SPATIO-TEMPORAL DISTRIBUTION AND SPACE-TIME CLUSTER ANALYSIS OF LUMPY SKIN DISEASE OUTBREAKS IN CATTLE IN SELECTED ZONES OF OROMIA REGION, ETHIOPIA

Abstract

Lumpy Skin Disease (LSD) is a viral infection that affects cattle, causing significant economic losses and posing a threat to food security. A study in the selected zones of Oromia Region, Ethiopia, spanning 15 years (2008-2022), aimed to identify spatiotemporal distribution, cluster of LSD outbreaks and to project potential future outbreaks for the years from 2023-2027. The study analyzed LSD outbreak spatial cluster (Moran I, Getis Ord Gi and ST-model), time series data using classical additive and STL decomposition, and four forecasting models (ARIMA, SARIMA, ETS, and SL+random walk) were used. The findings revealed 457 outbreaks, with 879 fatalities among 50,025 recorded cases. The mortality, morbidity, and case fatality rates stood at 0.034%, 1.9%, and 1.8% respectively. Notably, seven hotspots emerged predominantly in the northwestern and eastern segments of the Arsi district. The study further pinpointed two spatial and twelve spatiotemporal clusters, including all distinct temporal cluster between January 1, 2009, and December 31, 2009, characterized by a relative risk (RR) of 2.68, a log likelihood ratio (LLR) of 16.23, and a statistically significant P-value of 0.001. The research underscored pronounced disparities in LSD outbreaks across various zones, with the Arsi district bearing 57.5% of the outbreaks within the survey timeframe. Seasonal trends indicated that LSD peaks during the wet months from September to December and low in cold dry March to May period. Among the forecasting methodologies evaluated, the SARIMA (1, 1, 1) (0, 2, 3) [12] model was best fit its counterparts, as reflected by the lowest RMSE, MA, and MASE, suggesting enhanced forecast accuracy for LSD outbreaks from 2019 to 2024. In contrast, the STL + random walk model ineffective for this specific data set. These findings was provide valuable insights into the dynamics of the disease within the study area. These insights can inform the development of effective strategies for LSD control and prevention, in cattle in the study zones. However, further research will needed on LSD serosurveillance and molecular characterization of the virus strains in the study area.