- Research article
- Open Access
A test of native plant adaptation more than one century after introduction of the invasive Carpobrotus edulis to the NW Iberian Peninsula
BMC Ecology and Evolution volume 21, Article number: 69 (2021)
Although the immediate consequences of biological invasions on ecosystems and conservation have been widely studied, the long-term effects remain unclear. Invaders can either cause the extinction of native species or become integrated in the new ecosystems, thus increasing the diversity of these ecosystems and the services that they provide. The final balance of invasions will depend on how the invaders and native plants co-evolve. For a better understanding of such co-evolution, case studies that consider the changes that occur in both invasive and native species long after the introduction of the invader are especially valuable. In this work, we studied the ecological consequences of the more than one century old invasion of NW Iberia by the African plant Carpobrotus edulis. We conducted a common garden experiment to compare the reciprocal effects of competition between Carpobrotus plants from the invaded area or from the native African range and two native Iberian plant species (Artemisia crithmifolia and Helichrysum picardii) from populations exposed or unexposed to the invader.
Exposure of H. picardii populations to C. edulis increased their capacity to repress the growth of Carpobrotus. The repression specifically affected the Carpobrotus from the invader populations, not those from the African native area. No effects of exposition were detected in the case of A. crithmifolia. C. edulis plants from the invader populations had higher growth than plants from the species' African area of origin.
We found that adaptive responses of natives to invaders can occur in the long term, but we only found evidence for adaptive responses in one of the two species studied. This might be explained by known differences between the two species in the structure of genetic variance and gene flow between subpopulations. The overall changes observed in the invader Carpobrotus are consistent with adaptation after invasion.
The large-scale alteration of species distributions is one of the most drastic types of disturbance to the biosphere that have occurred during the Anthropocene [1, 2]. Although global biodiversity is being eroded [3, 4], it may be increasing at smaller spatial scales due to the arrival of invasive species [5, 6]. The long-term ecological consequences of these invasions are unclear. While some non-native species can outcompete native species to extinction [7,8,9], others may cause no serious adverse impacts (as is often the case, at least in the short-term; see [10, 11], giving native species the opportunity to co-evolve with the invaders (reviewed in Oduor et al. ) and even to develop new mutualisms . In this way, invasive species might eventually become stably integrated in the new ecosystems [14, 15], increasing local biodiversity and reinforcing the services provided by these ecosystems or their resilience to further alteration (reviewed in Chapman et al. ; but see Kaiser-Bunbury et al. ).
Analyses of the evolutionary processes that could result in this final integration of the invasive species in the long term are relatively scarce , but short-term experimental findings consistent with evolutionary change in invasive species, leading to divergence in relevant adaptive traits from their source populations [19, 20], are accumulating [21, 22]. Such findings run from increases in invasive ability [23,24,25], to changes in interactions with other species [26, 27], and responses to abiotic factors [28, 29]. Invasive species have also been shown to induce short term evolutionary changes in native species [30,31,32,33,34,35,36], reviewed in Oduor et al. . However, these short-term changes may be poor guides to predicting the properties of future ecosystems [37, 38], because biological invasions may alter the physical ecosystem, species composition and abundance, favouring the establishment of other invasive species [39, 40] and triggering a cascade of coevolutionary, multi-species processes [18, 41], all of which may take some considerable time [37, 42, 43]. Thus, studies of biological invasions are most informative when they consider the possible changes in both the invasive species and native plant communities, and when a long time has elapsed since the introduction [44,45,46,47].
In this study, we explored if native dune species have evolved adaptative responses as a consequence of the interaction with the invasive South African species Carpobrotus edulis (L.) N.E.Br. (Aizoaceae) (hereafter Carpobrotus), introduced at least one century ago to NW Iberia. In a common garden experiment, we compared the reciprocal effects of competition between Carpobrotus plants from either European (invader) populations or from native African populations and the native Iberian species Artemisia crithmifolia L. (A. campestris L. ssp. maritima (DC.) Arcang.; hereafter Artemisia) and Helichrysum picardii Boiss. & Reuter (Helichrysum serotinun subsp. picardii (Boiss & Reuter) Galbany, L. Sáez & Benedí; hereafter Helichrysum). These are among the most representative endemic species from secondary or grey dunes, one of the main habitats invaded by Carpobrotus in NW Iberia. The sampled native plants were from populations that had either already been exposed to European Carpobrotus and have therefore had the opportunity to co-evolve with it (exposed populations) or they were from populations from the same region that had not been exposed to Carpobrotus. Considering the time elapsed since the introduction of Carpobrotus to the NW Iberian Peninsula, we hypothesized that it may have genetically changed to adapt to the new conditions. This same old invasion, along with the strong selection pressures exerted by an invader able to establish monodominant stands, may have resulted in natives' adaptions reducing the impact of the invader and easing the future development of a stable, biodiverse community.
No Carpobrotus or Artemisia plants died in the competition pots, whereas two exposed and two unexposed Helichrysum plants competing with the African Carpobrotus died, as did three exposed and three unexposed Helichrysum plants competing with the European Carpobrotus. We show in more detail the analyses corresponding to final whole plant dry mass, hereafter “growth” in Fig. 1 and Table 1 for the competition pots, and Fig. 2, Additional file 1: Table S1 and Additional file 2: Table S2 for the comparisons between competition and single plant pots. The corresponding results for shoot, root and whole plant dry mass were qualitatively similar to those for growth and are shown in Additional file 3: Table S3, Additional file 4: Table S4 and Additional file 5: Table S5.
