 We will continue our discussion on the Indian Ocean and the monsoon today and I said that it is important to understand how the variability of convection in the monsoon region depends on the variability of SST of the North Indian and Equatorial Indian Ocean. So I am going to first talk about an example which is the aberrant behaviour of the monsoon in June 2009. So what happened in June 2009 was very intriguing. After an early onset over Kerala on 23rd May, the advance of the monsoon over the Indian region was delayed by about two weeks with the monsoon restricted to the west coast and southern peninsula until 24th of June. So you can see what happened here see actually these are the dates of onset when the monsoon is supposed to come the mean dates by 15 June it should have come here and by 24th it should have come somewhere here instead of that by 24th June the northern limit of the monsoon was still this one and because of that see all these regions had very very high deficits in rainfall all these subdivisions because monsoon had not come there at all only in the south we got some rain and so what happened was that this resulted in a massive deficit in the All India rainfall of 54 percent of the long term average for this period. The All India rainfall for the month of June was rather close to the lowest recorded rainfall which is 50 percent in June 1926 since 1871. So it was close to a record in terms of deficit in June and of course this was not predicted by anybody. Now let us remind ourselves that the onset phase of the monsoon commences with the onset over Kerala in late May or early June and culminates with the establishment of a TCG over the Indian monsoon zone north of the peninsula in July we have already seen this the onset phase comprises one or more northward surges of the rain belt across the peninsula this also we have seen these are seen as northward propagations of cloud bands in satellite imagery stretching from the Arabian Sea across the Indian region to the Bay of Bengal. Now the variation of low LR band at 85 degrees east over Bay of Bengal we will look at it we have going to look at it for the May to July period for a season with typical advance of the monsoon India onset phase which is the season of 2008 along with this aberrant onset phase in 2009. So what you see here above is 2008 and here you see 2009 we got a reasonably good onset but after that you can see 85 degrees east is a longitude just across the Bay right it is slightly to the west of Kolkata it is going across the Bay so this is the first northward surge but you see what happens in a typical year is after the northward surge then the band hangs here in the monsoon zone that never happened after this the band kept on appearing over the Bay but for very very short periods two days three days and so on whereas you see typical year this is what would happen but you also see that there was some convective activity over the equatorial Indian ocean you see equator to 10 south you kept getting a seesaw between convection here and here you can see this band had become prominent then this got born it died this band was here then this got born it died and so it continued till about 24th of June. So this is the story which is very different from a typical onset phase of the monsoon so prominent feature of the variation of 2008 is a series of northward propagation from south of the equator to north of 25 north beginning with one in the first half of June after this propagation which was associated with the advance of the monsoon the convection persisted over the region 15 to 27 north for over a month except for a short gap of three days we have seen that once this advanced it persisted over this zone for several days except for a gap of very very few days. However in 2009 after the propagation in late May the convection disappeared after two three days we have seen this once it propagated here after two three days the convection just disappeared during 1 to 24 June the convection over the bay appeared intermittently over the region between 10 and 20 north we have seen this also see between 10 and 20 north is this region so the convection went on appearing and disappearing it appeared intermittently in this region and it was not sustained for more than three four days and no propagation occurred over the Indian longitudes of 1780 also the story was the same there were short spells of convection over the region between 10 and 20 thus there was a delay in the advance of the monsoon and until 24 June the monsoon rains were restricted to the region south of 20 north so clearly the question to be addressed is why was convection over the bay not sustained during 1st to 24 June 2009 now we know that in 1st to 24 June 2009 convection occurred frequently over the eastern equatorial Indian Ocean east of 80 degrees east in the 0 to 10 degree south belt whereas convection over this region occurred for very few days in 2008 this is something we have noted before that here between equator and 10 south hardly any days of convection but for 2009 you see large number of days with convection here so in fact we will I will show now that this aberrant behavior in June 2009 can be attributed to the SST of the bay vis-a-vis that of the eastern equatorial Indian Ocean so this is an example to show how variability of SST can have a direct impact on the monsoon now we know we have discussed this often that the CTCG is actually maintained by propagations from the equatorial ITCG or