The establishment of new botanic gardens in tropical regions highlights a need for weed risk assessment tools suitable for tropical ecosystems. The relevance of plant traits for invasion into tropical rainforests has not been well studied. 

Working in and around four botanic gardens in Indonesia where 600 exotic species have been planted, we estimated the effect of four plant traits and time since species were introduced on: a) naturalization probability of exotic species; b) the abundance (density) of naturalized species in adjacent native tropical rainforests; and c) the distance that naturalized exotics have spread from the botanic gardens.  

We found that specific leaf area (SLA) strongly differentiated 23 naturalized from 78 non-naturalized exotic species (randomly selected from 577 non-naturalized species) in our study. These trends may indicate that exotics with high SLA benefit from at least two factors when establishing in tropical forests: high growth rates and occupation of forest gaps. Exotic species that were present in the gardens for over 30 years and those with small seeds also had higher probabilities of becoming naturalized, indicating that plants can invade the understorey of closed canopy tropical rainforests, especially when invading species are shade-tolerant and have sufficient time to establish. On average, exotic species that were not animal dispersed spread 78 m further into the forests than animal-dispersed species. We did not detect relationships between the measured traits and estimated density of naturalized exotics in the adjacent forests.

Synthesis: Traits were able to differentiate exotic species that naturalized from botanic gardens from those that did not; this is promising for developing trait-based risk assessment in the tropics. We suggest tropical botanic gardens avoid planting exotic species with fast carbon capture strategies and those that are shade tolerant, to limit the risk of invasion and spread into adjacent native forests.

The dataset consists of specific leaf area (SLA), height, seed mass, dispersal method, and residence time. We measured traits for every individual of the naturalized exotic species we detected (a total of 23 species and 913 individuals) and for non-naturalized exotics (195 individuals with the average and minimum number of individuals per species being 6 and 2 respectively).

For SLA data of non-naturalized exotics, we sampled shaded leaves instead of fully sun-exposed as suggested by Pérez-Harguindeguy et al. (2013) to avoid bias toward environmental differences between shaded forests and open areas of botanic gardens.

Plant height (m) was measured following the method suggested by Pérez-Harguindeguy et al. (2013). Not all height data were obtained from direct measurement during surveys due to measurement difficulties in the field. Direct measurements were conducted for 40% of the species. For the remaining 60%, data were obtained from the ‘calibrated’ median height data available in botanical descriptions (e.g. http://www.efloras.org/ and http://hear.org). We transformed this information into height estimates as follows. First, we plotted measured heights against the corresponding median available in databases. Then, we fitted a model to describe the relationship between these two sources of information. We chose the best fitted model based on the largest r² value. Then, we ‘calibrated’ the height of the remaining 60% of exotic species based on the fitted model, using the median height secondary data from databases as the predictor variable.

We did not measure seed mass directly because of limited fruit availability in the study sites during the survey period. Secondary seed mass data for naturalized and non-naturalized exotics were obtained from the Kew Seed Database (<http://data.kew.org/sid/>) and defined as 1000 dry seed mass (mg). When we did not find seed mass data for a species in the database, we used the average data of the corresponding genus (29 species) or family (4 species) from the same database, using a minimum of 30 other species.

We categorized dispersal method as whether the plants are dispersed by animals or not.

We classified growth-form of exotic plants into herbs, shrubs, and trees (we excluded other growth forms, including ferns and vines, due to very low detection rates). Data on dispersal method, origin, and growth-form were collected from databases, including <http://data.kew.org/sid/>, http://www.ars-grin.gov/>, <http://www.gbif.org>, <http://www.pfaf.org/>, and <http://www.hear.org/pier/>. 

Information on minimum residence time, i.e. the number of years elapsed since the species was first known to be present in the botanic gardens to the date of the survey was obtained from botanic gardens’ catalogue collections (Cibodas: 1930, 1963, 1977, and 1988; Bali: 1989, 1999, and 2006) and planting date official records (Kuningan and Baturraden). We used minimum residence time in our analyses to account for the lag time between the introduction and establishment of an exotic species.