Details 2nd phase


These objectives are based on the research successfully undertaken at the Natural Risk Assessment Laboratory (NRAL) during 2010-2012, focused on the mechanisms of ocean-related extreme events in coastal zones. They  open now a new avenue of the research on climate extremes by:

(i) extending the focus of research from the coastal zones to climate extremes over the entire European continent, including hydroclimate, temperature and wind extremes as well as associated natural hazards and

(ii) moving from highly accurate diagnostics of climate extremes and the understanding of their mechanisms to improve their predictability on decadal and centennial time scales.

During 2010-2012 NRAL, established at the Faculty of Geography of Moscow State University, implemented a comprehensive research programme of ocean-related extreme events in costal zones, centred on understanding their non-linear nature and multifactor character. We developed a comprehensive catalogue of climate extremes over coastal zones of European Russia (Mukhametov et al. 2012) and performed high resolution diagnostic and modelling studies of different types of extreme events resulting in natural hazards, such as extreme wind wave storms, extreme precipitation and associated flash and river flooding, extreme temperature conditions and abrupt changes in the local geochemical balances. In particular, we understood that extreme wind waves may not necessarily follow mean climatological values of wind and wave height and may exhibit strong increases in magnitude even when the mean values are relatively stable, as in the Barents andKaraSeas(Grigorieva et al. 2011). Similarly, extreme precipitation over most European coastal zones originates from clustering of heavy rainfalls into prolonged wet spells, even when the number of wet days is not increasing or being relatively stable over time and nevertheless implying also the lengthening of dry periods (Zolina et al. 2012). The latter become not only longer but also hotter resulting in extremely long heat waves and severe droughts, like the record breaking hot summer of 2010 (Zveryaev et al. 2012). All these extremes result in critical shifts of geochemical balances and geomorphology of coastal zones, moving some of them to the point of no return (Kasimov et al. 2012) and implying a new task for the adaptation to the changed environment which is several orders of magnitude more complicated compared to the present mitigation and adaptation to climate change.

Our studies clearly demonstrated that most of the coastal hazards are associated with the compound nature of climate extremes, quantified through hydrological modelling using high resolution models of surface hydrology (Alexeevsky et al. 2012), synergizing the results of modelling of ocean dynamics (Kulikov et al. 2012), wave modelling (Arkhipkin et al. 2012) and non-hydrostatic atmospheric modelling (Gavrikov et al. 2012). To build an effective system which allows for the synthesis of all components – ocean dynamics, surface hydrology and atmospheric dynamics – we implemented at NRAL most advanced wave models (WAWEWATCH and SWAN), high resolution regional ocean model ROMS, a modelling system for surface hydrology MIKE-3 and the atmospheric high resolution non-hydrostatic model WRF. Never before have all these highly technological numerical tools been employed in a synergistic and holistic way, even at leading operational and forecasting centres. An example of the effective use of such system is a comprehensive diagnostic of extreme precipitation and flooding in the coastal regions of Krasnodar Kraj (Krymsk, Gelendzhik) on 6-7 July 2012 when the record breaking precipitation interacted with highly exposed flooding surface hydrology and orography resulting in catastrophic floods and inundations with more than 200 fatalities and hardly estimated economical loss.


Figure 1. Diagnostics of the extreme precipitation event on 6-7 July 2012 in the coastal area ofKrasnodarregion (Krymsk, Gelendzhik) with WRF model. (a) potential temperature at 500 hPa and surface wind on 00:00 07.07.21012, (b) accumulated precipitation during 24-hr period of 06-07. 07.2012 and aerologic diagram (potential temperature is in blue, equipotential temperature is black) on 02:00 07.07.2012.

Figure 1 shows the results of very high resolution (1-km) simulations of atmospheric conditions with WRF clearly indicating that interaction of the local advection of moisture from theBlack Seawith the local orography resulted in a critical change of condensation limits in the mid and upper troposphere provoking disastrous precipitation. This and numerous other diagnostics pose questions about the local and non-local mechanisms driving extreme events in coastal zones. In an attempt to answer these questions we performed a comprehensive analysis of characteristics of atmospheric cyclones and their role in transporting ocean moisture to the continents (Rudeva and Gulev 2011, Tilinina et al 2012). This required an involvement of NRAL in the international programme on cyclone diagnostics IMILAST (Nue et al. 2012) that allowed for the first time to quantify characteristics of cyclones responsible for coastal climate extremes. Surprisingly, these cyclones are not those generated in the Atlantic storm formation region over the Gulf Stream, as one would expect, but rather those generated or locally re-intensified over the Central and Eastern North Atlantic andWestern Europe(Rudeva and Gulev 2011, Rudeva et al. 2012).

Figure 2. The total number of cyclones in the Atlantic-European sector (a) as well as the number of cyclones (b) and the number of cyclone generation events (c) affecting European continent and European Russia as revealed by the ensemble of cyclone tracks in 7 reanalyses.

