CLIMATE WARMING: CONSEQUENCES FOR VITICULTURE
AND THE NOTION OF ‘TERROIRS’ IN EUROPE
Bernard Seguin and Inaki Garcia de Cortazar
Unité Agroclim, INRA, Site Agroparc, F-84914, Avignon cedex 9, France
Keywords: grapevine, climate change, climate, phenology, greenhouse effect, vintage.
Abstract
As for any crop, the impact of an anthropogenic greenhouse effect will
include the stimulating effect of increased CO2 concentration on photosynthesis,
which will result in increased dry matter production and may lead to noticeable
changes in cultural practices. However, it is more likely that the most significant
impact will result from rising temperature, even if other climatic variables, such as
rainfall also are considered. Apart from a significant displacement of the traditional
limits for grapevine cultivation, serious questions may arise concerning ‘terroirs’:
Will it be possible to keep the same cultivars by adjusting vineyard cultural and
enological practices? How might viticulture and wine-making respond to the
predicted increase in surface temperature? The warming trend of the last fifteen
years in most of Western Europe and especially in France may provide some clues.
The phenology of the grapevine has significantly advanced; one to two weeks for
flowering and almost one month for harvest date in the last 50 years. The advance
in harvest date also has been accompanied by changes in sugar and acid
concentrations. Expressed in terms of the Huglin index, the increase in temperatures
due to global warming will lead to vintages that are more uniform across years.
However, there may be a tendency in which the climatic variables of a particular
grape growing region will exceed the established limits for grape cultivars strongly
associated with that location (terroir). The 2003 growing-season, characterized by
very hot weather and drought (this being similar to the end of century climatic
scenarios) provides evidence that bioclimatic indices do not take into account the
possible natural adaptation of some grape cultivars to the predicted changes in
climate. Therefore, only the use of more sophisticated tools such as accurate crop
models, currently under development, may provide a more valuable perspective on
future viticulture.
INTRODUCTION
The general aspects of climate change impacts upon agriculture have been
presented in several documents these last years (Rosenzweig and Hillel, 1998; Reddy and
Hodges, 2000; IPCC, 2001). Although grapevine production is known to be very
dependent upon climate, it has been less studied than annual crops or pastures in the
context of climate change impacts. Few elements are given in the although very
comprehensive assessment of IPCC 2001, and more information only exists in the review
presented by Schultz (2000). From this date, the signs of observable climate warming in
the last years have been observed, especially in Europe, and they give presently more
elements to precise the main lines of potential impacts and discussing the ways of
adaptation.
Which Consequences for Vine Physiology?
In order to discuss the possible impact on general viticulture, we first have to
present what is known about the local scale effects on the vine plant ecophysiological
functioning. We will here sum-up the main elements, using the review by Schultz (2000)
and adding some new inputs from recent studies.
Apart the effects of strict climate change, we first have to notice that vine, as any
other vegetation, will be directly concerned by the increase of the atmospheric
concentration of carbon dioxide CO2, which is the main driver of the anthropic
accentuation of greenhouse effect. In the classical hypothesis of a doubling of CO2
concentration (compared to that in the years 90s’, thus corresponding to about 700 ppm),
the resulting induced stimulation of photosynthesis will amount to about 30 %, depending
upon varieties and physical environment (Bindi et al, 1996a). It will lead to a resulting
increase of 15 to 20 % for biomass production, taking into account the simultaneous
increase of respiration. It will be associated to an increase of water use efficiency by
about 10 % due to the increase of stomatal resistance.
At this stage, it is necessary to point out that the long- term exposure may have a
very different effect. Acclimation response may be substantial, and will vary between the
varieties and training modes.
These effects will be combined with those directly resulting from the modification
of climatic factors. The most significant will evidently be the temperature, which will act
by accelerating the phenology. It will both lead to a shortening of the development phases
and an advance in stages which will encounter climatic conditions, not only modified by
the climate change, but also by the shift in phenological calendar: in the case of
Languedoc vineyards, simple simulations of phenology changes for mean warmings of
2°C and 4°C display a moving of maturing period starting from 13 august now to
respectively 23 and 4 July (Lebon, 2002). The whole cycle will be subject to higher
temperatures. Outside 33° C in leaf temperatures (Schultz, 2000), the optimum for
photosynthesis will be overpassed, but threshold values are rather badly known.
