Grapevine ( Vitis spp.) leaves have diverse shapes and sizes ( Chitwood et al., 2014; Chitwood and Mullins, 2022), which are influenced by genetics ( Chitwood et al., 2014; Demmings et al., 2019), vine phytosanitary status ( Klein et al., 2017), environment ( Chitwood et al., 2016b; Chitwood et al., 2021), leaf age, node position ( Chitwood et al., 2016a; Chitwood et al., 2016b; Bryson et al., 2020), rootstock ( Migicovsky et al., 2019; Harris et al., 2021), and many other factors.

The primary photosynthetic organ of the plant, the leaf has many characteristics that impact photosynthesis. For example, increasing stomatal density enhances leaf photosynthesis ( Tanaka et al., 2013). Leaf colour may also impact photosynthetic rate, as shown in one study of red and green leaves, in which the green leaves had a 47 % higher maximum CO ₂ assimilation rate compared to red leaves from neighbouring plants ( Menzies et al., 2015). In addition, a larger leaf size will increase the photosynthetic potential of the plant; however, it may also have negative consequences: larger leaves with a thicker boundary layer can slow heat loss, increasing respiration at a rate greater than the increase in photosynthesis ( Givnish, 1987; Westoby et al., 2002). Both the size of an individual leaf and the combined size of all leaves, also known as the total leaf area, may influence plant growth and health. In grapevine, the higher water demand for transpiration associated with a larger total leaf area may increase vine water stress and as a result reduce yield ( Mirás-Avalos et al., 2017).

In addition to the temperature of individual leaves, it is important to consider canopy temperature, which we define as the average temperature taken across multiple leaves across the whole plant. Canopy temperature can be influenced by many factors, both external (environmental), such as net radiation and wind velocity, and internal (physiological), such as stomatal control of transpiration. The architecture of the canopy, which can be altered through management practices, as well as the spacing and orientation of rows, can also interact with the environment to impact the temperature of the canopy ( Boissard et al., 1990). For example, wide row spacing, a high number of shoots per hectare, and high vine vigour all increase within-canopy shading, altering the canopy microclimate and ultimately influencing traits such as yield and fruit composition ( Smart, 1985).

Canopy temperature can be measured using infrared thermometry, such as remote thermal imaging, which is non-invasive and non-destructive ( Leinonen and Jones, 2004; Giménez-Gallego et al., 2021). Temperature can influence many developmental processes in grapevines, with higher temperatures accelerating them (e.g., the timing of budbreak, bloom and onset of fruit ripening), which is of particular concern in the face of climate change ( Keller and Tarara, 2010; Parker et al., 2011). However, the relationship between temperature and development is not linear, and both low and high temperatures can be detrimental, limiting berry size and delaying ripening ( Keller, 2020b). While ambient temperature plays a critical role, canopy temperature is also important. For example, bud temperature, rather than air temperature determines the timing of budbreak ( Keller and Tarara, 2010).

In addition to the important role of temperature in vine development, it has a critical impact on vine transpiration. While transpiration increases with temperature, very high temperatures will lead to stomatal closure ( Keller, 2020a). Stomatal closure due to, for example, water deficit, will also decrease photosynthesis due to decreased CO ₂ availability ( Chaves et al., 2008). Transpiration contributes to the cooling of the leaves, and by extension the canopy as a whole, and can be particularly important in sunlit leaves ( Gates, 1964).

Many grapevines are irrigated and to efficiently manage water stress in both vines and other plants, the grower needs to know when water stress will begin and how much water to apply. Water status is also important in rain-fed vineyards where management practices are applied to minimise water stress. Canopy temperature can be used to assess plant water status using the crop water stress index (CWSI), which is calculated based on the difference between canopy and air temperature ( Cohen et al., 2005). CWSI may be used in grapevines to determine the need for and effect of irrigation ( Ahi et al., 2015). Thus, canopy temperature plays an important role in vine development and is a critical indicator of vine water status. How the shape and size of grapevine leaves interact with the environment to influence canopy temperature is poorly understood. Given that grapevine leaf shape is at least partly controlled by genetics ( Chitwood et al., 2014; Demmings et al., 2019), the potential positive impact of particular leaf shapes or sizes on canopy temperature (such as keeping vine temperature low in hotter climates and reducing the need for irrigation) could be a desirable target for grapevine breeders. Based on this premise, we examined the relationship between grapevine leaf shape or size and canopy temperature.

