 you connected? Yes. Hello, everybody. Hi, good afternoon. Here is good afternoon. Good morning. Good morning. So thank you very much for connecting here from in the morning there. So let me start with a few welcome remarks. In fact, Carolina was supposed to be here in person but she had to change her plans in the last minute. So I'm very pleased to welcome today, Professor Carolina Vera. She is a full professor, a distinguished climate scientist, a full professor at the School of Sciences of the University of Buenos Aires and principal researcher of Argentina National Council of Sciences, Konichet. She's currently the vice chair of the working group one bureau of the intergovernmental panel for climate change. She has served on a number of international panels like the scientific advisory panel of the World Meteorological Organization and many others. I will leave it to Filippo to give a more scientific introduction to her career. I will just mention one particularly important award that she has received. She was awarded by the American Meteorological Society with the Cleveland Abbey Award. And the quote says that for unselfish devotion to advancing and communicating climate science to decision makers and stakeholders in South America and across the world. And I'm very happy to inform you also that since this year Carolina is also a member of the Scientific Council of ICTP. So we are very glad to have her advice on matters, very important matters related to climate and the scientific priorities of ICTP. So this session will be moderated by our colleague, Professor Filippo Georgie, whom all of you know the head of the Red System Physics section. And she will be talking about the interplay between natural variability and anthropogenic climate change at regional scales. And as you know, it's a custom at ICTP that at the end of the colloquium, the students have the opportunity to interact with the some of our distinguished speakers in an interrupted and unsupervised way so that they can ask whatever questions they have about science and career. And usually everybody leaves the room now. Unfortunately, today, this is in hybrid mode, but we will conduct it in that manner at the end of this colloquium. So thank you very much. Welcome. And I give the floor to Filippo. Okay. Thanks, Ateesh. Hi, Carolina. What can I add? In terms of, well, first of all, I've known Carolina for many years now, because we have worked on IPCC related issues for a long time. And she actually, this is the ICTP on several occasions. And as Ateesh said, she's now really very happy, very fortunate that she's one of the she's in our scientific council IPICTP. I just wanted to add to her sort of international participation, international committees that she also was a member of the Future Earth Science Committee. This is Future Earth is a major initiative that so tries to understand where our earth is going in the next decades. And also wanted to mention that she was a member of the Joint Scientific Committee of the World Climate Research Program. This is probably the main program on climate issues. And this Joint Scientific Committee is sort of the scientific board of this major major committee show. So I mean, Carolina is probably one of the most renowned scientists from the Southern Hemisphere. She's an expert on climate variability, climate change, especially of the Southern Hemisphere of and within the Southern Hemisphere of South America. She's probably one of the main experts in South American climate issues. She started a career as many of us more on the physical side, working on climate variability issues, anthropogenic climate change issues and how they interact with each other, which is the main topic of today's presentation. But over the years, she's moved more towards the sort of more societal relevant questions and impact applications, climate services and so on. This is happening to a lot of us as we recognize the importance of climate research for society. Yeah, I just wanted to mention her role in the IPCC. She was a lead author for, I think, a couple of reports. And the last one, she's been a vice chair of Working Group One, I think, right, Carolina? I had this role, I think a couple of reports ago, and I can assure you, it's a lot of work. It takes up a lot of your time, a lot of your energy. But it's a very rewarding type of work, because everybody recognizes the importance of the work that the IPCC does. With all this, I think I'm going to stop here. And I think, Carolina, you can start your presentation and then we will have general question time. Also online, people can ask questions through the chat online that here James will record, and then there will be the session with the students. Florey, it's yours, Carolina. Okay, thank you. Thank you, Atish. Thank you, Elipo. It's a pleasure to be with you, at least virtually. It's a pity that COVID didn't allow me to participate, to visit the ICTP in person during this meeting of the scientific counseling, and also to give this call of duty. But I'm still happy that I can give you this presentation virtually. So I will start sharing my screen. I would like to know if someone can confirm me that everything is okay. So the title of this presentation, as you can see here is the interplay between the natural variability and anthropogenic climate change at the regional scales. I will concentrate my talk in two sources of knowledge. Just also I thought I like to do this when there are a particular students in the audience. And so it's it can be easier to find the literature that I will describe. And also because I decided to to introduce you with this very complex but also very relevant topic. So I will concentrate a part of them of the work that they will show you that that is from the last IPCC Working Group 1 report that was published last August. And in an unprecedented way, it devotes one third of the content to the regional related issues. And in addition, I would like to share with you part of the work of Julia Mindlin. Julia is a PhD student at the University of Buenos Aires, a head shaper of the University of Reading and myself, her advisors. She is at the third year of her PhD work and she has been doing a tremendous work regarding one of the approaches that I will describe you in this colloquium that the name is Storylines. I also I would like to mention that Julia is member of Yes. Yes, a community. It's an early career scientist of the Earth system that are gathered in an association that they organize themselves to develop a research to communicate, to essentially have a