Abstract
Background and goals Climate change induces shifts in suitable areas for the cultivation and phenology of grapevines, which regulate both complex sunlight patterns within hedgerows and canopy photosynthetic gain (CPG). However, suitability assessments of such areas have focused on climate indices (e.g., growing season temperature) but have ignored canopy photosynthesis, despite its close connection to berry yield and wine quality. Here, we aimed to develop a model of theoretical maxima of grapevine canopy photosynthesis under the clear sky without water- and nutrient-limitations that integrated canopy light interception, leaf thermal balance, and photosynthetic processes.
Methods and key findings Maps of CPG during berry development were obtained by running this model with observed hourly meteorological data collected from various sites worldwide and those under future climate with elevated air temperatures and atmospheric carbon dioxide (CO2) concentrations (+2 K warmer, 500 μmol/mol and +4 K warmer, 850 μmol/mol). In high-latitude regions expected to become newly suitable in response to the temperature rise, the CPGs were projected to be greater than those in lower latitudes mainly because of higher cumulative light absorption, irrespective of a difference in cultivar-dependent thermal requirements for berry maturation. The higher CO2 concentrations also increased canopy photosynthesis in these regions. By contrast, at some prestigious sites located at lower latitudes, the projected leaf temperatures exceeded the optimal range for photosynthesis, resulting in reduced gains despite the CO2 fertilization effect.
Conclusions and significance Our results clarify significant spatiotemporal variations in grapevine canopy photosynthesis, with particular emphasis on the north-south gradient, and support the decision-making of grapegrowing stakeholders.
Introduction
Wine is an agricultural product closely connected to cultural and economic activities. Climate change threatens the leading winemaking regions in southern Europe (Fraga et al. 2016, Sgubin et al. 2023) and may require the relocation of vineyards to cooler sites at higher latitude or elevation (Hannah et al. 2013, Jones and Schultz 2016). Comprehensively understanding the suitability of viticulture to define leading grapegrowing sites is becoming increasingly important. Previous studies have proposed various agroclimate indices and analyzed the suitability in response to climate change at regional and global scales (e.g., Schultz and Jones 2010). These studies have focused on the feasibility of grapevine cultivation (e.g., the thermal requirements for berry maturation, water availability during the growing period, and/or winter survival). Another concern is the canopy photosynthetic gain (CPG) during the growing period. The CPG before anthesis regulates fruit set and berry yield (Poni et al. 2008), and the gain during berry development affects berry qualities such as soluble sugar content (Smith et al. 2019).
Earlier studies on modeling of crop primary production have expressed canopy photosynthesis as a product of two factors: light absorption and utilization efficiency of the absorbed light (Monteith 1977, Long et al. 2006). This concept has also been used in process-based growth models applied to viticultural simulations (Moriondo et al. 2015). The simplest approach for estimating canopy photosynthesis involves the use of the “big-leaf” model: estimating canopy photosynthesis considering the canopy to be a single leaf, and calculating the photosynthetic rate from its light absorption and a photosynthetic light response curve. Bindi et al. (1996) used this approach and simulated grapevine growth in Bologna, Italy in historical and future climates with different air temperatures and carbon dioxide (CO2) concentrations. Several modeling frameworks developed for major crops have been modified for grapevines and successfully simulated canopy dry matter and/or berry yield from local meteorological data (Poni et al. 2006, Cola et al. 2014, Fraga et al. 2015). By projecting such models to future climate scenarios, the effect of climate change on the yield, phenology, and water balance of grapevines has been evaluated in Europe (Fraga et al. 2016). In these studies, using the big-leaf approach, leaf photosynthetic responses to CO2 concentration and air temperature were sometimes considered using empirical coefficients, or ignored, leading to difficulties in estimating the interaction between the two factors and in understanding the biochemical limitations of assimilation.
Moreover, these models were run at a daily time-step, ignoring the diurnal variations in canopy light environments and photosynthetic activities originating from the hedgerow architecture common in vineyards (Smart 1973, Intrieri et al. 1998, Hunter et al. 2016). This point is particularly important in the context of climate change adaptation of viticulture. Shifts in the growing period due to the expected advancement of phenology (Webb et al. 2007) and relocation along latitude would differentiate temporal light environments in hedgerows and invalidate pre-parameterized models specialized for a certain region. Furthermore, the diurnal pattern of leaf temperature may also modify photosynthetic light response curves by affecting photosynthetic processes (Zufferey et al. 2000, Schultz 2003, Greer 2018). Recent studies have used three-dimensional models to quantify the temporal dynamics and intracanopy heterogeneity of grapevine photosynthesis and account for complex light profiles within the canopy (López-Lozano et al. 2011, Zhu et al. 2018, Prieto et al. 2020). For example, a study combined 3-D ray tracing and photosynthetic characteristics mapped on a 3-D reconstruction of canopy architecture digitized from an actual grapevine canopy in Montpellier and detailed diurnal courses of leaf- and canopy-level photosynthetic performance (Prieto et al. 2020). These models have provided valuable insights for comprehensively understanding canopy photosynthesis, although the large requirement of computational resources may hinder its application in global- and regional-scale analyses. Therefore, the effects of climate change on canopy photosynthesis of grapevines at a global scale has not been assessed with a model that incorporates complex light environments specific to the hedgerow cultivation system.
