Factors influencing scrotal temperature in the bull during simulated heat treatments and thermoneutral conditions as determined by multimodel inferencing.

The creation of any model is complex requiring vast amounts of data, typically gathered over a series of experiments. Specifically the temperature humidity index (THI) and heat load index (HLI) are used as management tools to implement mitigation strategies during hot climatic conditions. Exposure of the testes to hot climatic conditions has a negative impact on spermatogenesis in the bull, and other species. Despite the bull having effective scrotal thermoregulatory ability, these mechanisms can breakdown when the bull is under heat stressed causing scrotal temperature to increase. The ability to understand the effect climatic conditions have on scrotal temperature could be a valuable tool for producers. Six Wagyu bulls were exposed to series of heat treatments, acute and chronic with thermoneutral periods in between, in climate controlled rooms. Abdominal and scrotal temperature were continuously recorded via surgically implanted data loggers. Ambient temperature (AT) and relative humidity (RH) throughout the same periods were monitored. An information theoretic approach was used to analyse the data. Data were split into three groups, one for each of the ambient conditions of acute, chronic and thermoneutral. A global model for each of these groups was constructed using generalised additive models and included all parameters of interest; abdominal temperature, AT, RH and THI. Additional candidate models consisting of subsets of these global models were run for each acute, chronic and thermoneutral. Models were statistically compared and multimodel inferencing used to determine the effect these parameters had on scrotal temperature. The best model for acute included abdominal temperature, RH and an autoregressive factor (AR(1)) whilst the best model for both chronic and thermoneutral included just abdominal temperature and AR(1). Adding abdominal temperature to the model for the acute and chronic treatments improved the model strength with an increase in abdominal temperature increasing scrotal temperature. In contrast, the addition of abdominal temperature to the thermoneutral models did not consistently improve their strength. Despite a small sample size, the data indicates that scrotal temperature can be modelled. The addition of future observations to these models will continue to strengthen models, improving their usability.