The effect of salinity on branch water relations and stem hydraulic vulnerability in two co-occurring mangrove species ARC DP180102969
Plant mortality due to extremes in drought and salinity is frequently attributed to hydraulic failure (Adams et al., 2017; McDowell et al., 2022). However, hydraulic risk may be managed by flexibly altering storage and release of water within the plant. For instance, when drought or salinity conditions increase, adjustment of water relations parameters such as water storage and hydraulic capacitance can reduce mangrove hydraulic vulnerability (Bryant et al., 2021; Coopman et al., 2021). The role of such water relations parameters in the maintenance of hydraulic function under varied environmental conditions remains poorly resolved, yet is fundamental to understanding species’ survival under increasing drought and salinity (Guadagno et al., 2017; McDowell et al., 2022). This dataset investigates how plasticity in branch water relations and hydraulic vulnerability maintain turgor and stem hydraulic function in two co-occurring mangrove species: Aegiceras corniculatum and Rhizophora stylosa growing at low and high salinity.
We hypothesised that 1) as salinity increases the proportionate contribution of foliage water storage and capacitance to whole-branch water relations will decrease, 2) stem capacitance will increase with the proportion of living tissue (inner bark, cortex, ray parenchyma) in the stem, 3) P12 and P50 will occur at more negative water potentials in plants growing at high salinity compared to plants growing at low salinity, and 4) hydraulic safety margins will be smaller in plants grown at high than low salinity due to a greater change in estuarine water potential than plasticity in P50 as salinity increases.
For full methods see publication: Holly A. A. Beckett, Callum Bryant, Teresa Neeman, Maurizio Mencuccini, and Marilyn C. Ball, (2023). Plasticity in branch water relations and stem hydraulic vulnerability enhances hydraulic safety in mangroves growing along a salinity gradient. Plant, Cell & Environment.
Abridged methods:
Species and site: Sun-exposed, terminal canopy branches less than 1 m in length of Aegiceras corniculatum and Rhizophora stylosa growing naturally along the banks of the Daintree River, Daintree National Park, Far North Queensland, Australia (16.1700° S, 145.4185° E) were collected mid-dry season, August 2019, at two salinities from three estuarine sites: Upper and Middle estuarine sites (low salinity), and Lower estuarine site (high salinity). The low salinity sites had a salinity of 0 parts per thousand (ppt) while the high salinity site had a salinity of 24 ppt measured from surface water samples using a hand-held refractometer (A.S.T. Co. Ltd., Japan). Branches were defined as comprising of foliage (the collection of leaves on the branch) and stem (woody canopy growth to which foliage is attached).
Branch water relations: One branch from each of five trees of each species were collected from high salinity, while one branch from each of 10 trees of A. corniculatum and one branch from each of 7 trees of R. stylosa were collected from low salinity. Branches were recut under 1% seawater perfusion solution for R. stylosa and 5% seawater perfusion solution for A. corniculatum and allowed to rehydrate overnight. Branch water release curves were constructed using a modified bench drying method (Gleason et al., 2014; Bryant et al., 2021) and analysed according to methods described by Bryant et al. (2021). Branch, foliage and stem capacitance and water storage were calculated from both the initial linear slope prior to the water potential at the leaf turgor loss point, describing decline in water content with decline in water potential during drying, and from the total water lost between first to last measurement.
Stem anatomy: From each individual measured for branch water relations, an additional branch was collected for paired stem anatomy. Transverse sections 100 µm thick in A. corniculatum and 200 µm thick in R. stylosa were cut using a sledge microtome G.S.L.1 (S.Lucchinetti, Schenkung Dapples, Zurich, Switzerland) from three stem developmental points. Developmental points were identified based on maturity of bark development, distance from shoot tip and stem diameter to capture variation in anatomy with stem development. Sections were observed under a ZEISS Axiostar Plus epifluorescence microscope (ZEISS, Germany). The Halide Mark II Pro Camera app on an iPhone 8 was used to capture RAW format images. One transverse section from each developmental point was analysed for radial stem tissue layer thickness and area, and vessel lumen area (measured on each vessel within a 0.25 mm² area of xylem tissue) using FIJI image analysis software (Schindelin et al., 2012).