To facilitate the implementation of adaptation measures against climate change and to support viticultural stakeholders in decision-making, we simulated the potential photosynthetic gains of hedgerow grapevine canopies during the growing period using air temperatures and atmospheric CO2 concentrations according to climate change scenarios, at more than 1000 sites worldwide (Supplemental Figure 1). To achieve this, we developed a simplified canopy photosynthesis model for hedgerow grapevines, which incorporates several key factors identified through previous simulations and experiments: 1) the diurnal patterns of light absorption of hedgerows, which are influenced by hedge dimensions, site geography, and solar positioning; 2) the energy balance of leaves, which determines leaf temperature under specific meteorological conditions; and 3) the leaf-scale mechanistic responses of photosynthesis to environmental factors, based on a model combining biochemical CO2 fixation processes and stomatal responses.
Materials and Methods
The present model requires hourly environmental variables (e.g., air temperature, relative humidity [RH], direct and diffuse solar irradiance, and atmospheric CO2 concentration), leaf photosynthesis-related parameters (e.g., stomatal conductance, maximum rates of carboxylation and electron transport, and their temperature dependencies), and geometrical properties of the hedgerows. These parameters are described in the following sections and listed in Supplemental Tables 1 and 2. The schematic diagram of the present model is shown in Supplemental Figure 2. All analyses, data handling, and visualization were performed using the R statistical software (ver. 4.2.2; R Core Team 2022).
Meteorological data under present and future climates
Daily minimum and maximum air temperatures observed worldwide between 2018 and 2022 were sourced from a web page operated by the Japan Meteorological Agency (ClimatView; https://www.data.jma.go.jp/cpd/monitor/dailyview/index.php). In total, 1266 sites with more than three years of data were selected such that they uniformly covered land areas, as well as major wine areas (Supplemental Figure 1). The mean values of the daily temperatures from three to five years were used to estimate hourly air temperatures via interpolation (Parton and Logan 1981). This method separately simulates day and night temperature patterns using sinusoidal and exponential curves, respectively. Empirical coefficients of the diurnal pattern of 1.5 m air temperature (Parton and Logan 1981) were used for the interpolation. For simplicity, it was assumed that the air temperature was minimum at sunrise. The time of sunrise and day length were estimated from the site latitude and day of the year (Forsythe et al. 1995). The RH was calculated from the temperatures according to the Food and Agriculture Organization of the United Nations' protocol for estimating missing climatic data (Allen et al. 1998). This protocol assumes that water vapor saturates when the air temperature is minimum and that the water vapor pressure is constant throughout the day.
To obtain environmental data for future conditions, a large ensemble climate simulation database called d4PDF (Mizuta et al. 2017, Fujita et al. 2019) was used. The hourly values of air temperature and RH were corrected by referring to the projected changes. The changes were calculated by comparing the means from 30 ensemble members (= 300 years) of the present (HIST; 2001 to 2010 of historical climate simulations) and future (+2 K; 2041 to 2050 of +2 K warmer future simulations, and +4 K; 2100 to 2110 of +4 K warmer future simulations) climates. Hourly temperatures under future climates were calculated by adding the differences in the daily mean air temperatures at the corresponding grid cells. From the monthly values of mean air temperatures and RH at a height of 2 m in the historical, +2 K, and +4 K future experiments, the monthly mean vapor pressure in the three experiments was calculated using the Tetens’ equation (Tetens 1930). Differences in vapor pressure between the present and future climates at a grid that covers a specific site were added to estimate the values under future climatic conditions. Based on these results, the RH was calculated for each scenario and site. Atmospheric CO2 concentrations of 400, 500, and 850 µmol/mol were used for simulations under the HIST, +2 K, and +4 K conditions, respectively.
Growing degree days (GDD), phenology, and chilling requirement
Flowering and maturity dates were determined from the GDD as follows:
Eq. 1
where Tmean,i (°C) is the daily mean air temperature on day i and Tbase is the base temperature fixed at 10°C. Integration was initiated on 21 Dec and 21 June (i.e., winter solstice) in the northern and southern hemispheres, respectively. The GDD criterion for flowering, 330°C d, corresponds to the temperature required to reach flowering in several early cultivars (van Leeuwen et al. 2008). Three GDD criteria were defined to implement cultivar differences in thermal demand for maturity (1250, 1500, and 1750°C). These criteria were based on a previous phenological study covering many grapegrowing regions (van Leeuwen et al. 2008). Note that van Leeuwen et al. (2008) used harvest dates to estimate GDD values required for maturation.
In addition to this thermal requirement for maturation, grapevines require a certain chilling hour during winter for normal bud growth (Erez 2000). Chilling hour was defined as the number of hours exposed to cold air with temperatures of 0 to 7.2°C during winter (October to March and April to September for the northern and southern hemispheres, respectively), and the threshold was set as 650 hrs, based on a previous study (Rahemi et al. 2021).
Light absorption, energy balance, and photosynthesis of grapevine leaves and CPG
A simple canopy model consisting of hedges approximated by thin boxes on a uniform slope was used (Supplemental Figure 3) to simulate cumulative CPG during the growing period (flowering to berry maturation). The model divides leaves on the hedge surface into light leaves (illuminated by direct sunlight) and shade leaves (which received diffuse radiation owing to shade provided by a neighboring hedge). For simplicity, we did not consider mutual shading (i.e., shade provided by other leaves of the same plants). Light absorption, energy balance, and photosynthesis were calculated separately. From the hedge dimensions, slope angle, and solar positioning (as a function of time), the hourly irradiance on the light and shade leaves was estimated (Supplemental Equation 1 [Equations 1 to 8] and Supplemental Figure 3). Although there can be a significant increase in canopy leaf area (e.g., Williams and Ayars 2005), for simplicity, the total leaf area and hedge dimension (i.e., canopy height and width) were considered to be constant throughout the growing period.