network among the Earth system still related students or early career scientists. So I wanted to also mention that because in case that there are people that doesn't know this kind of associations. So I would like to start my presentations with this slide, this figure that is from the chapter 10. I will use a lot of the material of the chapter 10 of this report because it has been, it's a new chapter that in which we many scientists, particularly the authors of the chapter, but also the Bureau and others, have discussed concepts, ways, methodologies of elaborate regional climate change information. Given the many types of regional climates, the broad range of the spatial and temporal scales, and also we have to consider the diversity of the user needs. So a variety of methodologies and approaches needs to be used to construct regional climate change information. The sources include global and regional climate model simulations, methods like statistical downscaling and bias adjustment methods to deliver skillful information from models to regional local scales. Also regional observations with their associated challenges, right, because there are many regions of the world that still are under observes, even that we have hundreds of years of climate observations. But these regional observations are a key source of the regional climate information construction projects. We need high quality observations that enable monitoring of the regional aspect of climate. And they are used also to adjust inherent model biases and are the basis for assessing model performance. In addition, we use process understanding and attribution of observed changes to large and regional scale anthropogenic and natural drivers and foreseeing are also important sources. All these sources that you can see in this graphic are used when available to distill regional climate information. So we distill information from multiple lines of evidence. The resulted climate information can then be integrated, synthesized, and can be delivered in terms of data or in terms of storylines. I will talk about that in terms of graphics. But as I will show you these synthesis, this climate information will always be associated with levels of uncertainty that need to be assessed and communicated. Regional climate, regional climate change refers to a change in climate in a given region identified by changes, for example, in the mean or higher momentum or moments of the probability distribution of a certain climate variable. And this change should persist for a few decades or longer. It can also refer to a change in temporal properties such as a persistence and frequency of occurrence of weather and climate extreme events. Variability in regional climate arise from both natural and anthropogenic forces as well as from internal variability, including the local expression of large-scale remote drivers and the feedback between them. Due to the many possible drivers of variability and change, quantifying the interplay between internal variability and any externally forced component is then crucial in attempts to attribute causes of regional climate change and to increase the confidence of future regional climate change information. It's important to have an idea of the spatial and temporal scales that are relevant for regional climate change. Regional scales are divided or I define it, sorry, it can be defined as ranging from the size of the local scale, let's say a coastline, a mountain range, a city towards subcontinental areas, for example the Mediterranean basis. So this special length scales range from a few kilometers to a few thousand kilometers. And the relevant driving modes and processes at regional scales, we can summarize them and are displayed in this figure that has on the axis the spatial scales and the y-axis, sorry, the temporal scales. You can see here, I will not enter into details, that it can be evident there is a close relationship between spatial and temporal scales. For example, an individual convective storm may exhibit scales of variability ranging from meters and seconds, two kilometers and hours. But for example, El Niño Southern Oscillation phenomena, ENSO, the scales of variability are regional to hemispheric and even global in extent from months to multi-years in length. These scales interact. The interactions are represented in climate models, although the ability of current model to simulate the regional phenomena and even large-scale climate drivers still live through for improvement, limits their capability to represent interactions across spatial and temporal scales. In addition, process interaction in space is pervasive, sorry, which means that special scales, small special scales often have an influence on the larger scales. Depending on the context, for example, a region may refer to a larger area such as Monsoor region, but the smaller areas such as coastal, mountain range or cities have particularly a combination of these multi-scale and multi-scale problems. And users understood as the people that incorporate this climate information into their activities to take decisions often request climate information for this range of scales, the local ones. Since they are operating an adaptation decision, the decision scales that they consider range from the local to subcontinental level. This is the complexity of the problem. Regarding the sources of regional climate variability and change can be grouped in the external forcing controlling the regional climate and the internal drivers of regional climate variability. Regarding the external forcing, we have the anthropogenic and the natural ones. They are important differences in the process associated with each of them. For example, there are important differences in the processes affected by greenhouse gases, GAs of a land and ocean. For example, this leads to preferential warming of the land region, which are themselves cute towards the northern hemisphere. I will give you some examples because all these topics require a lot of time. But regarding the variations in the solar forces, they could influence the regional climate through its modulation of the circulation patterns. Although this research field is still hampered by large observational modeling uncertainties. The variation of the solar force in our society with the 11 years of life cycle that has been identified as affecting the leading atmospheric circulation most, for example, of the North Atlantic region. Both natural and anthropogenic aerosols are often emitted at regional scale, but have a short atmospheric lifetime from few