 So you can use this. Is this a laser plan? Here? OK. So good morning. I'm Josephina, and I work at the Federal University of Campina Grande de Brasil. And so what I'm going to discuss here with you today are results from work that I've been doing with my students. And they're preliminary. And some of them are quite recent. And I wanted to bring them to an audience who could, I think, contribute a lot. So this is my intention today. So the topic is, does the Canary's current appalling system affect the seasonal migration of the Atlantic ITCZ? And the motivation, well, this is where I work now around here. It's in the semi-arid northeast of Brazil. And this region has one short rainy season that's centered around March-April. These are these two towns here shown. And this short rainy season is due to the approximation of the Atlantic ITCZ. So March-April, that's when the ITCZ attains its southernmost position, because that's when the waters of the South Atlantic are warmest. And that's what brings the rainfall. Because this region is very close to the southernmost limit of the migration of the ITCZ, the rainfall season is very sensitive to the interannual variability in that migration. And so, yeah, so you could say it's similar to the Sahel. So it's like the Brazilian Sahel. So predicting the rainfall season is a big deal for this region. And so I just put two examples of satellite images for two different years, just one specific day. But here we can see this was a dry year, 2015. And the ITCZ is slightly further north than would be desirable. And the northeast is dry. And here on a wet year, the ITCZ came further southwards. And it's a good year. And so actually, the value of the prediction is what is done with the prediction is a whole other matter. Because many times, a lot of the people who live in the interior are subsistence farmers. And often, there is no plant B. So they either plant or they plant. Because they need to run the risk. It's all probabilistic at the end of the day. And they need to run the risk of having a harvest. But they still use the forecast to build resilient strategies so within the families. And the social tissue is a whole other matter. But still, it would be nice to provide good information. So why is the northeast semi-arid if it's downstream from the trade winds all year round? So there's always moisture, a lot of moisture coming in. These figures were not done for this talk. So none of this is the average for the rainy season in itself. But still, we can see that November to March, and then April to June, it's a little bit more humid. And then most of the time, it's quite dry. And there's a lot of moisture coming in from the ocean all the time. And this moisture goes on and converges and produces rainfall further downstream. So why? And then one of the reasons is, I think the main reason is probably subsidence. So this region suffers from subsidence from the Hadley cell. The tip of the South Atlantic subtropical high often is often over that region. And so this is something like the Hadley cell, so long-term in meridional circulation. What averaged over would be the northeast from 45 west to 35 west. And we can see a lot of subsidence close to the equator. So this is why there is a, this is the driest season, but there's a pronounced, there's a prolonged dry season. And this is a situation for March, April, May, which is somewhat better, but due to the approximation of the ITCZ, which is this whole organized system where you have everything together, convergence. So you have a moisture supply, you have upward motion, and then making for lowering the pressure in the low levels and then promoting further convergence of the trade winds and moisture supply and everything. A positive feedback loop that can, how can I say, overcome this subsidence situation. And there's also subsidence, there's often also subsidence from Walker-type circulations because of the intense rainfall in the Western Amazon. So the rain from the northeast depends on the seasonal migration of the ITCZ. And how does this seasonal migration work? So it's the cloud band that goes together with a band of warmest waters over the ocean, the equatorial trough, and the trade wind convergence. All of these features, they don't exactly coincide, but they do move around together. And the rationale is pretty much, as I understand it, is that the warmest waters reduce convective inhibition than pretty much just the wind. Generated turbulence is enough to lift the air to the free convection level, and then it triggers convection and that lowers the low level pressure and forces the trade wind convergence, which provides moisture to convection and so feeds back. So the question is, what controls the extent of the southward migration of the Atlantic ITCZ? And so the standard answer is SST, because SST would look like it's the outer forcing, what is able to reduce the convective inhibition and trigger spontaneous rainfall. And then, OK, so what controls SST? And that would control then the extent of the seasonal migration of the ITCZ. So over the Atlantic, so you can imagine, well, I tend to think of it like if you have two bowls of soup and they're at the same time, or if