Fire‐released seed dormancy ‐ a global synthesis

Abstract

Seed dormancy varies greatly between species, clades, communities, and regions. We propose that fireprone ecosystems create ideal conditions for the selection of seed dormancy as fire provides a mechanism for dormancy release and postfire conditions are optimal for germination. Thus, fire-released seed dormancy should vary in type and abundance under different fire regimes. To test these predictions, we compiled data from a wide range of fire-related germination experiments for species in different ecosystems across the globe. We identified four dormancy syndromes: heat-released (physical) dormancy, smoke-released (physiological) dormancy, non-fire-released dormancy, and non-dormancy. In fireprone ecosystems, fire, in the form of heat and/or chemical by-products (collectively termed 'smoke'), are the predominant stimuli for dormancy release and subsequent germination, with climate (cold or warm stratification) and light sometimes playing important secondary roles. Fire (heat or smoke)-released dormancy is best expressed where woody vegetation is dense and fires are intense, i.e. in crown-fire ecosystems. In such environments, seed dormancy allows shade-intolerant species to take advantage of vegetation gaps created by fire and synchronize germination with optimal recruitment conditions. In grassy fireprone ecosystems (e.g. savannas), where fires are less intense but more frequent, seed dormancy is less common and dormancy release is often not directly related to fire (non-fire-released dormancy). Rates of germination, whether controls or postfire, are twice as fast in savannas than in mediterranean ecosystems. Fire-released dormancy is rare to absent in arid ecosystems and rainforests. The seeds of many species with fire-released dormancy also possess elaiosomes that promote ant dispersal. Burial by ants increases insulation of seeds from fires and places them in a suitable location for fire-released dormancy. The distribution of these dormancy syndromes across seed plants is not randomcertain dormancy types are associated with particular lineages (phylogenetic conservatism). Heat-released dormancy can be traced back to fireprone floras in the 'fiery' mid-Cretaceous, followed by smoke-released dormancy, with loss of fire-related dormancy among recent events associated with the advent of open savannas and non-fireprone habitats. Anthropogenic influences are now modifying dormancy-release mechanisms, usually decreasing the role of fire as exaptive effects. We conclude that contrasting fire regimes are a key driver of the evolution and maintenance of diverse seed dormancy types in many of the world's natural ecosystems.

Ancient Rhamnaceae flowers impute an origin for flowering plants exceeding 250-million-years ago

“…This is associated with heat-released seed dormancy for the Ziziphoid ancestor at 259 Ma (p = 0.81) but not for Rhamnoideae-Ampeloziziphoideae (p = 0.26). The fire-prone environment but rarity of seed dormancy in this latter clade implies a monsoon-dominated past ( Pausas and Lamont 2022 ) consistent with the savanna setting of many extant species assessed here ( Figure S1 ). These results now add an important fire-related trait to the growing list of ancient adaptations to fire ( He et al., 2016 ; Lamont 2022 ).…”

“…Fire-released seed dormancy, accompanied by arillate seeds that are ant-dispersed/buried, is well-established among the Ziziphoideae ( Pausas and Lamont 2022 ) and our ancestral trait reconstructions show a strongly fire-prone history for this subfamily, with a probability (p) of 0.95 ( Figure S1 ). This provides extra support that the Phylicieae occupied the fire-prone shrub stratum of open vegetation (fynbos), 110−100 Ma, as suggested by Shi et al., 2022 on the basis of charcoal concurrently in amber of similar ages to their fossils.…”

“…The estimates of Belcher et al., 2010 that the probability of ignition of plant matter reached an all-time high of 0.90–1.00 in the 250–340 Ma period is consistent with our findings. A prerequisite for the evolution of fire-type, heat-released seed dormancy are hard (impermeable) seeds ( Pausas and Lamont 2022 ). Such a likelihood is as strong as for fire- proneness in the Rhamnaceae, essentially on the Ziziphoid side (p = 0.95) ( Figure 1 ).…”

“…Fire-related traits (presence/absence) for each species in the molecular phylogeny were collated from data in Pausas and Lamont (2022) and numerous websites (identity of these additional sources available on request). Hard-seededness was gauged as impermeable to water or evidence of a sclerified seed coat (these were sometimes referred to the endocarp and not the testa), heat-released dormancy is given as fire-stimulated germination in the results and required seeds treated with fire-type heat to have exceeded germination of the controls, and fire-proneness refers to occurrence of the species in vegetation likely to burn within its lifetime.…”

Fire and summer temperatures interact to shape seed dormancy thresholds

“…Its function is to prevent germination until specific external stimuli signal that conditions are favourable for seedling establishment ( Thompson and Ooi, 2010 ; Baskin and Baskin, 2014 ). There is extensive evidence that both heat shocks and smoke from wildfires break seed dormancy and stimulate germination in a wide range of species ( Keeley and Fotheringham, 2000 ; Moreira et al , 2010 ; Pausas and Lamont, 2022 ). Heat-released dormancy occurs in physically dormant seeds that form long-lived, fire-resistant soil seed banks, and have temperature requirements for breaking dormancy that typically match fire heat ( Moreira and Pausas, 2012 ; Pausas and Lamont, 2022 ).…”