Native and exotic invasive plants have fundamentally similar carbon capture strategies

Summary

1. Leaf trait relationships of native and exotic invasive species from a range of habitats were compared to assess consistency across habitats and the role of disturbance.

2. One hundred and twenty-two native and exotic species were sampled in five habitats in eastern Australia. Specific leaf area, foliar nitrogen (N_mass), assimilation rate (A_mass) and dark respiration (R_mass) were measured for each species. Plants were classified into four types: native undisturbed, native disturbed, exotic invasive undisturbed and exotic invasive disturbed.

3. All traits were positively correlated and slopes were homogeneous within habitats. Significant differences between plant types in slope elevation were found in only two of 18 cases. There were significant shifts in group means along a common slope between plant types within habitats. These shifts were associated with disturbed vs. undisturbed areas, with plant types from disturbed areas having higher trait values.

4. Synthesis. Exotic invasive and native species do not have fundamentally different carbon capture strategies. The carbon capture strategy of a species is strongly associated with disturbance, with species from disturbed sites having traits that confer capacity for fast growth. Thus, differences between exotic invasives and natives may reflect differences in the environmental conditions of the sites where they occur rather than differences between exotic invasives and natives per se.

Introduction

There has been considerable progress in the last decade in understanding strategies of plants in relation to leaf carbon economics (Reich, Walters & Ellsworth 1997; Westoby et al. 2002; Wright et al. 2004, 2005a,b). This work has identified variation in leaf traits that are central to the carbon fixation strategy of plants as one of the major spectrums of variation in plant ecological strategies. A fundamental trade-off in carbon fixation is between specific leaf area (SLA) and leaf life span (LL), where SLA can be thought of as the investment that is deployed per unit of light capture surface (leaf area per mass) and LL gives the expected duration of the return from that investment (Grime et al. 1997; Westoby 1998). Species with low SLA generally have longer LL than high-SLA species and also tend to have lower photosynthetic capacity, leaf nitrogen concentration and dark respiration rates (Field & Mooney 1986; Reich, Walters & Ellsworth 1997; Reich et al. 1998). Consequently, species with high SLA have a greater potential for fast growth than low SLA species. The work by Reich, Walters & Ellsworth (1997), Reich et al. (1999), Wright & Westoby (2002), Wright et al. (2004, 2005a,b) and Wright, Reich & Westoby (2001) has shown that scaling slopes in leaf trait relationships tend not to differ between growth forms and habitats, suggesting tight trade-offs and trait coordination, evolving repeatedly in many different lineages and places.

The success of exotic invasive species is often attributed to their capacity for fast growth, particularly when resources are not limiting (Pyšek & Richardson 2007). It is widely agreed that increases in resource availability result in invasion (Davis, Grime & Thompson 2000; Davis & Pelsor 2001; Gurvich, Tecco & Diaz 2005) and that increases in resource availability associated with disturbance differentially increase the performance of invaders over that of natives (Daehler 2003). Previous studies comparing leaf traits of native and exotic invasive species have shown that invasives have larger SLA than native species (Baruch & Goldstein 1999; Gulias et al. 2003; Leishman et al. 2007). Similarly, within alien species invasiveness has been shown to be associated with high SLA (Grotkopp, Rejmanek & Rost 2002; Hamilton et al. 2005; Burns 2006; Grotkopp & Rejmanek 2007). Exotic invasive species have also been shown to have higher relative growth rate (Pattison, Goldstein & Ares 1998; Grotkopp, Rejmanek & Rost 2002; Burns 2006; Grotkopp & Rejmanek 2007; but see Bellingham et al. 2004), foliar nutrients (Baruch & Goldstein 1999; Durand & Goldstein 2001; Craine & Lee 2003; Leishman et al. 2007) and photosynthetic capacity (Baruch & Goldstein 1999; Durand & Goldstein 2001; McDowell 2002; Leishman et al. 2007). Thus, there appears to be consistent evidence that exotic invasive species have leaf traits that confer the capacity for rapid growth, and this is likely to contribute to their success in environments that are not resource-limited.

Fewer studies have examined leaf trait relationships of exotic invasive compared with native species. Recently, Funk & Vitousek (2007) argued that exotic invasives may have higher resource-use efficiency (RUE) than natives, allowing them to successfully invade resource-poor environments. They studied leaf traits of 19 pairs of phylogenetically related exotic invasive and native species from Hawaii and found no significant difference between exotic invaders and natives in water use efficiency, but found higher photosynthetic nitrogen- and energy-use efficiency of exotic invaders in light and water-limited environments. However, Leishman et al. (2007) examined scaling relationships of foliar nitrogen (N_mass) and assimilation rate (A_mass) of 40 native and 57 exotic invasive species from the global literature and found no significant difference in slope or intercept of the relationship, indicating no difference in nitrogen-use efficiency between natives and exotics. They suggested that scaling relationships are a more robust method for examining different strategies of resource acquisition as differences in ratios such as RUE are affected by both the slope and intercept of the line describing the relationship. The few other studies that have examined leaf trait relationships have reported that exotic invasives have higher A_mass for a given SLA or N_mass (Gulias et al. 2003 for Mediterranean species; Leishman et al. 2007 for species on the nutrient-enriched Hawkesbury Sandstone-derived soils of the region around Sydney, Australia). These results suggest that exotic invasives can achieve a greater carbon gain for a given investment in leaves, resulting in even faster growth compared with natives.