Exposure of native Iberian species to Carpobrotus
Exposure was not significant in either the native or Carpobrotus plants (Table 1). However, in the case of Carpobrotus growth, this was due to heterogeneity of results across Carpobrotus origin and native species. We found a significant Exposure × Origin of Carpobrotus effect of Helichrysum on Carpobrotus, but not of Artemisia. The African Carpobrotus grew more when competing with the previously exposed Helichrysum: analysis of data from pots containing Helichrysum/African Carpobrotus detected a significant (LRT P = 0.024) and positive effect of Exposure. The difference occurred in the opposite direction in the corresponding analysis for European Carpobrotus (Fig. 1, LRT P = 0.174). This variation resulted in a significant (LRT P = 0.021) Exposure × Origin of Carpobrotus interaction in an analysis restricted to pots shared by either Carpobrotus with Helichrysum. Thus, the exposed Helichrysum suppressed more the growth of the European Carpobrotus with which it had the opportunity to co-evolve. The corresponding analysis for Artemisia did not detect any such interaction (LRT P = 0.600), and the difference between native species for double interactions resulted in a significant triple Exposure x Origin of Carpobrotus × Native species interaction in the full model (i.e., in the joint analysis of all competition pots in the experiment; Fig. 1 and Table 1). Thus, the potentially coevolved Helichrysum had stronger effects on the growth of European Carpobrotus than the potentially coevolved Artemisia. No effects of Exposure or the corresponding interactions were observed in these comparisons of native plants in competition and single plant pots (Fig. 2 and Additional file 1: Table S1).
Origin of Carpobrotus
The African Carpobrotus grew less than the European Carpobrotus. This was shown by the analysis of the competition pots (Fig. 1 and Table 1), the competition/single Carpobrotus plant pot comparisons (LRT P = 0.002, Additional file 2: Table S2 and Fig. 2) and also by the comparison of the African and European Carpobrotus grown in single plant pots (LRT P < 0.001). The Origin of Carpobrotus had no significant overall effect on the growth of the competing native plants, due to the heterogeneous growth of these plants. Helichrysum grew relatively more (on average for exposed and unexposed plants) in the presence of the African Carpobrotus than Artemisia (Fig. 1), as indicated by the significant Origin of Carpobrotus × Native species interaction (Table 1).
The Artemisia plants grew more than the Helichrysum plants, as seen in the competition pot comparisons (Fig. 1 and Table 1) and confirmed by the comparisons between competition and single plant pots (Fig. 2 and Additional file 1: Table S1) and by the direct comparison of growth of each species in the single plant pots (LRT P < 0.001). The effect on Carpobrotus growth was also significant: the Carpobrotus plants competing with the larger Artemisia grew less than those competing with the smaller Helichrysum (Fig. 1 and Table 1).
Both African and European Carpobrotus plants grew more when competing with Helichrysum, the smallest of the two native species studied (see Methods section), and therefore that expected to generate less competition for resources in the pots. This was consistent with competition limiting plant growth in this experiment. The bidimensional representation of the least square means from the analysis of the competition pots (Fig. 1) would support this view in the case of Helichrysum: the estimated correlation between mean growth of Helichrysum and Carpobrotus in the same pot was negative. This contrasted markedly with the positive sign of the corresponding estimate for Artemisia. As the power to detect four-point correlations is low, these two estimates were not significantly different from zero when tested separately. However, randomizing the allocation of pairs of standardized least square means to the two species resulted in only 144 replicates on 10,000 with larger than observed between-species differences in correlation, showing that the correlations between Carpobrotus and native plants were significantly (P = 0.014) different for the two species considered.
The residuals obtained after fitting the analytical model to the competition pots (Fig. 1) showed that some within-pot competition remained after correcting for the effect of Exposure, Origin of Carpobrotus and Native species. The correlation between the residuals of the Carpobrotus and native plant analyses were − 0.553 (P = 0.005, 22 d. f.) for Artemisia and − 0.558 (P = 0.048, 11 d. f.) for Helichrysum. The effect was not homogeneous in Helichrysum (Fig. 1). A separate estimate of this correlation for the competition pots with African Carpobrotus was positive (r = 0.176, P = 0.705, 5 d. f.), and another for the competition pots with European Carpobrotus was negative (r = − 0.883, P = 0.020, 4 d. f.), consistent with stronger competition between Helichrysum and the European Carpobrotus. A test randomizing the allocation of plant pair means to the two groups (of competition pots with one Helichrysum and one African or European Carpobrotus) detected only 211 of 10,000 replicates with more extreme differences in correlation. While this test was not planned a priori, the difference between the two correlations was remarkable. We found no evidence of such heterogeneity in the Artemisia pots, and the correlations were − 0.716 (P = 0.009, 10 d. f.) and − 0.278 (P = 0.381, 10 d. f.) in the pots with competing African and European Carpobrotus. The randomization test detected 1985 replicates of 10,000 with more extreme differences in correlation than observed between the two groups of Artemisia pots. There were no significant differences between the residual's correlations of exposed and unexposed weights in any species (P = 0.142 in a joint randomization test for both native species). The two native species’ contrasting correlations with Carpobrotus, both for lsmeans and residuals, suggest that their patterns, and possibly mechanisms of competition with Carpobrotus were different.
The Presence of Carpobrotus significantly depressed the growth of the native plants in the comparison of competition and single plant pots, whereas Presence of native plants did not significantly affect the growth of Carpobrotus (Fig. 2, Additional file 1: Table S1; this comparison did not use the unexposed natives; see Methods). In the same analysis, the significant interaction (LRT P = 0.044) between Presence of Carpobrotus and Native species again supports increased competition between Helichrysum and Carpobrotus.
The total biomass in the competition pots (i.e., the sum of the weight of native plants and Carpobrotus plants) was greater in the pots containing European Carpobrotus (Origin of Carpobrotus, LRT based on the same model as used for the weights of native Iberian species and Carpobrotus P < 0.001; Fig. 3) but there were no differences between competition pots containing exposed and unexposed native plants.
The two native species showed different patterns of Exposure × Origin of Carpobrotus interaction, which was consistent with differences in adaptative responses to the invader. We had no evidence of such interaction, and therefore response, in Artemisia. Our experiment compared in a common greenhouse environment plants from two populations of each native species, one exposed and the other unexposed to Carpobrotus. It would be parsimonious to attribute any overall differences between these two populations to the random sampling of each species' interpopulation variation, and this variation could be the result of many processes, like local abiotic adaption or genetic drift, besides adaptation to Carpobrotus presence. However, we found no such overall differences -i.e., no significant main factor Exposure- either for the competitive effect of the native plants on Carpobrotus, or their response to Carpobrotus. The only effects of Exposure were specific for each origin of the competing Carpobrotus, African or invader. They were detected by the significant interaction Exposure × Origin of Carpobrotus, where the strong interaction for Helychrysum prevailed over the small or inexistent interaction for Artemisia, and the triple interaction Exposure × Origin of Carpobrotus × Native species, reflecting this heterogeneity between native species for the Exposure × Origin of Carpobrotus interaction. The specificity of these Exposure effects for the two origins of Carpobrotus makes the mere sampling of interpopulation variation a less parsimonious interpretation and clearly suggests adaptation of Helichrysum to the Carpobrotus invasion.
