Impacts of Climate Variability and Change on Wine

Home , Grape, Terroir, Tom Stevenson, Vintage

GEOSCIENCE CANADA 1

PAPER 4 FROM THE “TERROIR,GEOLOGY AND WINE:A TRIBUTE TO SIMON J. HAYNES”SESSION HELD AT THE GEOLOGICAL SOCIETY OF AMERICAN ANNUAL MEETING,SEATTLE,WASHINGTON,NOVEMBER 2, 2003

for threshold issues in which too much Earth- and human-based systems. From warming will potentially alter traditional controlling vegetation patterns and geo- wine styles and/or varieties planted, and logical weathering characteristics, to likely bring about spatial shifts in viticul- influencing water resources and agricultural viability. tural productivity, climate is at the heart of the delicate equilibrium that exists on SUMMAIRE Earth. Climate is a wide-ranging factor Cet élément du concept de terroir qu'est in virtually all forms of agriculture, from le climat détermine la viabilité d'une influencing spatial variations of crop région à permettre le mûrissement d'une viability to largely determining year-to- variété définie de raisin ainsi que l'élabo- year yield variability; climate is an impor- Climate and Terroir: Impacts ration du style du vin qu'on en tire. La tant issue in determining where and how variabilité et le changement du climat crops are grown. Climate-related impacts of Climate Variability and affectent toute forme d'agriculture, mais on agriculture are generally related to cli- Change on Wine les effets sont rarement plus évidents mate variability over the short term (i.e., que sur la production de grands vins où intra-annual and inter-annual) and cli- Gregory V. Jones l'optimisation de la qualité nécessite une mate change over the long term (i.e., Department of Geography, Southern Oregon viticulture en des zones climatiques aux decades to centuries or longer). University, 1250 Siskiyou Blvd, Ashland, conditions très circonscrites, d'où les Understanding such relationships is sel- Oregon 97520, U.S.A., [email protected] risques. Bien que les climats aient dom more important than today, as vari- changé dramatiquement dans l'histoire ations and/or changes in climate can SUMMARY de la culture des vignes vinicoles, les greatly impact the ability to produce a As part of the concept of terroir, cli- changements climatiques récents ainsi given crop. mate is a factor that influences both the que les changements projetés dans un general viability of a region to ripen a avenir rapproché ont été beaucoup The Climate Component of Terroir specific variety of grapes and the result- étudiés en fonction de leurs impacts sur Grapevines are some of the oldest culti- ing wine style. Climate variability and l'actuelle industrie vinicole. Les compte vated plants and have been historically change affect every form of agriculture rendus des recherches montrent que les associated with Mediterranean climates and are seldom more evident than with changements climatiques, tant à courts et (e.g., Italy). Today, however, grapevines the production of high quality wines qu'à longs termes, ont amené des saisons for wine production (note that all occur- where narrow climate zones for opti- viticoles plus chaudes et plus longues rences of grapes or grapevines refer to mum quality place them particularly at comportant moins de risques de gel, et grapes for wine production, unless stat- risk. While climates have changed dra- cela dans nombres des meilleures ed otherwise) are grown in many types matically during the history of the culti- régions vinicoles du monde. Bien que of climates throughout the mid lati- vation of grapes for wine, the changes ces changements aient été associés à une tudes: Mediterranean, marine west coast in the recent past and those projected in amélioration de la production et de la (e.g., Oregon), humid subtropical (e.g., the near future have received a great deal qualité, les projections climatiques mon- eastern Australia), and semi-arid conti- of attention for their potential impacts trent qu'il est possible qu'un seuil soit nental climates (e.g., eastern Washington on today’s wine industry. Research has atteint, au-delà duquel ce réchauffement state) (Fig. 1). The climates of these shown that both recent short and long- pourrait altérer le style les vins tradition- regions are an integral part of the term climate changes have resulted in nels et/ou les variétés utilisées, entraî- notion of terroir, the French concept in generally warmer and longer growing nant des changements spatiaux proba- which a complex interplay of physical seasons with less frost risk in many of bles de la viabilité viticole. factors (Wilson, 1998; Haynes, 1999; the world’s best wine regions. While Meinert and Busacca, 2000) and cultural these changes have been related to INTRODUCTION influences (Vaudour, 2000) interact to greater production and quality, future cli- Climate is a very complex, highly vari- define the wine styles and quality that mate projections indicate the potential able, and pervasive factor in our natural come from any site or region. Climate is 2 one part of the continuum that includes 2003). While the average climate struc- Although the climate-based aspects of the physical landscape influences of soil ture is very important for grape growth terroir are related to various factors that and terrain, which in combination large- and wine quality, weather and climate operate from the macroscale to the ly determine the grape variety that can factors on daily and hourly time scales microscale, measures of temperature on be grown in a given region. Once grape are critical, and include: solar radiation, daily, weekly, monthly, and growing sea- variety-site characteristics are considered, heat accumulation, temperature extremes son time scales have most often been the remaining viticulture (the science of (including high temperatures during the employed to define the spatial differ- the cultivation of grapevines) and enolo- summer, winter freezes and spring and ences between regions. In general, grow- gy (the science of the making of wines) fall frosts), diurnal temperature ranges, ing season length and temperatures are aspects, which include regional associa- precipitation (especially during flowering critical aspects because of their major tions and cultural traditions, result in the and ripening stages), wind, and extreme influence on grape ripening and fruit defining wine style a region produces. weather events such as hail. Owing to quality, and therefore varietal adaptation Of all of the site factors, climate inadequate weather station locations, to a specific terroir. It is in its ideal cli- arguably exerts the most profound effect poor data sharing between countries, mate that a given grape variety can on the ability of a region or site to pro- and disagreement on what weather and achieve optimum ripening profiles of duce quality grapes. Worldwide, the aver- climate factors are most important, the sugar, acid, and flavour to maximize a age climatic conditions of a wine region climate component of terroir has been given style of wine and the vintage qual- determine to a large degree the grape studied largely through spatial and tem- ity. varieties that can be grown there, where- poral averaging, and by comparing new The growing season necessary as wine production and quality are wine region climates to analogous cli- for wine grapes varies from region to chiefly influenced by site-specific factors, mates in historical regions. region but averages approximately 170- husbandry decisions, and short-term cli- 190 days (Mullins et al., 1992). The mate variability (Jones and Hellman, Temperature and Growing Season length of the frost-free season is impor-

