Circulation Type Classification Research Articles

Circulation-Type Classification over Mexico by the Era-Interim Database and Cost733

In this paper an approach of a synoptic classification by cluster k-means (CKM) and the European Gro&#223wettertypes (GWT) was performed and executed in COST733 package. The methodology used a CKM with nine clusters and GWT with 16. The COST733 evaluated a dataset of 30-years since 1986 to 2015. The variables selected were mean sea level pressure (mslp), geopotential height (z500 and z850), wind speed and direction (u10, v10 and u, v at 850 hPa) and relative vorticity (vo) with a 0.75° × 0.75° resolution of the data grid at 00:00, 06:00, 12:00 and 18:00 UTC and 0 steps. These results were evaluated using COST733 to find the quality of measurements by the explained variance (EV) or reduction ratio in error and pseudo F value (PF) to determine the certainty of the results. GWT-16 showed better yearly values in the evaluation with 32.7 (EV) and 354.3 (PF) against the CKM-9 of 54.2 (EV) and 1621.8 (PF). Finally, it was concluded that GWT-16 could be used for classification of synoptic systems over Mexico and the analysis of meteorological phenomena triggers on increases or decreases of atmospheric pollution in areas over Mexico.

Circulation type classifications for temperature and precipitation stratification in Italy

ABSTRACTDuring the last few decades, weather circulation type classifications (CTCs) have been widely used to gain insight into processes at the synoptic scale, but also for studying the relationship between atmospheric circulation and surface climate variability. This study aims to evaluate the best performing CTCs based on COST733 software for the stratification of daily ground‐level precipitation and surface air temperature across Italy by means of four statistical metrics. Six classification methods belonging to the four COST733 groups (threshold‐based, PCA‐based, leader algorithms and optimization algorithms) were investigated on 32 and 26 data time series derived from Italian weather stations for daily mean temperature and daily mean precipitation, respectively. CTCs were computed using gridded mean sea level pressure and geopotential height at 500 hPa derived from the NCEP Reanalysis 2 dataset between 1979 and 2015 and tested on three different numbers of classes (8/9, 11/12 and 18 circulation types). Evaluation metrics showed an evident seasonal variability and high‐spatial heterogeneity reflecting the geographical complexity of the Italian territory. The study points out that the best classification, both for temperature and precipitation, is strongly dependent on the classification variable (mean sea level pressure and geopotential height at 500 hPa) showing relevant differences between surface temperature and precipitation. A low number of circulation types (8/9) resulted as the most appropriate grouping for the Italian domain and the Principal Component Transversal and Simulated Annealing were the best performing classification procedures for ground‐level precipitation and temperature stratification, respectively.

Variability of UTCI index in South Warsaw depending on atmospheric circulation

The aim of the study is the assessment and analysis of the variability of human-bioclimatic conditions in southern part of Warsaw, depending on the atmospheric circulation in the long-term (1998–2015). The Universal Thermal Climate Index UTCI was applied to evaluate the human-bioclimatic conditions using meteorological data for the 12:00 UTC obtained from the WULS Ursynów weather station. In order to determine the relationship between the UTCI and its corresponding synoptic situation, the circulation type classification proposed by Lityński was used. It was found that the period from the second decade of April to the first decade of October is the most favourable one for recreation and relaxation. The situations not caused by thermal loads of humans are the most frequent and the values vary from 42.9% in October to 66.6% in August. Whereas weather conditions causing the ‘strong’ and ‘very strong heat stresses’ and the ‘strong’ and ‘very strong cold stresses’ were relatively rare. The analysis of influence of atmospheric circulation on bioclimatic conditions has shown that the most favourable conditions with no heat load (class 0) of the body occurred during cyclonic circulation in the summer (70%) mainly at advection from the North and South. Particular analysis of selected most strenuous days of the heat wave and the cold wave confirmed the occurrence of a high frequency of conditions stressing the human body with ‘strong’ and ‘very strong heat’ as well as ‘strong cold’ for most of the day during anticyclonic circulation in both cases.

