Journal of Plant Hydraulics To -8 MPa and back: in-situ annual water relation in non-irrigated olive trees 1 Volcani Center, Institute of Soil, Water and Environmental Sciences, ARO, Rishon Lezion, 7505101 Israel; 2 R. H. Smith Institute of Plant Science and Genetics in Agriculture, Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, Rehovot 76100, Israel; 3Department of Plant and Environmental Science, Weizmann Institute of Science, Rehovot, Israel; 4Gilat Research Center, Institute of Plant Sciences, ARO, Gilat, 85280 Israel Barzilai Oded 1 2 Avraham May 1 Wagner Yael 3 Dag Arnon 4 Hochberg Uri 1 * Date of publication: 30.10.2022 Abstract

Olive trees are known for their capability to endure and recover from low stem water potentials (Ψstem < -6 MPa) and even maintain substantial gas exchange under such conditions. In the current experiment, we tested the physiological responses of mature olive trees in-situ while enduring such Ψstem under field conditions for seven months of rainless Israeli summer. To characterize olive trees' seasonal response to drought, we followed the annual patterns (a measurement day once a month) of Ψstem, gas exchange, osmotic adjustment, and hydraulic conductance of mature bearing olive trees that were disconnected from irrigation. We found that through the dry summer, the trees gradually dehydrated to a minimum of -8.1 MPa in October. An increase of 1320 mmol/kg in the leaf osmotic concentration through the summer was part of the tree's acclimation to the low Ψstem. Surprisingly, despite the low Ψstem that have probably led to substantial xylem cavitation, the trees maintained stomatal conductance of at least 0.023 mol m-2 s-1 and positive assimilation throughout the year. Following the winter rains, the plants recovered their Ψstem, gas exchange, and hydraulic conductance to maximal values. The results demonstrate olives' remarkable ability to tolerate Ψstem while maintaining above minimal gas exchange and producing fruits.

Introduction

Olives (Olea europaea), perhaps the most emblematic crop of the Mediterranean (Sofo et al., 2008), is one of the most important tree crops in the world. Covering 10.5 million hectares and producing 20.5 million tons, olives are a market of 20 billion dollars per year (FAO stat 2018). Most of the production areas are located around the Mediterranean basin, with Spain, Italy, Greece, Tunisia, and Turkey being the top producers (IOC2020). Despite the typical rainless summer in these production areas, the large majority of olive orchards are grown without irrigation (77% based on the IOC survey from 2016-2017). The inability to irrigate, combined with the climate change processes that are threatening most of the Mediterranean (Nasrallah and Balling, 1995; Lionello et al., 2014), emphasizes that understanding olives responses to drought is of critical importance.

Olives are considered a drought-tolerant crop that is capable of sustaining low water potential (Ψstem) while maintaining positive assimilation (Sofo et al., 2008; Fernández, 2014). Under typical Mediterranean environmental conditions, when mature olive trees are grown in non irrigated orchards, they dehydrate to stem water potential of -3 to -3.5 MPa. These Ψstem values allow them to maintain substantial gas exchange, which is critical for the oil accumulation in the fruit toward the end of the dry season (Giorio et al., 1999; Tognetti et al., 2007; Kitsaki and Drossopoulos, 2005; Diaz‐Espejo et al., 2007). When orchards were dried to lower Ψstem (-5 < Ψstem < -4 MPa), stomatal conductance was still higher than 0.05 mol m-2 s-1 (Naor et al., 2013; Torres-Ruiz et al., 2015). Even when olive trees were dried to Ψstem < -6MPa, they were able to retain pre-stress functionality upon irrigation due to the high resistance of their hydraulic system and photosynthetic apparatus (Angelopoulos et al., 1996; Moriana et al., 2003; Ennajeh et al., 2008; Perez-Martin et al., 2014). Furthermore, olive trees showed a noteworthy ability to accumulate osmolites, shifting their turgor loss point ΨTLP by more than 2 MPa through the dehydration, which allows them to maintain turgor even when dried to Ψstem < -6 MPa (Ennajeh et al., 2008; Sofo et al., 2008; Dichio et al., 2006a; Dichio et al., 1996). In some experiments, olives were even capable of maintaining substantial gas exchange (gs > 0.05 mol m-2 s-1) under Ψstem lower than -6 MPa (Ennajeh et al., 2008; Tugendhaft et al., 2016), but not in others (Angelopoulos et al., 1996; Sofo et al., 2004; Perez-Martin et al., 2014). Based on such findings, it appears that even if olives will dry to significantly lower Ψstem than those that they typically experience under non-irrigated conditions, they will not suffer irreversible damage and possibly might even be able to maintain assimilation. Having said that, nearly all of the experiments in which olives were dried to Ψstem < -6 MPa were conducted on potted trees (except for Moriana et al., 2003) and the seasonal dynamic of gas exchange and osmotic adjustment of mature bearing trees under such stress is still vague.

There are major differences between short dehydration of trees in pots versus field-grown trees under a natural drought that might result in different outcomes for the trees. On the one hand, the length of a whole summer drought accompanied with high ET in that period (more than six months in a typical Israeli summer) might present a bigger challenge than a short exposure to low Ψstem. For example, sustaining low assimilation through an entire summer might deplete the tree's carbon storage that is required for osmotic adjustment and respiration (Hasibeder et al., 2015). On the other hand, field-grown plants normally have substantial root volume with a larger water holding capacity, which results in slower dehydration. Accordingly, field-grown plants have much more time to acclimate to the gradually intensifying drought conditions.

In the current study, we followed the water relations of mature olive trees in-situ. The trees were grown under rainfed conditions in Israel over the course of a year. We wanted to find how mature olives sustain the long summer drought of the region where they were first domesticated (Langgut et al., 2019). The goal was to evaluate their acclimation to drought and see if the drought tolerance that olives exhibited in pots under short-term dehydration will hold under a whole summer drought. Specifically, we were curious to see which drought severity eliminates assimilation.