Figure 2 clearly demonstrates that cyclones affecting Europe are mostly generated locally or in the eastern part of the North Atlantic with just few originating in theWestern Atlantic. Specifically these cyclones form the major continental storm tracks (figure 2b) which are practically invisible in the climatology of cyclone activity. However, the total number of cyclones in general and the number of European cyclones in particular do not exhibit any clear statistical link with the ocean signal in theWestern Atlantic(see, e.g. Gulev et al. 2002, Rudeva and Gulev 2011). Thus, we hypothesize that the link between the characteristics of these cyclones, particularly their moisture budget and strength and the ocean-atmosphere fluxes providing diabatic sources for their development should be

(i) investigated in terms of ocean impact on the atmospheric circulation through the extreme of surface fluxes (Gulev and Belyaev 2012),

(ii) the ocean signal should be detected over the whole midlatitudinal and the Eastern Atlantic rather than in the Western Atlantic where most cyclones are formed and

(iii) likely the response is evident on time scales longer than several decades and associated with the Atlantic Multidecadal Oscillation (AMO).

To test these hypotheses we built a new long-term reconstruction of surface heat fluxes in the Atlantic (Gulev and Belyaev 2012) which clearly demonstrated a distinct link with the ocean temperature signal, for the first time confirming on a quantitative level the Bjerknes (1964) conjuncture (Gulev et al. 2012). Our model experiments (Zuev et al. 2012, Shelekhova et al. 2012) using the advanced climate models which were also implemented at NRAL, clearly demonstrated that the processes of air-sea exchanges in the midlatitudinal North Atlantic and associated northward heat transports may be indeed responsible for triggering atmospheric midlatitudinal jets at its eastward end resulting in the generation of systems further responsible for the local climate extremes.

Summarizing, now, we can state that on one hand, we know how different types of climate extremes are generated on a short time scale and which mechanisms are responsible for the strength of these events. On the other hand, we know which ocean signals are likely responsible for affecting these mechanisms and on which time scales these signals can be identified. This makes it possible to approach for the first time predictability of these extreme events and to clearly answer the question which of them are potentially predictable and which are not using state of the art climate models and methodologies of diagnostics. Our working assumption is based on the critical role of the ocean signal in forming anomalies of atmospheric circulation and on the strong impact of regional high resolution processes on these anomalies, which being analysed together in a holistic way, will help for the first time to directly engage the predictive potential of the ocean in the prediction of climate extremes and associated natural hazards. This justifies our objectives for the period 2013-2014 stated above.

Being still focused on climate extremes and natural hazards, we will implement novel research avenues of which the most important are the following:

1) We will expand our focus to the climate extremes which are not exclusively associated with coastal zones. This is justified by the strong interconnection between the coastal extremes and hazards and climate extremes over other regions. Furthermore, the analysis of predictability unavoidably requires the analysis of scaling of extremes since there might be particular space-time scales on which different extremes are most predictable. This, in turn, implies the necessity to engage e.g. remote temperature signals on glacier and snow melting working either hand in hand or out of phase with precipitation and surface hydrology to force the local flooding.

2) We will move from the local diagnostics and short-term forecasting and now-casting of climate extremes to their predictability on longer (decadal to centennial) time scales. This will require the use of results from advanced climate models accounting for all mechanisms responsible for changing intensity and frequency of extreme events to a full extent. Being focused now on predictability we will for the first time address the issue of predictability of risks of natural hazards, which requires a synergy between the occurrence and magnitude of climate extremes as revealed by climate predictions with the exposure of the infrastructure and human to these extremes in specific regions.

3) We will make a revolutionary step from the advanced statistical methodologies for quantifying extreme events which were developed during the first stage of the project to the measures of extremeness relevant for specific regions and potentially predictable. This step requires the development of multi-dimensional statistical distributions for many of which the mathematical concepts are not yet designed. For instance, for the accurate prediction of the flooding, duration-intensity distributions for precipitation should be merged with the water level extreme distributions. Similarly, probability densities of cyclone deepening rates should be analysed together with wind and wave distributions from exponential families.

These directions will make it possible to undertake new research bringing new knowledge for climate science and  tangible deliverables for society of which the assessment of the end-to-end predictability of climate extremes on different space-time scales and associated risks for natural hazards will become the most valuable outcome of the project.

A paradigm of predictability and the role of the ocean (WP1, WP2).

During the period of 2013-2014 the research project will target in a holistic manner the predictability of climate extremes over European Russia by accounting for the compound nature of extremes, their strong scaling in different regions and by establishing new effective measures and indicators of climate extremes which are on one hand most predictable and on the other hand are most effective in characterization of the impacts of climate extremes. By linking the global climate projections with regional strongly localized extremes we will be able to quantify the scales and event types which are most predictable and those which cannot be predicted. Both will be then transposed into the assessment of risks for natural hazards – the former to the risks which can be predicted, and thus mitigated by applying actions planned well in advance – and the latters which cannot (or can hardly be) predicted with the present level of our understanding and thus require adaptive actions including  preparedness. In many respects the quantification and regionalization of the unpredictable risks is of the same or even of a higher value than that for the predictable risks since it allows to avoid unjustified strategic changes in the economic infrastructure, helps to effectively distribute the resources between the mitigation and adaptation actions and minimizes overall losses.

