Alien plant invasions can occur in a range of nutritional environments including low-resource environments (Funk & Vitousek, 2007). This is also true of Australian acacias, which are often highly competitive in nutrient-poor Mediterranean-type ecosystems such as the CFR and Portuguese dune systems (Groves & di Castri, 1991). Invasive Australian acacias are able to effectively acquire nutrients and have been shown to have greater leaf N concentrations than native species in invaded regions, while P concentrations are slightly more variable (Table 5). Musil (1993) found that A. saligna also exhibited greater concentrations of K, Ca and Mg than native fynbos species. Considering that Australian acacias themselves originate from some of the most nutrient-poor soils in the world (Young & Young, 2001), it is not surprising that these species are able to effectively compete for nutrients, leading us to question whether these plants possess particular traits or mechanisms that enhance their competitive ability for nutrient acquisition and conservation.
Nutrient acquisition by plants is influenced by three major factors: root structure (including biomass, surface area and spatial distribution), soil nutrient availability and the ability of the plant to form specialized associations for nutrient acquisition (Lambers et al., 2008a). Root biomass can be preferentially allocated to enriched shallow soils and/or towards growing deeper roots to tap unused nutrient resources (Jobbágy & Jackson, 2001; Lambers et al., 2008a; Craine, 2009). As discussed earlier, invasive Australian acacias allocate a greater percentage of biomass to both deep and shallow roots in comparison with native species in invaded regions (Table 3). Acacia roots may also be more plastic in response to soil nutrient availability than other species. For example, the RMR of A. longifolia almost doubled when nutrient concentrations were reduced compared with native Mediterranean dune species Halimium halimifolium and Pinus pinea (Peperkorn et al., 2005).
The availability of soil nutrients to a plant is dependent on soil moisture and the ability of the plant to increase available nutrient concentrations through the use of root exudates. Soil moisture strongly influences the diffusive flux of nutrients into the rhizosphere. Plants can alter soil moisture by redirecting available water resources via hydraulic redistribution (Burgess et al., 1998; Hawkins et al., 2009), potentially increasing the solubility and hence availability of nutrients to plant roots (Jackson et al., 2000; Ryel, 2004; Hawkins et al., 2009). However, very little direct evidence for nutrient acquisition via hydraulic redistribution is available (Lambers et al., 2006), and this remains an untested possibility for invasive Australian acacias. Furthermore, transpirational water use by plants also drives nutrient mass flow (Barber, 1995), and transpiration is thus partially regulated by nutrient availability, particularly N (Raven et al., 2004; Cramer et al., 2008, 2009; Cernusak et al., 2010). Mass flow of nutrients requires adequate soil water to supply transpirational demand and hence operates at the expense of WUE (Barber, 1995; Tinker & Nye, 2000; Raven et al., 2004; Cramer et al., 2009). For many species, a decrease in nutrient availability decreases WUE (Raven et al., 2004), as has also been observed for A. longifolia (Peperkorn et al., 2005) suggesting that a water-nutrient trade-off may occur. The fact that water is required for both diffusive and mass-flow mobility of nutrients in soil provides a powerful explanation for the interaction of these two resources in determining plant growth.
Soil nutrient concentrations can also be altered by plants actively extracting nutrients that are not readily available, through the release of root exudates such as carboxylates and phosphatases (Lambers et al., 2008a). In the highly invaded South African CFR, well-represented families such as Proteaceae and the Restionaceae commonly produce specialized cluster roots, which increase surface area for diffusion and exudate release (Lamont, 1982; Lambers et al., 2006). Cluster roots are efficient at acquiring nutrients, particularly P from low-concentration and sparingly soluble sources (Lambers et al., 2006). Invasive Australian acacias lack cluster roots and are thus unlikely to be able to access these more recalcitrant forms of soil P. Despite this, invasive Australian acacias still compete effectively for nutrients in the intrinsically nutrient-poor soils of the CFR (Table 5). This competition may be enhanced through the ability of plants to form symbiotic mycorrhizal associations for nutrient acquisition (Lambers et al., 2008a).
