Bioclimatic modelling of steep-sided valleys in northeastern France

PI : Jean-Luc Dupouey (UMR 1434 SILVA)

Co-applicants : Anna Schmitt, Sandrine Chauchard et Vincent Badeau (UMR 1434 Silva)

Collaborations : Christophe Randin (Université de Lausanne, Département Ecologie et Evolution)

________________________________________________

Context — The lowlands of northeastern France contain numerous scattered climatic microrefugia, located in steep-sided valleys. These particular landforms, so-called “cold valleys”, display specific habitats of high environmental value, linked to the presence of different contrasting microclimates that coexist on short distances. Microclimates of the “cold valleys” are decoupled from the regional climate. On the south-facing slopes, climate is sub-Mediterranean whereas on the north-facing slopes and in the thalweg, climate is sub-montane. Consequently “cold valleys” exhibit large variations of climate at fine scale (below 100m) and constitutes microrefugia for both cold-adapted and thermophilous species. Thus, they could contribute to mitigate climate change effects, but are also under threat of regression.

Objectives — To protect them, we need to improve our knowledge of their location and to finely characterise their vegetation and their microclimates in relationship with topography. More widely, cold valleys represent ideal and appropriate case studies for i) understanding the climatic basis of the habitats distribution of relict and/or specialized plant species and ii) challenging and refining species/community distribution models.

Approaches — In a first step, we will develop a modeling framework in order to detect the distribution of cold valleys in the landscape by using a set of topographic variables: slope, aspect, confinement, topographical position and geology. In a second step, we will generate a set of improved climate-based predictors for species and community distribution models in order to better predict the spatial and temporal variations of local and micro-climate conditions within cold valleys. For this, we will first reconstruct the long-term temporal variation of temperature and moisture at sites with temperature and humidity loggers and based on a recently-developed framework. The reconstructed temperature and moisture variability will then be spatially interpolated by using multiple-regression techniques. In a third step, solar radiation and cold air pooling will be introduced in the models to refine them. Finally, this new set of predicting variables will be tested in species and community distribution models.

Expected results and impacts — To provide better tools for the identification of potential sites of presence of rare species, and thus help conservation programs. To anticipate the future distribution of these rare microclimates and associated habitats in a context of climate change, and thus help adapt the management of this high conservation value sites.