Author: Richardson, Sarah J; Peltzer, Duane A; Allen, Robert B; McGlone, Matt S
Date published: September 1, 2010
A long leaf lifespan is considered advantageous in nutrient-poor environments because leaf lifespan determines the residence time of resources retained in leaves, such as nitrogen (N) and phosphorus (P) (e.g. Monk 1966; Chabot & Hicks 1980; Eckstein et al. 1999; Aerts & Chapin 2000). The mean residence time (MRT) of nutrients in leaves is a critical aspect of whole-plant function; it determines individual plant performance by controlling rates of leaf photosynthesis and respiration, and also underpins the impact of a plant on ecosystem processes through litter quality and rates of litter decomposition. Leaf lifespan should increase as soil nutrient availability declines, and support for this hypothesis comes from both among- and withinspecies responses to soil nutrient availability. Comparisons among species have commonly demonstrated that plant communities on infertile soils typically support more evergreen species, or species with longer leaf lifespans, than communities on fertile soils (e.g. Monk 1966; Escudero et al. 1992). Further evidence has been derived from consistent relationships among species between leaf lifespan and leaf nutrient concentrations (e.g. Reich et al. 1997) that can be used to argue that declining leaf nutrient concentrations along soil fertility gradients should be accompanied by increasing investment in leaf structural carbohydrates and leaf lifespan. These predictions are based on the idea that the construction costs of leaves is higher in low-fertility environments and thus plants should protect their investment in leaves as nutrient availability declines (e.g. Lambers & Poorter 1992; Aerts & Chapin 2000; Grime 2001).
Evidence for within-species responses of leaf lifespan to soil fertility comes from natural fertility gradients, fertilisation experiments, and combinations of the two. Cordell et al. (2001) fertilised two sites with contrasting soil fertility along the Hawaiian soil chronosequence, and measured the response of leaf lifespan in the dominant tree species Metrosideros polymorpha. Fertiliser addition decreased leaf lifespan at the younger, fertile site as predicted, but leaf lifespan was unresponsive to fertiliser addition at the older, more infertile site. Furthermore, leaf lifespan of unfertilised trees was longest at the younger, fertile site counter to predictions that leaf lifespan should be shortest on the most fertile sites. In this paper, we examine within-species variation in leaf lifespan along a longterm soil chronosequence that represents a strong fertility gradient at Franz Josef, New Zealand, to test whether leaf lifespan increases within species in response to declining soil fertility. The Franz Josef chronosequence is an outstanding natural fertility gradient; many common New Zealand woody species are widespread across this gradient, providing the opportunity to examine how key plant traits respond within species to changes in fertility (e.g. Whitehead et al. 2005). We determine how leaf lifespan varies within six common woody species along this gradient, combine those data with previously published estimates of nutrient resorption (Richardson et al. 2004, 2005) to estimate mean residence time (MRT) of N and P, and conclude by positioning our leaf lifespan data in global leaf trait space (GLOPNET; Wright et al. 2004) and discussing the range of leaf lifespans sampled by our six study species.
Materials and methods
The Franz Josef soil chronosequence is a series of postglacial surfaces in southern New Zealand (43° S, 170° E) that range widely in fertility (Stevens 1968). Richardson et al. (2004) described soil chemistry on nine surfaces (sites) and in this study data were gathered from eight sites that support tall forest, ranging in age from 60 to c. 120 000 years. Available soil P (ratio of organic P to total C) declines along the chronosequence from 5 mg kg^sup -1^ (60 years) to 1 mg kg^sup -1^ (120 000 years; Spearman rank correlation between available soil P and soil age r = -0.92, P < 0.005). Available soil N (aerobic mineralisable N) declines from 110 mg kg^sup -1^ (60 years) to 15 mg kg^sup -1^ (120 000 years; Spearman rank correlation between available soil N and soil age r = -0.89, P < 0.005). Mean annual temperature (1971-2000) is 10.0° C (Hessell 1982) and mean annual precipitation varies from 6.5 m (at sites 60-12 000 years old) to 3.6 m (sites 60 000-120 000 years old; site-specific precipitation given in Richardson et al. 2004).
