 presentation. Hello, everyone. My name is Sergei Borodko. I'm from Belarusian State University in Minsk, Belarus, and most specifically the National Laws on Monetary Research and Education Center. It is one of the research centers of the university, and I'm going to tell a bit about one of our ongoing topics of research, which is related to analysis and modeling of ozone anomalies. But actually, this is maybe not the primary, so our primary area of research in our research center is developing actually instrumentations for ozone measurements, which is mainly optical passive remote sensing. So this is the instrument, more instrument hardware-oriented part of our research center. But another part, which I'm belonging to, is doing analysis and numerical modeling in the last year as well. So actually, yeah, we do something related to stratosphere-perspirant interactions and ozone climate connections in this. This has started maybe 20 years ago. I mean, the center was founded, and we have, from this theoretical analysis point of view, have analyzed some things like ozone trends, connections between ozone and atmospheric general circulation. And like you can see, analyzing, comparing a number of days from the observational analysis and comparing them to certain types of circulation, this was mainly analysis of observational data. And during the years, one interesting task pointed out, which was related to, is there a way to develop a way of forecasting total ozone content in a very short time on the basis of the meteorological information that you can have. And one of the basis for that was one of the results, which was related to statistical analysis between the daily measurements of total ozone content over the given point and to the surface temperature. So after a number of analysis techniques, it has been found that you can find a very beautiful correlation between the total ozone content over the given point and, say, if it is your station observations over the given city and the surface temperature. But it works only for the extended summer period. And also, you have to do some averaging of the parameters. It should be not instantaneous, like, but a very small, like, two, three or four days. Averaging gives you time series that can provide you with such a nice correlation. So it can be given some interpretation in terms of atmospheric masses changing over the given point. But anyway, it works only in the extended summer period in our region in mid-latitudes. And it was used as a base for a statistical forecasting techniques for the total ozone content for, like, for the next day, for the next few days. But there is one important exception where it is more difficult to use such techniques. And it is this not very, you know, not very numerous cases or numerous days when you have such events as a local ozone anomaly. So, which makes forecasting much more different. Actually, we can say a few words in more general perspectives that, just as everybody knows from textbooks, that we are now speaking about the stratospheric ozone, not the near-surface tropospheric ozone layer. And you know that it is main components of the atmosphere responsible for the unique temperature profile for the very existence of the stratosphere. And it's, so, it's just a very schematic description of the vertical distribution of ozone in the atmosphere. And we can expect that it will be dependent on the processes in the troposphere, that the ozone distribution, the stratosphere will be dependent on tropospheric weather phenomena and vice versa, that we can even expect that there should be a possible influence on the tropospheric weather and climate on the circulation on, large and small scales, depending on the processes in the ozone layer. So, it is a very interesting area of research regarding the question which are the predominant directions of interactions, how strong they are. While we can say that we can imagine changes in the stratospheric ozone distribution coming from the surface weather, on the other hand, if we can point a number of different reasons which may cause variability in the ozone, which are not related to the troposphere, so we can have changes in the stratospheric ozone due to other regions and then point out if you have some change, some effects from then in the surface weather and climate. So, such obvious, the most popular, maybe the most popular issue is like influence of solar activity on the ozone and then the propagation of the signal down to the lowest stratosphere in the troposphere. In fact, the system is endlessly complicated and more specifically for the topic of this workshop, we know that we can also speak about the role of the stratospheric and stratospheric ozone. Of course, in connection with the El Nino Southern installation and other season long scale processes related to tropospheric weather and climate. Yeah, but here we can just say that a lot of external things can be sources of the variability in the stratosphere and so another interesting point of, another interesting area of research is trying to find the top-down propagation and the top-down site of this stratosphere-troposphere interaction, some of the ozone influence on tropospheric weather and climate. This is also what we have been doing for some times and our approach included analysis of correlations between the changes in the stratospheric ozone distribution which can be due to some external reasons not related to a tropospheric phenomenon like what we mentioned above, like solar processes, energetic particle precipitations like that and we can analyze the relations between it, add between large-scale features of the tropospheric circulation which we can analyze by such, you know, such features as station or upper-level frontal zones or the boundary. So the tropospheric amasses, so this is a way of analyzing