Réf. Schmocker-Fackel & Naef 2010 - A

Référence bibliographique complète

SCHMOCKER-FACKEL, P., NAEF, F. 2010. More frequent flooding? Changes in flood frequency in Switzerland since 1850. Journal of Hydrology, 381, 1-8.

Abstract: Severe flood events have occurred in many Swiss catchments in the last decade. Have flood frequencies changed over the last 150 years in Switzerland? And is the high frequency observed recently a nationwide phenomenon? To answer these questions, the authors analysed streamflow data from 83 stations with a record length of up to 105 years, complemented with data from historical floods dating back to 1850. Multiple trend analysis of the annual flood series showed only few negative trends. The number of stations with positive trends was especially high, when the period of 2001–2007 was included into the analysis. The temporal and spatial distribution of flood events with return periods larger than 10 years, and the large scale flood events of the last 150 years were analysed as well. Periods rich in floods alternated with periods poor in floods, showing large regional differences especially between northern and southern Switzerland. The second half of the 19th century was rich in floods, both in northern as well as in southern Switzerland. In southern Switzerland and the northern Grisons, flood frequency was high again between 1940 and 1960, a period poor in floods in northern Switzerland. Here flood frequencies increased again only after 1968. The recent increase in flood frequency and discharge has been most pronounced along the central and western northern flank of the Alps. Our data suggest that, since 1900, periods with many floods in northern Switzerland have corresponded to periods with few floods in southern Switzerland and vice versa. The differences also suggest that changes in large scale atmospheric circulation might be responsible for the fluctuations in flood frequency. The fluctuations in flood frequency should be considered in defining design floods for flood protection measures.

Flood frequency; Switzerland; Climate change; Trend analysis

Organismes / Contact

• Mountain Hydrology and Torrents, Swiss Federal Research Institute WSL, Switzerland - petra.schmocker-fackel@alumni.ethz.ch
• Institute for Environmental Engineering, Swiss Federal Institute of Technology Zürich, ETHZ, Switzerland

(1) - Paramètre(s) atmosphérique(s) modifié(s)
(2) - Elément(s) du milieu impacté(s)
(3) - Type(s) d'aléa impacté(s)
(3) - Sous-type(s) d'aléa

Pays / Zone
Massif / Secteur
Site(s) d'étude
Période(s) d'observation

(1) - Modifications des paramètres atmosphériques








Informations complémentaires (données utilisées, méthode, scénarios, etc.)


(2) - Effets du changement climatique sur le milieu naturel








Sensibilité du milieu à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)



(3) - Effets du changement climatique sur l'aléa



Trend analysis:
In a multiple trend analysis the percentage of stations with significant negative and positive trend (p 6 0.05) was determined for different time periods and plotted against the beginning and ending years of each period. Only few negative trends could be found, a slight increase could be detected for the period 1936–1965 with a maximum of 16%. Contrary to this, up to 42% of the stations show a significant positive trend for periods starting before 1966 and ending after 2000. Especially high values were obtained when the two last periods of 2001–2005 and 2006–2007 were included as well. There are only few positive trends for periods starting 1966 or later or for time series ending before the 1980s. These results suggest a decrease in floods till the 1960s and a subsequent increase.

Changes in observed flood frequency:
An overview in time and space over the large floods in the last 100 years in Switzerland is given in the paper. It shows for 51 long times series (starting before 1935) the years in which flood damage or discharges exceeding HQ10 have occurred, sorted according to region. The three largest events (Top 3 events with rank 6 3) of each time series are marked in red. Obviously, large floods have occurred at all times. However, there has been an accumulation of floods in the last ten years in the regions Thur, west, east and central and less pronounced also in the Jura region. In southern Switzerland, many floods occurred between 1987 and 2000.