Materials and Methods

Experimental setup

The experiment took place between April 2019 and March 2020 and consisted of four fruit-bearing olive trees (Olea europaea cv. Arbequina), 20 years old, growing in the experimental farm of Volcani center (31.986, 34.823). The trees were growing in the middle of a row of 28 trees, which is the remainder of an orchard that was cut down two years prior to the experiment. Only three rows of the orchard were left, and we monitored trees from the central one. The spacing between trees in that row was two meters. The soil texture (at 30 cm depth) is sandy loam and composed of 62% sand, 25% silt and 13% clay. Soil porosity is 45%. From planting and until one year prior to our study, the orchards received irrigation of around 400 mm/year during the dry months. Irrigation had stopped a year and a half before we started our measurements, meaning that we measured the second summer with no irrigation.

Weather data was taken from the Beit Dagan meteorological weather station (https://ims.gov.il/) located 2.26 km from the experimental site. The winter before the measurements had started (winter of 2018-2019) had 703 mm, which is 38.5% higher than the 2008-2018 average for the area (507.9 mm). In the year in which the experiment took place, precipitation was 684 mm, as described in Figure S1. The annual potential evapotranspiration (ET0) between March 2019 to March 2020 was 214% (1466.3 mm) higher than precipitation at the same time. Maximal and Minimal temperatures for that period (42.3 and 3.8 OC, respectively), as well as ET0, are presented in Figures S1 and S2 and were well in the range of the last ten years (0.3 - 43.5 OC).

Field measurements

Stem water potential and gas exchange were measured once a month throughout the year (April 2019 – March 2020), at three hours intervals, five times a day, from predawn till dusk. On the same days, the leaves' osmotic content was evaluated at predawn and 16:00. The specific dates which were measured are: 3/4/2019, 5/5/2019, 3/6/2019, 2/7/2019, 4/8/2019, 29/8/2019, 3/10/2019, 4/11/2019, 2/12/2019, 6/1/2020, 5/2/2020, 30/3/2020.

Measurements were conducted only on clear days. All measurements were made on leaves that were formed in February and March 2019, meaning that they were 1-2months old when we started the experiment and 13-14 month-old when we terminated it. To understand the effect of leaf maturation on the measured parameters, in the last time point (30th of March 2020), we measured both 1-2 and 13-14 month-old leaves.

The stem water potential (Ψstem; MPa) was measured with a pressure chamber (1505D, PMS, Albany, OR, USA) on two twigs per tree (east and west facing) at five time points a day. Twigs with 4-8 leaves were bagged with aluminum foil bags for at least 30 minutes before excision and transported to the lab for measurement. The time span between excision and measurement was up to 30 minutes, during which the leaves were double-bagged to prevent dehydration. In a pre-experiment trial, we found that maintaining double-bagged leaves for 30 minutes in room conditions resulted in less than 0.05 MPa drop.

Stomatal conductance (gs; mmol m-2 s-1) and net assimilation (AN; µmol m-2 s-1) were measured using a commercial gas exchange system (Licor6400, Lincoln, NE) on one leaf per tree at the same time as Ψstem sampling. We only measured wide sun-exposed leaves that were able to cover the 2cm2 cuvette. The chamber parameters were set to CO2 concentration of 400 µmol m-2 s-1 with ambient PAR, humidity, and temperature. The ambient conditions in the measurement days appear in Figs S1 and S2.

To evaluate the leaf osmotic concentration (π; mmol/kg), leaves were collected at predawn and the afternoon (15:30-17:00, depending on day length) into sealed bags and frozen within 10 minutes. At both predawn and 16:00, we averaged two samples for each tree, one containing west-facing leaves and the other containing east-facing leaves. Osmolite could accumulate passively due to dehydration of the leaf. To reveal the active osmotic adjustment and account for the leaf dehydration, another set of leaves (two for each tree) was used to measure the osmotic concentration at full hydration (π100). These leaves were frozen only after their twigs were cut under water and hydrated for 10-120 minutes. By measuring every 10 min the Ψstem of other twigs collected from the same trees and going through the same hydration protocol, we made sure that the hydration time was the minimal time sufficient to achieve Ψstem > -0.2 MPa. This way, we made sure all leaves are near their maximal water content while avoiding potential oversaturation of the leaves. Generally, in the early spring, rehydration was short (10 min), while later in the summer, longer rehydration was required to account for the low leaf conductance. To extract the leaf sap, the leaves were thawed and centrifuged (Eppendorf 5417R, Hamburg, Germany) at 12000g for three minutes. The sap osmotic concentration (π) was evaluated in a vapor osmometer (VAPRO 5600, WESCOR, Puteaux, France). The leaf chemical potential (Ψπ;; MPa) was calculated based on π, according to the Van't Hoff equation. Essentially, 400 mmol/kg were converted into -1 MPa.

To estimate the leaf's turgor pressure (Pturgor; MPa), we subtracted the chemical potential from the stem water potential. Because the stem water potential is higher than the leaf water potential, our calculation possibly overestimated Pturgor by a few bars. More critical for the calculation of Pturgor is the effect of the apoplastic fraction (Af), which contains a very low solute content (McCully, 1999). Because the evaluation of Af is not straightforward (Callister et al., 2006), we simulated the effect of a range of Af on the Pturgor, assuming the apoplastic solute concentration is 0:

`

P t u r g o r = Ψ s t e m - Ψ π 1 - A f

Hydraulic conductance

The twig hydraulic conductance (Ktwig; mmol m-2 s-1 MPa-1) was measured in the second week of May 2020, using the leaf rehydration technique (Brodribb and Holbrook, 2003) on twigs with 4–8 leaves. Long branches (~1 m) were allowed to dry in the lab for up to four days to develop various ΨS. Before measurements, the branch was covered with a plastic bag for 30 min to ensure equilibrium. Then, one twig was used to determine the initial water potential (ΨI) while another was cut underwater to rehydrate. Rehydration lasted for 60-160 seconds, depending on the ΨI (more time for lower ΨI). The rehydrated leaf was covered with a plastic bag for another 30 min before its final water potential (ΨF) was determined. Ktwig was calculated from the leaf rehydration kinetics (Brodribb and Holbrook, 2003):

K t w i g = - l n ⁡ ( Ψ F Ψ I ) × C t

where C is the twig capacitance, and t is the rehydration time in seconds. C (mmol m−2 MPa−1) was measured as the average slope of the pressure-volume relationship, performed as described in (Turner, 1988). At the end of the pressure-volume analysis, the leaf area of the twigs was estimated using ImageJ to calculate their C. Different C was calculated for Ψ'sthat werehigher or lower than the turgor loss point.