hours to several days compared to the greenhouse gases that can live in the atmosphere for a dozen of years to 100 of years. So the aerosols are dispersed regional and affect the climate at the regional scale, mainly through radiative pooling and heating and cloud microphysical effects. And the volcanic eruptions, for example, load atmosphere with large amounts of sulfur that is transformed through chemical reactions and microphysics process into aerosols. If the plume reaches the stratosphere, the aerosols can remain there for months and years. About two or three years for large volcanic eruptions and even can be transported through the stratospheric circulation cells. If the eruption, for example, occurs in the tropics, the plume can be dispersed across the earth in a few years. While the eruption occurs in the high latitudes, aerosols mainly remain in that atmosphere. So there are regional differences of the impact of these different forces. In addition, something other forcing that is worth of mention is the land use change. We have an IPCC report that was approved in 2019 depoted on land and it shows the global extent that the land use has degradated the surface of our planet and land cover change and land management can influence the climate locally. In the cities we can mention the human heat islands. Also the changes in the rainfall associated with the cities, but also the deforestation can alter for example in the Amazon, the location and the timing of the convex storms during the wet season. Regarding internal variability, the internal drivers of regional climate variability, they affect substantially the regional scales from seasonal to multi-decay scales. This variability so arrives at large scales from internal modes of atmospheric and ocean variability, like the couple modes, like the Nino-Sauden oscillation or the dipole of the of the Indian ocean, but in addition they drive processes at a regional time scales and they can interact with large scale modes of variability with the response of the climate system to external force forcing. The consequential response is useful to understand the relationship between the large and regional scales. Let's say for example the N so that has the source of its variability in the tropical Pacific, the teleconnections that induce the situational anomalies can influence regions that are thousands of kilometers far from the Pacific. And so these atmospheric modes of variability they have also seasonally dependent regional effects. So the knowledge, a profound knowledge of the season variations and the regional features that this mode can take are very relevant information even for the climate change. And I would also to mention the decadal variability modes, those associated with the Pacific decadal variability, the Atlantic multi-decadal variability that are key drivers of regional climate change on a multi-annual to multi-decadal time scales. Could you please explain what are these acronyms like? Yes, I was mentioned then for example ENSO means El Nino Southern Oscillation that I mentioned that is the source in the Pacific, the IOD is the dipole associated with the Indian Ocean variability PBB and AMB is the Pacific decadal variability and Atlantic multi-decadal variability that also mentioned before and I forgot to mention thanks the annular modes. There are two atmospheric modes. I will concentrate on the southern hemisphere one later in my colloquium. The degree of confidence in climate simulations and in the resulting climate informations depends on the identification of the sources and role of the uncertainties. Regional climate change information projections as well as the global climate change information projections are affected by imperfect knowledge of current and past conditions because of the limitation of the observing systems, imperfect knowledge and implementation of the response of the climate system to external portions as described by the models the internal variability that has limited a limited, we have limited we have a profound knowledge of the internal variability but still limited and also there are limits to its predictability right because of the part of the chaotic nature of our climate system and regarding the future there is the uncertainty of how the future scenarios will be in terms of how the society will evolve in the future. So these sources of uncertainty need to be treated and considered to when we address or when we develop regional climate information. As the title of my colloquium says I will concentrate in the uncertainty associated with internal variability but at the same time I will touch some of the uncertainty associated with modeling and with observations always with focus on the regions but I will add that I also focus my talk in the uncertainty associated with the regional climate information associated with precipitation which is a very challenging variable as you can imagine. How we can manage these uncertainties in regional climate projections or simulations can be simulations of the past or present conditions or these simulations to project into the future. I would like to point it out that original climate projection based on a single simulation from a single model alone will inevitably be affected by not considering the internal variability because we have one we would have one realization of the climate evolution. So that impact of the internal variability is mainly due to the dominant influence of the chaotic atmospheric circulation on regional climate variability and in particular admit to high latitudes. I bring here an example of my region South-Eastern South America that has experimented a positive trend in the summer precipitation but strongly masked by the internal variability. In this example on the right an ensemble of 100 simulations made with a global model of the Max-Tan Institute MPI was used to project this trend. So on the left we have the trend map that we obtain Aberashi the ensemble members with the lowest that is the driest trend and on the right the same but for the ensemble members with the highest or wettest trend it is quite evident that the result change depending on which simulations we consider even from the same model. On the right we have the time series of the observed precipitation anomalies until 2014 and the anomalies represented by the ensemble depicted in gray for the next decades. In green we have the trend of the members with the wettest trend and in brown for those with the driest trend. When we want to project the global mean precipitation trends that is the first graphic in the