Other factors like solar radiation, air humidity and wind velocity will also have to
be considered, but they may be considered as second-order factors when compared to
rainfall. The most recent scenarios give more and more reliable informations on it, but
there are still some questions as their predictions result from the correct assessment of the
whole water cycle. For the case of Europe, they do not indicate a major change for
northern regions, but the tendency to predict the extension of summer droughts in the
South has to be seriously considered for these areas. The question of water availability,
also reinforced by the increase of water needs by the elevation of potential
evapotranspiration PET, will contribute to the resulting panel of possible effects.
On the whole, and adding the possible additional impact of UV-B increase
(Schultz 2000), the synthetic overall impact is difficult to predict in a simple way. If it is
clear that the higher production of biomass, associated to the higher water use efficiency,
will potentially lead to an ecophysiological functioning significantly different from actual
conditions, it is still difficult to assess the consequences for vine growing and to define
revised technical practices. Only the use of crop simulation models (as in Bindi et al,
1996b) could allow to cope with the complexity of involved processes and of their
interactions. However, they are still presently at the stage of development (as for STICSVigne, see Brisson et al, 2000).
Which Consequences for General Viticulture in Europe?
These basic effects on physiology only consider a given vine plant or vineyard in
its present state, when submitted to the future atmosphere and climate. As for all the
agricultural activities, it is obvious that the adaptation to a changed climate will involve,
not only changes in physiology at the crop scale, but also of the whole cropping system at
the local or regional scale, and more largely in the areas of production on a global scale.
We will now focus on these larger scales (defined in Fig.1), where significant evolutions
may be predicted.
Starting on the global scale, and restricting us to the example of Europe, the first
evident change will be observed in the northern limits. They were identified by Branas
(1946) as related to an heliothermic index (Fig.2).
It is clear that these limits will be moved to more northern latitudes by about 500
km for a mean warming of 3 °C, knowing that a warming of 1°C corresponds to an
approximative shifting of 180 km towards the north (Moisselin et al, 2001). This
prediction may be supported by the historical analysis of Le Roy Ladurie (1983) and
Legrand (1978) which reported the main evolutions of harvest dates from the middle-age
and mentioned the disappearance of vine cultivation in England between the years 1000
and 1200, partly due to a cooling of some °C (Fig.3)
The possible evolution of southern limits (in Maghreb or Spain) is more difficult
to predict, as it will depend upon not well-known thresholds in elevated temperatures and
also rainfall changes, which could however be compensated by irrigation. If the predicted
increase of drought around the mediterranean basin is effective, it could also lead to a
more systematic use of irrigation in areas as southern France, presently severely restricted
in high-quality productions.At this same global scale, the extension of vine cultivation
could also be possible in the eastern parts of Europe, as predicted by Kenny and Harrison
(1993) for the cultivation of Muller-Thurgau in Poland and Ukraine.
Which Consequences for Terroirs in Europe?
Apart from these possible significant changes in potential zones for vine
cultivation, the climate changes raises serious questions for the well-known link of wine
production to the notion of terroirs, as expressed by the classification of AOC in France
or DOC in Italy, and so forth.(see Vaudour, 2003). From the point of view of physical
factors (evidently completed by the history and long-term traditions), it is based upon a
straight balance between three elements: soil, climate, cropping system (Fig.4).
As soil is not movable, a significant climate change may threaten this straight
balance.
Contrary to most of the other crops, which could move to follow the climate,
geographical shifts (except some small-scale displacement using topography) are
impossible to face in this context (the AOC cannot be delocalized !!). To which limits
will they be able to adapt to this change, and if not which changes in the third component
(cropping system) are to be suggested?
The answer to this last point, which could also involve local scale adjustment of
microclimate by soil management, adjustment of canopy height, etc.., is still difficult to
precise, as it probably requires the use of crop models, as previously mentioned.
