Invasive alien plants cause severe global problems; therefore, determining the factors that lead to the success or failure of invasion is a critical question in the field of invasion ecology. In this study, we aimed to determine the factors underlying differences in the distribution range of alien plants in Japan by investigating why Lolium multiflorum thrives in a wide range of habitats while L. rigidum is mainly distributed on sandy beaches. We initially evaluated environmental niche suitability through species distribution modelling and subsequently examined whether species traits influence the differences in range expansion between the two species. We used MaxEnt modelling to identify potential environmental niches for both species. The analysis revealed that L. rigidum was considerably less suited to the Japanese climate compared to L. multiflorum, with high summer precipitation in Japan identified as one of the climatic factors limiting the distribution of L. rigidum. Given that these winter annual plants remain dormant as seeds during summer, in subsequent experiments, we buried seeds in paddy field soil and sandy beach sand during summer and evaluated their survival rate in autumn. The survival rate of L. rigidum seeds was significantly lower than that of L. multiflorum, particularly in paddy soil. Factors contributing to seed mortality may include the decay or early germination of L. rigidum seeds under Japan’s high rainfall conditions. This study emphasises the importance of considering local environmental factors alongside climate niche modelling in the risk assessment of invasive species. Moreover, the integration of species distribution modelling for large-scale evaluations and manipulation experiments for fine-scale assessments proved effective in identifying climatic conditions and species traits influencing the success or failure of alien species invasion.

Coastal environment, invasion success, local environmental condition, MaxEnt, seed burial experiment, seed decay, species distribution modelling, summer rainfall

Invasive alien plants cause issues such as loss of biodiversity, reduced crop yield, and health risks (Vitousek 1990; Pyšek and Richardson 2010; Vilà et al. 2010). However, not all introduced species become invasive in a new habitat. It has been hypothesised that only about 10% of species successfully progress through the consecutive steps of the invasion process. Specifically, approximately 10% of species transported beyond their native range will be released or escape in the new regions, about 10% of these introduced species will successfully establish themselves, and about 10% of established species will become invasive (tens rule; Williamson and Fitter 1996). Therefore, extensive long-term research has been conducted to identify which environments facilitate the establishment of invasive species and which species’ characteristics contribute to the successful establishment (Richardson and Pyšek 2006; Hayes and Barry 2008; Hui et al. 2016).

The impact of climate on determining invasion success or failure has been extensively investigated using the species distribution modelling (SDM) approach. This approach can estimate suitable climate conditions for a target species based on historical climate and species occurrence data and predict the distribution suitability of the species across geographic and temporal scales. This approach has revealed that specific climate factors can influence the potential geographic distribution of a species (Phillips et al. 2006; Phillips and Dudík 2008; Wiens et al. 2009). SDM is widely used in the risk assessment of invasive species, largely because the climatic niche of invasive species in their new destination areas often resembles that in their source areas (Liu et al. 2020). Although SDM is a robust tool that combines statistical modelling and geographic information systems to gain insights into the potential environmental factors influencing species distributions (Phillips et al. 2004), it cannot definitively establish causal relationships between environmental factors and species distribution (Merow et al. 2013). To identify the drivers of invasion processes in destination areas, experiments manipulating environmental factors are necessary; however, few studies have combined both SDM and experimental manipulation approaches.

The genus Lolium includes two outcrossing annual species: L. multiflorum Lam. and L. rigidum Gaudin, which are native to the Mediterranean region (Terrell 1968). They have been introduced as a forage crop and turfgrass in numerous countries across the globe (Humphreys et al. 2010), resulting in their escape from controlled areas and becoming problematic weeds. L. multiflorum is widely distributed throughout Europe, North America, South America, northern and eastern Africa, Australia, Central Asia, and eastern Asia (GBIF 2022a) (Fig. 1a). It directly reduces crop yield by spreading as a weed in agricultural fields (Liebl and Worsham 1987; Sønderskov et al. 2020) and serving as a habitat for rice-ear bugs, which are important pests of rice (Yoshioka et al. 2011). L. rigidum is distributed throughout Europe, Australia, North America, South America, South Africa, East Asia, and West Asia (GBIF 2022b) (Fig. 1b). This species has developed tolerance to multiple herbicides, posing a serious challenge (Heap 2023), particularly in wheat cultivation in Australia (Owen et al. 2014). In Japan, L. multiflorum has predominantly been introduced as forage and revegetation materials but has spread as weeds in universal environments such as farmland and riverbanks. Conversely, L. rigidum has mainly been introduced to Japan through imported grains contaminated with L. rigidum seeds and subsequently became established (Shimono et al. 2015; Higuchi et al. 2017; Hirata et al. 2023). Originally a weed in agricultural fields, it has only been specifically established in local environments, such as sandy beaches, in Japan (Hirata et al. 2023).

Figure 1.

The distribution of A L. multiflorum and B L. rigidum. This distribution map was made using the occurrence data from GBIF (GBIF 2022a, 2022b).

In this study, we aimed to investigate environmental factors and species traits that contribute to the distribution expansion of alien plants in Japan using both SDM and manipulation experiments, focusing on congeneric species of the genus Lolium with different distribution ranges in Japan. Specifically, we investigated why L. multiflorum thrives in a wide range of habitats whereas L. rigidum has not been able to spread to agricultural fields. We initially examined the suitability of the climate in Japan for the spread of these Lolium species using the MaxEnt model (Phillips et al. 2006), which is the most common SDM approach. MaxEnt analysis revealed summer precipitation as a key climatic factor limiting the distribution of L. rigidum in Japan. In addition, as L. rigidum and L. multiflorum are winter annuals that exist as seeds during the summer, we evaluated the seed survival rates of both Lolium species in the soil from summer to autumn.

