Ecological Niche Modeling: An Introduction for Veterinarians and Epidemiologists

Abstract

Most infectious diseases in animals are not distributed randomly. Instead, diseases in livestock and wildlife are predictable in terms of the geography, time, and species affected. Ecological niche modeling approaches have been crucial to the advancement of our understanding of diversity and diseases distributions. This contribution is an introductory overview to the field of distributional ecology, with emphasis on its application for spatial epidemiology. A new, revised modeling framework is proposed for more detailed and replicable models that account for both the biology of the disease to be modeled and the uncertainty of the data available. Considering that most disease systems need at least two organisms interacting (i.e., host and pathogen), biotic interactions lie at the core of the pathogen's ecological niche. As a result, neglecting interacting organisms in pathogen dynamics (e.g., maintenance, reproduction, and transmission) may limit efforts to forecast disease distributions in veterinary epidemiology. Although limitations of ecological niche modeling are noted, it is clear that the application and value of ecological niche modeling to epidemiology will increase in the future. Potential research lines include the examination of the effects of biotic variables on model performance, assessments of protocols for model calibration in disease systems, and new tools and metrics for robust model evaluation. Epidemiologists aiming to employ ecological niche modeling theory and methods to reconstruct and forecast epidemics should familiarize themselves with ecological literature and must consider multidisciplinary collaborations including veterinarians to develop biologically sound, statistically robust analyses. This review attempts to increase the use of tools from ecology in disease mapping.

Environmental Drivers and Potential Distribution of Schistosoma mansoni Endemic Areas in Ethiopia

“…The 95 endemic sites were treated as ‘occurrence’ points and were georeferenced ( Figure 1 ) for subsequent analyses. However, some ENM techniques used in this study require a sufficient number of both presence and absence data to develop the models [ 16 ]. To follow widely accepted practices in the field of ecology, we augmented the presence data with pseudoabsence data points for the niche modeling [ 17 , 18 ].…”

Enceladus as a Potential Niche for Methanogens and Estimation of Its Biomass

“…Nonetheless, we can envision ENMs at multiple spatiotemporal scales, as long as these are recognized and properly interpreted [25]. For example, the niche of a tick might be assessed within the skin of a host (fine-scale) or across biomes within a continent (coarse scale) [26], or the spores of Bacillus anthracis, the causative agent of anthrax, may respond to environmental cues occurring at micro scales [27], yet it also shows associations with environmental variation at coarse scales [28], which can be characterized and used to predict areas of potential occurrence [29]. In this study, we apply the same principle to methanogens on the Earth, while characterizing its niche at a coarse-scale using correlative models that look for associations between environmental variables and the occurrence of organisms.…”

Ecological niche modeling predicting the potential distribution of African horse sickness virus from 2020 to 2060

“…These characteristics can be combined with disease data to reconstruct or anticipate the regional spread of environmental (e.g., anthrax), vector-borne (e.g., Bluetongue disease), and directly transmitted (e.g., rabies) diseases. On the other hand, these investigations necessitate a fundamental knowledge of Geographic Information Systems, spatial statistics, and a thorough grasp of the biology of the disease system to be modeled 6 .…”