DOWNSCALING FUTURE CLIMATE SCENARIOS TO VINEYARD SCALE DESCENTE D’ÉCHELLE DES SCÉNARIOS DE CHANGEMENT CLIMATIQUE À L’ÉCHELLE DES TERROIRS VITICOLES

LE ROUX, Renan1*; DE RESSÉGUIER, Laure²; QUÉNOL, Hervé1; VAN LEEUWEN, Cornelis² 1 LETG-RENNES, UMR 6554 CNRS, Université de Rennes 2, Place du Recteur Henri Le Moal,35000 Rennes, France; 2 EGFV, Bordeaux Sciences Agro, INRA, Univ. Bordeaux, 33140 Villenave d’Ornon, France *Corresponding author: renan.leroux@univ-rennes2.fr

Abstract: The IPCC has created several climate scenarios evaluating the impacts and consequences of climate change. All agricultural sectors will be concerned, in particular grape and wine production. Climate, especially temperature, has a key role in grapevine development. Temperature variability can be significant at vineyard scale. Current climate change models are not accurate enough to take into account temperature variability at the vineyard scale. This study therefore proposes a method for downscaling future climate scenario at vineyard scale. The study focuses on the Saint-Emilion and Pomerol production area in France. A network of 90 temperature data loggers was used to collect air temperatures inside the vine canopy. These measurements allowed the creation of fine scale geostatistical models of temperature and fine scale maps of daily temperature for 4 successive years (2012, 2013, 2014 and 2015). Temperature data for these years are also available at 9 km resolution, using data from SAFRAN (Météo-France). Climate change model data are available for France at a 9 km resolution for all new IPCC scenario (DRIAS portal, from ALADIN model). For the years 2012, 2013 2014 and 2015, nine temperature patterns at regional scale were identified by a recognition algorithm. Fine scale daily maps were associated to these nodes, then for each day of the futures scenarios, the same regional patterns are identified and associated with the corresponding fine scale maps, creating fine scale maps for each day of the period 2005 – 2100 for RCP 4.5 and RCP 8.5. The results demonstrate, according to the actual futures scenarios, a rise of the temperature over the study site. However, the temperature distribution of the study site does not show major variation at mid term (2031-2050) and long term (2081-2100) whatever the scenario used.

Keywords: climate change, fine-scale modeling, regional modeling, phenology, downscaling
Réservé aux membres / Members only

Devenir membre / Membership