Introduction

Avocado is a commercially important evergreen fruit tree. It is grown in a number of countries with tropical and subtropical climates (Alcaraz et al. 2013). The black-skinned Guatemalan–Mexican hybrid avocado cv. Hass dominates global production and market share (Silva and Ledesma 2014). Despite its growing popularity, the geographical distribution and expansion of ‘Hass’ are restricted by its high sensitivity to extreme climate—and especially cold—events (Charrier et al. 2015). The latter are categorized into two groups: (i) extreme cold events, where the temperature falls below the freezing point of water, referred to as frost events (Perry 1998); (ii) milder cold events, where the minimum temperature comes close to, but is still higher than 0 ℃, referred to as chilling events (Jouyban et al. 2013). Frost events can cause plant tissue damage through direct mechanical injury from ice that forms in the intercellular matrix. In addition, water potential decreases in the extracellular space due to the ice formation, which causes water to flow out of the cells, resulting in protoplast shrinkage and dehydration (Jones 2006; Verslues et al. 2006; De Melo-Abreu et al. 2016). Moreover, excessive excitation of the respiratory and photosynthetic electron transport systems may result from freezing temperatures and high morning light intensity, thereby increasing the level of reactive oxygen species in the plant tissue (Janda et al. 2003). In avocado, severe frost may damage foliage, floral buds, flowers and fruit, thus reducing avocado crop yields (Bar-Noy et al. 2019). In addition, under exposure to controlled frost stress, the ratio of variable to maximum fluorescence (Fv/Fm) in ‘Hass’ leaves was found to be significantly reduced, indicating frost-induced photoinhibition (Weil et al. 2019). CO2-assimilation rates and stomatal conductance were also found to be significantly reduced by exposure of potted ‘Hass’ avocado trees to overnight freezing temperatures followed by daytime high-light conditions (Joshi et al. 2019). Studies show that chilling events may also damage avocado trees; for instance, ‘Hass’ trees exposed to chilling temperatures had lower maximum CO2-assimilation rates, light-saturation points, CO2-saturation points and quantum yields than non-stressed trees. Fv/Fm was also ~ 50% lower for leaves of trees exposed to chilling temperatures than non-stressed trees, suggesting that the leaves had experienced wintertime chill-induced photoinhibition (Whiley et al. 1999). Fv/Fm in young avocado ‘Reed’ and ‘Pinkerton’ tress was also reduced following exposure to chilling temperatures (Chernoivanov et al. 2022; Rubinovich et al. 2023). Nevertheless, the adverse effects of chilling temperatures in avocado are much less described than those of freezing temperatures, as plant tissues show almost no apparent damage.

Extreme cold occurrences are becoming more frequent as climate change intensifies, even in regions where they were previously rare (Zhang et al. 2012). Determining ways to reduce cold damage in both mature and young avocado orchards is therefore imperative. Various methods have been evaluated for their effectiveness in preventing cold damage to fruit crops, e.g., the effectiveness of wind machines and over-the-canopy water sprinklers were assessed in various crops, including avocado (Ribeiro et al. 2006; Ghaemi et al. 2009; Bar-Noy et al. 2019). Shading nets are also frequently used to protect agricultural crops from damaging environmental factors, including hail, wind, and excessive sunlight (Shahak et al. 2008; Shahak 2014; Manja and Aoun 2019). Teitel et al. (1996) suggested that high-density shading nets can reduce the risk of frost damage in crops. Indeed, banana plants covered with an Aluminet 50% shading net were considerably less damaged by frost than control plants protected by a Crystal 10% net (Zait et al. 2020). Other studies have suggested the use of shading nets to alleviate chilling stress as well. For instance, mid-winter Fv/Fm rates, chlorophyll content and trunk diameter were significantly higher in young avocado ‘Reed’ trees covered with a high-density Silver 70% net during cold winters without frosts, compared to uncovered trees (Chernoivanov et al. 2022). In a study conducted over three successive winters without frosts, avocado ‘Pinkerton’ trees were covered with Silver 50% or 70% shading nets, with non-covered trees serving as a control (Rubinovich et al. 2023). Fv/Fm was significantly higher in trees covered with the Silver 50% or 70% shading nets compared to the controls. Both net treatments had a favorable impact on fruit yield. However, as air temperature in these two studies did not drop below 0 ℃ and the minimum air temperature under the shading nets was similar to that for the controls, the positive effects of the high-density shading nets were attributed to the mitigation of chilling adverse effects via a reduction in irradiation levels during the winter.

