Impact of Environmental Conditions on Wood Anatomical Traits of Green Alder (Alnus alnobetula) at the Alpine Treeline
1. Introduction
Mountain ecosystems are particularly vulnerable and are expected to undergo considerable transformations due to global change [1,2,3]. However, within the European Alps, land abandonment and decreasing grazing pressure may have a greater impact on the treeline ecotone in the coming decades than rising temperatures [4,5]. There are indications that shrubs may benefit more from land use changes than trees and take over abandoned pastures [6,7,8]. During the last decades, in particular, green alder (Alnus alnobetula (Ehrh.) K. Koch = Alnus viridis (Chaix) DC.) has spread rapidly across the Alps [7,9,10,11]. A. alnobetula is a cold-resistant and moderately shade-tolerant species of the subalpine and subarctic zones of the northern hemispheres, growing in mountains, tundra, and river valleys [12]. Due to clonal growth, high seed production, and symbiosis with N2-fixing actinobacteria and ectomycorrhizal fungi, A. alnobetula is spreading rapidly after land abandonment. As green alder forms closed thickets, with canopy heights up to 4 m, it drives former N-poor grassland into nitrogen-saturated-species-poor shrubland and suppresses tree establishment within the treeline ecotone [7].
Although the occurrence of green alder was restricted to north-facing slopes and avalanche gullies, with high water availability [10,13,14], this species is now expanding into moderately steep well-drained subalpine grasslands, as well as sun-exposed sites with shallow soils and impaired water availability [9,15]. Green alder is known to be an anisohydric species, keeping its stomata open even under high vapor pressure deficits [16,17,18], a strategy that can be risky under drought conditions [19]. Nevertheless, the former restriction of A. alnobetula to moist habitats may be due to former land use patterns, while A. alnobetula is able to adapt to drier condition [15].
A. alnobetula is a diffuse-porous species, keeping vessel diameters constant throughout the annual ring [20]. However, drought-induced changes in vessel development in diffuse-porous species have been reported before [21,22,23]. Size, number, and distribution of vessels are closely linked to tree hydraulic conductivity and drought resistance [24,25], but still little is known about intraspecific adaptations in hydraulic properties. The intraspecific studies available to date have shown the effects of water availability on conduit traits [26,27] where decreasing water availability resulted in narrower conduits and higher conduit density in different species [28,29,30,31]. However, the adaptation of wood anatomical traits might vary between species and there may be different adaptation strategies for ring-porous and diffuse-porous angiosperms [32,33].
Furthermore, there is increasing awareness that aside from conductive cells, parenchyma tissues in the secondary xylem also play a critical role in the water relations of woody plants [34,35,36,37]. It is well known that parenchyma in the secondary xylem stores carbohydrates, which can subsequently be used for growth, establishment of freeze tolerance, and protection against or recovery after infestations [38,39,40,41]. Increasing drought stress in the course of global warming emphasizes the role of xylem parenchyma in maintenance of the water transport and embolism repair [33,42,43], as well as water storage and circulation between xylem and phloem [34,36,40,44,45]. Axial parenchyma is reported to be more plastic than ray parenchyma [39,40,46] and its fractions vary with environmental conditions at the intra- and interspecific level [47,48].
Analysis of functional adaptations in the wood anatomy of A. alnobetula provides insights into the ecological limits of this pioneer species. We, therefore, evaluated vessel and axial parenchyma properties and distribution in A. alnobetula at a south-facing windward and a north-facing leeward site at the alpine treeline, differing in soil water availability and evaporative demand. We tested the hypotheses that reduced soil water availability and enhanced evaporative demand on the south-facing site (i) reduces vessel diameter, (ii) increases vessel density, (iii) changes vessel distribution within the annual ring leading to a semi-ring-porous arrangement of vessels, and (iv) causes an increase in axial parenchyma cells.
2. Materials and Methods
2.1. Study Area and Sample Plots
The study was performed within the treeline ecotone of Mt. Patscherkofel in the Central European Alps, Tyrol, Austria, (47°12′ N, 11°27′ E) where A. alnobetula stands primarily spread out in avalanche gullies on leeward north-facing slopes between 1950 and 2150 m asl. Even so, during the last decades, A. alnobetula stands spread out to windward sites on south- to southeast-facing slopes (Oberhuber et al., 2022 [9]). Within the study area, the bedrock is dominated by gneisses and schists [49], and the soils are classified as haplic podzols [50,51].
