Abstract
Background Although fires can cause tree mortality or reduce post-fire growth in trees of all ages, and models exist that predict fire-induced mortality in mature trees, the development of predictive models of how fires impact younger trees has received less attention.Aims To assess whether inclusion of fire behaviour metrics alongside pre- and post-fire sapling morphological traits improve the prediction of fire-induced tree mortality as compared to existing models.Methods In this study, we subjected Pseudotsuga menziesii (Mirb.) Franco var. glauca (Beissn.) and Pinus monticola var. minima Lemmon saplings to increasing levels of fire intensity and evaluated models to predict immediate and delayed post-fire mortality.Key results For Pinus monticola, the optimal model relied on the post-fire crown volume scorched, while for Pseudotsuga menziesii the optimal model used flame height and fire radiative energy. We show that while Pinus monticola saplings exhibit immediate fire-induced mortality, Pseudotsuga menziesii saplings are prone to delayed fire-induced mortality.Implications Even in younger trees, crown volume scorched and related metrics remain consistent predictors of fire-induced tree mortality. Future studies should track mortality over extended periods to ensure that developed models better represent delayed fire-induced tree mortality.