Heat Load Forecasting

Considerable research has been devoted to forecasting heat load for animals in feedlots [84–86]. Heat load forecasting within the Australian feedlot industry has evolved over a period of two decades [87]. These developments have been used by the sea transport industry to generate a computer model that aims to assess the risk of heat stress and to contain mortality levels on livestock ships below certain arbitrary limits. The Heat Stress Risk Assessment (HSRA) model 'HotStuff' model was developed for the Australian livestock export industry to estimate and minimise the incidence of heat stress mortality in livestock during voyages to the Middle East [26]. The model has been in operation since 2003 and has had several refinements and reviews [51,83]. The HSRA model provides a framework from which to address heat stress and heat load. The model factors in the weather (both predicted and actual for the destination ports), the type, class and body weight of animals, and ship factors such as ventilation design and airflow. The latter requires input from the vessels' design specifications and is adjusted for each vessel prior to loading [26].

However, despite the sophistication of the model, it often fails to accurately predict a heat load event [87]. The accuracy of the HRSA model is currently being questioned, and the ability to prevent animals experiencing harmful heat load when travelling to the Middle East may be limited by the ventilation capacity of the vessels. Given mounting recent evidence of ongoing heat load events affecting exported sheep, it is argued that a review of the settings used for the input of data in addition to a review of the risk settings should be undertaken [14]. The current settings used is a 2% probability of a 5% mortality due to heat stress, and was chosen by industry [51,83]. It has recently been suggested that the risk setting should be replaced by the likelihood of an animal experiencing heat stress, not mortality, in order to achieve improved welfare outcomes [14].

It has also been suggested that revision of the HSRA model should include consideration of the different heat load thresholds, with examination of the appropriateness of using 0.5◦C rise in body temperature, the WBT at which that occurs, and the duration of that episode, as an arbitrary threshold [13]. Phillips [12] criticised the HST values used within the HSRA model as being above those suggested by animal studies. It is important to note that some of the discussion in recent reviews regarding heat stress thresholds (HSTs) do not appear to fully appreciate the distinctions between how different researchers have expressed the thresholds. For example, Stockman [20] described three different HSTs. The HSRA further iterates ML for different classes of stock, from a base animal of specified type, weight, and normal body temperature. Ferguson, et al. [51] did not criticise the methodology or base values in their review of the model.

The HSRA model does not ye<sup>t</sup> have the capacity to deal with the effects of cumulative heat load. It would be advisable for future revisions of the model to use expertise from the feedlot industry to consider the influence of duration of heat exposure and the capacity for respite to influence effects of heat load.