Mycorrhizal associations occur in 82% of higher land plants (Brundrett, 2002) and enhance nutrient (particularly P) acquisition (Lambers et al., 2008b; Smith & Read, 2008). Both arbuscular mycorrhizas (AM) and ectomycorrhizas (EM) are able to take up soluble P from the soil, but only EM are able to chemically release P from sorbed and organic complexes (Smith & Read, 2008). Most Australian acacia species are able to form AM and possibly also EM associations (Reddell & Warren, 1987). However, the relative importance of these associations for P uptake in acacias remains unclear. Hoffman & Mitchell (1986) showed a positive correlation between AM colonization with plant biomass accumulation and P content of A. saligna seedlings in the CFR. In contrast, Rodríguez-Echeverría et al. (2009) found that despite significant colonization of A. longifolia roots by AM fungi in Mediterranean dune systems, no advantage in P acquisition was conferred. The benefits of EM and AM associations in Australian acacias must depend on the form and availability of P in the soil. The formation of mycorrhizal associations and the lack of cluster roots are likely to restrict the invasive Australian acacias to dependence on organic P and the more soluble forms of inorganic P. This inability to acquire the sparingly soluble forms of P that cluster-rooted species (particularly Proteaceae, Restionaceae and Fabaceae) of the invaded CFR do may serve to limit invasions of Australian acacias on some extremely nutrient-impoverished sandstone-derived soils of the CFR.
Australian acacias are well known for their N2-fixation abilities (Levine et al., 2003). N2-fixing associations occur in most Australian acacias (Lawrie, 1981; Lee et al., 2006), which usually nodulate with common, but slow-growing Bradyrhizobium species (Lafay & Burdon, 2001; Rodríguez-Echeverría et al., 2011). Associations with other nodulating species have also been reported, including Rhizobium, Ensifer, Mesorhizobium, Burkholderia, Phyllobacterium and Devosia species (Marsudi et al., 1999; Lafay & Burdon, 2001; Hoque et al., 2011). Associations between acacias and their nodulating symbionts are highly complex and can be influenced by several biotic and abiotic factors (Thrall et al., 2000, 2007; Murray et al., 2001; Rodríguez-Echeverría et al., 2011). Nonetheless, invasive Australian acacias nodulate readily in both their native and non-native regions (reviewed in this volume by Rodríguez-Echeverría et al., 2011) and are considered prolific N2-fixing species (Lawrie, 1981). In coastal dunes of Portugal, A. longifolia was more efficient at forming symbiotic associations with bacteria and fixed greater amounts of N than other co-occurring N2-fixing legumes (Ulex eurpaeus and Cytisus grandiflorus; Rodríguez-Echeverría et al., 2009). Similarly, comparing the δ15N of N2-fixing plants to others with N2 fixation disrupted by O2 fumigation, Stock et al. (1995) found that A. saligna in the CFR relied almost completely on symbiotic N2 fixation, while A. cyclops growing on slightly more nutrient-rich soil obtained only 51% of its N budget from N2 fixation. The long-term post-fire persistence of invasive Australian acacias in the CFR is somewhat puzzling because few native N2-fixing legumes (especially reseeders) persist beyond their post-fire dominance (Kruger, 1983; Hoffmann et al., 1987; Cocks, 1994; Cramer, 2010). This lack of indigenous legume reseeder persistence has been ascribed to the post-fire decline in P availability (Power et al., 2010). These authors suggested that deep roots and excessive water consumption may contribute to Australian acacia persistence.
The N2-fixing capabilities of Australian acacias and their ability to persist in invaded regions result in a substantial inputs of N-enriched litter, leading to an elevated soil N status (Table 6). However, the ability of an invader to fix N2 in itself does not necessarily translate to immediate alteration of the invaded system’s nutrient cycling (Corbin & D’Antonio, 2004). Instead, Yelenik et al. (2007) demonstrated that with Australian acacias, the combination of N2 fixation coupled with the slow decomposition rates associated with sclerophyllous phyllodes led to elevated soil N pools with long-term impacts for ecosystem nutrient cycling. Australian acacias are thus strong ecosystem engineers, and the lasting legacy of increased soil N following Australian acacia invasion often results in reinvasion by the same or other alien species (Stock et al., 1995; Marchante et al., 2004, 2008, 2009; Yelenik et al., 2004).
