Short-term responses of growth and leaf physiology of the sclerophyllous Cryptocarya alba (Molina) seed sources to water restriction

ABSTRACT

There are concerns about the vulnerability of some dominant sclerophyllous species in Mediterranean zones to water restriction, especially at outplanting, and much of the efforts for active restoration are on identifying proper seed sources. This study aimed to determine the short-term responses of growth, biomass partitioning, and leaf-physiological traits of Cryptocarya alba provenances to water restriction in nursery conditions. Three seed sources from xeric sites (northern distribution) and one from a more mesic site (southern distribution) were subjected to water restriction for three months and then evaluated in increments in root collar diameter and height, biomass components (leaves, shoots, roots), gas-exchange and fluorescence parameters, and carbon (Δ¹³) and oxygen (δ¹⁸O) isotopic ratios. Water restriction decreased growth and leaf biomass as a short-term response, but this effect was not expressed in stem and root biomass. Similarly, water restriction decreased the photosynthetic rate (A_sat), stomatal conductance (g_s), and maximum efficiency of PSII. Moreover, water restriction also increased the intrinsic water use efficiency (WUE_int), but it had no effect on Δ¹³ or δ¹⁸O. Seed sources did not differ in any traits except in stem and total biomass. The results did not confirm a better performance of seed sources from xeric than mesic origin sites to water restriction.

KEYWORDS:

• Water restriction
• morpho-physiological responses
• provenances
• sclerophyllous species

Introduction

Although plants in Mediterranean ecosystems have evolved to grow with low water and nutrient availability (Sardans and Penuelas 2013), the expected higher temperatures and drought frequency due to climate change represent a threat to the plant growth and survival of multiple species (Bussotti and Pollastrini 2020). The Mediterranean-type ecosystems of central Chile have been experiencing a mega-drought since 2010 (Garreaud et al. 2017), a condition that might prevail in the future, producing a higher forest decline compared with other Mediterranean zones in the world. Because of this scenario and the current ecosystem degradation, there are concerns about the responses of some dominant sclerophyllous species to increasing water restrictions and how to restore their natural populations. Typically, local seed sources are recommended for restoration projects, but species with fragmented and isolated populations might be at higher risk under the changing climate because of lower phenotypic plasticity (Valladares et al. 2007). Thus, the early screening of seedlings from different genetic sources to water restriction is relevant to understanding their ability to cope with climate change and to improve the guidelines for active restoration with these species.

Plant species may differ in their strategies to cope with water restriction, but some common biochemically short-term responses are an increase of foliar proline, ABA, and reactive oxygen species, which are later down-regulated in the long term (Xu et al. 2010; Niether et al. 2020). These fast biochemical changes also trigger physiological responses, including an increase in stomatal closure and then in CO₂ assimilation and plant growth (Osakabe et al. 2014). As the water restriction persists (long-term effect), the carbon partitioning within the tree is optimized, with a higher investment of carbon to roots than to shoots (Ledo et al. 2018). Overall, growth is reduced at moderate and low water restrictions, while mortality is induced when water restrictions are more severe (Niether et al. 2020). In this context, intraspecific variation (i.e. seed source differences) has been observed in morphological traits such as root-to-shoot ratio, stomatal conductance, and intrinsic water use efficiency, which contribute to sustaining high CO₂ assimilation rates under drought conditions of different populations (Junker et al. 2017). Thus, identifying traits that respond rapidly to water restriction may help screening and select suitable seed sources for drier site conditions and address management decisions.

Cryptocarya alba (Molina) Looser. (Lauraceae family, common name Peumo) is one of the dominant species of the sclerophyllous forest of central Chile. It is an endemic species distributed from arid (31° S) to humid climates (40° S) (Rodríguez et al. 1983). The species is classified as shade-tolerant and shallow-rooted, so it grows better in mesic sites (Cabello and Donoso 2013; Ovalle et al. 2015). Compared with other sclerophyllous species, part of the susceptibility of C. alba to drought is likely due to its shallow-rooting habit, being unable to access groundwater. Under water restrictions, C. alba decreases leaf area and exhibits osmotic adjustment to maintain turgor without modifying the root-to-shoot ratio (i.e. the ratio between the area that absorbs water to the transpiring area) (Donoso et al. 2011; Ovalle et al. 2015). Similar behavior has been observed in Lithraea caustica Hook. et Arn., a species that coexists with C. alba in shaded and humid areas (Peña-Rojas et al. 2018). C. alba exposed to severe water restriction also showed significant recovery after rehydration (Donoso et al. 2011; Espinoza et al. 2021), which might be attributed to the higher production of fine roots (Ovalle et al. 2015). Even though the morphological and physiological responses of C. alba to water restriction are dependent on the seed source origin (Alvarez-Maldini et al. 2020; Espinoza et al. 2021), it is still unclear whether seed sources from more xeric locations might perform better under water restriction. We hypothesized that xeric seed sources of C. alba are more adapted to water restriction, which is influenced by a faster short-term acclimation to that condition when compared with non-xeric sources. In this study, we established a nursery experiment with 2-year-old C. alba seedlings, subjected to two water restriction levels for three months. We aimed to assess the short-term acclimation responses of C. alba seedlings from different genetic sources to severe water restriction.