you have two bowls of soup in a still room, the warmest one will heat the environment more because there will be a larger temperature difference between the liquid and the environment, and so there will be more heat transfer. But if you start to blow on one of the bowls of soup, then after a while, it can be cooler than the other and still transfer more heat because of the turbulent flexes. So you have two mechanisms. One is the larger the SST, the more heat flux you will have to the atmosphere, and the other one is the larger the wind, the more heat flux you will have to the atmosphere. So over the Pacific, it's generally there's ocean dynamics controlling the SST, and then that controls the heat transfer to the atmosphere. Whereas in the Atlantic, where the El Nino is not so pronounced, you generally have the intensity of the trade winds playing a very big role in controlling the SST. And then Chang in 1997, he put this into a context where if you were to have opposing anomalies of SST in the tropical North Atlantic and the tropical South Atlantic, this would promote trade wind anomalies that would make them stronger in the South. This would be the anomalous flow, but this would make the trade winds stronger in the South and weaker in the North. So weaker trades in the North would mean less heat transfer from the ocean to the atmosphere and therefore warmer ocean. And in the South, the more intense trades would mean more heat transfer and therefore cooler ocean. So you would have the SST maybe starting to disturbance. The trade winds responding in a way in which to perpetuate and intensify the disturbance. So there's a lot of talk about the tropical Atlantic SST dipole. And what this paper says is that, well, there is a statistical dipole, there's a statistical dipole signal. But in terms of physical mechanisms, all you have is a positive feedback. If you do have SST disturbances, the wind disturbances will tend to perpetrate and accentuate them. So then, but you do need, you don't have a mechanism to invert the signal. So you don't have a physical mechanism that's capable of inverting. If you have, by chance, an opposite situation in SST, then the mechanism works the other way around. So statistically, you end up getting a dipole, although there's no physical dipole. It's called wind evaporation feedback. Yeah, I think so, yeah. And well, so, but actually, when you look at the correlation, so this is the correlation between wind and the SST gradient and with different lags. And what you see here is that it's all positive. So that's the idea. There's no negative feedback. It's always positive. But the atmosphere, the correlation is larger when the atmosphere precedes the ocean by a month. So generally speaking, these situations will be triggered by the atmosphere. So the atmosphere will lead. There will be anomalies in the trade winds, and then that will start the feedback loop with the ocean responding. So the question becomes, what controls the wind intensity? And then, OK, so this is a summary of a lot of work that was done and a lot of it was done by Haast and Raph. And so one of the things that he, the main thing that he proposes is a control by the Pacific Ocean on the North Atlantic SST, Tropical SST. And that goes through what he calls an atmospheric bridge. So it's a wave train that starts from anomalous convection in the East Pacific. And then what it does is it promotes high level convergence over the Atlantic in a position such that it will weaken the subtropical high. And it's close to its southern boundary, so then it weakens the pressure gradient, and therefore weakens the trade winds. So that's how a warm Pacific would produce a warm Northern Atlantic. And so with the warm Northern Atlantic, this would tend to increase the southern trades and then give a cold Southern Atlantic and shift the position of the ITCC to the North, meaning less rainfall for the Northeast. And then the opposite is also true. When you have a cold phase in the Eastern Pacific, this, the teleconnection is such that you get a cold phase in the tropical North Atlantic. And this, because the North Atlantic's subtropical high becomes stronger, you get stronger trades and then a colder ocean. And the pushes that the ITCC southwards, and you get weaker trades in the South and the warm ocean. So it all comes together. And this is his figure for the atmospheric wage. So here is the signal in 200 millibars geopotential. And this is divergent. So whereas this is Africa. So convergence and, yeah, I guess, divergence here close to over the tropics, which would mean downward motion. And then, yeah, I don't know what phase this is in. Whatever phase this is in. So you're supposed to get divergence in the high levels in order to get a weaker North Atlantic subtropical high and convergence in the high levels to get the stronger subtropical high. I can't really distinguish that well. OK, so this is Hastin-Rach. And then there's this other work by Saravanan and Chang. And what they did was they ran three experiments. They forced the atmosphere with observed assistees. And at the first moment, it was the global ocean forcing the global atmosphere. And then they removed. They used monthly