Les différentes approches méthodologiques de reconstitution des paléo-inondations : une revue de la littérature Different methodological approaches in paleoflood reconstruction: a literature review

Résumé

Les deux dernières décennies en recherche scientifique se caractérisent par la multiplication des travaux concernant le réchauffement planétaire et les conséquences appréhendées de ce phénomène dans les divers milieux du globe. Ces travaux de recherche touchent différents champs disciplinaires dont la géomorphologie, l'hydrologie, l'écologie et bien d'autres. Dans le contexte des changements climatiques, on voit apparaître un intérêt grandissant pour l'étude des inondations anciennes ou des paléo-inondations. On tente de reconstituer la chronologie des inondations du passé notamment en regard des changements climatiques anciens ou subactuels (derniers siècles). Devant l'émergence de nombreux travaux dans ce domaine de la paléohydrologie, il nous est apparu intéressant d'examiner à travers une revue de la littérature scientifique les différentes approches méthodologiques utilisées dans l'étude des paléo-inondations. On s'attarde à décrire les principales méthodes et techniques employées dans la reconstitution des anciennes inondations, en particulier pour les environnements fluviaux (rivières, fleuves). On passe en revue les différents indicateurs biophysiques utilisés, notamment l'analyse des séquences stratigraphiques et sédimentaires, l'analyse des macrorestes, les méthodes de datation radiocarbone (¹⁴ C) et les techniques comme la thermoluminescence optique (TLO) et la dendrochronologie. Aussi, on présente les principaux modèles mathématiques utilisés dans la reconstitution des paléo-inondations, lesquels servent notamment à évaluer les variations niveaux/débits et la fréquence des inondations anciennes.

Mots-clés:

• Paléohydrologie,
• Paléo-inondations,
• Changement climatique,
• Approche multidisciplinaire,
• Revue de la littérature

Abstract

The last two decades of scientific research have been characterized by an increase in the number of studies on global warming and its impact on the earth's various environments (e.g., terrestrial and aquatic ecosystems, river systems). The research concerns different fields such as geomorphology, ecology, hydrology and many others. In the context of climatic change, there is growing interest in the study of past floods or paleofloods. Researchers are attempting to reconstruct the chronology of past floods, especially with respect to past or subrecent (past centuries) climatic changes. The work involves using different methodological approaches borrowed from various disciplines including geology, geomorphology and ecology. The reconstruction of ancient hydrological events such as paleofloods in fact requires that different methods and techniques be combined in order to retrace the chronology of events as precisely as possible using different biological and physical indicators. The earliest research in this field was conducted in the United States, more specifically in the southern and southwestern parts of the country, which are regions characterized by an arid and semi-arid climate. Over the last few years, however, there have been an increasing number of studies from various parts of the world. Much of this research is based on previous work but with new elements of interpretation, which are mainly related to the many fluvial environments and climatic patterns associated with floods.

A wide variety of indicators are used in the chronological reconstruction of ancient fluvial environments, whether in humid, sub-humid or desert regions. These indicators involve analyzing stratigraphic sequences and sedimentary deposits, organic matter and macrofossil deposits, as well as using radiocarbon dating (¹⁴ C), thermoluminescence (TL), and even dendrochronology. Some research uses statistical and mathematical models to evaluate river flow rates, which are transposed and adapted to ancient flood events. In fact, most of the work pertaining to the reconstruction of the frequency and magnitude of ancient floods uses several methods and techniques to obtain the long-term chronology of flood events in relation to the specific conditions (e.g. climate, geomorphology) of a region or study area.

Many studies have analyzed slackwater deposits to reconstruct the paleoflood history of rivers (ELY and BAKER, 1985 ; JONES et al., 2001 ; KOCHEL and BAKER, 1982 ; WOHL et al., 1994). These kinds of deposits are mainly composed of fine-grained sediments (silt and fine sand) from river banks that were deposited during "large floods in areas of reduced flow velocity caused by ponding, eddying, or back-flooding up tributaries" (ELY and BAKER, 1985, p. 104). These deposits were found in different fluvial environments and for many researchers serve as adequate physical and geomorphologic indicators in the reconstruction of relic floods. Also, in many studies researchers used both slackwater deposits (SWD) and paleostage indicators (PSI) for the reconstruction of ancient floods. In addition to slackwater sediments, flood debris and silt lines (paleostage indicators) are often well-preserved along the river terraces (BAKER,1987; OSTENAA et al., 2002; WOHL et al., 1994) and facilitate the reconstruction of ancient flood events.

The main problem in reconstructing paleofloods essentially lies in finding layers of organic matter in the sedimentary sequences, which makes it difficult to date the flood events using radiocarbon methods (¹⁴C). In arid or semi-arid environments, these organic layers are often rare because the prevalent climatic conditions do not favor the formation of a thick vegetation cover. However, this problem is not limited to these environments but also characterizes humid regions (YANG et al., 2000). The frequent absence of such organic matter layers in sediments is either due to an overly short vegetation cover formation period between phases of flooding or the erosion of such layers through the action of different natural phenomena (e.g. streaming, gliding, bioturbation, fluvial erosion). Other methods through which flood events can be dated include dendrochronology and thermoluminescence (TL). This last method can be used to obtain relatively precise dating of archaeological artefacts but at present is not very reliable for dating mineralogical samples (e.g. quartz or feldspath grains). The various problems involved in dating ancient floods make us aware of the importance of using as many physical and biological field indicators as possible in order to reconstruct the chronology of flood events as precisely as possible.

Keywords:

• Paleohydrology,
• paleofloods,
• Climatic change,
• Multidisciplinary approach,
• Literature review