The influence of atmospheric circulation on the spatial diversity of air temperature in the area of Forlandsundet (NW Spitsbergen) during 2010–2013

ABSTRACTThe relationship between atmospheric circulation and climate in Svalbard has been described in dozens of studies. However, the data used for that purpose usually came from permanent stations on the coast. The influence of atmospheric circulation on topoclimatic diversity has not been explored so often, and hardly at all for other periods than the Arctic summer. In this article, the relationships between circulation and air temperature are described using daily data sourced from six sites located around Forlandsundet (NW Spitsbergen) during 2010–2013. The analysis was conducted independently for three seasons identified as: winter (Nov–Mar), spring/autumn (Apr–May and Sep–Oct) and summer (Jun–Aug) and also for three air temperature parameters: diurnal mean (Ti), maximum (Tmax) and minimum (Tmin) temperature. The atmospheric circulation in the studied area was described using Tadeusz Niedźwiedź's classification of diurnal circulation types for Svalbard. The influence of atmospheric circulation on the spatial pattern of air temperature is not uniform across the Forlandsundet region; in particular, important differences were observed between coastal and inland parts of the study area. Thus, generalization of relationships between air temperature and atmospheric circulation for the entire area of Spitsbergen based on data only from coastal stations is not appropriate. The influence of atmospheric circulation on the spatial pattern of air temperature in the Forlandsundet region also changes through the year. In the cold season (Sep–May) it differs significantly from that observed in summer (Jun–Aug), and this feature is also seen in analyses of the 10% highest (≥ 90th percentile) and lowest (≤ 10th percentile) thermal differences. In summer, the influence of atmospheric circulation on air temperature in the topoclimatic scale is definitely less stable than in the cold season.

The exceptionally wet year of 2014 over Greece: a statistical and synoptical-atmospheric analysis over the region of Thessaloniki

The highest rainfall totals (912.2 mm) and the largest number of raindays (133 days), since 1958, were recorded in Thessaloniki during the year of 2014. Extreme precipitation heights were also observed on a seasonal, monthly and daily basis. The examined year presented the highest daily rainfall intensity, the maximum daily precipitation and the largest number of heavy precipitation days (greater than 10 mm), and it also exceeded the previous amounts of precipitation of very wet (95th percentile) and extremely wet (99th percentile) days. According to the automatic circulation type classification scheme that was used, it was found that during this exceptionally wet year, the frequency of occurrence of cyclonic types at the near surface geopotential level increases, while the same types decreased at a higher atmospheric level (500 hPa). The prevailing type was type C which is located at the centre of the study area (Greece), but several other cyclonic types changed during this year not only their frequency but also their percentage of rainfall as well as their daily precipitation intensity. It should be highlighted that these findings differentiated on the seasonal-scale analysis. Moreover, out of the three teleconnection patterns that were examined (Scandinavian Pattern, Eastern Mediterranean Teleconnection Pattern and North Sea-Caspian Pattern), the Scandinavian one (SCAND) was detected during the most of the months of 2014 meaning that it was highly associated with intense precipitation over Greece.

Relative impacts of land use and climate change on summer precipitation in the Netherlands

Abstract. The effects of historic and future land use on precipitation in the Netherlands are investigated on 18 summer days with similar meteorological conditions. The days are selected with a circulation type classification and a clustering procedure to obtain a homogenous set of days that is expected to favor land impacts. Changes in precipitation are investigated in relation to the present-day climate and land use, and from the perspective of future climate and land use. To that end, the weather research and forecasting (WRF) model is used with land use maps for 1900, 2000, and 2040. In addition, a temperature perturbation of +1 °C assuming constant relative humidity is imposed as a surrogate climate change scenario. Decreases in precipitation of, respectively, 3–5 and 2–5 % are simulated following conversion of historic to present, and present to future, land use. The temperature perturbation under present land use conditions increases precipitation amounts by on average 7–8 % and amplifies precipitation intensity. However, when also considering future land use, the increase is reduced to 2–6 % on average, and no intensification of extreme precipitation is simulated. In all, the simulated effects of land use changes on precipitation in summer are smaller than the effects of climate change, but are not negligible.