Hourly leaf temperatures (Tleaf, °C) were estimated based on the energy balance using the tleaf function in the R package ‘tealeaves’ (Muir 2019). Among the input parameters, air temperature (Tair, °C), RH (%), and irradiance per unit leaf area (Ilight and Ishade, W/m2) were obtained using the aforementioned procedures. The characteristic leaf dimension (dleaf), cuticle conductance (gc), and wind speed (u) were fixed at 0.11 m, 5 mmol/m2/s/Pa, and 2.0 m/s, respectively. A fixed value of stomatal conductance at 0.10 mol/m2/s approximated from diurnal patterns of nonstressed grapevine leaves was provided to estimate leaf temperature. This value was used only for the estimation of leaf temperature, and the effect of stomatal conductance on gas exchange was simulated independently, as described in the following paragraph. This parameter was confirmed to exert a small effect (usually smaller than 0.5°C in light leaves) on the leaf temperatures.
The net photosynthetic rates of the light and shade leaves were estimated using the Photosyn function in the R package ‘plantecophys’ (Duursma 2015). This function calculates leaf gas exchange by coupling the Farquhar-von Caemmerer-Berry model of leaf photosynthesis (Farquhar et al. 1980) and a model of stomatal response to the environment (Ball et al. 1987, Leuning 1995) (see also Equations 9 to 22 in Supplemental Equation 2). Temperature-dependent photosynthetic and stomatal parameters of grapevine leaves were adopted from two studies (Schultz 2003, Prieto et al. 2012) and corrected by referring to Tleaf calculated based on the leaf energy balance. Mean values of the two photosynthesis models are shown otherwise noted. Photosynthetic down-regulation of the maximum Rubisco carboxylation rate (Vcmax), maximum electron transport rate (Jmax), and rate of CO2 evolution in the light (day respiration) (Rd) in response to high growing period CO2 concentrations (500 and 850 µmol/mol for +2 K and +4 K conditions, respectively) were corrected using the empirical factors obtained from a meta-analysis (Poorter et al. 2022). The parameters for light and shade leaves were adopted from sun and shade leaves (Schultz 2003), and from outer and inner leaves (Prieto et al. 2012), respectively. These models adopted different stomatal responses and temperature kinetics of Rubisco, a key enzyme involved in photosynthetic carbon assimilation.
In the model of Schultz (2003) parameterized using Zinfandel and Riesling, leaf photosynthetic rate was limited mostly by electron transport, because stomatal conductance and Rubisco affinity to CO2 tended to be high. An increase in atmospheric CO2 concentrations due to climate change is projected to marginally promote leaf photosynthetic rates in this model. In the model of Prieto et al. (2012) parameterized using Syrah, leaf photosynthetic rate was limited by both electron transport and carboxylation reactions under the present climate because of the milder openness of stomata and lowered Rubisco affinity to CO2. Under the future climates with higher atmospheric CO2 concentrations, the rate increased substantially, owing to the shift in the limiting step from the carboxylation reaction to electron transport. The estimated net photosynthetic rates were higher in Schultz’s model (2003) under moderate air temperatures (20 to 30°C), and in the model by Prieto et al. (2012) under low and high temperatures (Supplemental Figure 4). The boundary layer resistance was ignored; it was assumed that the CO2 concentration at the leaf surface was the same as the atmospheric CO2 concentration. Ground-area-basis daily CPG was estimated by integrating the instantaneous values (time-step = 1 hr) of light and shade leaves.
Results and Discussion
Interaction of row orientation and site latitude and its effect on canopy light absorption and photosynthesis
Light and photosynthetic profiles of grapevine leaves on north-south (NS) and east-west (EW) hedgerows on a 5° slope directed to the equator were simulated (Supplemental Figure 3). The daily total amount of absorbed photons was smaller in EW rows than in NS rows, particularly in low- to middle-latitude zones (Figure 1A and Supplemental Figure 5), consistent with the results of previous research (Iandolino et al. 2013, Campos et al. 2017). This variation in light availability, based on site latitude and row orientation, shaped the theoretical upper limits of CPG, which interacted with the meteorological conditions of the site (Figure 1B). The simulated gain was greater in the NS rows than in the EW rows, particularly in low- to middle-latitude zones, which was attributable to the differences in temporal patterns of light absorption, as well as the amount of light absorption. Light absorption of EW rows peaked at 1200 hrs, when leaf temperature often exceeded the optimal range for photosynthesis (∼30°C; Schultz 2003) and was above the light saturation point of photosynthesis (∼1000 µmol/m2/s; Zufferey et al. 2000), leading to lower photosynthetic light use efficiency (Supplemental Figures 6 to 10). In contrast, the NS rows captured low-intensity sunlight in the cooler morning and exhibited efficient photosynthesis. This light absorption pattern of NS rows contributed to the greater CPG, compared to EW rows. These results indicate that diurnal radiation pattern and row orientation are necessary for accurate estimation of CPG of hedgerow grapevines. Although daily global solar radiation on a horizontal surface is sometimes used to estimate photosynthetic gains of horizontally-uniform major crops, this modeling approach is not suitable for hedgerow grapevines. Our simulation demonstrates that NS rows and high-latitude sites have an advantage over EW rows and low-latitude sites, in terms of potential CPG.