Targeting predictability, we will focus on major large-scale factors of predictability, since the local predictability of climate extremes is limited by time scales spanning few hours to several days and is provided by the existing operational forecasting systems already. Extreme rainfalls and associated floods as well as the rapid development of severe wind conditions can be predicted quite successfully some 1-2 days in advance by operational weather prediction systems, although there is big room for progress in this area, of course. What cannot be predicted accurately is the occurrence of the situations pre-conditioning these extremes on time scales from a decade to several decades. Even knowing explicitly the mechanisms driving extreme events, we are never sure that these mechanisms are adequately replicated by the climate models used for the predictions and, thus, that the responses are really those associated with the processes considered as predictors and not with other processes. Considering decadal time scales, we deal with a complicated mixture of anthropogenic signals and natural variability which interact in an essentially non-linear way. On centennial scales, climate change signals likely dominate over natural modes making our very long-term predictions even more accurate than those for several decades (the so-called Type II predictability paradox). However, on the decadal scale, we are still lacking the skills of predictability even for the mean state of climate, and especially for climate extremes.


Figure 3. Schematic representation of the predictability skills of the climate system at different time scales. The tone of light yellow – red colour scale reflects the skills of predictability. Predictability is more successful for short time scales (Type I predictability, initial value problem) and for very long time scales (Type II predictability, boundary conditions and external forcing problem). The poorest predictability skills are identified for time scales from several years to several decades.

This paradox schematically represented in Figure 3 showing that time scales spanning the range from several years to several decades are characterized by very poor skills of predictability. Exactly on these time scales the ocean’s role in climate (natural modes) starts to be comparable with the magnitudes of anthropogenic signals that make, on the first glance, the prediction on these time scales very difficult. However, the progress recently made in climate modelling and particularly in resolving the oceanic components of the climate system (Taylor et al. 2011, Palmer et al. 2008, Penduff et al. 2011), ensures that the ensemble simulations of the future climate under the IPCC Fifth Assessment Report (IPCC AR5) in contrast to the previous one (AR4, Meehl et al. 2007) will contain much more predictability on decadal time scales. The question is just how to accurately quantify these signals and to associate them with extreme events in particular regions.


Developing new measures of extreme events relevant for climate prediction (WP2, WP4)

We argue that the analysis of extremes in climate model simulations lacks the consistency between the measures of extremes used and the model capabilities to replicate the extremes. Under IPCC AR4 and AR5 we tried to apply to the results of the model experiments the metrics of extremes which were developed on the basis of long-term time series of in-situ observations in present climate (e.g. Klein, Tank and Koennen 2003, Trenberth et al. 2007, Scaife et al. 2008). Firstly, this approach does not account for the fact that model probability density functions (PDFs), even being of the same mathematical type as those for observations, may exhibit a drastically different shape. Secondly, this approach does not account for the highly localized nature of climate extremes. For instance, precipitation in the model is estimated as grid-cell average which can be hardly compared to the local point measurements provided by in-situ data. This is especially critical for still relatively coarse resolution climate models. In other words, we try to evaluate model experiments with inappropriate tools that critically downgrade the predictive potential for extreme events (which is already in question due to the interplay between natural modes and anthropogenic signals mentioned above). As a result, model based estimates of climate extremes are hardly comparable with those revealed by observational data even for the present climate and consequently, the climate projections for extreme events are much less accurate than those for climate means.

In this project, in order to assess the limits of predictability of climate extremes, we are going to use an ensemble of CMIP5 simulations performed with the best presently available climate models  to  apply to these simulations advanced methodologies for quantifying the magnitude and frequency of climate extremes during the 21st century. For this purpose we will use advanced statistical approaches developed during the project term in 2010-2012 and will establish the transfer functions between characteristics extremes in the data and in the model simulations. This part of work will be done using the 50-year time period from 1960 to 2010 well covered by data and by CMIP5 model simulations of the present climate. For precipitation, temperature, extremes winds and other parameters we will provide comparative assessments of the model and observationally-based probability density functions and will justify regional thresholds and transition algorithms for designing model-relevant metrics of climate extreme events. In particular, we will consider continuous PDFs of the absolute and relative extremeness and duration of wet and dry periods for precipitation (Zolina et al. 2009, 2010, 2012), fetches, steadiness and durations of extreme winds as well as separate extreme statistics of wind sea and swell for the analysis of storminess (Grigorieva et al. 2011, Badulin and Grigorieva 2012), and area-integrated discharges and ground water recharges for the surface hydrology. When new model-relevant metrics are justified they will be applied to CMIP5 simulations of the future climate in order to build an accurate representation of extreme events in climate models.





Identifying major mechanisms driving variability in extreme events in climate models (WP3, WP4)

Climate models differ from the nature not only by the way they replicate the extremes but also in many other respects. In particular, the continental scale hydrological cycle in general and for the Eurasian continent in particular demonstrates significant differences when estimated from model results and observational data.

Figure 4. (a) Representation of the atmospheric moisture transport through 60N for the present climate conditions in the ERA-Interim reanalysis (grey) and the ECHAM5 climate model (red) as well as in the ECHAM5 climate model for the last 4 decades of the 21st century (2060-2100)) (after Bengtsson et al. 2012). (b) Storm tracks in ECMWF reanalysis showing that most cyclones enter the high-latitudinal regions (and provide most of transport) exactly in the area where the uncertainties (panel a) are the largest.

