He&Lamont 2019 Fire As A Key

Biol. Rev. (2019), pp. 000–000.
1
doi: 10.1111/brv.12544
Fire as a key driver of Earth’s biodiversity

Tianhua He1,2∗ , Byron B. Lamont1 and Juli G. Pausas3
1
School of Molecular and Life Sciences, Curtin University, Perth, Australia
2
College of Science, Health, Engineering and Education, Murdoch University, Murdoch, Australia
3
CIDE-CSIC, Valencia, Spain
ABSTRACT
Many terrestrial ecosystems are fire prone, such that their composition and structure are largely due to their fire regime.
Regions subject to regular fire have exceptionally high levels of species richness and endemism, and fire has been
proposed as a major driver of their diversity, within the context of climate, resource availability and environmental
heterogeneity. However, current fire-management practices rarely take into account the ecological and evolutionary
roles of fire in maintaining biodiversity. Here, we focus on the mechanisms that enable fire to act as a major ecological
and evolutionary force that promotes and maintains biodiversity over numerous spatiotemporal scales. From an
ecological perspective, the vegetation, topography and local weather conditions during a fire generate a landscape with
spatial and temporal variation in fire-related patches (pyrodiversity), and these produce the biotic and environmental
heterogeneity that drives biodiversity across local and regional scales. There have been few empirical tests of the
proposition that ‘pyrodiversity begets biodiversity’ but we show that biodiversity should peak at moderately high levels
of pyrodiversity. Overall species richness is greatest immediately after fire and declines monotonically over time, with
postfire successional pathways dictated by animal habitat preferences and varying lifespans among resident plants.
Theory and data support the ‘intermediate disturbance hypothesis’ when mean patch species diversity is correlated with
mean fire intervals. Postfire persistence, recruitment and immigration allow species with different life histories to coexist.
From an evolutionary perspective, fire drives population turnover and diversification by promoting a wide range of
adaptive responses to particular fire regimes. Among 39 comparisons, the number of species in 26 fire-prone lineages is
much higher than that in their non-fire-prone sister lineages. Fire and its byproducts may have direct mutagenic effects,
producing novel genotypes that can lead to trait innovation and even speciation. A paradigm shift aimed at restoring
biodiversity-maintaining fire regimes across broad landscapes is required among the fire research and management
communities. This will require ecologists and other professionals to spread the burgeoning fire-science knowledge
beyond scientific publications to the broader public, politicians and media.
Key words: fire regime, pyrodiversity, biodiversity, environmental heterogeneity, intermediate disturbance hypothesis,
patch mosaic burning.
CONTENTS
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
II. Fire regime and pyrodiversity explained . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
III. Fire as a driver of biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
(1) Individual fire components as drivers of biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
(2) Diversity through time: fire resets community dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
(3) The intermediate disturbance hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
(4) The pyrodiversity–biodiversity hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
(5) Mixed fire regime and intermediate disturbance hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
(6) Fire and biodiversity at broad scales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
IV. Fire as an evolutionary driver of biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
(1) Fire-driven evolution and adaptation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
(2) Fire and diversification of plant lineages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
* Author for correspondence (Tel.: +61 8 9457 7164; E-mail: [email protected])
Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

2 Tianhua He and others
(3) Mutagenic effects of heat and chemicals released by combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

V. Implications for biodiversity conservation and fire management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
VI. Future research directions and outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
VII. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
VIII. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
IX. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
X. Supporting Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
I. INTRODUCTION Lamont & Downes, 2011; He et al., 2012; He et al., 2016),