In order to evaluate the Ψstem in which 50% of Ktwig is lost (Ktwig50), the data were fit to an exponential sigmoidal regression using a slop coefficient (α) as follows:

K t w i g = 100 1 + e α ( Ψ s t e m - K t w i g 50 )

The complementary conductance to Ktwig is the hydraulic conductance between the soil and stem (Ks-s; does not include the leaf conductance; mmol m-2 s-1 MPa-1). Together they represent the entire hydraulic conductance of the liquid phase in the tree. The Ks-s for each measurement date was evaluated using Ohm's law analogy:

K s - s = E Ψ P D - Ψ s t e m - m i n

where transpiration (E) was measured with the Licor6400 (as described above) at the time the trees exhibited their lowest stem water potential (Ψstem-min; typically at 16:00) and divided by the gradient between the ΨPD and Ψstem-min.

Statistical analysis

Statistical analysis was done using RStudio environment (R Core Team 2021, RStudio Team 2021). To detect diurnal changes in physiological parameters (i.e. Ψstem, AN, gs; Figure. 1), each of the four examined dates was subjected to repeated measures ANOVA using the “ez” package (Lawrence, 2016), where the different trees represented the different subjects and the sampling time was considered as the within-subject variance. Then, values were subjected to Tukey HSD test using the “multcomp” package (Hothorn et al., 2008). Seasonal change of physiological parameters (Figure. 2) was examined using repeated measures ANOVA, with the campaign date considered as the within-subject variance and time of sampling considered as the between-subject variance. Tukey HSD test was conducted separately for pre-dawn and afternoon measurements. Seasonal change in π and π100 (Figure. 3) was tested the same way, where date was considered as the within-subject variance and sampling time as the between-subject variance for both the repeated ANOVA analysis and the Tukey HSD test. The confidence level for all conducted tests was pre-determined to be 95% (i.e. α = 0.05).

 Results

Stem water potential (Ψstem) and gas exchange

The entire set of raw data and its statistical analysis are presented in supplementary file S1. Our measurements revealed that the trees dehydrated down to a Ψstem of -8.1 MPa in the end of the summer while maintaining positive assimilation. All the measured physiological parameters had a clear daily and seasonal pattern. During most of the year, all the measured parameters decreased during the day (Figure 1). The maximal daily Ψstem measured at predawn (ΨPD) or maximal daily stomatal conductance and assimilation that were measured in the morning declined to minimal values in the afternoon (between 12:00 and 18:00). This was only different in winter when the maximal gs and AN were exhibited at midday (Figure 1), probably because low temperatures, rather than water availability, was the limiting factor for gas exchange. The physiological status of the trees changed dramatically throughout the season, mostly as a function of time from the last effective rain (Figure 2). In early April, two days after the last substantial rain (>10 mm), the trees exhibited minimal Ψstem of -0.9 MPa, maximal gs of 0.26 mol m-2 s-1 and maximal AN of 16 μmol m-2 s-1. From April 2019 onward, all measured parameters were gradually declining, reaching minimal values in October 1st (-8.1 MPa, 0.05 mol m-2 s-1, and 4.16 μmol m-2 s-1, for the minimal Ψstem, maximalgs, and maximalAN , respectively). The severe stress at peak summer is also manifested in the subcuticular CO2 concentration which dramatically increased (Figure S3). These low values were maintained until December 2nd (-6.5 MPa, 0.05 mol m-2 s-1, 4.9 μmol m-2 s-1) despite a total of 25.7 mm of precipitation in the two months prior to that measurement. Following the major rains in January 2020 (363 mm), all three parameters made a sharp recovery to maximal values by February 1st 2020 (Figure 2). In agreement with the Darcy equation, as gs and the transpiration decreased during summer, also the difference between ΨPD and the minimal Ψstem decreased from an average of 1.5 MPa in May and June to an average of 0.6 MPa between August and December (Figure 2). It's important to mention that the minimal gs was never lower than 0.023 mol m-2 s-1 (reached in October), even when the minimal Ψstem was -8.1 MPa. It seems that under a seasonal dehydration, the Ψstem at 90% stomatal closure (an important value for species comparison; StPaul et al., 2017) of olives is ca -8 MPa.

Figure 1: The daily patterns of water potential and gas exchange at different seasons.

Stem water potential (󠇣Ψstem; A) net assimilation (AN; B) and stomatal conductance (gs; C) as a function of daytime on four dates during the experiment. Data are means ± SE, n=4. Statistical analysis of the data is presented in supplementary file S1.

Figure 1: The daily patterns of water potential and gas exchange at different seasons.

Stem water potential (󠇣Ψstem; A) net assimilation (AN; B) and stomatal conductance (gs; C) as a function of daytime on four dates during the experiment. Data are means ± SE, n=4. Statistical analysis of the data is presented in supplementary file S1.

Figure 2: Annual pattern of water potential and gas exchange.

Stem water potential (󠇣Ψstem; A), Stomatal conductance (gs; B), and Assimilation (AN; C) measured at predawn (for Ψstem) or morning (for AN and gs), and the afternoon (for all parameters) along the experiment. Data are means ± SE, n=4. Statistical analysis of the data is presented in supplementary file S1.

Figure 2: Annual pattern of water potential and gas exchange.

Stem water potential (󠇣Ψstem; A), Stomatal conductance (gs; B), and Assimilation (AN; C) measured at predawn (for Ψstem) or morning (for AN and gs), and the afternoon (for all parameters) along the experiment. Data are means ± SE, n=4. Statistical analysis of the data is presented in supplementary file S1.

To understand what is the actual minimal gs that olives are capable of, we bench dehydrated 5 small branches for 24 hours under lab conditions until their Ψstem was -5.2 ± 0.06 MPa. We then measured the stomatal conductance using the Licor6400 of both current year leaves and previous year leaves. We found that the minimal gs of 3-month-old leaves was 3 ± 0.4 mmol m-2 s-1 and of 15-month-old leaves was 5.1 ± 0.1 mmol m-2 s-1.

