Determining factors for the occurrence of woody invaders in Neotropical forests

Amanda Carmes, Pedro Fiaschi and Michele Dechoum, Universidade Federal de Santa Catarina, Brazil, Rafael Barbizan Sühs, Universidade de São Paulo, Brazil

Background/Question/Methods
The regional pool of introduced species is determined by historical and human factors (i.e. introduction efforts, reasons for introduction); but the arrival of introduced species in natural areas can be influenced by different traits (i.e. dispersion syndrome, shade tolerance), environmental (i.e. climate, resource availability) and human factors (i.e. population density, landscape fragmentation). The number of studies documenting the presence of invasive non-native plants in tropical and subtropical forests has been increasing in the Neotropics over recent years. Forest fragmentation may facilitate invasion because resource availability is increased, especially on edges between forest and converted areas. Changes in land use sometimes lead to an increase in the use of introduced species, which can expose natural habitats to propagule pressure by non-native species present in surrounding degraded areas. The main question in this study is: Are human factors more important than environmental factors in explaining the presence of woody invasive plants in different types of Neotropical forest? We conducted a systematic review of floristic studies published between 1945 and 2019 in the Neotropics, covering all forest types. We checked the history of invasion of the woody non-native plants cited in each study using data from international invasive non-native species databases (i.e. CABI ISC, GISD). Only plant species considered invasive elsewhere were included in the analyses.
Results/Conclusions
A total of 1,116 studies were selected, but only 124 studies met the inclusion criteria and were considered in our final dataset. Twenty-three woody invasive species were selected from these studies (17 trees and 6 shrubs). We collected the following data on each study site using the geographic coordinates of each sampling unit: thermal domain, climate regime, forest type, and leaf deciduousness. The following human factors were calculated for a buffer zone of 1km around each study site: accessibility, human influence index (HII), gross domestic product (GDP), forest cover, annual precipitation, annual mean temperature, and minimum temperature of the coldest month. All data collected for each site and respective buffer zone were used as explanatory variables in Generalized Linear Models in which the response variable was the number of woody invasive plants in each site. A total of 16 models were run. The best model (lowest AIC) was the one with HII as the only explanatory variable. Our preliminary results show the relevance of human factors for explaining the presence of woody invasive plants in Neotropical forests.