top of the right one we could see that the uncertainty among the of the difference among the model simulation depicted by this gray band is narrow and all the trends shows a clear positive magnitude. But when we want we go to a region that is the middle panel Saudis de South America we can see that the gray band start to be wider and we can see the differences among the wettest and driest member and these uncertainties even grow when we go at the level of a city my city when the Saudis that you would see there that the uncertainties modern uncertainty quite large and even the differences in the trends. So internal variability is a complex source of uncertainty that we need to address in order to distill relevant climate change information from these simulations. Climate response uncertainties so can be represented not only by multi members of the same model that is of a single model but also can be represented and should be represented by ensemble from different models and so the use of multi model ensembles I will refer a lot during the rest of my presentation has been proved to be a quite powerful tool. Ensemble for climate simulation play an important role in quantifying uncertainties in the simulation output but in addition providing information on internal variability ensembles of simulations can also estimate scenario uncertainties and model uncertainties. The report of the IPCC includes an atlas an interactive atlas that I invite all of you to visit and to play with a different ensemble of models and you could build your own graphics and maps like the one I am showing on the left about the use of these multi model ensembles used in our report. But I will also to make a point about the use of change of models because this is something to consider at the regional climate informations. The change of models can range from the definition of a force in a scenario to global modeling and potentially to dynamic or statistical downscaling and bias adjustment and even if we want to go beyond the climate variables we can link with impact models and other types of models. The program is the propagation and potential accumulation of uncertainties along the change that has been called cascade of uncertainty. So this is there is a structural uncertainty of both of the different models associated with the change and particularly from the size of the climate of the global and the downscaling method the global models and the downscaling methods right that it needs to to be addressed. And another issue relevant for a regional climate projection is the selection of models because there are models that have large too large errors at regional scales and so it is possible to implement ensembles of models for a particular region selecting the models that realistically simulate the process for that given region. The models are improving have improved and continue to do so and becoming better at capturing complex and small scale processes. This figure for example compare the simulations from the three most recent generation of models the same three five and six the names what we use they were available from 2008 2013 and 2021 and they are available for the whole world and you can see their skills for three climate variables and the skill is represented as the correlation between the simulated and observed patterns with one represent the perfect agreement. We can see that these bars represent the range of correlation values associated with each of the multimodal ensembles. So we can see that the models are doing quite great in terms of near surface air temperatures. You can see regarding precipitation that the recent set the best model performance of the set the new the new set that is the one in purple are doing better than the previous one but the worst models are doing more or less the same. So in the last ensemble of models the uncertainty among models have increased. How we attribute regional climate change as to I would say anthropogenic influence human influence. So regarding the model limitation observing limitations and limited process understanding to attribute regional climate change we need to use all the available multiple lines of evidence. Again, let's continue with the example of the wetting of South East and South America. On the left you will have you can see two graphics of the positive trend in summer December to February from two different observational data sets. You could see that they they both show these positive trends but with difference in the location and even the the magnitude of the of these trends. So as I mentioned before limitation the observing system particularly for precipitation in use is observation discrepancy is not only in South America but in many regions of the world. The other thing is how the models represent that trend. You will see in the top right the estimation of these precipitation trends from the observations with two X and then with different sets of models. For example in red is the the multimodal semi-six multimodal ensemble that they show in the previous slides. So you will see that most of the models of the model ensembles that are represented here are able to represent the positive trends but in general they are representing the weakest trend and associated with large model uncertainty. And this with only it is you not only to model limitations but also due to the inter-annual and inter-decade dial natural variability of the climate of this region. So this in order to reduce these uncertainties we also include what we know about the mechanism contributing to this particular trend. I'm showing this example but in the in the report there are others also from other regions and you can see that you can identify which are the large scale drivers of the particular trend. Let's say you could see that the green how gases portion is one of them. But also the stratospheric ozone depletion but as well as large scale modes of variability like the multidicatal variability. Those large scale drivers have a relevant signal in the dynamical conditions of the region that also teleconnection associated and then local processes and that produce a local change. So these different lines of process understanding have been addressed by different scientists from different parts of the world. Some of these lines have been addresses of my particular route. So it is now we have compiled in this graphic all the main process that we know that have a role in addressing this explaining this trend. So a way to reduce the mobile uncertainty observation and services is to pay attention about the how these