Daily light absorption (A) and daily photosynthetic gain of grapevine canopy under the present (B1, B4, B7: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios (B2, B5, B8: +2 K [2041 to 2050 of +2 K warmer future simulations], 500 µmol/mol and B3, B6, B9: +4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol) as a function of latitude zones in three timeframes: spring (A1 and B1 to B3: 10 May and 10 Nov in the northern and southern hemispheres, respectively), around summer solstice (A2 and B4 to B6: 20 June and 20 Dec), and summer (A3 and B7 to B9: 10 Aug and 10 Feb). Mean values with east-west (EW) and north-south (NS) hedges are shown with small dots representing the values of individual sites (n = 1266; see also Figure 2 and Supplemental Figures 1 and 10).
Effects of climate change on canopy photosynthesis and its spatial distribution
In simulations with elevated air temperature and non-elevated atmospheric CO2 concentration (i.e., +4 K, 400 µmol/mol), CPG was projected to decrease in low- to middle-latitude zones (up to ∼50°) during the summer in NS rows (Figure 2D, 2E, 2G, and 2H) and EW rows (Supplemental Figure 11D, 11E, 11G, and 11H). This is because leaf temperature would be outside the optimal range for gross photosynthesis. In high-latitude regions, increase in air temperature is predicted to enhance leaf photosynthesis, particularly in spring, when low air temperatures may limit photosynthesis (Figure 2A and 2B and Supplemental Figure 11A and 11B). When both elevated air temperature and increased CO2 concentration are considered (i.e., +4 K, 850 µmol/mol versus HIST, 400 µmol/mol), it is expected that net CPG will be greater or similar to current levels in most locations and seasons (Figure 2A, 2C, 2D, 2F, 2G, and 2I and Supplemental Figure 11A, 11C, 11D, 11F, 11G, and 11I) because of the CO2 fertilization effect (Long et al. 2004). It was projected that cumulative respiration during the growing period may increase from 1 to 3% and from 5 to 10% under +2 K, 500 µmol/mol and +4 K, 850 µmol/mol scenarios, respectively (Supplemental Data 1); this cost was mostly offset by the CO2 fertilization effect. However, in middle-latitude zones covering elite vineyards, the effect of climate change on photosynthesis may be positive or negative, depending on local meteorological conditions. In some hot summer sites, leaf temperatures may sometimes reach 40°C (e.g., Supplemental Figure 9); thus, the negative effect of rising temperatures may outweigh the positive effect of increased CO2. These sites include several European grapegrowing regions, which are predicted to undergo intensive warming in the future (Supplemental Figure 12). By contrast, cooler summer sites can benefit considerably from climate change, because the negative effect of rising temperatures is expected to be small and the positive effect of CO2 enrichment can be completely realized.
Site-dependent variations in daily canopy photosynthetic gain. Daily gains of north-south hedges on three different timeframes under the present (A, D, G: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios, +4 K (2100 to 2110 of +4 K warmer future simulations) warmer climates with and without CO2 enrichment (C, F, I: +4 K, 850 µmol/mol and B, E, H: +4 K, 400 µmol/mol, respectively), are shown. Representative sites are highlighted with labels and other sites are shown in gray.
We further simulated and compared cumulative CPGs during the growing period to consider probable shifts in grapevine phenology under future climatic conditions (Webb et al. 2007, Ramos et al. 2018). The growing period was determined based on GDD, which is calculated as the sum of daily average air temperatures above a baseline temperature of 10°C (Amerine and Winkler 1944). Cumulative CPGs were calculated by integrating daily gains from flowering to maturity for early (e.g., Pinot noir and Chardonnay), middle (e.g., Merlot and Cabernet Sauvignon), and late-ripening cultivars (e.g., Grenache and Cinsaut). Maps of cumulative CPGs revealed the spatial distributions and transitions in the photosynthetic potentials (Figure 3). In accordance with previous studies that used agro-climate indices (e.g., Hannah et al. 2013), suitable areas with cumulative temperatures during the growing period that are sufficient for maturity are expected to expand in high latitude zones because of climate change (Supplemental Table 3 and Supplemental Data 2). The present analysis further demonstrated that sites located at the high-latitude forefront, rather than those located at the middle of suitable areas, facilitated greater cumulative CPGs.
Spatial distributions of cumulative canopy photosynthetic gains over the (A) European, (B) North American, and (C) Oceanian regions. Estimated gains under the present climate (left panels: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios (central panels: +2 K [2041 to 2050 of +2 K warmer future simulations], 500 µmol/mol; right panels: +4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol), and those for early (top panels), middle (center panels), and late ripening cultivars (bottom panels) are shown. The gains are mapped on symbol colors when the site meets both heat and chilling requirements (circles) or the heat requirement only (triangles); sites without sufficient heat requirement are indicated with crosses.