Materials and methods

Plant material

The plant material used in this study was provided by the Centro de Semillas y Árboles Forestales (CESAF), Universidad de Chile, Chile, and corresponds to seedlings from four seed sources (Cuesta La Dormida, Antumapu, Cantillana, and Cayumanque) spanning the species distribution range (Table 1). All seedlings were containerized (volume of 100 cm³) except the ones from Antumapu provenance, which were grown in black polyethylene bags (volume of 396 cm³). In December 2019, after two years, all seedlings were transplanted to polyethylene bags of volume 4,500 cm³ and grown in a nursery located in the Colina locality until October 2020 (33°04’19,5” S 70°43’52,1” W), Metropolitan Region, Chile. Nursery conditions during the experiment included a cover with an 80% black polyethylene mesh (Raschel®, Santiago, Chile) and ambient temperature. The substrate was an operational mixture of local soil, coconut fiber, and sand (2:2:1 v/v), which was sterilized in an oven at 105°C for two hours. Water was provided by an automatic sprinkler irrigation system every other day, maintaining the substrate at field capacity, which was monitored before irrigation with a portable soil moisture sensor Teros 12 (Meter Group Inc., Pullman, WA).

Table 1. Climatic and geographic data for provenances Cuesta La Dormida (CD), Antumapu (AN), Cantillana (CA), and Cayumanque (CY).

Provenances (number of mother trees)	Climatic classification	Latitude/longitude	Elevation (m a.s.l.)	M.A.R.^1 (mm)	M.A.T.^2 (°C)
Cuesta la dormida CD	Suprathermal warm temperate, semi-arid humid regime (Csb2Sa)	33°04’17’“/70°58’00”’	743	429	14.4
Antumapu AN	Temperate Mediterranean with dry hot summer (Csb)	33°34’19’“/70°37’53”’	629	371	14.5
Cantillana CA	Suprathermal warm temperate, semi-arid humid regime (Csb2Sa)	33°52’07’“/70°55’20”’	373	543	12.0
Cayumanque CY	Temperate, warm summers and cold winters (Cfb)	36°42’31’“/72°31’52”’	733	1.035	13.4

MAR = mean annual rainfall, MAT = mean annual temperature.

Experimental design

At the end of October 2020, seedlings were arranged in a split-plot design with three replications. The main plot corresponded to the factor irrigation regime with two levels (i.e. well-watered (WW) and water restriction (WR)), while the subplot corresponded to the factor seed sources with four levels as the subplot (i.e. Antumapu, Cantillana, Cayumanque, and La Dormida). Each replicate had five seedlings in treatment WR and ten in treatment WW. In this phase, seedlings were irrigated manually, applying 400 cm³ of water every other day in the WW treatment and 150 cm³ once a week in the WR treatment. During the experiment, soil moisture content was monitored with a sensor Teros 12 (Meter Grpoup Inc., Pullman, WA) (Figure 1). Previously, to impose water restriction treatment, the substrate moisture content was maintained close to saturation (0.4 m³ m⁻³). Then, treatments of WW and WR were maintained approximately at 0.27 and 0.13 m³ m⁻³, representing values close to the field capacity and permanent wilting point, respectively. The watering treatments were applied until January 2021 (i.e. three months). Plant position within the plots was continuously changed to avoid potential border effects.

Figure 1. Maximum, mean, minimum temperatures (a), and soil moisture content (b) per watering treatment during the study period. Asterisks indicate significant differences among temperature (A) and soil moisture content (B) at each measurement date at a significant level of 0.05.

