Using in-canopy misters to mitigate the negative effects of heatwaves on grapevines

Abstract

Heatwaves in South Australian viticultural regions are becoming more frequent and lasting longer. To address this environmental challenge, a study was undertaken in the Riverland and Adelaide regions on Sauvignon Blanc and Cabernet Sauvignon grapevines, respectively, during the 2016-17 and 2017-18 growing seasons. Grapevine canopies were cooled using water-based evaporative cooling either in the canopy with fine mist applications (ICM) when temperatures exceeded 35 °C, or with under canopy sprinklers (UCS) or supplemental drip irrigation (SI) during warm nights preceding heatwaves. Canopy temperatures were 5 °C cooler with lower vapour pressure deficits (VPD) than ambient in all cooling treatments. Cooling resulted in improved vine water status, physiological performance (gas exchange), and yield in many of these treatments, but especially in the UCS treatment at the Riverland site for Sauvignon Blanc. During heatwaves, the SI treatment with additional irrigation maintained respective under vine and midrow temperatures approx. 2 °C and 7 °C lower than controls. SI and UCS vines consistently had superior water status and gas exchange than controls and maintained these levels irrespective of prevailing environmental conditions. SI vines also had the highest yields and crop loads in both seasons. Differences between treatments were not as evident at the Adelaide site with Cabernet Sauvignon. The Riverland fruit under misting had greater levels of volatile compounds but had lower wine polyphenols compared to controls. UCS and SI treatments had higher juice and wine acidity. A simulation model of canopy cooling based on misting accurately predicted canopy, leaf and bunch temperatures resulting from mist applications under various environmental conditions and can be incorporated in the future into automated mist controllers in the vineyard. It was found that the method of water application was critical for effective cooling and improved vine performance.  

Summary

Climate change in South Australian viticultural regions has prompted much research in the areas of drought stress mitigation, and, more recently, heat stress mitigation in vineyards. With an increasing incidence, severity and duration of heatwaves over the years, practical strategies are being sought by the industry to manage grapevines under these adverse conditions and to improve their resilience to future similar events. Very few studies have investigated the efficacy of water-based evaporative cooling in vineyards and the effects of cooler/warmer grapevines on wine quality. Strategies such as overhead sprinklers are currently being utilised in vineyards in Australian inland regions, and although effective, they are large consumers of water and much of the water is lost through evaporation. We aimed to evaluate similar but more efficient approaches of water-based cooling in vineyards based on in-canopy misting and undervine sprinklers, and to compare these approaches to additional water applications prior to and during heatwaves.

Heatwaves are defined by SA-BOM as being five or more days above 35 °C or three or more days above 40 °C, but this definition was not specific to winegrapes or agriculture. Key physiological processes such as photosynthesis become compromised beyond 37 °C. This threshold led to us re-defining the temperature threshold for grapevines for the purposes of this study to 35 °C, and heatwaves being defined as three or more consecutive days at or above this temperature. Prolonged periods of even moderate temperatures (30-32 °C or higher) can result in impaired performance as heat stress, often in conjunction with water stress, can impair vine performance. With this new threshold defined, our objectives in the present study were to assess the efficacy of in-canopy misting (ICM), undercanopy sprinklers (UCS), and supplemental irrigation with drippers (SI) on: (i) canopy microclimate, (ii) vine physiology, (iii) fruit development, (iv) yield and vine vigour, and (v) juice and wine composition. The study used two sites: (i) Yalumba Oxford Landing Estates in Qualco, SA (Riverland) planted with mature Sauvignon Blanc vines grafted on Ramsey rootstock, and (ii) Alverstoke vineyard, Waite Campus, Adelaide, SA planted with own-rooted Cabernet Sauvignon vines. Canopy microclimate, vine performance and fruit composition was monitored over two growing seasons, 2016-17 and 2017-18. Wines made at the end of each season underwent compositional analysis including flavour and aroma compounds.

In both seasons, ICM had a dramatic effect on canopy and leaf temperatures, decreasing temperatures by between 5-12 °C compared to control (uncooled) vines. Bunch temperatures were only about 3 °C cooler due to their high thermal mass. Vine water status and physiology were not altered as a result of mist applications. With misting, yields and pruning weights were slightly higher in the Riverland site in the first season but not in the second season. No changes in fruit and wine composition, or cell vitality were observed due to the misting with the exception of increases in certain volatile compounds such as linalool, isoamyl acetate and β-damascenone, which contribute fruity aromas to the wine. Fruit from the misted vines also had lower polyphenols compared to controls. In terms of water application rates, misting accounted for approx. 3-4% of total water used for irrigation in the Riverland block and 5-17% in the Waite block.

Undercanopy water applications were conducted at night for two hours preceding a heatwave. Each treatment – sprinkler and supplemental irrigation – accounted for 7-11% of the total season irrigation, or between 0.5-1.1 ML/ha. Both treatments were effective in reducing canopy temperatures and VPDs making conditions more favourable for vine physiology during heatwaves. While soil moisture levels were unaltered with the undercanopy water applications, vine water status improved compared to uncooled treatments. This resulted in higher net photosynthesis and stomatal conductance values even after the heatwave indicating better vine recovery with these treatments. In many cases, no decreases in gas exchange were observed even on the warmest days (T > 40 °C) calling into question the temperature thresholds for grapevine that were previously proposed. It therefore appears that a favourable canopy microclimate with high humidity and even perhaps cooler soils and roots may allow vines to continue performing at their optimum even on the warmest days. Undercanopy cooling with the SI treatment resulted in slightly higher yields in both seasons, but no changes were observed in juice and wine composition except for higher TA and lower polyphenols. In Cabernet Sauvignon (Alverstoke), the undervine cooled vines were more balanced with Ravaz Index values between 4-5 compared to the control vines that were undercropped (Ravaz Index ~ 1.2).
In comparing the Riverland and Waite vineyards, the Riverland site tended to be more responsive to the cooling treatments compared to the Waite location, with a number of reasons possibly explaining this. First, the Riverland experiences more heatwaves that tend to be warmer and longer in duration as warm air moves south from the centre of Australia. Second, the Waite vineyard is organically managed with a permanent fescue on the floor that likely moderates temperatures and retains soil moisture allowing roots to stay cooler than the Riverland site, which has a bare floor. Finally, vine size and crop load differences between the two vineyards could have contributed. The Waite Cabernet Sauvignon vines were undercropped (Ravaz Index ≤ 1.5) while the Riverland Sauvignon Blanc vines were overcropped (Ravaz Index > 16).

One of our goals was to develop an automated system to trigger and control the mister-based cooling system that would regulate the canopy temperature to 35 °C or below. In order to achieve this, a simulation model was developed using the Scilab programming language. The model was successfully developed and validated versus real vineyard data of canopy, leaf and bunch surface temperatures. Based on the mister settings used in the Riverland location, the model accurately predicted a temperature drop of the canopy (air) by 10-15 °C, leaf temperature drops of 20 °C and bunch temperature drops by 5 °C. A scenario analysis of various mister settings and environmental parameters determined that the efficacy of cooling is greater when (i) the frequency of misting is increased, and (ii) misting is applied until the leaves and bunches are wet thereby allowing greater evaporative cooling from those surfaces to draw out sensible heat. The model is now fully deployable on a portable Raspberry Pi microcontroller with capability to communicate with future control systems for automated operation of misting solenoids.