equatorial tropical conversion zone on to the land we have also noted that the relationship between CTCG or the tropical conversion zone in the latitude of the monsoon zone and the oceanic TCCG is very complex on the one hand the TCCG over the ocean helps in maintenance of CTCG by moving northward but on the other hand it also competes with it and this is something we have seen now we have also seen in our discussion of inter-annual variation as well as intracesional variation that is to say active and breaks that convection over the eastern equatorial Indian Ocean is unfavorable for the monsoon so that is the region over which if a TCCG appears it competes with the monsoonal TCCG while that over the western part is favorable for the monsoon and you can see that here this is the correlation between OLR and ISMR all India monsoon rainfall and you can see that there is a inverse correlation between the rainfall over India and here whereas rainfall over the monsoon region depends on the rain over the bay and the rain over the bay is negatively correlated to the rain over the eastern equatorial and central equatorial Indian Ocean whereas it is positively correlated with the western part this is for the seasonal scale on the intracesional scale also we have seen that if we look at the break composite the biggest signal is that when there is a break there is a flare up of convection here over eastern and central equatorial Indian Ocean and in an active phase convection over eastern and central equatorial Indian Ocean is suppressed so the competition between convection over the bay and the monsoon zone is with the eastern part eastern equatorial Indian Ocean and not with the western part in fact western part is favorable so we know this and we saw it also actually in the pictures that we saw here here also we could see that in 2009 as I mentioned this died and then this was born this flared up again after this died but again this flared up and this died and so on and so forth so you see a seesaw in activity here this also suggests that there is a competition between the convection over the bay between 10 and 20 north and the convection over equator to 10 south over the same longitude. So, the question to be addressed is now so we will now see why is it that the equatorial TCG was hyperactive in June 2009 at the cost of the TCG over the Bay of Bengal and we think we believe the answer is in the sea surface temperature distribution and this is the sea surface temperature SST and you see this is the eastern equatorial Indian Ocean this is the western equatorial Indian Ocean and this is the Bay of Bengal which we are concerned with this is the region over which disturbances form and then move on to the Indian region. Now you can see that the eastern equatorial Indian Ocean is definitely warmer than the Bay of Bengal and in fact this is a manifestation of the fact that you had a lot of cooling over the bay that year in a cyclonic storm was there in May which led to a lot of cooling here so that there is negative SST anomalies over the bay and positive SST anomalies over EIO. So, you have a situation in which the eastern equatorial Indian Ocean is warmer than the Bay in the month in which the onset phase occurs which is the June month. Now let us look at we have two data sets Reynolds here and TMI here TMI is the more modern one and it is always a good idea to see whatever result you get from one also holds for the other data set it should not be dependent on the data set itself. Now what you see here these hairs they correspond to every year for which data are available and they are what are they I have already shown you the regions which we are interested in this is EIO and this is the Bay of Bengal and we believe at least the pictures of low ELA region have shown us that there seems to be a competition between these two regions. So, we look at the SSTs of those two regions so this is the SST of EIO, this is the SST of BB Bay of Bengal and this is the difference in the SST of the two EIO minus BB. Now what we have done is we have plotted the mean which is climatology we have plotted 2008 which is blue and 2009 which is red. So, let us see what happens see climatologically you see that right from April onwards this is plotted from April onwards the Bay of Bengal SST builds up and it slowly starts decreasing with the onset of the monsoon this is climatology. Now what happened in 2008 the Bay was a little bit cooler in the early part of May but then made up and then was slightly cooler than the climatology. But what happened for 2009 is that you saw that after mid May sudden cooling occurred and till about 24th of June or so it was definitely cooler than climatology. So, Bay is definitely cooler than climatology in this year on the other hand eastern equatorial Indian ocean has been warmer than climatology throughout June, July and even part of August. So, whereas 2008 it was colder so we have a case in which EIO is warmer than usual throughout and Bay of Bengal is colder particularly in the critical time when the advance has to occur in the onset phase. Now what what does the difference show as we expect in fact the EIO is warmer than the Bay of Bengal throughout 2009 whereas it is colder than the Bay of Bengal throughout 2008 and climatology says that by and large up to about third week of June EIO tends to be colder than the Bay afterwards they are very comparable. So, during the advance phase the mean SSD gradient is such that the Bay is warmer