.5 m averaged over the ice sheet with the timeseries mean (dotted black line). Red circles mark the years for the 3 top GBI values.

Ice sheet melt is calculated using the energy balance fed by air temperature, humidity, wind speed, precipitation and downward shortwave and longwave radiation in a multilayer snowpack model (Best et al., 2011, Shannon et al., 2019). Albedo, which represents the ratio of reflected to incoming solar radiation on the surface of the ice, is parameterised in the model using the density, age and grain size of the snowpack and varies in value between that of ice and fresh snow. The three years where the summer GBI is highest coincide with those of the three lowest albedo values (see figure 2). There is also a strong correlation between the two with a coefficient of –0.81 and p-value < 1.0e-6. Atmospheric blocking can affect cloud cover which impacts solar radiation reaching the ice sheet. This may explain the correlation between GBI and albedo and is backed up by a study by Hofer et al. (2017) that found a strong, negative correlation between GBI and cloud cover since the mid-1990s using a combination of satellite observations of clouds and regional climate modelling. They suggest increased summer blocking is reducing cloud cover over southern Greenland allowing more shortwave radiation to reach the surface. Longwave radiation also increases from cloud cover over the northeast. Both of which drive the observed accelerated ice sheet melt since the mid-1990s (Hahn et al., 2020). In the present-day model run the cloud fraction anomaly is negative for 1991 and 2004 (see figure 2) with reduced cloud over most of the ice sheet, but in 2011 there is increased cloud over the interior and west in the northern half and over the southern tip. Also, in 2004 incoming shortwave radiation is not anomalous, but the longwave radiation anomaly is positive everywhere. Clearly, there is no consistent pattern with cloud in the model run. There is only a weak correlation between GBI and cloud fraction timeseries with coefficient –0.52 and p-value = 0.001 and there is a positive trend in cloud fraction since 1990 that contradicts satellite observations showing decreasing cloud. More importantly, there is no significant correlation between meltrate and cloud fraction timeseries in the model. It may be that the location of cloud cover is more relevant rather than the average amount over the whole ice sheet.

How is albedo related to GBI if not through cloud cover? The connection between albedo and GBI could be made indirectly through melt as GBI is correlated to net shortwave radiation, the amount of shortwave radiation absorbed by the ice, with coefficient 0.72 and p-value < 1.0e-5 rather than to incoming shortwave radiation. The mean summer meltrate and albedo have a strong correlation with coefficient of –0.83 and p-value < 1.0e-6. This can be explained by melt-albedo feedback where meltwater decreases albedo by increasing the grainsize of the ice and surface melt increases when the ratio of absorbed solar radiation increases (Box et al., 2012).

Atmospheric blocking, depending on the location of the block, can also advect warm air on to the ice sheet that will drive surface melting. Figure 2 shows that the three peaks in meltrate correspond to peaks in air temperature 1.5 above the surface of the ice. There is a very strong positive correlation between two with a coefficient of 0.92 and a p-value < 1.0e-6, although the correlation between summer air temperature and GBI is weaker with a coefficient of 0.67 and p-value < 1.0e-4. Figure 3 shows the mean summer 500 hPa geopotential height for 1985, when both the GBI and mean meltrate are both a minimum, 1991, 2004 and 2011. In 1985 there is low pressure off the northwest coast of Greenland bringing air down from the north in cyclonic (anti-clockwise) flow, where the flow moves along the pressure contours. In the high melt years there is a high over southern Greenland in 1991 which migrates off the southeast coast in the other two years leaving anti-cyclonic flow to the west to advect air from the south.

Figure 3: Mean summer geopotential height at 500 hPa in metres with white contours every 25 m for years 1985, 1991, 2004 and 2011. GBI is calculated inside the region marked by magenta lines. Black contours mark the edge of land.

It is perhaps not surprising that GBI is high when the air over the ice sheet is relatively warm given that geopotential height can be expressed as a function of virtual temperature of the air column. However, the air temperature at 1.5m and meltrate in 2011 appear to be outliers when plotted against GBI. The meltrate and temperature are much higher than the linear fit with GBI predicts. In 2011 the ice sheet is relatively warm almost everywhere and the meltrate is also above average in the interior and not just at the margins. Clearly, warm air is the dominate driver of melt, but the anti-cyclonic flow brings cloud and rain, and alters the sensible heat flux that increases melt too. Rain has a positive trend since the 1990s and reaches a maximum in 2011 where the anomalies are located over the ablation zone along the western margin in the northern half of the ice sheet and southern tip. Sensible heat flux is enhanced along a section of southern half of the ablation zone.

Anomalous atmospheric blocking in summer is linked to enhanced surface melt over Greenland in the UKESM1.ice present-day run. We have rejected the idea that clearer skies due to blocking are responsible for the enhanced melt. We found that GBI is most strongly correlated to albedo. It is likely that albedo is at its lowest when GBI is highest simply because there is more meltwater around, though this does not explain why the correlation is higher than the correlation between GBI and meltrate. Near surface air temperature is the dominant driver of meltrate and it appears that atmospheric blocking is bringing it on to the ice sheet. Blocking also modifies spatial patterns of cloud cover, rain and sensible heat flux that can contribute to the enhance melt. The findings in this article are based on one model run. We need to analyse more runs to demonstrate that the link is more than a happy coincidence. We also need to separate out the recent warming to identify the natural variability of the model using long control runs with and without coupling to the ice sheet. Should the link between blocking and melt prove robust in the model then we can investigate its effect along with possible orographic effects (Hahn et al., 2020) from an evolving ice sheet on sea-level rise projections from emissions scenarios. Some caution is needed though as projected total melt from Greenland is very sensitive to cloud microphysics (Hofer et al. 2019). Almost every aspect of the phenomenon may change in the future: the frequency and/or the location of the block, the temperature of the air and the shape of Greenland.

