Variance in plant functional traits in Eucalyptus baxteri and E. obliqua – a pilot study to assess utility for inferring climate vulnerability

IntroductionDieback is affecting forests of the stringybark eucalypt species, Eucalyptus obliqua and E. baxteri, in the Mount Lofty Ranges region of South Australia. This pilot project examined variation in plant functional traits across eight sites spanning a gradient of mean annual rainfall of approximately 700–1100 mm. The aim was to test the utility of functional traits for inferring relative climate vulnerability and to inform the level of replication needed in sampling within and between individuals and sites to capture variation.<br>MethodsWe measured key plant functional traits (relating to leaf temperature balance, water use efficiency, maximum photosynthetic rate and climatic tolerances) in stringybarks sampled across climatic gradients. Traits included were:<br>1. Stem specific density (SSD; negatively correlated with lumen fraction and linked to hydraulic capacity, tolerance to low xylem water potential, and resistance to cavitation; Preston et al. 2006)2. Leaf area (linked to leaf temperature and therefore photosynthetic rates and need for evaporative cooling; Wright et al. 2017)3. Leaf length:width ratio (linked to boundary layer resistance and therefore rates of convective and evaporative heat loss; Pérez-Harguindeguy et al. 2016)4. Specific Leaf Area (SLA; fresh leaf area divided by dry mass; linked to resource investment and maximum photosynthetic rate; Lowry and Smith 2018).<br>We sampled three–five individual trees per site and collected 1x stem samples and 5x leaf samples per individual for analysis. Stem samples were taken from young wood sections approximately 1.5 cm diameter (Pérez-Harguindeguy et al. 2016), selected straight sections with no branching where possible. Young but fully mature sun leaves with minimal insect damage were selected as leaf samples. <br>To calculate SSD, we calculated fresh volume as the cross-sectional area of the stem segment multiplied by its length and also, for comparison, we measured the volume by placing each sample into a measuring cylinder full of water. The displacement of water when the sample is added (final volume- initial volume) gave the volume of the sample. Samples were then air dried and weighed on a 0.1 mg balance after oven drying at 100°C. SSD is equal to oven dry mass divided by fresh volume. <br>To calculate leaf area, length:width ratio and SLA, we pressed and dried leaf samples and scanned them with a scale bar. ImageJ was used to measure one-sided leaf area, length and width. A recent pilot analysis in a set of eucalypt species showed that that amount of eucalypt leaf shrinkage is small (~10%) and consistent across samples (Morgan et al. 2021). Therefore, the use of dry leaf area for SLA is valid as it is not biased by shrinkage, and can be converted to fresh leaf area with minimal error. Leaves were subsequently oven dried at 70°C for 72 hours before being weighed on a 0.1 mg balance. SLA was calculated as dry leaf area divided by oven dry mass. <br>ResultsSignificant variation was observed across all four traits measured (leaf area, leaf length:width ratio, SLA and SSD). Variation among species and sites across a rainfall gradient would be useful to explore across a wider sampling of sites and individuals to infer drought vulnerability. Variance calculations based on sums of squares show important variation split among species, sites, and individual trees, with high within-individual variance in leaf traits suggesting that sampling of multiple individuals and leaves is required to compare differences at site and species level.<br>SLA and L:W ratio were higher in E. obliqua than E. baxteri, while SSD was lower. While site-to-site variation was high and somewhat site-specific, there was an overall increase in leaf area and SLA along the rainfall gradient, whereas L:W ratio and SSD did not show any overall pattern with rainfall across these sites. Site-level characteristics other than macro-rainfall are likely to explain some of the variation, for example, the relatively low SLA values for both species at the Bradwood site, which is high rainfall (&gt;1 m MAP) but has soil consisting of shallow sand over rock, thus limited in both nutrients and dry season moisture availability, both of which may limit leaf size (Wright et al. 2017). It is likely that site-specific influences would be less prominent in assessing trends across a wider range of sites. SSD shows promising variation but needs to be measured for more individuals and sites to better assess climatic patterns.<br><br>ReferencesMorgan, R., Martín‐Forés, I., Leitch, E. &amp; Guerin, G. (2021) Assessment of protocols and variance for specific leaf area (SLA) in 10 Eucalyptus species to inform functional trait sampling strategies for TERN Ausplots. The University of Adelaide. Dataset. https://doi.org/10.25909/14197298 <br>Pérez-Harguindeguy, N., Díaz, S., Garnier, E., Lavorel, S., Poorter, H., Jaureguiberry, P., et al. (2016) New handbook for standardised measurement of plant functional traits worldwide. Australian Journal of Botany 61, 167-234.<br>Preston, K.A., Cornwell, W. K. &amp; DeNoyer, J.L. (2006) Wood density and vessel traits as distinct correlates of ecological strategy in 51 California coast range angiosperms. New Phytologist 170, 807-818. <br>Wright, I.J., Dong, N., Maire, V., Prentice, I. C., Westoby, M., Díaz, S., et al. (2017) Global climatic drivers of leaf size. Science 357, 917-921. <br>Lowry, C.J. &amp; Smith, R.G. (2018) ‘Chapter 5 - Weed Control Through Crop Plant Manipulations, in Non-Chemical Weed Control’, Elsevier Inc, pp. 73–96.<br>