than the eastern equatorial Indian ocean. But what we see here is in fact opposite occurred in 2009 whereas in 2008 when we got very nice propagations and a very nice advance actually the eastern equatorial Indian ocean was colder than SSD. Notice that what I have talked about is so far for Reynolds also holds very much for the TMI again the same story and you see that eastern equatorial Indian ocean is warmer than the Bay in the critical period throughout that is to say from beginning of June throughout and this is what we believe has done the damage. So, at the beginning of June the SSD of the Bay was colder than almost all the years while the SSD of EIO was warmer than that of almost all the years. Consequently EIO in June 2009 was warmer than the Bay by almost one degree you see this if you look at June then this is almost one degree EIO was colder than Bay of Bengal by almost one degree in June. Note that while the SSD of EIO during June 2009 is the warmest SSD for June July 2008 is the coldest this also we have seen that while the SSD is the warmest here for EIO it is the coldest of all the years. So, it was an exceptional year in terms of equatorial warming. The difference in the SSD is positive and highest for 2009 and negative and almost the largest magnitude for 2008. Now, so what has happened is we have two tropical convergence zones competing with one another one is in the Bay between 10 and 20 north and the other is over the equatorial Indian Ocean between 0 and 10 south over the eastern part of the equatorial Indian Ocean. While normally the Bay is warmer and so the Bay TCC has an edge over EIO TCC. What has happened in 2009 is that EIO is warmer than the Bay in that critical time when advance of the monsoon should take place over the Bay and it is this high SSD of EIO related to that over the Bay which gave an edge to the TCC over the equatorial Indian Ocean in the competition with the TCC north of 10 degrees over the Bay in June 2009. So, this is a case in which we see a case of impact of SSD variability gradients of SSD reversing from east equatorial Indian Ocean to the Bay which have had a major impact on the TCC or cloud systems over the Bay and because of this the TCC or cloud systems over the Bay were therefore not sustained for longer than 3-4 days. No low pressure systems had been generated over the head Bay and westward propagation across the monsoon zone which is a characteristic of the onset phase of the monsoon did not occur. So, this is an example an interesting one in which the SSD field in June had a very large impact on the advance of the monsoon and hence on the monsoon itself because actually the deficit that was so huge in June was never made up till the end of the season because the rest of the season also turned out not to be very good for other reasons. So, this is just an example to say that all the effort we put in in trying to understand how SSD evolve and eventually we would also like to model how SSD evolves predict how what will be the SSD in the forthcoming season and so on. Why is it important because it can have a direct impact on the competition between the different TCCs and thereby on the monsoon itself. Now, I will not dwell too much more on the Bay of Bengal Arabian Sea and their relationship to the monsoon, but what I will now do is to go over to the equatorial Indian Ocean atmosphere system. See so far we have been talking of what is the North Indian Ocean comprising Arabian Sea and Bay of Bengal and we have looked at the mean patterns how they evolve and also tried to I tried to show you how the kind of impact these temperatures can have on the monsoon. But now let us go to the equatorial Indian Ocean system. So, first of all let us consider the variation of SSD and OLR over the equatorial region. Now, notice that this is now January and February notice that the equatorial Indian Ocean is warmer in the east than in the west and west is in fact rather cold it is below the threshold. So, along the equator if we look at then west is warmer I am sorry east is warmer than the west. Now, this is opposite of what we saw in the Pacific and we will keep seeing that the picture is somewhat similar to the Pacific except we have to invert east and west. Now, February again same thing east is much warmer than the west along the equator and this is the low OLR region here and you see that along the equator January and February the low OLR region is largely restricted to the south southern hemisphere and in fact also restricted to the eastern part of the basin and much more restricted in February. What you see here in red are contours of 27.5 and 28 the possible thresholds in this region and what you can see is that a large part of the ocean which with SSD which is warm enough to have convection in fact does not have convection this is true of January as well as February you see this part could have had convection, but does not have. Now, we come to March and April and these are lines that show 10 south to 10 north and in March and April everything has warmed up you see these are very very high temperatures here and so the northern hemisphere because the sun has moved now over the equator and in April it is slightly in the northern hemisphere vertically above and you see this is a large body of very warm water in April and this is the warm water in March. Now, what happens to the low OLR region? Low OLR region in this case again occupies a very small part of the warm ocean it occupies only the