Similar trends were found in other regions of the world (Supplemental Figures 13 to 15). These maps indicated that high cumulative CPG can be attained by cultivating grapevines at the coldest sites in suitable regions, suggesting the importance of improving and developing measures for cold-climate viticulture (e.g., measures against winter wilt and late-spring frost). This advantage of high latitude over low latitude was attributed to the greater amount of total light absorption, longer growing period duration (Supplemental Figure 16), and more efficient use of absorbed photons, as discussed in the previous section. Projected earlier ripening and shortening of the growing period under future climate conditions, which have already been detected in recent decades (Webb et al. 2011, Cook and Wolkovich 2016), would not decouple the correlation between cumulative photosynthetic gain and site latitude. Furthermore, at some low-latitude sites, the necessary duration of chilling required for normal spring bud development (∼650 hrs at an air temperature of 0 to 7.2°C [Rahemi et al. 2021]) may not be achieved because of warmer winters (Figure 3 and Supplemental Figures 13 to 15). This may increase risks of delayed and/or asynchronous spring budbreak (Erez 2000). Notably, violation of this criterion may not be fatal, as the chilling hour requirement represents significant genetic variation (Londo and Johnson 2014) and has not been met at several current grapegrowing sites with mild winters (e.g., Cape Town and San Francisco).
In our analysis, we found several sites at lower latitudes that exhibited exceptionally high cumulative CPGs, notwithstanding the relationship between site latitude and cumulative CPG (Figure 4). Most of these sites were located at high elevations and/or at sites with small seasonal temperature fluctuations (e.g., islands). Our simulation suggested that some sites in the southern hemisphere may exhibit a greater cumulative CPG during the growing period. This was partially because of the smaller Sun–Earth distance in the summer months in the southern hemisphere, leading to greater light availability during the growing period. It must be kept in mind that greater cumulative CPGs at some sites should lead to greater dry matter production and perhaps berry yields, while it can fuel undesirable vegetative growth, resulting in poor berry quality. Precise monitoring and control of the balance between vegetative and reproductive growth is important under future conditions in these regions.
Cumulative canopy photosynthetic gains from flowering to berry maturity. Cumulative gains as a function of site latitude under three climate scenarios and three heat requirements for maturity are shown. Estimated gains under the present climate (A, D, G: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios (B, E, H: +2 K [2041 to 2050 of +2 K warmer future simulations], 500 µmol/mol and C, F, I: +4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol right panels), and those for early (A to C), middle (D to F), and late ripening cultivars (G to I) are shown. Circles and triangles indicate sites that do or do not satisfy chilling requirements during winter, respectively. Representative sites are highlighted with labels and the other sites are shown in gray.
The present results suggest that projected changes in air temperature and CO2 concentration may not cause a drastic reduction in the potential cumulative CPG of the grapevine canopy at low latitude sites. However, in low-latitude regions projected to be vulnerable to drought, owing to a substantial reduction in precipitation (e.g., southern Europe; Fraga et al. 2016), potential gains would be rarely attained. Even if grapevine canopies exert their photosynthetic potential without experiencing water stress in these regions, their gains would be smaller than those at higher latitudes (Figures 3 and 4). This finding implies that irrigation alone may not be the panacea for vineyards in low-latitude regions. Continuous improvements in management practices and breeding techniques are required, particularly aimed at heat (Venios et al. 2020) and drought tolerance (Gambetta et al. 2020).
Model limitations and implication for further studies
Our canopy photosynthesis model has several limitations. This model does not consider damages due to cold weather conditions such as winter injury and late-spring frost damage (Rahemi et al. 2022). Grapevines at high-latitude sites can be damaged frequently by these events. In addition, the assumption of a constant canopy dimension during the growing period is not always satisfied in actual vineyards. This is particularly the case in high-latitude regions, where the increase in canopy leaf area is active around anthesis. These factors lead to optimistic estimates of the CPG in favor of high-latitude viticulture.
Furthermore, our model adopts a simple approach that separates the canopy into two parts (i.e., light and shade leaves), although the present model reproduced typical diurnal patterns of canopy photosynthesis in different row orientations (Intrieri et al. 1998). Several earlier studies developed 3-D models of grapevine canopy that simulate light intercept (Louarn et al. 2008, Iandolino et al. 2013) and leaf photosynthesis (Prieto et al. 2012, Zhu et al. 2018, Albasha et al. 2019). The use of these 3-D models in combination with climate scenario data sets would provide more accurate estimates of climate change effects on viticulture.
Another major limitation is that this model does not incorporate irreversible dysfunction to the photosynthetic systems caused by drought and heat stresses. Under high light with these abiotic stresses, leaf photosynthesis is suppressed (i.e., photoinhibited), as photodamage to photosynthetic apparatus exceeds the rate of its repair (Nishiyama and Murata 2014). Grapevine is a typical drought-resistant plant (Chaves et al. 2010) adapted to the dry Mediterranean climate, although some studies report the dysfunction caused by drought and heat stress under field conditions (e.g., Palliotti et al. 2009). Drought-induced embolism is uncommon in grapevines under historical climates in two major cultivation areas (non-irrigated Bordeaux and irrigated Napa); under the most extreme conditions observed over approximately a decade in these areas, the minimal water potential of stems never reached the point of no return for recovery (Charrier et al. 2018). The photosynthetic apparatus of grapevine leaves is rarely damaged unless stomatal conductance falls below a certain threshold value (∼0.10 to 0.15 mol/m2/s) (Flexas et al. 2004), which was rarely observed in the present simulation. However, more frequent and intensified drought hazards have been predicted by the latest climate projection models, with a high level of confidence (IPCC 2021). Because the present simulation shows theoretical maximal photosynthesis on clear days without any environmental stress, the risks associated with such extreme events should be considered separately.