The relative increase in growth for the African Carpobrotus competing with the exposed Helichrysum suggests that this native’s response to the European Carpobrotus involves costs that reduce its performance when the invader is absent. In another potentially costly adaptation, fitness of populations of the native Pilea pumila exposed to the invader Alliaria petiolata was maximal in sites with high densities of the invader and minimal in the low-density sites , indicating that adaptation to interaction with invasive species may become counterproductive when the invaders are rare or absent.
The change observed in the exposed Helichrysum does not fit the predictions of the Atwater's model  of plant invasion and the observations by Fletcher et al. . According to that model, of the two components of plant competitive ability defined by Miller and Werner , namely the ability of an individual plant to suppress competitors and the ability to tolerate them, competition among more than two individuals or species would favour the evolution of tolerance instead of suppression. This is because increased tolerance benefits only the species experiencing it, whereas increased suppression of some competitors would also benefit all no suppressed species in the competing assemblage. Consequently, the reduction in overall competition for the suppressor would be limited. The situation could be different in our experiment due to the asymmetry of the competition. Carpobrotus is a very successful invader and it could be difficult for native competitors to completely fill the void left by a suppression of its competitive effects. So, the Helicrysum plants could get a net benefit by trading Carpobrotus for other natives’ competition. However, no significant increase in growth was detected in the exposed Helichrysum competing with the European Carpobrotus.
The observed relative decrease in Carpobrotus growth by the exposed Helichrysum was modest and did not prevent the European Carpobrotus from becoming larger than the African plant. The larger mass was consistent with the trend of plants becoming taller and more vigorous when they grow in non-native environments  and with the evolution of increased competitive ability hypothesis : plants will show trade-offs in resource allocation to growth, reproduction and defence [53, 54], so that release from the biological enemies in their native environment will enable the invaders to increase their investment in other traits. In that case, it would be remarkable that Carpobrotus had maintained such release for so many years since its introduction. It is possible that the absence of native plants phylogenetically close (see ) to Carpobrotus in the NW Iberian Peninsula made it difficult for local phytophagous and pathogenic species to extend the range of exploited plants to encompass the newcomer.
The comparison of competition and single plant pots revealed high levels of competition in the pots containing two plants for both native species. Despite this competition, the greater dry mass of the European Carpobrotus was not obtained completely at the expense of the native plants, as the total biomass in the pots containing one native and one European Carpobrotus plant was higher than in those containing one native and one African Carpobrotus plant. This raises the possibility that the primary productivity of plant communities may have increased since the introduction of Carpobrotus. It must be noted however that conditions in the greenhouse cannot perfectly reproduce those in the field. For example, we tried to maintain all pots optimally watered thorough the experiment, thus excluding root competition for water, which could play some role in the interspecific competition in the field. Similarly, the regular arrangement of plants in the pots could not fully represent the irregular plant distribution and density observed in the dunes. However, as seen in Fig. 4, distances between plants in the field may be as short as in our experimental pots (uncharacteristically isolated plants were chosen for the pictures of non-exposed plants to improve visibility).
Two kinds of competitive interactions would have occurred in the pots. First, exploitation or scramble competition, where mineral soil resource availability to competitors is affected through resource depletion, and second, interference or contest competition, where interaction occurs through the production and release into the soil of chemicals that are toxic to other species or inhibit access of other roots to resources (allelopathy). Some previous studies have demonstrated that both mechanisms play a key role driving the competition between Carpobrotus and native species in the field . The correlation between the lsmeans of native and Carpobrotus plants in the same competition pot provided some evidence for interference. Because Artemisia is the largest of the two natives and the one suppressing Carpobrotus growth the most, it would be expected to be involved in stronger resource competition and more negative lsmeans correlations with Carpobrotus: pots with large Artemisia plants would be expected to sustain smaller Carpobrotus, and vice versa. But it was the reverse. The correlation was significantly less negative than that for Helichrysum. This reduced dependency of competition on plant size in Artemisia would be consistent with this species’ biology. While there is some evidence of allelopathic properties in the genus Helichrysum , direct comparison of allelopathic activities in plants of the Helicrysum and Artemisia genera  revealed a clear advantage in the activity of the latter. Allelopathy could thus explain a depression in Carpobrotus growth that is not dependent on the variation in size of Artemisia, as there could be differences in the regulation of plant growth and allelopathic activity. In fact, trade-offs between growth and production of allelopathic compounds have been found, at least in seaweeds .
Some evidence of competition remained after adjusting for the main effects Exposure and Origin of Carpobrotus and their interactions in the analytical model, as shown by the mainly negative correlations between the residuals for Carpobrotus and native species from the model adjusted to analyse their growth. These correlations were consistently negative for the Artemisia data, thus indicating that, although not dependent on the Exposure or Origin of Carpobrotus considered in that model, competition for limited resources in the pots occurred between Artemisia and Carpobrotus. In any case, the observation that the presence of Carpobrotus generally depressed the growth of Helichrysum more than that of Artemisia in the comparison of competition and single plant pots suggested more intense competition for resources. This could have resulted in stronger selection pressures on Helichrysum populations to adapt to the presence of Carpobrotus. Differences in the competitive impact of Carpobrotus across native species had already been observed in field studies of invaded areas [60, 61].