In this study, we compared the leaf trait relationships of co-occurring native and exotic invasive species from five distinctly different habitats in eastern Australia to assess whether there are differences in leaf trait relationships of natives and exotic invasives and whether these findings are consistent across habitats. We recognize that the definition of native and exotic invasive species can sometimes be problematic: for example, species that are native and non-invasive in their original range may be invasive in a novel range. In this study, this is true for several species including Acacia longifolia, Casuarina cunninghamiana, Eucalyptus camaldulensis and Pittosporum undulatum. Nevertheless, we hope that by taking a site-based approach we have minimized this terminology issue. We used standardized major axis (SMA) regression to compare scaling relationships between each possible pairwise combination of the traits SLA, N_mass, A_mass and dark respiration (R_mass). We tested for the following possibilities within each habitat: (i) leaf traits scale positively across all species, with differences between plant types seen as shifts along a common slope; (ii) leaf traits scale positively across all species, but the slope elevations differ between plant types; (iii) leaf traits scale positively across all species, but the scaling slopes differ between plant types; and (iv) there is no difference between plant types in slope, group means or slope elevation. In some of the habitats, the natives and exotic invasives co-occurred in a disturbed environment, while in others the exotic invasives occurred in a disturbed environment immediately adjacent to an undisturbed area dominated by native species. Thus, we were able to examine the influence of disturbance (and hence resource availability) on leaf trait relationships by comparing four plant types in total: native undisturbed, native disturbed, exotic invasive undisturbed and exotic invasive disturbed.

Materials and methods

Study sites, sampling and measurement protocol

We selected five sites that represented a broad range of vegetation types where native and exotic invasive species either co-occurred (disturbed sites) or occurred in adjacent sites (disturbed and undisturbed areas adjacent). All sites were located in New South Wales, Australia. The characteristics of each site are shown in Table 1, including location, climate, vegetation type, disturbance type and soil nutrients. At each site, we identified the most abundant native and exotic invasive plant species, ensuring that at least 10 species of each plant type were sampled. Species represented a range of families and growth forms (Table 2). For each species, we sampled five individuals. When it was unclear that individual plants were not connected underground (i.e. ramets), we ensured that we sampled ‘individuals’ that were as spatially separate as possible.

Table 1.   Characteristics of the five sites sampled

	Sydney sandstone woodland	Sydney Cumberland Plain woodland	Hunter Valley riparian	Dorrigo subtropical rain forest	Blue Mountains woodland
Vegetation type	Eucalypt woodland	Cumberland plain woodland	Riparian woodland	Subtropical rain forest	Eucalypt woodland
Location (Australia)	Pymble, Sydney region	Scheyville, Sydney region	Singleton, Hunter Valley	Dorrigo, mid-coast	Blackheath, Blue Mountains
Latitude (ºS)	33.75	33.61	32.57	30.32	33.63
Longitude (ºE)	151.14	150.88	151.17	152.71	150.28
Mean annual daily min–max temp (ºC)	11.1–22.6	10.5–23.9	11.6–24.3	12.0–25.6	7.9–16.6
Mean annual rainfall (mm)	1151.2	803.6	723.7	1517.8	1402.3
Elevation (m)	65.0	20.0	73.1	15.0	1030.0
Disturbance type	Nutrient enrichment (stormwater outlet)	Physical disturbance (road edge, powerline clearance)	Nutrient enrichment (adjacent agricultural land)	Physical disturbance (clearing and flooding)	Nutrient enrichment (stormwater outlet) and physical disturbance (adjacent to road)
Soil total P (mg kg^−1)	47 (undisturbed)	410 (disturbed)	460 (disturbed)	470 (undisturbed)	150 (undisturbed)
	240 (disturbed)			670 (disturbed)	220 (disturbed)
Soil total N (% w/w)	0.13 (undisturbed)	0.26 (disturbed)	0.21 (disturbed)	0.30 (undisturbed)	0.13 (undisturbed)
	0.44 (disturbed)			0.57 (disturbed)	0.20 (disturbed)
No. each plant type sampled
Native undisturbed	11	0	0	12	11
Native disturbed	10	12	11	0	0
Exotic invasive undisturbed	0	0	0	0	5
Exotic invasive disturbed	11	11	13	10	6

Table 2.   List of plant species, family, growth form and status for all plant species sampled at five sites representing five different habitats in eastern New South Wales, Australia