In this study, we build on the findings of Migicovsky et al. (2022b) to explore the consequences of leaf shape variation on vine canopy temperature across 388 unique accessions resulting from five biparental crosses. Each biparental cross had one parent with compound leaves and a different second parent with non-compound leaves, thus these accessions varied primarily in terms of lobing (Figure 1) and also differed in leaf area.

The first objective of this study was to account for the time of sampling on canopy temperature, as measured using a handheld infrared thermometer. To do this, we calculated a linear model for temperature ~ time of sampling, and determined the value of each vine on a particular date, based on the residuals from that model (Figure 2). The slope of the line differed between dates, indicating that both date and time of sampling influenced canopy temperature.

Each dot represents the temperature of a particular vine at a particular sampling day and time (n = 388). For each date, the measurements taken on that date are black, while the measurements from the remaining three dates are plotted in grey. The linear model for a particular date is shown using a red line.

On a global scale, smaller leaves are generally found at drier sites in warm regions in comparison to large-leaved species which are found in wet and hot environments. In wet and cold environments, species with smaller leaves predominate ( Wright et al., 2017). The difference in leaf sizes based on access to water is due to the thicker boundary layer of large leaves, which makes them more reliant on transpirational water loss for cooling ( Gates, 1968; Wright et al., 2017). In previous work, thermal imagery of 68 Proteaceae species was used to determine the effect of leaf size on leaf temperature, and temperature was found to increase significantly with leaf area ( Leigh et al., 2016). Similarly, in another study examining the difference between canopy temperature and air temperature using trees from 18 different species in an urban environment, trees with smaller leaves had lower canopy-to-air temperature differences, indicating they remained cooler during the day than trees with larger leaves ( Meier and Scherer, 2012). Both these studies provide evidence that smaller leaves are generally cooler than larger ones.

Although the climate in Madera County, California is dry and hot, vines were fully irrigated and thus water was not a limiting factor. The cooler temperature of the canopy for vines with larger leaves in our study may indicate that transpirational cooling was occurring at a higher rate, reducing the risk of high temperatures in comparison to vines with smaller leaves. In our study, the vines were fully irrigated, thus the larger leaves were not limited in terms of water, allowing more transpirational cooling to occur when necessary due to increased leaf area. This relationship seems particularly probable when considered in the context of the ambient temperatures on days of sampling: on the coolest day, 1 August 2019, the temperature was on average over 2 °C cooler, with a maximum temperature of 2.8 to 4.6 °C less than the other dates; this was the same sampling day on which no relationship between leaf size and temperature residuals was found, indicating that this trend is strongest on hotter days when the transpirational cooling benefits resulting from large leaves may be greater; by contrast, this may not occur on cooler days when the vines are under less stress from temperature. However, we did not measure transpiration rate in our study and, while temperature is known to influence the vapour pressure deficit and therefore transpiration, future work should measure transpiration rate directly.

In one study of an Australian heat wave, some vines were covered with a protective layer to reduce heating while others were exposed. In the exposed vines, transpiration increased nearly three-fold while photosynthesis decreased by 35 %, delaying ripening and causing a reduction in berry quality ( Greer and Weedon, 2013). In another study, controlling canopy temperatures resulted in a significant reduction in yield for vines with the warmed temperature treatment ( Greer and Weedon, 2018). These results clearly show the negative impact of excessive increases in canopy temperature. The ability to maintain a cooler canopy temperature on hot days is desirable for grape growers, and our preliminary findings indicate that this may be possible with larger leaves.