eastern part small region here, but notice that now it has also come to the northern hemisphere earlier in January February it was restricted to the southern hemisphere. Now, it is stretching across the equator and is a but is very much restricted to the eastern part. So, now we come to May and June and the SSDs are getting hotter and hotter and this is June by June of course, the onset has occurred and SSD particularly of the Arabian Sea has cooled a lot and Bay also has cooled somewhat, but in May the SSDs here are very high and the maximum SSD is of course, in the northern hemisphere now because this is the northern hemispheric summer. Now, let us see what has happened here is again we see that the low OLR region is restricted to the eastern part, but has spread somewhat to now come also to the southern tip of eastern Arabian Sea here. So, we have a low OLR region stretching all the way from 70 to 90, but the maximum intensity of convection is still around the east and this is the case for June again like I said for the other months only a small part well I would not say a small part only a fraction of the ocean with the SSD above the threshold has convection on it the rest of it does not have convection on it, but notice that the entire ocean is rather warm in May. Now, in June this upwelling begins to develop here. So, this begins to get cold and now you see a fairly large part of the warm ocean now is covered with low OLR covered with clouds. Now, you also see that in June the convection here is as strong as the convection in the bay. Now, we come to July and August and you see this typical pattern where you see ocean here is very warm equatorial Indian Ocean is warm and bay remains warm, but the Arabian Sea is cooling the warm pool region of the eastern Arabian Sea is shrinking from in July at least you see it here, but in August you do not see it anywhere near the northern part of this coast it is only restricted to the southern part of peninsula. So, SSD is cooling very fast what is happening to OLR both for July and August you see almost it looks like the dynamics is very favourable and that is why almost the entire warm pool is covered with deep convection as evins from the OLR maps that in July and August also almost both of these months the warm pool is covered with low OLR. Now, this is the climatology of SSD along the equator for September and October. Now, you see this has almost with the withdrawing of the monsoon and withdrawing of the clouds things slowly begin to warm up again and you can see by October quite a large pool of warm water has built up over the Arabian Sea here. So, and this is the low OLR region now you see particularly in October a large part of the Arabian Sea has no convection although it is above the threshold. So, although a lot of cooling of the Arabian Sea took place during July, August and we are taught in fact that northeast monsoon does not rain too much on this side because no systems form in the Arabian Sea because it has cooled but actually it has really not gone below the threshold. So, we have to attribute the lack of rain over this part during what is called the northeast monsoon or what we call the post monsoon to dynamics somehow again the low OLR region is getting restricted to the eastern part of the basin and this is November, December now a lot of warming has taken place but of course, cooling of the northern parts because the sun has gone to the south and again the low OLR region is restricted to the eastern part. So, what we have seen is that there is a warm pool region there is a region over the equatorial Indian Ocean and the West Pacific in fact, which remains about 28 throughout the year and this region is about 29. So, if you look at annual SST this is the entire warm pool of the Indo-Pacific region and this is the SST variation of the equatorial region remember we saw a similar picture for the Arabian Sea and Bay of Bengal now we are looking at the equatorial region. This is the region of Somalia where a lot of upwelling takes place. So, what you see is initially the SST builds up till April-May then it decreases very rapidly to become almost minimum in August and then builds up again. So, this is the SST in the western most part of the equatorial region whereas, in the central equatorial region you can see it is very, very dull SST is above the threshold and starts decreasing only after September and also in the eastern part just of Sumatra now again it is above 28 throughout the year, but is somewhat colder in October, November, December. So, this is the variation over the equatorial region. Now, we have been talking about the thermocline how deep is the thermocline and we have seen that for El Nino in the Pacific you know the thermocline was very shallow in the East Pacific and very deep in the West Pacific. Now, here opposite situation holds in January you see yellow colour means deeper thermocline what they have plotted is depth of the 20 degree centigrade isotherm. So, the deeper it is the larger the depth it means deep thermocline here. So, if you look at the equator then the thermocline is shallower over the West then it is over the East in January. Now, let us see what happens there are also winds there which we will come to these are the winds in June you can see this is the southwest monsoon this is the strong southwest flow that occurs and here also still the eastern part of the equatorial Indian ocean is deeper than the western part. And same story continues