Osmotic adjustments

The large reduction in Ψstem was accompanied by major osmotic adjustment (Figure 3). During summer, the leaf osmotic concentration at predawn (πPD) increased from 866 mmol/kg in April 1st to 2186 and 2082 mmol/kg in August and September, respectively. When π is plotted in respect to the time of year, it goes hand in hand with the Ψstem dynamics, with one major exception. Even after the major rains of January, π100 at predawnwas still 39% higher than its April 2019 value (1089 compared with 780 mmol/kg). This difference probably originates from leaf maturation, as young leaves in April 2020 showed similar π100 to the leaves sampled in April 2019 (780 compared with 858 mmol/kg, respectively). Part of the increase in π during the summer originated from the leaf dehydration. The active osmotic adjustment can be clearly viewed through the seasonal pattern of the osmotic concentration in rehydrated leaves (Figure 3B). The results suggest that most of the osmotic adjustment originated from active osmolyte accumulation. For example, the accumulation of osmotic concentration in rehydrated leaves at predawn explains 55% (754 of 1354 mmol/kg) of the total osmotic adjustment between April and August and 75% (825 of 1129 mmol/kg) of the total osmotic adjustment between April and September.

In addition to the annual osmotic adjustment, the trees also exhibited a daily osmotic adjustment, increasing by up to 578 mmol/kg between predawn and the afternoon (Fig. 3A). About half of that increase (on average 49%) could be attributed to active osmotic adjustment as exhibited by π100 , which increased by 93-233 mmol/kg between predawn and the afternoon (Figure 3B). Despite the assumption that the daily accumulation of osmolites should correspond to the photosynthetic rates, the active osmotic adjustment was not smaller in the dry months (103 mmol/kg between July and November) compared with the spring and early summer (93 mmol/kg between December and June).

Figure 3: Annual dynamics of the leaf osmotic concentration.

The leaf native osmotic concentration (π; A) or saturated osmotic concentration (π100; B) at predawn and the afternoon along the experiment. In the last measurement day (March 20th) we collected data for both mature leaves (the same type of leaves that were followed throughout the experiment) or young leaves that were formed in the same calendaric year. Data are means ± SE, n=4. Statistical analysis of the data is presented in supplementary file S1.

Figure 3: Annual dynamics of the leaf osmotic concentration.

The leaf native osmotic concentration (π; A) or saturated osmotic concentration (π100; B) at predawn and the afternoon along the experiment. In the last measurement day (March 20th) we collected data for both mature leaves (the same type of leaves that were followed throughout the experiment) or young leaves that were formed in the same calendaric year. Data are means ± SE, n=4. Statistical analysis of the data is presented in supplementary file S1.

Hydraulic conductance

Rehydration kinetics measurements suggest that olive twigs should lose 50% of their hydraulic conductance already at -3.2MPa and 88% of their hydraulic conductance at -5.3 MPa (Figure 4A). Such data suggest that at peak summer, the leaves should have lost most of their hydraulic conductance. Measurement of the hydraulic conductance from the soil to the stem (KS-S) (Figure 4B) suggests lower vulnerability as compared with Ktwig, decreasing from 6.4 mmol m-2 s-1 MPa-1 in spring to ~2 mmol m-2 s-1 MPa-1 through most of the summer and exhibiting minimal values of <1 mmol m-2 s-1 MPa-1 in December.

Figure 4: Hydarulic conductance as a function of time and water potential.

Twig hydraulic conductance (Ktwig; A) as a function of stem water potential (ΨS) along the experiment. Plant hydraulic conductance (Ks-s; B) along the experiment. Data are means ± SE, n=4.

Figure 4: Hydarulic conductance as a function of time and water potential.

Twig hydraulic conductance (Ktwig; A) as a function of stem water potential (ΨS) along the experiment. Plant hydraulic conductance (Ks-s; B) along the experiment. Data are means ± SE, n=4.

Discussion

Our results demonstrate olive trees' ability to sustain Ψstem of -8 MPa while maintaining substantial gas exchange throughout the dry summer and recovering full functionality without sacrificing leaves after the winter rains. The -8 MPa that we measured was similar to record lows Ψstem for olive trees under extreme drought (Angelopoulos et al., 1996; Ennajeh et al., 2008; Moriana et al., 2003a) and is a testament to olives' remarkable drought tolerance. The fact that -8 MPa was never surpassed, not even during the seven months drought the trees experienced in the current experiment, might suggest that -8 MPa is the limit and that olives won't be able to recover lower Ψstem.

Osmotic adjustment

The results join a large number of studies that showed olive trees' ability to accumulate high concentrations of osmolytes in response to dehydration (Dichio et al., 1996; Dichio et al., 2006; Ennajeh et al., 2008; Sofo et al., 2008). The similar dynamics of Ψstem and π100 along the season imply that π100 is mostly affected by the plant water status rather than phenological effects. The only exception is the measured increase in π100 upon leaf maturation, which is long known for other species like Soybeans (Glycine max), red and white oak (Quercus Rubra, Quercus Alba), and Carya (Carya tomentosa) (Zur et al., 1981; Parker et al., 1982). While we still don't know which osmolytes drive the changes during leaf maturation, (Fernandez-Escobar et al., 1999) showed that Calcium (a major osmolyte in olive leaves; (Dichio et al., 2009)) more than doubled its concentration during the first year of olive leaves maturation. This higher π of mature leaves implies that they are more resilient to water shortage and could explain why in the bench dehydrated shoots, the minimal gs of 3 months old leaves was lower than the 15 months old leaves.

The daily adjustment of π between predawn and the afternoon highlights olive trees' ability to regulate their turgor in respect to the daily fluctuation in water potential. Our finding that the active adjustment account for nearly half of the total daily change in osmotic concentration is similar to the 42% measured by (Dichio et al., 2006) in potted olives. A large daily osmotic adjustment was also measured in other species ((Bowman and Roberts, 1985; Gersony et al., 2020), implying that the phenomenon is probably common among taxa. It seems reasonable that most of the daily-accumulated osmolytes are photosynthetic products that build up during the day and are exported through the phloem at night (Quick et al., 1992; Gersony et al., 2020). In support, the major osmolytes found in olives are Glucose and Mannitol (Cataldi et al., 2000; Dichio et al., 2009). Even at the end of the summer, under extreme water stress that led to minimal assimilation, there was a daily accumulation of osmolytes, probably thanks to reduced phloem export from the leaves under water stress (Woodruff, 2014).

