Search

One of the most impressive flash floods of the 20th century in France as well as in Catalunia occurred in the Eastern part of the Pyrenees on October 1940. 47 people were killed in France during this extraordinary event and more than 100 in Spain. It caused tremendous damages to buildings and, in particular, destroyed the center of the resort town of Vernet-les-Bains on the slopes of the mount Canigou. The maximum observed 24-hour accumulated rainfall reached locally 1000 mm and this remains until a record value for the French European territory. The flood has been abundantly documented by the technical State services as well as some scientists of the time and a large part of this documentation has been archived.At the light of the recent advances in flash flood studies, this data set has been unearthed and the past analyses of the event have been deeply revisited. This revealed, in particular, that the peak discharge values, on which local risk assessment studies are based, had been largely over-estimated. This led to several mis-interpretations of the processes occurred during this flood such inundation dynamics or the driving factors of the tremendous observed scour and erosion volumes. Some particular features of this event could also be confirmed or revealed: (1) an impressive wave propagation of about four million cubic meters due to the breach of a natural dam due to a massive landslide in the upper part of the Tech river which occurred during the night, affected already destroyed areas and got therefore almost unnoticed, (2) a local major amplification of the inundation induced by the breach of a railway embankment, (3) the dynamics of scour and erosion. Beyond the conclusions drawn on this specific extreme flash flood, this study illustrates that our knowledge about extreme flood events is still limited by their poor and often biased documentation. New and adapted observation and documentation strategies, based for instance on the systematic collation and analysis of opportunistic data such as amateur videos, have to be implemented to enable a real progress in the science of extreme flash floods.

International audience; The paper presents conclusions from the seminar, which was held on November 20th, 2019, during the 2019 SHF days. The purpose of the meeting was to see how human and social sciences may contribute to improve the governance of water tension situations. Among the conclusions is the importance of historical studies, of the proper territorial scale to consider, of expertise credibility improvement, together with the cultural value of water in some situations. It was also shown that water crisis should be anticipated in order to protect the low-income people.; Le papier présente la synthèse des enseignements du séminaire qui s'est tenu le 20 novembre 2019 durant les Journées 2019 de la SHF. L'objet du séminaire était d'analyser l'apport des sciences humaines et sociales pour traiter les questions de gouvernance de l'eau en situation de tension sur la ressource. Les conclusions montrent notamment l'intérêt des approches historiques, ainsi que le besoin de conforter la crédibilité de l'expertise et de travailler à la bonne échelle territoriale. Elles montrent aussi sur certains cas l'importance de la valeur culturelle de l'eau, ainsi que le besoin d'anticiper les crises, pour protéger les plus démunis.

There are concerns that recent climate change is altering the frequency and magnitude of river floods in an unprecedented way. Historical studies have identified flood-rich periods in the past half millennium in various regions of Europe. However, because of the low temporal resolution of existing datasets and the relatively low number of series, it has remained unclear whether Europe is currently in a flood-rich period from a long-term perspective. Here we analyse how recent decades compare with the flood history of Europe, using a new database composed of more than 100 high-resolution (sub-annual) historical flood series based on documentary evidence covering all major regions of Europe. We show that the past three decades were among the most flood-rich periods in Europe in the past 500 years, and that this period differs from other flood-rich periods in terms of its extent, air temperatures and flood seasonality. We identified nine flood-rich periods and associated regions. Among the periods richest in floods are 1560–1580 (western and central Europe), 1760–1800 (most of Europe), 1840–1870 (western and southern Europe) and 1990–2016 (western and central Europe). In most parts of Europe, previous flood-rich periods occurred during cooler-than-usual phases, but the current flood-rich period has been much warmer. Flood seasonality is also more pronounced in the recent period. For example, during previous flood and interflood periods, 41 per cent and 42 per cent of central European floods occurred in summer, respectively, compared with 55 per cent of floods in the recent period. The exceptional nature of the present-day flood-rich period calls for process-based tools for flood-risk assessment that capture the physical mechanisms involved, and management strategies that can incorporate the recent changes in risk., This work was supported by the ERC Advanced Grant ‘FloodChange’ project (no. 291152), the Horizon 2020 ETN ‘System Risk’ project (no. 676027), the DFG project FOR 2416, the FWF projects I 3174 and W1219-N22, the Spanish Agency of Science and FEDER/UE projects CGL2016-75475/R, CGL2017-86839-C3-1-R, CGL2016-75996-R and CTM2017-83655-C2-2-R, the ICREA Academia programme, and project CZ.02.1.01/0.0/0.0/16_019/0000797, Ministry of Education, Youth and Sports of the Czech Republic.