Prominently visible are: the large floods of August 2005, in which 22 of the 83 catchments experienced a Top 3 event; the August 2007 flood with 19 Top 3 events and the June 1910 flood with 20 catchments with Top 3 discharges or extensive flood damage. In the last decade, there have been two or more large floods in 13 of the 19 catchments in the Jura, west and central region. The Emme river at Wiler, for example, produced two Top 3 flood events and one flood larger than HQ10 since 2005. The recent accumulation of floods seems to be especially concentrated in northern Switzerland. Region South experienced the largest floods in 1954, two floods in 1987 and floods in 1993 and 2000. To visualise how the number of floods varied from 1920 to 2007, the HQ10 floods were summed up in each region for every year and divided by the number of stations in operation. Then an 11-year running mean was calculated. The central and Thur regions, as well as the Jura and west regions were merged. In the Thur-central region, few floods occurred over nearly four decades between 1940 and 1975. Since 1975 flood frequency has increased, with a sharp rise in the 1990s. The south region is behaving differently. It has large fluctuations, with peaks between 1950 and 1960 and between 1980 and 1990. During the past century, periods of high flood frequency in the Thur-central region corresponded to periods of low frequency in the south region and vice versa.

In the Jura-west region, floods are more evenly distributed. A peak occurred around 1940, followed by a period of fewer floods until the 1980s and a sharp recent increase. The frequencies in the small catchments, not assigned to a region, have also increased since the 1960s, but do not yet show a sharp increase.

Analysis of large scale flood events:
To investigate a complete, nationwide and long time series of large scale flood events, the authors identified floods affecting a large area and causing considerable damage on a national and regional scale in Switzerland since 1850. For each event, they determined its spatial extent according to reports on flood damage or on the measured flood discharges. They identified 40 events and could classify their spatial distribution into three types: north-east, north-west and south (NE, NW and S).

While large events like the August 2005 flood affected all of the enveloping area of the three types [shaded area in fig. 6], smaller floods affected only part of it. The 2005 and 1910 floods belong to the NE floods which affect always the central region and sometimes also the Thur, west and east regions. NW floods like 1999 and 2007 extend from western Switzerland over the Jura mountains and the Swiss plateau to the Thur region, while the central region is only marginally affected and the East Region not at all. All winter floods are of this type. The S floods, like the flood of Aug 1954 and the two floods of 1987, affect southern Switzerland (upper Valais and Ticino) and often the Grisons and Uri. During the large S events in 1888 and 1897, even the Thur region was affected. The only known flood, which affected practically all regions in Switzerland, although with limited damage, occurred in August 1978.

All S floods, for which we have meteorological information, occurred due to flow over the Alps (wind direction at 500 hPa) from SW or S, while all NW floods occurred due to westerly flow (SW to NW). The NE floods occurred due to flow from NE or changing wind directions, also including the VB weather conditions (VanBebber, 1891) responsible for many large floods in Switzerland, Austria, the Czech Republic and Poland. In the Jura and south regions, only one flood type has occurred, the NW and S floods respectively, while in the Thur and Central regions all three flood types can be observed. Of the 40 known events, 17 are S floods, 14 NW floods and only 8 NE floods. S floods accumulated between 1888 and 1900, 1920 and 1960 and again between 1987 and 2000. Between 1888 and 1900, the east and central regions were frequently affected by S floods and the Thur region sometimes. During the two latter periods the Thur region was not affected and the central region only sometimes. Between 1987 and 2000, no S floods occurred at all in the east region, they were limited to the south region. It seems that frequency and extent of precipitation overlapping to the north of the Alps has decreased since the 19th century.

In northern Switzerland, large flood events accumulated between 1874 and 1881 and after 1968. Especially in the Thur and central regions, few large floods happened in-between. NE floods have occurred relatively regularly over the whole time series but overlapping of Type S floods to the two regions in northern Switzerland was less frequent and only four smaller NW floods occurred in this period, all during winter time and all restricted to the Jura and west regions. These results suggest that the changes in flood frequency in Switzerland are due to changes in atmospheric circulation.

Between 1999 and 2007, four large scale floods occurred in northern Switzerland. Three of them belong to the 4 largest events since 1900. However, a similar accumulation of large scale floods occurred in the second half of the 19th century.





