Skip to content

GitLab

Last edited by David Docquier
Page history

SST restoring experiments

Back to Home.

Go to Global experiments.

Go to Atlantic and Pacific SST experiments.

In this set of experiments, I restored the sea-surface temperature (SST) in a selected region to a certain value and compared the results of these sensitivity experiments to a control run. The control run starts from the year 2014 of the CMIP6 EC-Earth3-Veg r1 historical member (T255-ORCA1), uses a 2000-year constant atmospheric forcing and is 200 years long. The sensitivity experiments all start from the year 2130 of the control run (year 117 if we consider the starting year to be 1) and are 50 years long.

I used 3 different domains for the SST restoring experiments in the North Atlantic Ocean:

  1. North Atlantic, between 40N (latitude of Madrid) and 80N (latitude of Svalbard) and between 40W (longitude of Central Atlantic) and 20E (longitude of BSO)
  2. Northern North Atlantic, between 66N (latitude of Iceland) and 80N (latitude of Svalbard)) and between 22W (longitude of East Greenland) and 20E (longitude of BSO)
  3. Barents Sea Opening (BSO), between 70N and 77N and between 15E and 22E.

I have applied a SST restoring to the mean climatology of the period 2100-2129 (30 years) in my control run plus 1°C, 3°C and 5°C, respectively, for the 3 different domains described above, resulting in 9 different sensitivity experiments, which can be compared to the control run. In order not to have too sharp gradients in SST (and other resulting variables), I have applied a 9-point smoothing at the borders of the domains.

Based on the SST restoring information, the model computes the corresponding surface heat flux correction in the domain of interest. More information about the methodology can be found in Ruprich-Robert et al. (2017).

Additionally, I have applied the same methodology to 3 different domains in the North Pacific Ocean:

  1. North Pacific, between 30N (latitude of northern Mexico) and 66N (latitude of Bering Strait) and between 120E (longitude of eastern China) and 120W (longitude of western USA)
  2. Northeastern Pacific, between 40N (latitude of northern California) and 50N (latitude of southern Canada) and between 180W (longitude of Central Pacific) and 120W (longitude of western USA)
  3. Bering Strait, between 64N and 67N and between 172W and 166W.

For the 3 Pacific SST experiments, I have applied a SST restoring similar as for the Atlantic SST experiments using an increase in SST of 3°C.

Below, I focus on the 9 Atlantic SST experiments. To see the results comparing Atlantic and Pacific SST experiments, go to Atlantic and Pacific SST experiments.

Here are the different model experiments with the corresponding color in the plots below:

  • t613 (red solid curve, run by K. Wyser) is the CMIP6 EC-Earth3-Veg r1 historical member covering 1850-2014 (only the period 1950-2014 is shown here);
  • D000 (blue solid curve) is the control run starting from year 2014 of the t613 experiment with 2000-year constant atmospheric forcing; this control run is 200 years long; all the sensitivity experiments start in year 2130 of this control run (year 117 if we consider the starting year to be 1) and are 50 years long;
  • D012 (green dashed curve) starts in year 2130 of the control run (D000); the only change compared to the control run is that the SST is restored to the mean climatology increased by 3°C in the North Atlantic Ocean (40-80N, 40W-20E);
  • D013 (red dashed curve) is similar to D012, except that the SST is restored to the mean climatology increased by 1°C;
  • D014 (gray dashed curve) is similar to D012, except that the SST is restored to the mean climatology increased by 5°C;
  • D015 (orange dashed curve) starts in year 2130 of the control run (D000); the only change compared to the control run is that the SST is restored to the mean climatology increased by 3°C in the northern North Atlantic Ocean (66-80N, 22W-20E);
  • D016 (magenta dashed curve) is similar to D015, except that the SST is restored to the mean climatology increased by 1°C;
  • D017 (cyan dashed curve) is similar to D015, except that the SST is restored to the mean climatology increased by 5°C;
  • D018 (yellow dashed curve) starts in year 2130 of the control run (D000); the only change compared to the control run is that the SST is restored to the mean climatology increased by 3°C at the Barents Sea Opening (70-77N, 16-20E);
  • D019 (brown dashed curve) is similar to D018, except that the SST is restored to the mean climatology increased by 1°C;
  • D020 (purple dashed curve) is similar to D018, except that the SST is restored to the mean climatology increased by 5°C;

Results show that:

