 Here you have the wave activity fluxes computer from the regression of the stream functional anomalies and for the DJF season let's see first day zero which is when we have a positive phase of the pattern. So here is the southern tip of South America you see this cyclonic circulation here which would be enhancing convection in the South American region and then we get this positive anomaly to the north and if you see the evolution you can see mostly a subpoly waving here building up to this cyclonic anomaly in South America that will then be causing this positive phase of the pattern. There are also several other features which are more present in another month like for October, November you can see this intense anti cyclonic anomaly here in the west of the dotted peninsula that in previous studies has been shown to be a precondition of the precipitation here in South American so we also see that type of anomaly with the regression patterns and if you see the animation here you can see that these are quasi stationary patterns that oscillate from this cyclonic anomaly in South America towards day zero and then favor precipitation in the South East and South America region and this is more likely to what you said and that you can see the wave change not like really propagating but just like this standing oscillation there in the evolution. So regarding the rainy season and the interest of our view in 30 to 90 days we were wondering if this activity that we saw in the tropical region might be related to the Madden-Julian oscillation and for that we use another study in which we use the RMM index to characterize the MGO and we found similar results with the OMI index and in this case for each seasons we related the RMM index to rainfall in South America following Wheeler for 2009 and we computed composites of the probability of weekly of seven day running mean weekly rainfall exceeding the upper third side and we express that as a ratio to 0.33 so a value of one would mean that there's a nominal probability of exceeding the upper third side and value of 1.5 would mean that that probability increases in the 50% and so on. So this was what we called for DJF season so these colors would mean that the probability of exceeding this upper third side in the weekly average rainfall it would be higher than normal and then this below normal and we found after buying statistical significance using Monte Carlo techniques that there was a dipolar signal related to the different phases of the MGO mostly around phases 3 to 5 and what was shown in I think Tuesday this lecture that you showed how to use the chip's data set and the R&D library to make composites of the RMM phases I performed is like the same for this region here and you still get to see this dipolar signal between the SAX region and the South American region and the opposite phase maybe on R&D phases I and V. So there seems to be a there could be a relationship there so if you see the location of the convective centers between phases 3 to 5 you could see it around the Indian Ocean and moving towards the maritime continent and this would be related to a positive phase of the CIS pattern and in that positive phase on day zero what we saw is that the convection numbers were located actually in the same region as these phases so and this was the wave trend that was present in the same day as this so this wave trend doesn't necessarily mean that was generated by by this this location of the of the negative anomaly but it takes the time to build up and then to impact in South America and to test that from another point of view we used an MGO index computed following Jones of 2012 in this case they used the combined uf of 200 the Pascal and 800 the Pascal a son of wind between these anomalies and they were filtered in the 22 standard days and they used this type of index with to analyze the intracisional viability of the modern Julian oscillation and they say that the advantages respect to the Arameum index is that without as not taking convection as a part of the index and mostly analyze the circulation and that makes it less noisy and detect better the isolated MGO events and so you can analyze the PC one and PC two of this type of combined uf as you did for the Arameum one and two and using this type of index we defined MGO coherent events also following Jones in which the amplitude should be greater than 0.9 during the whole event it should be also a and is for propagation along the whole event in the Arameum diagram that would be a counterclockwise rotation and the eventual last at least 25 days that should define the MGO coherent debate and on the other hand we defined using the 1390 system index which was the principal component associated to the uf are positive and negative events that those are which last at least five consecutive days above or below one standard deviation. So an example for one season is that here we have the 1390 system index in black the MGO amplitude in the dashed line and shaded in yellow and then you see the MGO phase computed with this index here it's from 1 to 8 and then you see that these steps as the MGO changes of phase and propagates towards the east but it was a really nice example of the evolution of both indexes and what we tried to do was to see the simultaneous relation between the C-s index and the phases in which the MGO phases occur following what our previous studies and what we found are some diagrams like this so this is the would be like MGO phase diagram the yellow diamond would be the time when the positive C-s phase was maximum and then you get to see the evolution of the different days along this phase diagram what you could see is that in the positive C-s events they mostly occur during MGO phases three to six while the negative C-s events are produced between phases seven to two so this is coherent with what we were expecting from this type of work when we analyzed the local signal precipitation associated to the MGO anomalies