Hydraulic conductance

Under stress, the reduction in Ktwig was much more prominent than Ks-s, pointing that the hydraulic vulnerability segmentation hypothesis (Tyree and Ewers, 1991) is valid in olives. In support, other studies have also found that olive leaves are more vulnerable as compared with their stems (Torres-Ruiz et al., 2015; Rodriguez‐Dominguez et al., 2018). Both Ktwig and Ks-s measurements imply that by the end of the summer, under -8 MPa, the plants have lost most of their hydraulic conductance, probably as a result of xylem cavitation. It could be argued that our measurements are inaccurate due the importance of outer-xylary regulation of conductance or the effect of compartmentalized capacitance that could effect the rehydration kinetics measurements (Blackman and Brodribb, 2011). However, even more direct methods to quantify xylem cavitation in olives showed that 50% cavitation (P50) in stems or leaves was between -4.5 and -7 MPa (Ennajeh et al., 2008; Torres‐Ruiz et al., 2014; Torres‐Ruiz et al., 2017; Rodriguez‐Dominguez et al., 2018). The fact that the minimal Ks-s values were measured in November and December, when Ψstem were not at their minimum, is probably due to the low temperatures in these months that further reduced the plant hydraulic conductance thanks to their effect on viscosity and aquaporin expression (Murai-Hatano et al., 2008). The recovery of Ks-s in the spring was probably due to the differentiation of new xylem vessels that could have replaced the conductance that was lost at peak summer (Améglio et al., 2002; Hammond et al., 2019). These findings suggest that olives can tolerate and recover a high degree of xylem embolism, as was previously shown for beech, poplars and pines (Barigah et al., 2013; Hammond et al., 2019).

Stomatal conductance in the face of hydraulic failure

An interesting finding was that the trees maintained substantial gas exchange despite the low Ψstem. As observed by many others for nearly every studied tree species, most stomatal closure occurred above -4 MPa (Figure 5; Summarized by (Martin‐StPaul et al., 2017). However, the idea that from that point onward, the plants will minimize their transpiration (Blackman et al., 2016) is not supported by the current findings. Also, the notion that plants regulated their gs to a minimum in order to avoid Ψstem that might lead to xylem cavitation (Sperry et al., 1988) is called into question. The seasonal lowest gs was 23 mmol m-2 s-1, several folds higher than the minimal gs that was measured in the lab (3-5 mmol m-2 s-1) or the minimal gs reported for other species (1-10 mmol m-2 s-1; (Duursma et al., 2019). While the measurement of leaf vapor conductivity (termed in this paper stomatal conductance- gs) includes both the stomata pores and the cuticular conductance, it is unlikely that cuticular conductance had a substantial contribution to the measured gs.

Figure 5: Comparison of stomatal closure in bearing and non-bearing olive trees.

Stomatal conductance (gs) as a function of the stem water potential (ΨS) compared for potted young plants (CV. Souri) and mature bearing field grown trees (CV. Arbequina). The gs and ΨS were measured following the same protocols of the current paper during 34 days of dehydration. Complete experimental description is available in Barzilai et al. (2021). Data are means ± SE. For the field grown tree n=4 and for the potted tree n=6.

Figure 5: Comparison of stomatal closure in bearing and non-bearing olive trees.

Stomatal conductance (gs) as a function of the stem water potential (ΨS) compared for potted young plants (CV. Souri) and mature bearing field grown trees (CV. Arbequina). The gs and ΨS were measured following the same protocols of the current paper during 34 days of dehydration. Complete experimental description is available in Barzilai et al. (2021). Data are means ± SE. For the field grown tree n=4 and for the potted tree n=6.

It is difficult to say if the relatively high gs values that were measured under -8 MPa were kept thanks to turgor maintenance or despite negative turgor. Simulations of the leaf turgor using the apoplastic fractions that were previously measured for olives (from 10% in (Dichio et al., 1996) to 40% in (Arias et al., 2015)) do not clarify exactly if and when the trees lost turgor (Figure 6). However, it is more than likely that these relatively high gs were maintained despite substantial cavitation (discussed above).

In agreement with our results, several other studies have also found that plants under extreme drought maintained gs far above the minimal values. For example, (Odening et al., 1979; Odening et al., 1974) found stomatal opening and positive assimilation for L. tridentata at -8 MPa. Furthermore, (Hoffmann et al., 2011), who investigated forests under extreme drought conditions in southeastern USA, found that despite extremely low predawn water potentials (<-4MPa) that led to substantial wilting and leaf desiccation, most plants exhibited gs that was higher than 10 mmol m-2 s-1. Also, in other olives experiments under low Ψstem (<-4 MPa), significant stomatal conductance was measured (0.25 < gs <0.1 mol m-2 s-1; (Moriana et al., 2003a; Naor et al., 2013; Perez-Martin et al., 2014; Erel et al., 2014; Torres-Ruiz et al., 2015)). These findings remind us that the hydraulic signal is not the sole regulator of the stomata.

The high temperatures of the Mediterranean summer could increase minimal gs by their effect on the cuticule integrity (Schuster et al., 2016) or the leaf latent cooling requierments (Marchine et al., 2021). Cuticular damage typically occurs at >40 °C (Schuster et al., 2016; Billon et al., 2020), while the maximal leaf temperature during measurement days (measured by the licor6400; File S1) was always lower than 38°C. However, we didn't measure on the hottest days of summer, and it is probable that leaf temperatures crossed the 40°C at some point of the season (see Fig S1), possibly increasing the cuticle permeability. Another possibility is that the hot summer temperatures (>30 °C in the measuring days of August, September and October) forced the trees to relatively high minimal gs to maintain latent leaf cooling despite the cavitation danger.

Figure 6: Turgor pressure simulations for four different apoplast fraction along the experiment.

Data are means ± SE, n=4.

Figure 6: Turgor pressure simulations for four different apoplast fraction along the experiment.

Data are means ± SE, n=4.

