Thermal tolerance of the biological control agent Neolema abbreviata and its potential geographic distribution together with its host Tradescantia fluminensis in South Africa

In this era of global climate change, understanding how climate influences species distributions together with their subsequent interactions is a major priority for managers of both natural and managed ecosystems. Using the predictive algorithm Maximum Entropy (MaxEnt), we projected the current and potential distribution of Tradescantia fluminensis Vell (Commelinaceae) before using thermal tolerance indices of its biological control agent Neolema abbreviata (Coleoptera: Chrysomelidae) to simulate its potential distribution in South Africa. The mean temperature in T. fluminensis microhabitats, recorded using data loggers, was 24.7 +/- 0.1 degrees C (mean +/- SE) whilst temperatures ranged from -1 to 29.7 degrees C. Following 2 h exposure at low temperatures, N. abbreviata survival from 0 to 100% fell between -13 and -6 +/- 0.2 degrees C with 50% mortality occurring at -9.8 +/- 0.2 degrees C. For heat assays, temperatures above 45.8 +/- 0.2 degrees C resulted in 100% mortality whilst temperatures below 43 +/- 0.2 degrees C resulted in 100% survival with 50% mortality occurring at 44.2 +/- 0.2 degrees C. The mean critical thermal limits were 1.8 +/- 0.4 degrees C (CTmin) and 48.8 +/- 1.3 degrees C (CTmax). The derived warm and low temperature tolerance largely matched with current and projected climate in T. fluminensis microhabitats. Model evaluation for T. fluminensis distribution was based on mean Area Under the Curve (AUC), which was 0.85 +/- 0.017, suggesting good predictive performance of the model. Our results showed a limited capacity for geographic range expansion by T. fluminensis under future climate scenarios. This indicates good prospects for biological control of T. fluminensis using N. abbreviata in South Africa.