Species	Family	Growth form	Status
Sydney sandstone woodland
Acetosa sagittata	Polygonaceae	Climber	Exotic disturbed
Asparagus aethiopicus	Asparagaceae	Herb	Exotic disturbed
Cardiospermum grandiflorum	Sapindaceae	Climber	Exotic disturbed
Ehrharta erecta	Poaceae	Grass	Exotic disturbed
Lantana camara	Verbenaceae	Shrub	Exotic disturbed
Ligustrum lucidum	Oleaceae	Shrub/small tree	Exotic disturbed
Nephrolepis cordifolia	Davalliaceae	Fern	Exotic disturbed
Senna pendula	Fabaceae–Caesalpinioideae	Shrub	Exotic disturbed
Sida rhombifolia	Malvaceae	Herb	Exotic disturbed
Solanum mauritianum	Solanaceae	Shrub/small tree	Exotic disturbed
Tradescantia albiflora	Commelinaceae	Herb	Exotic disturbed
Acacia parramattensis	Fabaceae–Mimosaceae	Shrub/small tree	Native disturbed
Casuarina glauca	Casuarinaceae	Tree	Native disturbed
Commelina cyanea	Commelinaceae	Herb	Native disturbed
Dianella caerulea	Phormiaceae	Herb	Native disturbed
Imperata cylindrica	Poaceae	Grass	Native disturbed
Lomandra longifolia	Lomandraceae	Herb	Native disturbed
Oplismenus aemulus	Poaceae	Grass	Native disturbed
Pittosporum undulatum	Pittosporaceae	Tree	Native disturbed
Pteridium esculentum	Dennstaedtiaceae	Fern	Native disturbed
Sigesbeckia orientalis	Asteraceae	Herb	Native disturbed
Acacia longifolia subsp. longifolia	Fabaceae–Mimosaceae	Shrub	Native undisturbed
Acacia suaveolens	Fabaceae–Mimosaceae	Shrub	Native undisturbed
Banksia serrata	Proteaceae	Shrub/small tree	Native undisturbed
Banksia spinulosa	Proteaceae	Shrub	Native undisturbed
Corymbia gummifera	Myrtaceae	Tree	Native undisturbed
Dodonaea triquetra	Sapindaceae	Shrub	Native undisturbed
Grevillea buxifolia	Proteaceae	Shrub	Native undisturbed
Persoonia levis	Proteaceae	Shrub/small tree	Native undisturbed
Persoonia pinifolia	Proteaceae	Shrub	Native undisturbed
Smilax glyciphylla	Smilacaceae	Climber	Native undisturbed
Xanthorrhoea sp.	Xanthorrhoeaceae	Herb	Native undisturbed
Sydney Cumberland Plain woodland
Anredera cordifolia	Basellaceae	Climber	Exotic disturbed
Asparagus asparagoides	Asparagaceae	Climber	Exotic disturbed
Bidens pilosa	Asteraceae	Herb	Exotic disturbed
Cotoneaster glaucophyllus	Malaceae	Shrub	Exotic disturbed
Eragrostis curvula	Poaceae	Grass	Exotic disturbed
Lantana camara	Verbenaceae	Shrub	Exotic disturbed
Olea europaea subsp. cuspidata	Oleaceae	Shrub/small tree	Exotic disturbed
Plantago lanceolata	Plantaginaceae	Perennial herb	Exotic disturbed
Rubus ulmifolius	Rosaceae	Shrub	Exotic disturbed
Senecio sp.	Asteraceae	Herb	Exotic disturbed
Solanum nigrum	Solanaceae	Herb	Exotic disturbed
Acacia falcata	Fabaceae–Mimosaceae	Shrub	Native disturbed
Brunoniella australis	Acanthaceae	Herb	Native disturbed
Bursaria spinosa	Pittosporaceae	Shrub	Native disturbed
Dichondra repens	Convolvulaceae	Herb	Native disturbed
Einadia hastata	Chenopodiaceae	Herb	Native disturbed
Eremophila debilis	Myoporaceae	Shrub	Native disturbed
Eucalyptus crebra	Myrtaceae	Tree	Native disturbed
Eucalyptus moluccana	Myrtaceae	Tree	Native disturbed
Glycine microphylla	Fabaceae–Faboideae	Climber	Native disturbed
Hardenbergia violacea	Fabaceae–Faboideae	Climber	Native disturbed
Microlaena stipoides	Poaceae	Grass	Native disturbed
Pratia purpurascens	Lobeliaceae	Herb	Native disturbed
Hunter Valley riparian
Anredera cordifolia	Basellaceae	Climber	Exotic disturbed
Cestrum parqui	Solanaceae	Shrub	Exotic disturbed
Foeniculum vulgare	Apiaceae	Herb	Exotic disturbed
Galenia pubescens	Aizoaceae	Herb	Exotic disturbed
Ipomoea purpurea	Convolvulaceae	Climber	Exotic disturbed
Medicago polymorpha	Fabaceae–Faboideae	Herb	Exotic disturbed
Olea europaea subsp. cuspidata	Oleaceae	Shrub/small tree	Exotic disturbed
Panicum spp.	Poaceae	Grass	Exotic disturbed
Ricinus communis	Euphorbiaceae	Shrub	Exotic disturbed
Rumex crispus	Polygonaceae	Herb	Exotic disturbed
Salix sp.	Salicaceae	Tree	Exotic disturbed
Schinus areira	Anacardiaceae	Tree	Exotic disturbed
Sonchus oleraceus	Asteraceae	Herb	Exotic disturbed
Acacia parvipinnula	Fabaceae–Mimosaceae	Shrub	Native disturbed
Acacia salicina	Fabaceae–Mimosaceae	Shrub	Native disturbed
Casuarina cunninghamiana	Casuarinaceae	Tree	Native disturbed
Dichondra repens	Convolvulaceae	Herb	Native disturbed
Eucalyptus camaldulensis	Myrtaceae	Tree	Native disturbed
Melia azedarach	Meliaceae	Tree	Native disturbed
Microlaena stipoides	Poaceae	Grass	Native disturbed
Persicaria decipiens	Polygonaceae	Herb	Native disturbed
Pratia purpurascens	Lobeliaceae	Herb	Native disturbed
Rumex brownii	Polygonaceae	Herb	Native disturbed
Urtica incisa	Urticaceae	Herb	Native disturbed
Dorrigo subtropical rain forest
Ageratina riparia	Asteraceae	Herb	Exotic disturbed
Araujia sericifera	Apocynaceae	Climber	Exotic disturbed
Lantana camara	Verbenaceae	Shrub	Exotic disturbed
Phyllostachys spp.	Poaceae	Shrub	Exotic disturbed
Ricinus communis	Euphorbiaceae	Shrub/small tree	Exotic disturbed
Rosa rubiginosa	Rosaceae	Shrub	Exotic disturbed
Senna sp.	Fabaceae–Caesalpinioideae	Shrub	Exotic disturbed
Solanum mauritianum	Solanaceae	Shrub/small tree	Exotic disturbed
Tradescantia albiflora	Commelinaceae	Herb	Exotic disturbed
Unknown	Unknown	Herb	Exotic disturbed
Callicoma serratifolia	Cunoniaceae	Tree	Native undisturbed
Cordyline stricta	Asteliaceae	Shrub	Native undisturbed
Duboisia myoporoides	Solanaceae	Tree	Native undisturbed
Eupomatia laurina	Eupomatiaceae	Shrub/small tree	Native undisturbed
Ficus coronata	Moraceae	Shrub/small tree	Native undisturbed
Glochidion ferdinandi	Phyllanthaceae	Shrub/small tree	Native undisturbed
Guioa semiglauca	Sapindaceae	Tree	Native undisturbed
Homalanthus populifolius	Euphorbiaceae	Shrub/small tree	Native undisturbed
Marsdenia rostrata	Apocynaceae	Climber	Native undisturbed
Neolitsea dealbata	Lauraceae	Shrub/small tree	Native undisturbed
Ripogonum album	Ripogonaceae	Climber	Native undisturbed
Rubus moluccanus var. trilobus	Rosaceae	Climber	Native undisturbed
Blue Mountains woodland
Crocosmia × crocosmiiflora	Iridaceae	Herb	Exotic invasive disturbed
Cytisus scoparius	Fabaceae–Faboideae	Shrub	Exotic invasive disturbed
Lonicera japonica	Caprifoliaceae	Climber	Exotic invasive disturbed
Populus alba	Salicaceae	Tree	Exotic invasive disturbed
Ranunculus repens	Ranunculaceae	Herb	Exotic invasive disturbed
Vinca major	Apocynaceae	Herb	Exotic invasive disturbed
Cotoneaster franchetii	Malaceae	Shrub	Exotic invasive undisturbed
Hedera helix	Araliaceae	Climber	Exotic invasive undisturbed
Ilex aquifolium	Aquifoliaceae	Shrub/small tree	Exotic invasive undisturbed
Pinus radiata	Pinaceae	Tree	Exotic invasive undisturbed
Prunus laurocerasus	Amygdalaceae	Shrub	Exotic invasive undisturbed
Acacia rubida	Fabaceae–Mimosaceae	Shrub	Native undisturbed
Dianella caerulea	Phormiaceae	Herb	Native undisturbed
Eucalyptus sieberi	Myrtaceae	Tree	Native undisturbed
Grevillea laurifolia	Proteaceae	Shrub	Native undisturbed
Hakea dactyloides	Proteaceae	Shrub	Native undisturbed
Leptospermum polygalifolium	Myrtaceae	Shrub/small tree	Native undisturbed
Mirbelia platylobioides	Fabaceae–Faboideae	Shrub	Native undisturbed
Olearia asterotricha	Asteraceae	Shrub	Native undisturbed
Persoonia myrtilloides subsp. myrtilloides	Proteaceae	Shrub	Native undisturbed
Pittosporum undulatum	Pittosporaceae	Shrub/small tree	Native undisturbed
Symphionema montanum	Proteaceae	Subshrub	Native undisturbed