Table 6. Litter biomass, litter N concentrations and soil N concentrations of Australian acacia invasions compared to uninvaded native vegetation in Portugal and in the Cape Floristic Region (CFR). Data for longer (20+ years) and shorter (10 years) invasion periods are shown for Portugal. A + indicates a significantly (P < 0.05) greater value associated with Australian acacias in comparison with native vegetation, 0 indicates no significant difference and ND indicates no data available.
|A. cyclops||CFR||+||+||+||Witkowski, 1991b|
|CFR||ND||ND||+||Stock et al., 1995|
|A. longifolia||Portugal (20+ years)||+||+||+||Marchante et al., 2008|
|Portugal (10 years)||+||+||0|
|Portugal||ND||ND||+||Rodríguez-Echeverría et al., 2009|
|A. saligna||CFR||+||+||+||Witkowski, 1991b|
|CFR||ND||ND||+||Stock et al., 1995|
|CFR||+||+||+||Yelenik et al., 2004, 2007|
The sclerophyllous nature of Australian acacia phyllodes translates to long-lived leaves and evergreen trees (Loveless, 1961; Turner, 1994a). The evolutionary drivers for this adaptation, whether drought tolerance or nutrient conservation, have been subject to much debate (Givnish, 1979; Turner, 1994b; Pasquet-Kok et al., 2010). In nutrient-rich environments, the common drought-tolerance adaptation is drought deciduousness (Mooney & Dunn, 1970). However, in nutrient-poor environments, drought deciduousness would lead to the costly loss of limited nutrients. Thus, it is thought that sclerophyllous, long-lived phyllodes evolved to enhance nutrient conservation in response to nutrient limitations (Beadle, 1966; Specht & Rundel, 1990) with drought tolerance and unpalatability being associated with the sclerophyllous nature of phyllodes.
Extended leaf longevity of Australian acacias would, however, not be a marked advantage when invading other sclerophyllous vegetation with similar nutrient-retention characteristics. For example, leaf longevity of Australian acacias (mean years ± SE; 1.84 ± 0.28; Wright et al., 2002) did not differ significantly (P > 0.05) from that of native CFR vegetation (mean ± SE; 2.62 ± 0.31; Midgley & Enright, 2000). Sclerophylly, although not different from that of the invaded flora, when coupled with other traits such as N2 fixation may contribute to the success of Australian acacias. Interestingly, the non-phyllodinous and relatively non-sclerophyllous (i.e. high SLA) invasive Australian acacia, A. mearnsii, has particularly long-lived bipinnate leaves, which turn brown during drought but recover subsequent to the onset of rain (Orians & Milewski, 2007) possibly acting to conserve nutrients over multiple seasons.
Heteroblasty thus confers the advantage of different growth strategies between juvenile and adult life stages and between different environmental circumstances (Pasquet-Kok et al., 2010). As young seedlings, acacias benefit from the high relative growth rate associated with bipinnate leaflets (Witkowski, 1991b; Hansen, 1996; Evans et al., 2000; Pasquet-Kok et al., 2010). The phyllodinous species then switch to slower-growing, longer-lived and hence nutrient-conserving phyllodes (Ullmann, 1989; Orians & Milewski, 2007; Pasquet-Kok et al., 2010). Using acacia invasions in South Africa as a case study, the distinct advantage of phyllodes in nutrient-poor and summer-drought regions can be inferred by the relative success of phyllodinous species in the mediterranean climate and nutrient-poor fynbos biome (Rouget et al., 2004; Table 7, e.g. A. pycnantha). In contrast, the non-phyllodinous species (e.g. A. mearnsii and A. dealbata) are more successful as invaders in more mesic environments or along water courses (Rouget et al., 2004; Table 7) where nutrients and water are not as limiting.
Table 7. Percentage of records of the eight most widespread invasive Australian acacia species found in each biome in South Africa. Species are ranked from most prevalent to least prevalent according to the percentage of quarter degree squares occupied, as recorded in the South African Plant Invaders Atlas (SAPIA; Henderson, 2007). The percentage of the total records that were found along water courses is also listed. Foliage indicates whether adult plants have leaves (L) or phyllodes (P). The biome in which each species had the highest occurrence is in bold.
Plants can also conserve nutrients through the remobilization of limiting nutrients prior to leaf abscission (Eckstein et al., 1999; Wright et al., 2002), acting to increase the mean residence time of nutrients in the plant. Australian acacias remobilize nutrients prior to leaf abscission, especially when the specific nutrient is limiting in the system (Witkowski, 1991a). In the South African CFR, A. saligna remobilized a large proportion (71%) of its leaf P, an amount significantly greater than that of the comparison native species Leucospermum parile (48%; Witkowski, 1991a). However, studies assessing remobilization efficiencies of these plants in comparison with natives in invaded regions are scarce. Specht (1981) and Langkamp & Dalling (1982) showed that remobilization of nutrients by invasive Australian acacias was not particularly different to that of other Australian species from nutrient-impoverished areas (e.g. Banksia ornata and Acacia holosericea) and is thus not a trait unique to the invasive Australian acacias.