Figure 1. General geographical extent of the world’s main viticulture regions (adapted from de Blij, 1983). Contours represent the mean annual 10°C and 20°C isotherms as a proxy for the latitudinal limits of the majority of the world’s grape growing areas. Neither latitude nor isotherms fully accounts for the location or quality of individual vineyard regions, however. The solid dots represent the wine regions studied by Jones et al. (2005). GEOSCIENCE CANADA 3 tant to the onset of bud break, flower- 2000 degree-days. However, this index oped a simple classification of viticultur- ing, and the timing of harvest. Prescott has been found to be less appropriate al climates that uses five dimensions of (1965) noted that in most areas suitable for determining viticultural viability out- mean temperatures, continentality for grape production the mean tempera- side of California (Gladstones, 1992; (defined as the difference between the tures of the warmest and coldest Spellman, 1999; Jones and Davis, 2000). average mean temperature of the months are more than 18.9ºC and warmest and coolest months), sunlight -1.1ºC, respectively. Winter and spring Bioclimatic Indices hours, aridity (based upon the difference temperatures are also important in that Other bioclimatic indices have been between rainfall and evaporation), and frost and low temperatures can cause used to characterize a region’s potential relative humidity. Gladstones (1992) injury to grapevines. Research has also for viticulture and are mostly developed developed a classification similar to shown that there is a minimum winter on the basis of temperature. Various Amerine and Winkler (1944) but refined temperature that the grapevines can forms of a heliothermal index (Branas, it by imposing an upper limit on mean withstand. This minimum ranges from 1974; Huglin, 1978) and a latitude-tem- temperatures, a correction factor for lati- -5ºC to -20ºC, and is chiefly controlled perature index (Jackson and Cherry tude, and a correction for each month’s by microscale climate variations which in 1988; Kenny and Shao, 1992) have been temperature range. In a maturity classifi- turn reflect location and topography used to help define the suitability of a cation, Jones et al. (2005) related climate (Amerine et al., 1980; Winkler et al., region to the planting of certain grape and ripening potential for different vari- 1974). Temperatures below these thresh- varietals. Smart and Dry (1980) devel- eties based on average growing season olds will damage plant tissue by the rup- turing of cells, the denaturing of enzymes by dehydration, and the disrup- Table 1: Wine region average growing season temperatures as analyzed by Jones et tion of membrane function (Mullins et al. (2005) sorted into their respective climate maturity groupings as depicted in al., 1992). Another important influence Figure 2. on winter damage is vine health prior to Region Growing Seasona Climate winter dormancy. Diseases such as mois- Tavg (°C) Maturity Groupingb ture-induced downy mildew that ultimately cause defoliation and reduced Mosel Valley 13.0 sugar accumulation, can affect winter Alsace 13.1 Cool hardiness (Amerine et al., 1980). Champagne 14.5 Rhine Valley 14.9 Heat Summation Index N. Oregon 15.2 In France during the mid nineteenth Loire Valley 15.3 century, A.P. de Candolle observed that Burgundy-Côte 15.3 vine growth started when the mean daily Burgundy-Beaujolais 15.8 temperature reached 10ºC. This led to Chile 16.3 Intermediate the notion of a heat summation above a E. Washington 16.5 base temperature that could define vine Bordeaux 16.5 growth and grape maturation. Amerine C. Washington 16.6 and Winkler (1944) followed this con- Rioja 16.7 cept and developed a heat summation S. Oregon 16.9 index for California that is widely used as a guide for selecting appropriate grape C. California 17.0 varieties and for determining a given South Africa 17.1 area’s suitability to produce quality wine N. California 17.4 grapes. In the Northern Hemisphere the N. Rhône Valley 17.6 heat summation index is calculated for N. Portugal 17.7 Warm the period of April 1 through October Barolo 17.8 31 by summing each day’s average tem- S. Rhône Valley 18.2 perature above a base of 10ºC, the mini- Margaret River 18.6 mum temperature at which vine growth Chianti 18.8 occurs. Using the index, five climatic regions were defined, Winkler regions I- Hunter Valley 19.8 V, from which a minimum and maxi- Barossa Valley 19.9 Hot mum threshold of 950 and 2900 degree- S. Portugal 20.3 days, based on ºC respectively, was cal- S. California 20.4 culated for the cultivation of grapes. Note that the growing season average temperatures depicted here are derived from a 0.5° x 0.5° grid Although it varies by grape variety and and not from any one station [see Jones et al. (2005) for details]. a The growing season is Apr-Oct in the Northern Hemisphere and Oct-Apr in the Southern many different terroir factors, the best Hemisphere. quality wine typically is produced from b The climate maturity groupings are based upon the average growing season temperatures and the regions that experience between 1400- ability to ripen a given variety [see Figure 2 and Jones et al. (2005) for details]. 4 temperatures – cool, intermediate, sum of all influences in the terroir The parameters included plant effects warm, and hot climates – that were debate, Van Leeuwen et al. (2004) may that influence quality (i.e., phenology, derived from the benchmark climates of have come the closest to date. Studying shoot growth cessation, vine water sta- the world’s premium quality wine pro- numerous plots in the St. Émilion and tus, etc.) and yield components such as ducing regions (Table 1; Fig. 2). Pomeral regions of Bordeaux, France, fruit weight, sugar and acid levels, and the authors examined the effect of vin- phenolics (e.g., anthocyanin). Their Importance of Climate: the Bordeaux tage (climate), soil, and grape variety results showed that climate was the Example (Cabernet Sauvignon, Cabernet Franc, dominant factor, accounting for over While no one study can ever capture the and Merlot) on wine quality parameters. 50% of the variation when averaged across all quality parameters. Soil type and structure was the next most important factor, accounting for a quarter of the resulting wine quality. Varietal differences, while not as important as climate or soil, still accounted for 10% of the variation in quality parameters. While the study controlled for many cultural practices (e.g., pruning type and timing, trellising, and harvest timing), it would appear in this case that the cultural component of terroir roughly accounts for only 15-20% of the variation in important wine quality parameters. If results from this study are applicable elsewhere, it would appear that climate plays the most significant role in wine quality.