For each individual, we measured assimilation rate (A_max) and dark respiration (R_mass) using a LI-COR LI-6400 portable photosynthesis system. For assimilation rates, ambient CO₂ concentration was maintained at 500 μL L⁻¹, CO₂ reference at 375 p.p.m., relative humidity at 20–50%, leaf temperature at 20 °C and PAR at 1500 μmol m⁻² s⁻¹. For dark respiration rates, ambient CO₂ concentration was maintained at 300–500 μL L⁻¹, CO₂ reference at 375 p.p.m., relative humidity at 20–50%, leaf temperature at 20 °C and PAR at 0 μmol m⁻² s⁻¹. When an individual leaf filled the chamber, measurements were taken on one young fully expanded leaf. If leaves were too small to fill the chamber then multiple leaves were used where possible. For assimilation rate measurements, leaves were selected that were in full sun. For dark respiration measurements, twigs from each individual were collected and stored in the dark in moist conditions for at least 30 min (and no longer than 3 h) before measurements were taken. Measured leaves were collected, scanned to determine their area, dried and weighed. Photosynthetic and respiration rates were then calculated on a per-gram-dry-matter basis.

From each individual, we collected young- to medium-aged fully expanded leaves to calculate SLA (leaf area per unit mass) and determine foliar N. Projected leaf area was determined for five leaves using a flatbed scanner (HP Desk II scanner, Hewlett-Packard, Forest Hill, Vic., Australia) and image analysis software (Delta-T SCAN; Delta-T Devices, Burwell, Cambridge, UK). Leaves were then oven-dried at 60 °C for 48 h before weighing. Dried leaf material was ground and analysed for nitrogen content using a LECO C : H : N analyser.

Five soil samples (5 cm diameter and 10 cm depth) were collected from random locations at each site, bulked and oven-dried at 110 °C for 48 h, then sieved using a 2 mm sieve. These were then analysed for total phosphorus (total P) and total nitrogen (total N), using ICP-AES following aqua-regia digestion (total P) and cadmium reduction flow injection (Clesceri, Greenberg & Eaton 1999) (total N). All soil analyses were conducted at the National Measurement Institute, Pymble, NSW, Australia.