High-Resolution Dynamical Downscaling of ERA-Interim Using the WRF Regional Climate Model for the Area of Poland. Part 2: Model Performance with Respect to Automatically Derived Circulation Types

This paper presents the application of the high-resolution WRF model data for the automatic classification of the atmospheric circulation types and the evaluation of the model results for daily rainfall and air temperatures. The WRF model evaluation is performed by comparison with measurements and gridded data (E-OBS). The study is focused on the area of Poland and covers the 1981–2010 period, for which the WRF model has been run using three nested domains with spatial resolution of 45 km × 45 km, 15 km × 15 km and 5 km × 5 km. For the model evaluation, we have used the data from the innermost domain, and data from the second domain were used for circulation typology. According to the circulation type analysis, the anticyclonic types (AAD and AAW) are the most frequent. The WRF model is able to reproduce the daily air temperatures and the error statistics are better, compared with the interpolation-based gridded dataset. The high-resolution WRF model shows a higher spatial variability of both air temperature and rainfall, compared with the E-OBS dataset. For the rainfall, the WRF model, in general, overestimates the measured values. The model performance shows a seasonal pattern and is also dependent on the atmospheric circulation type, especially for daily rainfall.

Influence of the Boundary Forcing on the Internal Variability of a Regional Climate Model

The internal variability of a ten-member ensemble of the regional climate model REMO over Europe is investigated. It is shown that the annual cycle of internal variability behaves differently compared to earlier studies that focused on other regions. To gain better insight into the dependence of the internal variability on the boundary forcing variability, a circulation type classification is performed on the forcing data. It can be shown that especially in the winter season internal variability is dependent on the circulation type included in the boundary forcing, whereas in the summer season the level and pattern of internal variability is rather independent from the circulation type of the driving field. It is concluded that for Europe the internal variability of REMO in winter is governed by circulation patterns related to the North-Atlantic Oscillation, whereas in summer local processes play a bigger role.

Multiyear Aerosol Study Based on Lidar&Sunphotometer Measurements in Romania

This observational study focused on three-years time-averaged data set (January 2012–2015). An investigation of long-term trends was performed on two different data sets derived from active and passive remote sensing measurements in Magurele, Romania.Measurements of sun photometer aerosol optical depth (AOD) at 500 nm and 340 nm show the mean values of 0.230 ±0 .118 and 0.398 ± 0.185, respectively. The lidar AOD at 532 and 355nm has a mean of 0.271 ±.0.164 and 0.472 ± 0.165 respectively. The highest seasonal mean value was measured by the lidar during the summer of 2014 while the lowest seasonal value was measured by the sunphotometer in February 2012.The origin of atmospheric aerosols has been analyzed using both backtajectories of Hysplit and Circulation Type Classification (CTCs) methods.

Circulation types classification for hourly precipitation events in Lublin (East Poland)

AbstractThe paper presents an objective classification of circulation types for East-Central Europe with application to hourly precipitation events in Lublin (East Poland) from May to September. The development of the classification utilized sub-daily sea-level pressure values (at the main standard synoptic hours,

Overly persistent circulation in climate models contributes to overestimated frequency and duration of heat waves and cold spells

The study examines links of summer heat waves and winter cold spells in Central Europe to atmospheric circulation and specifically its persistence in an ensemble of regional climate models (RCMs). We analyse 13 RCMs driven by the ERA-40 reanalysis and compare them against observations over reference period 1971–2000. Using objective classification of circulation types and an efficiency coefficient with a block resampling test, we identify circulation types significantly conducive to heat waves and cold spells. We show that the RCMs have a stronger tendency to group together days with very high or low temperature and tend to simulate too many heat waves and cold spells, especially those lasting 5 days and more. Circulation types conducive to heat waves in summer are characterized by anticyclonic, southerly and easterly flow, with increasing importance of warm advection during heat waves. Winter cold spells are typically associated with easterly and anticyclonic flow, and the onset of cold spells tends to be linked to northerly and cyclonic flow with cold advection. The RCMs are generally able to reproduce the links between circulation and heat waves or cold spells, including the radiation-to-advection effect for heat waves and the opposite advection-to-radiation effect for cold spells. They capture relatively well also changes of mean temperature anomalies during sequences of given circulation types, namely the tendency towards temperature increase (decrease) during those types conducive to heat waves (cold spells). Since mean lengths of all circulation supertypes are overestimated in the RCMs, we conclude that the overly persistent circulation in climate models contributes to the overestimated frequency of long heat waves and cold spells. As these biases are rather general among the examined RCMs and similar drawbacks are likely to be manifested in climate model simulations for the twenty-first century, the results also suggest that climate change scenarios for spells of days with high or low temperatures should be interpreted with a view to the models’ limitations in simulating the recent climate.