Leaf lifespan was estimated in March 2002 from between two and six individuals per species per site (Table 1). These six species are all common along the sequence and account for 15-72% of total cover at each site, with the exception of the 60-year-old site where only one species (Griselinia littoralis) was sampled. Each species was sampled where it occurred along the chronosequence. Whole branches were sampled using orchard cutters or a shotgun and leaf lifespan was estimated by retrospective analyses of leaf scars (Lusk 2001; Cornelissen et al. 2003). Each branch was divided into annual segments within which we counted the number of living leaves and the number of leaf abscission scars. As leaves were sampled in autumn, we assumed that the youngest (current) segment was already 6 months old and that subsequent segments were progressively 12 months older than this. We used logistic regression with binomial errors (Crawley 2005) to model the relationship between segment age and leaf survival and to estimate the segment age when 50% of leaves had been lost, taken here as an estimate of median leaf lifespan. Leaf mass per unit area (LMA, g m^sup -2^) and leaf nitrogen concentrations (N) were also measured on current-year foliage from each branch sample (Richardson et al. 2004, 2005). Leaf area was measured using a LiCor Area Meter, Model Li-3100 (Lincoln, NA, USA), leaf mass was measured on material oven-dried at 60°C for 48 h, and leaf nitrogen was determined from dried material using the acid digest and colorimetric methods described in Blakemore et al. (1987). Mean residence times for N and P were calculated using the formula in Kazakou et al. (2007) and our estimates of leaf lifespan from this study in combination with previously published estimates of nutrient resorption efficiency in these species at these sites (Richardson et al. 2005).
A general linear model (GLM) with model contrasts was used to test for differences in leaf lifespan among species and for the effect of site age on leaf lifespan within a species. Similarly, a GLM was used to test for an effect of site age on MRT of N and P. Site age, log-transformed, was used in these models to integrate across several measures of soil fertility that decline along the sequence (i.e. total P, mineral P, aerobic mineralisable N, ratio of organic P to total C). Replicates were individuals within species within sites.
We compared leaf lifespan, N and LMA at Franz Josef with a global dataset of leaf traits (GLOPNET; Wright et al. 2004). Our first goal was to determine how much of the global 'trait space' was sampled by the six species sampled at Franz Josef. To achieve this we used the functional richness index of Mason et al. (2005) to express the proportion of trait space sampled at Franz Josef as a proportion of the total space sampled by GLOPNET. Our second goal was to describe which part of global trait space was occupied by species at Franz Josef. All analyses were conducted in R v.2.9.1 (R Development Core Team 2009).
Mean leaf lifespan varied approximately four-fold among the six species (GLM F^sub 5,168^ = 47.4, P < 0.001; Table 1). The oldest living leaves ranged from 30 months (3 years) in Coprosma foetidissima and Griselinia littoralis to >120 months (>10 years) in Prumnopitys ferruginea, Weinmannia racemosa, and Pseudopanax crassifolius. Leaf lifespan within a species typically varied between 40% and 50% across all samples taken (Fig. 1). Mean leaf lifespan was only significantly related to chronosequence site age in two species; leaf lifespan decreased with site age in Pseudopanax crassifolius (GLM F^sub 1,27^ = 15.7, P = 0.0005; Fig. 1) and Prumnopitys ferruginea (GLM F^sub 1,22^ = 5.0, P = 0.036). Mean residence time (MRT) of N and P varied greatly among species and among sites and was commonly greater for P than for N (Table 2). The MRT of N increased significantly with chronosequence site age in Metrosideros umbellata (GLM F^sub 1,26^ = 4.45, P = 0.045) and Pseudopanax crassifolius (GLM F^sub 1,27^ = 6.38, P = 0.018) and the MRT of P increased significantly in Coprosma foetidissima (GLM F^sub 1,23^ = 4.70, P = 0.04), M. umbellata (GLM F^sub 1,26^ = 7.81, P = 0.01), and Weinmania racemosa (GLM F^sub 1,28^ = 18.2, P < 0.001).
The six species sampled at Franz Josef sampled 12% of the full range of global leaf lifespans (Fig. 2) and occupied that part of the leaf economics spectrum characterised by conservative leaf traits; relatively long-lived leaves with low leaf nitrogen concentrations (Fig. 2). Indeed, the longest mean leaf lifespan estimated from an individual at Franz Josef (133 months, Pseudopanax crassifolius at the 250-year-old site) was towards the upper limit of leaf lifespans sampled globally (Fig. 2).
