1. Introduction
Over 9.9 × 106 ha of grape crops were planted worldwide in 2016, producing over 36.1 × 109 L of wine and accounting for 13% of global alcohol consumption by volume (Anderson et al. 2017). Despite the size of the industry, wine production has traditionally been concentrated in a few regions, primarily in Europe (Anderson et al. 2017). However, over the past 150 years, the industry has undergone a geographic shift toward new world wine regions such as Australia and the Americas. These regions, which were responsible for just 1% of global wine production during the 1860s, now produce approximately 29% of the world’s wine (Anderson et al. 2017). In Canada, the wine industry contributed approximately $9 billion (Canadian dollars) to the national economy in 2015 and $2.8 billion of this came from British Columbia (BC), Canada, making BC the second-largest provincial wine producer in the country, after Ontario (Wine Growers Canada 2017; Agriculture and Agri-Food Canada 2018a). Additionally, the BC wine industry employs approximately 12 000 people directly (Agriculture and Agri-Food Canada 2018a) and drives the tourism sector by welcoming over 1 million visitors annually (BC Wine Institute 2020a). Beyond economic impacts, BC wines and wineries have brought the province international recognition in the form of awards and honors (BC Wine Institute 2020b) despite being relatively young and small on a global scale.
Climate plays an important role in grape growth and the production of quality wines (Jones et al. 2005; Jones and Goodrich 2008). The effects of temperature on viticulture are well documented throughout history, with major examples of environmental change like the Medieval Warm Period and Little Ice Age resulting in observed shifts in both wine-producing regions and harvest dates for established vineyards (Jones 2007). Throughout the twentieth century, viticulture typically took place between the 35th and 50th parallels in the Northern Hemisphere and the 30th and 45th parallels in the Southern Hemisphere (van Leeuwen and Darriet 2016), with average growing season (GS) temperatures typically falling between 12° and 22°C (Jones and Schultz 2016). Nevertheless, exceptions to these ranges can be found with both cooler and warmer conditions, and anthropogenic climate change has already pushed the latitudinal boundaries poleward (Jones and Schultz 2016). Temperature extremes such as freeze damage during the dormant and shoulder seasons as well as heat stress during the growing season can also affect grape production and wine quality (Kriedemann 1968; Caprio and Quamme 2002; Belliveau et al. 2006; Holland and Smit 2014; Jones and Goodrich 2008) and can be limiting factors in some regions (White et al. 2006; Jones 2018). Precipitation is another critical climatic factor for viticulture and oenology, and although it is generally less impactful than temperature, it still has an observable impact on wine and wine quality (Caprio and Quamme 2002; Jones and Goodrich 2008; Moriondo et al. 2013; Tóth and Végvári 2015). Extremely dry conditions (i.e., drought) can threaten entire crops, and moderate water deficits can reduce yield (van Leeuwen et al. 2019). The other extreme, too much moisture, also poses a risk to grape crops by means of pest and fungal infestations (Belliveau et al. 2006) and can alter wine characteristics by diluting flavor and lowering alcohol content (Jones and Goodrich 2008; van Leeuwen et al. 2019).
While traditional wine-producing regions have developed where suitable climates for viticulture exist, climate change has the potential to alter global climatic conditions (Pachauri et al. 2014), threatening some existing wine regions while creating new opportunities in others (Jones et al. 2005; Jia et al. 2019). Climatic changes relevant to viticulture outlined in the IPCC’s Fifth Assessment Report include rising temperatures, changing precipitation patterns, more frequent extreme events, and impacted crop yields. Since the 1950s, observable warming has occurred across the planet, and this can be attributed to anthropogenic greenhouse gas emissions (Pachauri et al. 2014). In southern Canada, temperatures rose by approximately 1.7°C between 1948 and 2012 (Vincent et al. 2015, 2018). Over the twenty-first century, continued warming under two climate change scenarios, or representative concentration pathways (RCP; Moss et al. 2010; van Vuuren et al. 2011), is expected to increase global average temperature by between 1.7° and 4.8°C under RCP4.5 and RCP8.5, respectively (Pachauri et al. 2014). Northern regions, including Canada, are expected to experience greater warming than the lower latitudes (Romero-Lankao et al. 2014). Precipitation trends are less clear, but patterns are known to be changing and this, as well as factors such as melting ice and demographic shifts, will affect water availability in some regions (Pachauri et al. 2014; Jia et al. 2019). In BC, trends in total precipitation are insignificant, but rising temperatures have caused a shift in precipitation phase from snow to rain (Vincent et al. 2015, 2018). Some types of extreme events, including heavy precipitation, heat waves, and wildfires, are becoming more frequent (Pachauri et al. 2014; Jia et al. 2019), although many of these trends have not been explicitly linked to anthropogenic climate change in North America (Romero-Lankao et al. 2014). In BC, these extreme events are particularly concerning; lightning-caused wildfires are becoming increasingly common (Hanes et al. 2019), and have been identified, along with other extreme weather events such as heavy rain, as a concern by provincial grape growers (Belliveau et al. 2006). Climate change has already affected global agricultural yields, but these effects have not been unilateral (Pachauri et al. 2014). In Canada, rising temperatures historically have been credited with increasing major crop yields, while related extreme events, like droughts and heat waves, have caused crop losses (Romero-Lankao et al. 2014). Without considering the potential for adaptation, continued warming is expected to become detrimental to North American crop yields (Romero-Lankao et al. 2014).
Adaptation in the grape and wine industry has the potential to both minimize loss from the threatening impacts of climate change and maximize gain from the auspicious impacts (Jones et al. 2005; Diffenbaugh et al. 2011). Moderate changes to local climatology can often be addressed through changes to vineyard and winery operations, but more drastic climatic change necessitates greater planning and investment from producers (Diffenbaugh et al. 2011). Due to the relatively narrow ranges of climatic suitability for individual grape varieties, drastic change can necessitate moving the production of certain varieties to new locations and growing new heat-tolerant varieties at existing vineyards (Diffenbaugh et al. 2011). Planting new vineyards is an extremely expensive and time-consuming process, as vines typically take 5 years to become productive (Belliveau et al. 2006) and 10–20 years to reach maturity (Diffenbaugh et al. 2011), during which time the climate may continue to change. Consequently, the grape and wine industries require high-quality climatic information to make informed decisions with regard to adaptation strategies and they need this information well in advance of the projected changes. Regionally specific climate change impact assessments (CCIAs) on grape growth and wine production that include projections of future climatic change are therefore valuable resources for effective planning and successful adaptation within the industry.
This CCIA will attempt to determine and convey meaningful insights about the impacts of climate change to the grape growers and wine producers of a previously unstudied region, the Fraser Valley in British Columbia. The relevant peer-reviewed literature will be reviewed to identify effective and robust methods for evaluating the impacts of weather and climate on viticulture and oenology. This knowledge will be interpreted in the context of the local climatology in the Fraser Valley to tailor key indicators and critical thresholds for a grape and wine CCIA in this region. Historical observations of climatic conditions in the Fraser Valley will be used to identify historic trends among the chosen indicators and thresholds. To model future changes in these variables, two different statistical downscaling methods will be evaluated based on ability to reproduce historic conditions in the Fraser Valley, and the most effective method will be used to create local, daily climate change scenarios over the twenty-first century. The observed and projected trends will be discussed in the context of the grape and wine literature and used to inform regionally specific adaptation strategies.