That said, little is known about the effectiveness of shading nets in preventing damage from extreme cold in ‘Hass’ avocado. We hypothesized that high-density shading nets might mitigate cold damage in young ‘Hass’ avocado trees. Therefore, the main objective of this study was to evaluate the effects of covering young ‘Hass’ avocado trees with high-density shading nets during the winter. We decided to cover the trees only during the winter when cold events are likely to occur, as bee activity and pollination services in trees growing under shade netting during spring may be impaired (Mditshwa et al. 2019).

Materials and methods

Experimental site

The experiments were conducted between the end of 2021 and May 2023 in a 0.5-ha experimental ‘Hass’ avocado orchard in the Northern Agriculture R&D research farm in northwest Israel (lat. 33°15’N, long. 35°62’E, 72 m above sea level). The experimental site soil is composed of 54% clay, 28% silt, and 18% sand. All trees were grafted on the West Indian ‘Ashdot 17’ rootstock. The experimental site was chosen in a location with a high risk of extreme cold events. The trees were planted in 2020 with 6 m spacing between rows and 4 m between trees. The rows were oriented north/south. Irrigation rates for each of the experimental months are described in Table 1. Freezing and chilling temperatures occurred in the experimental orchard during the winters, from December to March of each year.

Table 1 Irrigation rates in the experimental orchard during the experimental period
Full size table

Treatments experimental design

From early January through mid-March 2022, and from early December 2022 through the end of March 2023, Silver-colored nets with 60% shading level (Silver 60%, Ginegar Plastic Products Ltd., Ginegar, Israel) were placed over the trees (Silver 60% trees), supported by a metal construction. The height of the nets was 5 m above the tree base. Trees from the control plot were not covered with shading nets (control trees). The experimental plots were completely randomized, with three replications for each treatment (control and Silver 60%, Fig. 1). The area of each replication was ~ 0.1 ha (~ 42 trees), and measurements were taken only from the four middle trees of each replication (experimental trees).

Fig. 1
figure 1

The experimental avocado orchard. Aerial photo taken by a drone in March 2022 of trees covered by shading nets and control, non-covered trees during the winter

Full size image

Air temperature measurements and meteorological conditions

Air temperatures in the field were measured with miniature waterproof, single-channel Hobo temperature data loggers (cat. No. UA-001–64; Onset Corp., Bourne, MA, USA), which were positioned 1.5 m above the ground and shielded from direct sunlight by the canopy. The air temperature was measured continuously at 10-min intervals from December or January until March in each year. Two temperature data loggers were placed for each treatment. Meteorological data including relative humidity, solar radiation, wind speed and precipitation was acquired from a nearby meteorological station in Kibbutz `Kfar Blum` (Israeli Meteorological Services, Fig. 2).

Fig. 2
figure 2

Meteorological data. A Relative humidity. B Wind speed. C Solar radiation. D Precipitation. Data was recorded from December 2021 to June 2023 at `Kfar Blum` station, close to the experimental site