the interaction of this kind. But the other side of the study, the other side of the kind of troposphere interactions is related to smaller scales and this is what I'm going to focus now on. This is such event, such phenomena as locals on animalists which we may expect to be more like the down-top side of interaction. This is, we may expect that this to be a result of the influence of surface weather on the stratospheric ozone distribution and we shall now see why. Because they universally recognize to be formations of predominantly dynamical origin, why? Because mainly of, you see, they have synoptic scale in both space and time and since they are synoptic scale deviations, we may expect that they should be related purely to the dynamical processes in the troposphere and in the troposphere, upper-troposphere or stratosphere region. They are occasionally named mini-holes for the negative and mini-heist positive, so, but we should know that they are holes but they are not this hole actually which is a very known, maybe their hole, I use the capital H for it because it is the seasonal, the true ozone hole that is forming is above Antarctica every polar winter and we know that it is due to chemical processes and we can speak about the importance of, in fact, of studies of the influence of this large ozone skull on the tropospheric circulation on the surface weather. And what we are talking about is not even this one because this is not, this was not above Antarctica but you remember that true the same, the same ozone hole of the same size approximately just as large as above Antarctica that we had above in the northern hemisphere in 2011. This was also due to chemical ozone destruction, it was also on the season of high scale related to large polar vortex, ozone, chemical ozone depletion and an interesting thing is that actually in both cases we had an important influence of ozone depletion on surface, on the troposphere circulation and on surface weather. We can name at least a few publications that analyze that when we were speaking about the classical the Antarctic ozone hole, we can just mention a few modeling studies and analysis studies that analyze how does this ozone depletion influence the circulation and has effects not just over Antarctica but over other regions like you see subtropical precipitation patterns but it's not just the Antarctic ozone hole because the Arctic record animal in 2011 and also regular lowering of the ozone content of Arctic which are not as large as it was in 2011 but nevertheless we know there is a lot of research relating to that and its influence on the circulation and its relation to the El Nino events and also in fact in the generality is a very important side of the atmospheric process especially for the seasonal forecast and which is what we call this meeting and in fact it is one of the important factors that are to be accounted in the model for seasonal forecast especially the interaction between the circulation, the ozone redistribution and its radiative heating. So what are we calling ozone mini-holes, ozone anomalies are events like that, this is kind of old, this is December 1987, January 1998, it was, yeah it is an optic time scale but it just happened so that it was exactly on the new year day and actually I like this case very much because we are speaking about this mini-hole, anomaly I like this case very much because it happened above my country, it's passed through Central Europe and Eastern Europe and finally passed Belarus and several other countries in Eastern Europe and in fact it is the record low event because it is responsible for the very low total ozone values like 163 drops in units it is comparable to that in the Antarctic, I guess in the Antarctic ozone hole, anyway it is very low for Europe and what we can see is that this is one of the kind of maps that are very often used when looking for ozone distribution so this is obviously from the very well known and used Canadian Environment Canada webpage so we can see, we are looking for total ozone now, we see that at the same moment we have, this is a typical distribution if not the fact that we have such a small amount of ozone here, such a profound ozone minima, so we have both ozone maximum and minima and it is in fact just a dynamical distribution of ozone without any chemical depletion but anyway it is still, ozone hole and it still leads to high ultraviolet exposure values but only for a few days when it takes place, it can be more obvious to look at it on the deviation map where we see deviations of the total ozone from its so called normal values and this is probably which is more often used than the total ozone maps but here one question that we may think of is what should be used as a normal distribution of ozone just for this calculation and it actually depends on who is doing this so in different places we may find that different climatologists, different distributions of ozone like different normal distributions of ozone are used but in fact what we have in the end depends on that because in this case obviously we can see the ozone anomaly both in the total ozone fields and in the deviations from the so called normal distribution but in not all cases are like that, you see that we can maybe speak about some other not so profound animals just here but it becomes important when we try to make some kind of statistic or climatology of local ozone anomalies and in this case results will be dependent on what is used as a normal distribution so and we have actually tried to make such kind of analysis and to go from the case by case analysis which is like when you see when you are looking for papers studying local ozone anomalies it normally analyze like one or two cases and they it is not always easy to just have a like a database of all say local ozone anomalies