 Okay, so my talk will be on the Equatorial Atlantic variability. So all the talk we've been having since morning on the Atlantic has been on the extra tropics. So what I try to show here is that there is some strong decayed variability in the tropical Atlantic. So I have to acknowledge some people that have contributed to this work and all of them except one is here. So you can also discuss this work further or ask questions to them. Okay, so I start by giving some background on what is called tropical Atlantic variability. There are two major modes. One occurs during spring and that is the so-called Atlantic meridional mode that is like a dipostructure. One just not adequate and then the other one south in the co-tongue but then a bit further south and this occurs in spring. And the peak periodicity is around eight years or so. There's been a lot of work on this and the relation between this on Atlantic tropical cyclone and all that. Then in summer there's a second mode. This so-called Atlantic Nino is like a smaller broader to the big in the Pacific Ocean. So yeah, I just said it in terms of time scales. The meridional mode is basically a sub-decader not really a decader of variability but then the Atlantic Nino is basically in Teranoa. So from different calculations the periodicity is between roughly two years to a little more than four years. So in terms of the mechanisms driving this, so I'm coming down to Atlantic Nino that is the equatorial mode which is what I would discuss further. So in terms of the mechanisms there are the standard theory is that this is controlled by the by the mechanism just like you have in the Pacific Ocean where you have the interaction between wind, SST and then thermocline. Then but recently there's been several other mechanisms proposed like Ingo Recta proposed what we call the New Atlantic Nino where you have a mixed layer temperature advection from Northern Pathotropic Atlantic to Earth's equatorial region. Also recently there is a second interest in waves, equatorial waves that can drive SST variability and all that. So all these are based on ocean dynamics. So the question we have tried to pose recently which is a motivation for the present world indicator variability is how important is when you combine all these different theories about ocean dynamics if you combine them together what is their total contribution to this Atlantic Nino variability. So if we go back to the basics, the SST variability is driven by a combination of ocean dynamics and heat flux is within the mixed layer. So given by this equation so we can the first equation it's just a simple expression of the heat budget in a fully coupled ocean model but then if we have we can remove the first part in blue there so that we are left with the second part in red which is the thermodynamic feedbacks that is heat flux and we can do the we can do the heat budget within this small box in the lower plug that is the Atlantic Nino box. So if we do this what we find using two different fully coupled models is that you look at the upper panels and you look at the upper panels you see that the you see that the contribution from the advection terms that is the dynamical terms in terms of the quantity are so much smaller compared to contributions from the heat fluxes that is the lower panels. Then in terms of the timing for example if you look at the GFDR model you find out that what actually drives the SST variability is heat flux at T minus two before the peak phase of the Atlantic Nino event while the dynamical part actually become important when the Atlantic Nino peak phase has already set. So and another way of looking at this is if we take the same exactly the same models the only difference is that one is coupled to just heat flux that is the so-called slab model and then the other one is coupled to the full ocean dynamics and we just take a measure of variability of the Atlantic Nino index as standard deviation and we plot and compare two of them. So what you see on the left panel is for the Atlantic Ocean so you see the pinkish color is for the just heat flux alone slab model then the blue is when you come to the full ocean dynamics so what you'll find is that in all the models so much of the variability come from slab model alone then if you compare with Pacific it is completely something different and yet I mean the understanding the theories we have about Atlantic is basically based on analogies we take from what we already know in the Pacific Ocean. So one thing to suspect once you see this you think maybe in the models the mist layer depth is not well represented so what we find actually is that the mist layer depth in these models is actually consistent with observation and in general at least in the Atlantic Nino region is basically there is east to the right of the red line there is basically less than 15 meters actually so yeah this is for the rest of the models the individual models I mean. So what you understand from this analysis is that if you have idea of the variability of Atlantic Nino from the slab model without ocean dynamics you can easily say what the variability would be if the model is fully coupled at least judging by the 12 similar models we studied. Again if you look at the planet to the right that is for the Pacific Ocean this is not so. So this background is the motivation for the present work which we are doing now which is basically still ongoing and the question is how important how important is atmospheric variability to equate with the Atlantic variability. So from what you already know and what I've said it is well known that this meridional mode is basically driven by the atmosphere the theory is the so-called worst feedback. So recently there's also a paper that just plotted without discussing so much that just I showed this spectrum of Atlantic Nino index and if you look at it you see something around 10 to 14 years or so you see something that looks like a peak. So the question is this peak is it real? Is it not real? And this peak is for summer. So what you've done here is if we take observation and calculate the spectrum of Atlantic Nino for the different months so for the different months on the x-axis then on the vertical axis what you have different periodicities. So you see that around around 10 years plus some few years you have strong variability starting from sometime around spring going into summer and if we try to look at the spatial pattern of this that is taking the UF of the box calculating the UF and then regressing it on global SST what we find is a well defined the meridional mode not south mode in spring that is the top panel but in summer the pattern changes and what we have is basically an equatorial mode that looks like Atlantic Nino. Sorry I have to say that this data is filtered so we are looking at 8 to 25 year time scale so every variability above 25 years is being cut off and everything less than 8 years is being cut off also. Sorry and the importance of this if we look at the correlations there if we take rainfall over the southern part of west Africa in summer which is the main rainy season for this place so we find very strong correlation both in real analysis and then if we repeat this with observation we still find very strong correlation but for the for the Sahel even though there is negative correlation for the box we consider the correlation is not so robust. Again if we take a box over northern part of South America up to from 10 south to 10 north we still get very strong correlation both in observation and in real analysis so at least this shows that this mode is important and could be useful to understand the decadent variability of rainfall in these places. Okay so from the literature actually this mode something like this mode has been described if you look at the pattern I showed in the previous slide what we have is like what Shia Anthony Motto called in 1998 Pan-Atlantic mode that extends well well north and then going back into the southern to south Atlantic ocean where you have bands of cooler thirsty, warmer thirsty and all that and this is the decadent mode which they described. However this is basically considered either a spring mode or a non-mode and not considered a summer mode. So if we look at the evolution of the SST anomaly now by taking the year for individual month starting from March to April to going down to the rest of the month so what we find is that in March and April the strong anomaly is basically in the north of Equator then in April this is basically the same starting from May the northern part of the anomaly begins to cool down begins to weaken while the southern part of it that is the Atlantic Nino region begins to warm and then by the time you get to June the northern part is already weakened and the southern part is very very strong now so this is basically the transition between this is sometime in May and June. So what we do now if you look at the spectrum now you'll find that the strongest variability is in May June and then again we've said that the transition is also in May June so what we've done here is to take the SST and then sea level pressure because of course in the tropical Atlantic you have this very strong coupling between the ocean and the atmosphere both in terms of the Atlantic Nino and in terms of the meridional mode also so take the two of them do a maximum covariance and analysis and then take the time series and regress it onto the global anomalies and what we find here is like a truly pan Atlantic mode where the SST anomalies both in the south in the tropical region and then in the north basically of comparative magnitude. Okay so just to conclude we have tried to describe or we are trying to describe a strong decadent variability in Atlantic ocean that occurs in during the Boreal summer and we have shown that this is related to ground fall over the nearby continent and we consider this mode as a part of the pan Atlantic as a part of the pan Atlantic mode and what happens is that the anomalies actually is related to both North Atlantic and also South Atlantic ocean which have not gone to show so much now so yeah thank you I stop here.