 Okay, so good morning everybody. So yeah, I will Go back to to Volcanic activity we've seen a very nice introduction yesterday and try to dig into possible impacts on the Atlantic right you know, but turn in circulation viability so a rapid impact of volcanic eruptions on the climate just to recall what this is a Just illustration from Driscoll et al We can see from from different reanalysis It induces quite rapidly an intensification of the stratospheric polar vortex surface and I would like anomalous atmospheric circulation and winter warming over Eurasia these these Impacts are very rapid one or two years after the eruption Can we think of a longer lasting impact of volcanic eruptions on the ocean? one of the first Clue for significant decadal scale impact of volcanic eruptions on the ocean has been given by church et al So this is a mixture of observations and simulations And you can see that both for global ocean heat content and global mean sea level You see some decadal variations in the time series that seems to be like happening Concomitantly with the the eruptions that occurred over the historical periods with a more Idealized experimental setting Stenchikov went a little step further in these issues. So here he shows the anomalous Sea surface temperature after a Tambora like eruption that occurred in 1816 and a more moderate eruption that occurred in 1991 the Pinatubo eruption you can see that in case of a very quite strong eruption here the surface at this the Signal at the surface of the ocean seems to persist for about 10 years It's maybe a bit shorter for weaker eruptions and if you look into the deep ocean. These are ocean heat content anomaly The top 300 meters in blue the whole depths in in green and in black Sorry, and the the below 300 meters in red you can see that it seems that Volcanic signal can persist for longer than the cali time scales in the deep ocean What happens for the moc? So this is a very brief literature review So in the same experimental setup that the same paper Stenchikov et al. He saw in both cases for both Weak and a little bit stronger eruptions and intensification of the moc Within the cali time scales. This is also consistent with this famous paper by otter a et al Well here they used a simulation of the last millennium and they look at composites of the moc response after several eruptions And they also see this intensification. This is also consistent with Last millennium simulations run with the echo g model. So all this seems to be fine We are happy this seems to be consistent. However, I also noted. Oh, yeah I'm sorry in most of these papers the mechanism associated to this intensification is linked to precisely the NAO inducing enhancement of the deep convection in the North Atlantic and intensification of the moc However, the literature also shows several of several different responses. We've already seen this this graph yesterday But I think it's interesting. This is the so here it's in an ensemble of simulation David noted the pre-eruption value of the moc against the well the post-eruption anomaly against the pre-eruption value And you can see a clear tendency that like at least a clear Impact of the initial state on on the response Another study performed with had CM3 showed an intensification of the moc here After the kakatao eruption, but not after the pinna tubo eruption So there seems to be a sensitivity to the eruption perhaps the magnitude of the eruption and finally this paper I think also have been have been discussed slightly yesterday In the CCSM model These these authors showed that after a sequence of decadally paced Intense volcanic event. There seems to be a persistent Response of the sea ice volume in the Arctic which then caps Deep convection in the in the northern North Atlantic and induces a weakening of the moc after these volcanic events so All this doesn't seem to be that clear now Volcanic eruptions in the past. So which are they this is a record of the illustration of the global average radiative forcing associated to volcanoes here in black as compared to greenhouse gases over the Last 40 years most famous eruptions are the agung el chichon and pinna tubo Eruptions and now if we look here I try to keep the same scale for the global average radiative forcing and this is what happens over the last millennium and I think I can convince you that the past Thousand years can be a relevant period for looking at the impact of volcanic eruptions First of all because there are much more and of course the period is longer and also because they are much stronger So this is what I'm going to do. I'm going to focus on the last millennium simulation and I will use Mostly to a family of two models that we run at the IPSL The former version that was used for see me three and the the newer version used that see me five The oceans are very similar. The atmosphere has a slightly improved increased resolution but still these are These two simulations of the last millennium are run independently So this is what we get for example just as an illustration. So here at the top I show the forcing in terms of solar variation The forcing in terms of volcanic eruptions and these are results the air temperature in the northern North Atlantic and the sea surface Temperature in the Atlantic just to convince you that these are really independent simulations Of course, it's the same family of models But we have this so you of course these time series They don't completely superpose because of internal variability We can try to compare them to proxy So this is a proxy taken here in the north north of Iceland reconstruction of SST from Alkenon And these are these are the time series that we get from the from the proxy So of course, it's very difficult to To try to match the wiggles I think one can at least see here this nice drop that seems to be concomitant with this very strong Samalah eruption in 1258 and Perhaps a drop here associated to the Tambora eruption Back to our topic today the MOC So this is the MOC maximum of the MOC that we get in the two simulations So again, is there a synchronization? Is there a phasing of the MOC at all? Is there a consistent response of the MOC to the external forcing? It's hard to tell from these So what I what I did? If it works. Yeah, what I did here is compute Correlation over a sliding windows between the two MOC time series And here is the significant level and you can see that there seems to be Periods over the last millennium where these the MOC are simulated by these two more or less independent Models or at least simulations the MOC seems to be synchronized or at least positively correlated And if you look a bit closer, I think one can say that this corresponds to periods of intense volcanic activity First here around the 13th century and then perhaps later here around the 17th century With a small effect of the school I eruption here perhaps Okay, is this robust at all so I leave this as an open question I tried to take the MOC from the last millennium simulations I could find on the on the ESG server and so I and I compare so I compare CCSM For model MPI model my both IPSL models and yeah, that's it So I have four of them and I try to correlate the MOC two by two to see if there is any Synchronization of the MOC I would be interested to discuss this with you I'm not sure about the response here. I