Bengtsson et al. (2012) clearly demonstrated that the advection of atmospheric moisture as revealed by reanalyses accurately replicating the present climate and by the model experiments for the present climate state are quite far away from each other with most differences in the Atlantic-European basin, where the major fraction of the moisture transport is provided (Figure 4). Remarkably, the projected moisture transport for the second half of the 21st century implies the changes in the transport magnitude which are less than the difference between the present climate estimates in model and in the data. This clearly shows that unless consistent metrics quantifying the transport of ocean-related signals to the Eurasian continent are established, it is difficult to build climate forecasts of extreme events. In order to provide these metrics under our project we will develop the analysis of cyclone activity with a particular emphasis on extreme cyclones. From our project results during 2010-2012 (Rudeva and Gulev 2011, Tilinina et al. 2012, Kravtsov and Gulev 2012, Gavrikov et al. 2012) we understood that the cyclones resulting in extreme events (e.g. in abundance precipitation) may not necessarily be extreme by nature, exhibiting for instance explosive deepening or extremely fast development. Many disastrous floods, especially in summer, are associated with quite ordinary cyclones, as for the event of 2002 flooding in Central Europe (Ulbrich et al. 2003) and for the Krasnodar flooding in 2012 (see, figure 1). Thus, the most relevant metric can be the transport of atmospheric moisture by cyclones and the evolution of cyclone moisture and energy balance. To implement these metrics we will use a methodology developed by Rudeva and Gulev (2011) and will enrich it with the direct estimation, besides the recycling ratio, of the lateral advection of moisture into the cyclone domain. Applied to the results of the model simulations, this methodology will provide accurate estimates of the role of cyclones in the advection of atmospheric moisture from the ocean to the continent.


Regionalization of extreme events in climate forecasts and projections for impacts (WP4, WP5, WP6)

In the next step it will be important to quantify the anomalous responses of European climate to the anomalously strong or weak atmospheric moisture transport. During 2010-2012 we analyzed the variability of the duration of wet and dry periods overEuropeand concluded that trends in the durations of dry spells are not always and everywhere opposite to trends in wet spell durations. Over most ofEastern Europeand European Russia both wet and dry spells extend in length, especially in the cold season, suggesting a regrouping of the wet days into more prolonged wet episodes and dry days into more prolonged dry episodes (Zolina et al. 2010, 2012). This process, comparable to a redistribution of beads on a necklace with a fixed total number of beads, represents one of the most remarkable climate phenomena of the hydrological cycle inEurope. To evaluate the presence of this phenomenon in the model simulations of the future climate we will use an advanced statistical formalism developed by Zolina et al. (2012) allowing for estimation of the fractional contribution of wet and dry periods of different durations to the total number of wet and dry days using the Fractional Truncated Geometric Distribution (FTGD).

When the effects associated with the changes in the wet and dry spell durations are quantified, it will be important to identify the mechanisms responsible for the observed changes in wet and dry period durations. Cold season increases in the duration of wet spells and a growing occurrence of extremely long wet spells could be attributed to changing cyclone activity in the Atlantic–European sector. Although some studies (e.g., Gulev et al. 2001; Trigo 2006; Wang et al. 2009; and others) found that the number of deep cyclones has an increasing tendency over Europe, to establish a direct prognostic link between cyclone activity and the wet spell durations the sole consideration of the number of cyclones may fall short. Prolonged wet periods are more likely linked to the occurrences of cyclone series (Mailier et al. 2006) or to particular regional airflow regimes (e.g., Maraun et al. 2010) that may, indeed, change the degree of intermittence of wet and dry days to a larger extent than just the number of cyclones. In the analysis of the model results this will be particularly important for understanding the strong cold season increase of the duration of wet periods overEurope. During the last few decades an enhanced poleward deflection of the Atlantic storm tracks was observed in this region and this tendency is expected to become more pronounced in future climate (Leckebusch and Ulbrich 2004; Löptien et al. 2008). Thus, will we anticipate a stronger intermittence of wet and dry regimes under this phenomenon? And even more specifically, are the cyclones populating the deflected (in the future climate) storm track will provide more extreme conditions? Answers to these questions will provide the end-to-end consideration of the mechanisms forming the hydroclimate and wind extremes over Europe and will allow for quantifying predictability, ensuring accurate prediction of the extremes (for the type of processes and regions which hold the predictability).

Even when the climate projections of regional extremes are developed, we still will not be capable of the precise forecasting of the localization and timing of extremes in the next several decades, be it extreme rainfall, flash flooding, drought, wind storm or extreme wave height. What can be drawn from the model experiments are the regional tendencies in the frequency and magnitude of climate extremes. However, the fact that for instance the number of events with prolonged rainfalls exceeding in magnitude a given threshold will increase during the period 2040-2060 in a particular region (for instance Southern Russia) does not indicate when and where exactly large inundation events will occur. This is what cannot be revealed from climate model runs and what is unreasonably to request. However, using climate model output, we can derive well founded regional statistics to further analyse selected cases of extremes which are qualitatively and quantitatively consistent with those observed in the present climate. The instruments for this analysis will be (i) a local mesoscale atmospheric model WRF, (ii) high resolution wave models WAVEWATCH and SWAN and (iii) the regional ocean dynamics model ROMS, already implemented at NRAL during the first stage of the project in 2010-2012. For the time periods 2040-2060 and 2080-2100 we will select specific extreme events (extreme rainfalls, flash and river flooding, local intensifications of winds, continuous periods of very high and very low temperatures) and will undertake a series of experiments with the abovementioned models to provide regionalization and downscaling of these events. This will allow for the quantitative estimation of the impacts of extreme climate events in the future climate, based on the projections of extreme events on climate models.