Morphological and physiological measurements

Plants were measured in root collar diameter (D) and height (H) immediately before and after applying the watering treatments. Then, growth increments were calculated as D (IncD) and H (IncH) differences between the two measuring dates. After the experiment, survival (Surv) was recorded as a categorical trait (1 = alive, 0 = dead), and five seedlings were selected for leaf-physiological measurements. Light-saturated photosynthesis rate (A_sat, μmol m⁻² s⁻¹), stomatal conductance (g_s, mmol m⁻² s⁻¹), intrinsic water use efficiency (WUE_int= A_sat/ g_s, μmol mmol⁻¹), and efficiency of PSII under light conditions (Fv’/Fm’) were measured in one fully expanded leaf in the upper part of the plant, using a portable gas exchange system LI 6800 (LICOR Inc., Lincoln, NE, U.S.A.). We used a light intensity of 90% red and 10% blue. For a proper interpretation of the fluorescence data, we prefer to use the fraction of light dissipated thermally (HD = 1- Fv’/Fm’) (Demmig-Adams et al. 1996). Chamber conditions were set up at ambient conditions during measurements, with a temperature of 20°C, CO₂ concentration of 400 ppm, relative humidity of 50%, vapor pressure deficit (VPD) of 1.8 kPa, and Photosynthetic active radiation (PAR) of 1,800 µmol m⁻² s⁻¹. Measurements were performed between 11.00 and 14.00 h at local time. Leaves with sizes smaller than 6 cm² were scanned and analyzed using image analysis software (ImageJ, Rasband/NIH, Bethesda, MD, U.S.A.) to obtain leaf area and apply the corresponding corrections to the leaf-physiological parameters. Afterward, three seedlings per plot were extracted from the substrate and then oven-dry at 65°C for 48 hours to determine biomass components. We determined the dry mass of leaves (LDM, g), shoots (SDM, g), roots (RDM, g), total dry mass (TDM, g), and root-to-shoot ratio (RDM/SDM).

Additionally, on three plants per plot, 1.2 mg of dried foliage was finely ground in a ball mill to determine the carbon (δ¹³C) and oxygen (δ¹⁸O) isotope composition using a Sercon INTEGRA2 elemental analyzer-isotope ratio mass spectrometer at the Universidad de Chile. The measurements were standardized to the relative values of Pee Dee Belemnite (PDB) and Vienna Standard Mean Ocean Water (VSMOW) for carbon and oxygen, respectively. The carbon isotope discrimination (Δ¹³C; ‰) was estimated according to Farquhar et al. (1989), considering an air¹³C isotopic composition of ˗8 ‰.

Statistical analysis

All traits were analyzed at the plant level using a mixed model containing the fixed effects of the watering treatment (W), seed source (P), interaction W×P, and the random effects of the whole- and split-plot. Because seed sources exhibited some growth differences before the watering treatment, we included the pre-treatment plant height (Hbefore) in the model as a covariate for growth increments and biomass traits. Survival was analyzed with a generalized linear model with a binomial distribution and a logit link function. For soil moisture data, we performed a simple repeated measured analysis that included the fixed effects of W, Date, and the interaction W×Date. In some traits, we used the log transformation to meet the assumption of normality and homoscedasticity. Post-hoc means comparisons were based on the Šidák´s test. Significant differences were considered at a probability level of 0.05. The statistical analyses were performed with the functions lmer and glmer from package lme4 of the R software (R Core Team 2013).

Results

Growth increments, biomass components, and survival

Table 2 shows the results from the analysis of variance and the means of growth and biomass traits per level of watering treatment and provenance. The covariate used in the analyses was significant for all biomass and growth increment traits (p < 0.05), except for the root-to-shoot ratio and increment in diameter. None of these traits was affected by the watering × seed source interaction (W×P) (p > 0.05) (Table 2). The water restriction treatment (WR) significantly decreased height and collar diameter increments, but these subtle differences were not reflected in the stem or total biomass (p < 0.05, Table 2). Treatment WR also had a higher impact on the increments in height (74% less) than collar diameter (53% less) relative to treatment WW. Treatment WR also decreased a 26% the leaf biomass but did not affect other biomass components. Overall, there were no differences among seed sources on any assessed traits, except on stem and total biomass (p < 0.05, Table 2). Provenances Antumapu and La Dormida tended to have higher stem biomass than the other provenances. Plot survival ranged from 40% to 100%, with an average of 87.5%. However, neither the watering treatments nor the seed source had an effect on this trait.