in August, but notice that in the Arabian sea you get a very very deep thermocline here and again in November this has become even deeper than before and this has become shallow. So, we have this East-West variation in the thermocline depth the thermocline being deeper in the East than in the West this is almost a mirror image of what happens in the Pacific which we had looked at. Now, let us look at a vertical section of temperature along the equator from the East coast of Africa that is from here to Central America right is right across here. So, first part is the Indian ocean and then is the Pacific ocean. And what we are going to do is take a vertical section all along the equator and see how temperature varies. And what you see here is the picture for April and this is the picture for October. Remember this is part of the Pacific here and this is the classic picture of the Pacific where the Pacific has a very very deep thermocline here and East Pacific it shows. In April we find that there is not too much East-West variation in the thermocline you know it is more or less flat, but you see that by October you see definitely that the slope of the thermocline is in opposite direction to that in the Pacific that it becomes deeper in the East. So, there is a lot of East-West variation in the equatorial Indian ocean which is of opposite sign to that in the Pacific that while the thermocline of the equatorial Pacific is shallow in the East and deep in the West that of the equatorial Indian ocean is shallow in the West and deep in the East also for almost the whole year we have seen that the East is warmer than the West equatorial Indian ocean. Now, the question is why is the thermocline deeper in the East? Again we had seen that in the Pacific also the answer lay in the nature of the winds it was the trade winds there. Now, here also we have the answer lies in the nature of the winds. Now, the equatorial Indian ocean experiences a somewhat different seasonal wind forcing and hence has a different response than the Arabian sea and Bay of Bengal. Since the Coriolis parameter vanishes at the equator East-West pressure gradients lead to zonal wind this is something we have to keep in the back of our mind that we are looking at the equator. So, it is like flow down a pipe the wind will go from high pressure to low pressure. Now, these are the wind stresses derived from encepre analysis and what you will see by enlarges that the winds are very very small and they are rather large south of 10 south and north of 10 north and they are somewhat large near the coast here in the equatorial region, but otherwise they tend to be very small. So, now what does the pressure gradient depend on? See pressure gradient will depend on the sea surface temperature gradient as well as the gradient in the atmospheric convection which is associated with metropospheric heating. So, if you have deep clouds over a place then because they have they are associated with heating at middle levels metropospheric heating the pressure they tend to decrease the surface pressure under that column of air. So, either because the sea surface temperature is high or and or because of metropospheric heating you get changes in surface pressure. Now, we have seen that the eastern equatorial Indian ocean is warmer than the western part and there is more convection over the eastern part than the western part. So, both these factors are acting to make the pressure over the east lower than the pressure over the west. So, the surface pressure is lower over the eastern part and that would mean that you have winds flowing winds which blow towards the east or westerly winds. So, winds over the equatorial Indian ocean particularly their zonal component are weak during the monsoon because during the monsoon most of the convection is actually over other latitudes. But relatively strong westerly winds appear in April may spring and then again in during the transition between the monsoon October November. See in these months what happens? So, this is the month of April and this is the month of October. Now, in these months if you look at this part of the basin then the east west SSD gradients are not that great in either of the months, but what is high is the east west gradient in convection atmospheric convection or in metropospheric heating. So, this will lead to actually strong winds which are eastward along the equator and they can attain speeds as much exceeding 1 meter per second and these are known as whirl key jets. So, now let us look at the zonal current. Now, we have gone to the ocean from the winds and the zonal current is in fact very high. This is April this is the mean current and this is April May when it is high and this is again October November when it is high and it is high over the central part of the basin. See this is 60 degrees east and this is 80 degrees east. So, this is very very high and in between it remains eastward or westerly, but the magnitude is not so high. These are known as whirl key jets after an oceanographer called Witki who discovered and described them in great detail. Now, this is in fact the annual component and the semiannual component which shows semiannual component is very large. So, it shows that twice a year you get very very large magnitudes of westerly of easterly flow I am sorry of westerly flow. Now, so what do these Witki jets do? Again the story is somewhat similar to the Pacific, but you have to replace west by east. In the Pacific