The mean annual temperature at the meteorological station on top of Mt. Patscherkofel was 0.8 ± 0.7 °C during the period from 1991 to 2020, with February being the coldest month (−6.6 °C) and July the warmest (8.9 °C). Despite the high altitude, air temperature can frequently reach maxima around 20 °C during summer. Mean annual precipitation was 889 ± 128 mm with precipitation maxima occurring in summer (long-term mean: 371 ± 74 mm, June to August) and winters being the driest season (132 ± 60 mm, December to February). The study area is also characterized by the frequent occurrence of strong southerly Föhn-type winds [52], which strongly influences snow depth and snow distribution—[53] and hence, the duration period of the permanent snow cover. At south-facing slopes, snow depth is generally
Within the study area, we selected two study plots at the upper edge of the treeline ecotone (Figure 1, Table 1): a north-exposed leeward (hereafter N-site) and a south-exposed windward site (hereafter S-site). Although the two plots were only 150 m apart in linear distance, they differed considerably with respect to slope exposure and soil depth (Table 1), soil water availability, and evaporative demand (Table 2, Figure 2). The canopy height of A. alnobetula was higher on the N-site, although shrubs were older on the S-site (Table 1).
2.2. Environmental Data
Environmental conditions were recorded at both study sites from June through September 2022. At each study plot, air temperature and relative air humidity (CS215 Temperature and Relative Humidity Sensor) and solar radiation (SP1110 Pyranometer Sensor) (all sensors, Campbell Scientific, Shepshed, UK) were monitored 2 m above ground, while soil temperature (T 107 temperature probe, Campbell Scientific) and moisture in 10 cm soil depth were monitored using four ThetaProbes ML2 (Delta-T Devices Ltd., Burwell, UK). Precipitation (ARG100 Rain Gauge) was recorded at the S-site. All the environmental data were recorded with a Campbell CR1000 data logger (Campbell Scientific, Shepshed, UK) programmed to record 30-minute averages of measurements taken every minute.
Daily mean air temperature averaged throughout the growing season did not differ significantly between the two study plots (Figure 2, Table 2). During the growing season, daily mean solar radiation and soil temperature in 10 cm soil depth, respectively, were significantly higher on the windward S-site than on the leeward N-site (Table 2). Conversely, soil water content in 10 cm soil depth was significantly lower on the S-site than on the N-site (Table 2), indicating that drier conditions prevail on the S-site compared with the N-site. This assumption is supported by significantly lower annual radial increments and a lower canopy height at the S-site compared with the N-site (Table 1).
2.3. Sample Collection, Preparation, and Wood Anatomical Analysis
Stem discs were sampled from 3 different stocks on the N-site (mean diameter of stem discs: 2.13 ± 0.21 cm) and 4 stocks and the S-site (mean diameter of stem discs: 2.08 ± 0.17 cm). Discs were taken 150 cm from the shoot tip as there is evidence that xylem anatomy in trees and shrubs is changing from tip to base [54,55,56,57], with the narrowest conduits found at the tip of the shoots. Stem discs were air-dried, and subsequently, rectangular pieces were cut from the discs to prepare the probes for microtome sectioning. After the samples were soaked in glycerin for one day to soften the wood, microsections of 10 μm thickness were produced using a sledge microtome (WSL-core microtome, WSL, Birmensdorf, Switzerland). The microsections were stained with safranin and astrablue to differentiate between lignified (red) and unlignified (blue) cells. The stained cross sections were observed under the microscope (Olympus Typ BX50, Olympus Corporation, Tokyo, Japan), and images were taken at 10× magnification with an HD-microscope camera (ProgRes GryphaxR, Jenoptik, Jena, Germany).
Wood anatomic variables were evaluated from the microscopic images using the open-source image analysis program Fiji (ImageJ2). Vessel number, vessel area, and diameter were evaluated for 10 annual rings (2011–2020). In total wood anatomy of 70 annual rings from 7 samples was determined. In total, 5677 and 4402 vessels were measured at the N- and the S-site, respectively. From the collected data, mean vessel area (MVA), vessel density (VD, number of vessels mm−2), and theoretic conductive area (TCA, i.e., percentage of vessel area to total area) were calculated for each sample. Kernel density estimations were used to compare vessel area distributions between sites. All individual vessel measurements were included in this analysis. To compare vessel area distributions related to wood surface area, all measured data were assigned to size classes and the number of vessels per size classes and xylem area was calculated.