In olives, perhaps more than in any other crop, fruit-bearing was shown to regulate stomatal conductance. (Bustan et al., 2016) used large (2.5 m3) lysimeters to show that fruit removal reduced transpiration by 30%. Additionally, stomatal response to water potential was shown to depend on the relative fruit load, where high fruit load led to substantial stomatal conductance even under very low Ψstem (Trentacoste et al., 2011; Ben-Gal et al., 2011; Naor et al., 2013). In the only other study where field-grown olives were dried to -8MPa (Moriana et al., 2003b), they exhibited gs higher than 25 mmol m-2 s1 in the 'on' year, but much lower gs and higher Ψstem in the 'off' year. When reviewing the existing literature on olives water relations, it becomes apparent that complete stomatal closure under low Ψstem occurred only in non-bearing trees (Angelopoulos et al., 1996; Sofo et al., 2004; Perez-Martin et al., 2014). Studies on fruit-bearing olive trees rarely measured stomatal conductance lower than 30 mmol m-2 s-1. To emphasize the difference between dehydration of potted young trees and mature bearing trees, Figure 5 combines data of gs in respect to Ψstem from the current study with a dataset that we recently collected for potted (10L) non-bearing olives (CV. Souri; Barzilai et al., (2021)). The figure portrays the "unexpected" behavior of mature-bearing trees compared with the minimal stomatal conductance (<10 mmol m-2 s-1) exhibited by the young potted trees from -4 MPa downward (Figure 5). One clear difference between potted and field trees is the root volume that when combined with low Ψstem enables sufficient suction to utilize deep water pools. Alternatively, these findings imply that gas exchange regulation is not strictly hydraulic and that other factors, such as temperatures and fruit load, should be considered when modeling water loss under extreme drought.

In conclusion, olive trees can sustain positive assimilation under the long and dry Israeli summer even when driven to a very low Ψstem, partially thanks to their ability to accumulate a large concentration of osmolytes. The trees lost most of their stomatal and hydraulic conductance during the dry period, but these were fully recovered after the winter rains. It seems that olive trees maintain their gs above the minimal values even when surpassing the threshold of xylem cavitation.

Supplemental data

File S1. The entire raw data and statistical analysis of the data presented in figures 1,2, and 3.

Figure S1. Precipitation and Penman values during the experiment

Figure S2. Daily maximun and minimum temperature values during the experiment.

Figure S3. Sub cutictular carbon concentration (Ci) during the experiment.

Figure S4. The hydraulic conductivity between the soil and the stem (Ks-s) as a function of stem water potential (Ψstem) along the experiment