Climate Variability, Change, and Agricultural Responses It is clear from the observational record that global climates have changed over both short and long time scales. Global observations from the 20th century indicate that the mean surface temperature of the planet has increased by 0.6 ± 0.2ºC with land areas warming more than oceans (Houghton et al., 2001). In most cases observed atmospheric warming trends have been found to be asymmetric with respect to seasonal and diurnal cycles, with greatest warming occurring during the winter and spring and at night (Karl et al., 1993). Precipitation changes have been much more spatially variable, but show 5-10% increases over broad areas of the Northern Hemisphere while other regions have seen declines of 5-15% (IPCC, 2001). Some regions have also seen an increased frequency of heavy precipitation events, while other regions have seen moderate changes in drought frequency and severity. In addition, evi- Figure 2. Climate maturity groupings based on average growing season tempera- dence from paleoclimatological studies tures and the estimated span of varietal ripening potential that occurs within and has revealed large fluctuations in global across the groups. While V. vinifera varieties are not typically used for table grapes and regional climate over the past mil- and raisins, they are included for comparison of warm to hot climate wine grape lion years. Furthermore, there are indica- production. Note that the average growing season temperatures are depicted in tions that over the last 2000-3000 years Table 1 and are derived from grids, not station data; therefore the values given may climates have shifted from cooler to deviate slightly from any one station in a given region (Jones et al., 2005). warmer periods, with the most recent GEOSCIENCE CANADA 5 century potentially being the warmest in water use efficiency in plants in the pursuit. Experience over thousands of the last millennium (Mann et al., 1999). absence of other climate-related changes years has resulted in grapes being grown Our understanding of the potential (Rosenberg, 1981a,b). Therefore, more and wine being produced in mostly mid- range of climate’s natural variation is still carbon dioxide can potentially lead to latitude regions that are roughly defined somewhat limited, however. greater production. However, many by their ability to ripen fruit to produce Contemporary global climate observational and modeling studies on certain wine styles (Fig. 1). Although change due to increased levels of atmos- the effects of increases in CO2 on vari- many new world wine regions are rela- pheric CO2 has received considerable ous plant systems (Reekie and Bazzaz, tively young, long-term historical records attention (Climate Research Board, 1979; 1991; Hänninen, 1995; Maytin et al., of European viticulture have been main- CO2/Climate Review Panel, 1982; 1995) indicate that increased carbon tained for nearly a thousand years. Carbon Dioxide Assessment Committee, uptake is partitioned, with more entering Records of harvest dates and yield were 1983; IPCC 2001; and others). The vegetative tissues and less entering initially maintained by monks during the effects have been observed in the gener- reproductive organs, which could spell a Middle Ages and later by merchants and al climate record (e.g., Jones et al., 1986) reduction in grain and fruit quality. More the prominent châteaux. These records and simulated through a variety of gen- complete research into changes in the indicate that the region has experienced eral circulation models (GCMs; see Cess quality of plant tissue and fruit along wide fluctuations of climate and viticul- et al., 1990; and Gates et al., 1998 for with better understanding of how other tural productivity in the past (e.g. good reviews). The observed trends in impacts (e.g., concomitant changes in Penning-Roswell, 1989). Gladstones temperatures have been related to agri- temperature) from CO2–induced climate (1992) depicted climatic variability in cultural production viability through change is needed. Europe from historical records and their impact on winter hardening poten- Agricultural productivity varies details that, during the medieval ‘Little tial, frost occurrence, and growing sea- with climate variations both between Optimum,’ temperatures were about 1ºC son lengths (Menzel and Fabian, 1999; and within years. The observed human above the present day and the warmth Carter et al., 1991; Easterling et al., 2000; activities that drive changes in atmos- lasted from about the eighth century to Nemani et al., 2001; Moonen et al., 2002; pheric composition and climate could the early fourteenth century. During this Jones, 2005). On the other hand, GCM therefore be expected to alter, perhaps time, vineyards were planted over most simulations – although far from perfect significantly, the relative levels of agri- of southern England and along the representations of climate reality – all cultural productivity in different regions coasts of the North and Baltic Seas. indicate that climate change will contin- of the globe. Agriculture systems typi- Subsequently the ‘Little Ice Age’ ue to occur in some manner owing to cally are chosen and developed based on brought an abrupt fall in temperatures anthropogenic influences. While the ini- the mean climatic conditions of a region and a prolonged cold period that lasted tial magnitude of climate change, as sim- and its variability, but it has been sug- from the early fourteenth century ulated by GCMs, has been reduced due gested that variations in the probability through the early 1800s. While there to improved parameterization (IPCC, and duration of extreme events are were some large fluctuations in weather 2001), climate change at any magnitude more important than changes in the during this period, the overall climate or in any direction will still have an mean (Waggoner, 1989; Katz and regime was extremely rough on all effect on agricultural productivity. The Brown, 1992). Extreme events including aspects of agriculture in Europe with overall impacts of climate change on floods, high temperatures, frost, failed crops and famine for many con- agriculture ultimately will depend on drought, hail, rainfall, and wind all affect secutive years. Vineyards throughout the plant physiological requirements and the agriculture in some way. If extreme British Isles and northern Europe died spatial variations, seasonality, and magni- events increase in frequency, plant sys- out and harvests did not occur for many tude of the warming (McCarthy et al., tems would tend to fail to recover over years in southern Europe and the 2001; Butterfield et al., 2000). the short term and/or adjust over the Mediterranean. Temperatures during the Although the forecast of tem- long term without some form of stress coldest decades of the Little Ice Age perature increases for individual regions and damage occurring. In addition, as an were about 2ºC lower than the warm is under debate owing to the scale prob- agricultural system becomes less periods of the Middle Ages, and at least lem associated with GCM projections resilient, it is more susceptible to 1ºC lower than today (Le Roy Ladurie, (Giorgi and Mearns, 1991; Jones et al., increased outbreaks in disease and pests. 1971). 1995), there is one area of no debate - Therefore, climatic variability and its Pfister (1988) used the recorded the atmospheric concentration of car- relation to climate change will play an dates of harvests and other vine devel- bon dioxide will continue to rise for at important role in the long and short- opmental stages to study the direct least a few decades to come (Climate term structure and yields of agriculture effects of climate variability on viticul- Research Board, 1979; CO2/Climate systems, including the growing of wine ture in Europe from the Middle Ages to Review Panel, 1982; Carbon Dioxide grapes. 1860. He inferred that temperatures dur- Assessment Committee, 1983; IPCC ing the growing season in the High 2001; and others). Under optimum Past Climate Variability, Change, and Middle Ages must have averaged 1.7ºC growing conditions, the atmospheric Viticulture: the European Example warmer than today, and that harvest concentration of CO2 is the limiting fac- Viticulture and enology combine to pro- dates began around the first of tor in photosynthesis, and increasing duce a very geographically defined and September compared to early to mid CO2 concentrations will increase the therefore climatically distinct agricultural October today. These grape harvest 6 records correlated well with other long- Analyzing a nearly 50-year record of decreases in summer vapor pressure term records of climate change as evi- grape and wine data for Bordeaux deficits. If the current trends in frost denced in glacial advances and retreats, (1949-1997), Jones and Davis (2000) frequency continue, Napa/Sonoma ice core analyses, palynological studies, found trends towards earlier phenologi- could become a frost-free climate in varve chronologies, and den- cal events (bud break, flowering, vérai- another decade or two. In a larger study drochronologies. Carrying the viticulture son, and harvest), shorter intervals covering the entire west coast of the record even further into the twentieth between events, and a lengthening of United States, Jones (2005) documented century, Ray (1985) compiled the dates the growing season. This study also climate variability during 1948-2002 for of the start of vintage harvest for a sin- found that the two main varieties grown the principal grape growing regions in gle château in Bordeaux (Lafite). While in Bordeaux, Merlot and Cabernet the states of California, Oregon, and he noted that some of the variation is Sauvignon, exhibited larger berry Washington. This research showed that due to biotic influences (mildew, fungus, weights and higher sugar to total acid on average, most regions have experi- pests, diseases, etc.), the dates corre- ratios, which resulted in higher vintage enced a decline in frost frequency, earlier spond very well with land-based clima- ratings and less year-to-year variability in last spring frosts, later first fall frosts, tologies of temperature (Jones et al., quality. Warmer temperatures during the longer frost-free periods, and warmer 1986). In addition, recent research by growing