We found no evidence that leaf lifespan increased within species as soil fertility declined. Leaf lifespan of Pseudopanax crassifolius decreased significantly with declining fertility but this response may be accounted for, in part, by heteroblastic shifts within the species. Individuals of P. crassifolius at the early chronosequence sites are younger individuals supporting longer, narrower, tougher pre-adult leaves ('transitional' sensu Gould 1993) that may have longer leaf lifespans than fully mature, adult leaves at the older sites (Dansereau 1964; Gould 1993). However, leaf lifespan of homoblastic Prumnopitys ferruginea also decreased significantly with declining soil fertility, counter to our predictions, suggesting that heteroblasty alone could not explain our results. Longer leaf lifespans on young, fertile sites have been reported from the Hawaiian chronosequence for Metrosideros polymorpha (Cordell et al. 2001) and the tree fern Cibotium glaucum (Walker & Aplet 1994). Cordell et al. (2001) suggested that other factors operating along the soil chronosequence, such as loss of soil structure and impeded drainage, may confound within-species responses of leaf lifespan to soil fertility, but these additional factors are poorly understood.
Previous studies at the Franz Josef chronosequence have demonstrated that both leaf and litter nutrient concentrations were highly plastic and exhibited strong directional responses to declining soil fertility (Richardson et al. 2004, 2005). When combined with the leaf lifespan estimates reported here, there was evidence that the MRT of both N and P increased in some species along the chronosequence (Table 2). These increases were largely driven by within-species shifts in resorption efficiency (Richardson et al. 2005). This result is unusual as the importance of leaf lifespan in determining intraspecific variation in MRT is thought to be equal to that of resorption (Eckstein et al. 1999) or substantially greater (Kazakou et al. 2007).
One explanation for the relatively invariant leaf lifespans reported here is that all the sites sampled are broadly 'nutrient-poor' (soil total P < 900 mg kg^sup -1^; Richardson et al. 2004) and that intraspecific responses by leaf lifespan to soil nutrients only occur between extremely fertile sites and these more widespread, 'nutrient poor' sites. McGlone et al. (2004) proposed that low soil fertility throughout New Zealand accounted for the low incidence of deciduousness and short leaf lifespans in the woody flora. We suggest that limited plasticity of leaf lifespan within species is an additional dimension to that putative constraint on leaf lifespan. A similar suggestion was made by Cordell et al. (2001) to account for the absence of a response by leaf lifespan to fertiliser addition at old, nutrient-poor sites in Hawai'i. The extent of trait plasticity on infertile soils could be explored at Franz Josef using fertilisation studies as part of a wider effort to determine how environment controls key plant functional traits, both within and among New Zealand species. Finally, the significance of these within-species responses to community-level shifts in leaf lifespan are unknown as data are not available for all species at all sites. Compositional turnover among sites and the sitelevel dominance of species with distinctively short (e.g. Aristotelia serrata and Melicytus ramiflorus) and long (e.g. Lagarostrobos colensoi and Phyllocladus alpinus) leaf lifespans at the youngest and oldest sites, respectively, may override the absence of within-species shifts reported here.
The New Zealand woody flora is overwhelmingly evergreen (McGlone et al. 2004). The six species sampled here are all common woody species found throughout New Zealand. If the leaf lifespans we estimated at this site are typical of values for other similar species, our data suggest a mean leaf lifespan for New Zealand trees of between 2 and 3 years. Few strictly comparable, quantitative estimates are available for other evergreen woody species in New Zealand, but descriptive estimates of leaf lifespan from Nothofagus spp. (c. 13 months for N. fusca, 1-2 years for N. solandri and N. solandri var. cliffortioides, and 3-5 years for N. menziesii; Russell 1936; Bussell 1968; Wardle 1984) correspond well with the range sampled at Franz Josef. While average leaf lifespans of 1-5 years are typical of temperate evergreen rainforest species (Wardle 1991; Lusk et al. 2003), the lifespans of individual leaves can be substantially greater. For example, the oldest living Pseudopanax crassifolius leaves sampled in this study were >11 years old (Table 1), despite the mean leaf lifespan being 3-4 years. Extraordinary individual leaf lifespans may be widespread among slow-growing species; Wardle (1963) estimated that some living leaves of the long-lived subalpine conifer species Halocarpus biformis could be >50 years old (600 months). These long-lived individual leaves must still be productive in order to be retained, suggesting they are well situated for photosynthesis and sufficiently robust to have avoided damage. However, robust leaves of high structural carbohydrate content would have low leaf-level productivity and thus the presence of very old leaves can only be characteristic of long-lived species with slow growth rates and low carbon gain requirements from individual leaves.