over Europe for a given time period and so it should be homologically done with something like objective feature extraction so we would like to have an objective identification and track an algorithm for all kinds of anomalies and then apply it to some for analysis or analysis fields it is we can use similar ideas like used another papers related to the general fields of application of objective feature identification algorithms for geoscientific data sets but the point is that different definitions of local ozone anomalies exist so what we are going to use is just a simple definition as deviation over the in the total ozone field from some specified normal distribution which is greater than some threshold value and also we needed to be a contiguous region on the map and also we would like consecutive time steps of the anomaly to overlap and so it would be in fact a continuous region of more than threshold of deviation that is above this threshold value which is a continuous region in the three dimensional space like latitude, longitude and time and also we need to do some subsequent filtering to filter out anomalies that are just not true anomalies but something some spurious things and we can start using just reanalysis data or entering for the extended time period and MACC for the shorter one and actually we would like to compare two ways of specifying the climatology of the ozone and we are taking to 20% deviation for grid points area corrected for the latitude cosine of latitude and the minimal duration of 24 hours and so smaller anomalies will be filtered out and what we get, we are looking only for the European region and what we actually get is some kind of statistic of special distribution of anomalies, it is not like how many days with the anomaly above it but it's actually the number of events so it means that in this region we had about 200 and something events of locals on anomalies more or less pronounced over the period so another, we can obviously see that they are practically very low anomalies according to the above mentioned definition in lower latitudes and also that we have this area which can be explained by that this type of temporal variability in the ozone field is more pronounced here in North Atlantic, yeah this was negative animals and these are the positive ones and we can compare it with what we get if we are using a different definition for the normal distribution of ozone and the results are somewhat different and also the distribution will be looking to be different when we are using an extended distribution, a longer time period from 1991 but it is similar to MACC with the same definition of climatological norm of ozone and another thing is that we can analyze just briefly the temporal distribution and in fact this is what one should expect that we have much more locals on anomalies in winter time than in summer and it comes actually both to negative and positive anomalies so there are virtually no such events in summer and probably what I mentioned before that the statistical forecasting techniques for the total ozone works well only in the extended summer period so we can relate this to that as well, it's a bit different for the different definition but actually coming back to the specified case we can see it's a very, the white actually is lower than 200 dobson units, well so it's really the lowest anomaly ever recorded over Belarus and I guess over Poland, over neighboring countries even by ground-based instrumentation so if you look at their analysis it shows the development of the anomaly over several days and what we are trying to do is check how easy it is to reproduce the ozone anomaly in a general circulation model because yes we will see it is more difficult than reproducing surface weather so the quality, the outcome will be much more different for the ozone than for the surface weather so we can see that just for this case if you are using at least the T255 resolution we can indeed reproduce the, and we actually try to use different initial dates but this is like five to seven days ahead of the anomaly minima, we indeed have some anomaly, this is formed by in the last time moment we don't have it in the model just for this case so we either disappear so move some different direction and if we compare the two cycles of open AFS we have, we can see a small difference and maybe the cycle 40 is more pronounced on the last time step on the model what is actually most interesting for me is the role of the dynamical ozone in the radiation scheme and actually on these time scales as probably one should expect we can see only minimal differences between the ozone taken from climatology to the radiation scheme and prognostic ozone taken to the radiation scheme but it is not always the same so on this we can see more difference on the just on the next time step so there is indeed some difference but it is just maybe a smallest one should expect on this kind of short synoptic time scales but I'm about to finish but I would like to show just another case of anomaly it is a local ozone anomaly over many of the UK and some other parts of Europe in March 2005 it is an example for springtime anomaly and it is peculiar that this one got simulated well as compared to the previous one maybe this is because of a better representation of ozone for these years in the initial data yeah we can see it developing in the analysis and also so yeah the minimum was above the UK exactly so it's kind of interesting case and there is some difference between the cycles of the model but not that much and they are it is in principle it is reproduced well in both model cases model cycles and also in this case both with the climatological and prognostic ozone the result is almost the same and it is much more similar to the analysis than those 1997-1998 case okay thank you