the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Dec 2025
North Atlantic response to a quasi-realistic Greenland meltwater forcing in eddy-rich EC-Earth3P-VHR hosing simulations
Abstract. The vast majority of studies examining the impact of freshwater from ice sheet melting on the Atlantic Meridional Overturning Circulation (AMOC) use climate models that cannot resolve mesoscale ocean processes and do not include an accurate spatio-temporal distribution of the freshwater forcing. These two factors critically affect the nature of the AMOC response. Our study fills that gap with a set of three hosing experiments performed with the global configurations of the eddy-rich climate model EC-Earth3P-VHR. The model is forced for 21 years with a spatial and monthly distribution of Greenland meltwater fluxes derived from observations, equal to 0.04 Sv on an annual average.
Within the first year, we observe a response of reduced salinity in the Greenland and Labrador currents. This is accompanied by an acceleration and a cooling along the currents that lead to a rapid weakening of the AMOC at subpolar latitudes. Around year 7, deep mixing in the Labrador Sea begins to weaken due to as freshwater anomalies accumulate through lateral exchanges with the boundary currents. This shallowing of the mixed layer further weakens the AMOC, resulting in a stronger reduction that reaches also the subtropical latitudes. By the end of the simulation, the AMOC has weakened by almost 3 Sv at subppolar latitudes (i.e. a decrease of around 20 %), with an average relative decrease of 10 % for the whole Northern Hemisphere. The reduction in the AMOC is strong enough for some global climate impacts to emerge, such as the “bipolar seesaw” temperature response.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(8836 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(8836 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-5882', Anonymous Referee #1, 03 Jan 2026
-
AC1: 'Reply on RC1', Eneko Martin-Martinez, 17 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5882/egusphere-2025-5882-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Eneko Martin-Martinez, 17 Apr 2026
-
RC2: 'Comment on egusphere-2025-5882', Anonymous Referee #2, 09 Jan 2026
In their manuscript “North Atlantic response to a quasi-realistic Greenland meltwater forcing in eddy-rich EC-Earth3P-VHR hosing simulations” Eneko Martin-Martinez and co-authors present results from a common Greenland hosing experiment though with a more realistic, moderate magnitude of 0.04 Sv and spatial distribution of the additional freshwater flux along the coast of Greenland. Most importantly, they use a climate model of very high resolution in ocean and atmosphere unprecedented for such experiment to my knowledge. Unfortunately, this limits the experiment length to only 21 year. However, they ran three ensemble members for estimating the imprint of internal variability on the response. The presentation of results is focused on the response by the AMOC and potential drivers of its decline.
The experiment is certainly timely as the discussion of Greenland ice sheet melting potentially impacting AMOC strength over the next decades is still ongoing. The experimental design is reasonable though slightly different to previous studies (of which numerous are referenced), which limits comparability and differences to previous studies cannot be attributed to model uncertainty unambiguously. On the one hand, the results align with previous findings, thus are supportive of the state of knowledge. Which is good. On the other hand, this means there is not much groundbreaking novelty inherent to the outcome. Maybe because of this, I have the impression that the authors do not really take full advantage of the unique part of their experimental setup, the model resolution. Overall the results section is largely descriptive, few connections are drawn, less are proven, and the analysis is mostly limited to surface fields. It would be helpful if the authors would connect the responses by indicating responsible processes, even if this is speculative (which should be noted then). And by putting more emphasis on the role of mesoscale features simulated by the high resolution of the model. Arguments related to explicitly resolved eddies, eddy mixing and sharper gradients at fronts are rather qualitative but could in principle be quantified. I wish there would be more specific analysis regarding, for instance, freshwater spreading and associated timescales, changes in surface fluxes (e.g. heat loss), and water mass transformation. How do these shape the initial decline of the AMOC?
The other main criticism I have is that there is little validation of the model provided. The authors use the OSNAP section in Figure 5 to show the depth structure of changes in the subpolar North Atlantic. This would be an opportunity to show the mean state of the reference run in comparison to OSNAP observation, at least in the supplementary material. Some discussion of ocean biases is found in Moreno-Chamarro et al. (2025), which is referenced in line 77. Please emphasize this. Nevertheless, a specific discussion of biases with respect to the subpolar North Atlantic would be a great addition to the paper since the model seems to perform rather well (judging from global plots in above reference).