and also with this analysis respect to the location of the convective centers or in this case they are this is the CPC figure of composites for the velocity potential anomalies following the tropics so then you get in phases four and five this the location of the maximum diversion is here on the barricade continent which is mostly what we found here you've seen the signal of the pc1 in this region and for the negative events you could see that this case was 24 days before the peak of the positive phase so that would be a period of around 50 days in the change of science and you could see that the convection has positive anomalies here in the barricade continent which would be related to these maximum conversions here so from another perspective another thing we did was we took all the values of the C-s index that occurred along the the season and then we categorized them according to the MGO phase that was observed in that day this all during the MGO coherent events and what we found in that way that is that in this case we are not considering only the C-s events but we're seeing all the values of the C-s index that for phases between three to six you have a higher probability of having a positive index in these phases whereas when you have an MGO phase from seven to two there is a higher probability of getting a negative C-s index in those phases and analyzing individual cases we could we could associate this mostly the most part of these negative values to an evolution of the C-s index that came from negative values maybe on phase three and then it had it peak on phases five and six as well as these negative values in these phases were associated to this the evolution of the indices that peak in the phases three and four remember that there's around the in the mean that you can spend the MGO spends around five to six days in each phase so the MGO propagates to the east and the phase evolves to a higher values then you get maybe six days differences between different phases in the mean so every MGO event is different but then you can see like a temporal evolution here so the highlights for the rainy season is that the leading pattern of interstitial variability is still a dipole explained around 21.5 percent of the variance and associated to the variability in South America we could see that there was really progression of MGO like tropical convection anomalies to the east and there was some seasonal differences even within the rainy season and we could also observe extra tropical wave trains seem to link that convective anomalies to the extra tropical activity in South America and through studying the MGO impact in South America we could associate these similarities to the MGO like progression anomalies to actually the MGO circulation and the positive six events in that sense from mostly between phases three to six so to address what I left here is that we need to analyze better these MGO signals to study which are the precursors of the CIS events and not only the simultaneous relation between the MGO phase and the CIS events and I also wanted to repeat this characterization of MGO events using the OMI index which is an index that uses only the OMI so you can see only the convective and the convective signal of the MGO without this influence of the circulation associated and also this is filtered so I've never seen another phase diagram of the OMI index I don't know how smooth it is as it's evolved but I hope it's less noisy than the RMM index so it's less tricky to define these intense MGO events so for the dry season we also get the UF leading pattern it's only one center has only one center of action in the same region of the system tonight the day by the viability and as I pointed out before we don't see a such an intense signal in the tropical convection of using for the rainy season but you can still see some anomalies there propagating around the tropics in the Indian Ocean during the phase the positive phase of the pattern where this pattern is reaching its maximum values you can see that there are convective inhibition here in the Indian Ocean and regarding the wave trains maybe we can see the animation here you can see the influence of not only the exotropics building up to these anomalies in South America but also of the tropical region here that you see the different signs of the wave trains also resembling the BSA patterns that they pop up not only the traditional timescale in the annual timescale and this also seems to be more like quasi stationary wave planes and not a propagating signal you can also see this splitting of the of the wave trains during the social winter season so what's left here is to analyze which is the exactly the source of these wave trains and you could see this formation of this positive center in the Indian Ocean that moves slowly towards the east as the pattern of American boats so just for competition I'm showing this figure though we are not such as conclusive as during the rainy season but for the positive six events we see that between phases six and one the you can get most of them but though they are less events of the MGO and also they seem to be shorter in this season at least as they are computed from this Jones index and then from the negative six events you see that mostly between phases two to five but you can still see some negative events in in those phases so that's not as conclusive as for the summer season for the positive six events really quick we see that the most most of them are going to be in phases seven and eight which will be these ones in the velocity potential regression and this couple of conversions might be related to these centers here in the tropical Indian Ocean so the highlights for this dry season is that the pattern is monopolized we've seen in the 2090 days variability they are not as clearly related to tropical convection as they were during the rainy season but you can still see a tropical wave transfer along which energy is propagated and also