processes are being represented and how these processes which is the role of each of these mechanisms in explaining the the trend. As I mentioned storylines is an excellent way of constructing and communicating regional climate information. They have a what are these storylines? Essentially are self consistent and plausible unfolding of a physical trajectory of the climate system or a weather or climate event on time scale from hours to multiple decades. I will show you now an example of an event storyline and then of a dynamical storyline. Events storyline is where the particular dynamical conditions during the event and the original warming are specified and control the hazards arising from the event. To give you an example I will show you the results of an article that Julia, my student with Linda van Gardner a PhD student of Helmholtz Centrum Herum have performed a very nice study in which they have conditionally attribute the climate change effect specific to a summer drought in South East America that happened in 2011 and 2012 that was pretty short but it has devastating effect on corn and soybean production. So understanding drought impacts in the region that exceeds the climatological what in time as I made as I showed you before is relevant for decision making and adaptation. Drought attribution attribution typically relies on statistical approaches that focus on changes in the frequency duration and severity. But in this case the use of story lines it's a completely different approach. In this case it is essentially an attribution of the thermodynamic parts of this event right that we call it conditional attribution. And to do that they use a particular method the name is spectral magic technique that combines global models with the analysis of observation and we will not enter into the details of the of the method but the nicest thing is allowed to develop three story lines. One about a word without climate change. So how how would be a drought event like this without climate change that is the counterfactual story lines. The were as we know the factual storyline and it were war by two degrees about pre-industrial. And so they apply this technique to develop these story lines that it is worth to say that the story lines do not predict the future. So they but instead they are able to answer questions like the ones that I am posing on how would be this drought without a human influence how would be this drought into two degree words. And so they is to answer that they study the changes of key thermodynamic variables temperature potentially about the transpiration precipitation. They found that this event was affected by the strong influence of linear although this one of the phases of the linear southern oscillation combine it with regional climate variability patterns. They found that temperature potentially about the transpiration that I am showing the figure are higher in the current climate and in the two degree warming world. But the precipitation show no difference among the story lines. So the paper conclude that the drought would have been there with or without climate change but that climate change costs the regions to be at the higher risk of drought. And that the climate change induce positive precipitation trend in the region is currently outweighing the increased temperature and potentially about transpiration effect. So in the future under a different level of warming remains unclear if the effect of climate change on droughts extreme would bypass the wetting background or not. So this is a very nice example of the outcome of a storyline. Another topic that we like to to briefly talk is the emergence. In climate science emergence refers to the appearance of a persistent change in the probability distribution or the temporal properties of the climate variable compared with that of the reference periods. Filippo have made some research on this topic. In the context of human influence on climate the objective of information studies is the search for the opinions of a signal characterizing and anthropogenically force change relatively to the climate variability of a reference periods defined as the noise. Precise definition of signal and noise as well as metric to measure the relative importance of the signal are key ingredients on the immersion framework depends on the framing question. Immersions is sensitive to spatial scales to noise definition and the choice and length of the reference period. Again, regarding regional precipitation regional precipitation future changes are much more impacted by internal variability than their temperature counterpart and relative to the mean temperature changes the larger influence of this internal variability on the precipitation contributes to a more delayed emergence of the force precipitation response. Here you have an example based on this 76 multimodal ensemble that we use in the APCC report and the emergence has been assessed for the mean precipitation forces changes as a function of global warming levels for all of the regions. The annual mean precipitation change is based on the difference between a 20-year average centering the global warming level minus the mean precipitation during the industrial period taking that reference. The approaches here assume that each grid points in each grid point the emergence can occur when the force change is considered robust and the change is considered to be robust when at least 66 of the models that is 30 out of the 45 that use have a signal to noise ratio greater than 1. The signal to noise ratio is estimated for each model from the ratio between the change and the standard and the function that depends of the standard deviation of the 20-year means to represent the internal variability. And you can see that at a global warming level of 1 degrees emergence that are represented for the dark green or the dark brown, right? So at the global warming of 1 degree emergence only appears in high latitude emissions albeit only with small area fraction that you can see that are still like the greens or like brown. Robust changes in tropicals and subtropical regions only appears for global warming levels of 1.5 Even in other regions and sub-tropical and mid-latitudes the emergence appears with higher levels of global warming levels 3 or 4 degrees. And even at this high global warming levels there are still a large number of these regions with robust changes covering less than 50% of their areas, right? So projected changes in mean climate and extreme are even amplified what is generated by the internal variability. This is