The difference in the response of the native species to the introduction of Carpobrotus could be related also to the genetic structure of the populations of these species. Artemisia displays very limited genetic variation, both between and within populations, in the study region, probably due to its ability to disperse over long-distances at high rates, and to initiate new populations from very small propagules, in a series of founder events . The low variation will limit the potential of exposed subpopulations to adapt to competition from the invasive Carpobrotus. By contrast, Helichrysum italicum has been shown to maintain considerable genetic differences between subpopulations at distances of only tens of kilometres in Sardinia, probably due to limited mobility of pollinating insects , and considerable variation within populations, at least in the Western Mediterranean . Similar partial isolation could facilitate the local evolution of resistance in areas of the NW Iberian Peninsula invaded by Carpobrotus. Interestingly, Helichrysum italicumm subsp. picardii was the second most abundant native species in a study of 8 sites invaded by Carpobrotus in the sand dune systems of the western coast of Portugal (with an average cover of 6.0%, compared with 6.6% for Corema album, 13.9% for Carpobrotus edulis and the only Artemisia species mentioned (Artemisia campestris ssp. maritima) was in eighth position, with 1.7% ). A large population size favours the maintenance of genetic variation, which would also facilitate the evolution of resistance to Carpobrotus. However, it is not clear whether the large populations of Helichrysum in the Carpobrotus-invaded sites are the cause or the consequence of that evolution, as we are not aware of any comparison of Helichrysum abundance in invaded and non-invaded sites.
These evolutionary considerations may be useful additions to the list of criteria for assessing the vulnerability of native species and ecosystems to biological invasions, on which to base the assignment of priorities for surveillance and protection interventions . These assessments tend to be based on ecological features (e.g. [67, 68], but our study suggests that vulnerability to biological invasions may also depend on the genetic structure of populations, the amount of genetic diversity and the gene flow patterns. The same factors could also be important for designing management plans for invasive species .
In conclusion, we found that native species can respond in order to reduce the ecological impact of invasive species, which would facilitate the integration of the latter into the invaded community. This result is consistent with previous studies of ancient introductions of mussel macroparasites (~ 70 years ), herbaceous plants (~ 150 years ) and trees (~ 170 years ). However, these changes may only occur in some native species, possibly depending not only on ecological, but also on evolutionary aspects such as population size and the amounts of genetic variance and gene flow. We propose that consideration of these aspects may be important in analysing the conservation impact of biological invasions. The heterogeneity in native plant responses to invasion might help to explain why Carpobrotus is still having a strong impact across its area of distribution in NW Iberia.
Introduction of Carpobrotus edulis
Carpobrotus edulis is a succulent perennial plant that has been introduced from its native range in South Africa  across all Mediterranean climate regions, including California, Australia and the Mediterranean basin . In Europe, this species has been grown for ornamental purposes since the beginning of the seventeenth century  and records of its presence in NW Iberia date back to the eighteenth century . Due to its ability to rapidly spread forming deep, dense mats, the species has been used to stabilize sand dunes and prevent soil erosion in this area since the early twentieth century and nowadays naturalized populations of C. edulis can be found elsewhere in coastal habitats , where it may have been co-evolving with native plants for more than 100 years. Its facultative C3-CAM physiology , high morphological and ecophysiological plasticity [76,77,78], flexible mating system , and an intense vegetative clonality [20, 73, 80], enable the plant to tolerate a wide range of ecological conditions. These characteristics along with the high rates of seed dispersal , are also important features explaining the effective colonization of dune habitats, where plants compete for space, light, water and nutrients  in such a way that C. edulis can reduce the growth, survival, and reproduction of some native species  and references therein]. Consequently, the release of Capobrotus in natural environments and protected areas is prohibited in several countries (e.g., Spain, Portugal, United Kingdom, Ireland, and Italy), although this taxon is not included in the Regulation (EU) no. 1143/2014 . In California, the plant poses a threat to several rare and endangered plant species and it is listed as CalEPPC List A-1 and as CDFA-NL (http://www.cal-ipc.org/); on the contrary it is not declared or considered noxious by any state government authorities in Australia .
We collected Carpobrotus plants from sand dune populations in their South African native range (Hawston beach, 34° 23′S, 19° 07′W, Western Cape, South Africa) in mid-January 2015 and in the invaded range (Praia de Moledo, 41º 51′N, 8º 51′W, Caminha, Portugal) in mid-April 2015. The African specimens of Carpobrotus were used as an experimental control as they share origins with the invasive European Carpobrotus, but not their recent adaptive story. The Artemisia crithmifolia and Helichrysum picardii plants exposed to Carpobrotus were collected from Caminha (Portugal) at the same time as the Carpobrotus plants.
The two native species selected in the experiment differ in several respects relevant for their evolutionary responses to biological invasion. Artemisia is a larger plant (see the two species’ descriptions in Castroviejo et al. ) with more allelopathic activity  and lower genetic variation [62, 63] and ground cover in the sampled sand dunes . Artemisia, but not Helichrysum, has rhizomatous structures allowing to optimize belowground resource uptake and storage, which could increase its competitive ability. The Artemisia and Helichrysum plants from populations unexposed to Carpobrotus were collected in mid-March 2016, in Camariñas (Praia do Trece, [43° 11′N, 9º 10′W], Galicia) and Porto do Son (Praia das Furnas [42° 38′N, 9º 02′W], Galicia), respectively (Fig. 4). The population of Artemisia in Porto do Son and the population of Helichrysum in Camariñas were too small to take a representative sample of both species from one site. Thereby, unexposed Artemisia and Helichrysum plants were collected in Camariñas and Porto do Son, respectively. The distance between these populations is about 76 km and the environmental conditions are quite similar. The monthly mean temperatures registered at the meteorological station of Camariñas ranged from 11 ºC (March 2016) to 18.5 ºC (August 2016) and monthly mean rainfall ranged from 397.1 L/m2 (January 2016) to 1.3 L/m2 (July 2016). The monthly mean temperatures registered at the meteorological station of Ribeira, a place very near to Porto do Son, ranged from 10.9 ºC (February 2016) to 21.5 ºC (July 2016) and monthly mean rainfall ranged from 265.7 L/m2 (January 2016) to 4.8 L/m2 (July 2016) (www.meteogalicia.es). No direct measures of the extent of Carpobrotus plant cover are available for the exposed sampling sites, but visual estimates based on pictures taken during the sampling were of about 60–70% in both sites.