Paramètre de l'aléa
Sensibilité des paramètres de l'aléa à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)

(Large) flood frequency

Regions used in the study and weather conditions:
Usually, Switzerland is divided into the following regions (Fig. 1, upper left): (1) the Jura Mountains along the north-western border to France, (2) the Swiss plateau (hilly area between the Jura and the Alps, (3) the Alps, a high mountain range running from east to west and (4) southern Switzerland. The Alps can be further subdivided into the northern flank, the inneralpine Valleys in Valais and Grisons and the southern flank (Ticino and southern valley of Grisons). Climatic variations are large between these regions. North of the Alps, a temperate middle European climate dominates, while the climate is Mediterranean south of the Alps. Often the Alps divide northern polar and southern tropical air masses. In addition, the topography directly affects precipitation through orographic uplift (Grebner, 1996). Therefore, even between neighbouring valleys or between valley and mountain ridge large differences in temperature and rainfall occur. These differences required a further differentiation of the Swiss Plateau and the northern flank of the Alps in west–east direction and of Grisons in north–south direction.

In the regions central, east and south, large floods occur mainly between June and September. In the Thur region floods can occur in every season, although most frequently in early summer. In the Jura region, floods in winter and spring time are especially of importance (Pfaundler, 2001). Although flood seasonality was not used for the region building, the regions reflect the large seasonal differences occurring in Switzerland. Three different types of synoptic circulation lead to extraordinary large scale precipitation events in Switzerland (Grebner, 1993). In the ‘‘north alpine” type, the centre of the precipitation event is located north of the Alps. It can occur under different weather conditions and flow directions over the Alps. The ‘‘south alpine” type is caused by southerly flow only and the centre of the precipitation event lies south of the Alps. Precipitation intensities and amounts are much higher than on the north side of the Alps and the precipitation is often intensified by convective precipitation cells (Grebner, 1993). Sometimes, the south alpine precipitation field is overlapping to the north side of Alps (overlapping type), causing floods in the Grisons, the Valais, Uri and sometimes even in central and north-eastern Switzerland.

To identify temporal fluctuations in the annual flood series of Switzerland, the authors used a multiple trend analysis (McCabe and Wolock, 2002). To investigate spatial differences, they examined the temporal distribution of floods with return periods of more than 10 years in six hydro-climatic Swiss regions in more detail. Besides analysing events in single catchments, they identified the temporal and spatial distribution of the large scale flood events in Switzerland since 1850, which hit one or several regions, and compared them with atmospheric flow conditions.

Study site and data:
The annual maximum flood series (FOEN, 2007c) of 83 stations in small- to medium-sized catchments (0.5–3350 km²) with record length of 30–105 years were used for this investigation. On rivers with several stations, only the station with the longest record was used, a second station was only considered, if the catchment area of the downstream station was at least twice the area of the upstream station. Most stations started operation in the 1930s, some only in the 1960s. To extend the time series back to 1900, we reconstructed the flood series, using the WSL historical flooddamage database and other historical sources (Röthlisberger, 1991; Hegg et al., 2000). As no exact discharge values could be given, the historical floods were rated as medium, large or extreme. Although the record of the historical floods is not complete, the majority of the large floods is contained in it. [...] The 10 year flood (HQ10) was estimated using the whole measurement period. For inhomogeneous series, the HQ10 values were estimated for the periods before and after dam construction or shifting of the station. In the paper, a flood event is defined as discharge larger than HQ10 or as an event that is mentioned in the historical records.

Trend test:
Of the 67 homogeneous annual flood series, four flood series are significantly (p 6 0.05) lag-one serially correlated with correlation coefficients r(1) between 0.25 and 0.44. They were excluded from the trend analysis. Trend tests were performed with the distribution free Mann–Kendall trend test (Mann, 1945 and Kendall, 1975). This test identifies monotone rising and falling trends and can deal with missing data in a time series. Multiple trend tests were conducted as suggested by McCabe and Wolock (2002) by varying the beginning and ending date of the time series in 5 year steps for the period 1931–2007. All periods with a minimum record length of 10 years were considered.