2. Methods
a. Study area
The Fraser Valley (Fig. 1) is an officially recognized appellation of origin for wine in British Columbia, with 25 wineries and 200 acres of vineyard (BC Wine Institute 2020a). The region is located on the British Columbian mainland’s southwestern coast along the Canada–U.S. border and follows the Fraser River from Hope, British Columbia, toward the greater Vancouver area (Fig. 1). Major climatic influences in this area include the Rocky Mountains to the east and the Pacific Ocean to the west, as well as the coastal mountains surrounding the valley, all of which contribute to the moderate temperatures and abundant precipitation in the region (Kelley and Spilsbury 1939). Soil in the Fraser Valley tends to be silty, slightly acidic with a pH of approximately 6.0, yellow-brown or red-brown as a result of high iron content, and high in organic matter (Kelley and Spilsbury 1939; BC Wine Institute 2020a,b). The Fraser Valley is extremely productive, producing over 50% of BC’s total agricultural output by value (Fraser Valley Regional District 2011), indicating strong potential for the expansion of local wine production. Major grape varieties produced in this appellation of origin (representing >10% of total production) include Bacchus (34%), Madeleine Angevine (14%), Siegerrebe (13%), and Pinot noir (12%) (BC Wine Grape Council 2019). Until now, no CCIA on the wine industry in the Fraser Valley has been published, which represents a knowledge gap for regional stakeholders.
(top) Wine regions of British Columbia, Canada (source: Vineyards.com 2020) and (bottom) the Fraser Valley growing area (source: Agriculture and Agri-Food Canada 2018b).
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
b. Climatic key indicators and critical thresholds
Within the existing CCIA literature on viticulture and oenology, a wide variety of key indicators and critical thresholds have been used to assess climatic suitability and climate risks for grape growth and wine production. While inconsistency between CCIAs makes comparisons between regions and studies more difficult (Pachauri et al. 2014), the diversity of microclimates in different wine regions requires local specificity to assess meaningful trends. The relevant grape and wine literature was reviewed to establish a reasonably comprehensive set of key indicators and critical thresholds to assess climatic suitability and climate risks for viticulture and oenology, with the selected variables (Table 1) tailored to the local climatology of the Fraser Valley. Brief explanations for the inclusion of each variable will be presented along with sources for the development of each variable or other grape and wine CCIAs in which they have been included.
Formulas of selected key indicators and critical thresholds.
The accepted GS in the Northern Hemisphere within the grape and wine literature is the period between 1 April and 31 October (Jones et al. 2005; White et al. 2006; Jones and Goodrich 2008; Holland and Smit 2014; Shaw 2017; Schultze and Sabbatini 2019; Hewer and Brunette 2020; Hewer and Gough 2020); all of the following GS variables will refer to this period. By extension, the accepted Northern Hemisphere dormant season (DS) is between 1 November and 31 March; all the following DS variables will refer to this period. The ripening period (RP) is less clearly established: some have defined it as 1 September–31 October (Holland and Smit 2014; Shaw 2017; Hewer and Brunette 2020; Hewer and Gough 2020) while others consider the period between 15 August and 15 October (Jones and Goodrich 2008). Discrepancies between ripening periods can be largely attributed to latitudinal differences between growing regions, explaining the use of the later RP in Canadian studies, and the earlier RP in American studies. Accordingly, the RP in the high-latitude Fraser Valley will most likely occur relatively late in the season. For this reason, and to allow comparison with regions at similar latitudes, RP variables in this study will refer to the 1 September–31 October period.
1) Climatic suitability
Average GS mean temperature has been used extensively in the grape and wine literature (Jones et al. 2005; White et al. 2006; Hall and Jones 2010; Holland and Smit 2014; Rayne and Forest 2016; Shaw 2017; Hewer and Brunette 2020; Hewer and Gough 2020) to evaluate climatic suitability, making it a useful variable for comparisons of global wine regions. Jones et al. (2005) developed this variable by defining regions ranging from cool to very hot (Table 2) that can be used to classify wine-producing regions based on the grape varieties that will thrive in the local conditions.
Classification systems for climatic indices.
Growing degree-days (GDD) in the context of viticulture are a measurement of heat accumulation above a base of 10°C. Cumulative GDD during the GS can be classified using the Winkler index and can be used to characterize a region based on the grape varieties that will thrive under the local conditions (Amerine and Winkler 1944) (Table 2). The Winkler index was originally developed based on Californian wine production but was later developed with reference to American and Australian viticulture by Jones et al. (2010) and Hall and Jones (2010). It has become one of the most widely used indices in the global grape and wine literature (Shaw 1999; White et al. 2006; Jones and Goodrich 2008; Jones 2012: Rayne and Forest 2016; Roy et al. 2017; Shaw 2017; Jones 2018; Schultze and Sabbatini 2019; Hewer and Brunette 2020) making it another useful metric for comparing the Fraser Valley with other wine-producing regions.
Although the Winkler index has been extensively used, it has also been criticized for failing to consider the effects of latitudinal differences and inconsistent growing seasons in different locations (Shaw 1999). The latitude–temperature index (LTI) (Jackson and Cherry 1988) accounts for these shortcomings by considering the mean temperature of the warmest month of the year and incorporating the latitude of the region in question. Four cultivar groups have been defined using the LTI, which, like previous indices, can be used to classify regions based on which grape varieties will thrive under local conditions (Table 2).
The cool nights index (CI) is another temperature-based index but is based on minimum temperatures rather than mean temperatures. In the Northern Hemisphere, this index is calculated by determining the average of daily minimum temperatures during the month of September (Tonietto and Carbonneau 2004) and it is used frequently in the grape and wine literature (Moriondo et al. 2013; Shaw 2017; Jones 2018; Hewer and Brunette 2020). The index was originally used alongside other viticulture indices to characterize wine regions and has a set of associated ranges that can be used to classify and compare regions ranging from very cool nights to warm nights (Tonietto and Carbonneau 2004) (Table 2).
The deMartonne aridity index (dMI) is used to quantify moisture conditions by considering both precipitation and temperature (deMartonne 1926). There are many indices that assess moisture (Zargar et al. 2011), but the simplicity of the dMI makes it more accessible than many moisture indices that require soil moisture data. Throughout the GS, some moisture is required for plant growth; however, during the RP, Jones and Goodrich (2008) found that higher temperatures have a positive effect on wine quality, while higher precipitation has a negative effect. Considering the formula for dMI (Table 1), it can be expected that lower dMI values during the RP will lead to higher quality wine. There is a system of classifications for dMI ranging from arid to extremely humid (Table 2) and although these classifications are not specifically related to viticulture and oenology, they can be useful for contextualizing and comparing regions. dMI classifications are based on an annual time scale so dMI values during the GS and RP will be calculated using an adjusted formula outlined by Croitoru et al. (2012) (Table 1). This method magnifies interannual variability and will particularly affect dMI values calculated over short time periods, like the RP.
2) Climate risks
Grapevine photosynthesis increases as temperature rises until a heat stress threshold is reached and further warming becomes detrimental (Kriedemann 1968). There is no consensus in the literature with regard to the exact threshold for heat stress (Table 3), so average maximum temperature is a useful measurement to consider along with the frequency of extreme heat days. Heat tolerance differs between grape varieties, and studies have incorporated extreme heat thresholds ranging from 30° (de Orduña 2010; Shaw 2017; Hewer and Brunette 2020; Hewer and Gough 2020) to 35°C (Belliveau et al. 2006; White et al. 2006). Vincent et al. (2018) define extreme heat events as those exceeding the 95th percentile of maximum daily temperatures, however, over the historic period (1970–2019) in the Fraser Valley, the 95th percentile of extreme heat during the GS is approximately 20.8°C, which is unlikely to cause damage to vines. Caprio and Quamme (2002) identified a 32°C threshold by analyzing crop yield data from the Okanagan Valley wine region, which provides a meaningful physiological basis for this threshold. Additionally, while the Fraser Valley and nearby Okanagan Valley are not identical climatologically, the 32°C could be considered the most applicable of the thresholds in the grape and wine literature due to the proximity of its source material to the study site.
Climate risk thresholds in North America and their sources.