To quantify phenological differences between vessels formed in different phases during the growing season, we determined the position within the annual ring for each vessel. Using this positioning data, we analyzed differences in the number and sizes of vessels in the first and second 50% of each annual ring (i.e., early and late growing season). MVA, VD, and TCA were calculated to gain insight into the plasticity of vessel formation throughout the growing period. In addition, vessel area distribution was analyzed for the first and the second 50% of the growth rings. Because some of the inner growth rings were curved, which made reliable positioning difficult, only the outermost 6-year rings (2015–2020) were used for the processing of the position data.
In a second series of measurements, we evaluated the number, area, and position of the diffuse parenchyma within the annual ring. Using these data, we calculated the percentage of the total area occupied by the axial parenchyma (PA). We also used the positioning data to calculate the percentage of PA for the first and the second 50% of the annual ring to get insight into the timing of the formation of diffuse parenchyma in the course of the growing period.
2.4. Data Analysis
Shapiro–Wilk tests and Q-Q plots were used to check for normal distribution of data. Most anatomical and climate datasets were not normally distributed, so we used nonparametric statistical tests for analyses. Because temperature data were consistently normally distributed, differences in air and soil temperature at the N-site and S-site were tested for significance using paired t-tests. For analyzing differences in MVA, VD, and TCA at the two sites, Mann–Whitney U tests were applied. For comparison of vessel and parenchyma properties in the first and second half of the annual rings and for analysis of soil moisture and solar radiation data, we used paired samples Wilcoxon tests. All statistical analyses were performed using SPSS Statistics, Version 26 (IBM, New York, NY, USA).
4. Discussion
Analysis of site-specific functional adaptations of A. alnobetula enables us to determine the effects of global warming on the limits of the expansion of this pioneer species. Intraspecific studies provide insights into the impact of environmental factors on species-specific anatomical traits, disentangling genotypic and environmental effects contributing to xylem development is a difficult task. There is evidence that provenances can have more impact on xylem anatomy than environmental conditions [58,59,60]. Therefore, anatomical variations in plants growing at faraway sites or along wide-ranging environmental gradients may result from genetic variations. However, in the presented study, the examined plants were growing in close proximity, and A. alnobetula has expanded from the north-facing to the south-facing site only during the last decades [9], ruling out differences in provenience. Even so, the two sites differed significantly with respect to environmental conditions. The S-site experienced significantly higher solar radiation and was more exposed to the prevailing southerly winds while soil water availability was significantly lower than at the N-site (Figure 2, Table 2; c.f. [53]). Moreover, in anisohydric A. alnobetula, stomata remain open even under high evaporative demand [16,17,18], which, in combination with frequently occurring strong winds reducing boundary layer resistance (e.g., [61,62]), may lead to drought stress during periods of reduced soil water availability at the S-site. This is confirmed by a lower canopy height (Table 1), a significantly lower radial growth (Table 2; [53]), and premature leaf wilting, occurring at the S-site in mid-August after a dry period in summer 2022. At the N-site, by contrast, leaf wilting did not occur until early October.
Xylem cell division and differentiation are driven by environmental and internal factors [63]. As a key factor, temperature controls cambial cell division and plays a major role in cell wall lignification [64,65]. Conversely, as a turgor-driven cell enlargement is mainly controlled by cell water status [66,67] and sugar availability acting as an osmotic agent [68]. Cell enlargement and cambial cell division are seriously hampered under drought, often leading to a decline in radial growth [69] and vessel diameter [70,71,72]. Air temperature did not significantly differ between the two sampling sites (Table 2), and due to a high surface roughness, A. alnobetula experienced strong aerodynamic coupling to the free atmosphere [16]. On the other hand, significantly higher solar radiation at the S-site, especially during cloudless periods, enhanced solar heating of the ground. Nevertheless, as samples were taken in the upper branch section, temperature effects are less likely to explain observed differences in anatomical traits between the N- and the S-site.