Regions used in the study:
The authors adopted the hydro-climatic regions of Switzerland as defined in several Swiss climatologic studies (e.g. Kirchhofer, 2000; Mäder, 1990; Schüepp, 1979). [...] Catchments which were lying in two hydro-climatic regions or close to the boarder of a region were assigned to the region to which they correlated best. In addition, the number of regions could be reduced to six since high correlations existed between neighbouring regions. As stations located in the Jura correlated high with those on the western Swiss plateau, these two regions were merged to the Jura Region (R1). The Jura (R1) and West Region (R2), which covers the western part of the northern flank of the Alps, were also merged for part of the regional analysis. Sometimes they were kept apart to allow a better geographical orientation. The same procedure was selected for the Thur Region (R3) and the Central Region (R4). The Thur region (R3) contains the stations on the eastern Swiss Plateau and along the eastern northern flank of the Alps. The Central Region (R4) includes stations on the western and central northern flank of the Alps. The East Region (R5) is a small region in north Grisons. The South Region (R6) is the largest region, including stations in upper Valais, northern Ticino and the central and southern Grisons, as well as three stations just north of the main ridge of the Alps (Reuss, Linth and Seez). The large extend of Region R6 is justified by a study of Schaub et al. (1990). They found that all large floods in these regions were caused by precipitation events south of the Alps, which were overlapping to the north causing floods in the Valais, the Ticino, central Grisons and the upper part of the Reuss valley. So we ended up with five regions in northern Switzerland and one region containing the inner alpine valleys and areas as well as southern Switzerland. Sixteen of the 33 catchments smaller than 80 km2 showed no correlation with any other catchment or only with a neighbouring small catchment. In these small, fast reacting catchments, floods are mainly caused by local, convective rainfall events. They were therefore excluded from the regional analysis.

Identification of large scale floods:
Besides analysing events in single catchments, we also identi- fied the temporal and spatial distribution of large scale flood events in Switzerland since 1850. Such events affected one or several regions, causing large damage. We then compared them with atmospheric flow conditions. A flood event was defined as large scale flood, when more than 10% of all stations experienced a flood larger than HQ10. Additionally, and for the period 1850–1900 exclusively, large scale flood events were determined from the WSL flood-damage database (Röthlisberger, 1991). For all events after 1945, the wind direction over the Alps at the 500 hPa pressure level could be determined from the Alpine weather statistics (MeteoSchweiz, 1985 and Schwarb, 1996). For events before 1945, the wind direction was estimated using reports of MeteoSwiss (Witterungsberichte, 1911–2007; Annalen, 1864–2007) whenever possible.

(4) - Remarques générales

In August 2005, an extreme flood occurred in Switzerland, killing six people and causing over 1.9 billion Euro of damage. Almost the whole northern flank of the Swiss Alps was affected and the largest peak discharges ever were measured in 32 catchments (FOEN, 2007a). The Lütschine in the Bernese Oberland and several other rivers experienced the third large flood event with a return period of about 100 years within a decade. In August 2007 another large flood occurred, causing over 400 million Euro of damage and again in several catchments largest peak discharges ever were measured. Is such an accumulation of floods unique in our flood records and is it only a local or a nationwide phenomenon? Spreafico and Stadler (1988) investigated trends in flood discharges in Switzerland up to 1984 and found no significant trends in the majority of catchments. Trends existed mainly in rivers which where influenced by newly constructed hydro-electric power plants. Brisan et al. (2005) investigated trends in seasonal runoff in Switzerland and found an increase in winter maximum streamflow in recent years.

It is expected that global warming will influence the precipitation regime (IPCC, 2008). In the last decade, several studies have investigated changes and trends in streamflow to test whether such changes can already be seen in measured time series. They arrived at different conclusions. Kundzewicz et al. (2005) detected upward and downward trends in annual maximum flow series in a worldwide dataset, but most of the series showed no significant trend. Svensson et al. (2005) found no general patter of increasing or decreasing numbers of floods or their magnitudes in 21 stations worldwide, while Milly et al. (2002) detected a significant increase worldwide in the monthly mean flood discharge of very large basins.