Data analysis

Measurements of individuals were averaged for each species at each site. Species were grouped into plant types based on their status as exotic invasive or native and on their occurrence in disturbed or undisturbed sites, resulting in four possible plant types: native undisturbed, native disturbed, exotic invasive undisturbed, exotic invasive disturbed. As we wanted to compare scaling relationships between plant types, we used SMA regression to describe the relationship between each possible pairwise combination of leaf-level traits. SMA slopes were fitted for each plant type and tested for homogeneity. A common slope was fitted when slopes did not differ significantly. Elevation differences between SMA slopes were tested for by comparing residual axis scores, which measure the spread of data away from the fitted common slope, using the WALD test. Shifts between plant types along the common fitted slope were tested for by comparing fitted axis scores, which measure the spread of data along the common slope, using the WALD test. All data were log-transformed for analysis and analyses were performed using (S)MATR software (Falster, Warton & Wright 2006).

Results

We first examined correlation coefficients for each pairwise trait combination within each plant type for the five sites. We examined whether the correlations between traits were significant at P < 0.10. All traits were positively correlated: most trait-pair combinations had significant correlation coefficients, with r² values ranging from 0.13 to 0.88 (Table 3). The relationships between SLA and A_mass and between A_mass and R_mass were the most consistent, with 10 of 12 possible cases being significant for both relationships. When species from all plant types from the five sites were combined, 19 of 24 pairwise relationships were significant.

Table 3.   Correlation coefficients for each pairwise trait comparison by plant type for each of the five sites studied. NS indicates that the correlation between the two traits was not significant (P > 0.1)

Site	Plant types	SLA vs. Amass	SLA vs. Rmass	SLA vs. Nmass	Amass vs. Rmass	Amass vs. Nmass	Rmass vs. Nmass
Sydney sandstone woodland	Native, undisturbed	r^2 = 0.45, P = 0.02	r^2 = 0.67, P < 0.01	NS	r^2 = 0.82, P < 0.01	NS	NS
	Native, disturbed	r^2 = 0.88, P < 0.01	r^2 = 0.60, P < 0.01	r^2 = 0.48, P = 0.03	r^2 = 0.76, P < 0.01	r^2 = 0.38, P = 0.06	r^2 = 0.34, P = 0.07
	Exotic, disturbed	r^2 = 0.56, P < 0.01	r^2 = 0.24, P = 0.01	r^2 = 0.34, P = 0.06	NS	r^2 = 0.75, P < 0.01	NS
Sydney Cumberland Plain woodland	Native, disturbed	r^2 = 0.79, P < 0.01	r^2 = 0.62, P < 0.01	r^2 = 0.42, P = 0.02	r^2 = 0.70, P < 0.01	r^2 = 0.56, P < 0.01	r^2 = 0.44, P = 0.02
	Exotic, disturbed	r^2 = 0.76, P < 0.01	r^2 = 0.87, P < 0.01	r^2 = 0.80, P < 0.01	r^2 = 0.72, P < 0.01	r^2 = 0.73, P < 0.01	r^2 = 0.67, P < 0.01
Hunter Valley riparian	Native, disturbed	r^2 = 0.75, P < 0.01	r^2 = 0.76, P < 0.01	NS	r^2 = 0.74, P < 0.01	NS	NS
	Exotic, disturbed	r^2 = 0.50, P = 0.01	r^2 = 0.34, P = 0.04	NS	r^2 = 0.49, P = 0.01	r^2 = 0.45, P = 0.01	NS
Dorrigo subtropical rain forest	Native, undisturbed	r^2 = 0.27, P = 0.08	NS	NS	r^2 = 0.45, P = 0.02	r^2 = 0.13, P = 0.01	r^2 = 0.34, P = 0.06
	Exotic, disturbed	NS	NS	NS	r^2 = 0.78, P < 0.01	NS	r^2 = 0.75, P = 0.01
Blue Mountains woodland	Native, undisturbed	r^2 = 0.41, P = 0.03	r^2 = 0.40, P = 0.04	NS	r^2 = 0.28, P = 0.09	NS	NS
	Exotic, undisturbed	r^2 = 0.75, P = 0.06	NS	r^2 = 0.80, P = 0.04	NS	NS	NS
	Exotic, disturbed	NS	NS	r^2 = 0.67, P = 0.04	r^2 = 0.54, P = 0.10	r^2 = 0.56, P = 0.09	NS
All sites	Native, undisturbed	r^2 = 0.27, P < 0.01	r^2 = 0.24, P < 0.01	r^2 = 0.26, P < 0.01	r^2 = 0.23, P < 0.01	r^2 = 0.12, P = 0.06	NS
	Native, disturbed	r^2 = 0.75, P < 0.01	r^2 = 0.65, P < 0.01	r^2 = 0.23, P < 0.01	r^2 = 0.73, P < 0.01	r^2 = 0.39, P < 0.01	r^2 = 0.22, P < 0.01
	Exotic, disturbed	r^2 = 0.52, P < 0.01	r^2 = 0.29, P < 0.01	r^2 = 0.31, P < 0.01	r^2 = 0.39, P < 0.01	r^2 = 0.57, P < 0.01	r^2 = 0.28, P < 0.01
	Exotic, undisturbed	r^2 = 0.75, P = 0.06	NS	r^2 = 0.80, P = 0.04	NS	NS	NS