Considering a major revision and sharpening of the arguments, maybe some additional analysis emphasizing more the role of the explicitly resolved mesoscale dynamics (beyond reducing biases), this paper could become a valuable contribution and suitable for publication in OS.
Minor comments by line
93 Separating lines of the six sub-drainage basins could be added to Figure 1b
109 “…by matching the years between …” I am not sure I understand this sentence. Do you mean that you compute differences by subtracting the experiment from control year-wise, i.e. not comparing the experiment to a long-term mean of control? Please rephrase. The interactive atmosphere would lead to independently evolving internal variability. The shortness of the simulations (21 years) suggests however, that low-frequent variability is likely still preserved. This could be noted here.
124 and 129 The latitudes given (33.8N and 60.2N), are these locations of max AMOC in control or of the maximum change in experiment minus control? This is somewhat unclear here. Please specify.
132 “The ensemble-mean difference …” Please provide numbers here. Judging from Figure 2 I’d say there is a significant difference with an AMOC weakening of 1.5 and 2 Sv respectively. Also, simply comparing the last year is likely misleading and an average weakening of at least the last 5 year should be given.
134 not sure which “ratio” is meant hear. Isn’t this a simple difference?
135 replace “happens” with “is located”
139 remove last part of sentence: “…, that is when the AMOC…” Due to the shortness of the runs, it remains unclear whether the AMOC response is already fully developed.
143 add “… by 1.4 Sv on decadal average” at the end of this sentence.
145 replace “lowest latitudes” with “subtropical region”
146 there appears to be something missing at the end of the sentence: “…, representing a 7%.” of what?
148 remove “the” from “…AMOC in the density space” and “… than in the depth space”.
151 Which water mass is simulated in this density range?
154 impact “on” not “in”
154 replace first occurrence of “reduction” with “weakening” to avoid repeat wording in same sentence
165 add “-“ to “boundary-interior”
167 The buoyancy loss could also happen over the path of the NAC in the eastern North Atlantic where the initial densification of the Atlantic water begins.
175 rephrase last part of sentence: “… Greenland coast, suggesting an absence of any local impact of the meltwater injection.”
178 suggest to rephrase: “This cooling is unlikely driven by export with the Labrador Current, …” and add “s” to “shows” then.
178 SST is not an ideal indicator of the heat content of the Labrador Current. However, I agree that the SST change in the NAC is likely indication of an AMOC weakening.
186 remove “depth” from “mixed layer depth (Fog. 4c).” and remove “,which implies that the impact is relatively high.” Leave such judgement to the reader.
188 “, where the mixing also weakens” should be dropped for readability of the sentence.
189 From the thin contour lines in Figure 4c it is very difficult to see whether the controls run has deep convection in the Irminger and Nordic Seas. Please state this more clearly.
190f The deceleration of the gyre by missing deep convection could be countered by a stronger density gradient to the boundary current, which freshened (but also cooled).
193 Please rephrase: “This pattern does not … of the Gulf Stream, as no significant changes in the mean latitudinal position of the current is found when estimating its position from SST and SSH gradients (not shown).”
196 “with a slowing of the Golf Stream” and “temperature and salinity in[!] the NAC”
197 “change of[!] the wind stress” and replace “discards” with “excludes”
198 “, which therefore may only be tied to a reduction…”
230 What is the origin or cause of this warming of the DWBC?
232 replace “since” with “from”
239 How are these numbers (0.4 psu and 1˚C) diagnosed, base don which years?
241 replace “since” with “from”
246 I agree with the current acceleration but this seems to have a different timescale than the response in temperature and salinity. T and S show a rather rapid adjustment during the first 6 years whereas the Labrador Current exhibits a gradual, cumulative increase in speed. Please explain, eventually speculate, and emphasize this difference.
250 “… 5 years of the simulation (marked by dots).”
253 and 258 Why is the NAC responding later than the Norwegian Current? Isn’t the Norwegian Current a continuation of the NAC and would thus lag the ANC response? Please explain otherwise this appears to be inconsistent.
266 replace “gets consistently reduced over time” with “is permanently reduced after year 6”
267 Shallow mixed layers in the Nordic and Irminger Seas are not worth mentioning since there is no deep convection there in control either – except for there would be an expectation for deep convection to occur under hosing. This would need to be explained though.