season and a reduction of growing seasons with greater heat accu- Chuine et al. (2004) and colleagues used ripening period rainfall were found to be mulation. contemporary grape harvest dates from the most significant climate factors in Burgundy to reconstruct spring-summer Bordeaux phenology, composition, and Climate Variability Impact: Large temperatures from 1370 to 2003 and, wine quality. Wine production trends in Scale while the results indicate that tempera- Bordeaux were found to be more vari- Examining larger scale influences of cli- tures as high as those reached in the able than phenology, composition, and mate variability, Rodó and Comín (2000) warm 1990s have occurred several times quality; however, the study revealed that detailed how the North Atlantic in the region since 1370, the extremely rainfall during the physiologically impor- Oscillation (NAO; a north-south seesaw warm summer of 2003 appears to have tant stages of flowering and maturation in the pressure fields of the North been higher than in any other year since tended to decrease production (Jones Atlantic) and the El Niño-Southern 1370. and Davis, 2000). Similar results were Oscillation (ENSO; a combined east- found for California’s Napa and Sonoma west seesaw in sea surface temperatures Climate Variability and Climate Valleys where, since the 1950s, grape and pressure fields in the tropical Change Impacts on Wine growers have seen dramatic increases in Pacific) impacted grape production and Recent research on the impacts of cli- premium wine quality (along with wine quality in Spain. Acting as a climate mate on wine factors has followed two decreased year to year variability in quali- “teleconnection”, where the impact of lines of analysis: 1) how short-term cli- ty), grape yield, and crop value (Nemani an atmospheric variation in one region is mate variability and trends have impact- et al., 2001). This research found that the conveyed to other regions; El Niño ed grapevine phenology, production, and climate of the region has experienced events (warmer than average sea surface quality (e.g., Braslavska, 2000; Jones and greater warming at night and during temperatures in the eastern tropical Davis, 2000; Esteves and Orgaz, 2001), spring. As a consequence of the asym- Pacific) were found to have significant and 2) how future climate change may metric warming, the diurnal temperature impact on wine quality, providing more potentially impact the spatial viability of range – the difference between daily rain during the growing season in Spain wine growing regions, vine and grape maximum and minimum temperatures – over the last thirty years. The NAO, on growth potential, and overall wine quali- declined by 1.9ºC over 47 years. the other hand, did not impact wine ty and styles (Kenny and Harrison, 1992; Although the average annual warming quality even though the condition is Bindi et al., 1996; Jones et al., 2005). trend was a modest 1.1ºC over the 47 closer to the wine producing regions of Owing to the fact that many viticulture years, there was a 71% decline in frost Spain. Esteves and Orgaz (2001) found climates are dry during the growing sea- frequency (20-day reduction) and a 25% similar results for the Viseu wine region son, most of the research conducted to increase in the length of the frost-free of Portugal. The authors conducted a date has focused on temperature-related growing season (65-day increase). The spectral analysis between wine quality impacts. decline in frost frequency was signifi- and teleconnection indices (ENSO and cantly correlated with the increase in NAO) and determined that wine quality Climate Variability Impact: Small vintage ratings. A possible explanation in Portugal follows a 3 to 7 year cycle Scale for such a relation could be that frosts that is similar to that of the ENSO Climate variability impacts agriculture may damage the primary buds on the cycle. Whereas the NAO had some and, in particular, grape and wine pro- vine, leaving secondary or tertiary buds impact on Portuguese wine quality, its duction are typically related to site-spe- that are less fruitful, along with delaying effect was much less than that of cific extreme events such as frost, hail, subsequent plant physiology and ulti- ENSO. Jones (1997) also found that the and heavy rain, or to broader scale inter- mately leading to uneven maturity and NAO was insignificant for Bordeaux annual cycles of warmer, colder, wetter, poor wine quality. Additionally, for the phenology, grape composition, yields, or drier conditions brought about by same time period, grape yields grew 34% and wine quality. The authors of each of recurring modes of atmospheric circula- and were significantly influenced by the papers noted above speculate that tion and/or sea surface temperatures. minimum temperatures in spring and the best explanation for these observa- GEOSCIENCE CANADA 7 tions is due to the fact that the NAO is these authors showed that the length of resulted in a composite 23 day reduction a dominant winter mode of circulation the growing season should expand over in the interval from bud break to har- variability, whereas ENSO can influence all of Europe with precipitation increas- vest, and that doubled concentrations of moisture levels throughout the year ing in the north and decreasing in the CO2 in the atmosphere resulted in a depending on its strength and length of south. Using mean climate values and 36% increase in yield. However, the activity. For the wine regions of the west Broadbent’s (1981) compilation of vin- combined effects of increases in tem- coast of the United States, Jones (2005) tage ratings for Bordeaux and perature and CO2 resulted in an increase analyzed short-term climate variability in Champagne, the authors noted that the in yield variability with the potential to relation to the Pacific Decadal climate variables explained 58 percent create greater economic risk. Pincus Oscillation (PDO, a measure of the and 63 percent of the vintage ratings, (2003) provided insight into the connec- dominant variability in sea surface tem- respectively. Their warmer world scenar- tions between wine, place, and identity – peratures in the North Pacific) and ios suggested that conditions would lead the heart of terroir – by examining how ENSO, finding that the PDO was the to improvements in wine quality, particu- climate change may make the “associa- most dominant influence on important larly in Bordeaux. From recent analyses tions between wine and place difficult or viticulture climate parameters. However, by Jones et al. (2005), where vintage rat- impossible to maintain.” Examining cli- neither the PDO nor ENSO was signifi- ings for Bordeaux and Champagne have mate projections to 2025 and discussing cantly related to vintage ratings for Napa trended higher with less year-to-year the results with grape growers and wine (Jones, 2005). The geographical location variability, the work of Lough et al. makers in various regions, Pincus (2003) of Australian and Chilean wine produc- (1983) appears to have been generally found that some form of adaptation – ing regions places each near one of the correct for these regions. including changing varieties, vineyard main centers of action of ENSO. For Tate (2001) discussed the general management, and regional identities – Australia, ENSO is the main contributor impact of future climate change on must occur for the industry to survive. of rainfall variability and El Niño years grape production and wine quality. Examining projected climate change sce- often result in moderate to severe Although no data were analyzed, this narios for 2030 and 2070 for Australian drought. A similar but opposite situation author described the potential conse- viticulture, McInnes et al. (2003) found is experienced in Chile where greater quences of climate change as affecting that temperatures are predicted to warm than average rainfall occurs during El where grapes can be grown (“ideal” by 1.0-6.0ºC by 2070, thus increasing the Niño years. Although direct studies of locations today ceasing to be so in the number of hot days and decreasing frost the impacts of ENSO on wine produc- future and more poleward locations risk, while precipitation changes are pre- tion and quality in Chile and Australia becoming viable), changing the distribu- dicted to be more variable, potentially are incomplete, climate modeling indi- tion and intensity of pests and diseases, resulting in greater growing season stress cates that ENSO conditions are expect- causing sea level changes that could alter on irrigation. These authors indicate that ed to increase in frequency and severity coastal climates, and increasing CO2 the challenges facing the Australian wine with climate change. Finally, the authors such that it might impact grape compo- industry include more rapid phenological in the studies noted above state that nents or the texture of oak wood used development, changes in suitable loca- since many of the circulation modes can in barrels. In another overview of tions for some varieties, a reduction in be assessed months ahead of a given potential climate change impacts, Schultz the optimum harvest window for high growing season, there may be some (2000) discussed how shifts in precipita- quality wines, excessive vegetative potential for seasonal prediction that tion would greatly affect the dry summer growth due to increased CO2 levels, and would allow a greater latitude of plan- regimes of most high quality wine greater management of already scarce ning for the impacts noted above. regions. In regions where irrigation prac- water resources. Clearly, more research needs to be done tices are controlled or completely forbid- In a multi-region analysis of the in the area of climate variability modes den (i.e., many parts of France) the ten- impacts of climate change on wine qual- and their impact on viticulture and wine dency to drier summers could be detri- ity, Jones et al. (2005) analyzed growing quality for other areas of the world. mental to yields and flavor profiles. season temperatures in 27 of arguably Schultz (2000) also detailed how changes the best wine producing regions in the Climate Change Impacts in CO2, temperature, and solar radiation world, in terms of being highly recog- Lough et al. (1983) examined scenarios likely would have direct impacts on nized for quality (Fig. 1). The authors of a warmer world in Europe using the yields, grape composition, and flavor used average growing season tempera- noted 20th century warming as an ana- development. tures as these values typically define the log. They compared thermal, moisture, Addressing the climate change climate-maturity ripening potential for and pressure changes over the two effects for a specific area and grape vari- varieties grown in cool, intermediate, coolest and warmest twenty year periods eties, Bindi et al. (1996) studied the warm, and hot climates (Fig. 2; Table 1). during the 1900s, and used the results to effects of increased CO2 levels and the For example, Cabernet Sauvignon is construct scenarios of the impacts of associated changes in climate on grown in regions that span intermediate future climate change on agriculture and Cabernet Sauvignon and Sangiovese to hot climates with growing season energy usage. While noting the difficulty grapes in Northern Italy. Using field averages that range from roughly 16.5- in assessing the consequences of tech- data from 1992-1994 and a model of 19.5ºC (e.g., Bordeaux or Napa). Results nological advances and the “fertilizer grapevine growth and yield, they found from the analysis of 1950-1999 revealed effect” of increasing CO2 concentration, that projected temperature increases that 17 of the 27 wine regions experi- 8 enced statistically significant warming Napa regions warmed by 1.8ºC and greater growing season temperature vari- during their respective growing seasons. 