Table 2. P-values from the analysis of variance and means (standard errors in parenthesis) per watering treatment (W) and provenance (P) on biomass, growth increment, and survival on C. alba in the nursery experiment. Provenances were Cuesta La Dormida (CD), Antumapu (AN), Cantillana (CA), and Cayumanque (CY). Different letters within a column indicate significant differences according to Šidák´s test. Cov(Hbefore) corresponds to the covariate effect of the initial height before imposing the water restriction.

Source of Variation	Biomass					Growth increment		Surv
	Leaf	Stem	Root	Total	Root: Shoot	IncD	IncH
Watering (W)	0.0151	0.1112	0.3591	0.1035	0.6002	0.0002	<0.0001	0.9475
Provenance (P)	0.1614	0.0091	0.0491	0.0306	0.1386	0.4573	0.7327	0.0527
W×P	0.0521	0.4486	0.4384	0.2156	0.261	0.5755	0.4727	0.5940
Cov(Hbefore)	<0.0001	0.0002	0.0007	<0.0001	0.1930	0.0999	0.0058
Means	(g)	(g)	(g)	(g)		(mm)	(cm)	%
WW	5.8 a (0.27)	4.5 a (0.21)	10.5 a (0.66)	20.8 a (1.09)	2.5 a (0.23)	1.5 a (0.12)	16.9 a (0.83)	80.1 a (5.12)
WR	4.3 b (0.28)	3.7 a (0.21)	9.6 a (0.67)	17.6 a (1.10)	2.6 a (0.23)	0.7 b (0.14)	4.3 b (1.05)	91.3 a (3.11)
La Dormida	4.7 a (0.28)	3.7 b (0.25)	11.9 a (0.94)	20.3 ab (1.22)	3.3 a (0.33)	0.9 a (0.19)	11.5 a (1.34)	86.6 a (9.54)
Antumapu	5.5 a (0.30)	5.1 a (0.26)	10.7 a (1.02)	21.5 a (1.29)	2.3 a (0.36)	0.9 a (0.19)	8.9 a (1.36)	86.6 a (4.59)
Cantillana	5.0 a (0.28)	3.8 b (0.25)	8.6 b (0.96)	17.6 b (1.23)	2.3 a (0.33)	1.2 a (0.19)	11.6 a (1.39)	80.2 a (8.56)
Cayumanque	4.9 a (0.28)	3.7 b (0.25)	8.8 b (0.95)	17.5 b (1.21)	2.5 a (0.33)	1.3 a (0.18)	10.2 a (1.28)	96.6 a (4.52)

Leaf-physiological parameters

Table 3 shows the results from the analysis of variance and the means of leaf-physiological traits per level of the watering treatment and provenance. Similarly, all the leaf-gas exchange parameters were significantly affected by the WR treatment (p < 0.05, Table 3). WR decreased A_sat (58%) and g_s (68%), but increased WUE_int (31%) relative to the treatment of WW. There was a subtle but significant effect of the water restriction on the fluorescence parameters (p < 0.05, Table 3), while we did not observe an effect on the carbon and oxygen isotopic composition. Treatment WR decreased Fv’/Fm’ (14%) and increased 1 - Fv’/Fm’ (7%) relative to the WW treatment. None of the leaf-physiological parameters assessed in this study differed between the provenances (p > 0.05, Table 3). Additional mean values per trait at each combination of provenance and watering treatment are presented in Table S1 (Suplementary material).

Table 3. P-values from the analysis of variance and means (standard errors in parenthesis) per watering treatment (W) and provenance (P) on leaf-physiological traits on C. alba in the nursery experiment. Provenances were Cuesta La Dormida (CD), Antumapu (AN), Cantillana (CA) and Cayumanque (CY). Different letters within a column indicate significant differences according to Šidák´s test.