282 Another reason could be internal variability of surface heat fluxes forced by the coupled atmosphere.
314 Suggest to summarize the main process of this teleconnection as explained in Diamond et al (2025). The wave train is mentioned in line 336. This is too late.
327ff It is not clear to me how this teleconnection works. As just mentioned, please provide some more insight here likely to be summarized from Diamond et al. (2025).
334 replace “previously see over” with “highlighted above for”
364 “…significant impacts on the short 20-year time scale of the experiments, like …”
368 I am missing an item summarizing the key findings using a VHR model. What is specifically enabled by having an eddy-rich ocean and correspondingly well resolved atmosphere? This is the unique selling point of the paper!
397 (Appendix A) “…, many global 3D temperature (but not SST) data files …” I assume here that SST was available and that this is the reason why your analysis is mostly surface focused, right?
407 “… that the intensity of the anomalies … anomalies are greater.” I do not understand this sentence. Please rephrase. Are these two different anomalies? What is smaller, what is greater?
Figures:
Figure 2: Please add bold dots for the start values of each of the three experiments in panel (a) to highlight the three different AMOC states chosen.
Figure 4 The thin contour lines indicating the control climatology are color coded. Please provide a color legend for this. Alternatively use black lines and provide few labels and an increment (in the caption).
Figure 4d Is the sea ice edge in the hosing experiment really that far south as the stipling indicates? This seems unrealistic.
Figure 7d Showing Irminger and Nordic seas here only makes sense if you also discuss a potential east- and northward migration of the deep convection.
Figure B2 Please increase line width of the contour lines indicating the control state for improved visibility. Similarly for Figure 4 only that in the latter there is less space for this.
Citation: https://doi.org/10.5194/egusphere-2025-5882-RC2 -
AC2: 'Reply on RC2', Eneko Martin-Martinez, 17 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5882/egusphere-2025-5882-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Eneko Martin-Martinez, 17 Apr 2026
-
EC1: 'Comment on egusphere-2025-5882', Sjoerd Groeskamp, 05 Feb 2026
My apologies for the late reply.
First I want to thank the reviewers for their effort reviewing this manuscript.Both reports provide positive feedback on this paper and consider it important work that will become worthy for publication after some substantial revisions.
I therefore invite the authors to provide detailed point by point response to the reviewers comments and upload a revised version of their manuscript for consideration for publication in Ocean Sciences.Citation: https://doi.org/10.5194/egusphere-2025-5882-EC1
Peer review completion
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-5882', Anonymous Referee #1, 03 Jan 2026
Please find attached my general recommendation along with my major and minor comments.
-
AC1: 'Reply on RC1', Eneko Martin-Martinez, 17 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5882/egusphere-2025-5882-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Eneko Martin-Martinez, 17 Apr 2026
-
RC2: 'Comment on egusphere-2025-5882', Anonymous Referee #2, 09 Jan 2026
In their manuscript “North Atlantic response to a quasi-realistic Greenland meltwater forcing in eddy-rich EC-Earth3P-VHR hosing simulations” Eneko Martin-Martinez and co-authors present results from a common Greenland hosing experiment though with a more realistic, moderate magnitude of 0.04 Sv and spatial distribution of the additional freshwater flux along the coast of Greenland. Most importantly, they use a climate model of very high resolution in ocean and atmosphere unprecedented for such experiment to my knowledge. Unfortunately, this limits the experiment length to only 21 year. However, they ran three ensemble members for estimating the imprint of internal variability on the response. The presentation of results is focused on the response by the AMOC and potential drivers of its decline.