1.2ºC from 1950-1999, respectively. Also ability. In the study, a large majority of Figure 3 provides an example of the note that Napa has a warmer average the U.S. and European wine regions saw warming where both the Bordeaux and growing season and that Bordeaux has significant temperature increases whereas the majority of the Southern Hemisphere locations changes were not significant. Averaged across all wine regions with significant trends, the warming was 1.3ºC. The most dramatic of these changes, confirmed by another observation-based climatology (Moisselin et al., 2002), occurred in the northern Rhone Valley of France where the growing season warmed by 4.1ºC. Jones et al. (2005) also examined the relationships between average growing season temperatures and vintage ratings as given by Sotheby’s and the Wine Enthusiast (Stevenson, 2002; Mazur, 2002). Ratings are commonly used to compare vintages and have a have strong influence on the economic success of a wine producing region (de Blij, 1983). Although vintage ratings are inherently subjective, the correlations between different rating systems typically are very high (r > 0.9; Jones, 1997) indicating their usefulness as a wine quality metric. Furthermore, in the absence of a worldwide systematic, consistent, and readily available set of compositional data, vintage ratings currently provide the best metric for global scale analyses of wine quality. Vintage ratings do have limita- tions, however, as they are often averaged over large regions, effectively mask- ing inter-region variations, and not all ratings systems cover all regions, nor have all regions received the publicity necessary to be more thoroughly rated. The Sotheby’s vintage ratings data analyzed by Jones et al. (2005) represent one of the most complete rating systems and, when combined with the Wine Enthusiast ratings for South Africa and Chile (Mazur, 220), cover 27 regions and 30 categories of wine over varying time periods during 1967-2000 (Table 2; Fig. 1). Some regions are divided into sub- regions or wine styles with separate ratings, whereas others are simply divided into ratings for red and white wines. For 25 of the 30 wine regions or categories Figure 3. Observed growing season average temperature anomalies for a) Bordeaux, of wine, vintage ratings have shown France, and b) Napa, California, as analyzed by Jones et al. (2005). The temperature trends of increasing overall quality with data are monthly values extracted from a 0.5° x 0.5° grid centered over the wine less vintage-to-vintage variation. As an producing regions for 1950-1999. Tavg is the average growing season temperature example, Figure 4 depicts the vintage (April-October in the Northern Hemisphere and October-April in the Southern ratings for 1963-2000 for red wines Hemisphere). Tstd is the standard deviation of monthly temperatures during the from the Médoc and Graves region of growing season, and the Trend is over the 50-year period. Bordeaux, and red wines from GEOSCIENCE CANADA 9

California: a general trend over time to ing rates and magnitudes detailed by the ability to ripen similar varieties. Note better quality, with less variability, can be Jones et al. (2005) for the next ~half that a wine region, on average, can be seen. Also note the greater relative vari- century have numerous potential positioned within the range of the cli- ability in Bordeaux vintages as compared impacts on grapevine growth, grape pro- mate maturity types based on its average to California, indicating the impact of duction, and wine production in the growing season average temperature greater growing season climate variability future. The impacts are not likely to be (Fig. 2). For example, if a region has an as seen in Figure 3. Examining the rela- uniform across all varieties and regions, average growing season temperature of tionships between climate and wine but are more likely to be related to a cli- 15ºC and the climate warms by 1ºC, that quality found that the majority of the matic threshold whereby any continued region is climatically more conducive to trends in vintage ratings were significant- warming would push a region outside ripening some varieties, while potentially ly related to average growing season temperatures. Whereas the effect varied Table 2: Wine regions and categories of wine analyzed by Jones et al. (2005). The by region, vintage ratings on average wine regions correspond to the locations shown in Figure 1. rose by 13.3 points (on a 100 point scale) for every 1ºC warmer the growing season. From 10-62% of the variation in Region Categories of Wines in Sotheby’s vintage ratings (32% average) was related Vintage Ratings to growing season temperature variations, with the most significant results C. Washington being found in the cooler climate E. Washington US - Pacific Northwest Red regions (e.g., the Mosel and Rhine Valley N. Oregon US - Pacific Northwest White of Germany). S. Oregon To examine future climate N. California change, Jones et al. (2005) used output US - California Red C. California US - California White from the HadCM3 general circulation S. California model (GCM; Gordon et al., 2000; Pope et al., 2000) from 1950-2049 for 25 grid N. Portugal Vintage Port cells encompassing the same wine S. Portugal No Specific Rating Provided regions as described above. A compari- Rioja Rioja Red son of the two periods, 1950-1999 and Barolo Barolo Red 2000-2049, suggested that mean growing season temperatures would warm by an Chianti Chianti Red average 1.2ºC over the 27 wine regions Rhine Valley Rhine Valley White studied with the differences for Mosel Valley Mosel-Saar-Ruwer Valley White Bordeaux and Napa 1.2ºC and 1.7ºC, respectively (Fig. 5). The projected N. Rhône Valley N. Rhône Valley Red changes are greater for the Northern S. Rhône Valley S. Rhône Valley Red Hemisphere (1.3ºC) than the Southern Loire Valley Loire Valley Red Hemisphere (0.9ºC). Examining the rate Loire Valley Sweet White of change projected for 2000-2049 revealed significant changes in each wine Alsace Alsace White region with trends ranging from 0.2ºC to Champagne Vintage Champagne 0.6ºC per decade. Overall changes dur- Burgundy - Côte D’Or Red ing the 2000-2049 period averaged 2.0ºC Burgundy-Côte Burgundy-Beaujolais Burgundy - Côte D’Or White across all regions with the smallest Burgundy - Beaujolais Red warming in South Africa (0.9ºC/50 years) and greatest warming in Portugal Bordeaux - Médoc and Graves (2.9ºC/50 years). For the Napa and Bordeaux Bordeaux - St. Émilion and Pomeral Bordeaux regions, decadal trends are 0.4 Bordeaux - Sauternes and Barsac and 0.5ºC while the overall change is 2.2 Hunter Valley Hunter Valley Red and 2.3ºC, respectively (Fig. 5). In addi- Hunter Valley White tion, the HadCM3 model predicted sig- Margaret River Margaret River Red nificant increases in growing season Margaret River White temperature variability and warming during the dormant season across most Barossa Valley Barossa Valley Red regions. Barossa Valley White While the observed warming of South Africaa Overall Vintage the last fifty years appears to have most- Chilea Overall Vintage ly benefited the quality of wine grown a Rating data for South Africa and Chile are from a different source than the other locations (see worldwide, the predicted regional warm- text for details). 10 less conducive to ripening for others. If varieties that a region can ripen will evidenced mostly through a more rapid the magnitude of the warming is 2ºC or expand in many cases, if a region is a plant growth and out-of-balance ripen- larger, a region may potentially shift into hot climate maturity type and warms ing profiles. If a region currently experi- another climate maturity type (e.g., from beyond what is considered viable, grape ences a maturation period (véraison to intermediate to warm, using the climate growing becomes challenging and may harvest) that allows sugars to accumu- scale of Jones et al. 2005 – cool, inter- even be impossible. The wine quality late, maintains acid levels, and produces mediate, warm, and hot – as noted issues related to climate change and the optimum flavor profile for that vari- above). While the range of potential shifts in climate maturity potential are ety, balanced wines result. In a warmer than ideal environment, the grapevine will go through its phenological events more rapidly, resulting in earlier sugar ripeness and, while the grower or winemaker is waiting for flavors to develop, the acidity is lost through respiration resulting in “flabbly” wines (high alcohol with little acidity for freshness). In addition, harvests that occur earlier in the summer, in a warmer part of the growing season (e.g., September instead of October in the Northern Hemisphere), will result in hot and potentially desiccat- ed fruit without greater irrigation inputs.