Hydraulic vulnerability curves: The pneumatic method was used to estimate hydraulic vulnerability (Pereira et al., 2016; Zhang et al., 2018; Sergent et al., 2020; Jansen et al., 2022). One sun-exposed canopy branch was collected from each of five trees of A. corniculatum and four trees of R. stylosa from high salinity, and from each of six trees of A. corniculatum and five trees of R. stylosa from low salinity. Trees sampled were 2-4 m tall. Measurement of hydraulic vulnerability and analyses were conducted according to methods described by Bryant et al. (2021).
Statistical analyses: R statistical software package (R Core Team, 2023 version 4.3.1) was used for all statistical analysis. The lmer() function in the lmerTest package (Kuznetsova et al., 2017) was used to run all linear mixed models. To analyse water storage and capacitance at the branch level, a linear mixed model with salinity and species as fixed effects and estuarine site as random intercepts was used. A linear mixed model was then used to analyse contributions to branch water storage and capacitance made by foliage and stem organs, with salinity, species and organ as fixed effects and estuarine site and individual as random intercepts. To assess the significance of main effects and interactions of linear mixed models at the p<0.05 level, the anova() function was used. To address the effects of stem anatomical traits on stem capacitance, a linear model with species as a fixed effect was used. Slopes for species and species by stem anatomical trait were considered significant at the p<0.05 level. Diagnostic residual plots and the Shapiro-Wilk test were used to assess model fit. To improve model fit where necessary, dependent variables were log transformed. Pairwise comparisons, model estimated marginal means, trendlines and standard error (SE) were produced using the emmeans package (Lenth, 2020).
The fitplc package (Duursma & Choat, 2017) was used to determine hydraulic vulnerability curves and key parameters including the water potentials
Plant mortality due to extremes in drought and salinity is frequently attributed to hydraulic failure (Adams et al., 2017; McDowell et al., 2022). However, hydraulic risk may be managed by flexibly altering storage and release of water within the plant. For instance, when drought or salinity conditions increase, adjustment of water relations parameters such as water storage and hydraulic capacitance can reduce mangrove hydraulic vulnerability (Bryant et al., 2021; Coopman et al., 2021). The role of such water relations parameters in the maintenance of hydraulic function under varied environmental conditions remains poorly resolved, yet is fundamental to understanding species’ survival under increasing drought and salinity (Guadagno et al., 2017; McDowell et al., 2022). This dataset investigates how plasticity in branch water relations and hydraulic vulnerability maintain turgor and stem hydraulic function in two co-occurring mangrove species: Aegiceras corniculatum and Rhizophora stylosa growing at low and high salinity.
We hypothesised that 1) as salinity increases the proportionate contribution of foliage water storage and capacitance to whole-branch water relations will decrease, 2) stem capacitance will increase with the proportion of living tissue (inner bark, cortex, ray parenchyma) in the stem, 3) P12 and P50 will occur at more negative water potentials in plants growing at high salinity compared to plants growing at low salinity, and 4) hydraulic safety margins will be smaller in plants grown at high than low salinity due to a greater change in estuarine water potential than plasticity in P50 as salinity increases.
For full methods see publication: Holly A. A. Beckett, Callum Bryant, Teresa Neeman, Maurizio Mencuccini, and Marilyn C. Ball, (2023). Plasticity in branch water relations and stem hydraulic vulnerability enhances hydraulic safety in mangroves growing along a salinity gradient. Plant, Cell & Environment.
Abridged methods:
Species and site: Sun-exposed, terminal canopy branches less than 1 m in length of Aegiceras corniculatum and Rhizophora stylosa growing naturally along the banks of the Daintree River, Daintree National Park, Far North Queensland, Australia (16.1700° S, 145.4185° E) were collected mid-dry season, August 2019, at two salinities from three estuarine sites: Upper and Middle estuarine sites (low salinity), and Lower estuarine site (high salinity). The low salinity sites had a salinity of 0 parts per thousand (ppt) while the high salinity site had a salinity of 24 ppt measured from surface water samples using a hand-held refractometer (A.S.T. Co. Ltd., Japan). Branches were defined as comprising of foliage (the collection of leaves on the branch) and stem (woody canopy growth to which foliage is attached).