Precipitation is a major contributor to soil moisture during the GS and has been included in several grape and wine CCIAs (Jones and Goodrich 2008; Jones 2012; Moriondo et al. 2013; Holland and Smit 2014; Tóth and Végvári 2015; Shaw 2017; Jones 2019; Hewer and Gough 2020; Hewer and Brunette 2020). Soil moisture in many BC wine regions is also heavily dependent on snowmelt (Belliveau et al. 2006), but winter temperatures in the Fraser Valley are generally high enough that snow cannot accumulate substantially, so DS precipitation was not considered. While sufficient precipitation for grape growth is most often assessed as a factor of climatic suitability, the high levels of precipitation experienced in the Fraser Valley suggest that water deficit will almost certainly not be a limiting factor in the local context. However, high precipitation during the GS has been identified as a risk by grape growers (Belliveau et al. 2006; Holland and Smit 2014) so this variable should be assessed for its potential detrimental effects. In addition to high levels of cumulative precipitation, extreme precipitation events can lead to disease, rot, or fungal infection and are frequently considered in the grape and wine literature (Shaw 2017; Jones 2019; Schultze and Sabbatini 2019; Hewer and Brunette 2020; Hewer and Gough 2020). Thresholds defining extreme precipitation are not consistent within the literature (Table 3), and other studies have used somewhat arbitrary thresholds of 10 mm (Shaw 2017; Hewer and Brunette 2020; Hewer and Gough 2020) or 50 mm (Schultze and Sabbatini 2019). Vincent et al. (2018) define days with heavy precipitation as those exceeding the 90th percentile of precipitation events, and using this percentile approach, extreme precipitation during the GS in the Fraser Valley will be defined as days with total precipitation greater than 18 mm.
During the GS, low minimum temperatures can prevent or delay vine growth and hinder production. Accordingly, high GS average minimum temperatures are associated with higher quality wines (Jones and Goodrich 2008) and the variable has been included in CCIAs of other cool climate wine regions in Quebec (Jones 2012, 2018) and Ontario (Holland and Smit 2014) in Canada. A more serious risk associated with minimum temperatures during the GS is frost, which was identified by grape growers as a problem in BC (Belliveau et al. 2006) and Ontario (Holland and Smit 2014). There is some debate over whether temperatures only slightly below freezing are low enough to damage growing vines, and tolerance can differ between grape varieties. This has led to the use of several different thresholds to define frost days (Table 3). To avoid overlooking potentially damaging conditions, days for which the minimum temperature falls below 0°C will be considered potential frost days and used to assess the risk of frost damage.
During the DS, low minimum temperatures can damage vines, although once again, the exact threshold for freeze damaged is not well established (Table 3). Most studies have used thresholds of −18°C (Schultze and Sabbatini 2019) or lower (Caprio and Quamme 2002; Belliveau et al. 2006; Holland and Smit 2014; Roy et al. 2017; Shaw 2017; Hewer and Brunette 2020; Hewer and Gough 2020); however, temperatures this low rarely occur in the Fraser Valley. Vincent et al. (2018) define extreme cold as minimum temperatures below the 5th percentile of daily minimum temperatures, which is −6.4°C during the DS in the Fraser Valley, but this threshold is likely too high to represent potential vine damage. It must also be considered that higher temperatures are known to damage vines that are not yet acclimatized to cold, such as during shoulder seasons (Belliveau et al. 2006; Holland and Smit 2014). Aney (1974) found that vine damage can occur at temperatures as high as −12°C before mid-November in Oregon, which is likely due to vines having not yet become acclimatized to cold. A similar threshold, −12.2°C, has also been used by White et al. (2006) and Diffenbaugh et al. (2011) who assessed warm wine regions in the United States. In the Fraser Valley, it could be the case that vines do not become acclimatized to cold at any point during the DS because of the high temperatures, making a threshold of −12°C a reasonable indicator for potential vine damage in this region.
c. Data
1) Historical observations
Observational data of daily mean, minimum, and maximum temperature and total precipitation were retrieved from Environment and Climate Change Canada weather stations during the 1970–2019 period (Environment and Climate Change Canada 2020). The primary weather station used was Abbotsford station (latitude: 49.03°N, longitude: −122.36°E) due to its central location in the Fraser Valley (Fig. 1) and the availability of consistent data dating back to 1970. Since the Fraser Valley runs in the zonal direction, latitudinal differences should not have a significant impact on the climatology of the valley, making a central location appropriate to represent the region. Of the 18 262 days of data gathered from the Abbotsford weather station, 167 (0.91%) were missing. In these cases, the data were completed using observations from the Mission West Abbey weather station (Fig. 1; latitude: 49.15°N, longitude: −122.27°E), located approximately 14.9 km northeast of the Abbotsford station as is common practice when producing long time series for CCIA (Environment and Climate Change Canada 2017).
2) Statistically downscaled climate change scenarios
Statistically downscaled climate change scenarios for all of Canada are available from the University of Victoria’s Pacific Climate Impacts Consortium (2019). The Pacific Climate Impacts Consortium (PCIC) data are based on CMIP5 general circulation models (GCMs) as well as historical daily gridded climate data for Canada and produce simulated daily data for the period 1950–2100 with 1950–2005 representing the historical baseline. More information on PCIC’s statistical downscaling methods can be found on their website (https://www.pacificclimate.org/data/statistically-downscaled-climate-scenarios). Model run r1i1p1 for 27 different models, each corresponding to a CMIP5 GCM, was retrieved for two emissions scenarios, one intermediate (RCP4.5) and one high (RCP8.5) (Moss et al. 2010; van Vuuren et al. 2011). The downscaled model outputs have a gridded resolution of roughly 10 km, and the modeled data corresponding to the Abbotsford weather station (latitude: 49.03°N, longitude: −122.36°E) cell were retrieved for analysis. The weather variables available from PCIC are minimum daily temperature at surface level Tmin, maximum daily temperature at surface level Tmax, and total daily precipitation Ptot.
The statistical downscaling model (SDSM) is a decision-centric tool that assesses local climate change impacts by reproducing observed weather conditions and applying change factors to them as specified by the user (Wilby et al. 2002). By applying GCM anomalies to synthetic weather generated with SDSM, localized daily datasets of climate variables representing future conditions in the Fraser Valley were produced. The variance of the generated weather was also artificially inflated within SDSM to match observed conditions more closely. Selective ensembles of seasonal CMIP5 GCM projections based on the approach of Hewer and Gough (2016a,b, 2019a,b, 2020) were created to produce seasonally specific change factors for each emissions scenario during three delta periods across the twenty-first century: the 2020s (2011–40), the 2050s (2041–70), and the 2080s (2071–2100). The change factors were applied to the simulated weather in SDSM to produce local, daily climate change scenarios for RCP4.5 and RCP8.5 throughout the twenty-first century.
3) Statistical analysis
Over the historic period, linear regression is used to determine the rate of change and statistical significance of trends for each variable (Table 4). The chi-square test is used to test the assumption of normality, and when violated, the nonparametric tests Spearman’s rho and Kendall’s tau are reported to reinforce the significance of Pearson’s R. The Durbin–Watson test is used to identify serial correlation and reported when statistically significant. Projections of average annual values of each variable (Table 5) are reported for the three delta periods during the twenty-first century—the 2020s, 2050s, and 2080s—for both RCP4.5 and RCP8.5.
Historic trends in key indicators and critical thresholds for viticulture and oenology in the Fraser Valley (1970–2019). Statistically significant values are shown in boldface type.
Projections of key indicators and critical thresholds for viticulture and oenology in the Fraser Valley during the twenty-first century.
3. Results
a. Climatic suitability
Average GS temperature (Fig. 2a) demonstrated a statistically significant positive trend (R2 = 0.626; P < 0.001) and rose at a rate of 0.045°C yr−1, or approximately 2.3°C over the 50-yr historic period (Fig. 2a). However, serial correlation was detected by the Durbin–Watson test (d = 1.154; P ≤ 0.05), which may have confounded the results. Regardless, rising temperatures have transitioned the Fraser Valley from a very cool climate classification at the beginning of the historic period to an intermediate classification during some years at the end of the historic period. GDD (Fig. 2a) also increased with a statistically significant positive linear trend (R2 = 0.640; P < 0.001) and rose at a rate of 8.1 GDD yr−1, or 405 GDD over the 50-yr historic period. GDD also had issues of serial correlation detected by the Durbin–Watson test (d = 1.201; P ≤ 0.05), which, again, may have confounded the results. Regardless, rising GDD have transitioned the Fraser Valley from Winkler region Ia at the beginning of the historic period to Winkler region Ib during some years at the end of the historic period.