along this tropical latitude resembling the PSA pattern as I mentioned before the relation is not clear between the season MGO as it was during the rainy season and we also need to address these precursors of the six events here though I don't know how given that the relation is not as clear I don't know what we could expect if they are coming from the MGO signal or not and also using the OMI index so for the short intracisional variability this was most similar to what Liebman in 1999 showed in the evolution of the SACs or van der Veel that they they were focused most without leaving out the tracheons in the two to ten that two to ten day period by in the scale while analyzing the short intracisional variability and the same way we computed the SIS pattern as the leading uf in this region and we see that there are some similar differences that we pointed out before but still we could we perform this regression evolution using the separate PC ones associated to these patterns and we saw a similar features that we could get when we use only one pattern across the whole season and then dividing the PC one and performing regressions from this pattern so not only for the sake of simplicity but also for making easier the task of monitoring tools that we are trying to develop we chose to stick to only one pattern and analyze the seasonal dependence using that one I think I have the animation so it's this evolution is for from 15 days before the peak of the positive phase in South America and as you can see there are some really interesting seasonal differences which is where somewhat expected if you see we've been talking about winter in this short intracisional time scale you still get these two patterns it and as you can see this seem more clearly to be propagating into South America and not just like a standing oscillation as the 3290 patterns and also you can see this splitting of the weight train evolution along the subpolar weight the subpolar weight train and the subtropical weight trains and in the Ovala evolution you can also trace back these centers towards the subtropical region associated to the South Pacific conversion so I don't know what would it would mean in a sense of a source but it was a difference related respect to the other seasons there it was also there in the spring season but not quite there in summer and March had been made so in this case it takes around 15 to 17 days for the pattern to evolve from a negative phase to another negative phase or from a positive phase to another positive phase and some of the here is that we got for explaining this evolution of variability in this case obviously we weren't expecting tropical convection numbers like MGO to be related to the evolution of this pattern but it was suggested to us that the SPCC could have an influence in what sense in a nonlinear sense as there were studies from Braub in 2008 and 10 in which they showed that there could be a nonlinear processes of resonance of equatorial waves and they analyzed the resonance between three different types of waves and when they use the convective couple wave there then you can get the resonance in the intracisional variability time scale that could evolve towards South America the signal and so they associated that process with tropical convection so that's one of the of the theories we have and then the leading pattern of variability as I've shown before shows this is a polar wave transacting in the pacific ocean and also it's a topical one mostly in spring and offshore winter and those end up enhancing a convection the South American region and they seem more like propagation of the patterns of the PSA patterns and not like a standing oscillation and then what we need to address in this variability band that it was mostly less static than the viability associated to the malangial end oscillation is how do nonlinear processes influence in the region which is the region of the four things also I don't know if analyzing separate case studies would allow us to avoid this moving and then detecting the sources of this variability more more I don't know how to say sorry and just to close up this is we're currently as I told you before trying to use this CISP pattern to monitor the evolution of the intracisional variability of rainfall in South America so we are currently in this this web page we have several tools of a of weekly monitoring and also a subsystem of forecast using the climate forecast system version two model for precipitation temperature upper level geopotential heights on OLR and we also modify the algorithm to compute the CIS index in a quasi real time way I don't know if you are familiar for working in real time with intracisional indexes that the that involved filtering is not quite trivial because you can't filter in real time using the band pass filter so it's tricky to adapt the methodology to to compute the real time index when you want to use an intracisional index to monitor variability so then in this page you can select each of the station the weather station located in this region to see the house evolution of precipitation there and you also get the quasi real time CIS index this was for the winters also winter that passed and this is well this is computed in real time you see that here in these positive values that occurred between the end of September and October you could see that there was an enhancement of consecutive days of precipitation in this case I think that the weather station is one around here that I selected for this for this so this is going to where we are going to trying to develop these monitoring and prediction tools but we also don't want to leave all the dynamics behind so we need to to currently address most of the influence that was suggested here so if you have another suggestion for different diagnostics that we can address or regress then there will be I will be taking those back home and then performing them to to continue study this evolution so thank you