The history of flood management in France forms part of the country’s broader history, marked by the affirmation and development of the State. Long treated as the primary responsibility of riparians, resident communities, cities and local authorities, this responsibility shifted in the 17th and especially the 18th Centuries as Central Government gradually took control of proceedings. This assumption of responsibility went on to follow a trajectory that involved technical, legal and administrative records. A journey which, in many ways, resembles a real conquest. Primarily it was a conquest for knowledge achieved through the development of skills and specific methods, whose implementation contributed to significant transformations of natural habitats, however this without providing definitive solutions for the prevention of catastrophic floods. Secondly, it was a political conquest as, in many ways, the fight against floods was an opportunity for the State to increase its legal and administrative jurisdiction on privately- or community-owned land and rivers.

In the face of regularly occuring floods, learning from past events seems to be an obvious solution. However, effective use of historical information in prevention processes and tools often raises many theoretical and practical questions. Regarding land use planning, the creation of protective structures and even emergency response planning, the empirical approach, which consists of basing flood risk prevention on only the highest waters that have occurred in the past, is not without problems. It leads to systematic upward reassessment of the reference hazard each time a significant new flooding event occurs and introduces heterogeneity between sites with available historical data and sites with no information. It is better to consider not only a single flooding event but the entire timeline or, at least, the most notable events. This is what makes it possible to estimate a flood value associated with a given return period (10-year, 100-year or 1,000-year flood) or with its annual exceedance probability (respectively, 10− 1, 10− 2, 10− 3). One of the methodological challenges, therefore, is obtaining sufficiently detailed timelines and being able to identify the most exceptional events during a particular period. For flood risk prevention plans (PPRI), the reference hazard is that which corresponds to the highest known waters, or the 100-year hazard, if this is greater. This choice raises several questions: which knowledge base should we use? How far back in time should we look? Are past high waters and their consequences always representative of the current risk?

The may-june 1856 floods have been catastrophic on the Loire and Rhone basins and in a lower measure also on Garonne and Seine Basins. Human losses have been insignificant. On the other hand urban areas, infrastructures and also agriculture have been strongly affected. A general losses board is presented. Concerning prevention, 1856 flood initiated a large public works program to protect fluvial towns (28th mai 1858 law). This event prepared also development of local floods warning system along french rivers.

During the last two centuries, France has been affected by severe floods, which are relatively well documented. The importance of these floods depends on their spatial extent and on their temporal dynamics. The flood of May-June 1856 is the largest one, covering almost two thirds of the total surface of France. This extreme flood results from the synchronized floods over the main french rivers : Loire, Rhone Garonne and upper Seine catchment (Yonne).The main objective of this work is to localize the different sources of information in order to gather and summarize the different studies conducted at the period of the flood. Then a statistical analysis of the rain and discharge measurements will be performed to quantify the extent and the dynamics of this event. By this way, its importance will be shown in comparison with other reference events over several basins, in terms of temporal and spatial scales.

Les grandes crues de mai-juin 1856 affecterent une grande moitie sud de la France. Leur impact fut d’autant plus fort que ces evenements faisaient suite a deux decennies d’inondations majeures (1836, 1840, 1846, 1851, 1852…). La situation de crise fluviale va etre l’occasion d’une mobilisation particuliere de l’Etat que traduisent ses engagements budgetaires. L’engagement financier en reparation et renforcement des infrastructures fluviales, routieres et surtout urbaines fut, en effet, sans precedent de la part de l’administration centrale (lois du 30 juillet 1856 et du 28 mai 1858), mais aussi des collectivites locales et des riverains. Sur un plan technique et scientifique, l’evenement de 1856 marque egalement une etape importante dans l’apprehension du phenomene inondation par les ingenieurs d’Etat et des Ponts et Chaussees en particulier. La grande enquete sur les crues diligentee par le ministere des Travaux Publics des l’automne 1856 peut etre consideree de ce point de vue comme le premier retour d’experience technique sur les inondations a l’echelon national.