Full size image

Leaf-level light intensity and gas-exchange measurements

Leaf-level photosynthetic photon flux density (PPFD), CO2-assimilation rate and transpiration rate were measured using a LI-6800 portable photosynthesis system (clear-top 9 cm2 chamber, LI-COR, Lincoln, NE, USA). CO2 fed into the leaf chamber was set to 415 ppm, the air flow into the chamber was ~ 700 μmol s−1, and boundary-layer conductance to water vapor was ~ 3 mol m−2 s−1. Chamber climatic conditions (temperature and relative humidity) were set to ambient. Measurements were performed in the orchard at midday on mature attached leaves that were facing the sun, and while the leaves were inside the chamber, care was taken to maintain the same orientation to the sun. Stomatal conductance to water vapor (gs) was calculated by the LI-COR device. Leaf measurements for the control and Silver 60% treatments were taken from 3 leaves per tree, at least 12 trees from each treatment (n = 12 trees). Measured leaves were selected from the southern face of the trees. It is important to note that during the experiment, fruit load on the trees of both treatments (control and Silver 60%) was similar and very low to null (data not shown).

Analysis of chlorophyll a fluorescence

Chlorophyll a fluorescence was measured in complete darkness with a FluorPen FP100 portable fluorometer (Photon Systems Instruments, Drasov, Czech Republic). The Fv/Fm was calculated (Schreiber et al. 1986; Kramer et al. 2004; Baker 2008). Measurements were taken from 3 leaves per tree, at least 12 trees from each treatment (n = 12 trees).

Flowering intensity estimation, tree trunk diameter and chlorophyll measurements

The flowering intensity was assessed as described previously (Ziv et al. 2014; Bar-Noy et al. 2019) during peak bloom in Apr 2022 and 2023; in a blind test, two surveyors assessed each experimental tree independently on a scale of 0 to 5, with 0 denoting no visible flowering and 5 denoting maximum flowering intensity. For tree trunk diameter measurements, the trunks were marked 1–2 cm above the grafting point, and their diameter was measured at the same height with a Vernier caliper once a month. Measurements were taken from at least 12 trees from each treatment (n = 12 trees). Leaf chlorophyll concentration index (CCI) was measured using a chlorophyll meter (Apogee MC-100, Apogee Instruments, Logan, UT, USA). Measurements were taken from 3 leaves per tree, at least 12 trees from each treatment (n = 12 trees).

Observations of foliage and floral bud damage

At the beginning of February 2022, following the significant frost event of January 2022, damage to foliage and floral buds was evaluated in the control and Silver 60% trees. For floral bud damage evaluation, swelled floral buds at stage 511 according to the extended BBCH scale (Alcaraz et al. 2013) from each tree were cut crosswise and visually assessed for browning. A floral bud was considered damaged if frost had browned more than 25% of the cross-sectioned tissue area. For each treatment, at least 20 floral buds were examined on at least 12 trees from each treatment (n = 12 trees). A blind test was used to evaluate the visual effects of foliage frost damage, and two surveyors separately rated each tree on a scale from 0 to 5, with 0 representing no apparent damage, i.e., all leaves green and vital, and 5 representing heavy damage, i.e., leaves turned brown by frost damage (Bar-Noy et al. 2019). Measurements were taken from at least 12 trees from each treatment (n = 12 trees).

Statistical analysis

All results from the same measurement time points were subjected to unpaired t-test using GraphPad Prism version 9.5.1 software (GraphPad Software, LLC).

Results

Effect of shading nets on air temperature

During the first year of the experiment, the minimum air temperature in the orchard during winter fell below 0 ℃ on eight different nights (Fig. 3A). The control plots’ air temperature on the coldest night reached − 2.49 ℃ on January 18, and the maximum air temperature reached 30.36 ℃ on February 23 (Fig. 3C). During the second year of the experiment, the winter was milder and the minimum air temperature in the orchard fell below 0 ℃ on five nights (Fig. 3B), and the control plots’ air temperature on the coldest night reached only − 0.89 ℃ on February 18. The maximum air temperature reached 39.4 ℃ on March 11 (Fig. 3D). In both years, minimum air temperatures under the Silver 60% shading nets were very similar to the control (Fig. 3). However, maximum air temperatures were generally lower under the shading nets than the control.