just to illustrate the temporal evolution of the emergence, right? And here we have for let's say on the left for the Mediterranean evolution. Let's pay attention to the very dark sorry, the brown line that is thicker and we could see that the internal or the signal not the anthropogenic signal in the Mediterranean precipitation emerged by mid-21st century the drying conditions of the Mediterranean even for the South Asia region that is the one on the right. On the other hand so this is South America that is the top right the fourth graphic the signal in distinguish distinguishing from the internal variability emerged by the end of the 21st century and in the region of the Western Africa they don't emerge it is not only due to the internal variability but also to the model difference the models difference are so large so they are contributing to the lack of emerge of an anthropogenic signal in the Western Africa precipitation. So just to finish I would like to give you an example of the dynamic storylines again from these dynamic storylines in this methodology the global warming level and the remote drivers are specified and control the long-term change in atmospheric dynamic and regional warms so in this way these storylines contribute to attributing uncertainties in regions in regional projections sorry uncertainties in changes of remote drivers which add the interpretation of such uncertainties so the physical climate storylines I will show you an example help to even physically explain contradicting regional projections and thus make the combined information a better representation of the 21st century remember the contradicting regional projection that I showed you some slide ago in Southeast South America which models projecting a wetting and other models projecting a dry right so I will describe a little bit to the annual mean response to climate change of the southern hemisphere on the right you have the change of the zonal winds at the lower levels of the atmosphere the trend projected into the future is that the western least of mid-latitudes will get strength and they will move forward and in association that will move forward the storm tracks that are the preferred regions in which the weather system developed and so the precipitation change is also an increase of the precipitation at mid and subpolar regions and a decrease of the precipitation at a subtropical latitude but this is the changes depicted by the multimodal ensemble mean but as I will show you in a minute the mechanism behind these changes are still poorly understood and the model exhibits a considerable spread in the zonal mean response as I mentioned before there are the annular modes in the southern annular modes that is the leading pattern of large scale circulation but I believe in the southern hemisphere which is associated with negative precipitation anomalies in the polar regions and sorry negative pressure anomalies in the polar region and positive pressure anomalies in subtropical latitudes this is in its positive phase this mode it has a strong inter-annual variability when it's in this positive phase the stratospheric polar vortex strength and this vortex also breakdown is delayed into the spring and the westerlies are stronger you can see an index that just picked the variability of this mode during the last decades and you can see that this inter-annual variability is associated with a positive trend this behavior of the sun has been projected into the future with models but you can see here you have for example the representation of the sun special pattern by two models you can see that the model can have very different special path special representation of this mode but also you can see the evolution into the future of the activity of this mode in the graphic on the right you would see that all models agree that there is a positive trend towards a more frequency of the positive phase of this mode but at the same time with the large uncertainty and so we can use these dynamical conditions to develop storylines about the future of the circulation anomalies in the southern hemisphere and also the regional climate change this is a basis on the paper that we developed with Julia a part of the HRPST work in this case we only consider as climate forcing the GAG increase in the atmosphere we are not considered the ozone the GAG induce changes in the large-scale remote drivers that we are considering two different possibilities the tropical warming or the changes in the startophenic polar vortex that in turn induce changes in the dynamical conditions of the circulation that in turn induce regional climate change this is a there are a mathematics strong mathematics behind this methodology I will not enter into the detail of this equation but I wanted you to know that there is a complex mathematics behind this methodology that I will show you essentially the results that it can be summarized in this figure this figure shows you how the two remote drivers that we are considering that is the changes in the tropical warming that can if you get the tropical warming a more positive trend the tropical warming that alter the habit itself toward the mid-latitudes and can have an impact in the southern hemisphere situation on the other hand changes in the vortex of this polar vortex of the southern hemisphere can also influence the location of the westerlies moving them towards the equator or towards the pole and even intensifying or weakening so we computed for the two drivers the changes in the future for each of the models of the multimodal ensemble that we consider and so there are models that represent there are four essentially four categories right in which the two drivers are positive and large or they are negative and large or they have opposite effects and so you can develop four storylines under these four different possibilities right and so I will show you how the circulation change can be depicted with these four storylines in the middle you have the change of the zonal mean wind in that I showed you before right that associated with westerlies or zonal winds that are stronger at sub polar latitudes and a weaker at the subtropical latitudes but you can see that the change in the winds is different if you have for example a high tropical warming with the late vortex breakdown that is the panel B compare with for example low tropical warming and early vortex breakdown so the response is can be very different considering the changes of the remote drivers into the story and so for a user if in the past we only deliver the change in the