So that the present analysis is restricted to the use of large-scale climate indices (a
comprehensive survey is given in Vaudour, 2003). Among the various available, we have
selected the Huglin index (1986), which was based on a fine analysis of main terroirs in
France. We have to point here that it only relies on temperature, which may be justified in
a first approach, as the general restriction on the use of indices to correctly express
possible changes with climate change equally applies for all these indices.
The use of climatic scenarios proposed by Meteo-France (Perarnaud et al, 2003),
with a spatial resolution of 50 km, for the period 2070-2100 points out a significant
change for both the climatic classification given by Huglin and the adequacy of varieties
he has established with the ‘past’ climate of the years 70 (see Fig.5 for the example of
Avignon, which would move from mild-hot to hot in terms of climate and indicates that
the usual varieties like Syrah and Grenache would have to be replaced by others more
adapted to a largely warmer climate).
Similar changes have been computed with the same Huglin index for Geisenheim
in Germany (Schultz, 2000) and other sites in Germany and Italy (Battaglini, 2003). They
clearly indicate threatening with the maintening of the traditional terroirs when submitted
to this climate warming, which is well perceived by wine growers (about 60% of
questioned wine growers in Europe, with almost 60 % ready to adopt new varieties in
order to face this changing in Germany, compared to about 45 % in Italy and only 30% in
France (Battaglini, 2003).
The main question is to know if these projections, mainly based on phenology
simple models and the Huglin index, are reliable enough to correctly predict the future of
terroirs for this century. One first element of validation may be deduced by making use of
recent climate warming in Europe, and especially of the exceptional summer 2003.
The Recent Warming and the Exceptional Summer of 2003
Effectively, effects of climatic change have been already noticed, in relation to the
factor temperature, evidently central in the context of global warming foreseen in climatic
scenarios. It has been identified in the climatic context of France, with an order of
magnitude of 0.9° C during the last century ( Moisselin et al 2002), with a recent increase
between 0.4 and 0.6 ° C during the last decade. This warming has induced noticeable
changes in the phenology of nearly all perennial crops, especially for flowering dates
(advanced by two to three weeks for vine as well as fruit trees, see Fig.6).
Harvesting dates have been also advanced by almost a month in fifty years, in all
the french vineyards (see the example of Côtes du Rhône, Fig.7 from Ganichot, 2002),
without any possible other explanation by the evolution of technical practices.
It is already possible to give some first elements by using intermediate tools taking
the form of indices, as this proposed by Huglin (1978). The retrospective analysis for the
last 30 years and for 7 sites where meteorogical data have been collected within the INRA
network (Angers, Avignon, Bordeaux, Colmar, Dijon, Montpellier, Valence) (see Fig.8
for Avignon and Dijon in Burgundy) clearly displays a tendancy to higher values for the
Huglin index, together with a lower interannual variability. The warmer and more regular
climate is evidently favorable to the wine quality, as attested by the observations from the
growers who confirm the higher sugar content and lower acidity (see Duchêne and
Schneider, 2004 for Colmar in Alsace). However, it also appears that this evolution
approaches the previously established limits in terms of adaptation for varieties, with a
problem for maintaining the usual equilibrium between sugar and acidity.
The warming has culminated with the summer of 2003, which was characterized
by a really exceptional episode of nearly 3 months (June to August) combining hot
temperatures (higher than normal by between 3 to 5 °C) and a severe drought. A detailed
analysis of this event, which has hit a part of Germany, Italy, Spain and Portugal, but was
mainly centered on France, may be found in Seguin et al (2004). Limiting us to the
consequences for wine, we may notice that it has led to an exceptional advance in harvest
dates (the most advanced ever noted in Burgundy from the analysis of historical series as
far back as 1370!!,), with very high degrees of alcohol and low acidity in agreement with
the exceptional high values of Huglin index observable in fig.7 and 8. The overall quality,
not yet totally analysed, appears to be highly variable and sometimes not in agreement
with the usual fine-scale variabilities within the terroirs. Which allows to state that their
limits of resilience regarding warmer conditions have been attained, but in any case not as
seriously overpassed as indicated by the Huglin index. As this year has closely
approached the climatic characteristics of summers predicted by the scenarios, this year
2003 allows to invalidate the predicting capacities of empirical indices like that of Huglin.
