Laboratory fire simulations overestimate germination responses compared to controlled burns in cactus species

Abstract Background Studies on plant germination responses to fire have primarily relied on laboratory simulations, typically using oven-based heat shock treatments and liquid smoke applications. However, the extent to which these approaches reproduce real fire conditions remains debated, and an alternative experimental approach involves the use of controlled biomass burning. In this study, we compared germination responses under controlled biomass burns with those under equivalent laboratory treatments, using cacti as a model system. Cacti are conspicuous components of arid and semiarid ecosystems across the Americas and are increasingly threatened by anthropogenic pressures, including changes in fire regimes. Despite much of their distribution occurring in fire-prone environments, few studies have evaluated their germination responses to fire. We assessed germination (percentage and mean germination time) of 16 cactus species using two experimental approaches across seven treatments: a control; four biomass burning treatments with different fuel loads (50 g, 50 g + ash, 100 g, and 100 g + ash); and two equivalent laboratory treatments (100 °C/5 min and 130 °C/5 min), both combined with liquid smoke. We also evaluated the effect of seed mass on germination. Results Germination responses differed markedly between experimental approaches. Under biomass burning treatments, germination was consistently very low—often null—across all species, whereas laboratory treatments elicited germination responses in most species. Under the moderate-intensity laboratory treatment (100 °C + smoke), responses did not differ significantly from the control. Mean germination time varied among species: in some cases it did not differ from the control, whereas in others it either increased or decreased under laboratory treatments. Seed mass had no significant effect on germination percentage, but significantly influenced mean germination time. Conclusions Overall, our results indicate that conventional laboratory experiments do not adequately reproduce real fire conditions and may substantially overestimate germination responses in cacti.