EGU2020-20368
https://doi.org/10.5194/egusphere-egu2020-20368
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.The North Atlantic Oscillation and the Greenland ice sheet in CMIP6
- 1Danish Meteorological Institute, Research and Development, København, Denmark (rum@dmi.dk)
- 2Institute for geodosey and geoinformation, University of Bonn
- 4Geodesy and Earth Observation, DTU-Space, Danish Technical University, Lyngby, Denmark
The North Atlantic Oscillation (NAO) is an important control on both northern European weather and Greenland ice sheet surface mass budget via the path of storm tracks that deliver precipitation, particularly in the winter, and by the strength and persistence of the Greenland blocking high that promotes melt in summer. Within CMIP5 models, atmospheric blocking was generally poorly represented regardless of location, we here examine an ensemble of 10 CMIP6 fully coupled earth system models (ESMs) that were available by Summer 2019 in order to examine if model improvements better represent the NAO in CMIP6.
We examine temperature over Greenland and the north Atlantic region as well as NAO position, persistence and strength in winter and summer for each model in the historical scenario. No single model performs well on all characteristics but the UKESM and EC-EARTH3 perform the best when compared to the ERA5 climate reanalysis.
We also show how the NAO is expected to change in 8 of these models under different future climate scenarios. The location of the Icelandic low in particular migrates northwards by varying amounts, likely related to Arctic sea ice changes within the models and with a consequent impact on precipitation.
Downscaling experiments carried out using the HIRHAM5 regional climate model over the Greenland ice sheet show the importance of accurately characterising the NAO in order to correctly estimate both winter accumulation and summer melt and the combination that gives the ice sheet mass budget. Our study emphasises the importance of assessing a range of different climate and weather variables when selecting models to downscale for obtaining ice sheet mass balance. We also note that while some progress has been made in better representing atmospheric blocking in ESMs, largely down to higher resolution in atmospheric models, there is still a substantial improvement required before ESMs can be said to accurately characterise the climate of the North Atlantic region with consequent impacts on ice sheet surface mass budget projections.
How to cite: Mottram, R., Ascheneller, S., Sauerland, F., Anker Pedersen, R., Thejll, P., Lang Langen, P., Boberg, F., Stendel, M., Hansen, N., and Yang, S.: The North Atlantic Oscillation and the Greenland ice sheet in CMIP6, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20368, https://doi.org/10.5194/egusphere-egu2020-20368, 2020
Comments on the display material
AC: Author Comment | CC: Community Comment | Report abuse
Dear Ruth,
This is a very interesting presentation. And I also like the fact that you mention that looking at the different members would be useful.
As I understand, you're looking at the CMIP6 historical and SSP scenario runs of 10 GCMs.
Could you clarify how you derived the Greenland Ice Sheet SMB in Slide 3 please?
Thanks,
David
HI David,
Thanks for your comment.
The SMB in slide 3 is taken from our daily polarportal SMB product (see: ) the atmosphere is from the numerical weather prediction model HARMONIE-AROME that DMI runs for the Greenalnd weather forecast. We use output from that to drive an SMB model offline. The SMB model is the same as that we use with the HIRHAM5 regional climate model and that gives the background climatology also shown on the polarportal.
The SMB model is described in detail in Langen et al., 2017 and the HARMONIE-AROME model in Bengtsson et al 2017 () though both have been updated slightly since. All our data is freely available for scientific use here: )
OK thanks for your reply and the information.
So I guess the right terminology for the models you used is AOGCM (and not ESM).
David