In addition, photosynthetic acclimation to elevated air temperature and CO2 concentration may introduce uncertainty. The temperature-response curves of Vcmax and Jmax have been reported to change in response to the growing period temperatures (Yamori et al. 2010). This acclimation was not considered in the present simulation. Gallo et al. (2021, 2022) imposed 1.5 to 3.6°C warming on two grapevine varieties using open-top chambers, and assessed shifts in the temperature-response curves. They found significant acclimation in Grenache and little effects in Syrah, suggesting intraspecific variation of strategies to cope with high temperatures. As well as the temperature acclimation, photosynthesis of leaves was reported to decrease in response to long-term CO2 enrichment, known as down-regulation (Ainsworth and Long 2005), although this effect has been corrected using empirical factors based on the latest meta-analysis (Poorter et al. 2022). While Vcmax is expected to decrease in response to higher atmospheric CO2 concentrations, acclimation to higher growth temperatures may shift the optimal temperature for carboxylation (Hikosaka et al. 2006) and partially offset this negative effect.
Several experiments have assessed the effects of the simultaneous elevation of CO2 concentration and air temperature on grapevine photosynthesis (Salazar-Parra et al. 2012, 2015, Martínez-Lüscher et al. 2015, Edwards et al. 2017). A three-year experiment using open-top chambers with a heating and CO2 supply system showed that CO2 enrichment enhanced the net photosynthetic rates of leaves under saturating light by 30 to 40% and did not reduce the rate over the season in the Shiraz canopy, which was managed using industry-best practice (Edwards et al. 2017). A greenhouse experiment with elevated CO2 and temperature using cuttings of Tempranillo also reported that the leaf photosynthetic rate under the growth conditions remained higher than that of the control plants until maturity (Martínez-Lüscher et al. 2015). Free-air CO2 enrichment experiments that continued for multiple seasons with Riesling and Cabernet Sauvignon canopies revealed that the CO2 enrichment treatment enhanced leaf photosynthesis throughout the growing periods and increased berry yields (Wohlfahrt et al. 2018); similar results were reproduced in different years (Kahn et al. 2022). In Vitis labrusca cv. Concord cuttings grown under normal and elevated CO2 concentrations, leaf photosynthetic rates were consistently higher in the treated plants after 24 days of treatment without clear down-regulation (da Silva et al. 2017). By contrast, Salazar-Parra et al. (2012, 2015) observed that plants grown under elevated CO2 and temperature transiently exhibited higher leaf photosynthesis, although this positive effect diminished within ∼20 days, indicating clear down-regulation in photosynthesis. Other studies have reported an increase in dry matter production in response to elevated CO2 concentration (Bindi et al. 2001) and the combination of elevated CO2 concentration and air temperature (Kizildeniz et al. 2018a, 2018b, Arrizabalaga-Arriazu et al. 2020), suggesting that climate change will possibly enhance canopy photosynthesis, although it is partially compensated by down-regulation.
Conclusion
Here, we propose a simple model of hedgerow grapevine canopy photosynthesis and provide a global perspective on grapevine canopy photosynthesis under changing climates. The amount of light absorbed by grapevine hedges during the growing period was greater in higher latitude than in lower latitude, and likewise higher in NS- than in EW-oriented hedges. Temperature rises were projected to satisfy the thermal requirement for berry maturation in high-latitude regions. These high-latitude sites would represent greater photosynthetic gains than lower latitudes because of the greater light absorption. The higher CO2 concentrations under the future climate conditions synergistically promote canopy photosynthesis in these regions. By contrast, in some sites located at lower latitudes, the projected leaf temperatures exceeded the optimal range for photosynthesis, leading to reduced gains despite the CO2 fertilization effect. Our analysis demonstrated spatial heterogeneity on the effects of climate change on cumulative photosynthetic gains in viticulture.
Supplemental Data
The following supplemental materials are available for this article in the Supplemental tab above:
Supplemental Table 1 List of parameters used in the present model.
Supplemental Table 2 Temperature dependency of the photosynthetic parameters.
Supplemental Table 3 Effects of climate change on the number of suitable sites for viticulture, compared across present (HIST, 2001 to 2010 of historical climate simulations) and two future (+2 K, 2041 to 2050 of +2 K warmer future simulations; +4 K, 2100 to 2110 of +4 K warmer future simulations) climates.
Supplemental Figure 1 Location of sites subjected to the present analysis. Site elevations are indicated by the color of the symbols (n = 1266).
Supplemental Figure 2 Schematic diagram of the present model consists of meteorology, phenology, and canopy photosynthesis modules. HIST, present climate, 2001 to 2010 of historical climate simulations; +2 K, future climate, 2041 to 2050 of +2 K warmer future simulations; +4 K, future climate, 2100 to 2110 of +4 K warmer future simulations; RH, relative humidity; GDD, growing degree day; DOY, day of year.
Supplemental Figure 3 Schematic representation of a hedgerow vineyard light simulation model. Perspective views of east-west (EW) (A and B) and north-south (NS) (C and D) hedges with (A and C) and without (B and D) shade by the neighboring hedges. Further details are in the Materials and Methods.