highlight the same periods as I highlighted for the IPSL CM 5 There are some some periods where there seems to be a synchronization for example here between IPSL CM 4 and CCSM 4 Perhaps slightly here one can see at least an intensification of the correlation Perhaps here CCSM 4 and IPSL CM 5 so CCSM 4 seems to be a nice friend of IPSL models But well, this is I don't really have the answer and I come back to to the IPSL model and in order to gain understanding in the mechanisms that could lead to this synchronization I compute composite analysis over selected eruptions And this is what we get so in face with the eruption This is the composite of the Ekman pumping term Together with the MOC immediate response. So this is a very rapid Adjustment what happens because of this NAO like response There is an Ekman anomalous Ekman pumping here in the northern North Atlantic around 50 degrees north and this leads to a very rapid Adjustment biotropic adjustment of the MOC simply as a response to the dynamical Well a dynamical response to the anomalous atmospheric circulation Okay This I just just to note that this is this has been found with other models So this seems to be a quite a robust issue Now at longer time scales what happens because of these by model correlation that I had seen in the sliding Correlation I had to cut this these periods of the last millennium into two two different periods so first if I concentrate on these intense and cumulative events rapidly rapid if each simulation following I mean Eruptions following each other very rapidly I see in both models a reduction of the MOC So if you remember what I showed at the very beginning This is quite opposite to what several other authors have found and this is due in fact to here I show a few years after the eruption the anomalous mix layer in colors and the winter ice cover in in in contours and you can see that quite a long time after the eruption there is a Weakening of the unknown a weakening of the anomalous winter mix layer So weakening of the deep convection due to sea ice capping persistent sea ice capping in the northern North Atlantic The eruptions are so strong and Following each other so rapidly that this that the cooling is very strong in the northern at North Atlantic the sea ice invades the Nordic seas and the Supola Atlantic Ocean and these skills or at least reduces the the deep convection Okay, and now if if now I concentrate on the weaker volcanic eruptions that are also more Spaced and so the system perhaps this is just on hypothesis has more time to adjust after each Eruption I have more or less a consistent response in both simulations with an intensification of the MOC the structures are not the same the response seems to be slightly dependent on the model perhaps on the eruption and But oops sorry for this figure But as supposed to show the anomalous mixed layer depth and the sst a few years in this model here IPSL Five and six years after the eruption and I hope you can see here in in in colors This is the anomalous mixed layer. There is an intensification a deepening of the of the mixed layer associated to a Surface well which induces a surface warming so it's really an intensification of the convection After this NaO like response that helps to induce this acceleration of the MOC We've where we've already talked about this this study yesterday It's it's the the study that I shown is also consistent over the whole last millennium if you only pick up Pinatubo like eruptions. So here we had a very Constraining criteria. We had to pick up Pinatubo like eruptions that were not followed or preceded by very strong volcanic eruptions So we only picked up five selected volcano and this is the composite response of the MOC in the same simulation And I think one can see a consistent Intensification of the MOC something like 10 to 15 years after the eruption So this is a consistent response Interestingly, we found a similar synchronization of the MOC in the historical ensemble of one of these model IPS LCM 5a So this is the MOC in an ensemble of five Control simulations. So absolutely no forcing you can see that no timescale or synchronization seems to appear here from these time series There is no forcing and here is the historical simulation So just the control plus all the observed forcing over the past Decades including volcanic eruptions as I've mentioned at the beginning. We have an eruption here in 1962 the Agoon and one in 1982 the El Chichon eruption and I think you are convinced that here There is there seems to be a synchronization in this ensemble Although again the the simulations are completely independent and there seems to be an Intensification of the MOC some 15 years after the the eruption This is this is quite consistent also with this simip 5 ensemble If you select a sub sample of the models that has a similar Viability timescale as the IPSL model in this case. So this is the red line This is the average of eight others see me five model, which we picked up because they have the similar preferred timescale in the North Atlantic viability and here you can see that There is a very nice synchronization in the historical simulations again It's important to say that here. There is no nudging. There is no observations except for external forcing and and there is quite a nice synchronization, I think with the IPSL model So with this I shall conclude So there is indeed a decal timescale response Decal timescale response. Sorry of the ocean temperature to volcanic events Simulated MOC over the last thousand years shows some synchronization among models. I would be interested to have your Feelings about this and I think if this happens, it's mostly during periods of volcanic activity Yet beyond the robust dynamical adjustment at very short timescale the response of the MOC at the scale time scale seems to depend strongly on the volcanic eruption magnitude, so During for other response to strong and cumulative eruptions. There is a weakening rather of the MOC I think linked to sea ice capping While as a response to more moderate eruptions either from the recent period or even from the last millennium There is rather an intensification of the MOC So I have two two two points as as challenges or perspective for the future. I think these simulations I mean these results call for a more systematic assessment so sensitivity Assessment of sensitivity of the response to magnitude season of the eruption So there is much more to do and and I think perhaps this will be Addressed within volmip. There is also we've discussed this very briefly a need for updated for things So this is a recent paper That that appeared so he in this in this graph here the the purple shading shows the response of the of the of the models to Volcanoes I mean the reconstructed response of the models to volcano over the northern hemisphere as a response as after this eruption in 1258 and And the the the lines here are the tree ring Reconstructions so you can see that the models they largely overestimate the response the northern hemisphere cooling after this eruption both after the 1258 eruption and the Tambora eruption and this is a new Module to that takes into account microphysical processes In the stratosphere for simulating the impacts of the volcano and you can see now a much more agreement between the models and the Proxies, thank you very much