Historical trends among factors affecting climatic suitability for viticulture and oenology in the Fraser Valley: (a) average GS mean temperature and GDD, (b) LTI and CI, and (c) GS dMI and RP dMI.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
Statistically significant increases in LTI (Fig. 2b) were observed at a rate of 0.61 yr−1 (R2 = 0.457; P < 0.001), or a total increase of approximately 31 LTI over the 50-yr historic period. This transitioned the Fraser Valley from cultivar group A to group B. Statistically significant increases in the CI (Fig. 2b) were also observed at a rate of 0.059°C yr−1 (R2 = 0.447; P < 0.001) or approximately 3.0°C over the entire historic period (Fig. 2b). Despite this increase, the Fraser Valley remains within the CI+2 classification (very cool nights).
No significant trends were detected for dMI (Fig. 2c) in either the GS (R2 = 0.045; P = 0.141) or RP (R2 = 0.008; P = 0.528). Observed dMI values were also extremely variable, ranging from a Mediterranean to extremely humid classification during the GS and from a semiarid to extremely humid classification during the RP.
Under RCP4.5, average GS temperature (Fig. 3a) is expected to continue increasing, surpassing the baseline average (14.4°C) by 1.2°C during the 2020s, 2.1°C during the 2050s, and 2.7°C during the 2080s. This would place the region in a warm climate classification by the end of the twenty-first century. Under RCP8.5, average GS temperature is expected to rise above the baseline average by 1.3°C during the 2020s, 2.9°C during the 2050s, and 4.9°C during the 2080s. This would transition the region to a warm climate classification by the middle of the century, and a hot climate classification by the end of the century. Under RCP4.5, average cumulative GDD (Fig. 3b) is expected to surpass the baseline average (1009 GDD) by 198 in the 2020s, 385 in the 2050s, and 501 in the 2080s. This would transition the Fraser Valley to Winkler region II during the 2050s. RCP8.5 increases GDD beyond the baseline average by 217 during the 2020s, 538 during the 2050s, and 948 during the 2080s. This increase would place the region in Winkler region II by midcentury, and Winkler region IV by the end of the century.
Projections of climatic suitability for viticulture and oenology in the Fraser Valley during the 2020s (2011–40), 2050s (2041–70), and 2080s (2071–2100) under intermediate (RCP4.5) and high (RCP8.5) emissions scenarios: (a) average GS mean temperature, (b) GS GDD, (c) LTI, and (d) CI.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
Under RCP4.5, LTI (Fig. 3c) is expected to surpass the baseline average (200.0) by 16.4 during the 2020s, 29.4 during the 2050s, and 38.1 during the 2080s. RCP8.5 projections increase the LTI by 19.0 above the baseline average during the 2020s, 39.8 during the 2050s, and 66.8 during the 2080s. Both emissions scenarios maintain the region’s classification in cultivar group B. Under RCP4.5, the projected increase in September minimum temperatures would raise the CI (Fig. 3d) beyond the baseline average (9.3°C) by 0.8°C during the 2020s, 1.8°C during the 2050s, and 2.6°C during the 2080s. This would leave the region at the high end of the CI+2 classification (very cool nights) by the end of the twenty-first century. RCP8.5 projections see the CI increasing beyond the baseline average by 1.1°C during the 2020s, 2.7°C during the 2050s, and 4.6°C during the 2080s. This suggests that the CI in the region would enter the CI+1 classification (cool nights) during the 2050s and remain in this classification throughout the century.
Despite little observed change in dMI since 1970, substantial decreases have been projected throughout the twenty-first century, and this is reasonable in the context of significant temperature rise. Under RCP4.5, GS dMI (Fig. 4a) is expected to decrease below the baseline average (43.7) by 3.3 during the 2020s, 6.7 during the 2050s, and 7.9 during the 2080s, maintaining the region’s very humid classification during the GS. Under RCP8.5, GS dMI is expected to decrease by 5.2 during the 2020s, 5.9 during the 2050s, and 12.1 during the 2080s, transitioning the GS into a humid classification by the end of the century. RCP4.5 projections lower the RP dMI (Fig. 4b) below the baseline average (61.0) by 14.8 during the 2020s, 19.7 during the 2050s, and 25.0 during the 2080s leaving the RP at the threshold between the humid and very humid classification by the end of the century. RCP8.5 projections lower the RP dMI by 20.2 during the 2020s, 17.6 during the 2050s, and 28.1 during the 2080s, leading to a humid classification for the RP by the end of the century.
Projections of climatic suitability for viticulture and oenology in the Fraser Valley during the 2020s (2011–40), 2050s (2041–70), and 2080s (2071–2100) under intermediate (RCP4.5) and high (RCP8.5) emissions scenarios: (a) GS dMI and (b) RP dMI.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
b. Climate risk
Average maximum GS temperature (Fig. 5a) increased at a rate of 0.033°C yr−1 between 1970 and 2019 (R2 = 0.353; P < 0.001), or approximately 1.7°C over the 50-yr period. The chi-square test revealed that the data did not meet the assumption of normality associated with linear regression (χ2 = 7.940; P = 0.047). However, the nonparametric tests Spearman’s rho (Rs = 0.596; P ≤ 0.05) and Kendall’s tau (T = 0.492; P ≤ 0.05) each indicated a statistically significant increasing trend, reinforcing the linear trend. The frequency of extreme heat days (Fig. 5a) increased at a rate of 0.44 days per year (R2 = 0.068; P = 0.067), or 22 additional days over the 50-yr period.
Historical trends among thermal climate risks for viticulture and oenology in the Fraser Valley: (a) average GS maximum temperature and frequency of GS extreme heat days and (b) average dormant season minimum temperature and frequency of dormant season extreme cold days.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
Average minimum DS temperatures (Fig. 5b) rose at a statistically significant rate of 0.034°C yr−1 (R2 = 0.222; P < 0.001) or approximately 1.0°C over the historic period. Correspondingly, the frequency of extreme cold days (Fig. 5b) decreased at a rate of −0.040 days yr−1 (R2 = 0.129; P = 0.010), or an average of 2 fewer extreme cold days per DS over the observational period. The chi-square test indicated that the frequency of extreme cold days did not meet the assumption of normality (χ2 = 73.046; P < 0.001), but the nonparametric statistics Spearman’s rho (Rs = −0.341; P ≤ 0.05) and Kendall’s tau (T = −0.264; P ≤ 0.05) both reinforced the significance of the linear trend.
Projections for RCP4.5 show average maximum GS temperature (Fig. 6a) surpassing the baseline average (19.8°C) by 1.3°C during the 2020s, 2.6°C during the 2050s, and 3.4°C during the 2080s. RCP8.5 would result in greater temperature rise; 1.5°C above the baseline average by the 2020s, 3.9°C by the 2050s, and 5.5°C by the 2080s. Extreme heat events (Fig. 6b) become increasingly common along with these temperature increases: RCP4.5 results in an average increase above the baseline average (3.2 days) by 3.0 days per year becoming 2.9% of the GS during the 2020s, 9.7 additional days per year or 6.0% of the GS during the 2050s, and 14.6 additional days per year or 8.3% of the GS during the 2080s. RCP8.5 projections result in 4.3 additional extreme heat days per year or 3.5% of the GS during the 2020s, 15.7 additional days per year or 8.8% of the GS during the 2050s, and 37.4 additional days per year or 19.0% of the GS during the 2080s.
Projections of thermal climate risks for viticulture and oenology in the Fraser Valley during the 2020s (2011–40), 2050s (2041–70), and 2080s (2071–2100) under intermediate (RCP4.5) and high (RCP8.5) emissions scenarios: (a) average GS maximum temperature, (b) frequency of GS extreme heat days, (c) average DS minimum temperature, and (d) frequency of DS extreme cold days.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
RCP4.5 is expected to increase average DS minimum temperatures (Fig. 6c) above the baseline average (1.4°C) by 0.6°C during the 2020s, 1.7°C during the 2050s, and 2.1°C during the 2080s. RCP8.5 is expected to increase average DS minimum temperatures above the baseline average by 1.3°C during the 2020s, 2.1°C during the 2050s, and 3.7°C during the 2080s. RCP4.5 projections lower the average annual frequency of DS extreme cold (Fig. 6d) below the baseline average (1.0 days) by 0.1 days during the 2020s, 0.6 days during the 2050s, and maintain this level during the 2080s. RCP8.5 projections lower frequencies below the baseline average even further; by 0.5 days during the 2020s, 0.6 days during the 2050s, and 0.9 days during the 2080s.