We may presume that it has been calibrated in the steady-state climatic conditions
prevailing in the years 70, and is not well adapted for a precise assessment of future
conditions, because it relies on spatial variations which underestimate the adaptative
capacities of both varieties and wine-making technologies. It may express some true
tendencies in vine production, but not enough precisely to correctly precise the future of
terroirs. The threatening with climate warming is effective, but requires more
sophisticated tools for assessing its consequences and the adaptation strategies, and in this
sense, the exceptional year of 2003 will be especially useful.
CONCLUSION
When looking to the far future (end of this century), we have to combine both
effects of climate change and CO2 increase, and to consider the interactions between
climate, soil, genotypes and cultural practices. Some partial elements may be already
considered, as the adverse effect upon quality in mediterranean regions which could result
from the shifting of maturing period earlier in summer with warmer temperatures (Lebon,
2002). But many other factors will have to be considered, as the modifications in rainfall
and water balance at the first range. Only the availability of a mechanistic model, with
climate inputs from regionalised scenarios, would be able to integrate the complexity of
possible interactions. The STICS-vigne model, in the way of elaboration (Brisson et al,
2003), appears as especially suited to the objective of defining the technical modalities for
the adaptation to thes new conditions.
For the near future, the prospective appears to be more difficult, because we have
now to take decisions for the years 2020 without a clear information about the climate of
this period coming from climate modelers. The only possible way consists in assuming
the following-up of the recent tendency, and to submit it in terms of the index of Huglin
and later outputs of STICS-vigne model. The main informations resulting from this
procedure will be assessed in the next two or three years in order to elaborate reasonable
prospective elements for this near future.
Literature Cited
Battaglini, A. 2003. Perceptions des changements climatiques par les viticulteurs
européens. Le lien de la vigne, Assemblée générale du 4 avril 2003, la viticulture
mondiale face à l’évolution du climat. www.vinelink.org/Home/Ang/DefaultAng.htm
Bindi, M., Fibbi, L., Gozzini, B., Orlandini, S. and Miglietta, F. 1996a. The effect of
elevated CO2 concentration on grapevine growth under field conditions. Acta Hort.
427: 325-330.
Bindi, M., Fibbi, L., Gozzini, B., Orlandini, S. and Miglietta, F. 1996b. Modelling the
impact of future climate scenarios on yield and yield variability of grapevine. Climate
Res. 7 : 213-224.
Branas J., Bernon G. and Levadoux, L. 1946. Eléments de viticulture générale. Dehan,
Montpellier (France).
Brisson, N., Gaudillère, J.P., Ramel, J.P. and Vaudour, E. 2002. Utilisation du modèle de
culture STICS pour renseigner les zonages viticoles. IVe symposium sur le zonage
vitivinicole. Avignon, Juin. CD edited by OIV.
Duchêne, E. and Schneider, C. 2004. Grapevine and climatic change: a glance at the
situation in Alsace. Submitted to Agronomie.
Ganichot, B. 2002. Evolution de la date des vendanges dans les Côtes-du-Rhône
méridionales. 6émes Rencontres Rhodaniennes. Ed. Institut Rhodanien. Orange,
France. p. 38-41.
IPCC. 2001. Climate change 2001: impacts, adaptation and vulnerability, Contribution of
Working Group II to the third assessment report of IPCC, Cambridge University
Press, Cambridge.
Kenny, G.J. and Harrison, P.A. 1992. The effect of climate variability and change on
grape suitability in Europe. J. of Wine Research. 3:163-183.
Huglin, P. 1978. Nouveau mode d’évaluation des possibilités héliothermiques de la vigne.
C.R. Acad. Agric. Fr.1117-1126.
Lebon, E. 2002. Changements climatiques : quelles conséquences prévisibles sur la
viticulture ? 6émes Rencontres Rhodaniennes. Ed. Institut Rhodanien. Orange, France.
p. 31-36.
Legrand J.P. 1977. Fluctuations météorologiques, vendanges et activité solaire. La
Météorologie. 9 :73-91.