Fig. 3
figure 3

Minimum and maximum daily air temperatures in the control plots and under the Silver 60% shading nets during the winters of 2021–2022 (A, C) and 2022–2023 (B, D). The mean temperature of each treatment is the average of two data loggers

Full size image

Frost protection of the shading nets

On 17 Jan 2022 and 18 Jan 2022, two frost events occurred in the experimental orchard; the latter was more significant, with the minimum air temperature in the control plot reaching − 2.49 ℃, and staying below 0 ℃ from 2150 h on January 17 until 0710 h on January 18, i.e., approximately 9 h (Fig. 4A). In the Silver 60% plots, the minimum air temperature reached − 2.03 ℃ and the duration of freezing temperatures was slightly shorter, from 2250 h on January 17 till 0650 h on January 18. Following these events, obvious and significant tissue damage to young shoots, flowering buds and flowers was observed in the trees from the control plot, but not in trees covered with the Silver 60% nets (Fig. 4B). In the control trees, 93% of the examined floral buds were severely damaged (Fig. 4C). In contrast, in the trees covered with the Silver 60% nets, only 4% of the examined floral buds were damaged, significantly (p < 0.0001) lower than in the control trees. Similar findings were observed regarding the damage to young vegetative shoots following the frost, which was assessed at 4.35 out of 5 in the control trees, significantly greater (p < 0.0001) than the damage assessed in trees covered with the Silver 60% nets (0.5 out of 5, Fig. 4D).

Fig. 4
figure 4

Effect of shading nets on air temperature and frost damage in avocado trees. A Daily air temperatures in the control plots and under the Silver 60% shading nets on 17 and 18 Jan 2022. The mean temperature of each treatment is the average of two data loggers. B Pictures were taken from the control and Silver 60% trees at the beginning of February 2022, following the significant frost event in January 2022. C Damage to floral buds and D young vegetative shoots in the control and Silver 60% trees at the beginning of February 2022. Values are means ± SE of 12 trees from each treatment (n = 12). Columns marked with an asterisk differ significantly (unpaired t test, ****p < 0.0001)

Full size image

Effect of shading nets on leaf-level light intensity and gas-exchange parameters

In measurements where the shading nets were deployed above the trees (Fig. 5, grey background), PPFD was significantly (p < 0.0001) lower (by ~ 60%) than the control (Fig. 5A). For all of the other measurements (when the shading nets were not deployed above the trees), PPFD was similar between the two treatments.

Fig. 5
figure 5

Effects of shading nets on leaf-level light intensity, CO2 assimilation and stomatal conductance measured and calculated by the LI-COR system. A PPFD—photosynthetic photon flux density; B A—CO2 assimilation; and (C) gs—stomatal conductance to water vapor. Parameters were measured in the orchard from December 2021 to May 2023 monthly at mid-day when Silver 60% nets were deployed (grey background) and not deployed (no background). Measurements were taken from trees not covered (control) or covered with the Silver 60% shading net (Silver 60%). Values are means ± SE of 12 trees (n = 12, at least 3 different leaves were measured from each tree). Treatments marked with an asterisk differ significantly (unpaired t test, *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; N.A. no data available)