The experiment is certainly timely as the discussion of Greenland ice sheet melting potentially impacting AMOC strength over the next decades is still ongoing. The experimental design is reasonable though slightly different to previous studies (of which numerous are referenced), which limits comparability and differences to previous studies cannot be attributed to model uncertainty unambiguously. On the one hand, the results align with previous findings, thus are supportive of the state of knowledge. Which is good. On the other hand, this means there is not much groundbreaking novelty inherent to the outcome. Maybe because of this, I have the impression that the authors do not really take full advantage of the unique part of their experimental setup, the model resolution. Overall the results section is largely descriptive, few connections are drawn, less are proven, and the analysis is mostly limited to surface fields. It would be helpful if the authors would connect the responses by indicating responsible processes, even if this is speculative (which should be noted then). And by putting more emphasis on the role of mesoscale features simulated by the high resolution of the model. Arguments related to explicitly resolved eddies, eddy mixing and sharper gradients at fronts are rather qualitative but could in principle be quantified. I wish there would be more specific analysis regarding, for instance, freshwater spreading and associated timescales, changes in surface fluxes (e.g. heat loss), and water mass transformation. How do these shape the initial decline of the AMOC?
The other main criticism I have is that there is little validation of the model provided. The authors use the OSNAP section in Figure 5 to show the depth structure of changes in the subpolar North Atlantic. This would be an opportunity to show the mean state of the reference run in comparison to OSNAP observation, at least in the supplementary material. Some discussion of ocean biases is found in Moreno-Chamarro et al. (2025), which is referenced in line 77. Please emphasize this. Nevertheless, a specific discussion of biases with respect to the subpolar North Atlantic would be a great addition to the paper since the model seems to perform rather well (judging from global plots in above reference).
Considering a major revision and sharpening of the arguments, maybe some additional analysis emphasizing more the role of the explicitly resolved mesoscale dynamics (beyond reducing biases), this paper could become a valuable contribution and suitable for publication in OS.
Minor comments by line
93 Separating lines of the six sub-drainage basins could be added to Figure 1b
109 “…by matching the years between …” I am not sure I understand this sentence. Do you mean that you compute differences by subtracting the experiment from control year-wise, i.e. not comparing the experiment to a long-term mean of control? Please rephrase. The interactive atmosphere would lead to independently evolving internal variability. The shortness of the simulations (21 years) suggests however, that low-frequent variability is likely still preserved. This could be noted here.
124 and 129 The latitudes given (33.8N and 60.2N), are these locations of max AMOC in control or of the maximum change in experiment minus control? This is somewhat unclear here. Please specify.
132 “The ensemble-mean difference …” Please provide numbers here. Judging from Figure 2 I’d say there is a significant difference with an AMOC weakening of 1.5 and 2 Sv respectively. Also, simply comparing the last year is likely misleading and an average weakening of at least the last 5 year should be given.
134 not sure which “ratio” is meant hear. Isn’t this a simple difference?
135 replace “happens” with “is located”
139 remove last part of sentence: “…, that is when the AMOC…” Due to the shortness of the runs, it remains unclear whether the AMOC response is already fully developed.
143 add “… by 1.4 Sv on decadal average” at the end of this sentence.
145 replace “lowest latitudes” with “subtropical region”
146 there appears to be something missing at the end of the sentence: “…, representing a 7%.” of what?
148 remove “the” from “…AMOC in the density space” and “… than in the depth space”.
151 Which water mass is simulated in this density range?
154 impact “on” not “in”
154 replace first occurrence of “reduction” with “weakening” to avoid repeat wording in same sentence
165 add “-“ to “boundary-interior”
167 The buoyancy loss could also happen over the path of the NAC in the eastern North Atlantic where the initial densification of the Atlantic water begins.
175 rephrase last part of sentence: “… Greenland coast, suggesting an absence of any local impact of the meltwater injection.”
178 suggest to rephrase: “This cooling is unlikely driven by export with the Labrador Current, …” and add “s” to “shows” then.
178 SST is not an ideal indicator of the heat content of the Labrador Current. However, I agree that the SST change in the NAC is likely indication of an AMOC weakening.
186 remove “depth” from “mixed layer depth (Fog. 4c).” and remove “,which implies that the impact is relatively high.” Leave such judgement to the reader.
188 “, where the mixing also weakens” should be dropped for readability of the sentence.
189 From the thin contour lines in Figure 4c it is very difficult to see whether the controls run has deep convection in the Irminger and Nordic Seas. Please state this more clearly.
190f The deceleration of the gyre by missing deep convection could be countered by a stronger density gradient to the boundary current, which freshened (but also cooled).
193 Please rephrase: “This pattern does not … of the Gulf Stream, as no significant changes in the mean latitudinal position of the current is found when estimating its position from SST and SSH gradients (not shown).”