Source of Variation	Asat	gs	WUEint	Fv’/Fm’	1 - Fv’/Fm’	Δ^13	δ^18O
Watering (W)	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	0.5541	0.1783
Provenance (P)	0.0925	0.1231	0.3279	0.6931	0.6931	0.6762	0.0684
W×P	0.7519	0.8563	0.5548	0.1532	0.1532	0.0549	0.1713
Means	µmol m^−2 s^−1	mol m^−2 s^−1	µmol mol^−1			‰	‰
WW	6.9 a (0.25)	0.062 a (0.002)	118 a (5.4)	0.407 a	0.598 a (0.011)	20.6 a (0.25)	25.7 a (0.66)
WR	2.9 b (0.25)	0.02 b (0.002)	154 b (5.6)	0.348 b	0.639 b (0.011)	20.3 a (0.25)	27.2 a (0.65)
La Dormida	5.8 a (0.35)	0.051 a (0.004)	136 a (7.6)	0.379 a (0.012)	0.610 a (0.015)	20.5 a (0.27)	27.3 a (0.62)
Antumapu	4.5 a (0.37)	0.035 a (0.004)	143 a (7.9)	0.378 a (0.012)	0.607 a (0.017)	20.4 a (0.25)	25.4 a (0.61)
Cantillana	4.3 a (0.37)	0.036 a (0.004)	136 a (8.6)	0.366 a (0.012)	0.662 a (0.017)	20.6 a (0.26)	26.9 a (0.61)
Cayumanque	5.1 a (0.35)	0.043 a (0.004)	129 a (7.9)	0.387 a (0.012)	0.594 a (0.015)	20.2 a (0.25)	26.0 a (0.66)

Discussion

Effect of water restriction on growth increments, biomass components, and survival

In this study, we applied a moderated soil water restriction as a control treatment (0.27 m³ m⁻³), a soil moisture level found in irrigated plants in restoration projects and compared this against a more severe water restriction treatment (0.13 m³ m⁻³) (Figure 1). In our study, the short-term morphological response of C. alba to the water restriction was a decrease in stem growth and leaf biomass. However, this decrease in stem growth did not reflect differences in stem biomass. The reduction in leaf biomass (25%) due to the water restriction could be more attributed to leaf senescence than to a lower leaf construction. Leaf senescence has been reported to decrease plant transpiration and adjust the water balance (Munné-Bosch and Alegre 2004). These responses agree with those found in a similar experiment in L. caustica, another dominant species of the sclerophyllous forests of central Chile (Peña-Rojas et al. 2018), which coexist with C. alba.

The water restriction treatment did not affect the root, total biomass, and root-to-shoot ratio (Table 2). A higher carbon partitioning to roots than shoots (i.e. higher root-to-shoot ratio) is considered a long-term response to drought to reduce leaf transpiration and respiration (avoidance strategy) (Brunner et al. 2015). Therefore, despite the root-to-shoot ratio remaining unaffected between watering treatments, leaf senescence could be a short-term avoidance strategy of sclerophyllous species, such as C. alba, which needs further research. Thus, the experimental period was likely not enough to affect these complex traits. Similarly, the water restriction treatment did not affect survival, which was over 80% at the end of the experiment in this treatment. These results showed some disagreement with other studies with the same species and similar levels of water restriction (see Alvarez-Maldini et al. 2020; Espinoza et al. 2021), and seems to be related to the duration of water withholding than to the level of water restriction applied (close to the permanent wilting point). For instance, after three re-watering cycles of 15 days, Espinoza et al. (2021) found that water restriction strongly decreased root biomass and survival but with no effect on the stem and leaf biomass. Their results would contradict the fact that minor root biomass should be a long-term response to drought than the expected leaf biomass. Otherwise, our results agree with Donoso et al. (2015), who reported significantly less growth only in leaf biomass components after a 30-days cycle of water restriction. Nevertheless, both studies reported a significant recovery of the species after re-watering. Thus, C. alba might tolerate a short period of drought by osmotic adjustment (Donoso et al. 2015) and respond faster to irrigation due to its higher fine root turnover (Ovalle et al. 2015), which might explain the higher survival found in our study.

Effect of water restriction on leaf-physiological parameters

The photosynthetic rate and stomatal conductance found in this study and their response to water restrictions are comparable to those found in other sclerophyllous species, such as Quillaja saponaria and L. caustica (Donoso et al. 2011; Peña-Rojas et al. 2018). Water restriction induced the stomatal closure limiting the photosynthetic capacity, but also affected the maximum photochemical efficiency (i.e. Fv’/Fm’) as has been reported in other studies (Wang et al. 2018). Moreover, we observed a slight increase in the energy dissipated thermally (i.e. 1 - Fv’/Fm’) due to the water restriction (Table 3). This vulnerability of the PSII to water restriction is expected to be higher in shade-tolerant species such as C. alba than sun-tolerant species because they require lower light levels to saturate photosynthesis, which is also consistent with the trade-off principle between shade and drought tolerance (Kupers et al. 2019). Plants in our experiment were shaded at 80%. Thus, the species would likely be more susceptible to increased photoinhibition under water limitation and high irradiation (Valladares et al. 2005; Galmés et al. 2007). The WR treatment also increased WUEi relative to the WW treatment, which agreed with the study by Alvarez-Maldini et al. (2020) and Espinoza et al. (2021). However, it did not affect Δ¹³, which is a more integrated measure of intrinsic water use efficiency, and consequently is considered a more long-term response (Altieri et al. 2015). The impaired results on WUEi and Δ¹³ suggest that the carbon assimilated during the treatment period was invested more in the rapid turnover molecules of leaves than in their structure (Brendel 2001).

