The spatiotemporal distribution of lumpy skin disease virus
Oral communication in GeoVet 2019. Novel spatio-temporal approaches in the era of Big Data
October 8th, 2019
Machado G., Korennoy F., Alvarez J., Picasso C., Perez AM. and VanderWaal K.
Lumpy Skin Disease (LSD) is a vector-borne disease of cattle caused by Capripoxvirus (Poxviridae family), referred to as LSD virus (LSDV). The morbidity rate varies between 10% and 20%, whereas the mortality rate does not usually exceed 5% (OIE, 2018). The main route of transmission often relates with mosquitoes and biting flies (Chihota et al., 2001; Tuppurainen et al., 2011; Tuppurainen & Oura, 2012). Therefore, LSDV transmission and spread linked to warm and humid weather conditions that are associated with high population densities of biting arthropods (Ochwo et al., 2018). Long-distance disease dispersal may also be facilitated by windborne carriage of vectors and their transportation via vehicles carrying hay and straw (Klausner et al., 2017).
Even though the global distribution of LSDV has been restricted to a few regions, the identification of potential geographic distributions has not yet been determined, therefore is a need to calculate the likelihood of this virus reaching free areas. One way to approximate LSDV’s potential distribution is by correlating environmental abiotic conditions with disease occurrence location, via ecological niche models, alternatively Bayesian hierarchical statistical approaches allow the inclusion of spatially and temporally explicit features in regression models and capture how disease risk can be related to proximity with areas experiencing outbreaks (Lawson, 2018). In this study, we developed a novel and integrative approach, by combination of ecological niche modeling and fine spatiotemporally explicit Bayesian hierarchical model on LSDV outbreak occurrence data to estimate of the underlying LSDV risk.
We retrieved and curated LSDV outbreak data from Middle Eastern, Central Asian, and Eastern European countries in 2014-2016. Outbreaks were defined as the detection of one or more cases of LSDV per location. We used ecological niche modeling and Bayesian space-time model in combination to estimate the LSDV spatial risk.
Variables related to the average temperature, precipitation, wind speed, as well as land cover and host densities were important drivers explaining the observed distribution of LSDV in both modeling approaches. Areas of elevated LSDV risks were identified mainly in Russia, Turkey, Serbia, and Bulgaria (Fig. 1). Results suggest that, if current ecological and epidemiological conditions persist, further spread of LSDV in Eurasia may be expected (Fig. 2).
The ecological niche model developed in this study demonstrated the ability to estimate the past distribution of LSDV. Further studies to confirm the capacities of ecological niche model to forecast the potential distribution of LSDV are needed. Future models should also include future climate scenarios in order to identify potential shifts in the disease distribution especially in areas of transitional elevation, which may become suitable for the occurrence of potential vectors. The spatiotemporal patterns of LSDV occurrences modeled by the Bayesian hierarchical approach were heterogeneous, with temperature and precipitation increasing the relative risk and stronger winds reducing risk. With this study, we identified hotspot areas by using two modeling approaches, demonstrating how integrated approaches can better guide disease control and active surveillance efforts
 | Department of Population Health and Pathobiology. College of Veterinary Medicine. North Carolina State University (NCSU). |
 | Federal Governmental Budgetary Institution - Federal Center for Animal Health (FGBI ARRIAH). |
 | Department of Veterinary Population Medicine. College of Veterinary Medicine. University of Minnesota (UMM). |
Related link:
The spatiotemporal distribution of lumpy skin disease virus
Link to GeoVet 2019. Novel spatio-temporal approaches in the era of Big Data