remember the warm water near the surface of east specific gets transported by the trades towards the west. Now, here Witki jets transport warmer upper layer water towards the east which accumulates near the boundary causing the thermocline in the east to be deeper than in the west. Just like in the Pacific the thermocline is deeper in the west where the surface warm water is pushed by the winds towards the west specific. Here it is carried by these Witki jets to the east and so that makes the thermocline deeper. The thermocline slope becomes greater during spring and fall and the volume of warm water and the heat content of the Indian ocean is larger on the eastern side than the west. When the thermocline is shallow winds have to do lesser work to bring cooler water into the mixed layer than when it is deep. That is for the same wind strength a shallow thermocline can facilitate cooling of the mixed layer and SST whereas a deeper thermocline may not. Thus Witki jets dictate the shape of the thermocline in the equatorial Indian ocean. Climatologically since the eastern equatorial Indian ocean is warmer it supports a more convective atmosphere than in the west. Now this is the mean picture we have been talking about looking at mean monthly patterns but in some years this picture changes dramatically of you know most east being warmer than the west in terms of SST east having a deeper thermocline than the west and in the atmosphere above east having more convection in the west. This is the mean picture but this changes drastically. So how does it change we get negative SST anomalies over the eastern equatorial Indian ocean and positive SST anomalies over the western equatorial Indian ocean. So where it is warm it becomes colder eastern Indian ocean and the western equatorial Indian ocean becomes warmer and the anomalies can be so large that the gradient itself can change SST gradient can change sign. Now this is the case of 94 and we generally look at June to September because that is the summer monsoon period and that is what we are concerned with most. So this is the SST anomaly and this is the climatology I am sorry this is the SST climatology and this is the actual SST in 94 okay and what you see is climatology says that the east is warmer than the west but actually now in 94 west was warmer than the east and this of course is because you had huge negative SST anomalies here and positive SST anomalies here. So the gradient now has become towards the west as far as SST is concerned and now let us look at OLR what happens. Now convection is suppressed over EIO that is to say positive OLR anomalies and enhanced over WEIO. So now this is the west equatorial Indian ocean box and this is the east equatorial Indian ocean box. I will define the longitude and latitude later and what you see is this is the actual OLR. So instead of convection being by and large restricted to east and central part of this region it is now restricted to the western part. So the convection has also switched and of course you get very large positive OLR anomalies here showing that convection is suppressed over EIO and enhanced over WEIO. Now what does that mean we have we know that along the equator things just go along from high pressure to low pressure convection is suppressed here enhanced here and actually convection is occurring here and SST is also higher here which means pressure is lower here and you will get anomalies which are easter events. So you get easter events here and if the gradients are strong enough it is just not easterly anomalies but the winds themselves change direction and instead of flowing from west to east they start flowing from east to west. So this is the situation that occurs in some years. So this is opposite of what we have seen in the climatology or the main picture. Saji et al recognized anomalous conditions such as 94 and 97 which I will show you this is the case of 97 and here again you see cold anomaly over EIO, warm anomaly over WEIO, convection very much suppressed over EIO this is the OLR anomaly and enhanced over WEIO. So these are two major events 94 and 97 and Saji et al said that these are in fact manifestations of a mode of the tropical Indian Ocean which they called the Indian Ocean dipole mode. Why did they call it a dipole mode because typically we see that this mode is associated with anomalies of the opposite sign SST anomalies of the opposite sign in the east and west. So this looks like a dipole and in terms of OLR anomalies also it looks like a dipole. So these people Saji et al suggested the name Indian Ocean dipole for this mode Indian Ocean dipole mode and characterized by low SST of Sumatra and high SST in the western equatorial Indian Ocean accompanied by wind and rainfall anomalies. Now highlighting 97 98 event Webster et al also have suggested that anomalous conditions present during this period was due to an internal mode of the coupled system of the Indian Ocean coupled Indian Ocean atmosphere system mode of this is called the Indian Ocean mode. Now if we have to see if this is a major mode of climate of SST variability over the Indian Ocean how much of the variability is explained by this kind of a pattern or this kind of a mode. Now if we look at SST variability of the SST field over the Indian Ocean and analyze it then the dominant mode which explains 30 percent of the variance corresponds to SST anomalies of the same sign over the entire region. So you have pictures like this now this is 97 where