mean climatological assistees for everywhere outside of 30 degrees North and 30 degrees South. So inter-annual variability was only present in the tropics. And then they used that to force the atmosphere. And lastly, they used climatological, well, seasonal monthly mean, long-term mean assistees. Because you still keep the seasonal cycle, but you get rid of the inter-annual variability. So they observed assistees for the tropical Atlantic and then climatological assistees everywhere else. And so they ran these three experiments. And one of the things that they saw was that they then correlated assistees with rainfall over the Northeast. And so the pattern is pretty much identical when you do global ocean, global atmosphere, and tropical ocean, tropical atmosphere. Sorry, tropical ocean, global atmosphere. But when you do the tropical Atlantic global atmosphere, you get the same signs but weaker correlations, especially in the North Atlantic. Now, the atmospheric bridge mechanism is obviously it's still there because you're using observed assistees in the Atlantic. So when the atmospheric bridge was working and it affected the tropical Atlantic assistee that's recorded in the data set, so you're still forcing rainfall with the atmospheric bridge mechanism. But what this points to is that what this suggests is that there's probably another mechanism, too. So you still keep the atmospheric bridge working for the SST. But if there's a direct mechanism from Pacific SST to rainfall without having to go through the Northern Atlantic, then you lose that. And so they wouldn't vary together as well, SST and the rainfall. And so what they propose is when they compare the experiment where the Pacific is also forced with the one where it's only the Atlantic, what they see, it's kind of hard to see. But they see this classical georesponse to a tropical heat source in the atmosphere with the two anticyclones, North and South from the equator. And then they say, OK, so if we have this, we're bound to have subsidence further to the east. And then that would be a promoting drought over the Northeast. So you can think of it as two controls. I think this, like, you can think of the problem as two different controls, although they sometimes interlap, which would be whether the ITCZ gets close enough or doesn't. And then whether you have stronger than usual subsidence in the Northeast or not, because there are other minor systems which will also promote rainfall. And in that case, they would be inhibited. So high-level psychotic fortices. And then later in the year, the easterly waves. So you have these things in play, too. And they would be inhibited by stronger subsidence. OK, so this is the scenario. And then it's well-rounded, and it all makes sense. But then the forecast is really, really bad. I think the forecast is pretty bad. Because, say, this is supposed to be a dry year. Well, and then the forecast is also bad for the other region. So you could argue that, you know. But anyway, this is what called my attention that this is supposed to be a dry year in the Northeast. And still you have, like, a 60% chance of rainfall, average or above average rainfall. So that's uncomfortable. And there is also a political issue. Because you cannot generate panic. If you have a drought forecast for the Northeast, it has a strong emotional effect. But it's still pretty bad. Because when you go to these meetings, these seasonal forecast meetings, you start to notice that it's not just a question of the models not, OK, maybe this is a difficult mechanism for the models. It's very involved. There are different stages. The model has to get a lot of stuff right. But what you see when you go to the meetings is that the forecasters don't really have much more to go on. So they don't have an opinion that can differ from the model very much. All they do is look at the MJO and forecast if it's going to be a signal of downward air motion or upward air motion. And there's not much to do beyond that. So what was our approach? So we thought, OK, maybe if this is all accounted for and it's all understood, maybe the question is, how do these different mechanisms combine and which one prevails in different situations? So maybe that could be helpful. And so what we decided to do was look at individual years as case studies, so individual rainy seasons as case studies, and then test these different mechanisms for the strength and to see which one was prevailing. And so we calculated things like anomalies of rainfall and SST and high-level divergence and heat fluxes and all of these kinds of anomalies to have an idea of what was going on. And so what we found was the following. This is probably the best situation. This is like, so this was 1985. It was actually a strong Blaninha year. And yes, this is anomalous 100 millibar winds. So we have a much stronger North Atlantic subtropical high. And it does seem like it's cooling the tropical Atlantic, right? Yes? Yeah, what did I say? Oh, sorry, 1,000. And it does seem like it's cooling the tropical Atlantic. But then we have this really strong cold tongue. And it's coming from the coast of, I don't know, I'm not sure. Sorry? OK. OK, from the coast. But this is