In particular, to quantify the local mechanisms underlying extreme events in the Nordic and Arctic shelf seas, numerical experiments with the regional atmospheric model WRF, the wave model WAVEWATCH and the regional ocean dynamics model ROMS will be performed to simulate specific extreme events such as wind waves, forced for specific case studies with both reanalysis (for present climate) and CMIP5 model (for the future climate) data. The use of the WRF system will be two-fold in these experiments. First, the WRF non-hydrostatic model in a very high resolution (6 km down to 750 m) will be used to adequately simulate the highly localized wind patterns responsible for extreme marine storms and storm surges. Second, the model will be employed to quantify local changes of the energetics of atmospheric synoptic transients causing the extremes. This holistic approach will provide explicit diagnostics of the regional extremes and will help to come up with an end-to-end understanding of the nature of these extremes. In a further step, the same model suite driven with short time periods of transient simulations from the subset of the CMIP5 models will be used to downscale simulated individual future events to a very high resolution.

Using these accurate hindcasts of extreme climate events in reanalyses and high resolution experiments with WRF and other mesoscale modelling tools, we will analyse the magnitude and the occurrence of these events in climate model simulations. The purpose of this activity is to obtain and quantify characteristics of climate extremes in long-term model runs and to quantify the extent to which models are capable of replication the observed magnitudes and frequency of extreme climate events. For this purpose we will use long-term climate simulations with climate models of the CMIP5 ensemble and will identify climate extremes using the peak-over-threshold approach. In the next step, these model case studies will be analysed using the high resolution WRF system in order to obtain quantitative characteristics of extremes and to compare them to the observed magnitudes. This will allow for establishment of the scaling ratios between the model-derived extreme events and those observed and for quantification of predictability of these events in the future climate. For quantifying occurrences and magnitudes of extremes in the model runs we will use well established statistical approaches based on different types of extreme value distributions (e.g. Zolina et al. 2009, 2010, 2012). For example for extreme precipitation, besides standard metrics, of a special interest will be the duration of wet and dry spells and associated precipitation intensities. Similarly, for the analysis of wind storms we will provide directional statistics of extreme winds and waves (e.g. Gulev and Grigorieva 2006).


Addressing the scaling problem (WP2, WP5)

This activity addresses the problem that the global climate models in general still have a coarse horizontal resolution (usually coarser than 100 km) and do not provide information about local scale circulation dynamics and feedback mechanisms that ultimately might cause extreme events; this argument holds in particular for hydrologic and wind extremes. Furthermore the impacts of extreme events are often experienced locally on scales not resolved by global climate models. Increasing the model resolution may help to solve this problem to some extent (Wehner et al. 2010), but such experiments are still rare and expensive. To close the gap between coarse resolution global climate models and local scale requirements, different dynamical and statistical downscaling approaches have been developed (e.g., Maraun et al, 2010b). Despite the improved resolution, highly localised extreme events are still poorly represented by most downscaling techniques (e.g., Maraun et al., 2010). Here, very high resolution regional climate model simulations are required or statistical approaches that make explicit use of extreme value theory (Vrac and Naveau, 2007; Kallache et al., 2010; Maraun et al., 2011). Regional climate model simulations have the advantage of providing physically and spatially consistent simulations, but show considerable biases compared to observed climate variables (Christensen et al., 2008). As a consequence they are often not directly applicable for impact studies. Therefore, current research focuses on combining the advantages of dynamical and statistical downscaling by a statistical correction of regional climate model simulations using so-called bias correction or model output statistics. However, no specific bias correction methods have been developed yet that explicitly address extremes by extreme value theory.


Figure 5. Total number of northern hemisphere cyclones per winter as a function of spectral resolution (Tn) for four different cyclone intensity classes: (a) all cyclones and (b) minimum sea-level pressure (SLP) less than 980 hPa, Results are shown for ERA-40 re-analysis data (circle), seasonal integrations at different resolutions (solid) and operational analysis data truncated at different total wave numbers (dashed). Also shown are the 95% confidence intervals. (After Jung, Gulev and Rudeva 2006).