Between 1970 and 2019, average GS minimum temperatures (Fig. 7a) rose at a rate of 0.056°C yr−1, or 2.8°C over the 50-yr historic period and the linear trend was statistically significant (R2 = 0.730; P < 0.001). However, positive serial correlation was detected within the data (d = 1.116; P ≤ 0.05), which may have confounded the results of the linear trend analysis. The frequency of potential frost days (Fig. 7b) decreased at a rate of −0.129 days per year, or an average of approximately 6.5 fewer days over the 50-yr period, and this trend was statistically significant (R2 = 0.363; P < 0.001). The chi-square test revealed that the frequency of potential frost days did not meet the assumption of normality (χ2 = 35.522; P < 0.001), but the nonparametric statistics Spearman’s rho (Rs = −0.622; P ≤ 0.05) and Kendall’s tau (T = −0.481; P ≤ 0.05) both reinforced the significance of the linear trend.
Historical trends among physical climate risks for viticulture and oenology in the Fraser Valley: (a) average GS minimum temperature and frequency of GS potential frost days and (b) total GS precipitation and frequency of GS extreme precipitation days.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
Precipitation trends were less distinguished than temperature trends in the Fraser Valley during the observational period. Total precipitation during the GS (Fig. 7c) displayed a slight decrease between 1970 and 2019 of −0.793 mm yr−1 or approximately 39.7 mm over the historic period, but this trend was not statistically significant (R2 = 0.008; P = 0.528). The frequency of heavy precipitation days (Fig. 7d) also decreased slightly, at a rate of −0.032 days per year, or 16 fewer days over the historic period, but again, the trend was not statistically significant (R2 = 0.023; P = 0.290).
Under RCP4.5, average GS minimum temperature (Fig. 8a) is expected to surpass the baseline average (8.9°C) by 0.7°C during the 2020s, 1.7°C during the 2050s, and 2.3°C during the 2080s. Under RCP8.5, warming is expected to be more drastic, surpassing the baseline average by 0.9°C during the 2020s, 2.4°C during the 2050s, and 3.3°C during the 2080s. Under RCP4.5, the frequency of potential GS frost days (Fig. 8b) is expected to decrease below the baseline average (2.5 days) by 1.1 days during the 2020s, 1.5 days during the 2050s, and 1.9 days during the 2080s. Under RCP8.5, potential frost days during the average GS are expected to fall below the baseline average by 1.2 days during the 2020s, 1.9 days during the 2050s, and are projected to be eliminated during the 2080s.
Projections of physical climate risks for viticulture and oenology in the Fraser Valley during the 2020s (2011–40), 2050s (2041–70), and 2080s (2071–2100) under intermediate (RCP4.5) and high (RCP8.5) emissions scenarios: (a) average GS minimum temperature, (b) frequency of GS potential frost days, (c) total GS precipitation, and (d) frequency of GS extreme precipitation days.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0145.1
Inconclusive trends over the historic period set little precedence for precipitation changes in the Fraser Valley in response to climate change. Nonetheless, modest decreases in total GS precipitation (Fig. 8c) were projected over the twenty-first century. Projections of the frequency of extreme precipitation days (Fig. 8d) were indistinct, although slight increases were projected for all periods and emissions scenarios except RCP4.5 during the 2050s.
4. Discussion and conclusions
This CCIA assessed both historic and projected trends in key indicators and critical thresholds for grape growth and wine production in the Fraser Valley of BC in Canada. Previous CCIAs on viticulture and oenology have helped inform the selection of these indicators and thresholds and tailor the analysis to the local climate. No prior grape and wine CCIAs have been conducted for this appellation of origin, and few other CCIAs of comparable cool climate wine regions have considered robust projections of future climate scenarios (White et al. 2006; Roy et al. 2017; Schultze and Sabbatini 2019; Hewer and Brunette 2020; Hewer and Gough 2020). The trends identified in the Fraser Valley are generally similar to those of other cool climate wine regions including BC’s Okanagan Valley (Rayne and Forest 2016), Ontario (Shaw 2017; Hewer and Brunette 2020; Hewer and Gough 2020), Quebec (Jones 2012; Roy et al. 2017; Jones 2018, 2019), the northern United States (Jones and Goodrich 2008; Schultze and Sabbatini 2019), and the United Kingdom (Nesbitt et al. 2016). In all these regions, warming trends have been observed historically (Jones and Goodrich 2008; Jones 2012; Holland and Smit 2014; Rayne and Forest 2016; Roy et al. 2017; Shaw 2017; Jones 2018; Hewer and Brunette 2020; Hewer and Gough 2020). This reflects the statistically significant warming trends observed in the Fraser Valley between 1970 and 2019. Where projections have been made, warming is also projected to continue across the Canadian wine regions (Roy et al. 2017, Hewer and Brunette 2020; Hewer and Gough 2020) as well as in Europe (Fraga et al. 2013; Moriondo et al. 2013; Tóth and Végvári 2015; Nesbitt et al. 2016; Cardell et al. 2019) over the remainder of the twenty-first century. This analysis suggests that the Fraser Valley is no exception. GS precipitation patterns across high-latitude wine regions in Canada, the United Kingdom, and the northwestern United States have generally not shown clear trends historically (Jones and Goodrich 2008; Jones 2012; Nesbitt et al. 2016; Shaw 2017; Jones 2019; Hewer and Brunette 2020; Hewer and Gough 2020) and historic trends in the Fraser Valley are consistent with this. Only Schultze and Sabbatini (2019), Hewer and Brunette (2020), and Hewer and Gough (2020) have considered projections of precipitation in nearby cool-climate wine regions and did not project substantial changes in the northern United States or Ontario. The modest decreases projected for the Fraser Valley are questionable based on a lack of historic precedence for precipitation change in response to climate change, and precipitation levels will still be relatively high for a wine region. The projected decrease in dMI is also questionable by lack of historic precedence, but since dMI is based partially on temperature, anticipated warming makes lower dMI values more reasonable. During the RP, lower dMI could lead to improved wine quality (Jones and Goodrich 2008).
Unsurprisingly, higher greenhouse gas (GHG) emissions associated with RCP8.5 are projected to result in greater increases in temperature indicators for viticulture than RCP4.5, as well as greater increases in the frequency of extreme heat and decreases in the frequency of extreme cold. Variables that did not show consistent differences in trends projected under the different emissions scenarios include GS dMI, RP dMI, total GS precipitation, and the frequency of GS extreme precipitation; all variables for which no significant trend was detected under historic climate change. Differences between projections based on the two scenarios are greatest at the end of century, which is consistent with results of similar viticulture and oenology CCIAs (Hewer and Brunette 2020; Hewer and Gough 2020) and reasonable given the differences in expected emissions levels and GHG concentrations during the late century under either emissions scenario (van Vuuren et al. 2011).
Projections of GS potential frost days suggest that frost will decline or even be eliminated as a risk factor in the Fraser Valley. However, another possibility is that bud break of grapevines will occur earlier in the season, making crops vulnerable to frost that may occur outside the bounds of the accepted GS (Meier et al. 2018). Meier et al. (2018) found that either scenario is possible, depending on several factors including location and emissions scenario. Therefore, the potential for phenological changes to preserve the risk of frost in the Fraser Valley represents a knowledge gap that could be filled by phenological research beyond the scope of this study.
