Environmental Science
Climate Change Set to Worsen Facial Eczema Risk in New Zealand
Original Authors: Wada, Sagarasaeranee, Cogger, Marshall, Cuttance, Macara, Sood, Vallee
A new study leveraged machine learning to uncover the critical climate variables influencing *Pseudopithomyces chartarum* spore counts—a primary risk factor for facial eczema in livestock—and to forecast these counts in New Zealand up to 2100. Researchers integrated extensive longitudinal spore count data (6,975 counts) collected from 862 paddocks spread over 102 North Island farms between 2010 and 2017 with historical and projected climate information. This robust dataset allowed for the application of advanced models, including random forest, to identify key environmental drivers and project future disease risk.
The sophisticated random forest model demonstrated an impressive 80% accuracy in classifying low or high facial eczema risk. It pinpointed several climate factors as significantly important, including soil temperature at 10 cm depth, solar radiation, potential evapotranspiration, vapour pressure, soil moisture, and minimum temperature. Crucially, the model projects a concerning future: as lead author Wada notes in the paper, "Over the next 80 years, our model predicted an increase in the seasonal mean spore counts in the study farms by a mean of 17% (min 6, max 30%) under the high-end greenhouse gas emission scenario (representative concentration pathways (RCP) 8.5)." Furthermore, the probability of high-risk spore counts (> 20,000) is expected to rise by 14-22% each decade under moderate to high emission scenarios.
These findings indicate a widespread increase in peak spore counts, particularly across the entire North Island and three districts in the South Island, by 2100. Such precise predictions offer invaluable insights for stakeholders, enabling them to pinpoint high-risk geographical areas and implement targeted mitigation or adaptation measures against facial eczema. This study not only highlights the escalating environmental health challenges posed by climate change but also underscores the immense value of ecological data and predictive modeling in enhancing preparedness and resilience for future environmental disease risks.