Partenaires

CNRS


Rechercher


Accueil du site > LIFE ADVICLIM (2014-2019) LIFE13 ENV/FR/001512 > WORKPACKAGE

A1 : Observation and spatial conceptualisation at vineyard scale

Leader : Kees Van Leuwen and Valérie Bonnardot

par Hervé Quénol - publié le

1.1 Climatic and phonological observations at fine scale

National climate monitoring networks are frequently of insufficient spatial resolution to provide a clear picture of the temperature patterns in wine regions with a complex terrain. The originality of the ADVICLIM project lies in a unique network of thermal sensors and weather stations established since 2008 within the framework of the ANR-TERVICLIM and GICC-TERADCLIM programs. The network of thermal sensors consists of 400 sensors located in more than 33 wine estates in 14 wine producing countries. This fine scale network will provide valuable data to assess the spatial variability in each site, and to perform multi-criteria spatial modeling. Every experimental sites consist of an imbricated network of climate monitoring. The choice of the locations for temperature data loggers depends on topography (need to represent all parameters such as aspect, altitude, etc...) ; on soil types (need to represent the difference in soil texture, depth, etc.). Moreover, the number of temperature data loggers will depend on the environment (diverse or not) and the surface of the estate/district as it needs to cover it regularly. An electronic system of communication between the different data loggers and a website platform is required. The thermal sensors (Data Loggers type) are located over a few square km surface in different wine producing regions worldwide. To date, these sensors are programmed and downloaded in the field with a laptop every month, resulting in important travelling costs and high consumption of time. The aim is to create a connection between each sensor (per site) in order to download real time data continuously via a Global System for Mobile Communications (GSM) and thus examine them via Internet. We will also define the protocols of observations of the main stages of phenology (bud burst, flowering, veraison) to be observed in each experiment. Those protocols will define main stages and frequency. Data obtained in each vineyard will be stored in a common database in order to be available for the scientific community after the end of the project.

1.2 Modeling of climate variability at fine scales

Modeling at fine scale will include (i) the output from numerical EURO-CORDEX models with a kilometre resolution (ii) the spatial modeling of climatic data from the measurement networks using multicriteria modeling at very high resolution (90m), and (iii) the future climate simulations using meso-scale climatic model ran under different scenarios of climate change. (i) The coarse resolution output from numerical climate models require downscaling. We use the downscaling output of EURO-CORDEX. It will provide knowledge and understanding the climate variability at meso-scale in the different studied European wine regions. Climatic data from national weather station networks will be used to validate the modelled outputs data. (ii) In order to construct fine-scale spatial temperature fields, the multicriteria modelling will be used. This approach takes environmental factors into account. Indeed, the role of topographic factors in the spatial variability of temperatures at fine scales, in addition to the influence of geographical location (latitude/longitude) at larger scale has already been demonstrated. This type of modeling will make use of the climatic data provided by the fine scale networ. (iii) We use simulations of climate change scenarios (for Europe) carried out CORDEX program. These data are integrated into statistical models. Combining the outputs of regionalized CORDEX with multi local models models will perform simulations of bioclimatic indices 2041-2050.Combining the outputs of regionalized CORDEX with multi local models will perform simulations of bioclimatic indices 2041-2050.

Process-based phenological models for the grapevine work on the assumption that phenological development is mainly regulated by temperature. These models are driven by a temperature summation from a defined date and above a minimum temperature (threshold) until the appearance of a phenological stage (often judged at 50% level of appearance). Among phenological models for the grapevine, the Winkler Index (Winkler et al., 1974) allows classifying vineyards worldwide into climatic zones. A different model was proposed by Huglin, 1978. This author also published heat requirements for a set of grapevine varieties, allowing adjusting grapevine varieties to local climatic conditions. Although of interest, the limit of these two models is the fact that their construction was based on a limited dataset, collected in a limited number of sites. Recently, a new Grapevine Flowering Veraison Model (GFV) was published by Parker et al. (2011). This model is based on an extensive dataset (over 4,000 phenology observations collected in 123 sites) and advanced modelling techniques (PMP modelling platform, Chuine et al., 2003). It allows precise prediction of the timing of major phenological stages (flowering and veraion) for approximately one hundred cultivars of Vitis vinifera (Parker et al., 2013). This model was validated at a regional scale with data collected in classic weather stations. In the ADVIDCLIM project, the GFV model will be tested at a very refined scale, inside winegrowing regions, with temperature data collected with sensors placed within the vineyards. To this purpose, phenological observations will be collected close to the temperature sensors. These phenological observations include flowering, veraison and ripening. For grape ripening, grape samples will be taken weekly from veraison to ripeness. Sugar accumulation will be modelled as a function of local temperatures. The local use of the GFV model will allow the transformation of the temperature maps in spatialized predictions of the occurrence of phenological stages. It is likely that growers will have to change grapevine varieties in the future due to changing climatic conditions. The phenology maps, coupled to established heat requirements for grapevine varieties, will allow growers to optimize the adjustment of varieties to local climatic conditions.