The investigations of Lindström and Bergström (2004) in Sweden and Black and Burns (2002) in Scotland suggest that periods with few floods alternated with periods with frequent floods during the last century. Studies from Benito et al. (2003), Brázdil et al. (2006), Camuffo and Enzi (1996), Glaser and Stangl (2004), Mudelsee et al. (2003) and Pfister (1998), which considered also historical floods, found large fluctuations in flood frequencies during past centuries or even millennium as well. As periods with few and frequent floods seem to alternate, the results of trend calculations depend on the time period used. This requires a systematic approach to detect trends (Burn and Hag Elnur, 2002; Dixon et al., 2006; Lins and Slack, 1999; McCabe and Wolock, 2002). And studies covering different climatic regions found spatial differences in the occurrence and direction of trends (e.g. Burn and Hag Elnur, 2002; Burn et al., 2008; Kahaya and Kalayci, 2004; Lins and Slack, 1999; Zhang et al. 2001).

(5) - Syntèses et préconisations

Discussion and conclusions:
In Switzerland, periods with frequent floods have alternated with quieter periods during the last 150 years. In northern Switzerland, numerous floods were recorded between 1874 and 1881 and since 1968, while few floods have occurred in-between, especially along the northern flank of the Alps and in the Thur region. The multiple trend analysis shows a large increase in number of stations with significant upward trend, when the last decade is included in the analysis. Since around 1900, three of the four largest large scale flood events in northern Switzerland (floods 1999, 2005 and 2007) have all occurred within the last ten years. However, a similar accumulation of large floods has already been observed in the second half of the 19th century.

In regions south and east an accumulation of floods occurred in the second half of the 19th century and between 1920 and 1960, a period that was quiet in northern Switzerland. The largest flood event in both regions is the flood of 1954. There is no marked recent increase in flood frequency in the south region, however, several large flood events have occurred between 1987 and 2000. Since 1900, flood rich periods in northern Switzerland corresponded to quiet periods in southern Switzerland and vice versa. The recent increase in flood frequency and flood discharge has been especially severe along the central and western northern flank of the Alps (regions west and central).

Studies about changes in precipitation frequencies in Switzerland come to similar conclusions. Bader and Bantle (2004) investigated precipitation events with more than 70 mm in 48 h between 1864 and 2001. In northern Switzerland, periods with a high frequency of such events alternated with periods of low frequencies. One period of high frequencies was in the second half of the 19th century, followed by a relatively low period between the beginning of the 20th century and 1975. Since the late 1970s, frequencies are increasing again. Courvoisier (1998) also found a strong increase of precipitation events (>70 mm in 24 h, 1901–1996) in northern Switzerland since 1970. This is concurrent with the present findings about observed flood frequencies in northern Switzerland, although the precipitation amount required to produce a large flood is much larger. In August 2005, for example, precipitation amounted to 130 mm in 24 h and 220 mm in 48 h for large parts of the northern flank of the Alps, with some stations recording up to 240 mm in 24 h (FOEN, 2007a). No study about frequency changes of such extreme precipitation events has yet been published.

In southern Switzerland, the frequencies of precipitation events (70 mm in 24 h) also oscillated. A peak was reached in the 1950s and no recent increase has been observed since. This is concurrent with the high number of flood events in region south between 1940 and 1960. For a threshold of 100 mm precipitation in 24 h, an increase in precipitation events since the late 1970s in southern Switzerland is found (Bader and Bantle, 2004). Although our study has shown that the number of small floods (>HQ10) has not further increased in southern Switzerland in the last two decades, four catastrophic floods (two in 1987, 1993 and 2000) with exceptional high peak discharges (top 3 events) have occurred. As rainfall amounts during flood events in southern Switzerland are even higher than in northern Switzerland, amounting to over 300 mm in 24 h, the thresholds used in the precipitation investigations are rather small compared to the flood relevant precipitation.

No change in the seasonal distribution of precipitation events (>70% occurs in summer) and of the seasonal flood distribution could be observed. However, the observed increase in temperature since the late 1970s in Switzerland might also lead to chances in precipitation (Bader and Bantle, 2004 and Schmidli and Frei, 2005).