Full size image

At the beginning of the measurements, before the deployment of the shading nets (December 2021), CO2 assimilation was similar between the trees from the two treatments (p > 0.05, Fig. 5B). In most other measurements, the CO2-assimilation rate in Silver 60% tree’s leaves was higher than, or similar to the control. In January 2022, CO2-assimilation rate was relatively low in both plots but significantly higher (p < 0.001) in the Silver 60% trees (by 134%) compared to the control. A similar trend was found in February 2022, where the CO2-assimilation rate was 56% higher in the Silver 60% trees compared to the control, but the difference was insignificant (p > 0.05). In March, May and June 2022, the CO2-assimilation rate was similar between the trees of the two treatments. Interestingly, In July and August 2022, the CO2-assimilation rate was significantly higher (p < 0.05) in the Silver 60% trees, by 28% and 18%, compared to the control, respectively. A similar trend was found in September, October and November 2022, where the CO2-assimilation rate was higher in the Silver 60% trees by 16%, 15.9% and 7% compared to the control, respectively. An opposite trend was found in December 2022, where the CO2-assimilation rate was significantly higher (p < 0.05) in the control trees (by 26%) compared to the Silver 60% trees. Notably, this was the only time when the shading nets were deployed (grey background) and the CO2-assimilation rate was lower in the net-covered trees compared to the control. In January and February 2023, the CO2-assimilation rate was higher in the Silver 60% trees by 20% (p < 0.05) and 30% (p < 0.01) compared to the control, respectively. An opposite trend was found in March 2023, when the CO2-assimilation rate was 12% higher in the control trees compared to the Silver 60% trees, but the difference was insignificant (p > 0.05). In April 2023, the CO2-assimilation rate was 15% higher in the Silver 60% trees compared to the control trees, but this difference was insignificant (p > 0.05). In May 2023, the CO2-assimilation rate was significantly higher (p < 0.05) in the Silver 60% trees, by 17%, compared to the control trees.

In December 2021, gs was similar between trees from the two treatments (p > 0.05, Fig. 5C). In most other measurements, gs in the Silver 60% trees’ leaves was higher than, or similar to the control. During January 2022, February 2022, January 2023 and February 2023, gs was significantly higher (p < 0.05, p < 0.01, p < 0.0001, p < 0.01, respectively) in the Silver 60% trees, by 63%, 96%, 65% and 44% compared to the control, respectively. In July 2022, gs was also significantly higher (p < 0.01) in the Silver 60% trees, by 30%, compared to the control. In March, May, June, August, September, October and November 2022, gs was insignificantly higher (p > 0.05) in the Silver 60% trees, by 13%, 6%, 29%, 24%, 11%, 34% and 8% compared to the control, respectively. In December 2022, March 2023 and April 2023, gs was insignificantly higher (p > 0.05) in the control trees by only 5%, 9% and 2% compared to the Silver 60% trees, respectively. In May 2023, gs was insignificantly higher (p > 0.05) in the Silver 60% trees, by 28%, compared to the control.

Effect of shading nets on maximum quantum yield (Fv/Fm), chlorophyll content and trunk diameter

Fv/Fm rates were similar between the trees from both treatments (p > 0.05, Fig. 6A) in December 2021. In most other measurements, Fv/Fm rates in the Silver 60% trees’ leaves were higher than, or similar to the control. For most measurements taken when the shading nets were deployed (gray background), i.e., in February 2022 and from December 2022 to February 2023, Fv/Fm was significantly higher (p < 0.05, p < 0.05, p < 0.05, p < 0.001, respectively) in the Silver 60% trees compared to the control. In January 2022, Fv/Fm was insignificantly higher (p > 0.05) in the Silver 60% trees compared to the control. Fv/Fm was also significantly higher in the months immediately following the removal of the shading nets, i.e., March 2022 (p < 0.01), March 2023 (p < 0.01) and April 2023 (p < 0.05). There were no significant differences in any of the other measurements (p > 0.05) between Fv/Fm rates of trees from both treatments.

Fig. 6
figure 6

Effects of shading nets on maximum quantum yield, leaf chlorophyll content and trunk diameter. A Maximum quantum yield (Fv/Fm). B Leaf chlorophyll content (CCI). C Trunk diameter. Parameters were measured monthly from December 2021 to May 2023 in the orchard when Silver 60% nets were deployed (gray background) and not deployed (no background). Measurements were taken from trees not covered (control) or covered with the Silver 60% shading net (Silver 60%). Values are means ± SE of 12 trees (n = 12, at least 3 different leaves were measured from each tree). Treatments marked with an asterisk differ significantly (unpaired t test, *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001)