We sampled intensively the native and invasive Carpobrotus populations. To have a more comprehensive representation of the genetic variability of the species in each area, we selected 36 separated clumps per area. The minimum separation between sampled clumps was of 25 m. Carpobrotus forms compact clumps  and it is reasonable to assume that each separated clump represents a different genotype. Thus, we would have collected a total of 72 genotypes. From these, we randomly selected the genotypes used in our experiment. Our sampling protocol has been described in detail in Roiloa et al. . Likewise, the populations of the two native species were extensively sampled in order to gather the greatest genetic diversity inside each population.
The plant taxa nomenclature follows standard Iberian floras . The native species were collected following current Spanish regulations. No specific permissions were required. The invasive C. edulis was collected from natural populations and propagated under the permission from the Spanish Ministry of Agriculture, Food, and the Environment, and complied with the Convention on the Trade in Endangered Species of Wild Fauna and Flora.
We found no historical records (including records about past eradication campaigns, which mainly began in the twenty-first century in NW Iberia; ) of the presence of Carpobrotus in our unexposed locations of Camariñas and Porto do Son. In any case, a previous, undocumented presence of Carpobrotus in the locations -both currently free of Carpobrotus - would imply that the species had become extinct, which is unlikely given its invasive nature. Collected plants were washed and maintained in a climate-controlled greenhouse at the University of Santiago de Compostela until the start of the experiment in April 2016.
The experiment was carried out in a greenhouse at the University of Santiago de Compostela (Galicia, Spain). We used 5L plastic pots filled with a growing substrate similar to that in natural conditions, i.e., a 1:1 mixture of potting compost and dune sand. The environmental conditions were identical for all species, grown under a natural day/night light cycle between April 2016 and April 2017. Monthly global irradiance ranged from 15.3 MJ m−2 day−1 (April 2016) to 21.3 MJ m−2 day−1 (April 2017), although photosynthetically active radiation was reduced by about 12% inside the greenhouse with respect to full sunlight outdoors (measured with a LI-190SA Quantum Sensor, LI-COR, Lincoln, Nebraska, USA). The temperature inside the greenhouse ranged from 15 ℃ to 22 ℃. The plants were watered according to their requirements (once or twice per week) in order to prevent hydric stress. Additionally, to avoid confounding effects of pot position within the greenhouse, these positions were randomized monthly.
The basic units in our experiment were competition pots containing one native plant and one Carpobrotus genet. Each genet, obtained from different donor plants, was composed by the three-vegetative most apical ramets (i.e., modules sensu Harper ), to guarantee that all of the material was at the same development stage. The plant pairs in these competition pots were arranged in a factorial design (Fig. 5) considering the effects of prior exposure of native plants to Carpobrotus (Exposure: exposed/unexposed), origin of Carpobrotus plants (African/European), native Iberian plant species (Artemisia/Helichrysum) and their interactions on the competition between the two plants (six replicate pots per combination of factors: 6 × 2 × 2 × 2 = 48 pots; 96 plants). The two-plant competition pots were complemented with pots containing single plants (six pots for African Carpobrotus, six for European Carpobrotus, and six for each Exposure × Native species combination: 36 pots and plants). Comparisons between single plant and competition pots made possible to confirm the existence, and measure the intensity, of competition experienced by Carpobrotus and native plants in the competition pots.
Before the start of the experiment, the initial fresh weight of each Carpobrotus, Artemisia and Helichrysum plant was measured to the nearest 0.0001 g (Mettler AJ100, Mettler-Toledo, Greifensee, Switzerland). The plants were grown for twelve months under the experimental conditions and were then harvested, washed, cleaned and dried at 60 °C to constant weight. Each plant was separated into shoots (including leaves and stolons in the case of Carpobrotus) and roots, and the final dry weights of each fraction (total, above ground and root, dry mass) were recorded.
All statistical analyses were based on linear models. The data sets were unbalanced, as some plants did not survive up to the end of the experiment, and we used the R  package stats’ “glm()”, “drop1()” and “emmeans()” functions to carry out likelihood ratio tests (LRT) and calculate least square means (hereafter “lsmeans”) for the mass data. We complemented the LRTs with more intuitive, Akaike weight-based calculations of the probability of one model favouring the other . The difference in AIC between model i and the best model, i.e., that with minimum AIC, is
∆i(AIC) = AICi – minAIC
and the weight of model i:
where K is the number of models considered. The normalized probability that model 1 is preferred (i.e., it is better in terms of Kullback–Leibler discrepancy; see ) over model 2 is
Because the numbers of parameters in the statistical models considered were large relative to sample sizes, we replaced AIC with its small sample version AICc  in all calculations shown.
Both the analyses of the measures from the native plants and the Carpobrotus plants in the competition pots considered the effects of Exposure, Origin of Carpobrotus, Native species and their interactions. This was because, in a competition situation, the characteristics of one plant may affect the plant competing with it. For example, African and European Carpobrotus could have different effects on the native plant in the same pot. For the same reason, both analyses considered as covariables the initial fresh mass of the native and the Carpobrotus plants.
The native Iberian species growing with South African Carpobrotus were not considered in the comparisons between the native plants in competition and single plant pots, because this would have yielded heterogeneous and difficult-to-interpret levels for the origin of Carpobrotus factor (European, African and none). The effects considered in these analyses were Exposure to Carpobrotus, Native species and the Presence (presence/absence) of Carpobrotus in the pot. Only the initial fresh mass of the native plants was used as a covariable here, as only half of the pots (those containing two plants) contained a Carpobrotus plant for which an initial weight was available. Similarly, the Carpobrotus plants growing with non-exposed native plants were not considered in the comparisons between Carpobrotus in competition and single plant pots, to prevent heterogeneous and difficult-to-interpret levels for the Exposure factor (exposed, unexposed and no native plant). Only the initial fresh mass of Carpobrotus was used as a covariable in these analyses, along with Origin of Carpobrotus and Presence of native Iberian plants.
Availability of data and materials
The dataset supporting the conclusions of this article is included within the article (and its Additional file 6).
Ricciardi A. Are modern biological invasions an unprecedented form of global change? Conserv Biol. 2007;21:329–36.
Richardson DM, Allsopp N, D’Antonio CM, Milton SJ, Rejmanek M. Plant invasions-the role of mutualisms. Biol Rev Camb Philos Soc. 2000;75:65–93.