  • Fig. 1: the mean Atlantic OHT increases north of 60N and decreases south of 60N compared to the control run in the experiments where the SST restoring is imposed in the North Atlantic and northern North Atlantic (no clear change with the restoring occurring only at the BSO); the OHT increase north of 60N is logical as the SST is getting warmer; the reason for the decrease south of 60N is probably linked to a weakening of the ocean circulation at latitudes lower than 60N following the increase in SST (so decrease in density) (see Fig. 8);
  • Fig. 2: the total OHT to the Arctic through all Arctic straits (BSO, Fram, Bering, Davis) clearly increases with increasing SST; the increase in OHT is larger with higher SST and is clearly larger with the North Atlantic domain compared to the 2 other domains; with the BSO domain, the OHT increase is relatively small; so the larger the domain for the SST increase, the higher the resulting OHT increase;
  • Fig. 3: this is the OHT through the BSO, which is the most important OHT contributor to the Arctic; results are broadly consistent with Fig. 2, except that the increase in OHT does not linearly increase with the increase in SST when the BSO domain (smallest domain) is considered;
  • Fig. 4: there is an overall OHT increase at Fram Strait with SST when using the 2 first (larger) domains and not really when using the BSO domain; however, the OHT increase is much lower than for the BSO and the increase occurs in the first 20 years of the simulation before decreasing to values close or below the control run;
  • the OHT at the Bering Strait also increases with increasing SST for all 3 domains, but the strength of the increase is much lower compared to BSO and Fram; and the changes in OHT at Davis Strait are very small in all experiments (not shown);
  • Figs. 5 and 6: the total Arctic-sea ice area (SIA) and volume (SIV) decrease in all experiments with higher SST, in agreement with the OHT increase; the SIA decrease is larger with increased SST for the 2 first domains, but not for the BSO domain (where the strength of the SIA decrease is relatively similar between the 3 experiments);
  • Fig. 7: the spatial extent of the SST increase follows the domain used for the SST restoring, with a larger spatial extent in the SST increase with the larger domain;
  • Fig. 8: the mixed layer depth in the Labrador, northern North Atlantic and Greenland-Iceland-Norway (GIN) Seas decreases with higher SST for the 2 larger domains (North Atlantic and northern North Atlantic); this probably explains the OHT decrease south of 60N, following a weakening of the overturning circulation;
  • Fig. 9: the mean March sea-ice concentration clearly decreases in the Barents and GIN Seas in the North Atlantic and northern North Atlantic experiments (top and middle rows) and this decrease is much less pronounced in the case of the BSO experiment (bottom row); furthermore, increasing the SST in the North Atlantic experiment (top row, from left to right) clearly decreases the sea-ice concentration beyond the Barents Sea (through the Kara Sea) and also in the Bering Strait, interestingly.
  • Fig. 10: the mean March sea-ice thickness decreases everywhere in the Arctic in all experiments, especially in the North Atlantic experiments (top row); as for sea-ice concentration, increasing the SST in the North Atlantic experiment (top row, from left to right) clearly decreases the sea-ice thickness, but this sea-ice thinning clearly spans the whole Arctic;
  • Fig. 11: the Atlantic meridional stream function (AMSF) strongly decreases in the North Atlantic experiments (top row), decreases in the northern North Atlantic experiments (middle row) and does not substantially change in the BSO experiments (bottom row); in the North Atlantic and northern North Atlantic experiments, the higher the SST increase, the stronger the decrease in AMSF.

OHT_50yrs_SST Fig. 1: Latitudinal transect of mean Atlantic OHT averaged over 50 years (except 2000-2014 for the CMIP6 r1 member to compare to TF2017); OHT estimates from Trenberth and Fasullo (TF2017) are plotted for reference; the number in brackets is the difference in mean OHT between the experiment and the control (CTRL)

OHT_Total_timeseries_SST Fig. 2: Time series of annual mean OHT through all Arctic straits (BSO + Fram Strait + Bering Strait + Davis Strait); the number in brackets is the difference in mean OHT between the experiment and the control (CTRL)

OHT_BSO_timeseries_SST Fig. 3: Time series of annual mean OHT at the Barents Sea Opening (BSO); the number in brackets is the difference in mean OHT between the experiment and the control (CTRL)

OHT_Fram_timeseries_SST Fig. 4: Time series of annual mean OHT at the Fram Strait; the number in brackets is the difference in mean OHT between the experiment and the control (CTRL)

SIA_Cycle_SST Fig. 5: Mean seasonal cycle of Arctic sea-ice area averaged over 50 years for the control run and sensitivity experiments and over 1979-2014 for the CMIP6 r1 member and satellite observations; the numbers in brackets are the differences in mean March and September Arctic sea-ice areas between the experiment and the control (CTRL)

SIV_Cycle_SST Fig. 6: Mean seasonal cycle of Arctic sea-ice volume averaged over 50 years for the control run and sensitivity experiments and over 1979-2014 for the CMIP6 r1 member and PIOMAS reanalysis; the numbers in brackets are the differences in mean March and September Arctic sea-ice volumes between the experiment and the control (CTRL)

Maps_March_SST_diff Fig. 7: Maps of difference in mean March SST between the different sensitivity experiments and the control run, averaged over 50 years

Maps_March_MLD_diff Fig. 8: Maps of difference in mean March mixed layer depth between the control run and the different sensitivity experiments, averaged over 50 years

Maps_March_siconc_diff Fig. 9: Maps of difference in mean March sea-ice concentration between the different sensitivity experiments and the control run, averaged over 50 years

Maps_March_sivol_diff Fig. 10: Maps of difference in mean March sea-ice thickness between the different sensitivity experiments and the control run, averaged over 50 years

zomsfatl_50yrs_diff_SST Fig. 11: Vertical profiles of difference in mean Atlantic meridional stream function (AMSF) between the different sensitivity experiments and the control run, averaged for each latitudinal band and over 50 years

Important note: These figures were produced by D. Docquier and constitute preliminary results (not published in papers).