Full size image

CCI was similar between the trees from both treatments (p > 0.05, Fig. 6B) in December 2021. For most other measurements, CCI in the Silver 60% trees’ leaves was higher than, or similar to the control. For most measurements when the shading nets were deployed, i.e., February 2022, January 2023 and February 2023, CCI was significantly higher (p < 0.001, p < 0.0001, p < 0.0001, respectively) in the leaves of Silver 60% trees compared to the control, respectively. These differences were noticeable, especially in January and February 2023, when CCI was higher by 43% and 30% in the leaves of Silver 60% trees compared to the control, respectively (Fig. 6B, Online Resource 1). CCI was also significantly higher in the months following the removal of the shading nets in 2023, i.e., March, April and May 2023 (p < 0.05, p < 0.05, p < 0.01, respectively). There were no significant differences for any of the other measurements (p > 0.05) between CCI rates of trees from both treatments.

The trunk diameter of trees from both treatments was similar in December 2021 (Fig. 6C). From February 2022 through all other measured time points, the trunk diameter of the Silver 60% trees was larger than that of the control trees. In May 2023, which was the final measured time point, the trunk diameter of the Silver 60% trees was 6% larger than that of the control trees. However, this difference was insignificant (p > 0.05).

Effect of shading nets on flowering intensity

In April 2022 and 2023, flowering intensity was high in both treatments (Fig. 7). Nevertheless, in April 2022, flowering intensity in the Silver 60% trees was rated 4.5 out of 5, significantly greater (p < 0.0001) than in the control which was rated 4 out of 5 (Fig. 7A). In the following year, the differences in flowering intensity of the trees from both treatments were insignificant (p > 0.05) and were rated 4 and 3.7 out of 5 in the control and Silver 60% trees, respectively (Fig. 7B). Fruit yield was not examined, as the trees were young (before full bearing) and did not bear fruit during this study.

Fig. 7
figure 7

Flowering intensity in the control and Silver 60% trees was assessed and scored in April 2022 (A) and April 2023 (B) on a scale of 0–5, with 0 representing no apparent flowering and 5, maximum bloom. Values are means ± SE of 12 trees from each treatment (n = 12). Columns marked with an asterisk differ significantly (unpaired t-test, ****p < 0.0001; ns not significant)

Full size image

Discussion

We examined the effects of covering young ‘Hass’ avocado trees during the winter with Silver 60% high-density shading nets on frost mitigation and tree performance. The experimental plot was chosen because of its high risk of extreme cold. Indeed, during the 2 years of the experiment, 13 different frost nights resulted in mild or severe cold damage to the trees, depending on the severity of the events (Fig. 2). In the first year of the experiment, the significant frost event that occurred in January 2022 caused visible external damage to the control trees (Fig. 4). There was obvious tissue damage to young vegetative shoots, and the floral buds, which are considered more frost-sensitive (Bar-Noy et al. 2019), were severely damaged. At the same time, the net-covered trees almost did not show any obvious external damage. Similar results have also been observed in banana, where a high-density 50% Aluminet net reduced frost damage compared to the control 10% Cristal Leno shading net (Zait et al. 2020). Under the former net, both daytime light intensity and maximum temperature were lower, whereas the minimum air temperature was higher (by more than 2 ℃) than under the latter net during frost events (Zait et al. 2020). However, in the present study, while the Silver 60% shading nets also significantly reduced daytime light intensity and maximum air temperatures, they only slightly elevated minimum night temperatures (Figs. 3, 5). Similar results were shown in two different studies conducted on ‘Reed’ and ‘Pinkerton’ avocado trees, where Silver 50% and 70% shading nets reduced daytime light intensity, but had no effect on minimum night temperatures (Chernoivanov et al. 2022; Rubinovich et al. 2023). Thus, it seems that net composition, rather than its shading percentage, affects its ability to increase nighttime temperatures. Interestingly, in the present study, although the shading nets did not elevate air minimum temperatures, they still had positive effects on cold-damage reduction and tree performance (Figs. 3, 4, 5, 6, 7). As high light intensity has a significant influence on overall stress during extreme cold events (Wise 1995), it is possible that adverse effects of the cold were reduced in the net-covered trees due to the reduction of daytime light intensity, rather than an increase in nighttime temperature. Also, as the shading nets reduced day-time maximum temperatures (Fig. 3) it is possible that slowing the warming process following a night with freezing temperatures reduced frost damage by reducing the thawing rate (Yoshida and Sakai 1968; Gusta and Fowler 1976; Snyder and Melo-Abreu 2005).

