PS 64-2 - Integrating climate, connectivity, and trade data layers to understand the risk of Mediterranean fruit fly invasions in Haiti and the Caribbean

Friday, August 16, 2019
Exhibit Hall, Kentucky International Convention Center
Joubert Fayette1,2, Xing Yanru1,2, Wanita Dantes1,2, James C. Fulton1,2, Margareth Divers3, Nicolas Carvil4 and Karen A. Garrett1,2, (1)Institute for Sustainable Food Systems, University of Florida, (2)Plant Pathology, University of Florida, (3)USAID AREA Project, Haiti, (4)Faculte d'Agronomie et de Medecine Veterinaire, Universite d'Etat d'Haiti, Haiti
Background/Question/Methods

Fruit flies (Diptera: Tephritidae) are one of the most important pests in the world. Haiti is among the twenty greatest mango producers globally. Since 2008, a nation-wide program has been implemented to detect and control fruit flies, to protect the mango export economy in Haiti and to prevent the introduction of the Mediterranean fruit fly, Ceratitis capitata. The objective of this study is to guide management practices that could limit or delay the introduction of C. capitata in Haiti using a risk analysis framework based on a species distribution model, cropland connectivity, and trade. For ecological niche modeling, two modeling methods, the Biomod2 package and maximum entropy species distribution modeling (Maxent) were used to predict the potential areas of suitability of C. capitata in the Caribbean, using distribution records from the Global Biodiversity Information Facility (GBIF) and environmental predictor variables (WorldClim). The cropland connectivity risk index (CCRI) is based on a weighted mean of four measures of node centrality in network analysis. It evaluates the connectivity between areas of mango production in the Caribbean. Trade networks were evaluated based on United Nations Commodity Trade Statistics Database summaries. These three geographic risk layers were integrated in a risk summary.

Results/Conclusions

A significant portion of mango production in Haiti in located in the north. The results of the species distribution modeling suggest that there are patches of suitable environment for C. capitata toward the north and toward the south in Haiti. For mango cropland connectivity, Haiti is an important location, with high CCRI and a potential role as a bridge location connecting neighboring countries. These risk layers were integrated in a risk analysis framework, along with other important risk factors including trade. The major ports of Haiti are Port-au-Prince, Cap Haitien, Saint Marc, and Gonaive, with corresponding risk for pest introduction. We evaluate how these ports are linked to mango exporting countries. Combined risk maps can inform national strategies for disease and pest management in formal pest risk assessments (PRAs) in advance of potential introduction and establishment events. Focusing on the higher risk areas can make efficient use of limited resources for sampling and mitigation of pest and pathogen invasions.