Recent summer Arctic atmospheric circulation anomalies in a historical perspective

Abstract. A significant increase in the summertime occurrence of a high pressure area over the Beaufort Sea, the Canadian Arctic Archipelago, and Greenland has been observed since the beginning of the 2000s, and particularly between 2007 and 2012. These circulation anomalies are likely partly responsible for the enhanced Greenland ice sheet melt as well as the Arctic sea ice loss observed since 2007. Therefore, it is interesting to analyse whether similar conditions might have happened since the late 19th century over the Arctic region. We have used an atmospheric circulation type classification based on daily mean sea level pressure and 500 hPa geopotential height data from five reanalysis data sets (ERA-Interim, ERA-40, NCEP/NCAR, ERA-20C, and 20CRv2) to put the recent circulation anomalies in perspective with the atmospheric circulation variability since 1871. We found that circulation conditions similar to 2007–2012 have occurred in the past, despite a higher uncertainty of the reconstructed circulation before 1940. For example, only ERA-20C shows circulation anomalies that could explain the 1920–1930 summertime Greenland warming, in contrast to 20CRv2. While the recent anomalies exceed by a factor of 2 the interannual variability of the atmospheric circulation of the Arctic region, their origin (natural variability or global warming) remains debatable.

A new circulation type classification based upon Lagrangian air trajectories

A new classification method of the large-scale circulation characteristic for a specific target area (NW Iberian Peninsula) is presented, based on the analysis of 90-h backward trajectories arriving in this area calculated with the 3-D Lagrangian particle dispersion model FLEXPART. A cluster analysis is applied to separate the backward trajectories in up to five representative air streams for each day. Specific measures are then used to characterise the distinct air streams (e.g., curvature of the trajectories, cyclonic or anticyclonic flow, moisture evolution, origin and length of the trajectories). The robustness of the presented method is demonstrated in comparison with the Eulerian Lamb weather type classification. A case study of the 2003 heatwave is discussed in terms of the new Lagrangian circulation and the Lamb weather type classifications. It is shown that the new classification method adds valuable information about the pertinent meteorological conditions, which are missing in an Eulerian approach. The new method is climatologically evaluated for the five-year time period from December 1999 to November 2004. The ability of the method to capture the inter-seasonal circulation variability in the target region is shown. Furthermore, the multi-dimensional character of the classification is shortly discussed, in particular with respect to inter-seasonal differences. Finally, the relationship between the new Lagrangian classification and the precipitation in the target area is studied.

Downscaling of monthly PM10 indices at different sites in Bavaria (Germany) based on circulation type classifications

Atmospheric circulation affects local concentrations of particulate matter with an aerodynamic diameter of 10μm or less (PM10) in different ways: Via the determination of local meteorological conditions favoring or suppressing the formation and the accumulation of PM10, and through its control on short–and long–range transport of particles and precursors. The quantitative assessment of the connections between the large–scale atmospheric circulation and local PM10 is relevant not only for the understanding of observed variations in PM10 concentrations. It is even more important for estimating the potential effects of projected future changes in large–scale atmospheric circulation on PM10. In this contribution, daily atmospheric circulation types (CTs), resulting from variants of three different classification methods, and their monthly occurrence frequencies have been utilized in three different downscaling approaches for estimating monthly indices of PM10 for the period 1980–2010 at 16 locations in Bavaria (Germany). All variants of approaches have been evaluated via a leave–one–out cross validation procedure in order to attain reliable performance ratings to detect the most suitable downscaling approaches. Results indicate that the highest performance of downscaling approaches is achieved in winter when the best performing models explain on average roughly 50% of the observed PM10 variance. From this it can be concluded that classification–based approaches are generally suitable for the downscaling of PM10, particularly during winter when PM10 concentrations in Bavaria reach maximum values. As preferable settings of the downscaling approaches, the usage of rather small spatial domains and a relatively high number of classes for circulation type classification and furthermore the utilization of multiple linear regression analyses or random forest analyses for relating CTs to PM10 have been ascertained. These findings provide the basis for further enhancements of the classification–based downscaling of monthly PM10 that will be realized in successive investigations.