Since 1850, 40 large scale flood events have occurred in Switzerland and could be classified into the three types NW, NE and S. All S floods occurred during southerly flow (flow direction over the Alps at 500 hPa pressure level), the NW floods during westerly flow and the NE floods during north-easterly flow or changing flow directions including the VB flood events of 1910, 2002 and 2005. The NE floods are relatively evenly distributed over time. NW floods have occurred less often in the first seven decades of the 20th century and did not affect the northern slope of the Alps during this period. Overlapping of precipitation from south of the Alps to northern Switzerland was also less frequent. This might explain the low frequency of large scale flood events along the northern slope of the Alps and in the Thur region between 1900 and 1968.

The present data suggest that changes in atmospheric circulation might be responsible for the changes in flood frequency in the different Swiss regions. (Frei et al., 2001) also suggest an interrelationship between changes in extreme precipitation in Switzerland and large scale atmospheric circulation. Relationships between atmospheric circulation patterns and flood frequency were also found for winter floods in central Europe (Jacobeit et al., 2003) of for floods in the upper Mississippi valley (Knox, 2000). The authors think that further research in this direction might help to better understand how climate change influences flood frequency in the different regions of Switzerland.

The large scale flood events in Switzerland were caused by large scale advective rainfall events of long duration, sometimes intensified by convective precipitation. Small catchments are more likely to produce floods during local convective rainfall events with high rainfall intensities. The flood frequencies of small catchments showed an increase with time, too. However, no recent sharp increase could be observed. The likelihood and magnitude of extreme events influences the planning of flood protection measures. Merz and Blöschl, 2008a and b identified many uncertainties connected with this approach. The present study highlights a further complication in showing that the reoccurrence probability of a specific discharge is not constant over time in Switzerland. The study of Redmond et al. (2002) suggests that this is not a local phenomenon. They note that ‘‘significant climatic variability occurs on time scales of decades to centuries and over areas the size of typical river basins and lead to changes in extreme event likelihood”. In Switzerland, the flood frequency cycles vary regionally. Especially in northern Switzerland, runoff measurements started in a period of low flood frequency. These time series, used for flood frequency analysis, lead to an underestimation of the actual frequencies. Future flood protection measures should, therefore, take into consideration the observed cyclic behaviour of flood frequency.

Références citées :

Abdi, H., 2007. Kendall rank correlation. In: Salkind, N.J. (Ed.), Encyclopedia of Measurement and Statistics. Sage, Thousand Oaks (CA).

Bader, S., Bantle, H., 2004. Das schweizer klima im trend. Temperatur – und Niederschlagsentwicklung 1864–2001. Veröffentlichung der MeteoSchweiz Nr. 68, 45p.

Benito, G., Díez-Herrero, A., Fernández de Villalta, M., 2003. Magnitude and frequency of flooding in the Tagus basin (central spain) over the last millennium. Climatic Change 58, 171–192.

Black, A.R., Burns, J.C., 2002. Re-assessing the flood risk in Scotland. The Science of the Total Environment (294), 169–184.

Brázdil, R., Dobrovolny, P., Kakos, V., Kotyza, O., 2006. Historical and recent floods in the Czech Rebublic: causes, seasonality, trends, impacts. In: Schanz, J. et al. (Eds.), Flood risk Management Hazards, Vulnerability and Mitigation Measures. Springer, pp. 247–259 (Chapter 20).

Brisan, M.V., Molnar, P., Burlando, P., Pfaundler, M., 2005. Streamflow trends in Switzerland. Journal of Hydrology 314, 321–329. [Fiche biblio]

Burn, D.H., Fan, L., Bell, G., 2008. Identification and quantification of streamflow trends on the canadian prairies. Hydrological Sciences Journal 53 (3), 538–548.

Burn, D.H., Hag Elnur, M.A., 2002. Detection of hydrologic trends and variability. Journal of Hydrology. 255, 107–122.