CONCLUSIONS Cultivation of grapevines for wine production has a rich geographical and cultural history of development. This history has helped shape the notion of terroir, which describes the interrelation- ships among climate, soil, landscape, and the people who cultivate the grapevine for the production of wine. While there is much debate as to what is the most important aspect of terroir (e.g., Seguin, 1986; and others), environmental change has the potential to impact the balance that exists in any region. Today’s viticultural regions for premium quality wine production are located in narrow climatic zones that put them at particular risk from both short term climate variability and long term climate change. This was clearly evident during the Little Ice Age when viticultural viability was threatened throughout much of Europe. The warming of the last century or more appears to have largely benefited grape growing and wine production through the expansion of viable growing regions, providing longer growing seasons, earlier phenological development, and more optimum ripening leading to better overall quality. It is clear that advances in viticultural practices such as irrigation, nutrition, trellising, and pest/disease control, and more knowledge and expe- Figure 4. Vintage ratings for (a) red wines from the Médoc and Graves regions of rience in winemaking techniques, cer- Bordeaux and (b) red wines from California as analyzed by Jones et al. (2005). The tainly have contributed to larger yields ratings are from Sotheby’s (Stevenson, 2001) and are based on a 0-100 scale. A and better quality. In spite of such LOWESS filter is applied to indicate the underlying pattern in the ratings. advances, however, grape growers and GEOSCIENCE CANADA 11 winemakers generally believe climate varieties come to maturity too early in plex owing to the interactive effects with plays the most significant role in deter- the season and the typical balance changes in temperature and moisture mining the overall quality and style of between sugar and acids cannot be availability. Observations and modeling wine from a given region, and that year- achieved. This suggests increasing indicate that photosynthesis and water- to-year variations in the quantity and potential economic risks for grape grow- use efficiency (ratio of photosynthesis to quality of vintages are controlled by cli- ers and winemakers (Bindi and Fibbi, water consumption) is stimulated by mate variability. 2000). Changes in growth and quality increased CO2 and that grapevine pro- Future climate change issues for due to increases in CO2 are more com- duction should increase without causing the wine industry are mostly related to changes in temperature, precipitation, and CO2 concentrations and their interactive effects on the spatial viability of wine producing regions, plant growth timing, production, and quality. The climate change scenarios predicted for grape growing regions point to the potential for regional changes in viticultural viability owing to changes in growing season temperatures and precipitation seasonality and distribution (Pincus, 2003; McInnes et al., 2003; Jones et al., 2005). Regions with cool growing season temperatures today (e.g., the Mosel and Rhine Valleys of Germany – more poleward locations) theoretically would have reduced year-to-year vintage quality variations and potentially could ripen warmer climate varieties (Schultz, 2000). Other regions, currently with warmer growing seasons (e.g., the Iberian Peninsula and Chianti – more subtropical regions) may become too warm for the existing varieties grown there, while the hottest regions may lose production viability altogether. While growing season changes clearly are important, projected dormant period temperature changes would also affect viticulture by reducing winter freeze damage in some regions (e.g., eastern Washington state), while other regions (e.g., parts of California and Australia) would have very mild winters where the hardening of latent buds may not occur and pests limited by winter minimum temperatures may increase in number or severity. Although grape growing typically requires less water demand than many other crop systems, changes in seasonal- ly dependent snowmelt or rainfall could also place added stress on vines in water-limited regions. Given the Figure 5. Modeled growing season average temperature anomalies for a) Bordeaux observed and modeled acceleration of and b) Napa as analyzed by Jones et al. (2005). The modeled temperature data are vegetative and reproductive growth of from the HadCM3 climate model on a monthly time scale extracted from a 2.5° x grapevines in a warmer climate, a gener- 3.75° grid centered over the wine producing regions for 2000-2049. The anomalies al trend of increased yields and higher are referenced to the 1950-1999 base period from the HadCM3 model. Note that the sugar contents is predicted for several difference between the 1950-1999 growing season average temperature in Figure 3 growing regions and varieties. However, and the 1950-1999 growing season average temperature shown here are due to the thresholds may be reached by which larger size of the grid square used in the HadCM3 model. Trend values are given as grape quality could be jeopardized when an average decadal change and the total change over the 50-year period. 12 negative influences on the quality of Ough, C. S., Singleton, V. L. and Webb, Dioxide and Climate: A Scientific grapes and wine (Bindi et al., 2001). A. D., 1980, The Technology of Wine Assessment (U.S. National Research In most of the wine producing Making, (4th ed.): AVI Publishing Council): National Academy Press, regions of the world, especially those Company, Inc., Westport, Connecticut. Washington, D.C. 22 p. with the longest history, both physical 795 p. CO2/Climate Review Panel, 1982, Carbon Bindi, M., Fibbi, L., Gozzini, B., Orlandini, Dioxide and Climate: A Second and cultural landscapes and local S., and Miglietta, F., 1996, Modeling the Assessment (U.S. National Research economies are shaped by wine produc- Impact of Future Climate Scenarios on Council): National Academy Press, tion. In these regions the wine industry Yield and Variability of Grapevine: Washington, D.C. 72 p. drives regional development and domi- Climate Research, v. 7, p. 213-224. de Blij, H. J., 1983, Geography of viticulture: nates many economic sectors from pro- Bindi, M. and Fibbi, L., 2000, Modeling cli- rationale and resource: Journal of duction to trade to tourism. While much mate change impacts at the site scale on Geography, v. 82, n. 3 p. 112-121. uncertainty still exists in the magnitude grapevine: in Downing, T.E, and Easterling, D. R., Evans, J. L., Groisman, P. and rate of