引言：南澳大利亚洛夫蒂山山脉区域的纤维桉（stringybark eucalypt）物种——斜叶桉（Eucalyptus obliqua）与巴克斯桉（Eucalyptus baxteri）林分正受枯梢病影响。本试点项目沿年平均降雨量约700–1100 mm的梯度，在8个样地中开展了植物功能性状变异研究。研究旨在验证功能性状用于推断相对气候脆弱性的实用性，并明确在个体内、个体间及样地间采样时所需的重复水平，以充分捕捉性状变异。

方法：本研究针对沿气候梯度采样的纤维桉，测定了与叶片温度平衡、水分利用效率、最大光合速率及气候耐受性相关的关键植物功能性状，所测定性状包括：
1. 茎比密度（Stem Specific Density, SSD）：与管腔分数呈负相关，与水力容量、低木质部水势耐受性及抗空化能力相关（Preston等，2006）；
2. 叶面积：与叶片温度相关，进而影响光合速率与蒸发冷却需求（Wright等，2017）；
3. 叶长宽比：与边界层阻力相关，进而影响对流与蒸发散热速率（Pérez-Harguindeguy等，2016）；
4. 比叶面积（Specific Leaf Area, SLA）：为鲜叶面积除以干质量，与资源投资及最大光合速率相关（Lowry与Smith，2018）。

每个样地采样3–5株个体树木，每株个体采集1份茎样本与5份叶样本用于分析。茎样本取自直径约1.5 cm的幼木质段（Pérez-Harguindeguy等，2016），优先选取无分枝的直段。叶样本则选取年轻但完全成熟的向阳叶片，且虫害损伤较轻。

为计算SSD，本研究通过茎段横截面积乘以其长度计算鲜体积；同时为进行对照，将每份样本放入装满水的量筒中，通过排水体积（终体积-初体积）得到样本体积。随后将样本风干，并在100℃烘箱中烘干后，使用精度为0.1 mg的天平称量干质量。SSD计算公式为烘干干质量除以鲜体积。

为计算叶面积、叶长宽比与SLA，本研究将叶样本压制干燥后，结合比例尺进行扫描。使用ImageJ软件测定单侧叶面积、叶长与叶宽。此前针对一组桉树物种的试点分析显示，桉树叶片的收缩率较低（约10%）且样本间一致性较好（Morgan等，2021）。因此，使用干叶面积计算SLA是合理的，其不会因收缩产生偏差，且可通过极小误差转换为鲜叶面积。随后将叶片在70℃烘箱中烘干72小时，使用精度为0.1 mg的天平称量干质量。SLA计算公式为干叶面积除以烘干干质量。

结果：本研究测定的4项性状（叶面积、叶长宽比、SLA与SSD）均存在显著变异。沿降雨梯度的物种间与样地间变异，可通过扩大样地与个体采样范围进行探索，以推断干旱脆弱性。基于平方和的方差分析显示，物种、样地与个体树木间均存在重要变异；叶片性状的个体内方差较高，这表明在比较样地与物种水平的差异时，需要对多个个体与叶片进行采样。

斜叶桉的SLA与叶长宽比高于巴克斯桉，而SSD则更低。尽管样地间变异程度较高且具有一定样地特异性，但叶面积与SLA整体沿降雨梯度呈现上升趋势，而叶长宽比与SSD未呈现随降雨变化的整体规律。除宏观降雨外，样地的其他特征或许可以解释部分变异，例如Bradwood样地的两个物种的SLA值均相对较低：该样地降雨量较高（年平均降雨量＞1000 mm），但土壤为浅覆沙岩层，养分与旱季水分有效性均受限，这两者均可能限制叶面积（Wright等，2017）。在更大范围的样地中评估趋势时，样地特异性影响可能会减弱。SSD显示出具有潜力的变异特征，但需要对更多个体与样地进行测定，以更好地评估其气候模式。

参考文献
Morgan, R., Martín‐Forés, I., Leitch, E. & Guerin, G. (2021) 10种桉树比叶面积（SLA）的测定方案与方差评估：为TERN Ausplots的功能性状采样策略提供参考. 阿德莱德大学. 数据集. https://doi.org/10.25909/14197298
Pérez-Harguindeguy, N., Díaz, S., Garnier, E., Lavorel, S., Poorter, H., Jaureguiberry, P., 等. (2016) 全球植物功能性状标准化测定新手册. 澳大利亚植物学杂志 61, 167-234.
Preston, K.A., Cornwell, W. K. & DeNoyer, J.L. (2006) 51种加州海岸被子植物的木材密度与导管性状：作为生态策略的独立关联因子. 新植物学家 170, 807-818.
Wright, I.J., Dong, N., Maire, V., Prentice, I. C., Westoby, M., Díaz, S., 等. (2017) 叶面积的全球气候驱动因子. 科学 357, 917-921.
Lowry, C.J. & Smith, R.G. (2018) 第5章 - 通过作物植株调控进行杂草防控，载于《非化学杂草防控》, 爱思唯尔公司, 第73–96页.