The model resolution problem is not only critically important for extreme event statistics, but also for quantifying atmospheric cyclone characteristics. Figure 5 shows how the number of cyclones critically depends on the spectral resolution of the models. It is important to stress that the effect is quite different for all and for very deep cyclones with the latter being less dependent on the resolution. However, given that continental extremes are not necessarily associated with extreme cyclones (see above), resolution may critically affect the further representation of extremes in model runs. It is important to disentangle the dynamical/physical effects of horizontal resolution from those due to pure spectral truncation. Figure 5 shows the number of cyclones obtained for the high-resolution analysis (the ‘truth’) along with those obtained from the spectrally truncated SLP analyses (dashed lines). For the most intensive cyclones the truncation effect cannot explain the sensitivity of the total number of extratropical cyclones described above. For shallow cyclones, on the other hand, the truncation effect clearly dominates over the dynamical/physical effects of horizontal resolution. Given the fact that most continental cyclones are predominantly shallower than those over oceanic storm tracks we do anticipate a strong impact of model resolution on the representation of extremes and will address this issue in our project.

To study the influence of model resolution on the representation of extreme events, scaling experiments with different resolutions will be carried out with ECHAM5/OM and some other models from the CMIP5 ensemble comprising T31 (approx. 400×400 km2), T42, T63, T106 and T213 (approx. 50×50 km2) and even higher resolutions covering a period of 50 years. In order to avoid spurious differences due to different evolutions of long-term climatic modes of variability, the AGCMs will be forced by identical observed oceanic boundary conditions. As a byproduct, this experiment will yield an optimal atmospheric resolution to represent extreme events, which will be used for further sensitivity studies.

Synthesis and assessment of risks in climate prediction of extreme events (WP6, WP7)

At the final stage of the project we will use the developed projections of extreme events in order to assess the risks of impacts of these events onto ecology and infrastructure of different regions exposed to climate extremes. For instance, exploration of the natural resources on the Arctic shelf as well as developments of pipelines in the Black and Baltic seas are strongly affected by climate factors, making these regions highly vulnerable due to climate change and to associated natural hazards, the key concerns being sea ice reduction, sea level rise, land loss, changes in maritime storms and harsh weather events, responses to sea level rise, implications for permafrost degradation and regional water resources. Furthermore, the integration of coastal structures into the economy goes far beyond coastal regions per se and stresses that the economy and life conditions of the coastal zones crucially impact on the economy and life conditions in the inland regions. Observed and projected climate changes imply changes in the depth, type and area of permafrost. This may critically affect the whole infrastructure of high latitudinal regions. For many years the stability of this infrastructure largely relied on the stability of the permafrost and with climate change this stability is under threat. To meet these challenges reliable estimates of the key environmental variables in the nearest future are very important, since they impose changed requirements for the strategic planning of the socio-economic development. The proposed project set-up will aim at addressing these issues relevant to society.

For instance, intensifying warming and associated rapid Arctic sea ice retreat during the last decades has a major impact on the Arctic marine navigation and shelf exploration development. The prolonged open water conditions in the Arctic marginal seas may lead to new perspectives for the socio-economical development of Polar Regions and functionality of the Northern Sea Routeand Northwest Passagein the 21st century. At the same time, the extreme storm events and high sea waves may enhance hazards for marine navigation, fishery and shelf exploration system. Design criteria for ships, maritime infrastructure, exploration and exploitation infrastructure and shore-based facilities will have to be adapted to changed environmental conditions. Therefore, the model assessments of extreme storm events and associated sea wave activity have very important implications for the possible projection of the shelf exploration and marine navigation in the Arctic basin.

In a final step we will apply the regional projections of atmospheric forcing to high-resolution terrestrial models to estimate, on the regional and local scales, hazards and risks. These models comprise high resolution topographical and bathymetrical data, detail land use such as agriculture, settlements, and infrastructure and population numbers. Using different scenarios for the forcing, the risk analysis will result in identifying areas of different degrees of risk which will be mapped. In applying present and future scenarios for land use and economic development alternatives will be developed to allow studying the impact of changing risks onto the economic indicators (IPCC-SREX, 2012 a, b). In a further step a methodology will be developed in co-operation with regional and local authorities and agencies to identify mitigation strategies and measures.

Results from estimating the risk of natural hazards in key regions of European Russia, such as the White Sea, the Baltic, Black, Azov andCaspianSeasin the first phase of the NRAL Laboratory are available to be extended to these risk assessments. State-of-art assessment developments will be together with co-operation partners (Mazzorana et al, 2011) adapted to NRAL requirements. Combining physical data with socio-economic data we will develop comprehensive risk assessment information leading to regional hazard and risk atlases. One key area where this development is already underway is Krasnodar Kraj where substantive analysis and projections have been started already before the Krymsk floods in July, 2012. This approach will be used for developing canonical information products, applicable to other key regions. In parallel we will adapt and improve statistical methods to minimize the uncertainties of risk calculations. Recommendations of the European Community (CEC, 2004, 2010) will be used as guidance to develop strategies consistent within a larger context.


Алексеевский Н.И., Айбулатов Д.Н. Динамика гидрографической сети и морского края дельты Волги с 1800 по2010 г. Россия, Москва Вестн. Моск. ун-та. Сер.5. География. №2. 2011. С. 96–103

Зуев В.В., Семенов В.А., Шелехова Е.А., Гулев С.К., Колтерманн П. (2012) Оценки влияния океанического переноса тепла в Северной Атлантике и в Баренцевом море на климат Северного полушария. Доклады РАН, т. 445, № 5., с. 585–589.