Díaz S, Settele J, Brondízio ES, Ngo HT, Agard J, Arneth A, et al. Pervasive human-driven decline of life on Earth points to the need for transformative change. Science. 2019;366:eaax3100. https://doi.org/10.1126/science.aax3100.
Bongaarts J, Casterline J, Desai S, Hodgson D, MacKellar L. Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. Popul Dev Rev. 2019;45:680–1. https://doi.org/10.1111/padr.12283.
Thomas CD, Palmer G. Non-native plants add to the British flora without negative consequences for native diversity. Proc Natl Acad Sci U S A. 2015;112:4387–92.
Su G, Logez M, Xu J, Tao S, Villéger S, Brosse S. Human impacts on global freshwater fish biodiversity. Science. 2021;371:835–8.
Bright C. Life out of bounds. New York: WW Norton and Co; 1988.
Ellis BK, Stanford JA, Goodman D, Stafford CP, Gustafson DL, Beauchamp DA, Chess DW, Craft JA, Deleray MA, Hansen BS. Long-term effects of a trophic cascade in a large lake ecosystem. Proc Natl Acad Sci U S A. 2011;108:1070–5.
Fritts TH, Rodda GH. The role of introduced species in the degradation of island ecosystems: a case history of Guam. Annu Rev Ecol Syst. 1998;39:113–40.
Bartley TJ, McCann KS, Bieg C, Cazelles K, Granados M, Guzzo MM, et al. Food web rewiring in a changing world. Nat Ecol Evol. 2019;3:345–54.
Thomas CD. Local diversity stays about the same, regional diversity increases, and global diversity declines. Proc Natl Acad Sci U S A. 2013;110:19187–8.
Oduor A. Evolutionary responses of native plant species to invasive plants: a review. New Phytol. 2013;200:986–92.
Workman RE, Cruzan MB. Common mycelial networks impact competition in an invasive grass. Am J Bot. 2016;103:1041–9.
Schilthuizen M, Pimenta LPS, Lammers Y, Steenbergen PJ, Flohil M, Beveridge NGP, et al. Incorporation of an invasive plant into a native insect herbivore food web. PeerJ. 2016. https://doi.org/10.7717/peerj.1954.
Vizentin-Bugoni J, Tarwater CE, Foster JT, Drake DR, Gleditsch JM, Hruska AM, et al. Structure, spatial dynamics, and stability of novel seed dispersal mutualistic networks in Hawai’i. Science. 2019;364:78–82.
Chapman PM. Benefits of invasive species. Mar Pollut Bull. 2016;107:1–2.
Kaiser-Bunbury C, Mougal J, Whittington AE, Valentin T, Gabriel R, Olesen JM, et al. Ecosystem restoration strengthens pollination network resilience and function. Nature. 2017;542:223–7.
Strauss SY, Lau JA, Carroll SP. Evolutionary responses of natives to introduced species: what do introductions tell us about natural communities? Ecol Lett. 2006;9:357–74.
Alfaro B, Marshall DL. Phenotypic variation of life-history traits in native, invasive, and landrace populations of Brassica tournefortii. Ecol Evol. 2019;9:13127–41.
Roiloa SR, Retuerto R, Campoy JG, Novoa A, Barreiro R. Division of labor brings greater benefits to clones of Carpobrotus edulis in the non-native range: evidence for rapid adaptive evolution. Front Plant Sci. 2016;7:349. https://doi.org/10.3389/fpls.2016.00349.
Felker-Quinn E, Schweitzer JA, Bailey JK. Meta-analysis reveals evolution in invasive plant species but little support for evolution of increased competitive ability (EICA). Ecol Evol. 2013;3:739–51.
Colautti RI, Lau JA. Contemporary evolution during invasion: evidence for differentiation, natural selection, and local adaptation. Mol Ecol. 2015;24:1999–2017.
Bertelsmeier C, Keller L. Bridgehead effects and role of adaptive evolution in invasive populations. Trends Ecol Evol. 2018;33:527–34.
Cruzan MB. How to make a weed: the saga of the slender false brome invasion in the North American west and lessons for the future. Bioscience. 2019;69:496–507.
Stastny M, Sargent RD. Evidence for rapid evolutionary change in an invasive plant in response to biological control. J Evol Biol. 2017;30:1042–52.
Agrawal AA, Hastings AP, Bradburd GS, Woods EC, Züst T, Harvey JA, et al. Evolution of plant growth and defense in a continental introduction. Am Nat. 2015;186:E1–15.
Stuart YE, Campbell TS, Hohenlohe PA, Reynolds RG, Revell LJ, Losos JB. Rapid evolution of a native species following invasion by a congener. Science. 2014;346:463–6.
Colautti RI, Barret SC. Population divergence along lines of genetic variance and covariance in the invasive plant Lythrum salicaria in eastern North America. Evolution. 2011;65:2514–29.
Ziska LH, Tomecek MB, Valerio M, Thompson JP. Evidence for recent evolution in an invasive species, Microstegium vimineum, japanese stiltgrass. Weed Res. 2015;55:260–7.
Callaway RM, Maron JL. What have exotic plant invasions taught us over the past 20 years? Trends Ecol Evol. 2016;21:369–74.
Carroll SP, Dingle H, Famula TR, Fox CW. Genetic architecture of adaptive differentiation in evolving host races of the soapberry bug, Jadera haematoloma. In: Hendry AP, Kinnison MT, eds. Microevolution Rate, Pattern, Process. Contemporary Issues in Genetics and Evolution, vol 8. Dordrecht: Springer; 2001. p. 257–272.
Lankau RA. Coevolution between invasive and native plants driven by chemical competition and soil biota. Proc Natl Acad Sci U S A. 2012;109:11240–5.
Mealor BA, Hild AL. Post-invasion evolution of native plant populations: a test of biological resilience. Oikos. 2007;116:1493–500.
Whitney KD, Gabler CA. Rapid evolution in introduced species, “invasive traits” and recipient communities: challenges for predicting invasive potential. Divers Distrib. 2008;14:569–80.