Améglio T. , Bodet C. , Lacointe A. , Cochard H. , 2002. Winter embolism, mechanisms of xylem hydraulic conductivity recovery and springtime growth patterns in walnut and peach trees. Tree Physiology, 22(17), 1211-1220. https://doi.org/10.1093/treephys/22.17.1211. Angelopoulos K. , Dichio B. , Xiloyannis C. , 1996. Inhibition of photosynthesis in olive trees (Olea europaea L.) during water stress and rewatering. Journal of Experimental Botany, 47(8), 1093-1100. https://doi.org/10.1093/jxb/47.8.1093. Arias N. S. , Bucci S. J. , Scholz F. G. , Goldstein G. , 2015. Freezing avoidance by supercooling in O lea europaea cultivars: the role of apoplastic water, solute content and cell wall rigidity. Plant, Cell & Environment, 38(10), 2061-2070. https://doi.org/10.1111/pce.12529. Barigah T. S. , Charrier O. , Douris M. , Bonhomme M. , Herbette S. , Améglio T. , Fichot R. , Brignolas F. , Cochard H. , 2013. Water stress-induced xylem hydraulic failure is a causal factor of tree mortality in beech and poplar. Annals of Botany, 112(7), 1431-1437. https://doi.org/10.1093/aob/mct204 . Ben-Gal A. , Yermiyahu U. , Zipori I. , Presnov E. , Hanoch E. , Dag A. , 2011. The influence of bearing cycles on olive oil production response to irrigation. Irrigation Science, 29(3), 253-263. https://doi.org/10.1007/s00271-010-0237-1 . Blackman C. J. , Pfautsch S. , Choat B. , Delzon S. , Gleason S. M. , Duursma R. A. , 2016. Toward an index of desiccation time to tree mortality under drought. Plant, Cell & Environment, 39(10), 2342-2345. https://doi.org/10.1111/pce.12758 . Bowman W. D. , Roberts S. W. , 1985. Seasonal changes in tissue elasticity in chaparral shrubs. Physiologia Plantarum, 65(3), 233-236. https://doi.org/10.1111/j.1399-3054.1985.tb02388.x. Brodribb T. J. , Holbrook N. M. , 2003. Stomatal closure during leaf dehydration, correlation with other leaf physiological traits. Plant Physiology, 132(4), 2166-2173. https://doi.org/10.1104/pp.103.023879. Bustan A. , Dag A. , Yermiyahu U. , Erel R. , Presnov E. , Agam N. , Kool D. , Iwema J. , Zipori I. , Ben-Gal A. , 2016. Fruit load governs transpiration of olive trees. Tree Physiology, 36(3), 380-391. https://doi.org/10.1093/treephys/tpv138. Callister A. N. , Arndt S. K. , Adams M. A. , 2006. Comparison of four methods for measuring osmotic potential of tree leaves. Physiologia Plantarum, 127(3), 383-392. https://doi.org/10.1111/j.1399-3054.2006.00652.x. Cataldi T. R. , Margiotta G. , Iasi L. , Di Chio B. , Xiloyannis C. , Bufo S. A. , 2000. Determination of sugar compounds in olive plant extracts by anion-exchange chromatography with pulsed amperometric detection. Analytical Chemistry, 72(16), 3902-3907. https://doi.org/10.1021/ac000266o. Diaz‐Espejo A. , Nicolás E. , Fernández J. E. , 2007. Seasonal evolution of diffusional limitations and photosynthetic capacity in olive under drought. Plant Cell and Environment, 30(8), 922-933. https://doi.org/10.1111/j.1365-3040.2007.001686.x. Dichio B. , Margiotta G. , Xiloyannis C. , Bufo S. A. , Sofo A. , Cataldi T. R. I. , 2009. Changes in water status and osmolyte contents in leaves and roots of olive plants ( Olea europaea L.) subjected to water deficit. Trees-Structure and Function, 23(2), 247-256. https://doi.org/10.1007/s00468-008-0272-1. Dichio B. , Nuzzo V. , Xiloyannis C. , Celano G. , Angelopoulos K. , 1996. Drought Stress-induced Variation of Pressure-volume Relationships in Olea Europaea L. cv." Coratina". Paper presented at the II International Symposium on Irrigation of Horticultural Crops 449, 401-410. https://doi.org/10.17660/ActaHortic.1997.449.56. Dichio B. , Xiloyannis C. , Sofo A. , Montanaro G. , 2006. Osmotic regulation in leaves and roots of olive trees during a water deficit and rewatering. Tree Physiology, 26(2), 179-185. https://doi.org/10.1093/treephys/26.2.179. Duursma R. A. , Blackman C. J. , Lopez R. , Martin-StPaul N. K. , Cochard H. , Medlyn B. E. , 2019. On the minimum leaf conductance: its role in models of plant water use, and ecological and environmental controls. New Phytologist, 221(2), 693-705. https://doi.org/10.1111/nph.15395 . Ennajeh M. , Tounekti T. , Vadel A. M. , Khemira H. , Cochard H. , 2008. Water relations and drought-induced embolism in olive (Olea europaea) varieties 'Meski'and 'Chemlali'during severe drought. Tree Physiology, 28(6), 971-976. https://doi.org/10.1093/treephys/28.6.971. Erel R. , Ben-Gal A. , Dag A. , Schwartz A. , Yermiyahu U. , 2014. Sodium replacement of potassium in physiological processes of olive trees (var. Barnea) as affected by drought. Tree Physiology, 34(10), 1102-1117. https://doi.org/10.1093/treephys/tpu081 . . Fernández J. , 2014. Understanding olive adaptation to abiotic stresses as a tool to increase crop performance. Environmental and Experimental Botany, 103, 158-179. https://doi.org/10.1016/j.envexpbot.2013.12.003 . Fernandez-Escobar R. , Moreno R. , Garcıa-Creus M. , 1999. Seasonal changes of mineral nutrients in olive leaves during the alternate-bearing cycle. Scientia Horticulturae, 82(1), 25-45. https://doi.org/10.1016/S0304-4238(99)00045-X . Gersony J. T. , Hochberg U. , Rockwell F. E. , Park M. , Gauthier P. P. , Holbrook N. M. , 2020. Leaf carbon export and non-structural carbohydrates in relation to diurnal water dynamics in mature oak trees. Plant Physiology 183(4), 1612-1621. https://doi.org/10.1104/pp.20.00426 . Giorio P. , Sorrentino G. , d'Andria R. , 1999. Stomatal behaviour, leaf water status and photosynthetic response in field-grown olive trees under water deficit. Environmental and Experimental Botany, 42(2), 95-104. https://doi.org/10.1016/S0098-8472(99)00023-4 . Hammond W. M. , Yu K. , Wilson L. A. , Will R. E. , Anderegg W. R. , Adams H. D. , 2019. Dead or dying? Quantifying the point of no return from hydraulic failure in drought‐induced tree mortality. New Phytologist, 223(4), 1834-1843. https://doi.org/10.1111/nph.15922. Hasibeder R. , Fuchslueger L. , Richter A. , Bahn M. , 2015. Summer drought alters carbon allocation to roots and root respiration in mountain grassland. New Phytologist, 205(3), 1117-1127. https://doi.org/10.1111/nph.13146. Hoffmann W. A. , Marchin R. M. , Abit P. , Lau O. L. , 2011. Hydraulic failure and tree dieback are associated with high wood density in a temperate forest under extreme drought. Global Change Biology, 17(8), 2731-2742. https://doi.org/10.1111/j.1365-2486.2011.02401.x . Kitsaki C. , Drossopoulos J. , 2005. Environmental effect on ABA concentration and water potential in olive leaves (Olea europaea L. cv "Koroneiki") under non-irrigated field conditions. Environmental and Experimental Botany, 54(1), 77-89. https://doi.org/10.1016/j.envexpbot.2004.06.002. . Langgut D. , Cheddadi R. , Carrión J. S. , Cavanagh M. , Colombaroli D. , Eastwood W. J. , Greenberg R. , Litt T. , Mercuri A. M. , Miebach A. , 2019. The origin and spread of olive cultivation in the Mediterranean Basin: The fossil pollen