There may be other explanations for the reduction of frost damage under the shading nets. For instance, plant and soil water status, which are known to be affected by shading nets and other over-the-canopy covers, may affect overall plant physiological conditions and response to environmental stress (Snyder and Melo-Abreu 2005; Brito et al. 2021; Boini et al. 2022, 2023). Higher leaf water content in trees under the shading nets may alleviate cell dehydration during frost events and provide freeze protection (Pearce 2001; Snyder and Melo-Abreu 2005). Still, it is less likely that in rainy winters, where soil water content is relatively high, the shading nets would have a significant effect on leaf water content. In addition, shading nets may increase leaf and bud temperature by reducing the loss of infrared radiation emitted by plants and the ground to the sky during radiative frosts, thus reducing tissue damage (Zait et al. 2020).It is important to note that the effect of the shading nets on the above-mentioned parameters and their relation to frost damage were not examined in this study and therefore should be further investigated in future studies.

Following the significant frost event which occurred in the first year of the experiment, the visible damage in the control trees was restricted to flowering buds and young vegetative shoots, whereas no external damage was evident in mature tissues, such as the mature leaves and stems (Fig. 4). This may be because, as already noted, avocado flowers and flowering buds are considered more frost-sensitive than mature tissues (Bar-Noy et al. 2019). In the following spring of that year, flowering intensity was significantly higher in the net-covered trees than in the controls, possibly because more floral buds managed to develop properly in the former (Fig. 7). However, in the second year of the experiment, with no significant frost events, there were no significant differences in flowering intensity between the two treatments. Similar findings have also been shown in previous studies, where ‘Reed’ flowering intensity during the spring in young trees which had been covered with Silver 50% shading nets during cold winters without frost events was insignificantly lower compared to that of control, uncovered trees (Chernoivanov et al. 2022).

During all measurements when the shading nets were deployed, chlorophyll content was higher in the net-covered trees compared to the control (Fig. 6, Online Resource 1). It is likely that the control trees’ decreased chlorophyll content was brought on by cold stress, as decreased chlorophyll content in the leaf can reduce its light-harvesting capacity and photoinhibition occurrence (Kyparissis et al. 1995). For instance, as an adaptive reaction to excessive light capture during cold events, the leaf chlorophyll content of the evergreen tree Quercus ilex was reduced by about 30% after exposure to cold stress (Oliveira and Peñuelas 2000). It is also plausible that in the net-covered trees, shading caused an increase in leaf chlorophyll content which would have increased photosynthetic efficiency under low-light conditions (Figs. 5, 6). For example, a field experiment carried out in Florida revealed that shaded avocado leaves contained more chlorophyll than leaves that were exposed to the sun (Reed et al. 2012). Another study revealed that apple trees grown in the shade had higher chlorophyll levels than those grown in full sunlight (Liu et al. 2019). As leaf chlorophyll content indicates the physiological status of the plant and its photosynthetic capacity and dry matter production (Croft et al. 2020), it is possible that in the present study, the higher chlorophyll content in the net-covered trees had a positive effect on vegetative growth, e.g., tree trunk diameter (Fig. 6). However, this assumption should be further investigated, as the reduction in chlorophyll content following cold stress in avocado is not well-described.