Interdisciplinary frameworks for studying natural hazards and their temporal trends have an important potential in data generation for risk assessment, land use planning, and therefore the sustainable management of resources. This paper focuses on the adjustments required because of the wide variety of scientific fields involved in the reconstruction and characterisation of flood events for the past 1000 years. The aim of this paper is to describe various methodological aspects of the study of flood events in their historical dimension, including the critical evaluation of old documentary and instrumental sources, flood-event classification and hydraulic modelling, and homogeneity and quality control tests. Standardized criteria for flood classification have been defined and applied to the Isère and Drac floods in France, from 1600 to 1950, and to the Ter, the Llobregat and the Segre floods, in Spain, from 1300 to 1980. The analysis on the Drac and Isère data series from 1600 to the present day showed that extraordinary and catastrophic floods were not distributed uniformly in time. However, the largest floods (general catastrophic floods) were homogeneously distributed in time within the period 1600-1900. No major flood occurred during the 20th century in these rivers. From 1300 to the present day, no homogeneous behaviour was observed for extraordinary floods in the Spanish rivers. The largest floods were uniformly distributed in time within the period 1300-1900, for the Segre and Ter rivers., Une approche interdisciplinaire sur l`étude des phénomènes climatiques et de leur variabilité temporelle présente un intérêt certain pour l`analyse des risques naturels, de l`aménagement du territoire et de la gestion durable des ressources en eau. Cet article traite des aspects méthodologiques liés à la reconstitution et l`analyse d`événements de crue sur le dernier millénaire. Il aborde les questions de la critique des sources documentaires historiques, de la classification des crues et de la modélisation hydraulique des crues anciennes, des tests statistiques sur l`homogénéité des valeurs. Une échelle standardisée de classification des crues a été définie et appliquée aux crues de l`Isère et du Drac en France, de 1600 à 1950, et du Ter, du Llobregat et du Segre en Espagne, de 1300 à 1980.L`analyse des séries chronologiques du Drac et de l`Isère, de 1600 à aujourd`hui, montre que les crues fortes à exceptionnelles ne sont pas réparties de façon homogène dans le temps. Aucune tendance n`est décelable pour les crues exceptionnelles de 1600 à 1900. Aucune crue majeure de ce type n`a été observé sur ces deux rivières au cours du XXe siècle. En Espagne, de 1300 à aujourd`hui, l`occurrence des crues n`est pas stationnaire pour les crues fortes à exceptionnelles. Les crues exceptionnelles, de 1300 à 1900, sont par contre réparties de façon homogène pour le Segre et le Ter.

This paper presents the Historisk-Ardeche project, based on the collaboration between historians, archivists, hydraulicians and hydrologists. It provided the inventory of very extreme rainfall and flood values during the two last centuries on the Ardeche catchment. A long discharge data series (1644-2000) on the St-Martin-d'Ardeche area is issued from an hydraulic modelling, with an estimate of the relative errors. A flood frequency analysis shows a good agreement between the historical approach and an hydrometeorological approach (Gradex model).

La recrudescence des inondations que connait la plaine de Grenoble entre la fin du XVI e siecle et le milieu du XIX e siecle, fait l'objet d'une mobilisation avant tout en terme de Travaux Publics. Sous la houlette de l'Etat et de ses ingenieurs l'entreprise de pacification fluviale s'organise des la seconde moitie du XVII e siecle. Le developpement du service des Ponts et Chaussees permet a la fois un encadrement plus strict des travaux et la mise en place de budgets plus reguliers. Pour autant l'endiguement ne resout pas tout. Le retour des grandes inondations amene les autorites a envisager d'autres formes de prevention tournees vers la preparation a la situation d'urgence (defense des lieux strategiques, secours aux habitants). A partir des annees 1850, la surveillance du cours d'eau permet aux demarches preventives de s'adosser a une connaissance en temps reel de la riviere (annonce des crues).

