How do species reach the limits of their distribution and what prevents their continued expansion beyond these ranges? Exotic plant species represent a natural experiment to answer these questions. If climate is the limiting factor, then one would expect a matching between the observed range limit for a species and the range limit predicted by its climatic niche. If there is no matching, then other factors such as dispersal limitation, competition or facilitation come into play. In this work, the predicted and observed range limits for eight exotic legume species were compared: Acacia dealbata, Acacia melanoxylon, Cytisus striatus, Teline monspessulana, Ulex europaeus, Lotus corniculatus, Trifolium suffocatum and Vicia villosa, in a latitudinal gradient in Chile. For the estimation of the observed range limit (North and South), absence/presence data were obtained from 30° to 43.1° south latitude. For the estimation of the predicted range limits, GBIF presence data were used to construct the global climatic niches, identifying suitable climatic zones (presences) and unsuitable climatic zones (absences). With this information, presence probability models were constructed with hierarchical Huisman-Olff-Fresh (HOF) regression, from which the predicted range limits (North and South) were obtained. Our results suggest that the species Acacia dealbata and Cytisus striatus have reached their predicted edge at the northern and at the southern end of the gradient. The rest of the species have not yet reached this limit across both geographic edges. At the southern end of the gradient, most species have not reached the limit predicted by the climatic niche; except for Cytisus striatus whose observed range limit is higher than predicted. Factors other than climate are discussed to explain the discrepancies between observed and predicted range limits.

Biogeography, climate, legumes, niche requirements, range limits

Charles Darwin (1957) provided the first hypothesis to understand the factors that set limits to species geographic distribution; he proposed that abiotic factors (climate) are important at the abiotic stressful limits of distribution while species biotic interactions become more important at the more benign extreme of the distribution. Nowadays, this framework plays a central role in modern biogeography (Gaston 2003; Sexton et al. 2009) and is assumed a ubiquitous biogeographic pattern (Louthan et al. 2015). New reformulations have proposed to understand range limits of species (Ettinger and HilleRisLambers 2017; Sirén and Morelli 2020; Paquette and Hargreaves 2021), however, they maintain the basic concepts proposed by Darwin.

Hargreaves et al. (2014) proposed a hierarchical approach to understand range limits using the climatic niche of species. This framework predicts the climatic niche limits (CNL) i.e. the geographic limits predicted from climatic niche, and compares it with the range limit (RL), i.e. the observed limits in the extremes of distribution. If RL - CNL = 0, populations located at the limit range, perfectly match between climatic niche requirements and the extreme of distribution; if RL - CNL < 0, beyond RL there are suitable habitats, but they cannot be colonized by species due to dispersal limitation, introduction time or negative biotic interactions (competition, predation); if RL – CNL > 0, in RL populations are occupying unsuitable habitats, they are sink populations which must be continuously supplemented by individuals from source populations in order to persist. Another explanation is that species establish positive interactions with other species, allowing expansion due to facilitation with other plant (Badano et al. 2007; Arredondo-Núñez et al. 2009). Biological invasions provide useful insights to learn about dynamics of species range limits (Sexton et al. 2009). Given that invasive species may have colonized habitats quite different to those existing in their native ranges, we have a natural experiment to examine whether the new range limits depart from expectations from climatic requirements (Keane and Crawley 2002; Gaston 2003; Goncalves et al. 2022).

Climatic niche analysis has been prolifically used to understand the biogeography of biological invasions (Peterson 2003; Thuiller et al. 2005; Broennimann et al. 2007; Reed et al. 2008; Gallagher et al. 2010; Peña-Gómez et al. 2014; Cabra-Rivas et al. 2016). The global climatic niche enables exploration of the climatic potentialities of species using the totality of occurrences recorded for the species across invaded regions (Gallien et al. 2012; Taucare-Ríos et al. 2016). In this study, we assessed if RL is within the variation of CNL using global climatic niche. These methodologies were applied to a set of eight exotic leguminous plants species, invasive in central Chile, and in other regions of the world (Quiroz et al. 2009). Finally, we also discussed its applicability to other biogeographic situations, and to what extent the hypothesis proposed by Hargreaves et al. (2014) give us clues about the geographic dynamics of these eight species.

Acacia species (A. dealbata and A. melanoxylon) presents the broadest latitudinal range in Chile (Table 1); A. dealbata was the species with a higher number of presence records in the field (Table 1). On the other hand, Trifolium suffocatum was the species with the lowest latitudinal range and one of the species with lowest number of presence records in the field work in Chile (Table 1).

For the northern distribution of Acacia dealbata and Cytisus striatus, a matching between the observed and predicted north range limit was detected (Table 2), while for the rest of the species the observed north range limit was significantly lower than predicted (Table 2, Figs 3, 4). For the southern distribution of Acacia dealbata and Cytisus striatus, matching between observations and predictions was founded; for Acacia melanoxylon, observed RL was lower than expected, and for the rest of the species, the observed southern range limit was significantly higher than predictions (Table 3).

Figure 3.

Bootstrap distribution of Northern and Southern CNL, obtained from global niche models, for Teline monspessulana, Cytisus striatus, Acacia melanoxylon and Acacia dealbata, central Chile.

Figure 4.

Bootstrap distribution of Northern and Southern CNL, obtained from global niche models, for Trifolium suffocatum, Vicia villosa, Ulex europaeus and Lotus corniculatus, central Chile.