Branch water relations: One branch from each of five trees of each species were collected from high salinity, while one branch from each of 10 trees of A. corniculatum and one branch from each of 7 trees of R. stylosa were collected from low salinity. Branches were recut under 1% seawater perfusion solution for R. stylosa and 5% seawater perfusion solution for A. corniculatum and allowed to rehydrate overnight. Branch water release curves were constructed using a modified bench drying method (Gleason et al., 2014; Bryant et al., 2021) and analysed according to methods described by Bryant et al. (2021). Branch, foliage and stem capacitance and water storage were calculated from both the initial linear slope prior to the water potential at the leaf turgor loss point, describing decline in water content with decline in water potential during drying, and from the total water lost between first to last measurement.
Stem anatomy: From each individual measured for branch water relations, an additional branch was collected for paired stem anatomy. Transverse sections 100 µm thick in A. corniculatum and 200 µm thick in R. stylosa were cut using a sledge microtome G.S.L.1 (S.Lucchinetti, Schenkung Dapples, Zurich, Switzerland) from three stem developmental points. Developmental points were identified based on maturity of bark development, distance from shoot tip and stem diameter to capture variation in anatomy with stem development. Sections were observed under a ZEISS Axiostar Plus epifluorescence microscope (ZEISS, Germany). The Halide Mark II Pro Camera app on an iPhone 8 was used to capture RAW format images. One transverse section from each developmental point was analysed for radial stem tissue layer thickness and area, and vessel lumen area (measured on each vessel within a 0.25 mm² area of xylem tissue) using FIJI image analysis software (Schindelin et al., 2012).
Hydraulic vulnerability curves: The pneumatic method was used to estimate hydraulic vulnerability (Pereira et al., 2016; Zhang et al., 2018; Sergent et al., 2020; Jansen et al., 2022). One sun-exposed canopy branch was collected from each of five trees of A. corniculatum and four trees of R. stylosa from high salinity, and from each of six trees of A. corniculatum and five trees of R. stylosa from low salinity. Trees sampled were 2-4 m tall. Measurement of hydraulic vulnerability and analyses were conducted according to methods described by Bryant et al. (2021).
Statistical analyses: R statistical software package (R Core Team, 2023 version 4.3.1) was used for all statistical analysis. The lmer() function in the lmerTest package (Kuznetsova et al., 2017) was used to run all linear mixed models. To analyse water storage and capacitance at the branch level, a linear mixed model with salinity and species as fixed effects and estuarine site as random intercepts was used. A linear mixed model was then used to analyse contributions to branch water storage and capacitance made by foliage and stem organs, with salinity, species and organ as fixed effects and estuarine site and individual as random intercepts. To assess the significance of main effects and interactions of linear mixed models at the p<0.05 level, the anova() function was used. To address the effects of stem anatomical traits on stem capacitance, a linear model with species as a fixed effect was used. Slopes for species and species by stem anatomical trait were considered significant at the p<0.05 level. Diagnostic residual plots and the Shapiro-Wilk test were used to assess model fit. To improve model fit where necessary, dependent variables were log transformed. Pairwise comparisons, model estimated marginal means, trendlines and standard error (SE) were produced using the emmeans package (Lenth, 2020).
The fitplc package (Duursma & Choat, 2017) was used to determine hydraulic vulnerability curves and key parameters including the water potentials
Published to:
- Australian National University
- Australian National Data Service
- hasAssociationWith:
Prof. Marilyn Ball [anudc:5998] - hasAssociationWith:
Mr. Callum Bryant [anudc:6001] - hasAssociationWith:
Prof. Maurizio Mencuccini [anudc:6098] - hasAssociationWith:
Dr Teresa Neeman [anudc:6113] - isOwnedBy:
Dr Holly Beckett [anudc:6101] - hasPrincipalInvestigator:
Dr Holly Beckett [anudc:6101]