ntbox: An r package with graphical user interface for modelling and evaluating multidimensional ecological niches

Abstract

1. Biodiversity studies rely heavily on estimates of species' distributions often obtained through ecological niche modelling. Numerous software packages exist that allow users to model ecological niches using machine learning and statistical methods. However, no existing package with a graphical user interface allows users to perform model calibration and selection based on convex forms such as ellipsoids, which may match fundamental ecological niche shapes better, incorporating tools for exploring, modelling, and evaluating niches and distributions that are intuitive for both novice and proficient users. 2. Here we describe an r package, NicheToolBox (ntbox), that allows users to conduct all processing steps involved in ecological niche modelling: downloading and curating occurrence data, obtaining and transforming environmental data layers, selecting environmental variables, exploring relationships between geographic and environmental spaces, calibrating and selecting ellipsoid models, evaluating models using binomial and partial ROC tests, assessing extrapolation risk, and performing geographic information system operations via a graphical user interface. A summary of the entire workflow is produced for use as a stand-alone algorithm or as part of research reports. 3. The method is explained in detail and tested via modelling the threatened feline species Leopardus wiedii. Georeferenced occurrence data for this species are queried to display both point occurrences and the IUCN extent of occurrence polygon (IUCN, 2007). This information is used to illustrate tools available for accessing, processing and exploring biodiversity data (e.g. number of occurrences and chronology of collecting) and transforming environmental data (e.g. a summary PCA for 19 bioclimatic layers). Visualizations of three-dimensional ecological niches modelled as minimum volume ellipsoids are developed with ancillary statistics.

Geographic potential of the world’s largest hornet,Vespa mandariniaSmith (Hymenoptera: Vespidae), worldwide and particularly in North America

“…S1 ). All analyses were done in R; specifically, raster processing was completed using the packages raster ( Hijmans et al, 2020 ), rgeos ( Bivand et al, 2020b ), and rgdal ( Bivand et al, 2020a ); PCA was performed using the ntbox package ( Osorio-Olvera et al, 2020 ).…”

Geographic abundance patterns explained by niche centrality hypothesis in two Chagas disease vectors in Latin America

“…We measured distances between observations of population abundances ( S2 Table ) and the niche centroid. To estimate the niche and its centroid, we used the minimum-volume ellipsoid (MVE) approach with the ellipsoid_selection function of the R package ntbox [ 53 ]. The ellipsoid_selection function has a model calibration and selection protocol that allows us to select niche models that are statistically significant and have good performance.…”

Influence of montane altitudinal ranges on species distribution models; evidence in Andean blow flies

“…After disposing of spatial autocorrelation with the R package ntbox ( Osorio-Olvera et al., 2020 ), the complete occurrence database included 228 records of presence, including asynanthropic and medium lowland (1,100–2,600 m.a.s.l) distribution Blepharicnema splendens (Diptera: Calliphoridae) (22) and high land (2,000–3,127 m.a.s.l) distribution Sarconesia roraima (Diptera: Calliphoridae) (15); hemisynanthropic and lowland (0–1,000 m.a.s.l) Chloroprocta idioidea (Diptera: Calliphoridae) (59) and medium lowland (1,100–2,700 m.a.s.l) distribution Lucilia purpurascens (Diptera: Calliphoridae) (30); eusynanthropic and high land (2,200–2,800) distribution Calliphora vicina (Diptera: Calliphoridae) (19) and lowland (0–1,400 m.a.s.l) Cochliomyia macellaria (Diptera: Calliphoridae) (83) ( Table S1 ). Presence records by species were obtained from the geographical information on specimens revised in eleven entomological museums, and recent literature on blow flies published since 2000.…”

“…The fine-tuned MaxEnt models were made by finding the lowest delta value of Akaike’s information criterion corrected for small sample sizes (AICc) among candidate models, which reflects both models, goodness-of-fit and complexity, providing the most conservative results ( Basanta, Rebollar & Parra-Olea, 2019 ) ( Table 1 ). Using the R package ntbox ( Osorio-Olvera et al., 2020 ), a ROC (Receiver Operating Characteristic) was estimated to assess the performance of the model ( Peterson, Papes & Sober, 2008 ). The model was assessed through calibration, randomly selecting 25% of the distribution data and comparing it with the threshold of the area under the curve (AUC).…”