Camuffo, D., Enzi, S., 1996. The analysis of two bi-millennial series: Tiber and Po river floods. In: Jones, P.D., Bradley, R.S., Jourzel, J. (Eds.), Climate Variations and Forcing Mechanisms of the Last 2000 Years, NATO ASI series, vol. 141. Springer- Verlag, Berlin, pp. 433–450.

Courvoisier, H.W., 1998. Statistik der 24 – stündigen Starkniederschläge in der Schweiz 1901–1996. Arbeitsberichte der Schweizerischen Meteorlogischen Anstalt. MeteoSchweiz.

Dixon, H., Lawler, D.M., Shamseldin, A.Y., 2006. Streamflow trends in western Britain. Geophysical Research Letters (33), L19406.

FOEN – bundesamt für umwelt, 2007a. Ereignisanalyse hochwasser 2005, Teil 1 – Prozesse, Schäden und erste Einordnung. Bern, 215 pp.

FOEN – Bundesamt für umwelt, 2007c. Hochwasserwahrscheinlichkeiten (Jahreshochwasser). Schweizerische Eidgenossenschaft, Bern.

Frei, C., Davis, H.C., Gurtz, J., Schär, C., 2001. Climate dynamics and extreme precipitation and flood events in central Europe. Intergrated Assessment 1, 281–299.

Glaser, R., Stangl, H., 2004. Climate and floods in central Europe since AD 1000: data, methods, results and consequences. Surveys in Geophysics 25, 485–510.

Grebner, D., 1996. Räumlich – zeitliche verteilung von niederschlägen. In: Instabile Hänge und Andere Risikorelevante Natürliche Prozesse. Birkhäuser Verlag Base, Monte Verità, pp. 141–161.

Grebner, D., 1993. Synoptische zirkulationen während extremniederschlägen in der nordalpinen Schweiz. Aktuelle Aspekte in der Hydrologie, Zürcher Geographische Schriften, Heft 5e, pp. 39–48.

Hegg, Ch., Gerber, D., Röthlisberger, G., 2000. Unwetterschaden–datenbank der Schweiz. Interpraevent 2000, Villach, Band 1, pp. 37–48.

IPCC, 2008. Climate change and water. Intergovernmental panel on climate change, IPCC technical paper VI, 200pp.

Jacobeit, J., Wanner, H., Luterbacher, J., Beck, C., Philipp, A., Sturm, K., 2003. Atmospheric circulation variability in the north – Atlantic-European area since the mid-seventeenth century. Climate Dynamics 20, 341–352.

Kahya, E., Kalayci, S., 2004. Trend analysis of streamflow in Turkey. Journal of Hydrology. 289, 128–144.

Kendall, M.G., Babington Smith, B., 1939. The problem of m rankings. The Annals of Mathematical Statistics 10 (3), 275–287.

Kendall, M.G., 1975. Rank correlation methods. Charles Griffin, London.

Kirchhofer, 2000. Climatological atlas of Switzerland. Meteoswiss. Federal Office of Topography swisstopo.

Knox, J.C., 2000. Sensitivity of modern and Holocene floods to climate change, Quaternary Science Reviews. 19, 439–457.

Kundzewicz, Z.W., Graczyk, D., Maurer, T., Pinskwar, I., Radziejewski, M., Svensson, C., Szwed, M., 2005. Trend detection in river flow series: 1. Annual maximum flow. Hydrological Sciences Journal 50 (5), 797–810.

Lindström, G., Bergström, S., 2004. Runoff trends in Sweden 1807–2002. Hydrological Sciences Journal 49 (1), 69–83.

Lins, H.F., Slack, J.R., 1999. Streamflow trends in the United States. Geophysical Research Letters 26 (2), 227–230.

Mann, H.B., 1945. Nonparametric tests against trend. Economietrica 13, 71–84.

McCabe, G.J., Wolock, D.M., 2002. A Stepp increase in streamflow in the conterminous United States. Geophysical Research Letters 29 (24). doi:10.1029/2002GL0159999. Mäder, F., 1990. Klima und Wetter IV. In: Spiess, E. (Ed.), Atlas der Schweiz, 13a.