almost the entire region is warm and this is 99 where almost the entire region is cold. So there is one mode in which the entire region is either warm or cold or the variation of SST is coherent across the region. So examples are in fact this mode is associated with ENSO and you see 87 was an El Nino year and this you see is a signal of La Nina. So during El Nino you get warm SST anomalies and during La Nina cold SST anomalies but uniformly so. So this mode of the entire basin becoming warmer or colder accounts for about 30 percent of the variance. Now IOD mode is the second most important mode of SST variability and accounts for 12 percent of the variance. So overall it does not account for much of the variance but the special feature of this mode is that it has very intense events which we call positive IOD events which have a huge impact on the climate of the region. Now Saji et al actually suggested an index for the IOD mode which is called the dipole mode index and this is defined as SST anomaly of the west equatorial Indian Ocean which I showed you minus SST anomaly of EEO. So normalized by the standard deviation where this is where we define WEIO 50 to 70 is 10 out to 10 naught and EEO is 90 to 110 is and 0 to 10 south. So this is EEO 90 to 110 is 0 to 10 south and these boxes were actually defined by Saji et al and these are the boxes that have been continued to be used for looking at equatorial Indian Ocean dynamics and particularly for Indian Ocean dipole mode. Now what are the phases? The phase of the Indian Ocean dipole mode in which the SST anomaly is negative and OLR anomaly positive over EEO like the 94 case and SST anomaly is positive and OLR anomaly negative over WEIO is called the positive phase of IOD and an event in which amplitudes of these anomalies are large such as in 94 is called a positive dipole event. And we have seen that here this is a case where you have very large anomalies and convection is enhanced over the west and suppressed over the east and you also have west is warm relative to the east positive SST anomaly over west negative over the east. This is the positive phase defined as the positive phase of IOD and if the amplitude of the anomalies is large we call it a positive IOD event. Now the reverse negative IOD also occurs when would that occur that would occur if the SST of the east is higher than average that is to say you have a positive SST anomaly on the east and negative on the west and enhanced convection on the east and suppressed convection on the west. So, this is the opposite sign of anomalies to the positive one. But remember that climatology or the mean state is such that convection is generally restricted to the eastern ocean. What negative IOD phase means is that is further enhanced in the east and decreased in the west. So, negative IOD is an intensification of the climatological or the normal state whereas positive IOD represents conditions nearly opposite to the normal. Instead of convecting on the east the atmosphere is convecting on the west instead of being warmer in the east the ocean is warmer in the west. So, this is a major change in the gradients east west gradients pretty much the way El Nino also occurs. So, this is why P IOD events are special negative IOD is just an intensification of climatology this is something to be born in mind. Therefore, major focus of attention has been on positive events and what you see here is plot of DMI which I defined the dipole mode index. When it is positive these are the events of 94 and 97 was a huge positive index. And then afterwards recently we have had 2006 as a positive index and earlier 61 was also a positive IOD event. So, there have been several P IOD events in a called history and the composite anomaly patterns of several of these events show you how they develop. So, this is actually the pattern for composite in a mature phase of the IOD and this is where SST is given in terms of shades here and the contours correspond to depth of the 20 degrees isotherm. So, the contours are all negative here which shows that the depth of the thermocline is decreasing markedly over the east and is increasing over the west and west is warmer than the east convection is suppressed over the east and is enhanced over the west. Now, next we look at there is one more feature to look at that associated with this is very, very strong easterly anomaly of the wind which goes from east to west which we have mentioned. So, now we will see how these things are consistent because strengthening of if we go back here the winds imply that all along here you have strengthening of winds which are parallel to the Sumatra coast remember this is the southern hemisphere. So, if you have winds like this that will lead to Ekman drift in this direction and therefore, away from the coast that means they will be upwelling in addition to that these winds that come from east to west will also lead to upwelling because they are going to move water away from the coast. So, both these factors are going to lead to upwelling and that means cooling of the sea cooling of the eastern equatorial Indian Ocean and so they are consistent with the negative SST anomalies there. Now, I am going to talk about evolution of IOD also look at atmospheric component of IOD separately in the next lecture and then we will look at how the these modes or the coupled Indian Ocean dipole mode its atmospheric and oceanic components are they in any way linked to variability of the monsoon. Thank you.