middle now we can reduce the uncertainties associated with that including these four storylines that add the changes of the remote drivers into this story this can also be applied to precipitation changes again I will not enter into the details here we have the precipitation change in the three main land areas of the southern hemisphere South America South Africa and Australia with the market market in red you have the precipitation changes depicted by the multimodal ensemble mean but we can add the precipitation changes associated with each of the four storylines and the conclusions are that the two drivers have significant explanatory power in the three regions and so we can deliver tailored regional storylines so just to be concluding I would like to mention to highlight some of the points that I discuss it from one hand we can conclude that global and regional climate models are important sources of climate information at regional scales global models by themselves provide a useful line of evidence for the construction of regional climate information like the storylines that I have just shown you through the attribution of projection of course change or the quantification of the role of internal variability while human influence has contributed to multi-decadal mean precipitation changes in several regions internal variability can delay the emergence of the anthropogenic signal in long-term precipitation changes in many land regions at regional scales internal variability is stronger and uncertainties in observation models and human influence and are all larger than at the global scale precluding a robust assessment of the relative contribution of the different anthropogenic climate forces so the use of diverse methods multiple model ensemble types process understanding and different observed data sets strengthens the robustness of results of regional climate change studies in particular storylines help attributing uncertainties in regional projections to uncertainties in changes of remote drivers that aid the interpretation of uncertainties in climate projections so what else with this I give you just a general idea of what is important to develop climate information including uncertainty assessment what else is this is just the beginning the beginning for developing user-relevant climate information the user-oriented climate information construction cannot only be made by the scientist by the technician should be integrated following a co-production process involving both the users and the information producers into a user contact context sorry that often is already taking into account when constructing the regional climate information in fact the distillation process leading to the climate information should consider the specific context of the user questions at stake the values of both the user and the producers and the challenge of communicating across different the different communities thank you okay thanks very much so I think we will take first questions from the floor Danny or or from the maybe just to open the the debate I think it was clear from from your presentation that it's difficult enough to attribute changes and trends for mean annual precipitation but as you well know what people are interested in are really higher-ordered you know moments extremes and so on okay you maybe give some thoughts as we'll be ever able to who so to identify trends in extreme events that you can really say that these are due to climate change versus just variability because the variability is really huge on the other hand some physical processes are more obvious I mean that extreme events will get more intense in some ways from the physical point of view it's probably more it's probably easier to to find than mean precipitation so it's a complex discussion but I would like to hear your thoughts about this yeah so I fully agree with that the attribution detection attribution community in the world has increased quite a lot this is good and the number of methodology the models and observations used to do the attribution study have also increased also important to to mention is something that you know but maybe part of the audience doesn't know that the the climate science community is well coordinated and well organized across the whole world I think that one of the things that also like to mention is the IPCC was awarded with the peace Nobel Prize no the science as a award in the concept in which this community works together and so in the the detection attribution attribution community significant effort have been made in order to make research studies of specific event attributions and in general the conclusion is that this particular event attribution like the one that I show you about the drought the conclusion is that event attribution regarding temperature extremes or temperature related features can get a significant result in many many regions of the world even in in the latitudes of southern hemisphere like event it hit the wave attribution in my region for example but the precipitation related event are still very challenging in many regions of the world and particularly because of the the consideration that I made in my presentation but in the report if you want only to read the summary for policymakers there are very nice graphics that we have done that synthesize in which regions and for which variables we have been able to assess or to identify the human influence on the change extremes and meaning thanks there's a question I'd have a question thanks Carolina I got intrigued by by this precipitation trend in South America and you you hinted some possibilities like greenhouse gas forcing I mean it's observed trend right you showed this in the last 50 years of the 20th century but then the models seem to indicate we have to wait like 100 years before this becomes really very evident and above the noise level and on the other hand I've seen in the in the southern Alamo trend there was some strange wiggle reaching that region so I wonder if you have any if there was any attribution any more detailed attribution done to what this trend is really due if it's greenhouse gas increase or zone zone related or anything else thanks thanks for the question because it allowed me to to give more detail that I couldn't do it because of the lack of time as you mentioned essentially there are three main large scale drivers that have been identified that have a role in inducing these positive trends one is the multi-decay that variability associated with the Pacific decay the variability the Atlantic decay the variability