We then examined relationships of the pairwise traits where a significant correlation existed. First, we examined whether the slopes describing each pairwise trait combination for each plant type differed significantly within sites. Slopes for plant types within sites were found to be homogeneous in all cases (P > 0.05, Table 4). We then fitted a common slope for plant types within each site and tested whether there was a shift in elevation of the slope for each pairwise trait relationship. Within the Sydney sandstone woodland, Sydney Cumberland Plain woodland, Dorrigo subtropical rain forest and Blue Mountains woodland, there were no significant differences (P < 0.05) in slope elevation between plant types (Table 4). In the Hunter Valley riparian site, there was no significant difference in elevation of the common slope between native and exotic invasive species (both from disturbed sites) for the relationship between SLA and R_mass. However, there were significant differences in elevation of the common slope between natives and exotic invasives for the relationships between SLA and A_mass and between A_mass and R_mass. Exotic invasive species had a higher A_mass for a given SLA and a lower R_mass for a given A_mass. Overall, of the 18 trait combinations where significant correlations enabled comparisons of pairwise trait relationships, 16 showed no difference in the elevation of a common slope between exotic and native plant types (Table 4).

Table 4.   Results of standardized major axis regression analysis for all pairwise trait combinations for five sites from NSW, Australia. Significant results are shown in bold. NA indicates that there were insufficient significant correlations between the traits among plant types (Table 3) to justify analysis

Trait pair	Site	Plant types	Slope homogeneity (P)	Shift in elevation (P)	Shift along slope (P)
SLA vs. Amass	Sydney sandstone woodland	Native, undisturbed	0.680	0.272	<0.001
		Native, disturbed
		Exotic, disturbed
	Sydney Cumberland Plain woodland	Native, disturbed	0.304	0.990	0.350
		Exotic, disturbed
	Hunter Valley riparian	Native, disturbed	0.447	0.042	0.111
		Exotic, disturbed
	Dorrigo subtropical rain forest	Native, undisturbed	NA	NA	NA
		Exotic, disturbed
	Blue Mountains woodland	Native, undisturbed	0.392	0.063	<0.001
		Exotic, undisturbed
		Exotic, disturbed
SLA vs. Rmass	Sydney sandstone woodland	Native, undisturbed	0.330	0.371	<0.001
		Native, disturbed
		Exotic, disturbed
	Sydney Cumberland Plain woodland	Native, disturbed	P = 0.425	0.117	0.587
		Exotic, disturbed
	Hunter Valley riparian	Native, disturbed	P = 0.352	0.558	0.477
		Exotic, disturbed
	Dorrigo subtropical rain forest	Native, undisturbed	NA	NA	NA
		Exotic, disturbed
		Native, undisturbed
	Blue Mountains woodland	Exotic, undisturbed	NA	NA	NA
		Exotic, disturbed
SLA vs. Nmass	Sydney sandstone woodland	Native, undisturbed	0.237	0.570	<0.001
		Native, disturbed
		Exotic, disturbed
	Sydney Cumberland Plain woodland	Native, disturbed	0.613	0.363	0.523
		Exotic, disturbed
	Hunter Valley riparian	Native, disturbed	NA	NA	NA
		Exotic, disturbed
	Dorrigo subtropical rain forest	Native, undisturbed	NA	NA	NA
		Exotic, disturbed
	Blue Mountains woodland	Native, undisturbed	0.087	0.235	<0.001
		Exotic, undisturbed
		Exotic, disturbed
Amass vs. Rmass	Sydney sandstone woodland	Native, undisturbed	0.481	0.095	<0.001
		Native, disturbed
		Exotic, disturbed
	Sydney Cumberland Plain woodland	Native, disturbed	0.849	0.168	0.168
		Exotic, disturbed
	Hunter Valley riparian	Native, disturbed	0.784	0.012	0.166
		Exotic, disturbed
	Dorrigo subtropical rain forest	Native, undisturbed	0.144	0.621	<0.001
		Exotic, disturbed
	Blue Mountains woodland	Native, undisturbed	0.917	NA	NA
		Exotic, undisturbed
		Exotic, disturbed
Amass vs. Nmass	Sydney sandstone woodland	Native, undisturbed	0.235	0.641	<0.001
		Native, disturbed
		Exotic, disturbed
		Native, disturbed
	Sydney Cumberland Plain woodland	Exotic, disturbed	0.169	0.875	0.538
	Hunter Valley riparian	Native, disturbed	NA	NA	NA
		Exotic, disturbed
	Dorrigo subtropical rain forest	Native, undisturbed	NA	NA	NA
		Exotic, disturbed
	Blue Mountains woodland	Native, undisturbed	NA	NA	NA
		Exotic, undisturbed
		Exotic, disturbed
Rmass vs. Nmass	Sydney sandstone woodland	Native, undisturbed	NA	NA	NA
		Native, disturbed
		Exotic, disturbed
	Sydney Cumberland Plain woodland	Native, disturbed	0.351	0.804	0.833
		Exotic, disturbed
	Hunter Valley riparian	Native, disturbed	NA	NA	NA
		Exotic, disturbed
	Dorrigo subtropical rain forest	Native, undisturbed	0.682	0.913	<0.001
		Exotic, disturbed
	Blue Mountains woodland	Native, undisturbed	NA	NA	NA
		Exotic, undisturbed
		Exotic, disturbed

Finally, we examined whether there were significant (P < 0.05) shifts between plant types along the common fitted slope for each pairwise combination of traits within sites. In the Sydney sandstone woodland site, there were significant shifts along the common slope for all pairwise trait combinations where comparisons were possible (n = 5). For the SLA and A_mass relationship, native undisturbed species were significantly different from exotic invasive disturbed species (P < 0.001) and native disturbed species were significantly different from exotic invasive disturbed species (P = 0.035), with native undisturbed species having the lowest trait values, exotic invasive disturbed species the highest trait values and native disturbed species intermediate trait values (Fig. 1a). The pattern was the same for all other pairwise relationships in the Sydney sandstone woodland site, except that all three plant types were significantly different from each other (Figs 2a, 3a, 4a, 5a and 6a).