Merz, R., Blöschl, G., 2008a. Flood frequency hydrology: 1. Temporal, spatial, and causal expansion of information.. Water Resources Research 44, W08432. doi:10.1029/2007WR006744.

Merz, R., Blöschl, G., 2008b. Flood frequency hydrology: 2 Combining data evidence. Water Resources Research 44, W08432. doi:10.1029/2007WR006745.

MeteoSchweiz. 1911–2007. Witterungsbericht der Meteoschweiz. Schweizerische Meteorologische Anstalt, Zürich.

MeteoSchweiz, 1864–2007. Annalen der meteoschweiz. Schweizerische Meteorologische Anstalt, Zürich. MeteoSchweiz, 1985. Alpenwetterstatistik. Beschreibung der Einzelnen Parameter. Abteilung Forschung. Unpublished.

Milly, P.C.D., Wetherald, R.T., Dunne, K.A., Delworth, T.L., 2002. Increasing risk of great floods in a changing climate. Nature 415, 514–517. [Fiche biblio]

Mudelsee, M., Börngen, M., Tetzlaff, G., Grünewald, W., 2003. No upward trends in the occurrence of extreme floods in central Europe. Nature 425, 166–169.

Pfaundler, M.U., 2001. Adapting, analyzing and evaluation a flexible index flood regionalization approach for heterogeneous geographical environments. Diss. ETH No. 14253. Zürich, ETH Zürich, 189pp.

Pfister, Ch., 1998. Raum–zeitliche rekonstruktion von witterungsanomalien und naturkatastrophen 1496–1995. Schlussbericht NFP 31. Vdf , Zürich.

Redmond, K.T., Enzel, Y., House, P.K., Biondi, F., 2002. Climate variability and flood frequency at decadal to millennial time scales. In: House, P.K., Webb, R.H., Baker, V.R., Levish,D.R. (Eds.), Ancient FloodsModernHazards: Principles andApplications Of Paleoflood Hydrology.Water Science and Application, vol. 5. pp. 21–54.

Röthlisberger, G., 1991. Chronik der Unwetterschäden in der Schweiz. WSL Bericht 330.

Schaub, D., Horat, P., Naef, F., 1990. Die Hochwasser der reuss im 18. und 19. Jahrhundert und ihr einfluss auf die hochwasserstatistik. Wasser, Energie, Luft 3/4.

Schmidli, J., Frei, Ch., 2005. Trends of heavy precipitation and wet and dry spells in Switzerland during the 20th century. International Journal of Climatology. 25, 753–771. [Fiche biblio]

Schüepp M, 1979. Witterungsklimatologie. klimatologie der schweiz, band III. Beiheft zu den annalen der schweizerischen meteorologischen anstalt (Jahrgang 1978). Schweizerische Meteorologische Anstalt, Zürich.

Schwarb, M., 1996. Witterungsanalyse anhand der alpenwetterstatistik mit bezug auf den niederschlag, Periode 1945–1994. Berichte und skripten Nr. 58, Geographisches Institut ETH Zürich.

Spreafico, M., Stadler, K., 1986. Hochwasserabflüsse in schweizerischen Gewässern., vol. I. Mitteilungen Nr. 7, Landeshydrologie und Geologie, Bern.

Spreafico, M., Stadler, K., 1988. Hochwasserabflüsse in schweizerischen Gewässern., vol. II, Mitteilungen 8, Landeshydrologie und Geologie, Bern.

Svensson, C., Kundzewicz, Z.W., Maurer, T., 2005. Trend detection in river flow series: 2 Flood and low – flow index series. Hydrological Sciences Journal 50 (5), 811–824.

VanBebber, W.J., 1891. Die wettervorhersage: eine praktische anleitung zur wettervorhersage auf grundlage der zeitungswetterkarten und zeitungswetterberichte: für alle Berufsarten. Im Auftrag der Direktion der Deutschen Seewarte, Enke, Stuttgart, 171pp.

Zhang, X., Harvey, K.D., Hogg, W.D., Yuzyk, T.R., 2001. Trends in canadian streamflow. Water Resources Research. 37 (4), 987–998.