Семенов В.А., Мохов И.И., Латиф М. (2012) Влияние температуры поверхности океана и границ морского льда на изменение регионального климата в Евразии за последние десятилетия. Изв. РАН Физика атмосферы и океана. Т. 48. № 4. С. 403–421

Золина О.Г., 2011: Изменения продолжительности дождливых периодов в европе с 1950 по 2008 и их связь с экстремальными осадками. Доклады академии Наук, 436, 2, 279-283, doi: 10.1134/S1028334X11020206

Bengtsson, L., K.I.Hodges, S. Koumoutsaris, M. Zahn, N. Keenlyside, 2012: The changing atmospheric water cycle inPolar Regionsin a warmer climate. Tellus, 64, 1–14, DOI: 10.1111/j.1600-0870.2011.00534.x

Bjerknes, J., 1964: Atlantic air–sea interaction. Advances in Geophysics, 10, Academic Press, 1–82.

Blanchard, G., M. Kawanabe, M. Sugiyama, V. Spokoiny, K.-R. Muller, 2006: In search of non-Gaussian components of a high-dimensional distribution, J. Mach. Learn. Res., 7, 247–282.

Budgell, W.P., 2005: Numerical simulation of ice-ocean variability in the Barents Searegion, Ocean Dynamics, DOI 10.1007/s10236-005-0008-3.Badulin, S. I., V. G. Grigorieva: On discriminating swell and wind-driven seas in Voluntary Observing Ship data Journal of Geophysical research, VOL. 117, C00J29, 13 PP., 2012 doi: 10.1029/2012JC007937

CEC: Commission of the European Communities (2004) Communication from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of the Regions—Flood risk management—Flood prevention, protection and mitigation,                                                                                                                         http://eur-lex. Access 27 October 2010

CEC: Commission of the European Communities (2007) Directive 2007/60/EC of the European parliament and of the council of 23 October 2007 on the assessment and management of flood risks.                                                                                      http://eur-lex. Access 12 October 2010

Chen, Y., W. Hardle, V. Spokoiny, 2010: GHICA – Risk analysis with GH distributions and independent components, J. Empirical Finance, 17, 255–269.

Grama, V. Spokoiny, 2008: Statistics of extremes by oracle estimation, Ann. Statist., 36, 1619–1648.

Grigorieva, V.G, S.K. Gulev and K.P. Koltermann, 2011: Extreme wind waves in the marginal RussianSeas. Geography, Environment and Sustainability, No 02 (v. 04): 22-29.

Gulev, S.K., and K.P. Belyaev, 2012: Probability distribution characteristics for surface air-sea turbulent heat fluxes over the global ocean. J. Climate, 25, 184-206, doi: 10.1175/2011JCLI4211.1

Gulev, S. K., M. Latif, N. Keenlyside, and K. P. Koltermann, 2012: North Atlantic Ocean Control on Surface Heat Flux at Multidecadal Timescales, Nature,    Nature, 499, 464–467, (25 July 2013), doi:10.1038/nature12268

IPCC-SREX, 2012a: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change [Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D. Mastrandrea, K.J. Mach, G.-K. Plattner, S.K. Allen, M. Tignor, and P.M. Midgley (Eds.)].CambridgeUniversityPress,Cambridge,UK, andNew York,NY,USA, 582 pp.

IPCC-SREX, 2012b: Summary for Policymakers. In: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation [Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D. Mastrandrea, K.J. Mach, G.-K. Plattner, S.K. Allen, M. Tignor, and P.M. Midgley (Eds.)]. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change.CambridgeUniversityPress,Cambridge,UK, andNew York,NY,USA, pp. 1-19.

Kallache M, Vrac M, Naveau P, and Michelangelo, P.-A.: Nonstationary probabilistic downscaling of extreme precipitation. J. Geophys. Res. 116, D05113, 2011.

Kasimov, Nicolay S., Alexandre N. Gennadiev, Maria S. Kasatenkova, Michail Y. Lychagin, Salomon B. Kroonenberg, Peter Koltermann. Geochemical Changes in the Caspian Salt Marshes Due to the Sea Level Fluctuations, Earth Science Research, Vol 1, No 2 (2012), pp. 262-278

Kharin, V.V., and F.W. Zwiers, 2005: “Estimating extremes in transient climate change simulations.” J Climate, 18, 1156-1173.

Klein Tank A.M.G., and G. P. Koennen, 2003: Trends in indices of daily temperature and precipitation extremes in Europe, 1946–99. J. Climate, 16, 3665–3680.

Löptien, U., O. Zolina, S. K. Gulev, M. Latif, V. Soloviov, 2008: Cyclone life cycle characteristics over the Northern Hemisphere in coupled GCMs. Clim. Dynamics, DOI 10.1007/s00382-007-0355-5.

Mailier, P. J., D. B. Stephenson, C. A. T. Ferro, and K. I. Hodges, 2006: Serial clustering of extratropical cyclones. Mon. Wea. Rev., 134, 2224–2240.

Maraun D, Rust HW, and Osborn TJ: Synoptic airflow and UKdaily precipitation extremes. Development and validation of a vector generalised linear model, Extremes 13, 133-153, 2010a.