The geographic position of RL was quite similar between the northern and southern range; at the northern range, in only two species (Acacia dealbata and Cytisus striatus), the RL was explained by climate. For the rest of species, the RL was explained by other factors such as dispersal limitation or negative biotic interactions. At the southern range, for Acacia dealbata and Cytisus striatus, the RL matched with climatic expectations as well; in one species (Acacia melanoxylon), RL was explained by facilitation and for the rest of species, RL was explained by negative biotic interactions and dispersal limitation.

For A. dealbata and C. striatus, climate explained RL either at the northern or at the southern border. This matching may indeed be attributed to their high ornamental value and widespread cultivation in parks, gardens and road borders, with strong anthropogenic subsidies in terms of resources and conditions (Van Kleunen et al. 2018; Beaury et al. 2023). This increased human-mediated dispersal and cultivation likely enhance their chances of expansion, allowing them to reach the maximum extent of their distribution without dispersal limitation. A. dealbata is regarded one of the most successful exotic trees in central Chile with significant impacts on native biodiversity (Fuentes-Ramírez et al. 2011). Our results give support to this invasive success. Firstly, its niche requirements match with prevailing climatic conditions; secondly, there is no dispersal limitation; thirdly it is a strong competitor over native plants due to allelopathic effects on the germination and growth of seed and seedlings (Aguilera et al. 2015), and a faster growth rate relative to native trees (Fuentes-Ramírez et al. 2011). However, there are some ecological constraints that can limit further expansion because it has resulted in becoming the most attacked exotic plant by herbivorous insects in invaded ranges which, in turn, can be used by biological control in different parts of the world (Wilson et al. 2011; Wilgen et al. 2023).

At the southern range, Cytisus striatus also matched RL with climatic niche. This result may be attributed to a low cold or freezing resistance of this species, as it has been documented in the northern hemisphere (Beans et al. 2012; Thomas and Moloney 2013; Winde et al. 2020).

At the northern range, the abiotic environments are relatively hostile to plant species. Under these conditions, plant-plant facilitation should be promoted, according to theory (Bertness and Callaway 1994; Lortie and Callaway 2006); however, we did not find evidence of such mechanism. We suggest that the mismatch detected between observation and expectation for six species (Table 1) can be attributed to dispersal limitations (low propagule pressure) due in part to a relatively low human settlement. The southern end of the climatic gradient in turn, hosts a high plant species diversity in Chile as well as an increase of the forest biomass (Bannister et al. 2012). This increased diversity may render these communities less invasible due to biotic resistance or competitive mechanisms that limit the establishment of invasive species (Levine et al. 2004; Guo et al. 2023). Our results are consistent with Callaway’s hypothesis (Bertness and Callaway 1994; Lortie and Callaway 2006), who suggests that less hostile environments, such as the southern border, may induce higher competitive pressure on exotic plants, thus resulting in constrained expansion of exotic plants further south. In summary, we propose that dispersal limitation may play an influential role at the northern border (lower latitude), while competition is more important at the southern border (higher latitude). Further field experiments are needed to test these biogeographic hypotheses.

In the southern range, Acacia melanoxylon, exhibited its RL beyond predictions from climatic niche. Mechanisms such as facilitation by human use, potential nurse species interactions, or local adaptation and expansion of tolerance ranges may explain these patterns. For example, a study by Turner et al. (2015) about invasive thistle (C. diffusa) suggests that the physiological tolerances of C. diffusa may have expanded in the invaded range. Invasive species tend to present adaptive plasticity and niche expansion (Moran and Alexander 2014; Pack et al. 2022). These hypotheses warrant further investigation through transplant experiments to elucidate the underlying mechanisms driving these distribution patterns.

The study of the causal factors that explain RL along environmental gradients has proven to be a fruitful research program linking biogeography, ecology and evolution (Holt and Keitt 2005; Sexton et al. 2009; Louthan et al. 2015). Most efforts have been addressed to designing proper field experiments to discern the microevolutionary and ecological factors which are responsible for such limits (Geber 2011; Hargreaves et al. 2014; Sexton and Dickman 2016); however less effort has been devoted to inferring RL from climatic niche using statistical techniques. The method applied in this study proposes a methodology to infer statistically, the expected range limits based on climatic niche requirements; this method establishes the geographic position species range limits, so allowing us to dispose with precision where to put transplants’ experiments to test biogeographic hypotheses proposed by Hargreaves et al. (2014).

In conclusion, our study sheds light on the interplay between observed range limits and the global climatic niche for leguminous invasive plants in central Chile. While the climate-based limitation hypothesis is partially supported, with only two species showing concordance between niche and distribution at the northern and southern edge, our findings suggest that climatic conditions alone do not fully explain distribution patterns. This discrepancy between niche and distribution is particularly notable in areas with favorable climatic conditions, such as the southern extreme of the climatic gradient. Moreover, the idiosyncratic responses of species at both ends of the gradient highlight the importance of species-specific attributes and invasion processes that may influence distribution patterns. The complex interplay between climate, human activities, and ecological factors underscores the need for further research, particularly experimental studies, to validate and elucidate the underlying mechanisms shaping invasive plant distributions in mountainous regions like central Chile. Understanding these mechanisms is crucial for effective management and mitigation strategies aimed at controlling the spread and impact of invasive species in these ecosystems.