Despite only sampling six species at Franz Josef, we captured 12% of the total variation in leaf lifespan reported globally. These data from Franz Josef highlight that despite large-scale constraints on leaf construction by climate, local-scale mechanisms generate and maintain high trait diversity in a single ecosystem. Leaves at Franz Josef appeared to have low N content relative to their lifespans or their LMA, when compared with a global dataset (Fig. 2). Wright et al. (2001) demonstrated that leaves from high rainfall (1220 mm) environments in southern Australia had low leaf N concentrations relative to their LMA when compared with leaves from low rainfall (387 mm) environments. This was interpreted as a water conservation mechanism; a high leaf N concentration enables plants in dry environments to rapidly achieve a high internal CO2 concentration at a low stomatal conductance. This interpretation could be used to argue that species at Franz Josef, operating under exceptionally high rainfall (3000 - >7000 mm) and constantly perhumid conditions, can achieve high internal CO2 concentrations at comparatively low leaf N concentrations.
We thank the Department of Conservation for permission to collect samples, David Wardle for introducing us to the chronosequence, Melissa Brignall-Theyer and Jenny Bee for field assistance, Ian Wright and Peter Reich for permission to use the GLOPNET data, and Ian Dickie for critical review. Research was funded by the New Zealand Foundation for Research, Science and Technology Ecosystem Resilience Outcome-Based Investment (Contract C09X0502), the Marsden Foundation of the Royal Society of New Zealand, and Landcare Research's retained earnings.
Aerts R, Chapin III FS 2000. The mineral nutrition of wild plants revisited: A re-evaluation of processes and patterns. Advances in Ecological Research 30: 1-67.
Allan HH 1961. Flora of New Zealand. Vol. 1. Wellington, Government Printer. 1085 p.
Blakemore LC, Searle PL, Daly BK 1987. Methods for chemical analysis of soils. Rev. edn. NZ Soil Bureau Scientific Report 80. Lower Hutt, Department of Scientific and Industrial Research. 103 p.
Bussell WT 1968. The growth of some New Zealand trees. 1. Growth in natural conditions. New Zealand Journal of Botany 6: 63-75.
Chabot BF, Hicks DJ 1980. The ecology of leaf life spans. Annual Review of Ecology and Systematics 13: 229-259.
Cordell S, Goldstein G, Meinzer FC, Vitousek PM 2001. Regulation of leaf life-span and nutrient-use efficiency of Metrosideros polymorpha trees at two extremes of a long chronosequence in Hawaii. Oecologia 127: 198-206.
Cornelissen JHC, Lavorel S, Garnier E, Díaz S, Buchmann N, Gurvich DE, Reich PB, ter Steege H, Morgan HD, van der Heijden MGA, Pausas JG, Poorter H 2003. A handbook of protocols for standardised and easy measurement of plant functional traits worldwide. Australian Journal of Botany 51: 335-380.
Crawley MJ 2005. Statistics. An introduction using R. Chichester, UK, John Wiley. 327 p.
Dansereau P 1964. Six problems in New Zealand vegetation. Bulletin of the Torrey Botanical Club 91: 114-140.
Eckstein RL, Karlsson PS, Weih M 1999. Leaf life span and nutrient resorption as determinants of plant nutrient conservation in temperate-arctic regions. New Phytologist 143: 177-189.
Escudero A, del Arco JM, Sanz IC, Ayala J 1992. Effects of leaf longevity and retranslocation efficiency on the retention time of nutrients in the leaf biomass of different woody species. Oecologia 90: 80-87.
Gould KS 1993. Leaf heteroblasty in Pseudopanax crassifolius: functional significance of leaf morphology and anatomy. Annals of Botany 71: 61-70.
Hessell JWD 1982. The climate and weather of Westland. New Zealand Meteorological Service Miscellaneous Publication 115(10). 44 p.
