Maximum Entropy Niche Modelling to Estimate the Potential Distribution of Phytophthora megakarya (Brasier & M. J. Griffin) in Tropical Regions

Abstract

Background: Phytophthora megakarya is an invasive pathogen endemic to Central and West Africa. This species causes the most devastating form of black pod disease. Despite the deleterious impacts of this disease on cocoa production, there is no information on the geographic distribution of P. megakarya. Aim: In this study, we investigated the potential geographic distribution of P. megakarya in cocoa-producing regions of the world using ecological niche modelling. Methods: Occurrence records of P. megakarya in Central and West Africa were compiled from published studies. We selected relevant climatic and edaphic predictor variables in the indigenous range of this species to generate 14 datasets of climate-only, soil-only, and a combination of both data types. For each dataset, we calibrated 100 candidate MaxEnt models using 20 regularisation multiplier values and five feature classes. The best model was selected from statistically significant candidates with an omission rate ≤ 5% and the lowest Akaike Information Criterion corrected for small sample sizes, and projected onto cocoa-producing regions in Southeast Asia, Central and South America. The risk of extrapolation in model transfer was measured using the mobility-oriented parity (MOP) metric. Results: We found an optimal goodness-of-fit and complexity for candidate models incorporating both climate and soil data. Predictions of the model with the best performance showed that nearly all of Central Africa, especially areas in Gabon, Equatorial Guinea, and southern Cameroon are at risk of black pod disease. In West Africa, suitable environments were observed along the Atlantic coast, from southern Nigeria to Gambia. Our analysis suggested that P. megakarya is capable of subsisting outside its native range, at least in terms of climatic and edaphic factors. Model projections identified likely suitable areas, especially in Brazil and Colombia, from southwestern Mexico down to Panama, and across the Caribbean islands in the Americas, and in Sri Lanka, Indonesia, Malaysia, and Papua New Guinea in Asia and adjacent areas Conclusion: The outcomes of this study would be useful for developing measures aimed at preventing the spread of this pathogen in the tropics.