In the present study, the MaxEnt results indicated that L. rigidum was less suited to the Japanese climate than L. multiflorum. The Japanese climatic conditions that deviated considerably from the predicted suitability range for L. rigidum were temperature seasonality and summer precipitation. Temperature seasonality is a determinant influencing the northern limits of plants and animals in the Northern Hemisphere (Wiens et al. 2006; Qian et al. 2022). L. rigidum is found in temperate zones, and as Japan is not located at the northern limit, it is currently unclear what ecological significance can be ascribed to temperature seasonality as a climatic variable that explains the distribution of Lolium species in Japan.

MaxEnt predicted that high summer precipitation in Japan renders the environment unsuitable for L. rigidum. The burial seed experiment conducted during summer revealed a higher mortality rate for L. rigidum seeds than for L. multiflorum seeds. Potential reasons for seed mortality in this experiment include seed decay or premature germination. Additionally, in real-world field conditions, various factors such as predation and fungal infections, which were not accounted for in our experiment, could further reduce seed survival rates (Ranganathan and Groot 2023). To better evaluate the fitness, the survival rate of seedlings after germination in each field should be investigated in future studies. Nevertheless, comparing our experimental findings with L. multiflorum, which has successfully expanded its distribution across a wide range of environments, suggests that seed decay due to heavy summer rains likely reduces the fitness of L. rigidum in the field.

Water availability is the primary limiting factor for terrestrial plant production (Lambers and Oliveira 2019), and soil hydrological properties at a fine scale effectively determine plant distribution (Silvertown et al. 1999). As sandy beaches are generally arid environments (Brakenhoff et al. 2019), our experimental results confirmed that under the same precipitation conditions, soil moisture content was lower in sandy soil than in paddy soil. This lower moisture content suggests that seeds of L. rigidum are less likely to decay in such areas, and consequently, the establishment of L. rigidum in Japan is possibly locally limited to sandy beaches.

L. rigidum is native to the Mediterranean region (Terrell 1968) and has emerged as a major weed problem, especially under the Mediterranean-type climate of Western Australia (Owen et al. 2014). Japan, with its monsoon climate, receives considerably higher precipitation than the aforementioned regions (annual precipitation of 1,668 mm/year in Japan versus 733 mm/year in the Perth metropolitan area, Australia) (Commonwealth of Australia 2023; The World Bank 2023). In addition, as a consequence of the widespread cultivation of rice paddies throughout the country, the soil moisture content in Japanese agricultural land is generally high. Given this context, seed decay can be considered a weed control method in Japan (Kida and Asai 2006; Aoki et al. 2012). Although the drought sensitivity and flood tolerance of seedlings have been previously assessed to understand plant distribution patterns (Engelbrecht and Kursar 2003; Jansen et al. 2005), seed moisture tolerance has mostly been neglected and underestimated.

Sandy coasts are typically arid, nutrient-poor, and highly susceptible to salt spray, sand deposition, and strong winds, all of which are limiting factors for plant establishment (Maun 1994). Dryland salinity is a major problem in agricultural areas in Australia (Briggs and Taws 2003; George et al. 2006), and L. rigidum, a weed that thrives in such environments, may be highly tolerant to drought and salinity stress. Weeds that have become problematic in arid agricultural areas, not only in Australia, may have the potential to establish themselves on sandy beaches in wetter areas because of their drought and salt tolerance.

In the present study, niche modelling solely based on climate variables indicated that L. rigidum was not well-suited for distribution in Japan; however, it is actually expanding its distribution on sandy beaches. This suggests that climate niche modelling is insufficient for fine-scale predictions and underestimates the invasion risk of alien species in some specialised habitats. When applying SDM to finer scales, local predictors, such as soil conditions and topography, must be considered (Pearson and Dawson 2003). However, acquiring these variables over a large spatial extent is challenging (Bradley et al. 2012).

In summary, we conducted large-scale niche modelling to identify environmental factors predicted to limit the distribution of invasive species. Subsequently, we examined whether these factors actually affect the fitness of these species through manipulative experiments. Although manipulative experiments alone cannot account for all factors, focusing on environmental factors suggested by niche modelling and estimating the causality of invasion success through these experiments are highly valuable.

In this study, we focused on the alien L. rigidum, which has only expanded locally on sandy beaches in Japan, and L. multiflorum, which has been successfully established across various environments in Japan. We investigated the limiting factors for the distribution expansion of L. rigidum. Through manipulative experiments, we found that L. rigidum had a higher seed mortality rate, especially in paddy soil, than that of L. multiflorum. This result aligns with a suggestion from SDMs that the summer rainfall in Japan may be excessive for L. rigidum.

Predictions based solely on climate variables using SDM revealed that L. rigidum is not suitable for the Japanese environment. However, L. rigidum is actually expanding its distribution on Japanese sandy beaches. This indicates that niche modelling based on specific climate variables alone may underestimate the invasion risk of alien species. The combined use of large-scale niche modelling and manipulative experiments, as conducted in this study, demonstrates the importance of this approach for assessing the invasion risk of species in both regional and local environments.

We thank Tohru Tominaga and Shunji Kurokawa for their valuable discussions. We would like to thank Editage (www.editage.jp) for English language editing.