the cannery's current upwelling system. It's quite, I have a picture for it. It's quite wide in terms of latitudes. And this is a part of it. And so it does seem like, so the anomalies are kind of, they're appearing from the coast. And they do seem to get affected and to affect the whole basin, right? And if you notice, the trade winds, yes, they're stronger. And the water is cooler over the whole tropical Atlantic. But it's really from this, from the southern part of these anomalies that the trade winds diverge, that there's anomalous divergence. And then they converge in the South Atlantic where it's warmer. And you see that the rainfall response is very clear. So the correspondence with the rainfall anomalies is very clear. So, yeah. The rainfall blues, am I going to be able to go? Yeah. Oh, because it doesn't, yeah, I'm sorry, yeah. Maybe. Just all you have to say is blue is hot. Yeah, I will, I would, because there's a scale. So, you know, it could be more quantitative. But it's just, it's behind this other figure. Maybe at some other point we'll be able to see the scale. Yeah, so this is dry. And this is wet, right? And this is cold, and this is warm. So, yeah, so this is, it's always March, April, May. So it's always the rainy season for the Northeast. It's always when the ITCZ should be at its, it should be somewhere around here, sorry, here or here. And so these anomalies will tell us how it's shifted. It's shifts around its southernmost position always. Okay, so, and then these are the, doesn't, okay, oh, maybe, okay. Yeah. And here are the heat flux. This is all the heat fluxes together. So, sensible, and latent, and also irradiative. It's everything together. The convention is downward heat flux is positive. And of course, then the field and the tropics will be negative, because it's always the ocean heating the atmosphere. And so if we have a negative anomaly, it means that it's less, sorry, more heat being given from the ocean to the atmosphere, right? And if you have a positive anomaly, that means there's less heat being given from the ocean to the atmosphere. So what we expect, as we said earlier, is if there's, if the mechanism, if the winds are driving the SST anomalies, then we expect upward heat fluxes to be associated with cooler SSDs. And if it's the SST anomalies that are driving the heat flux ones, then you expect cooler SSDs to be related to less upward heat transfer, transport. So over most of the tropical Atlantic, we see probably the winds driving the SST anomalies. As much as this, we cannot, I don't, I'm not that sure how trustworthy this field is, really from the reanalysis. And then, but over exactly where the canneries upwelling anomalies appears to be, we have cooler SSDs associated with less heat being transferred into the atmosphere. So it's not the wind that's cooling the SST over this region, right? Okay, so also this, so this upwelling, one of the important mechanisms is ECMEN transport and it's forced by northeastern trades. So this upwelling, the southern, it's quite wide and the southern part of it has a strong seasonality and it only exists around March, April, a little bit, it's a little bit wider time gap, but it's around March, April, because that's when the SSD, ITCZ is further south and that's when the trade winds in this equatorial region, let's say, have a northeasterly component and then the ECMEN transport will drive the water away from the coast, right? So it does fit well, but then I have to go on and calculate the ECMEN transport, I suppose. But it does seem that it fits well, that we have more upwelling with this anomaly, so this northeastward anomaly. So what could be happening is we could have, yes, the South Atlantic subtropical high influencing the position of the ITCZ, but with an intermediate step, where it forces anomalous upwelling in the cannery's current system and that cools the ocean and that's where the wind diverges from, right? And that's where you get the dry anomalies, so the shift of the ITCZ. This is a picture of the upwelling, so it goes from like 22 north to 10 north, I guess, and they're so divided into different regions, yeah, because they have slightly different mechanisms. But this, I think, this region is the one, this band has the strongest seasonality and it depends on the migration of the ITCZ. But then the question becomes, well, there are several questions, so I'm gonna show you a number of other years. So I had like a 29-year sample, I think it was from 1983 to 2011. And then I extracted from this sample the years where we can see this working in some way, either because the, but as I said, these are preliminary results, so with my student, we had gone from doing a whole bunch of correlations, statistics, and this and that, and not really getting anywhere new with the problem to doing some case studies, so it was, and then we did like five different case studies with these different fields. But then this showed up and then I decided to see if it would be recurrent and then I have figures for these fields for every year, but not for all of the other fields that we were looking at too, so I don't have the heat fluxes for every year, unfortunately. But okay, so this is 1983, this