Le Roy Ladurie, E. 1983. Histoire du climat depuis l’an mil, Flammarion, Paris.
Moisselin, J.M., Schneider, M., Canellas, C. and Mestre, O. 2002. Les changements
climatiques en France au XX° siècle : étude des longues séries homogénéisées de
température et de précipitations. La Météorologie. 38 : 45-56.
Perarnaud, V., Seguin, B., Malezieux, E., Déqué, M. and Loustau, D. 2004.
Agrometeorological research and applications needed to prepare agriculture and
forestry adapt to 21st century climate change. To appear in Climatic change.
Reddy, K.R. and Hodges, H.F. 2000. Climate change and global crop productivity. CABI
Publishing, Wallingford.
Rosensweig, C. and Hillel, D. 1998. Climate change and the global harvest. Oxford
University Press, Oxford.
Schultz, H.B. 2000. Climate change and viticulture: a european perspective on
climatology, carbon dioxide and UV-B effects. Australian Journal of grape and wine
research. 6:1-12.
Seguin, B. 2003. Relation entre climat et terroir à différentes échelles spatiales : apport de
nouveaux outils méthodologiques. IVe symposium sur le zonage vitivinicole,
Avignon, Juin. CD ed by OIV.
Seguin, B., Baculat, B., Baret, F., Brisson, N., Huard, F. and Ruget, F. 2004. An overview
of the consequences of the summer 2003 for agriculture in France. VIII ESA congress:
European agriculture in a global context. Copenhagen, 11-15 july 2004. To appear in
the proceedings.
Vaudour, E. 2003. Les terroirs viticoles. Définitions, caractérisation et protection. Dunod,
éditions La vigne, Paris.
Figures
Fig.1 The spatial scales of climate
Septentrional limit of the vine in Europe
Northern limit of the vine
Isoheliothermic 2.6
Isoheliothermic –1 °C in january
Fig.2. The northern limits of vine cultivation in Europe (adapted from Branas,1946)
Cultivation of the
vine in England
on
iati
var
Hot period
d
ose
p
Sup
Regression time of
the vine in latitude.
Fresh
period
Fig.3 (adapted from Legrand 1978)
Fig.4. The physical component of terroirs (from Ecole Polytechnique Fédérale de
Lausanne)
Fig.5. The Huglin index computed for Avignon (France south-east) with
temperature data as simulated by Meteo-France for the years 2070 to 2100
23 23-juil
- july
13 13-juil
- july
03 03-juil
- july
2323-juin
- june
1313-juin
- june
0303-juin
- june
2424-mai
- may
Bordeaux
1414-mai
- may
Colmar
Marseillan
19
70
19
72
19
74
19
76
19
78
19
80
19
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
98
20
00
20
02
0404-mai
- may
Fig.6. Trends in flowering dates of Chasselas variety in Bordeaux, Colmar and
Marseillan-Montpellier (data from the Phenoclim database).
/Agrosystems
Agronomy
years
100
year
month
day
hour
1
min.
s.
Hydrology
Pedology
Physiolo
Physical
Turbulen
Ecophysi
Phytocli
Agronom
Geostatis
Microclim
Continent
Hydrolog
Transfer
/field
Ecology
Climate
Global
gy
chemistry
ce
ology
mate
tics
ate
al
y
Trends of wine harvest in Chateauneuf du Pape since 1945
11-oct
6-oct
1-oct
26-sept
21-sept
16-sept
11-sept
6-sept
1-sept
19 4 5
19 5 0
19 5 5
19 6 0
19 6 5
19 7 0
19 7 5
19 8 0
19 8 5
19 9 0
19 9 5
20 0 0
Fig.7. Trends of wine harvest in Chateauneuf-du-Pape (from Ganichot 2002)
Avignon (1965 – 2003)
Dijon (1965 – 2003)
Very Hot Cli mate
Mil d -Hot Climate
Mil d Cli mate
Huglin Index
Hot Cli mate
Col d Cli mate
Very Col d Climate
Fig.8. The consequences of the recent warming for the Huglin Index in Avignon
(left) and Dijon (right).