While the WR treatment had a lower net photosynthesis rate and stomatal conductance, it is important to take into account that measurements were made at the end of the experiment, when stress was particularly intense. For this reason, the instantaneous gas exchange measurements are not necessarily incompatible with the results obtained from the oxygen and carbon leaf stable isotopes. There were no differences in the integrated intrinsic water use efficiency estimated from Δ¹³C (Farquhar et al. 1982) in the remaining leaf tissue of the WR treatment compared to WW, possibly due to a coordinated decrease in both g_s and A_n over time, according to the dual isotopic model (Scheidegger et al. 2000). Moreover, the defoliation experienced by C. alba in WR may have compensated for the lower availability of water, allowing an adequate water supply to the remaining leaves during the experimental period. This would allow for maintaining similar gas exchange per unit leaf area between treatments or even a higher photosynthetic capacity per unit of stomatal conductance as was observed in Eucalyptus globulus (Pinkard et al. 2007; Eyles et al. 2009) experiencing a transient increase in the photosynthetic rate as a compensation mechanism against leaf shedding. This is coherent with the no differences in leaf δ¹⁸O between treatments because whenever the isotopic composition of irrigation water and atmospheric demand are the same between conditions, δ¹⁸O is inversely proportional to the time-integrated stomatal conductance (Farquhar et al. 2007; Hasselquist et al. 2010).

Differences in short-term responses to drought among provenances

In this study, we found no differential response of the C. alba provenances to the watering treatment in any of the traits as was expected (i.e. interaction effect) (Tables 2 and 3). Moreover, provenances did not exhibit differences in their growth and physiology, excepting they have some differences in carbon partitioning to the stem. Thus, there is no agreement in this species about a potential better response to drought from C. alba provenances from more xeric sites as was hypothesized. Hence, it would not validate the use of these types of seed sources in drought-prone restoration sites. This agrees with Broadhurst and Boshier (2014) in that local seed is not always better and claims for long-term and reciprocal transplant experiments. In our study, seed sources differed only in stem and total biomass. Seed sources Antumapu and La Dormida tended to have higher stem biomass than the other provenances. Nonetheless, these seed source differences were not associated with the effect of the provenance origin site, as the southern seed source Cayumanque did not differ from the northern provenances in most of the traits. Although the studies by Alvarez-Maldini et al. (2020) and Espinoza et al. (2021) tended to conclude that the northern seed sources acclimate better to drought, those results applied only to some specific seed sources, which did not necessarily correspond to the local ones. Local provenances should be better adapted to cope with climatic variations in these habitats, but according to our results, this ability might not be expressed as a short-term response.

As C. alba grows better under shaded environments and in more humid sites, our short-term experiment implies that the restoration of the species in semiarid environments must necessarily be complemented with watering. The lack of water during the first growing season will seriously compromise C. alba’s survival (Ovalle et al. 2015). Future research must consider monitoring the plant water stress through the leaf water potential, and a more intense assessment of morpho-physiological traits in a more extended period to better characterize traits that represent a better short- and long-term response to drought. Moreover, if possible, we recommend using a higher number of provenance per condition (i.e. xeric versus mesic origin) to better represent the distribution range of the species.

Conclusions

This study showed that the short-term morphological responses of C. alba to water restriction were a decrease in shoot growth and foliage biomass without altering other biomass components. Water restriction considerably decreased photosynthetic rate and stomatal conductance and increased the intrinsic water use efficiency. However, the experimental period did not affect carbon discrimination on leaves. All C. alba seed sources had a similar performance under water restriction. Thus, exploring provenance differences will require more extended periods and including more populations before prescribing local and non-local seed sources for restoration projects. Moreover, there is a need to characterize the variation within provenance, such as maternal effect.

Acknowledgements

We thank AngloAmerican Los Bronces operation for supporting this research and the project 045/2020 “Consideraciones Genéticas y Silvícolas para la Restauración de Bosques Hidrófilos de Quebrada en la Región Metropolitana” of the Fondo de Investigación del Bosque Nativo, Chile.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/13416979.2024.2346009