Callaway RM, Ridenour WM, Laboski T, Weir T, Vivanco JM. Natural selection for resistance to the allelopathic effects of invasive plants. J Ecol. 2005;93:576–83.
Rowe CJ, Leger EA. Competitive seedlings and inherited traits: a test of rapid evolution of Elymus multisetus (big squirreltail) in response to cheatgrass invasion. Evol Appl. 2011;4:485–98.
Crooks JA. Lag times and exotic species: the ecology and management of biological invasions in slow-motion. Ecoscience. 2005;12:316–29.
Simberloff D, Gibbons L. Now you see them now you don’t!-population crashes of established introduced species. Biol Inv. 2004;6:161–72.
Green PT, O’Dowd DJ, Abbott KL, Jeffery M, Retallick K, Mac NR. Invasional meltdown: invader–invader mutualism facilitates a secondary invasion. Ecology. 2011;92:1758–68.
Simberloff D, Von Holle B. Positive interactions of nonindigenous species: invasional meltdown? Biol Invasions. 1999;1:21–32.
Thorpe AS, Aschehoug ET, Atwater DZ, Callaway RM. Interactions among plants and evolution. J Ecol. 2011;99:729–40.
Dostál P, Müllerová J, Pyšek P, Pergl J, Klinerová T. The impact of an invasive plant changes over time. Ecol Lett. 2013;16:1277–84.
Iacarella JC, Mankiewicz PS, Ricciardi A. Negative competitive effects of invasive plants change with time since invasion. Ecosphere. 2015;6:art123. https://doi.org/10.1890/ES15-00147.1.
Hawkes C. Are invaders moving targets? The Generality and persistence of Advantages in size, reproduction, and enemy release in invasive plant species with time since introduction. Am Nat. 2007;170:832–43.
Kurr M, Davies AJ. Time-since-invasion increases native mesoherbivore feeding rates on the invasive alga, Sargassum muticum (Yendo) Fensholt. J Mar Biol Assoc UK. 2018;98:1935–44.
Moran EV, Alexander JM. Evolutionary responses to global change: lessons from invasive species. Ecol Lett. 2014;17:637–49.
Strayer DL, Eviner VT, Jeschke JM, Pace ML. Understanding the long-term effects of species invasions. Trends Ecol Evol. 2006;21:645–51.
Atwater DZ. Interplay between competition and evolution in invaded and native plant communities. Ph.D. dissertation, University of Montana; 2012.
Fletcher RA, Callaway RM, Atwater DZ. An exotic invasive plant selects for increased competitive tolerance, but not competitive suppression, in a native grass. Oecologia. 2016;181:499–505.
Miller TE, Werner PA. Competitive effects and responses between plant species in a first-year old-field community. Ecology. 1987;68:1201–10.
Crawley MJ. What makes a community invasible? In: Gray AJ, Crawley MJ, Edwards PP, eds. Blackwell Scientific Publications Oxford; 1987. p. 429–53.
Blossey B, Nötzold R. Evolution of increased competitive ability in invasive nonindigenous plants: a hypothesis. J Ecol. 1995;83:887–9.
Bazzaz FA, Chiariello NR, Coley PD, Pitelka LF. Allocating resources to reproduction and defense. Bioscience. 1987;37:58–67.
Coley PD, Bryant JP, Chapin FS. Resource availability and plant antiherbivore defense. Science. 1985;230:895–9.
Strauss SY, Webb CO, Salamin N. Exotic taxa less related to native species are more invasive. Proc Natl Acad Sci U S A. 2006;103:5841–5.
Novoa A, González L, Moravcováb L, Pyšek P. Constraints to native plant species establishment in coastal dune communities invaded by Carpobrotus edulis: Implications for restoration. Biol Conserv. 2013;164:1–9.
Araniti F, Sorgonà A, Lupini A, Abenavoli M. Screening of Mediterranean wild plant species for allelopathic activity and their use as bio-herbicides. Allelopathy J. 2012;29:107–24.
Mancini E, De Martino L, Marandino A, Scognamiglio MR, De Feo V. Chemical composition and possible in vitro phytotoxic activity of Helichrsyum italicum (Roth) don ssp italicum. Molecules. 2011;16:7725–35.
Rasher DB, Hay ME. Competition induces allelopathy but suppresses growth and anti-herbivore defence in a chemically rich seaweed. Proc R Soc B. 2014;281:20132615. https://doi.org/10.1098/rspb.2013.2615.
Jucker T, Carboni M, Acosta ATR. Going beyond taxonomic diversity: deconstructing biodiversity patterns reveals the true cost of iceplant invasion. Divers Distrib. 2013;19:1566–77.
Vilà M, Tessier M, Suehs CM, Brundu G, Carta L, Galanidis A, et al. Local and regional assessments of the impacts of plant invaders on vegetation structure and soil properties of Mediterranean islands. J Biogeogr. 2006;33:853–61.
García-Fernández A, Vitales D, Pellicer J, Garnatje T, Vallés J. Phylogeographic insights into Artemisia crithmifolia (Asteraceae) reveal several areas of the Iberian Atlantic coast as refugia for genetic diversity. Plant Syst Evol. 2017;303:509–19.
Melito S, Sias A, Petretto G, Chessa M, Pintore G, Porceddu A. Genetic and metabolite diversity of Sardinian populations of Helichrysum italicum. PLoS ONE. 2013;8:e79043. https://doi.org/10.1371/journal.pone.0079043.
Galbany-Casals M, Blanco-Moreno JM, Garcia-Jacas N, Breitwieser I, Smissen RD. Genetic variation in Mediterranean Helichrysum italicum (Asteraceae; Gnaphalieae): do disjunct populations of subsp microphyllum have a common origin? Plant Biol. 2011;13:678–87.
Maltez-Mouro S, Maestre F, Freitas H. Weak effects of the exotic invasive Carpobrotus edulis on the structure and composition of Portuguese sand-dune communities. Biol Inv. 2010;12:2117–30.
Probert AF, Ward DF, Beggs JR, Lin S, Stanley MC. Conceptual risk framework: integrating ecological risk of introduced species with recipient ecosystems. Bioscience. 2020;70:71–9.