is a very strong El Nino. And then what we see here is possibly a suppression of the upwelling, but we do see here a stronger equatorial upwelling in the Southern Hemisphere. This, I guess, should work a lot like the Pacific, so it would be driven by zonal winds, right? And so there's this cold anomaly and we see the, okay, the subtropical high is weaker than usual and maybe, yes, that does cool the tropical North Atlantic, the tropical North Atlantic, but it does seem that the relevant action takes place closer to the equator where you have the cool anomalies and then the warm anomalies and the divergence and the convergence of the wind. And here there's a clear signal of rainfall suppression, but then we don't get that much enhanced rainfall, but there's a very strong menu, so there could be subsidence. There's more things to check about this year. Oh, I do have the fluxes, yeah. So this region where we have these cold anomalies, they were of less, yeah, less heat transport to the atmosphere, which means to say that it wasn't, this was not wind produced, right? So likely upwelling really. Then there's 1986 and it's always the same story for these years that I've chosen. You do have stuff happening in the tropics, you know, that comes from the extra tropics, but then what really says how the ITCC is going to shift is what happens a lot closer to the equator because you see you have these trade wind anomalies, but they're kind of pretty much the same, the same, the same, and then they start to diverge and then they converge in the southern hemisphere and they seem to be responding to this gradient. This is very much equatorial SST gradient and the rainfall shifts. And then the situation can be the opposite. And so the message is this correspondence between equatorial SST gradient divergence and convergence of the trades in the equatorial strip and the shift in the rainfall. And it just, it works pretty well there for all of these cases, you know, either with the one sign or the opposite. So this looks like an Atlantic El Nino. There was not that much happening in the Northropical Atlantic, but then there's this divergence and it doesn't go very far. It just converges a little bit to the North. And then, yeah, somewhat like the opposite situation. Yeah. And here we just have, we seem to have a suppression of this upwelling. There's not so much happening in the Northern Hemisphere, but then the trade winds respond by flowing into this warmer water and the rainfall responds to. And here it's the same. So suppressed upwelling probably, and then enhanced upwelling. And yeah, I'm not sure, but could be that suppressed, I mean, could be suppressed upwelling here or it could just be the wind in this case, I'm not sure. And then from North to South, and then from South to North, this is clearly suppressed upwelling, I'd say. And then from North to South, and it just works for a number of years. So, I think I have the, yeah. Yeah, so this is suppressed upwelling at least up here. And then, what does the rainfall do? There is a reduced rainfall over, I guess where the upwelling was around climatology and then this was just not there. So the rainfall runs away from the South. Yeah, so it's from 29 years, it's like 17 of them, where you see this equatorial, it seems like it's more of an equatorial dipole than a tropical dipole. That's what it appears to be. And so there are a few questions and I would love any feedback on how to go about them. So one issue is, is this the real main mechanism? Is it an equatorial dipole? And in that case, maybe it's wind convergence as much as SST. And it does seem to work like that for El Nino, where oftentimes you have these very strong anomalies of SST, but when you look at the field itself, for some El Nino's, the gradient is very, very weak. There's not really, there's a warming in the central pacific state, and it just seems like an enormous pool of warm water, but the convection still shifts. So I always thought of it as being that the convection, that you get this spread out of the warm water because of weakening of the trades. And when you weaken the trades, what you also do is you shift the convergence and then you trigger, you just trigger convection at a different point, although there's no real strong SST gradient. And so on the inter-annual time scale, it does seem to, it just seems to happen that you will have a response of the ITCZ to this equatorial dipole, but could it also be that within, for every rainy season, these upwellings, they work as feedbacks where the cannery's current, the southern part of the cannery's current upwelling will exist when the ITCZ shifts to the south, because that's when you get the northeastern trades over the coast of Africa. And then because it appears, then it enhances convergence and convection in the SST when it shifts to the south. So just a part of the seasonal cycle as a positive feedback. And then lastly, it could be that in some cases, the cannery's current upwelling works as an intermediate step between the north Atlantic subtropical high and the ITCZ. So it would be in some cases necessary to have that in order to get the influence from the extra tropics to the equatorial region.