Kazakou E, Garnier E, Gimenez O 2007. Contribution of leaf life span and nutrient resorption to mean residence time: elasticity analysis. Ecology 88: 1857-1863.
Lambers H, Poorter H 1992. Inherent variation in growth rate between higher plants: A search for physiological causes and ecological consequences. Advances in Ecological Research 23: 187-261.
Lusk CH 2001. Leaf life spans of some conifers of the temperate forests of South America. Revista Chilena de Historia Natural 74: 711-718.
Lusk CH, Wright I, Reich PB 2003. Photosynthetic differences contribute to competitive advantage of evergreen angiosperm trees over evergreen conifers in productive habitats. New Phytologist 160: 329-336.
Mason NWH, Mouillot D, Lee WG, Wilson JB 2005. Functional richness, functional evenness and functional divergence: the primary components of functional diversity. Oikos 111: 112-118.
McGlone MS, Dungan RJ, Hall GMJ, Allen RB 2004. Winter leaf loss in the New Zealand woody flora. New Zealand Journal of Botany 42: 1-19.
Monk CD 1966. An ecological significance of evergreenness. Ecology 47: 504-505.
R Development Core Team 2009. R: A language and environment for statistical computing. Vienna, Austria, R Foundation for Statistical Computing. ISBN 3-900051-07-0, URL http://www. R-project.org.
Reich PB, Walters MB, Ellsworth DS 1997. From tropics to tundra: Global convergence in plant functioning. Proceedings of the National Academy of Sciences USA 94: 13730-13734.
Richardson SJ, Peltzer DA, Allen RB, McGlone MS, Parfitt RL 2004. Rapid development of phosphorus limitation in temperate rainforest along the Franz Josef soil chronosequence. Oecologia 139: 267-276.
Richardson SJ, Peltzer DA, Allen RB, McGlone MS 2005. Resorption proficiency along a chronosequence: responses among communities and within species. Ecology 86: 20-25.
Russell RS 1936. The mechanism of leaf-fall in certain New Zealand trees. Transactions and Proceedings of the Royal Society of New Zealand 65: 407-421.
Stevens PR 1968. A chronosequence of soils near the Franz Josef Glacier. Unpublished PhD thesis, University of Canterbury, Christchurch, New Zealand. 389 p.
Walker LR, Aplet GH 1994. Growth and fertilization responses of Hawai'ian tree ferns. Biotropica 26: 378-383.
Wardle JA 1984. The New Zealand beeches: ecology, utilisation and management. Wellington, New Zealand Forest Service. 447 p.
Wardle P 1963. Growth habits of New Zealand subalpine shrubs and trees. New Zealand Journal of Botany 1: 18-47.
Wardle P 1991. Vegetation of New Zealand. Cambridge University Press. 672 p.
Whitehead D, Boelman N, Turnbull M, Griffin K, Tissue D, Barbour M, Hunt J, Richardson S, Peltzer D 2005. Photosynthesis and reflectance indices for rainforest species in ecosystems undergoing progression and retrogression along a soil fertility chronosequence in New Zealand. Oecologia 144: 233-244.
Wright IJ, Reich PB, Westoby M 2001. Strategy-shifts in leaf physiology, structure and nutrient content between species of high and low rainfall, and high and low nutrient habitats. Functional Ecology 15: 423-434.
Wright IJ, Reich PB, Westoby M, Ackerly DD, Baruch Z, Bongers F, Cavender-Bares J, Chapin T, Cornelissen JHC, Diemer M, Flexas J, Garnier E, Groom PK, Gulias J, Hikosaka K, Lamont BB, Lee T, Lee W, Lusk C, Midgley JJ, Navas M-L, Niinemets Ü, Oleksyn J, Osada N, Poorter H, Poot P, Prior L, Pyankov VI, Roumet C, Thomas SC, Tjoelker MG, Veneklaas EJ, Villar R 2004. The worldwide leaf economics spectrum. Nature 428: 821-827.
Editorial Board member: Chris Lusk
Received 4 November 2009; accepted 25 March 2010
Sarah J. Richardson*, Duane A. Peltzer, Robert B. Allen and Matt S. McGlone
Landcare Research, PO Box 40, Lincoln 7640, New Zealand
*Author for correspondence (Email: email@example.com)
Published on-line: 13 May 2010