Interannual drought index variations in Central Europe related to the large-scale atmospheric circulation—application and evaluation of statistical downscaling approaches based on circulation type classifications

This contribution investigates the relationship between the large-scale atmospheric circulation and interannual variations of the standardized precipitation index (SPI) in Central Europe. To this end, circulation types (CT) have been derived from a variety of circulation type classifications (CTC) applied to daily sea level pressure (SLP) data and mean circulation indices of vorticity (V), zonality (Z) and meridionality (M) have been calculated. Occurrence frequencies of CTs and circulation indices have been utilized as predictors within multiple regression models (MRM) for the estimation of gridded 3-month SPI values over Central Europe, for the period 1950 to 2010. CTC-based MRMs used in the analyses comprise variants concerning the basic method for CT classification, the number of CTs, the size and location of the spatial domain used for CTCs and the exclusive use of CT frequencies or the combined use of CT frequencies and mean circulation indices as predictors. Adequate MRM predictor combinations have been identified by applying stepwise multiple regression analyses within a resampling framework. The performance (robustness) of the resulting MRMs has been quantified based on a leave-one-out cross-validation procedure applying several skill scores. Furthermore, the relative importance of individual predictors has been estimated for each MRM. From these analyses, it can be stated that model skill is improved by (i) the consideration of vorticity characteristics within CTCs, (ii) a relatively small size of the spatial domain to which CTCs are applied and (iii) the inclusion of mean circulation indices. However, model skill exhibits distinct variations between seasons and regions. Whereas promising skill can be stated for the western and northwestern parts of the Central European domain, only unsatisfactory skill is reached in the more continental regions and particularly during summer. Thus, it can be concluded that the presented approaches feature the potential for the downscaling of Central European drought index variations from the large-scale circulation, at least for some regions. Further improvements of CTC-based approaches may be expected from the optimization of CTCs for explaining the SPI, e.g. via the inclusion of additional variables in the classification procedure.

Regional Differentiation in Probability of Ice Days Occurrence in Poland

Abstract This paper aims at recognizing spatial differentiation in probability of ice days occurrence with reference to atmospheric circulation, using regional classification of circulation types for Poland. Daily maximum air temperatures measured at nine meteorological stations were used to recognize the ice days occurrence in the period from January 1951 to March 2014. The relations between the ice days occurrence and atmospheric circulation were analysed using catalogues of circulation types compiled for nine grid boxes within Poland. Linkage between the frequency of ice days and atmospheric circulation was recognized in the period of January 2001 - March 2014. A decreasing tendency in the frequency of the occurrence of winter ice days was found. The occurrence of ice days in Poland is clearly related to atmospheric circulation. In the entire area of Poland their occurrence is favoured by eastern and south-eastern advection of air during anticyclonic conditions (types Ea nad SEa). Regional variability is also noticeable.

Do global warming-induced circulation pattern changes affect temperature and precipitation over Europe during summer?

Future climate change projections are not limited to a simple warming, but changes in precipitation and sea level pressure (SLP) are also projected. The SLP changes and the associated atmospheric circulation changes could directly mitigate or enhance potential projected changes in temperature and precipitation associated with rising temperatures. With the aim of analysing the projected circulation changes and their possible impacts on temperature and precipitation over Europe in summer [June–July–August (JJA)], we apply an automatic circulation type classification method, based on daily SLP, on general circulation model (GCM) outputs from the Coupled Model Intercomparison Project phase 5 (CMIP5) database over the historical period (1951–2005) and for climate under two future scenarios (2006–2100). We focus on summer as it is the season when changes in temperature and precipitation have the highest impact on human health and agriculture. Over the historical observed reference period (1960–1999), our results show that most of the GCMs have significant biases over Europe when compared to reanalysis data sets, both for simulating the observed circulation types and their frequencies, as well as for reproducing the intraclass means of the studied variables. The future projections suggest a decrease of circulation types favouring a low centred over the British Isles for the benefit of more anticyclonic conditions. These circulation changes mitigate the projected precipitation increase over north-western Europe in summer, but they do not significantly affect the projected temperature increase and the precipitation decrease over the Mediterranean region and eastern Europe. However, the circulation changes and the associated precipitation changes are tarnished by a high uncertainty among the GCM projections.