evidence. The Holocene, 29(5), 902-922. https://doi.org/10.1177/0959683619826654 . Lionello P. , Abrantes F. , Gacic M. , Planton S. , Trigo R. , Ulbrich U. , 2014. The climate of the Mediterranean region: research progress and climate change impacts. Regional Environmental Change, 14(5), 1679-1684. https://doi.org/10.1007/s10113-014-0666-0. Martin‐StPaul N. , Delzon S. , Cochard H. , 2017. Plant resistance to drought depends on timely stomatal closure. Ecology Letters, 20(11), 1437-1447. https://doi.org/10.1111/ele.12851. McCully M. E. , 1999. Root xylem embolisms and refilling. Relation To water potentials of soil, roots, and leaves, and osmotic potentials of root xylem Sap. Plant Physiology, 119(3), 1001-1008.https://doi.org/10.1104/pp.119.3.1001. Moriana A. , Orgaz F. , Pastor M. , Fereres E. , 2003a. Yield responses of a mature olive orchard to water deficits. Journal of the American Society for Horticultural Science, 128(3), 425-431. https://doi.org/10.21273/JASHS.128.3.0425. Moriana A. , Orgaz F. , Pastor M. , Fereres E. , 2003b. Yield responses of a mature olive orchard to water deficits. Journal of the American Society for Horticultural Science, 128(3), 425-431. https://doi.org/10.21273/JASHS.128.3.0425. Murai-Hatano M. , Kuwagata T. , Sakurai J. , Nonami H. , Ahamed A. , Nagasuga K. , Matsunami T. , Fukushi K. , Maeshima M. , Okada M. , 2008. Effect of low root temperature on hydraulic conductivity of rice plants and the possible role of aquaporins. Plant and Cell Physiology, 49(9), 1294-1305. https://doi.org/10.1093/pcp/pcn104. Naor A. , Schneider D. , Ben-Gal A. , Zipori I. , Dag A. , Kerem Z. , Birger R. , Peres M. , Gal Y. , 2013. The effects of crop load and irrigation rate in the oil accumulation stage on oil yield and water relations of 'Koroneiki'olives. Irrigation Science, 31(4), 781-791. https://doi.org/10.1007/s00271-012-0363-z. Nasrallah H. A. , Balling R. C. , 1995. Impact of desertification on temperature trends in the Middle East. Environmental Monitoring and Assessment, 37(1-3), 265-271. https://doi.org/10.1007/BF00546894. Odening W. , Oechel W. , Strain B. , 1979. Non-positive turgor potential and growth in Larrea. Larrea, , 151-163.. Odening W. , Strain B. , Oechel W. , 1974. The effect of decreasing water potential on net CO2 exchange of intact desert shrubs. Ecology, 55(5), 1086-1095. https://doi.org/10.2307/1940358. Parker W. C. , Pallardy S. G. , Hinckley T. M. , Teskey R. O. , 1982. Seasonal changes in tissue water relations of three woody species of the Quercus-Carya forest type. Ecology, 63(5), 1259-1267. https://doi.org/10.2307/1938853. Perez-Martin A. , Michelazzo C. , Torres-Ruiz J. M. , Flexas J. , Fernández J. E. , Sebastiani L. , Diaz-Espejo A. , 2014. Regulation of photosynthesis and stomatal and mesophyll conductance under water stress and recovery in olive trees: correlation with gene expression of carbonic anhydrase and aquaporins. Journal of Experimental Botany, 65(12), 3143-3156. https://doi.org/10.1093/jxb/eru160. Quick W. , Chaves M. , Wendler R. , David M. , Rodrigues M. , Passaharinho J. , Pereira J. , Adcock M. , Leegood R. , Stitt M. , 1992. The effect of water stress on photosynthetic carbon metabolism in four species grown under field conditions. Plant, Cell & Environment, 15(1), 25-35. https://doi.org/10.1111/j.1365-3040.1992.tb01455.x. Rodriguez‐Dominguez C. M. , Murphy M. R. C. , Lucani C. , Brodribb T. J. , 2018. Mapping xylem failure in disparate organs of whole plants reveals extreme resistance in olive roots. New Phytologist, 218(3), 1025-1035. https://doi.org/10.1111/nph.15079. Sofo A. , Dichio B. , Xiloyannis C. , Masia A. , 2004. Lipoxygenase activity and proline accumulation in leaves and roots of olive trees in response to drought stress. Physiologia Plantarum, 121(1), 58-65. https://doi.org/10.1111/j.0031-9317.2004.00294.x . Sofo A. , Manfreda S. , Fiorentino M. , Dichio B. , Xiloyannis C. , 2008. The olive tree: A paradigm for drought tolerance in Mediterranean climates. Hydrology and Earth System Sciences, 12, 293-301. https://doi.org/10.5194/hessd-4-2811-2007. Sperry J. , Donnelly J. , Tyree M. , 1988. A method for measuring hydraulic conductivity and embolism in xylem. Plant, Cell & Environment, 11(1), 35-40. https://doi.org/10.1111/j.1365-3040.1988.tb01774.x. Tognetti R. , d'Andria R. , Sacchi R. , Lavini A. , Morelli G. , Alvino A. , 2007. Deficit irrigation affects seasonal changes in leaf physiology and oil quality of Olea europaea (cultivars Frantoio and Leccino). Annals of Applied Biology, 150(2), 169-186. https://doi.org/10.1111/j.1744-7348.2007.00117.x. Torres‐Ruiz J. M. , Cochard H. , Mayr S. , Beikircher B. , Diaz‐Espejo A. , Rodriguez‐Dominguez C. M. , Badel E. , Fernández J. E. , 2014. Vulnerability to cavitation in Olea europaea current‐year shoots: further evidence of an open‐vessel artifact associated with centrifuge and air‐injection techniques. Physiologia Plantarum, 152(3), 465-474. https://doi.org/10.1111/ppl.12185. Torres‐Ruiz J. M. , Cochard H. , Choat B. , Jansen S. , López R. , Tomášková I. , Padilla‐Díaz C. M. , Badel E. , Burlett R. , King A. , Lenoir N. , Martin‐StPaul N. K. , Delzon S. , 2017. Xylem resistance to embolism: presenting a simple diagnostic test for the open vessel artefact. New Phytologist, 215(1), 489-499. https://doi.org/10.1111/nph.14589 . Torres-Ruiz J. M. , Diaz-Espejo A. , Perez-Martin A. , Hernandez-Santana V. , 2015. Role of hydraulic and chemical signals in leaves, stems and roots in the stomatal behaviour of olive trees under water stress and recovery conditions. Tree Physiology, 35(4), 415-424. https://doi.org/10.1093/treephys/tpu055. Trentacoste E. R. , Sadras V. O. , Puertas C. M. , 2011. Effects of the source: sink ratio on the phenotypic plasticity of stem water potential in olive (Olea europaea L.). Journal of Experimental Botany, 62(10), 3535-3543. https://doi.org/10.1093/jxb/err044. Tugendhaft Y. , Eppel A. , Kerem Z. , Barazani O. , Ben-Gal A. , Kadereit J. W. , Dag A. , 2016. Drought tolerance of three olive cultivars alternatively selected for rain fed or intensive cultivation. Scientia Horticulturae, 199, 158-162. https://doi.org/10.1016/j.scienta.2015.12.043. Turner N. C. , 1988. Measurement of plant water status by the pressure chamber technique. Irrigation Science, 9(4), 289-308. https://doi.org/10.1007/BF00296704. Tyree M. T. , Ewers F. W. , 1991. The hydraulic architecture of trees and other woody plants. New Phytologist, 119(3), 345-360. https://doi.org/10.1111/j.1469-8137.1991.tb00035.x . Woodruff D. R. , 2014. The impacts of water stress on phloem transport in Douglas-fir trees. Tree Physiology, 34(1), 5-14. https://doi.org/10.1093/treephys/tpt106 . Zur B. , Boote K. , Jones J. , 1981. Changes in internal water relations and osmotic properties of leaves in maturing soybean plants. Journal of Experimental Botany, 32(6), 1181-1191. https://doi.org/10.1093/jxb/32.6.1181.