Fast-growing species like Populus sp. show high phenotypic plasticity and are used as model species to examine environmental impacts on xylogenesis [73]. In poplar, drought often initiates a reduction in the vessel area [23,29,74,75]. In our study, however, MVA and HD did not differ significantly between the S- and N-site. However, tracheid enlargement might be based on turgor control in gymnosperms [66], while in more complex angiosperms, cell differentiation is probably actively regulated by endogen mechanisms rather than being a result of passive processes [76]. Anatomical traits are often under strong genetic control [58,73] suggesting species-specific adaptations to environmental impacts. In beech, for example, like in our study, no change in MVA under drought was detected [22,77,78]. Despite significant differences in environmental conditions, plant height, and annual ring width, Kernel density estimations of vessel areas (Figure 4a) showed no differences between the two sampling sites, confirming our assumption that vessel size is a static trait in green alder. Nevertheless, the control of cell formation and the patterns of cell distribution and size are poorly understood and the role of auxin and other morphogens are still under debate [24,79,80].
The number of vessels per size class and wood surface area (Figure 4b) showed a lower absolute number of small and midsize vessels (up to 1200 µm2) at the dry S-site, which is in accordance with the significantly lower VD and TCA at the S-site. On the contrary, an increase in VD and TCA under drought conditions has been reported in several studies [21,58,77]. However, Arnic et al. [21] and Giagli et al. [22] pointed out that drier summer conditions resulted in narrower annual rings and, consequently, higher VD and TCA. Environmental and provenance effects on anatomical traits often result from stem height and ring width [58]. In beech, VD was positively related to the above-ground biomass and increased with tree height [77]. Plant height and crown size might often explain anatomical properties like VD and TCA better than climatic conditions [81], and leaf area has been found to be correlated with stem hydraulic conductivity [74]. In our study, the green alder stand at the N-site reached greater height, had a denser crown and, thus, higher leaf area compared with the S-site. Considering this, higher stem hydraulic conductivity was needed at the N-site to adjust the balance between the water supply and transpiring surface, which explains higher VD and TCA.
A. alnobetula is classified as a diffuse-porous species, with constant vessel diameters throughout the annual ring [20]. Nevertheless, anatomical parameters and density curves in our study estimate a near semi-ring-porous vessel distribution at both sampling sites. Environmentally induced changes in vessel distribution in diffuse-porous species have been reported—e.g., a near semi-ring-porous distribution of vessels in dry years and continuous decrease in vessel size under dry conditions have been reported for diffuse-porous beech [21,22,23]. As in diffuse-porous species, wide and narrow vessels can be formed during the entire growing period, and short-term adaptations to changing environmental conditions are possible in these species [82,83]. The bimodal shape of density curves over several years might be another reference for frequent semi-ring-porous vessel distribution at our treeline site [84]. Schreiber et al. [29] linked the bimodal distribution of vessel diameters in trembling aspen, which was found at a boreal site, to rough environmental conditions and a short growing period. Therefore, the harsh conditions at the altitudinal limit of A. alnobetula and not water deficiency might be the reason for hampered vessel formation in the second half of the growing season.
In addition to vascular tissue, axial parenchyma seems to play a key role in hydraulic optimization [33,44]. Xylem parenchyma is known to modulate xylem flow and hydraulic resistance through osmotic exudation into the xylem [34,39] and is involved in in embolism repair [34,85]. Moreover, water stored in xylem parenchyma cells [86,87] buffers a decline in water potential to sustain water transport under drought stress [34,88]. Aritsara et al. [44] found that species with more axial parenchyma were close to their hydraulic limits and [38] hypothesized that axial parenchyma fractions potentially keep vessels hydrated during drought periods. We, therefore, assume that the significantly higher amount of axial parenchyma at the S-site might be an adaptation to drought conditions.
However, there are still gaps in knowledge, especially when it comes to diffuse axial parenchyma in temperate species as most of the available studies concentrate on paratracheal parenchyma and species from subtropical and tropical regions (e.g., [38,44,89,90,91]). An accumulation of diffuse apotracheal parenchyma in latewood has been described for several conifer species [92] and European beech and pedunculate oak [20,93]. Nevertheless, in our study, the accumulation of diffuse parenchyma is very pronounced and not restricted to latewood. Therefore, we suggest that the higher fraction of parenchyma cells in the second half of the growing season, when vessel production is declining, indicates an enhanced investment in carbohydrate storage triggered by harsh environmental conditions at the upper distributional limit of the species.