Serial correlation was identified in three variables: average GS temperature, GDD, and average minimum GS temperature by the Durbin–Watson test, which could have confounded the results of the linear and nonparametric tests. Despite vulnerability to serial correlation, rising temperatures across Canada and BC are well established in the climate change literature (Romero-Lankao et al. 2014; Vincent et al. 2015, 2018). Ultimately, issues with serial correlation in these temperature-derived variables may have inflated R2 values, but the presence of positive trends is still very likely. Nonetheless, serial correlation can be addressed using several methods, such as the incorporation of a lagged covariate predictor in linear regression (Ludlow and Perez 2018), which could be an area of improvement for future research.
This CCIA has focused primarily on key indicators and critical thresholds for the growing season, dormant season, and ripening period. This is common in the grape and wine literature, but there is a great deal that can be learned from assessing the same variables over different months, seasons, or phenological stages. Belliveau et al. (2006) report that grape growers consider the specific timing of events like heavy rain and frost to be critical, and Caprio and Quamme (2002) even identified specific weeks during which weather events are known to affect grape harvests, such as the timing of heavy precipitation or drought. Consequently, analysis of trends and projections with higher temporal resolution could shed more light on climate change impacts in the Fraser Valley. Future research could therefore focus on more specific time periods such as bloom period (Jones and Goodrich 2008) or individual months (Shaw 2017; Jones 2019). Another factor that limited the climatic indicators and thresholds that could be included in this study was the nature of the statistically downscaled modeled data. The statistical downscaling method chosen to produce temperature data is not suitable for assessing diurnal temperature range. Consequently, several climatic indicators that are relevant to viticulture and oenology could not be assessed, including diurnal temperature range, the Huglin heliothermal index (Huglin 1978), and many more modern and complex moisture indices (Zargar et al. 2011), each of which would provide valuable information to the local industry if assessed in the future.
Grape and wine CCIAs have incorporated crop yield data in the Okanagan Valley (Caprio and Quamme 2002), and wine quality data in several well-known global wine regions (Jones et al. 2005; Jones and Goodrich 2008). These studies have helped to inform the selection of key indicators and critical thresholds for the Fraser Valley, but analyses specific to the region could have further validated the chosen variables. Several organizations have begun collecting and reporting grape crop data in BC, including the BC Grape Growers Association and the BC Wine Grape Council, but it will be years before datasets of useful length for climate research are available, and some of the information is withheld from the public due to business propriety. Nonetheless, the incorporation of this data could improve the analysis of future CCIAs and allow researchers to make more informed recommendations to grape growers and wine producers. In this regard, the grape and wine industries could gain critical insights for adaptation planning and climate risk management by sharing data with the academic community.
Adaptation options for grape growers and wine producers include altering vineyard design or management, adjusting the wine-making process, planting different grape varieties, and changing the location of vineyards (Diffenbaugh et al. 2011; Jones and Schultz 2016; van Leeuwen et al. 2019; Naulleau et al. 2021). Because of its current classification as a very cool climate region, or Winkler region I, the Fraser Valley is likely to have abundant opportunities for adaptation to climate change as compared with regions that are already considered hot (Diffenbaugh et al. 2011). It is also likely that the total area suitable for viticulture in BC will expand in the context of climate change as wine regions are expected to shift northward in both Europe (Fraga et al. 2013; Moriondo et al. 2013; Tóth and Végvári 2015; Cardell et al. 2019) and North America (White et al. 2006; Hannah et al. 2013; Roy et al. 2017). Emerging wine regions have already been identified north of the Okanagan Valley in BC (Fig. 1) (Jones and Schultz 2016). Nonetheless, the warming trends observed and projected over the twenty-first century in the Fraser Valley would transition the region into warmer viticultural classifications that, in the absence of other adaptation strategies (see Naulleau et al. 2021), will be suitable for different grape varieties. To interpret these results, the grape varieties recommended by Amerine and Winkler (1944) for the production of high-quality wines in each anticipated Winkler region have been reported in Table 6. Additionally, spatial analogs of well-known wine regions have been reported based on the 1971–2000 period GDD climate normals reported by Jones et al. (2010). It is important to note that these spatial analogs are based on historic conditions in each wine region; therefore, insights can be gained from the wines traditionally grown in these regions as well as traditional viticultural and oenological practices, rather than current varieties and practices that may have already been altered by climate change. Since the Fraser Valley is projected to progress from Winkler region II to IV between the 2050s and 2080s under RCP8.5, grape variety recommendations and spatial analogs have also been reported for Winkler region III, which will likely represent the Fraser Valley during the late 2050s and early 2080s in this emissions scenario. GDD alone cannot define a wine region comprehensively (Dunn et al. 2017) so other factors like soil conditions, extreme temperatures, and precipitation should also be taken into account when considering these recommendations and analogs.
Potential wine grape varieties for cultivation in the Fraser Valley and worldwide spatial analogs to 1971–2000 conditions based on GDD projections throughout the twenty-first century. Note that since no distinction was originally made between Winkler regions Ia and Ib, suitable grape varieties as determined by Amerine and Winkler (1944) are the same for each of these classifications.
While transitioning to new grape varieties is a feasible adaptation strategy, the long time to productivity and maturity for grapevines makes this approach expensive and time-consuming (Belliveau et al. 2006; Diffenbaugh et al. 2011). As a result, grape growers in the Fraser Valley need to plan well in advance for the conditions that the region is projected to experience, reinforcing the importance of this CCIA and others like it. White et al. (2006) and Moriondo et al. (2013) found that area suitable for viticulture shifts not only northward but upward in elevation as temperatures rise, meaning the mountainous areas surrounding the Fraser Valley could become increasingly well-suited for viticulture. However, the same challenges associated with grapevines reaching maturity when vineyards are replanted would apply to relocating vineyards to higher elevations, and the expenses of acquiring new land for viticulture would be significant. For this reason, it is more likely that existing vineyards will choose to grow different grape varieties and the nearby higher-elevation regions where cool-climate varieties can thrive will be potential sites for expansion.
Climate change will impact not only climatic suitability for grape varieties but the quality of wine each variety produces. Jones et al. (2005) found that rising temperatures tend to increase wine quality, but they also introduced the concept of optimal temperatures after which quality could decrease. Optimal temperatures can be different for each grape variety, and the recommendations of Amerine and Winkler (1944) are intended to incorporate these differences. It is important to note, however, that suitability based on wine quality is dependent at least in part on subjective taste. This means that the appropriate grape varieties in any regional classification may change as the public perception of high-quality wine changes. This nuance has been identified as a limitation for assessing the relationship between climate and wine quality (Jones et al. 2005; Jones and Goodrich 2008) and should be acknowledged when interpreting the recommendations from Amerine and Winkler (1944). Interestingly, the influence of subjective taste has also been identified by Diffenbaugh et al. (2011) as an adaptation strategy. Diffenbaugh et al. (2011) proposed that minor climatic changes, and the resulting changes in wine characteristics, could be adapted to through commercial marketing. Accordingly, public perception of high-quality wine, as well as the accepted characteristics of specific wine varieties, could also begin to change organically as respected wine regions begin to produce wines with flavors altered by climate change.
Acknowledgments
The authors thank the anonymous reviewers for their thorough and valuable feedback that improved the quality of this paper. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability statement
Data analyzed in this study came exclusively from public resources, which are openly available at locations cited in the reference section.
REFERENCES
Agriculture and Agri-Food Canada, 2018a: B.C. wine facts. Government of Canada, accessed 20 August 2020, https://www.canada.ca/en/agriculture-agri-food/news/2018/04/bc-wine-facts.html.
Agriculture and Agri-Food Canada, 2018b: Fraser Valley geographical indication. British Columbia Wine Authority, accessed 17 August 2020, http://www.bcvqa.ca/wp-content/uploads/2019/03/Fraser_Valley_GI-v-June-4-2018.pdf.
Amerine, M. A., and A. J. Winkler, 1944: Composition and quality of musts and wines of California grapes. Hilgardia, 15, 493–675, https://doi.org/10.3733/hilg.v15n06p493.
Anderson, K., S. Nelgen, and V. Pinilla, 2017: Global Wine Markets, 1860 to 2016: A Statistical Compendium. University of Adelaide Press, 582 pp.