Maraun D, Wetterhall F, Ireson IM, Chandler RE, Kendon EJ, Widmann M, Brienen S, Rust HW, Sauter T, Themessl M, Venema VKC, Chun KP, Goodess CM, Jones RG, Onof C, Vrac M, and Thiele-Eich I: Precipitation Downscaling under climate change. Recent developments to bridge the gap between dynamical models and the end user. Rev. Geophys. 48, RG3003, 2010b.

Mazzorana, B., F. Comiti, S. Fuchs (2011): A structured approach to enhance flood hazard assessment in mountain streams, Nat Hazards, DOI 10.1007/s11069-011-9811-y

Meehl GA, Stocker TF, Collins WD, Friedlingstein P, Gaye AT, Gregory JM, Kitoh A, Knutti R, Murphy JM, Noda A, Raper SCB, Watterson IG, Weaver AJ, Zhao Z-C (2007) Global Climate Projections. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (Eds) Climate change 2007: The Physical Science Basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change.CambridgeUniversityPress,Cambridge.

Neu, U., M. G. Akperov, N. Bellenbaum, R. Benestad, R. Blender, R. Caballero, A. Cocozza, H. F. Dacre, Y. Feng, K. Fraedrich, J. Grieger, S. Gulev, J. Hanley, T. Hewson, M. Inatsu, K. Keay, S. F. Kew, I. Kindem, G. C. Leckebusch, M. L. R. Liberato, P. Lionello, I. I. Mokhov, J. G. Pinto, C. C. Raible, M. Reale, I. Rudeva, M. Schuster, I. Simmonds, M. Sinclair, M. Sprenger, N. D. Tilinina, I. F. Trigo, S. Ulbrich, U. Ulbrich, X. L. Wang, H. Wernli, 2012:  IMILAST – a community effort to intercompare extratropical cyclone detection and tracking algorithms: assessing method-related uncertainties. Bull. Amer. Meteor. Soc. 93, in print.

Palmer, T. N., F. J. Doblas-Reyes, A. Weisheimer, and M. J. Rodwell, 2008: Toward seamless prediction calibration of climate change projections using seasonal forecasts. Bull Amer Meteor Soc, 89, 459-470.

Penduff, T., et al., 2011: Sea-level expression of intrinsic and forced ocean variabilities at interannual time scales. J Climate, 24, 5652-5670.

Rudeva, I.A., and S.K. Gulev, 2011: Composite analysis of the North Atlanticextratropical cyclones in NCEP/NCAR reanalysis. Mon. Wea. Rev., 139, 1419-1436.

Tilinina, N.D., Sergey K Gulev, Irina A Rudeva, K. Peter Koltermann, 2012: Comparing cyclone life cycle characteristics and their interannual variability in different reanalyses, J Climate, under revision.

Scaife, A. A., C. K. Folland, L. V. Alexander, A. Moberg, and J. R. Knight, 2008: European climate extremes and the North Atlantic Oscillation. J. Climate, 21, 72–83.

Skamarock, W. C., and J. B. Klemp, 2007: A time-split nonhydrostatic atmospheric model for weather research and forecasting applications. Journal of Computational Physics, 227, 7, 3465-3485.

Tolman, 2010: WAVEWATCH III (R) development best practices Ver. 0.1. NOAA/NWS/NCEP/ MMAB Technical Note 286, 19 pp.

Trenberth, K. E., and Coauthors, 2007: Observations: Surface and atmospheric climate change. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds.,CambridgeUniversityPress, 235–336.

Trigo, I.F., 2006: Climatology and interannual variability of stormtracks in the Euro-Atlantic sector: A comparison between ERA-40 and NCEP/NCAR reanalyses. Climate Dyn., 26, 127–143.

Vrac M, and Naveau P: Stochastic downscaling of precipitation: From dry events to heavy rainfalls, Wat. Resour. Res., 43, W07402, 2007.

Wang, X. L., F. W. Zwiers, V. R. Swail, and Y. Feng, 2009: Trends and variability of storminess in the Northeast Atlanticregion, 1874–2007. Climate Dyn., 33, 1179–1195.

Zolina, O., C. Simmer, K.P. Belyaev, A. Kapala, S. K. Gulev, 2009: Improving estimates of heavy and extreme precipitation using daily records from European rain gauges. J. Hydrometorol., 10, 701-716.

Zolina, O., C. Simmer, S. K. Gulevand S. Kollet, 2010: Changing structure of European precipitation: longer wet periods leasing to stronger extremes. Geophys. Res. Lett., VOL. 37, L06704, doi:10.1029/2010GL042468.

Zolina, O., C. Simmer, K.P. Belyaev, S.K. Gulev, K.P. Koltermann, 2012: Changes in the duration of European wet and dry spells during the last 60 years. J. Climate. doi:10.1175/JCLI-D-11-00498.1

Zuev, V.V., V. A. Semenov, E. A. Shelekhova, S. K. Gulev, and P. Koltermann, 2012: Evaluation of the Impact of Oceanic Heat Transport in the North Atlantic and Barents Sea on the Northern Hemispheric Climate. Doklady Earth Sciences, 2012, Vol. 445, Part 2, pp. 1006–1010.

© 2019