Poster De Conférence Année : 2024

A methodological approach to elaborate a modern reference in order to evaluate the application of the Landscape Reconstruction Algorithm in the Northern French Alps

Résumé

In recent years, studies based on modern analogues have been developed with the aim to test and calibrate the methodological advances in palaeoenvironmental sciences. Some of these advances, such as the development of models based on palynological studies to assess the spatial extent of vegetation across the Holocene, have required validation with modern or historical analogue studies. The Landscape Reconstruction Algorithm (LRA) allows vegetation to be reconstructed at known regional and local scales by considering pollen production and dispersal of each taxa and site-type dependent deposition mechanisms. Although the application of these model schemes is now commonly applied over Europe, there is still a critical need to understand how they perform in mountain environments where only few studies have been done. The aim of the present study is to reconstruct the spatial and temporal dynamics of vegetation in the northern Alps, at both regional and local scales, by applying the LRA to pollen records covering the last five millennia. In order to optimize the reliability of the LRA estimates for past vegetation reconstruction, the application of the LRA in mountain regions has to be tested by using modern analogue studies. Such approach requires a succession of methodological steps, each of them are important to improve the LRA estimates. In this poster, we present our methodological approach to establish a modern reference for evaluating the performance of the LRA in mountain environments. For this purpose, we cored 28 lakes and 16 bogs in the northern French Alps area to obtain modern pollen assemblages. Mountain lakes might present very different sedimentation rates. This feature requires to adapt the thickness of sediment samples in order to get a similar time period from a lake to another. To deal with this issue, chronologies of each top core sedimentary samples were assessed combining measurements of atmospheric lead pollutions (Avaatech XRF core scanner) with few short-lived radionuclide measurements (210Pb, 137Cs), when available. Based on the reconstructed chronologies, two pollen samples were taken and analysed in each core, corresponding to the periods post 70’s (modern and sub-modern pollen assemblages), and between the 50’s and 70’s (before the phase of reforestation caused by the agricultural decline) to check the proposed chronology. To compare the LRA results with modern analogues, a vegetation map was created and designed to get formatted vegetation composition data for comparison with the pollen types used in the LRA. The modern vegetation map is based on information from three different sources, i.e. Corine Land Cover, CNRS maps of potential vegetation, and CBNA (Conservatoire Botanique National Alpin) habitat map. To attribute plant compositional information to the different vegetation units defined from the maps, ca. 5000 botanical surveys (CBNA database and field work) and IFN (Inventaire forestier national) inventories have been used. Based on these two key methodological improvements, we are now able to apply the LRA to our modern pollen assemblages from mountains and compare the results to current vegetation (abundance/composition).
Fichier principal
Vignette du fichier
Poster_Q14-aj.pdf (85) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04952309 , version 1 (17-02-2025)

Identifiants

  • HAL Id : hal-04952309 , version 1

Citer

Andréa Julien, Charline Giguet-Covex, Laurent Marquer, Erwan Messager. A methodological approach to elaborate a modern reference in order to evaluate the application of the Landscape Reconstruction Algorithm in the Northern French Alps. Q14, Feb 2024, Rennes, France. 2024. ⟨hal-04952309⟩
0 Consultations
0 Téléchargements

Partager

More