Supplemental Figure 4 Net photosynthetic rates of light and shade leaves in response to photosynthetic photon flux density (PPFD), air temperature, and relative humidity (RH) of two models adopted in the present study.
Supplemental Figure 5 Diurnal patterns of canopy light absorption. Hourly values of leaf photon absorption per unit ground area of east-west (EW) and north-south (NS) hedges (red and blue lines, respectively) of five representative sites. The patterns in three timeframes are shown: spring (top panels: 10 May and 10 Nov in the northern and southern hemispheres, respectively), around summer solstice (middle panels: 20 June and 20 Dec), and summer (bottom panels: 10 Aug and 10 Feb). Numbers at the top of panels are daily integral absorption and absorbed fractions relative to incident radiation on horizontal surface (gray lines).
Supplemental Figure 6 Diurnal patterns of canopy photosynthesis and temperatures in Cape Town, South Africa (33.97°S). Hourly values of net canopy photosynthetic rates based on two photosynthesis models (blue, solid lines based on Schultz 2003 and blue, dashed lines based on Prieto et al. 2012), temperatures of leaves exposed to direct radiation or shaded by the neighboring hedge (dark red and orange lines, respectively), and air temperature (gray lines) of east-west (EW) and north-south (NS) hedges in three timeframes under (A) the present climate (HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and (B) the future climate (+4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol) are shown. Numbers at the top of panels are daily integral photosynthetic gains per unit ground area.
Supplemental Figure 7 Diurnal patterns of canopy photosynthesis and temperatures in San Francisco, CA (37.62°N). Hourly values of net canopy photosynthetic rates based on two photosynthesis models (blue, solid lines based on Schultz 2003 and blue, dashed lines based on Prieto et al. 2012), temperatures of leaves exposed to direct radiation or shaded by the neighboring hedge (dark red and orange lines, respectively), and air temperature (gray lines) of east-west (EW) and north-south (NS) hedges in three timeframes under (A) the present climate (HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and (B) the future climate (+4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol) are shown. Numbers at the top of panels are daily integral photosynthetic gains per unit ground area.
Supplemental Figure 8 Diurnal patterns of canopy photosynthesis and temperatures in Sapporo, Japan (43.05°N). Hourly values of net canopy photosynthetic rates based on two photosynthesis models (blue, solid lines based on Schultz 2003 and blue, dashed lines based on Prieto et al. 2012), temperatures of leaves exposed to direct radiation or shaded by the neighboring hedge (dark red and orange lines, respectively), and air temperature (gray lines) of east-west (EW) and north-south (NS) hedges in three timeframes under (A) the present climate (HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and (B) the future climate (+4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol) are shown. Numbers at the top of panels are daily integral photosynthetic gains per unit ground area.
Supplemental Figure 9 Diurnal patterns of canopy photosynthesis and temperatures in Burgundy, France (47.27°N). Hourly values of net canopy photosynthetic rates based on two photosynthesis models (blue, solid lines based on Schultz 2003 and blue, dashed lines based on Prieto et al. 2012), temperatures of leaves exposed to direct radiation or shaded by the neighboring hedge (dark red and orange lines, respectively), and air temperature (gray lines) of east-west (EW) and north-south (NS) hedges in three timeframes under (A) the present climate (HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and (B) the future climate (+4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol) are shown. Numbers at the top of panels are daily integral photosynthetic gains per unit ground area.
Supplemental Figure 10 Diurnal patterns of canopy photosynthesis and temperatures in Oslo, Norway (59.93°N). Hourly values of net canopy photosynthetic rates based on two photosynthesis models (blue, solid lines based on Schultz 2003 and blue, dashed lines based on Prieto et al. 2012), temperatures of leaves exposed to direct radiation or shaded by the neighboring hedge (dark red and orange lines, respectively), and air temperature (gray lines) of east-west (EW) and north-south (NS) hedges in three timeframes under (A) the present climate (HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and (B) the future climate (+4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol) are shown. Numbers at the top of panels are daily integral photosynthetic gains per unit ground area.
Supplemental Figure 11 Site-dependent variations in daily canopy photosynthetic gain. Daily gains of east-west hedges on three different timeframes under the present climate (A, D, G: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and +4 K (2100 to 2110 of +4 K warmer future simulations) warmer climates with and without CO2 enrichment (C, F, I: +4 K, 850 µmol/mol and B, E, H: +4 K, 400 µmol/mol, respectively) are shown. Representative sites are highlighted with labels and other sites are shown in gray.
Supplemental Figure 12 Projected rise in growing season temperature. Differences in mean values of air temperatures at a height of 2 m between the present (HIST, 2001 to 2010 of historical climate simulations) and +4 K (2100 to 2110 of +4 K warmer future simulations) climate scenarios from April to October for the northern hemisphere, and from November to March for the southern hemisphere, sourced from d4PDF, are shown.
Supplemental Figure 13 Spatial distributions of cumulative canopy photosynthetic gains in South America. Estimated gains under the present climate (left panels: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios (central panels: +2 K [2041 to 2050 of +2 K warmer future simulations], 500 µmol/mol; right panels: +4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol), and those for early (top panels), middle (central panels), and late ripening cultivars (bottom panels) are shown. The gains are mapped on symbol colors when the site meets both heat and chilling requirements (circles) or the heat requirement only (triangles); sites without sufficient heat requirement are indicated with crosses.