Aney, W. A., 1974: Oregon climates exhibiting adaptation potential for vinifera. Amer. J. Enol. Vitic., 25, 212–218.
BC Wine Grape Council, 2019: Annual crop assessment 2019 vintage. Accessed 6 August 2020, https://bcwgc.org/crop-report/2019-crop-assessment.
BC Wine Institute, 2020a: Economic impact of BC wine industry. Accessed 2 October 2020, https://winebc.com/industry/media/quick-facts/.
BC Wine Institute, 2020b: The Fraser Valley. Accessed 4 August 2020, https://winebc.com/discover-bc-wine-country/fraser-valley/#top.
Belliveau, S., B. Smit, and B. Bradshaw, 2006: Multiple exposures and dynamic vulnerability: Evidence from the grape industry in the Okanagan Valley, Canada. Global Environ. Change, 16, 364–378, https://doi.org/10.1016/j.gloenvcha.2006.03.003.
Caprio, J. M., and H. A. Quamme, 2002: Weather conditions associated with grape production in the Okanagan Valley of British Columbia and potential impact of climate change. Can. J. Plant Sci., 82, 755–763, https://doi.org/10.4141/P01-160.
Cardell, M. F., A. Amengual, and R. Romero, 2019: Future effects of climate change on the suitability of wine grape production across Europe. Reg. Environ. Change, 19, 2299–2310, https://doi.org/10.1007/s10113-019-01502-x.
Croitoru, A., A. Piticar, A. M. Imbroane, and D. C. Burada, 2012: Spatiotemporal distribution of aridity indices based on temperature and precipitation in the extra-Carpathian regions of Romania. Theor. Appl. Climatol., 112, 597–607, https://doi.org/10.1007/s00704-012-0755-2.
deMartonne, E., 1926: Aréisme et indice d’aridité. C. R. Acad. Sci., 182, 1395–1398.
de Orduña, R. M., 2010: Climate change associated effects on grape and wine quality and production. Food Res. Int., 43, 1844–1855, https://doi.org/10.1016/j.foodres.2010.05.001.
Diffenbaugh, N. S., M. A. White, G. V. Jones, and M. Ashfaq, 2011: Climate adaptation wedges: A case study of premium wine in the western United States. Environ. Res. Lett., 6, 024024, https://doi.org/10.1088/1748-9326/6/2/024024.
Dunn, M., M. D. Rounsevell, F. Boberg, E. Clarke, J. Christensen, and M. S. Madsen, 2017: The future potential for wine production in Scotland under high-end climate change. Reg. Environ. Change, 19, 723–732, https://doi.org/10.1007/s10113-017-1240-3.
Environment and Climate Change Canada, 2017: Climate data: Homogenized surface air temperature data. Government of Canada, accessed 16 August 2020, https://www.canada.ca/en/environment-climate-change/services/climate-change/science-research-data/climate-trends-variability/adjusted-homogenized-canadian-data/surface-air-temperature.html.
Environment and Climate Change Canada, 2020: Historical data. Government of Canada, accessed 15 June 2020, https://climate.weather.gc.ca/historical_data/search_historic_data_e.html.
Fraga, H., A. C. Malheiro, J. Moutinho-Pereira, and J. A. Santos, 2013: Future scenarios for viticultural zoning in Europe: Ensemble projections and uncertainties. Int. J. Biometeor., 57, 909–925, https://doi.org/10.1007/s00484-012-0617-8.
Fraser Valley Regional District, 2011: Regional snapshot series: Agriculture, agricultural economy in the Fraser Valley Regional District. Accessed 16 August 2020, https://fraseropolis.files.wordpress.com/2016/03/2012-fraser-valley-agriculture-snapshot.pdf.
Gough, W. A., 2008: Theoretical considerations of day-to-day temperature variability applied to Toronto and Calgary, Canada data. Theor. Appl. Climatol., 94, 97–105, https://doi.org/10.1007/s00704-007-0346-9.
Hall, A., and G. V. Jones, 2010: Spatial analysis of climate in winegrape-growing regions in Australia. Aust. J. Grape Wine Res., 16, 389–404, https://doi.org/10.1111/j.1755-0238.2010.00100.x.
Hanes, C. C., X. Wang, P. Jain, M. Parisien, J. M. Little, and M. D. Flannigan, 2019: Fire-regime changes in Canada over the last half century. Can. J. For. Res., 49, 256–269, https://doi.org/10.1139/cjfr-2018-0293.
Hannah, L., and Coauthors, 2013: Climate change, wine, and conservation. Proc. Natl. Acad. Sci. USA, 110, 6907–6912, https://doi.org/10.1073/pnas.1210127110.
Hewer, M. J., and W. A. Gough, 2016a: The effect of seasonal climatic anomalies on zoo visitation in Toronto (Canada) and the implications for projected climate change. Atmosphere, 7, 71, https://doi.org/10.3390/atmos7050071.
Hewer, M. J., and W. A. Gough, 2016b: Assessing the impact of projected climate change on zoo visitation in Toronto (Canada). J. Geogr. Geol., 8, 30–48, https://doi.org/10.5539/jgg.v8n2p30.
Hewer, M. J., and W. A. Gough, 2019a: Assessing the impact of projected climate change on the future of grape growth and wine production in the Niagara Peninsula (Canada). J. Wine Res., 31, 6–34, https://doi.org/10.1080/09571264.2019.1699781.
Hewer, M. J., and W. A. Gough, 2019b: Lake Ontario ice coverage: Past, present and future. J. Great Lakes Res., 45, 1080–1089, https://doi.org/10.1016/j.jglr.2019.10.006.
Hewer, M. J., and M. Brunette, 2020: Climate change impact assessment on grape and wine for Ontario, Canada’s appellations of origin. Reg. Environ. Change, 20, 86, https://doi.org/10.1007/s10113-020-01673-y.
Hewer, M. J., and W. A. Gough, 2020: Using a multiyear temporal climate-analog approach to assess climate change impacts on park visitation. Wea. Climate Soc., 11, 291–305, https://doi.org/10.1175/WCAS-D-18-0025.1.
Hewer, M. J., N. Beech, and W. A. Gough, 2021: Development and validation of the Climate Model Confidence Index (CMCI): Measuring ability to reproduce historical climate conditions. Theor. Appl. Climatol., 144, 1059–1075, https://doi.org/10.1007/s00704-021-03581-5.
Holland, T., and B. Smit, 2014: Recent climate change in the Prince Edward County winegrowing region, Ontario, Canada: Implications for adaptation in a fledgling wine industry. Reg. Environ. Change, 14, 1109–1121, https://doi.org/10.1007/s10113-013-0555-y.
Huglin, P., 1978: Nouveau mode d’évaluation des possibilités héliothermiques d’un milieu viticole. C. R. Acad. Agric. Fr., 64, 1117–1126.
Jackson, D. I., and N. J. Cherry, 1988: Prediction of a district’s grape-ripening capacity using a latitude–temperature index (LTI). Amer. J. Enol. Vitic., 39, 19–28.
Jia, G., and Coauthors, 2019: Land–climate interactions. Climate Change and Land: An IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems, P. R. Shukla et al., Eds., Cambridge University Press, 131–248.
Jones, G. V., 2006: Climate and terroir: Impacts of climate variability and change on wine. Fine Wine and Terroir—The Geoscience Perspective, R. W. Macqueen and L. D. Meinert, Eds., Geological Association of Canada, 247 pp.
Jones, G. V., 2007: Climate change: Observations, projections, and general implications for viticulture and wine production. Climate and Viticulture Congress, Zaragoza, Spain, International Organization of Vine and Wine, https://chaireunesco-vinetculture.u-bourgogne.fr/colloques/actes_clima/Actes/Article_Pdf/Jones.pdf.
Jones, G. V., and G. Goodrich, 2008: Influence of climate variability on wine regions in the western USA and on wine quality in the Napa Valley. Climate Res., 35, 241–254, https://doi.org/10.3354/cr00708.