196 “with a slowing of the Golf Stream” and “temperature and salinity in[!] the NAC”
197 “change of[!] the wind stress” and replace “discards” with “excludes”
198 “, which therefore may only be tied to a reduction…”
230 What is the origin or cause of this warming of the DWBC?
232 replace “since” with “from”
239 How are these numbers (0.4 psu and 1˚C) diagnosed, base don which years?
241 replace “since” with “from”
246 I agree with the current acceleration but this seems to have a different timescale than the response in temperature and salinity. T and S show a rather rapid adjustment during the first 6 years whereas the Labrador Current exhibits a gradual, cumulative increase in speed. Please explain, eventually speculate, and emphasize this difference.
250 “… 5 years of the simulation (marked by dots).”
253 and 258 Why is the NAC responding later than the Norwegian Current? Isn’t the Norwegian Current a continuation of the NAC and would thus lag the ANC response? Please explain otherwise this appears to be inconsistent.
266 replace “gets consistently reduced over time” with “is permanently reduced after year 6”
267 Shallow mixed layers in the Nordic and Irminger Seas are not worth mentioning since there is no deep convection there in control either – except for there would be an expectation for deep convection to occur under hosing. This would need to be explained though.
282 Another reason could be internal variability of surface heat fluxes forced by the coupled atmosphere.
314 Suggest to summarize the main process of this teleconnection as explained in Diamond et al (2025). The wave train is mentioned in line 336. This is too late.
327ff It is not clear to me how this teleconnection works. As just mentioned, please provide some more insight here likely to be summarized from Diamond et al. (2025).
334 replace “previously see over” with “highlighted above for”
364 “…significant impacts on the short 20-year time scale of the experiments, like …”
368 I am missing an item summarizing the key findings using a VHR model. What is specifically enabled by having an eddy-rich ocean and correspondingly well resolved atmosphere? This is the unique selling point of the paper!
397 (Appendix A) “…, many global 3D temperature (but not SST) data files …” I assume here that SST was available and that this is the reason why your analysis is mostly surface focused, right?
407 “… that the intensity of the anomalies … anomalies are greater.” I do not understand this sentence. Please rephrase. Are these two different anomalies? What is smaller, what is greater?
Figures:
Figure 2: Please add bold dots for the start values of each of the three experiments in panel (a) to highlight the three different AMOC states chosen.
Figure 4 The thin contour lines indicating the control climatology are color coded. Please provide a color legend for this. Alternatively use black lines and provide few labels and an increment (in the caption).
Figure 4d Is the sea ice edge in the hosing experiment really that far south as the stipling indicates? This seems unrealistic.
Figure 7d Showing Irminger and Nordic seas here only makes sense if you also discuss a potential east- and northward migration of the deep convection.
Figure B2 Please increase line width of the contour lines indicating the control state for improved visibility. Similarly for Figure 4 only that in the latter there is less space for this.
Citation: https://doi.org/10.5194/egusphere-2025-5882-RC2 -
AC2: 'Reply on RC2', Eneko Martin-Martinez, 17 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5882/egusphere-2025-5882-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Eneko Martin-Martinez, 17 Apr 2026
-
EC1: 'Comment on egusphere-2025-5882', Sjoerd Groeskamp, 05 Feb 2026
My apologies for the late reply.
First I want to thank the reviewers for their effort reviewing this manuscript.Both reports provide positive feedback on this paper and consider it important work that will become worthy for publication after some substantial revisions.
I therefore invite the authors to provide detailed point by point response to the reviewers comments and upload a revised version of their manuscript for consideration for publication in Ocean Sciences.Citation: https://doi.org/10.5194/egusphere-2025-5882-EC1
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,958 | 1,415 | 170 | 3,543 | 189 | 163 |
- HTML: 1,958
- PDF: 1,415
- XML: 170
- Total: 3,543
- BibTeX: 189
- EndNote: 163
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Eneko Martin-Martinez
Eduardo Moreno-Chamarro
Fraser William Goldsworth
Jin-Song von Storch
Cristina Arumí-Planas
Daria Kuznetsova
Saskia Loosveldt-Tomas
Pierre-Antoine Bretonnière
Pablo Ortega
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(8836 KB) - Metadata XML
Please find attached my general recommendation along with my major and minor comments.