Supplemental Figure 14 Spatial distributions of cumulative canopy photosynthetic gains in Asia. Estimated gains under the present climate (left panels: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios (central panels: +2 K [2041 to 2050 of +2 K warmer future simulations], 500 µmol/mol; right panels: +4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol), and those for early (top panels), middle (central panels), and late ripening cultivars (bottom panels) are shown. The gains are mapped on symbol colors when the site meets both heat and chilling requirements (circles) or the heat requirement only (triangles); sites without sufficient heat requirement are indicated with crosses.
Supplemental Figure 15 Spatial distributions of cumulative canopy photosynthetic gains in South Africa. Estimated gains under the present climate (left panels: HIST [2001 to 2010 of historical climate simulations], 400 µmol/mol) and two future climate scenarios (central panels: +2 K [2041 to 2050 of +2 K warmer future simulations], 500 µmol/mol; right panels: +4 K [2100 to 2110 of +4 K warmer future simulations], 850 µmol/mol), and those for early (top panels), middle (central panels), and late ripening cultivars (bottom panels) are shown. The gains are mapped on symbol colors when the site meets both heat and chilling requirements (circles) or the heat requirement only (triangles); sites without sufficient heat requirement are indicated with crosses.
Supplemental Figure 16 Probability densities of growth period duration (A1 to A3), cumulative photon absorption (B1 to B3), and cumulative photosynthetic gain (C1 to C3) of sites that meet both heat and chilling requirements. The densities are shown when sufficient data are available (n > 20). HIST, present climate, 2001 to 2010 of historical climate simulations; +2 K, future climate, 2041 to 2050 of +2 K warmer future simulations; +4 K, future climate, 2100 to 2110 of +4 K warmer future simulations.
Supplemental Equation 1
Supplemental Equation 2
Supplemental File 1 Diurnal patterns of light absorption and photosynthetic gain of grapevine canopy at five representative sites. site_name, name of site; site_latitude, latitude of site; site_longitude, longitude of site longitude; scenario, climate scenario used for the simulation; CO2conc, atmospheric CO2 concentration [µmol/mol]; hedge_direction, row orientation of hedgerows; photosynthesis_model, photosynthesis model used for the simulation; photosyn_light_leaf, net photosynthetic rate of leaves in the light part [µmol/m2/s]; temp_light_leaf: temperature of leaves in the light part [°C]; photosyn_shade_leaf, net photosynthetic rate of leaves in the shade part [µmol/m2/s]; temp_shade_leaf, temperature of leaves in the shade part [°C]; PPFD, photosynthetic photon flux density on horizontal surface [µmol/m2/s]; PPFD_light_top, photosynthetic photons absorbed by the light part attributable to the radiation on the canopy top [µmol/m2/s]; PPFD_shade_top, photosynthetic photons absorbed by the shade part attributable to radiation on the canopy top [µmol/m2/s]; PPFD_light_direct, photosynthetic photons absorbed by the light part attributable to direct radiation on the canopy side [µmol/m2/s]; PPFD_light_diffuse, photosynthetic photons absorbed by the light part attributable to diffuse radiation on the canopy side [µmol/m2/s]; PPFD_shade_diffuse: photosynthetic photons absorbed by the shade part attributable to diffuse radiation on the canopy side [µmol/m2/s].
Supplemental File 2 Cumulative photosynthetic gain of grapevine canopy during the growing period at 1266 sites. site_id, site ID provided by the ClimatView (https://www.data.jma.go.jp/cpd/monitor/dailyview/index.php); site_latitude, latitude of site latitude; site_longitude, longitude of site; chill_hour, the number of hours exposed to cold air with temperature of 0 to 7.2°C during winter [hr]; scenario, climate scenario used for the simulation; CO2conc, atmospheric CO2 concentration [µmol/mol]; ripeness, varietal difference in thermal requirement for maturation; days_flower, days to reach flowering [d]; days_mature, days to reach maturation [d]; hedge_direction, row orientation of hedgerows; photosynthesis_model, photosynthesis model used for the simulation; cumulative_photosynthetic_gain, cumulative photosynthetic gain of the canopy [mol/m2], cumulative_respiration, cumulative respiratory loss of the canopy [mol/m2]; cumulative photon absorption, cumulative light absorption of the canopy [mol/m2].
Footnotes
K.M. discloses support for the research of this work from JSPS KAKAENHI (grant number 19H00963). This study was partially supported by Hokkaido Wine Sustainability Promotion Project (K.M. and M.N.) and Cabinet Office grant in aid “Evolution to Society 5.0 Agriculture Driven by IoP (Internet of Plants),” Japan (K.M. and M.N.). This study used d4PDF produced with the Earth Simulator jointly by science programs (SOUSEI, TOUGOU, SI-CAT, DIAS) of the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), Japan. This manuscript has been posted to a preprint server, bioRxiv (https://doi.org/10.21203/rs.3.rs-2622978/v1).
Murakami K and Nemoto M. 2024. Global maps of canopy photosynthesis of grapevines under a changing climate. Am J Enol Vitic 75:0750015. DOI: 10.5344/ajev.2024.23039
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received May 2023.
- Accepted February 2024.
- Published online June 2024
This is an open access article distributed under the CC BY 4.0 license.
References
Vol 75 Issue 1
Jump to section
Related Articles
Cited By...
More from this TOC section
Similar Articles