Grainger TH, Levine JM, Gilbert B. The Invasion Criterion: a common currency for ecological research. Trends Ecol Evol. 2019;34:925–35.
Ward D, Morgan F. Modelling the impacts of an invasive species across landscapes: a stepwise approach. Peer J. 2014;2:e435. https://doi.org/10.7717/peerj.435.
Leger EA, Espeland EK. Coevolution between native and invasive plant competitors: implications for invasive species management. Evol Appl. 2010;3:169–78.
Feis ME, Goedknegt MA, Thieltges DW, Buschbaum C, Wegner KM. Biological invasions and host-parasite coevolutionary trajectories along separate parasite invasion fronts. Zoology. 2016;119:366–74.
Huang F, Lankau R, Peng S. Coexistence via coevolution driven by reduced allelochemical effects and increased tolerance to competition between invasive and native plants. New Phytol. 2018;218:357–69.
Wisura W, Glen HF. The South African species of Carpobrotus (Mesembryanthema–Aizoaceae). Contr Bolus Herb. 1993;15:76–107.
Campoy JG, Acosta ATR, Affre L, Barreiro R, Brundu G, Buisson E, et al. Monographs of invasive plants in Europe: Carpobrotus. Bot Lett. 2018;165:440–75.
Codd LE, Gunn M. Additional biographical notes on plant collectors in Southern Africa. Bothalia. 1985;15:631–54.
Winter K, Holtum JAM. Facultative crassulacean acid metabolism (CAM) plants: powerful tools for unravelling the functional elements of CAM photosynthesis. J Exp Bot. 2014;65:3425–41.
Campoy JG, Roiloa SR, Santiso X, Retuerto R. Ecophysiological differentiation between two invasive species of Carpobrotus competing under different nutrient conditions. Am J Bot. 2019;106:1454–65.
Fenollosa E, Munné-Bosch S, Pintó-Marijuan M. Contrasting phenotypic plasticity in the photoprotective strategies of the invasive species Carpobrotus edulis and the coexisting native species Crithmum maritimum. Physiol Plant. 2017;160:185–200.
Traveset A, Moragues E, Valladares F. Spreading of the invasive Carpobrotus aff acinaciformis in Mediterranean ecosystems: the advantage of performing in different light environments. Appl Veg Sci. 2008;11:45–54.
Suehs C, Affre L, Médail F. Invasion dynamics of two alien Carpobrotus (Aizoaceae) taxa on a Mediterranean island: II reproductive strategies. Heredity. 2014;92:550–6.
Campoy JG, Retuerto R, Roiloa SR. Resource-sharing strategies in ecotypes of the invasive clonal plant Carpobrotus edulis: specialization for abundance or scarcity of resources. J Plant Ecol. 2017;10:681–91.
D’Antonio CM. Seed production and dispersal in the non-native, invasive succulent Carpobrotus edulis (Aizoaceae) in coastal strand communities of Central California. J Appl Ecol. 1990;27:693–702.
D’Antonio CM, Mahall BE. Root profiles and competition between the invasive, exotic perennial, Carpobrotus edulis, and two native shrub species in California coastal scrub. Am J Bot. 1991;78:885–94.
Regulation (EU) No 1143/2014 of the European Parliament and of the Council of 22 October 2014 on the prevention and management of the introduction and spread of invasive alien species.
Benedí C, Buira A, Rico E, Crespo MB, Quintanar A, Aedo C (eds.). S. Castroviejo (coord.). Flora ibérica. Plantas vasculares de la península ibérica e islas Baleares. Vol. XVI (III) Compositae (partim). Madrid: Real Jardín Botánico, CSIC; 2019.
Harper JL. Population biology of plants. London: Academic Press; 1977.
R Core Team. R: A language and environment for statistical computing R Foundation for Statistical Computing, Vienna, Austria. 2018. https://www.R-projectorg/.
Wagenmakers EJ, Farrell S. AIC model selection using Akaike weights. Psychon Bull Rev. 2004;11:192–6.
Burnham KP, Anderson DR. Kullback-Leibler information as a basis for strong inference in ecological studies. Wildlife Res. 2001;28:111–9.
The authors thank Margarita Lema for assistance with field sampling.
This research and publication costs were funded by the Spanish Ministry of Economy and Competitiveness and the European Regional Development Fund (ERDF) (grant Ref. CGL2013-48885-C2-2-R). The funding bodies played no role in the design of the study, collection, analysis, and interpretation of data, and in writing the manuscript. Any opinion, finding and conclusion or recommendation expressed in this publication is that of the authors and the funding bodies does not accept liability in this regard.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
: Table S1. Analysis of the final dry masses of the native plants in the comparison of pots containing one plant and two plants. Columns show the levels of the main factors in advantage for final mass or the estimated slopes for the covariables, and Likelihood Ratio Tests probability for each model term, AIC weight and the normalized probability that the model including that term will be selected. Number of residual degrees of freedom =27.
: Table S2. Analysis of the final dry masses of the Carpobrotus plants in the comparisons of pots with one plant and two plants. Columns show the levels of the main factors in advantage for the final mass or the estimated slopes for the covariable, and Likelihood Ratio Tests probability for each model term, AIC weight and the normalized probability that the model including that term will be selected. Number of residual degrees of freedom =16.
: Table S3. Likelihood Ratio test probabilities for mass-related variables in the comparisons of pots containing two plants.
: Table S4. Likelihood Ratio tests probabilities for mass-related variables for native Iberian species in the comparisons of pots containing one and two plants.
: Table S5. Likelihood Ratio tests probabilities for mass-related Carpobrotus variables in the comparisons of pots containing one and two Carpobrotus plants.
The full information on sampling locations, experimental treatments, initial and final fresh masses and final root and total dry masses for all native and Carpobrotus plants in the experiment.
About this article
Cite this article
García, C., Campoy, J.G. & Retuerto, R. A test of native plant adaptation more than one century after introduction of the invasive Carpobrotus edulis to the NW Iberian Peninsula. BMC Ecol Evo 21, 69 (2021). https://doi.org/10.1186/s12862-021-01785-x
- Artemisia crithmifolia
- Biomass production
- Co-evolutionary changes
- Helichrysum picardii