International audience; This paper presents different studies on the Durance river watershed, situated in the south part of the Alps. Our studies explore a large collection of historical archives in hydrology and hydrometry, found by a historian. This watershed is characterised by an incredible density of hydrological stations, with more than 150 stations and 4500years of hydrological records, since the beginning of the 20th century. As an example, the archives of 10 stations starting between 1900 and 1917 were very detailed, with all the gaugings, rating curves, and streamflows records. Archives also reveal the great scientific and technical involvement at that period to understand, develop streamflow measurements and estimate available water resources. Then, we present an estimation of streamflow uncertainty of historical series due to the height measurement strategy. Results show that long streamflow series are not homogeneous in time and that old streamflows series could be significantly biased during snowmelt.; Cette communication présente différents travaux que nous avons menés sur le bassin de la Durance. Ces travaux explorent l'incroyable patrimoine de données historiques en hydrologie et en hydrométrie de la Durance, où il a existé plus de 150 stations hydrométriques, avec un total de 4500 années de débits. Des séries de débits commençant au début du 20ème siècle ont pu être retrouvées, ainsi que de très nombreuses informations sur les stations hydrométriques, les techniques de jaugeages, les courbes de tarage, etc. Ce patrimoine historique met en évidence le très fort investissementdes personnels pour développer les techniques de mesure et faire progresser la connaissance des ressources en eau des bassins. Ensuite, nous présentons des travaux que nous avons menés pour caractériser une partie de l'incertitude des données anciennes, liée au mode d'échantillonnage des hauteurs. Ces travaux montrent que les longues séries ne sont généralement pas homogènes d'un point de vue de la mesure et que les séries anciennes peuvent être significativement biaisées pendant la période de fonte. Enfin, nous présentons des travaux sur la caractérisation de la variabilité hydrologiquedu bassin de la Durance au cours des 100 dernières années. Ces travaux confirment l'intérêt de ces séries séculairespour mieux connaître le fonctionnement des bassins versants.

Comme pour les crues, la mise en place et le developpement, a partir de la premiere moitie du XIXe siecle environ, d’instruments et bientot de reseaux de mesures du niveau des grands cours d’eau permet aujourd’hui de reperer assez facilement les annees a etiage severe. Ces donnees instrumentees ne sont toutefois pas sans poser probleme (lacunes spatiale et temporelles, absence de references techniques sur le mode d’observation, etc.). Il faut aller vers d’autres sources pour preciser le contexte hydrometeorologique : navigation/batellerie, agriculture/irrigation, travaux publics, moulinage, mais aussi journaux, chroniques locales, etc. Pour la periode anterieure au XIXe siecle, la collecte des donnees s’operent exclusivement a partir de ce type de documentation. Dans ce contexte les sites urbains fluviaux, les etablissements religieux controlant une portion de riviere, etc., ont un grand potentiel documentaire.

The paper describes the various features of the BDHI database (objects, functions, content). This document database provides document sheets on historical floods from various sources: technical reports from water authorities, scientific accounts (meteorology, hydrology, hydraulics...), post-disaster reports, newspapers or book extracts... It is complemented by fact sheets on flood events, which provide a summary text on significant past floods: location, date and duration, type of flood, extent, probability, adverse consequences. A search engine is provided for information search based on time (specific date or period), on location (district, basin, city) or thematic topic (document type, flood type, flood magnitude, flood impact...). We conclude by some future challenges in relation to the next cycle of the Floods Directive (2016-2022), with the inventory of past floods which had significant adverse impacts. What are the flood events that need to be integrated (new ones later than 2011 and/or previous floods that had not yet been selected)? How can the process of historical data integration be extended at a local scale, with an adequate process of validation? How to promote the use of BDHI database in relation with the development of the culture of risk?