Figure 1.  Standardized major axis regression relationships between assimilation rate (A_mass) and specific leaf area for five sites in NSW, Australia. Plant types are native undisturbed (open triangle, dotted line), native disturbed (closed triangle, dashed line), exotic invasive undisturbed (open circle, dash-dot line), exotic invasive disturbed (closed circle, solid line).

Figure 2.  Standardized major axis regression relationships between dark respiration rate (R_mass) and specific leaf area for five sites in NSW, Australia. Plant types are native undisturbed (open triangle, dotted line), native disturbed (closed triangle, dashed line), exotic invasive undisturbed (open circle, dash-dot line), exotic invasive disturbed (closed circle, solid line).

Figure 3.  Standardized major axis regression relationships between foliar nitrogen (N_mass) and specific leaf area for five sites in NSW, Australia. Plant types are native undisturbed (open triangle, dotted line), native disturbed (closed triangle, dashed line), exotic invasive undisturbed (open circle, dash-dot line), exotic invasive disturbed (closed circle, solid line).

Figure 4.  Standardized major axis regression relationships between dark respiration rate (R_mass) and assimilation rate (A_mass) for five sites in NSW, Australia. Plant types are native undisturbed (open triangle, dotted line), native disturbed (closed triangle, dashed line), exotic invasive undisturbed (open circle, dash-dot line), exotic invasive disturbed (closed circle, solid line).

Figure 5.  Standardized major axis regression relationships between foliar nitrogen (N_mass) and assimilation rate (A_mass) for five sites in NSW, Australia. Plant types are native undisturbed (open triangle, dotted line), native disturbed (closed triangle, dashed line), exotic invasive undisturbed (open circle, dash-dot line), exotic invasive disturbed (closed circle, solid line).

Figure 6.  Standardized major axis regression relationships between foliar nitrogen (N_mass) and dark respiration rate (R_mass) for five sites in NSW, Australia. Plant types are native undisturbed (open triangle, dotted line), native disturbed (closed triangle, dashed line), exotic invasive undisturbed (open circle, dash-dot line), exotic invasive disturbed (closed circle, solid line).

In the Cumberland Plain woodland and the Hunter Valley riparian sites, there were only two plant types (native disturbed and exotic invasive disturbed) and no significant shifts between plant types along the common slope were found for any pairwise trait combination (Table 4, Figs 1b,c, 2b,c, 3b,c, 4b,c, 5b,c and 6b,c).

For the Dorrigo subtropical rain forest site, there were significant shifts along the common slope between the two plant types ‘native undisturbed’ and ‘exotic invasive disturbed’ for the relationships A_mass vs. R_mass and R_mass vs. N_mass (Table 4). For both relationships, native undisturbed species occurred at the low-trait-value end of each slope while exotic invasive disturbed species occurred at the high-trait-value end of each slope (Figs 4d and 6d).

There were three plant types in the Blue Mountains woodland: native undisturbed, exotic invasive undisturbed and exotic invasive disturbed. Pairwise trait relationships could be compared for SLA vs. A_mass, and for SLA vs. N_mass only and in both cases there were significant shifts along the slope among plant types (Table 4). For both relationships, there were significant shifts along the slope between the native undisturbed and exotic invasive disturbed species, and between the exotic invasive undisturbed and exotic invasive disturbed species (P < 0.001 in all cases, Figs 1e and 3e), but not between the native undisturbed and exotic invasive undisturbed species (SLA vs. A_massP = 0.618, SLA vs. N_massP = 0.945).

Discussion

This study has shown clearly that exotic invasive species share the same fundamental carbon capture strategy as native species, and this result was consistent for comparisons of plant types across different habitats. Leaf traits such as SLA, N_mass, A_mass and R_mass scaled positively across native and exotic invasive species, as expected from previous studies on native species only (e.g. Reich, Walters & Ellsworth 1997; Wright et al. 2004) and comparisons of native and exotic invasive species within sites (Leishman et al. 2007). We found no difference between plant types within habitats in the slopes of the relationships describing each pairwise combination of leaf traits, suggesting that there are no differences between native and exotic invasive plant species in the functional relationships that underpin their carbon economic strategy.

In general, differences between plant types were seen as shifts along the common slope describing each pairwise trait relationship, rather than as differences in slope elevations. We found differences in slope elevation in only two cases: in the Hunter Valley riparian site, exotic invasives had higher A_mass for a given SLA and lower R_mass for a given A_mass. In both these cases, this would result in greater carbon returns for exotic invasive species compared with natives. Previous studies have also shown that where slope elevations differ between natives and exotic invasives, the rate of return of carbon is always higher for exotic invasives (Gulias et al. 2003; Leishman et al. 2007).