Atmospheric circulation influence on climatic trends in Europe: an analysis of circulation type classifications from the COST733 catalogue

ABSTRACTWe studied the influence of changes in the frequency of atmospheric circulation types (CTs) on seasonal trends of daily maximum and minimum temperature and precipitation at several European stations in the period 1961–2000. The 24 circulation classifications used were created within the COST733cat database using eight classification methods; each applied on daily sea‐level pressure fields over Europe and 11 smaller European domains in three variants with fixed numbers of types (9, 18, and 27). This allows us to study how different spatial scales of circulation and varying numbers of CTs affect the circulation–climate relationship. Significant trends in the frequency of CTs took place mainly in winter, and these clearly reflect the tendency towards positive phase of the North Atlantic Oscillation: the frequency of westerly types increased in the central latitudinal belt from the British Isles east to Ukraine, whereas the frequency of days with cyclonic circulation increased over Iceland and decreased in central Mediterranean. Seasonal temperature and precipitation trends can be only partly explained by the changing frequency of CTs, the link being the strongest in winter. In spring, summer, and autumn, the observed climatic trends are not forced by changes in the frequency of CTs but rather by changing climate within these types, which confirms the previously reported nonstationarities in the relationship between atmospheric circulation and local climate. There are marked differences among the results obtained using 24 parallel, fully comparable objective catalogues of CTs, but neither good or bad result nor any preferred classification method can be discerned. In winter and spring, small‐scale circulation influences the observed climatic trends more than large‐scale circulation on a majority of stations except for Iceland and Scandinavia. Classifications with more CTs usually explain a larger proportion of the observed climatic trends.

An analysis of the synoptic and climatological applicability of circulation type classifications for Ireland

ABSTRACTCirculation type classifications compiled as part of the COST733 Action, ‘Harmonisation and Application of Weather Type Classifications for European Regions’, were evaluated based on their ability to describe variations in surface temperature (maximum and minimum) and precipitation across the Irish landmass. In all 16 different classification schemes, representative of four general approaches in synoptic typing (leader algorithm, optimization scheme, predefined types, eigenvector analysis) were considered. Several statistical measures variously quantifying performance in arranging daily observations into clearly defined homogenous groups were employed. Based on the results it was not possible to identify a single optimum classification or general approach in synoptic typing. This is related to inconsistencies in performance with respect to the specific target variable and statistical measures used; the results were also shown to be conditional on the number of circulation types (CTs) as well as spatiotemporal dependencies in performance. However, the study did indicate that those typing schemes based on predefined thresholds (Litynski, GrossWetterTypes, Lamb Weather Type) – along with the Kruizinga and Lund classifications – were better able to resolve surface temperature. With respect to precipitation those classifications derived using some optimization procedure (simulated annealing, Self Organizing Maps, k‐means clustering) were consistently among the best‐performing schemes. In capturing the relationship between synoptic‐scale circulation and precipitation the importance of incorporating some measure of vorticity was highlighted; in contrast the inclusion of discrete directional patterns was shown to be important for resolving variations in local temperature. The classifications generally performed best for winter, reflecting the closer coupling between circulation and surface conditions during this period. Spatial patterns in the synoptic–climatological relationship were more apparent for precipitation. In this case those more westerly/south‐westerly stations open to zonal airflow exhibited a stronger response to circulation variability.

Development and comparison of circulation type classifications using the COST 733 dataset and software

ABSTRACTIn order to examine correspondence between different methods for circulation type classification, a dataset of classification catalogs for 12 different European regions has been created using a specially developed software package. Twenty‐seven basic automatic classification methods have been applied in several variants to different input datasets describing atmospheric circulation. Together with six manual classifications a total of 33 methods are available for inter‐comparison.Pattern correlation, frequency time‐series correlation and the adjusted Rand index have been used for comparison. Highly significant correspondence has been detected only for two clustering techniques while the remaining classification methods show surprisingly low similarity. A Monte‐Carlo test with 1000 classifications of randomly defined types even shows that most of the methods are not more similar among each other than any arbitrarily chosen types.The predominant dissimilarity between the methods is interpreted to be a result of a lack of inherent structures of the input data. Only simulated annealing clustering and self‐organizing maps get nearly identical results because they can optimally fit the partitioning to the outer shape of the data cloud in the phase space. Also methods based on pre‐defined types come to very different results because small changes in the definition of thresholds may lead to large differences in the partitioning.It is concluded that because of the missing inner structure of the data there is no clear statistical reason to prefer any of the examined methods. For practice in synoptic climatology this means that finding a suited classification for a certain purpose may require a broad comparison of methods. The software package cost733class for development, comparison and evaluation of classifications which was developed and used in this study is available at http://cost733.geo.uni‐augsburg.de to facilitate this task.