Ce papier presente une analyse frequentielle des crues basee sur un echantillon de crues collecte sur une periode systematique et sur une periode historique. Elle est appliquee sur plusieurs petits bassins versants mediterraneens. Apres le recensement et la collecte des donnees sur les crues historiques, plusieurs modeles hydrauliques ont ete construits pour prendre en compte l'evolution geomorphologique des cours d'eau. Des courbes de tarage pour les periodes recentes et historiques ont ete construites et utilisees pour estimer les debits de crues avec leurs incertitudes. Ces incertitudes prennent en compte deux types d'erreurs: (a) une erreur aleatoire liee a la lecture de la hauteur d'eau, et (b) une erreur systematique liee a une sur ou sous estimation de la courbe de tarage. Un modele bayesien d'analyse frequentielle est developpe pour prendre en compte ces deux sources d'incertitudes. Il est montre que les incertitudes affectant les debits doivent etre prise en compte dans l'analyse frequentielle des crues car elles peuvent significativement modifier les intervalles de confiance des quantiles. Les quantiles de crues obtenus semblent concordant avec les estimations de formules empiriques pour deux des quatre bassins etudies.

Fitting statistical laws from a short time series does not give any guarantee of reliability on extreme flood estimation. Historical investigation through documentary sources can enlarge the record period. This paper presents a case study on the Ardèche river, based on collaboration between historians, archivists, hydraulic engineers and hydrologists for a better assessment of the flood risk. A list of historical flood levels from 1644 to the present has been drawn up and converted into discharge using hydraulic modelling. A sensitivity analysis provides error intervals on discharge estimates taking into account uncertainties on water level, roughness coefficient and channel geometry, and the impact of a non-permanent discharge or the backward effect. A flood frequency analysis using systematic and non-systematic data gives a practical example on how historical information can improve flood knowledge. It reduces the sampling uncertainty and shows also a good agreement with a hydrometeorological approach (Gradex model.; L'analyse probabiliste des crues extrêmes est délicate du fait de la taille réduite des chroniques de mesure. L'analyse des sources documentaires historiques permet d'élargir la fenêtre chronologique d'analyse. Nous présentons un cas d'étude sur le bassin de l'Ardèche, où une collaboration entre historiens, archivistes, hydrauliciens et hydrologues a permis de reconstituer une série de cotes de crues depuis 1644 jusqu'à aujourd'hui et de les convertir en débit par une modélisation hydraulique. Une analyse de sensibilité a été réalisée pour apprécier l'incertitude résultante, en considérant les erreurs sur le niveau des crues, l'estimation des pertes de charge hydraulique, la géométrie du lit de la rivière et l'effet des conditions aux limites (courbe de remous et effets de propagation. Une analyse fréquentielle, avec et sans données historiques, montre l'apport des informations anciennes sur la précision de l'estimation des crues rares. Une comparaison avec la méthode du Gradex montre par ailleurs une très bonne cohérence entre les deux approches.

The catastrophic floods recently occurring in Europe warn of the critical need forhydrologic data on floods over long-time scales. Palaeoflood techniques provideinformation on hydrologic variability and extreme floods over long-time intervals(100 to 10,000 yr) and may be used in combination with historical flood data (last1,000 yr) and the gauge record (last 30–50 yr). In this paper, advantages anduncertainties related to the reconstruction of palaeofloods in different geomorphologicalsettings and historical floods using different documentary sources are described.Systematic and non-systematic data can be combined in the flood frequency analysisusing different methods for the adjustment of distribution functions. Technical toolsintegrating multidisciplinary approaches (geologic, historical, hydraulic and statistical)on extreme flood risk assessment are discussed. A discussion on the potential theoreticalbases for solving the problem of dealing with non-systematic and non-stationary data ispresented. This methodology is being developed using new methodological approachesapplied to European countries as a part of a European Commission funded project (SPHERE)., This research was supported by the Spanish Committee for Science and Technology (CICYT) grant HID99-0858, FEDER Project 1FD97-2110-CO2-02, REN- 2001-1633 and by the European Commission (DG XII), through research contract number EVG1-CT-1999-00010 (Systematic, Palaeoflood and Historical data for the improvement of flood Risk Estimation, “SPHERE” Project). SPHERE Project Web page http://www.ccma.csic.es/dpts/suelos/hidro/sphere/home.html