climate change, any change Harrison, P.A., Butterfield, R.A., and Ya., Karl, T. R., Kunkel, K. E., and is likely to bring about cultural change Longsdale, K.G., (eds.): Climate Change, Ambenje, P., 2000, Observed variability where regional identities may shift with Climatic variability and Agriculture in and trends in extreme climate events: A the varieties and wine styles that can be Europe: An Integrated Assessment. Final brief review: Bulletin of the American Report: Environmental Change Institute, Meteorological Society. v. 81, p. 417-425. produced there. To prepare for the University of Oxford, p. 117-134. Esteves, M. A. and Orgaz, M. D., 2001, The future, the industry will most certainly Bindi, M., Fibbi, L., and Miglietta, F., 2001, influence of climatic variability on the need to integrate planning and adapta- Free Air CO2 Enrichment (FACE) of quality of wine: International Journal of tion strategies to adjust accordingly to grapevine (Vitis vinifera L.): II. Growth Biometeorology, v. 45, n. 1, p. 13-21. the predicted changes in climate. and quality of grape and wine in Gates, W. L., et al., 1998, An Overview of response to elevated CO2 concentrations: the Results of the Atmospheric Model ACKNOWLEDGEMENTS European Journal of Agronomy, v. 14, n. Intercomparison Project (AMIP I): I would like to thank the many 2, p. 145-155. Bulletin of the American Meteorological researchers from various fields of study Branas, J., 1974, Viticulture: Dehan, Society, v. 73, p. 1962-1970. who have delved into the notion of ter- Montpellier, 990 p. Giorgi, F. and Mearns, L. O., 1991, Braslavska, O., 2000, Trends in the grapevine Approaches to the simulation of regional roir and those that have studied climate- phenology - Slovakia, Dolne Plachtince, climate change: A review: Reviews of related impacts on the wine industry. period 1971- 2000: Slovak Geophysics. v. 29, p. 191-192. The data in the Jones et al. (2005) analy- Hydrometeorological Institute, Národny Gladstones, J., 1992, Viticulture and sis were provided by the Climate Klimaticky Program SR, v. 2000, n. 8, p. Environment: Winetitles, Adelaide, Impacts LINK Project (DEFRA 69-77. Australia. 310 p. Contract EPG 1/1/124) on behalf of Broadbent, M., 1981, The Great Vintage Gordon, C., Cooper, C., Senior, C. A., the Hadley Centre and U.K. Wine Book: Sotheby’s, London, 432 p. Banks, H., Gregory, J. M., Johns, T. C., Meteorological Office; and the Center Butterfield, R. E., Gawith, M. J., Harrison, P. Mitchell, J. F. B., and Wood, R. A., 2000, for Climatic Research, Department of A., Lonsdale, K. J. and Orr, J., 2000, The simulation of SST, sea ice extents Geography, University of Delaware. Modelling climate change impacts on and ocean heat transports in a version of Tom Stevenson, author of the New wheat, potato and grapevine in Great the Hadley Centre coupled model with- Britain: in Downing, T. E., Harrison, P. out flux adjustments: Climate Dynamics, Sotheby’s Wine Encyclopedia: A A., Butterfield, R. E. and Lonsdale, K. v. 16, p. 147-168. Comprehensive Reference Guide to the G., (eds.): Climate Change, Climatic Hänninen, H., 1995, Effects of climatic Wines of the World, is thanked for his Variability and Agriculture in Europe: An change on trees from cool and temperate insights and discussion on wine ratings. Integrated Assessment. Final Report: regions: an ecophysiological approach to This paper was critically read by L.D. Environmental Change Institute, modeling of bud burst phenology: Meinert and an anonymous reviewer: University of Oxford, 411 p. Canadian Journal of Botany, v. 73, p. comments from these individuals and Carbon Dioxide Assessment Committee, 183-199. from R.W. Macqueen improved the pres- 1983, Changing Climate (U.S. National Haynes, S.J., 1999, Geology and Wine 1. entation and are gratefully acknowl- Research Council): National Academy Concept of terroir and the role of geol- edged. Finally, thanks are due Simon Press, Washington, D.C., 496 p. ogy: Geoscience Canada, v. 26, p. 190- Carter, T. R., Parry, M. L. and Porter, J. H., 194. Haynes, whose interest in terroir and 1991, Climatic change and future agrocli- Houghton J.T. et al., 2001, Climate Change research provided the arena for matic potential in Europe: International 2001: The Scientific Basis, Contribution Geoscience Canada’s Geology and Wine Journal of Climatology. v. 11, p. 251-269. of the Working Group 1 to the Third series and the Geological Society of Cess, R. D. et al., 1990, Intercomparison and Assessment of the Intergovernmental America’s 2003 annual meeting tribute interpretation of climate feedback Panel on Climate Change: Cambridge session on terroir. processes in 19 atmospheric general cir- University Press, Cambridge, U.K., 944 culation models: Journal of Geophysical p. REFERENCES Research, v. 95, p. 16601-16615. Huglin, P., 1978, Nouveau mode d’evaluation Amerine, M. A. and Winkler, A. J., 1944, Chuine, I., Yiou, P., Viovy, N., Seguin, B., des possibilities heliothermiques d’un Composition and quality of musts and Daux, V., and Le Roy Ladurie, E., 2004, milieu viticole: C.R. Academy of wines of California grapes: Hilgardia, v. Grape Ripening as a Past Climate Agriculture in France (n. 111726). 15, p. 493-675. Indicator: Nature, v. 432, p. 289-290. Jackson, D. I. and Cherry, N. J., 1988, Amerine, M. A., Berg, H. W., Kunkee, R. E., Climate Research Board, 1979, Carbon Prediction of a district’s grape-ripening GEOSCIENCE CANADA 13

capacity using a latitude-temperature Horticultural Science, v. 67, n. 2, p. 239- Coughlan, J.C., 2001, Asymmetric climat- index: American Journal of Enology and 246. ic warming improves California vintages: Viticulture. v. 39, n, 1, p. 19-28. Le Roy Ladurie, E., 1971, Times of Feast, Climate Research, v. 19, n. 1, p. 25-34. Jones, G. V. and Davis, R. E., 2000, Climate Times of Famine: A History of Climate Penning-Roswell, E., 1989, Wines of influences on grapevine phenology, Since the Year 1000: Doubleday & Bordeaux, (6th ed.): Penguin Books, grape composition, and wine production Company, Inc., Garden City, New York. London/New York. 610 p. and quality for Bordeaux, France: 426 p. Pfister, C., 1988, Variations in the spring- American Journal of Enology and Lough, J. M., Wigley, T. M. L. and Palutikof, summer climate of central Europe from Viticulture, v. 51, n. 3, p. 249-261. J. P., 1983, Climate and climate impact the High Middle Ages to 1850, in Jones, G. V. and Hellman, E., 2003, Site scenarios for Europe in a warmer world: Wanner, H., and Siegenthaler, U., (eds.), Assessment: in Hellman, E. (ed.), Journal of Climate and Applied Long and Short Term Variability of “Oregon Viticulture” 5th Edition: Meteorology, v. 22, p. 1673-1684. Climate: Springer-Verlag, Berlin, p. 57-82. Oregon State University Press, Corvallis, M. Mazur, Wine Enthusiast’s 2002 Vintage Pincus, R., 2003, Wine, Place, and Identity in Oregon, p. 44-50. Chart. The Wine Enthusiast Magazine a Changing Climate: Gastronomica, v. 3, Jones, G. V., 1997, A Synoptic Climatological (http://www.winemag.com/vintage.cfm). n. 2, p. 87-93. Assessment of Viticultural Phenology: Mann, M. E., Bradley, R. S. and Hughes, M. Pope, V. D., Gallani, M. L., Rowntree, P. R. Dissertation, University of Virginia, K., 1999, Northern Hemisphere temper- and Stratton, R. A., 2000, The impact of Department of Environmental Sciences, atures during the past millennium: new physical parametrizations in the 394 p. Inferences, uncertainties, and limitation: Hadley Centre climate model — Jones, G. V., White, M. A., Cooper, O. R., Geophysical Research Letters, v. 26, p. HadAM3: Climate Dynamics, v. 16, p. and Storchmann, K., 2005 (in press), 759-762. 