Accurate prediction of wildland fire rate of spread (ROS) is essential for effective fire management, yet incorporating the effects of terrain slope remains a persistent challenge. In this study, a selection of five slope correction models in combination with five “basic” ROS models (i.e. models that do not account for the effect of slope) and their predictions are validated and comparatively assessed against a comprehensive dataset of 184 laboratory experiments conducted with pine needle litter on both horizontal and inclined terrain, found in the open literature. The analysis shows that predictions for horizontal spread are generally reliable, with most models performing within an acceptable error range. Incorporating fuel bed width and dimensionless descriptors that integrate fuel particle and bulk properties substantially improves predictive capability. For inclined terrain, model performance depends strongly on slope angle. At moderate slopes (≤ 20°), predictions are consistent and largely conservative. At steeper slopes (> 20°), however, all evaluated methods systematically under-predict ROS, with errors increasing by an order of magnitude. This deficiency reflects the inability of current formulations to capture key slope-driven physical mechanisms, including upslope-induced winds, flame tilt and entrainment effects, phenomena that have been reported in relevant studies. The findings highlight the need to refine ROS models by explicitly incorporating slope-related fire-flow interactions to ensure robust predictions. • Five basic ROS and five slope correction models are assessed against experiments • Identified key fuel bed descriptors improve horizontal ROS predictions • Slope correction models lose their predictive accuracy at slopes higher than 20 o -25 o • Fire-induced flow related effects must be incorporated in future models
Pallikarakis et al. (Sun,) studied this question.