Ensemble approach for potential habitat mapping of invasive Prosopispp.inTurkana,Kenya

Abstract

Aim:Prosopisspp.areaninvasivealienplantspeciesnativetotheAmericasandwell adapted to thrive in arid environments. In Kenya, several remote‐sensing studies conclude that the genus is well established throughout the country and is rapidly in‐ vadingnewareas.ThisresearchaimstomodelthepotentialhabitatofProsopisspp. by using an ensemble model consisting of four species distribution models. Furthermore, environmental and expert knowledge‐based variables are assessed. Location:TurkanaCounty,Kenya.Methods:Wecollectedandassessedalargenumberofenvironmentalandexpertknowl‐ edge‐based variables through variable correlation, collinearity, and bias tests. The varia‐ bleswereusedforanensemblemodelconsistingoffourspeciesdistributionmodels:(a) logisticregression,(b)maximumentropy,(c)randomforest,and(d)Bayesiannetworks.The modelswereevaluatedthroughablockcross‐validationprovidingstatisticalmeasures. Results: The best predictors for Prosopis spp. habitat are distance from water and built‐upareas,soiltype,elevation,lithology,andtemperatureseasonality.Allspecies distributionmodelsachievedhighaccuracieswhiletheensemblemodelachievedthe highest scores. Highly and moderately suitable Prosopis spp. habitat covers 6% and 9% of the study area, respectively. Main conclusions: Both ensemble and individual models predict a high risk of continued invasion,confirminglocalobservationsandconceptions.Findingsarevaluabletostake‐ holdersformanaginginvadedarea,protectingareasatrisk,andtoraiseawareness.