the other is the greenhouse gas forcing and the other is the trophatic ozone depletion regarding the greenhouse forcing there are several studies I published one of them in 2015 that confirm that you can only explain the magnitude the current magnitude of the observe trend including the greenhouse gases forcing on the other hand there are also in fewer studies that attribute the role of the stratospheric ozone depletion the current ability of the global models in depicting not only the precipitation related processes but also the stratospheric dynamics prevent us currently to assess which is the dominant forcing if the GAG forcing or the stratospheric ozone depletion but both has a considerable role the problem is in the future with the ozone recovery with the ozone recovery in the PhD thesis of Julia we have assessed the role of the the role of the ocean recovery in both the changes in regulation anomalies let's say the sun and the precipitation anomalies and we were able to confirm that the ozone recovery we will have an impact in the activity of the sun but we couldn't have significant results in the regional precipitation change let's take another from the flora Erika Carolina I just want to address the address of the climate skeptic and say okay you have shown a very complicated case in which you have to interpret and attribute the change and understand why the model disagree but as you know for the high moment of the distribution in the report we have been able to attribute and project a lot of extreme signal in many a region where we do not have observation either attribution would you like to comment on how we did that even if it's very difficult I would like you to comment that because you have a key role on that I was tempted to include that in my presentation but I concentrated my presentation in the physical aspects and not on that but please Erika you can mention how we did it and there is a very nice cross chapter that Erika has led to explain this just to follow up on what Filippo says it's very important to be able to look at the high moment of the distribution because what people are interested in what is relevant for life of people are really the extreme events and because in this report the shape of this report was so to be to have a great regional focus so we were able to include a lot of data not only from the past and present but also for the future so we were using a lot of model projection and this even if in some of the region increased the variability as Carolina was showing in some other helped us to really understand which was the confidence in the projection and even if there are regions of the world where we lack a lot of observation and a lot of study then there are still possible to attribute and to interpret the change senior because all the model agree and because we have a process understanding that is driving us so this was able to do so thanks to all this great availability of data that was never reached before let me add something to your great explanation that is my personal opinion is that currently although we have a lot of limitation because as physical scientists I notice that this is a physical scientific characteristic we put a lot of strength when we talk about what we don't know what we don't know what is still the certain because that motivates new research but when we dialogue with the users we realize that we have we know much more than in some regions they are able to use because as I show you the regional climate change information is the beginning is part of a dialogue that involves also people that is expert on the impact people that is expert on the vulnerability and people that will take the decision so sometimes in many regions let's say the energy sector might be very high well technically trained but in other urban planners if I give them four storylines you can use them how would you use this extra addition of information and sometimes they still don't have the tools for that so I think a lot of time we need to devote to the last part of my construction of additional information sorry I was thinking we have a couple of questions from how do we I have I can see two questions in the chat I can answer to Muhammad and Sebastian that they are interested in attribution studies storylines please send me an email I'm not an expert on attribution studies I am more from the dynamic part of the research but definitely I can put you in touch with different people that can also collaborate in these issues there is a question about storylines can you see that one in the Catalina yes yes the one of Sebastian ah can you read the question hi okay so Jill is asking once you get the four storylines what's the next step by design all storylines are equally relevant how to inform users on how possible are each of them this is something that is very new as I show you these are very recent papers to be honest we haven't interacted yet with users with this new tool of the four storylines but I have some ideas that is although the four are plausible they show they show the comparison the contrasting magnitudes gives information or extra information about the ranges of what would you expect in terms of future changes not only because until now you have the change represented by the median or the average and then the full multimodal ensemble dispersion but now we can think differently so you say okay you have these four possible futures but that gives you also an idea of the magnitude and a new idea of the magnitude of the change I don't think that we will talk with the user about the changes in the zone and wind but definitely we can do that in terms of the precipitation and also it's a valuable information to use in impact models to see if this new range of precipitation changes that depict the four storylines how sensitive are the let's say hydrological models to that change for example it will open a new line of dialogue with the users okay we don't have any more questions about the chat line so I would say we can probably stop here the general talk thanks very much Karolina now thanks everybody there will be the meeting with diploma students the group photo okay we have a group photo but I'm not sure how it works everyone everyone okay so everyone done here please everyone now we take our group photo everyone participants just please calm down and stay in front of the desk Karolina we are taking the group photo now with you just depending on us how about online people they cannot enter in the in the photo if they open their cameras they are not able to display just a question