Thus, we can consider that in general, native and exotic invasive species are positioned along a common axis of variation that describes their leaf carbon strategy, with native species generally at the lower end of the axis (i.e. having lower values of traits such as SLA, A_mass, R_mass and N_mass) compared with exotic invasive species. This is consistent with previous work by Meiners (2007) who showed that native and invasive plant species have very similar population dynamics, suggesting that they exploit similar ecological strategies within communities. However, this study has shown the importance of considering the context of disturbance, or resource availability, when comparing between plant types. We found clear differences in trait values of species occurring in undisturbed vs. disturbed locations within sites. In the Blue Mountains woodland, the Dorrigo subtropical rain forest and the Sydney sandstone woodland there were significant shifts along the common SMA slope between natives occurring in undisturbed locations and exotics occurring in disturbed locations. In the Blue Mountains and Dorrigo sites, it is clear that plant species that have traits that place them on the low end of the carbon capture strategy axis occur in undisturbed areas, irrespective of whether they are exotic or native. In the Sydney sandstone woodland, native species found in disturbed areas have traits that are significantly different from native species found in undisturbed areas. Some of these species are even considered to be invasive in other sites outside their normal range (e.g. Pittosporum undulatum and Acacia longifolia). Thus, it is the context of disturbance that is driving the trait values found in the species that occur at each location. In contrast, the Cumberland Plain woodland and Hunter Valley riparian sites included only disturbed areas, and in these sites there were no significant differences between native and exotic invasive species in the species’ leaf traits and hence their position along the carbon capture strategy axis.

Hobbs (1989) defines disturbance in the context of plant invasions as a process that creates patches of open ground or increased resource availability, where resources may be light, water or nutrients. It is generally recognized that increases in resource availability facilitate invasion (Davis, Grime & Thompson 2000; Daehler 2003) and that in situations where resource availability is high, plants that have the capacity for rapid growth will be advantaged (Grotkopp & Rejmanek 2007). Many previous studies have shown that exotic invasive species have higher leaf trait values contributing to capacity for fast growth than their native counterparts (Baruch, Ludlow & Davis 1985; Pattison, Goldstein & Ares 1998; Baruch & Goldstein 1999; Durand & Goldstein 2001; Grotkopp, Rejmanek & Rost 2002; McDowell 2002; Craine & Lee 2003; Gulias et al. 2003; Hamilton et al. 2005; Burns 2006; Grotkopp & Rejmanek 2007; Leishman et al. 2007; Pyšek & Richardson 2007; McAlpine, Jesson & Kubien 2008, but see Bellingham et al. 2004). Thus, alien species introduced into high-resource environments that have leaf traits that place them at the high end of the carbon capture strategy axis will be able to invade successfully. Similarly, native species with such attributes will be able to persist and may invade elsewhere where introduced. In such cases, relatively greater abundance of alien species may be due to the combination of disturbance, high-resource traits and reduction in natural enemies compared with coexisting native species (Blumenthal 2006).

Recent work by Funk & Vitousek (2007) suggested that the success of exotic invasive species in low-resource environments may be due to their higher resource use efficiency. In this study, we could compare photosynthetic nitrogen use efficiency (PNUE) of native and exotic species from two of the five different habitats in this study by examining the scaling relationships of N_mass and A_mass (Table 4). In the other three habitats, the correlations between N_mass and A_mass could not be examined because of non-significance of the slopes. In both the Sydney sandstone woodland and the Sydney Cumberland Plain woodland sites, there was no difference in either slope or elevation of the SMA line describing the relationship between N_mass and A_mass, suggesting that PNUE did not differ between the plant types. In the Sydney sandstone woodland vegetation, exotic species are unable to successfully invade the low-fertility soils and instead are confined to areas of nutrient enrichment from urban run-off (Lake & Leishman 2004). In the Cumberland Plain woodland vegetation, soil nutrient availability is higher (Table 1) and coexisting exotic and native species have similar traits. Thus, in these two vegetation types, higher PNUE does not explain the success of exotic species. Gulias et al. (2003) collated leaf trait data for endemic and non-endemic species of the Balearic Islands in the Mediterranean region. We re-analysed their N_mass and A_mass data (excluding crop species) using SMA regression and found that their results were consistent with ours, i.e. there was no difference between plant types in the slope of the relationship (P = 0.690) and no shift in elevation (P = 0.164) or along the common slope (P = 0.765) between plant types. Thus, across a range of habitats there is consistent evidence that PNUE does not differ between native and exotic species.

This study has shown that exotic and native species do not have fundamentally different carbon capture strategies but instead all fall along the same axis of variation that describes the leaf economics spectrum of plants. This study also showed that the position of a species along the leaf economics spectrum is strongly associated with the resource availability of the site it is found in, so that co-occurring exotic and native species have similar leaf traits. Any differences between exotics and natives may reflect differences in the environmental conditions of the sites where they occur. Sites with high resource availability will support invasion by species with traits associated with capacity for fast growth (high SLA, A_mass, R_mass and N_mass). Thus, the environmental conditions of the site, in association with the traits of the invaders, determine whether or not invasion will be successful. This idea is not new (see for example Lodge 1993; Higgins & Richardson 1998; and more recently Moles, Gruber & Bonser 2008); however, our study has provided consistent evidence from a range of habitats in support.

Acknowledgements

We thank Sarah Hill for help in the field and with species identification. We also thank NSW National Parks and Wildlife Service for advice and access to sites at Dorrigo and Scheyville National Parks, Blue Mountains City Council for advice and access to sites at Blackheath and Blue Mountains, and the Hunter Catchment Management Trust for advice and access to sites in the Hunter Valley. Two anonymous referees substantially improved the manuscript.