Effects of short-term climatic conditions on the risk of bovine respiratory disease in preweaned calves on California dairies

Abstract

Bovine Respiratory Disease (BRD) significantly impacts dairy calf health and economics, yet the relationship between short-term weather patterns and BRD risk remains poorly understood, particularly across different seasons and regions in California. We explored the effects of short-term climatic conditions on BRD in preweaned dairy calves using machine learning approaches. Data from 11,470 calves across five California dairy farms were analyzed using tree-based models to investigate associations between BRD status and climate variables, including temperature, humidity, and temperature-humidity index (THI) over one to three days prior to diagnosis. The gradient boosting model achieved the best performance with an area under the receiver operating characteristic curve of 0.720 (sensitivity: 0.71, specificity: 0.73). Maximum relative humidity two days prior and maximum THI three days before BRD diagnosis were the most important short-term climatic predictors of BRD risk along with the region of California and calf age. The analysis indicated that summer conditions in Southern California were associated with higher BRD risk 48 h post-exposure; maximum THI below 75 in winter increased BRD risk 72 h post-exposure across all regions; and calves aged 20–30 days in Southern California showed higher BRD risk compared to Northern California across all seasons. These findings provide novel insights into how short-term weather patterns influence BRD development, potentially supporting the development of region-specific prevention strategies.