In the 2 years of the experiment, when the shading nets were deployed over the tree canopy during the winter, gas-exchange parameters and Fv/Fm values were generally higher in the leaves of the net-covered trees compared to the controls (Figs. 5,6). Previous studies have shown that cold stress significantly reduces Fv/Fm, CO2-assimilation rate and stomatal conductance in potted avocado plants exposed to freezing temperatures (Joshi et al. 2019; Chung et al. 2022). Similar adverse effects were also observed in a field trial conducted in a banana plantation (Zait et al. 2020). Thus, our results suggest that covering the trees with the Silver 60% nets improves tree photosynthetic performance following extreme-cold events. Interestingly, some of the effects of the shading nets, which were deployed over the trees only during the winter, were maintained in the following months. In particular, CO2-assimilation rate, stomatal conductance and chlorophyll content generally remained higher in the net-covered trees compared to the controls from March to December of the two experimental years (Figs. 5,6). Thus, it is possible that the reduction in the adverse effects of cold events during the winter has a long-term beneficial effect on tree performance. This may lead to enhanced tree vegetative growth as seen in the present study, as well as in the previous study conducted with net-covered ‘Reed’ trees (Chernoivanov et al. 2022). This enhanced tree performance may also eventually contribute to yield parameters in mature avocado trees. Additional studies should be carried out to further support this assumption.

In addition to the occurrence of extreme cold events, climate change may result in extreme heat events. Data from another study conducted in Israel showed an increased frequency of extreme heat waves in the last 16 years (Alon et al. 2022). Interestingly, that study showed that covering the trees with high-density shading nets during extreme heat events improves photosynthetic performance in mature ‘Pinkerton’ avocado trees. Positive effects of shading nets on tree performance during heat stress have also been shown in other species. For example, covering potted `Honeycrisp’ apple trees with a blue 22% shading net during heat stress and high-light conditions decreased incoming light intensity, increased leaf-level photosynthetic light-use efficiency, and reduced symptoms of photoinhibition (Mupambi et al. 2018). A white 25% shading net reduced the effects of heat stress in mature ‘Washington’ navel orange plants grown in Egypt (El-Naby et al. 2020). Thus, using shading nets in ‘Hass’ avocado orchards should be considered a potential measure to reduce not only extreme cold damage but also the adverse effects of heat damage. Moreover, as the use of shading nets can reduce evapotranspiration and improve agricultural crops' water-use efficiency (Mditshwa et al. 2019; Mira-García et al. 2020; Boini et al. 2023), this factor should be investigated in future research, especially in areas with restricted water resources. However, on the other hand, excess shading may have negative effects on the trees. For instance, on days with typical fall temperatures, photosynthetic performance may be reduced under high-density shading nets (Alon et al. 2022). In addition, bee activity and pollination, which are imperative for maintaining high yields in avocado trees, may be reduced beneath shading (Mditshwa et al. 2019; Stern et al. 2021). Therefore, long-term follow-up studies with various avocado cultivars should be carried out to create optimal shading-management protocols that make use of shading nets and other shading methods, such as over-canopy solar panels (Trommsdorff et al. 2022). As light intensity during winter may differ between different regions across the globe, experiments should be carried out also in countries with different light regimes.

Conclusion

The results of this study indicate that covering young ‘Hass’ trees with Silver 60% shading nets during the cold winter mitigates frost damage and improves tree performance (Figs. 4, 5, 6, 7). The evaluation from a recent crop modeling study is that in most major avocado-producing countries, areas of highest suitability may decrease under the impact of future climate change (Grüter et al. 2022). That study also revealed that adaptation to climate change would be imperative for most major avocado-producing regions. Thus, using high-density shading nets during the winter may protect avocado trees from current and future adverse effects of climate change and may also enable the expansion of the geographical distribution of avocado production. Further studies on young and mature avocado trees in different countries should be carried out to validate this assumption.

Author contribution statement

Conceptualization—LR; Investigation—ML, EA and AC; Data analysis—LR; Writing—original draft preparation, LR; Writing review and editing—ML and AC; Funding acquisition, LR. All authors have read and agreed to the published version of the manuscript.