While we measured both leaf size and shape in our study, we did not measure canopy architecture, photosynthesis or water use efficiency, as the heterogeneity of individual seedling canopies is very high, which substantially complicates such observations on unreplicated seedlings. While overall

reduction in canopy temperature was found to occur with larger leaves, this may not be due to leaf size, but may instead be due to differences in the openness of the canopy. We also did not measure the distance from the ground that the temperature measurements were taken; trunk height has been shown to have an impact on temperature in previous work ( De Rességuier et al., 2023). Our measurements were taken on the north side of the vines; therefore, while all the vines were measured on the same side using the same method, and the vines were small and unlikely to cause shading, we may have detected stronger temperature differences, if we had measured from the south side due to the direct solar radiation received. Future work should also include measurements of total leaf area to determine if vines with smaller leaves have smaller total leaf area, allowing sunlight to penetrate the canopy in comparison to vines with larger leaves and a greater total leaf area. In addition, leaf area index is a useful metric of canopy density and may be altered through management practices such as shoot thinning and leaf removal, altering light interception ( Wang et al., 2019). As alternatives to making manual measurements of leaf area index, which is time-consuming, mobile digital devices ( Orlando et al., 2016) or remote sensing ( Illniyaz et al., 2022) can be used. Comment by Clémentine Pellissier: This reference is not mentioned in the reference list. Can you please provide the full rererence ? Otherwise, I’ll remove it from the main text.

There are also limitations to the accuracy of handheld infrared thermometers. For example, the distance that the thermometer is held from the target may impact accuracy, and if it is held too far away, the target may include areas not of interest, or external sources of light may impact the results; it should be noted that it was not possible to quantify the magnitude of this potential source of error in our study.

Future work making use of thermal remote sensing imaging ( Still et al., 2021) would be particularly useful in order to estimate the canopy temperature across numerous vines simultaneously, thus reducing the effects of timing and human error on the results. Indeed, thermal imaging paired with the CWSI could ultimately facilitate precision viticulture by assessing water stress and the need for irrigation; work in this area is ongoing ( Tanda and Chiarabini, 2019).

This temperature study built on our previous work, in which we determined that more highly lobed leaves compensated for what would otherwise result in a reduction in leaf area by having longer veins and a higher vein to blade ratio ( Migicovsky et al., 2022b). In this study, we determined that vines with larger leaves had cooler canopies than anticipated. Taken together, these findings indicate that it may be possible to select for large, highly lobed leaves in order to reduce canopy temperature and improve photosynthetic capacity, while still allowing light to permeate the canopy. However, due to the limitations of this work, future studies are still required in order to measure canopy density and transpiration rate in vines of differing leaf sizes and shapes. Ultimately, it may be possible for grape breeders to harness variation in grapevine leaf size for reduction in canopy temperature which would be valuable target for future cultivar improvement.

The research conducted for this study was supported by the National Science Foundation (NSF) Plant Genome Research Program 1546869. JFS was supported by an NSF Graduate Research Fellowship under Grant No. 1758713, and Saint Louis University. This research is also funded through the USDA National Institute of Food and Agriculture and Michigan State University AgBioResearch. This article was written, in part, thanks to funding from the Canada Research Chairs Program to ZM.

We acknowledge all of the individuals involved in maintaining the vineyard evaluated for this study. We acknowledge Leah Brand (Missouri State University), Julie Curless (Missouri State University), Dalton Gilig (University of Missouri), and Ilona Natsch (Saint Louis University) for assistance in sampling and landmarking of the leaves. We would also like to acknowledge Laszlo Kovacs (Missouri State University) for student supervisory support.

PC is employed by E. & J. Gallo Winery. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Any opinion, findings and conclusions or recommendations expressed in this material are those of the authors(s) and do not necessarily reflect the views of the National Science Foundation.

All data and code used in this study can be found on GitHub (https://github.com/zoemigicovsky/grape_leaf_temp). All original scans used in this study are available from Dryad ( Migicovsky et al., 2022a).

PC generated the seedlings and supervised the maintenance of the vineyard. ZM, JFS, PC, and DHC conceived of the initial idea for this study. ZM coordinated the research. ZM, JFS, ZH, LLK, AL, MM, and KW sampled the leaves for this study. AF, MK, AJM, PC, and DHC acquired the funding for this study and provided supervisory support. ZM performed the data analysis with input from DHC. ZM wrote the first draft of the manuscript, which all authors read, commented on, and edited.