References

• Best, M. J., Pryor, M., Clark, D. B., Rooney, G. G., Essery, R. L. H., Ménard, C. B., Edwards, J. M., Hendry, M. A., Porson, A., Gedney, N., Mercado, L. M., Sitch, S., Blyth, E., Boucher, O., Cox, P. M., Grimmond, C. S. B., and Harding, R. J.: The Joint UK Land Environment Simulator (JULES), model description – Part 1: Energy and water fluxes, Geosci. Model Dev., 4, 677–699, https://doi.org/10.5194/gmd-4-677-2011, 2011.
• Box, J. E., Fettweis, X., Stroeve, J. C., Tedesco, M., Hall, D. K., and Steffen, K.: Greenland ice sheet albedo feedback: thermodynamics and atmospheric drivers, The Cryosphere, 6, 821–839, https://doi.org/10.5194/tc-6-821-2012, 2012.
• Cornford, S. L., Martin, D. F., Graves, D. T., Ranken, D. F., Le Brocq, A. M., Gladstone, R. M., Payne, A. J., Ng, E. G., & Lipscomb, W. H.: Adaptive mesh, finite volume modeling of marine ice sheets. Journal of Computational Physics, 232(1), 529-549. https://doi.org/10.1016/j.jcp.2012.08.037, 2013.
• Hahn, L. C., T. Storelvmo, S. Hofer, R. Parfitt, and C. C. Ummenhofer: Importance of Orography for Greenland Cloud and Melt Response to Atmospheric Blocking. J. Climate, 33, 4187–4206, https://doi.org/10.1175/JCLI-D-19-0527.1, 2020.
• Hanna E., J.M. Jones, J. Cappelen, S.H. Mernild, L. Wood, K. Steffen, P. Huybrechts: The influence of North Atlantic atmospheric and oceanic forcing effects on 1900–2010 Greenland summer climate and ice melt/runoff. International Journal of Climatology, 33(4), 862-880, https://doi.org/10.1002/joc.3475, 2013.
• Hanna, E., T. E. Cropper, R. J. Hall and J. Cappelen: Greenland Blocking Index 1851-2015: a regional climate change signal. International Journal of Climatology, 36(15), 4847-4861, https://doi.org/10.1002/joc.4673, 2016.
• Hanna, E., X. Fettweis, and R.J. Hall: Brief communication: Recent changes in summer Greenland blocking captured by none of the CMIP5 models. The Cryosphere, 12, 3287–3292, https://doi.org/10.5194/tc-12-3287-2018, 2018.
• Hofer, S., A. J. Tedstone, X. Fettweis and J. L. Bamber: Decreasing cloud cover drives the recent mass loss on the Greenland Ice Sheet. Science Advances, 3, no. 6, e1700584
  DOI: 10.1126/sciadv.1700584, 2017.
• Hofer, S., Tedstone, A.J., Fettweis, X. and Bamber, J.L.: Cloud microphysics and circulation anomalies control differences in future Greenland melt. Nat. Clim. Chang. 9, 523–528, https://doi.org/10.1038/s41558-019-0507-8, 2019.
• Kuhlbrodt, T., Jones, C., Sellar, A., Storkey, D., Blockley, E., Stringer, M., Hill, R., Graham, T., Ridley, J., Blaker, A., Calvert, D., Copsey, D., Ellis, R., Hewitt, H., Hyder, P., Ineson, S., Mulcahy, J., Siahaan, A., Walton, J.: The Low-Resolution Version of HadGEM3 GC3.1: Development and Evaluation for Global Climate. Journal of Advances in Modeling Earth Systems, 10(11), 2865-2888, https://doi.org/10.1029/2018MS001370, 2018.
• Nowicki, S. M. J., Payne, A., Larour, E., Seroussi, H., Goelzer, H., Lipscomb, W., Gregory, J., Abe-Ouchi, A., and Shepherd, A.: Ice Sheet Model Intercomparison Project (ISMIP6) contribution to CMIP6, Geosci. Model Dev., 9, 4521–4545, https://doi.org/10.5194/gmd-9-4521-2016, 2016.
• Schiemann R., P. Athanasiadis, D. Barriopedro, F. Doblas-Reyes, K. Lohmann, M.J. Roberts,D.V. Sein, C.D. Roberts, L. Terray, and P.L. Vidale: Northern Hemisphere blocking simulation in current climate models: evaluating progress from the Climate Model Intercomparison Project Phase 5 to 6 and sensitivity to resolution. Weather Clim. Dynam., 1, 277–292, https://doi.org/10.5194/wcd-1-277-2020, 2020.
• Shannon, S., Smith, R., Wiltshire, A., Payne, T., Huss, M., Betts, R., Caesar, J., Koutroulis, A., Jones, D., and Harrison, S.: Global glacier volume projections under high-end climate change scenarios, The Cryosphere, 13, 325–350, https://doi.org/10.5194/tc-13-325-2019, 2019.
• Woollings, T., Barriopedro, D., Methven, J. et al. Blocking and its Response to Climate Change. Curr Clim Change Rep 4, 287–300. https://doi.org/10.1007/s40641-018-0108-z, 2018.