Jones, G. V., and H. R. Schultz, 2016: Climate change and emerging cool climate wine regions. Wine Vitic. J., 4, 51–53.
Jones, G. V., M. A. White, O. R. Cooper, and K. Storchmann, 2005: Climate change and global wine quality. Climatic Change, 73, 319–343, https://doi.org/10.1007/s10584-005-4704-2.
Jones, G. V., A. A. Duff, A. Hall, and J. W. Myers, 2010: Spatial analysis of climate in wine grape growing regions in the western United States. Amer. J. Enol. Vitic., 61, 313–326.
Jones, N. K., 2012: The influence of recent climate change on wine regions in Quebec, Canada. J. Wine Res., 23, 103–113, https://doi.org/10.1080/09571264.2012.678933.
Jones, N. K., 2018: An investigation of trends in viticultural climatic indices in southern Quebec, a cool climate wine region. J. Wine Res., 29, 120–129, https://doi.org/10.1080/09571264.2018.1472074.
Jones, N. K., 2019: Precipitation amounts and variability in a cool climate wine region, southern Quebec, Canada. J. Wine Res., 30, 322–334, https://doi.org/10.1080/09571264.2019.1652153.
Kelley, C. C., and R. H. Spilsbury, 1939: Soil survey of the lower Fraser Valley. Dominion of Canada—Department of Agriculture Tech. Bull. 650, 67 pp., https://sis.agr.gc.ca/cansis/publications/surveys/bc/bc1/bc1_report.pdf.
Kriedemann, P. E., 1968: Photosynthesis in vine leaves as a function of light intensity, temperature, and leaf age. Vitis, 7, 213–220, https://doi.org/10.5073/vitis.1968.7.213-220.
Ludlow, L., and S. Perez, 2018: Addressing autocorrelation in time series data: A comparison of four analytic methods using data from college course evaluations. Gen. Linear Model J., 44, 11–23, https://doi.org/10.31523/glmj.044001.002.
Meier, M., J. Fuhrer, and A. Holzkämper, 2018: Changing risk of spring frost damage in grapevines due to climate change? A case study in the Swiss Rhone Valley. Int. J. Biometeor., 62, 991–1002, https://doi.org/10.1007/s00484-018-1501-y.
Moriondo, M., G. V. Jones, B. Bois, C. Dibari, R. Ferrise, G. Trombi, and M. Bindi, 2013: Projected shifts of wine regions in response to climate change. Climatic Change, 119, 825–839, https://doi.org/10.1007/s10584-013-0739-y.
Moss, R. H., and Coauthors, 2010: The next generation of scenarios for climate change research and assessment. Nature, 463, 747–756, https://doi.org/10.1038/nature08823.
Naulleau, A., C. Gary, L. Prévot, and L. Hossard, 2021: Evaluating strategies for adaptation to climate change in grapevine production—A systematic review. Front. Plant Sci., 11, 607859, https://doi.org/10.3389/fpls.2020.607859.
Nesbitt, A., B. Kemp, C. Steele, A. Lovett, and S. Dorling, 2016: Impact of recent climate change and weather variability on the viability of UK viticulture—Combining weather and climate records with producers’ perspectives. Aust. J. Grape Wine Res., 22, 324–335, https://doi.org/10.1111/ajgw.12215.
Pachauri, R. K., and Coauthors, 2014: Climate Change 2014: Synthesis Report. Cambridge University Press, 151 pp., https://www.ipcc.ch/site/assets/uploads/2018/02/SYR_AR5_FINAL_full.pdf.
Pacific Climate Impacts Consortium, 2019: Statistically downscaled GCM scenarios—BCCAQ v2. University of Victoria, accessed 9 April 2020, https://data.pacificclimate.org/portal/downscaled_gcms/map/.
Rayne, S., and K. Forest, 2016: Rapidly changing climatic conditions for wine grape growing in the Okanagan Valley region of British Columbia, Canada. Sci. Total Environ., 556, 169–178, https://doi.org/10.1016/j.scitotenv.2016.02.200.
Romero-Lankao, P., J. B. Smith, D. J. Davidson, N. S. Diffenbaugh, P. L. Kinney, P. Kirshen, P. Kovacs, and L. Villers Ruiz, 2014: North America. Climate Change 2014: Impacts, Adaptation, and Vulnerability, Part B: Regional Aspects, V. R. Barros et al., Eds., Cambridge University Press, 1439–1498, https://www.ipcc.ch/site/assets/uploads/2018/02/WGIIAR5-Chap26_FINAL.pdf.
Roy, P., P. Grenier, E. Barriault, T. Logan, A. Blondlot, G. Bourgeois, and D. Chaumont, 2017: Probabilistic climate change scenarios for viticultural potential in Québec. Climatic Change, 143, 43–58, https://doi.org/10.1007/s10584-017-1960-x.
Schultze, S. R., and P. Sabbatini, 2019: Implications of a climate-changed atmosphere on cool-climate viticulture. J. Appl. Meteor. Climatol., 58, 1141–1153, https://doi.org/10.1175/JAMC-D-18-0183.1.
Shaw, A. B., 1999: The emerging cool climate wine regions of eastern Canada. J. Wine Res., 10, 79–94, https://doi.org/10.1080/09571269908718164.
Shaw, A. B., 2017: Climate change and the evolution of the Ontario cool climate wine regions in Canada. J. Wine Res., 28, 13–45, https://doi.org/10.1080/09571264.2016.1238349.
Tonietto, J., and A. Carbonneau, 2004: A multicriteria climatic classification system for grape-growing regions worldwide. Agric. For. Meteor., 124, 81–97, https://doi.org/10.1016/j.agrformet.2003.06.001.
Tóth, J., and Z. Végvári, 2015: Future of winegrape growing regions in Europe. Aust. J. Grape Wine Res., 22, 64–72, https://doi.org/10.1111/ajgw.12168.
van Leeuwen, C., and P. Darriet, 2016: The impact of climate change on viticulture and wine quality. J. Wine Econ., 11, 150–167, https://doi.org/10.1017/jwe.2015.21.
van Leeuwen, C., and Coauthors, 2019: An update on the impact of climate change in viticulture and potential adaptations. Agronomy, 9, 514, https://doi.org/10.3390/agronomy9090514.
van Vuuren, D. P., and Coauthors, 2011: The representative concentration pathways: An overview. Climatic Change, 109, 5–31, https://doi.org/10.1007/s10584-011-0148-z.
Vincent, L. A., X. Zhang, R. D. Brown, Y. Feng, E. Mekis, E. J. Milewska, H. Wan, and X. L. Wang, 2015: Observed trends in Canada’s climate and influence of low-frequency variability modes. J. Climate, 28, 4545–4560, https://doi.org/10.1175/JCLI-D-14-00697.1.
Vincent, L. A., X. Zhang, É. Mekis, H. Wan, and E. Bush, 2018: Changes in Canada’s climate: Trends in indices based on daily temperature and precipitation data. Atmos.–Ocean, 56, 332–349, https://doi.org/10.1080/07055900.2018.1514579.
Vineyards.com, 2020: Wine regions of British Columbia, Canada. Accessed 11 September 2020, https://vineyards.com/wine-map/canada/british-columbia.
White, M. A., N. S. Diffenbaugh, G. V. Jones, J. S. Pal, and F. Giorgi, 2006: Extreme heat reduces and shifts United States premium wine production in the 21st century. Proc. Natl. Acad. Sci. USA, 103, 11 217–11 222, https://doi.org/10.1073/pnas.0603230103.
Wilby, R., C. Dawson, and E. Barrow, 2002: SDSM—A decision support tool for the assessment of regional climate change impacts. Environ. Modell. Software, 17, 145–157, https://doi.org/10.1016/S1364-8152(01)00060-3.
Wine Growers Canada, 2017: Our industry. Accessed 20 August 2020, https://www.winegrowerscanada.ca/our-industry/economic-impact/.
Zargar, A., R. Sadiq, B. Naser, and F. I. Khan, 2011: A review of drought indices. Environ. Rev., 19, 333–349, https://doi.org/10.1139/a11-013.