123-147. Climate change and global wine quality: Maytin, C. E., Acevedo, M. F., Jaimez, R., Prescott, J. A., 1965, The climatology of the Climatic Change. Andressen, R., Harwell, M. A., Robock, vine (Vitis Vinifera L.) the cool limits of Jones, G. V., 2005 (in press), Climate Change A. and Azocar, A., 1995, Potential effects cultivation: Transcriptions of the Royal in the Western United States Grape of global climate change on the phenolo- Society of Southern Australia, v. 89, p. 5- Growing Regions: Proceedings of the 7th gy and yield of maize in Venezuela: 23. International Grapevine Physiology and Climatic Change, v. 29, p. 189-211. Ray, C., 1985, Lafite: The Story of Château Biotechnology Symposium, Davis, McCarthy, J.J. et al., 2001, Climate Change Lafite-Rothschild: Christie’s Wine California, to be published in Acta 2001: Impacts, Adaptation, and Publications, London, 135 p. Horticulturae. Vulnerability, Contribution of the Reekie, E. G. and Bazzaz, F. A., 1991, Jones, P. D., Bradley, R. S., Diaz, H. F., Kelly, Working Group 1 to the Third Phenology and growth of four annual P. M. and Wigley, T. M. L., 1986, Assessment of the Intergovernmental species grown in ambient and elevated Northern Hemisphere surface air tem- Panel on Climate Change: Cambridge CO2: Canadian Journal of Botany, v. 69, perature variations: Journal of Climate University Press, Cambridge, 1032 p. p. 2475-2481. and Applied Meteorology. v. 25, p. 161- McInnes, K. L., Whetton, P. H., Webb, L., Rodó, X. and Comín, F. A., 2000, Links 179. and Hennessy, K. J., 2003, Climate between large-scale anomalies, rainfall Jones, R. G., Murphy, J. M., and Noguer, M., change projections for Australian viticul- and wine quality in the Iberian Peninsula 1995, Simulation of climate change over tural regions: The Australian and New during the last three decades: Global Europe using a nested regional climate Zealand Grapegrower and Winemaker, Change Biology, v. 6, n. 3, p. 267-279. model. Part 1: Assessment of control cli- February 2003, p. 40-47. Rosenberg, N. J., 1981a, The increasing CO2 mate including sensitivity to location of Meinert, L.D., and Busacca A.J., 2000, concentration in the atmosphere and its lateral boundaries: Quarterly Journal of Geology and Wine 3: Terroirs of the implication on agricultural productivity: the Royal Meteorological Society. v. 121B, Walla Walla Valley appellation, southeast- effects on photosynthesis, transpiration p. 1413-1449. ern Washington, USA: Geoscience and water use efficiency: Climatic Karl, T. R., Jones, P.D., Knight, R.W., Kukla, Canada, v. 27, p. 149-172. Change, v. 3, p. 265-279. G., Plummer, N., Razuvayev, V., Gallo, Menzel, A. and Fabian, P., 1999, Growing Rosenberg, N. J., 1981b, The increasing CO2 K.P., Lindseay, J., Charlson, R.J., and season extended in Europe: Nature, v. concentration in the atmosphere and its Peterson, T.C., 1993, A new perspective 397, p. 659. implication on agricultural productivity: on global warming: asymmetric trends of Moisselin, J-M., Schneider, M., Canellas, C. effects through CO2 induced climatic daily maximum and minimum tempera- and Mestre, O., 2002, Climate change change: Climatic Change, v. 3, p. 239- ture: Bulletin of the American over France during the 20th century: A 254. Meteorological Society, v. 74, p. study of long-term homogenized data of Schultz, H. R., 2000, Climate change and 1007–1023. temperature and rainfall: La Météorlogie viticulture: a European perspective on Katz, R. W. and Brown, B. G., 1992, v. 38, p. 45-56. climatology, carbon dioxide, and UV-B Extreme events in changing climate: vari- Moonen, A. C., Ercoli, L., Mariotti, M. and effects: Australian Journal of Grape and ability is more important than averages: Masoni, A., 2002, Climate change in Italy Wine Research v. 6, p. 2-12. Climatic Change, v. 21, p. 289-302. indicated by agrometeorological indices Seguin, G., 1986, ‘Terroirs’ and pedology of Kenny, G. J. and Harrison, P. A., 1992, The over 122 years: Agricultural and Forest wine growing: Experentia, v. 42, p. 861- effects of climate variability and change Meteorology, v. 111, p. 13-27. 872. on grape suitability in Europe: Journal of Mullins, M. G., Bouquet, A. and Williams, L. Smart, R. E. and Dry, P.R., 1980, A climatic Wine Research. v. 3, p. 163-183. E., 1992, Biology of the Grapevine: classification for Australian viticultural Kenny, G. J. and Shao, J., 1992, An Cambridge University Press, Great regions: Australian Grapegrower and Assessment of a latitude-temperature Britain. 239 p. Winemaker, v. 17, p. 8-16. index for predicting climate suitability for Nemani, R. R., White, M. A., Cayan, D. R., Spellman, G., 1999, ‘Wine, weather and grapes in Europe: Journal of Jones, G. V., Running, S. W., and Climate’: Weather, v. 54, p. 230-239. 14

Stevenson, T., 2001, New Sotheby’s Wine Encyclopedia: A Comprehensive Reference Guide to the Wines of the World: Dorling Kindersley, London, 3rd Edition, 601 p. Tate, A. B., 2001, Global warming’s impact on wine: Journal of Wine Research, v. 12, n. 2, p. 95-109. Van Leeuwen C., Friant P., Chone X., Tregoat O., Koundouras S. and Dubourdieu D., 2004, The influence of climate, soil and cultivar on terroir: American Journal of Enology and Viticulture, v. 55, n. 3, p. 207-217. Vaudour, E., 2002, The quality of grapes and wine in relation to geography: Notions of terroir at various scales: Journal of Wine Research, v. 13, n. 2, p. 117-141. Waggoner, P. E., 1989, Anticipating the frequency distribution of precipitation if climate change alters its mean: Agricultural and Forest Meteorology, v. 47, p. 321-337. Wilson, J.E., 1998, Terroir: the Role of Geology, Climate, and Culture in the Making of French Wines: Mitchell Beazley, London, U.K., 336 p. Winkler, A. J., Cook, J. A., Kliewere, W. M. and Lider, L. A., 1974, General Viticulture, (4th ed.): University of California Press, Berkeley, 740 p.