Journal of Applied Entomology · 2019 · 19 citations · 41 references
Life SciencesEcologyEntomologyInsect EcologyPlant-microbe InteractionsTree-crop InteractionsDisease DynamicsDisease EcologyDisease Risk AssessmentInsect Disease EcologyPest DynamicsVector EcologyAlternative Host PlantsDamage Prediction ModelsDisease Spread DynamicsMaximum Entropy EstimationMaximum Entropy ModelingModel Performance AnalysisModel Performance EvaluationPrediction
Abstract The African citrus triozid (ACT), Trioza erytreae Del Guercio, is a destructive pest particularly on citrus, and vectors, “ Candidatus ” Liberibacter africanus (CLaf), which is the causal agent of the African citrus greening disease. Our study seeks to establish the distribution and host‐plant relationship of ACT across citrus production areas in Kenya. We also modelled the risk of spread using the maximum entropy modelling algorithm with known occurrence data. Our results infer that ACT is widely distributed and causes severe damage to four alternative host plants belonging to the family Rutaceae. The adults, immature stages (eggs and nymphs), galls and the percentage of infested leaves were significantly higher in shaded than unshaded trees. However, adult ACTs preferred Kenyan highlands to Victoria Lake and coastal regions. The average area under the curve of the model predictions was 0.97, indicating an optimal model performance. The environmental variables that most influenced the prediction were the precipitation of wettest quarter, precipitation of wettest month, mean diurnal range, temperature seasonality and mean temperature of the coldest quarter. The current prediction of ACT exceeded its existing range, especially in the Western, Nyanza, Central, Rift valley and Eastern regions of Kenya. The model predicted a contraction of suitable habitats for a potential spread in 2040 with an inland shift to higher altitudes in the cooler regions. The potential for further expansion to climatically suitable areas was more pronounced for the 2080 forecast. These findings provide relevant information to improve monitoring/surveillance and designing IPM strategies to limit its spread and damage.
41
R: A Language and Environment for Statistical Computing
R Core Team · 2000 · 353K citations · Full text
Natural Sciences, Computer Science, Statistical Computing +10
Novel methods improve prediction of species’ distributions from occurrence data
Jane Elith, Catherine H. Graham, Robert P. Anderson et al. · Ecography · 2006 · 9.1K citations · Full text