Accommodating scenarios of climate change and management in modelling the distribution of the invasive tree Schinus molle in South Africa.
Determining the potential range of invasive alien species under current conditions is important. However, we also need to consider future distributions under scenarios of climate change and different management interventions when formulating effective long-term intervention strategies. This paper combines niche modelling and fine-scale process-based modelling to define regions at high risk of invasion and simulate likely dynamics at the landscape scale. Our study species is Schinus molle (Peruvian pepper tree; Anacardiaceae), a native of central South America, introduced to South Africa in about 1850 where it was widely planted along roads. Localities of planted and naturalized trees were mapped along 5380 km of roads - a transect that effectively samples a large part of western South Africa. Correlative modelling was used to produce profiles of present and future environmental conditions characterizing its planted and naturalized ranges. A cellular-automata simulation model was used to estimate the dynamics of S. molle under future climates and different management scenarios. The overall potential range of S. molle in the region is predicted to shrink progressively with predicted climate change. Some of the potential range of S. molle defined based on current conditions (including areas where it is currently highly invasive) is likely to become less favourable. The species could persist where it is well established long after conditions for recruitment have deteriorated. Some areas where the species is not widely naturalized now (notably the fynbos biome) are likely to become more favourable. Our modelling approach allows for the delineation of areas likely to be invaded in future by considering a range of factors at different scales that mediate the interplay of climatic variables and other drivers that define the dimensions of human intervention such as distance from planted trees and the density of planted plants, both of which affect propagule pressure.