and has led to the development of an extraordinary diversity
Many terrestrial ecosystems are fire prone, and fire controls of plants (Rundel et al., 2016). Studies on fire-stimulated
their composition and structure (Bond, Woodward & responses under a wide range of fire regimes, and on
Midgley, 2005; Krawchuk et al., 2009; Pausas & Keeley, the role of fire in the ecology and evolution of non-plant
2009; Pausas & Ribeiro, 2013; Archibald et al., 2018; He & groups (fungi, insects, birds and mammals), are emerging
Lamont, 2018a). Where these ecosystems have a long history (McMullan-Fisher et al., 2011; Hutto et al., 2016; Stawski
of recurrent fire (over geological time scales), they harbour et al., 2017; Pausas & Parr, 2018; Carbone et al., 2019). The
exceptionally high levels of species richness and endemism long co-existence between these organisms and fire suggests
(Bond & Parr, 2010; Rundel et al., 2016; Fernandes et al., that they have adaptive traits allowing them to persist in
2018) and are considered biodiversity hotspots (Myers et al., fire-prone environments. Given that fire may kill organisms
2000). Environmental heterogeneity and resource availability while at the same time opening up new niches, our central
(climate, soils) have long been regarded as the most important question is: to what extent do different fire regimes shape
factors that determine species richness at the regional scale species diversity among and within biomes globally? Species
(O’Brien, 1998; Francis & Currie, 2003; Hawkins et al., diversity is shaped by multiple drivers under the umbrella of
2003; Kreft & Jetz, 2007). For example, the exceptionally environmental factors, disturbance, and species interactions
high species richness of the Cape flora in South Africa (e.g. Schnitzler et al., 2011; Stein, Gerstner & Kreft, 2014;
has been hypothesised to be the result of geographic Van der Niet, Peakall & Johnson, 2014; Cramer & Verboom,
and parapatric radiation in the region via a mosaic of 2017; Pausas & Lamont, 2018); their relative roles may
distinctive habitats due to localised soils, microclimates and depend both on their intensity and history. Here we focus
altitudinal variations (Linder, 2005; Schnitzler et al., 2011). on fire as an agent of disturbance and the mechanisms that
The relatively stable post-Miocene history and establishment enable recurrent fire to act as a significant ecological and
of semi-arid and highly seasonal (Mediterranean-type) evolutionary force that promotes and maintains biodiversity
climates in southwestern Africa have also been assumed to be over different spatiotemporal scales. To do so, we address a
major contributors to diversity (Goldblatt & Manning, 2000). range of factors, including time-since-fire, mixed fire regimes,
Although a role for the presence of wildfire as a stimulus for intermediate disturbance, and pyrodiversity, as they impact
species diversification was proposed long ago (Mutch, 1970), biodiversity at the landscape scale, through to fire’s direct
it has only received attention more recently (He, Lamont role as a mutagenic agent.
& Manning, 2016; Rundel et al., 2016, 2018) and has been
shown to have a dominant, even pre-emptive, role compared
with other likely environmental constraints (Hu et al., 2018;
Lamont, He & Yan, 2019b). In attempts to understand the II. FIRE REGIME AND PYRODIVERSITY
distribution and diversity of species, disturbance by fire as EXPLAINED
a key agent of selection has long been overlooked, and
continues to be so (Pausas & Lamont, 2018; Magadzire et al., Each particular fire, which we refer to as a fire event, has
2019; Pausas & Bond, 2019). However, there is increasing a unique set of properties characterised by its spatial and
evidence to suggest that prevailing fire regimes make a temporal dimensions (Bradstock, 2008). The fire event is
major contribution to explaining plant diversity at both the controlled by the type of vegetation and its moisture state,
community (Burkle, Myers & Belote, 2015; Hu et al., 2018) the topographic features of the fire-prone area, and the
and global (Pausas & Ribeiro, 2017) scales compared with local weather conditions during the fire event. A given fire
other possible constraints. event can be defined by: fire intensity (heat per unit time
Within the wide range of agents of selection available generated by the fire), size (total burnt area), its horizontal
(e.g. climate, soil, pollinators, microsymbionts, herbivores), (patchiness) and vertical (strata burnt) spread pattern, the
a role for fire in the diversification of floras has long been time since (the previous) fire, and the season in which it
postulated (Wells, 1969; Cowling, 1987). It is now known occurred (Fig. 1). These conditions define the type of fire (fire
that fire has been a leading determinant of species diversity in type), in terms of which of the two extremes of each of the six
fire-prone ecosystems over the last 100 million years (Simon components is exhibited by the fire, and it is independent of
et al., 2009; Bytebier et al., 2011; Crisp et al., 2011; He, the ignition source (Fig. 1). Note that not all combinations

Fire as key driver of biodiversity 3
fuel distribution, ignition source

vegetation structure, weather
patchy continuous
rai
n
Horizontal spread
su
w
nfa
e
cro
mm
tur
ll,
uc
er/
tio her
ign
d
au
str
a
e
itio
Se
at
tu
c
pre
rfa
n
m
we
ns
a
n
su
so
ls
a
ou
wi
t
ge
na
n te
i ca
rce
e
ac
ve
r/s
li
er t
urf
pri
y
b- s
ng
fire type/regime 1
V
su
)
ure
h
hig
sm
ity
r at
al l
F
t op
ns
pe
ad
i re
e
nt e
t em
og
ve
l lo
t ur
siz
rap n stru
g
uc
fue
eta
ei
,
fire type/regime 2
a ll
e
str
hy,
Fir
t io
er,
inf
ion
lar
(ra
at h
fue ture
low
g
t at
Fire interval
e
we
l lo
ate
ge
c
ad
clim
short long
ve
climate (seasonality, temperature)
vegetation structure, fuel load
Fig. 1. The six components (the two ‘core’ components in red) of individual fires (fire events) and environmental factors that control
the properties of each component to define the fire type. Ignition source (lightning, anthropogenic) may affect the properties of each
component but is not itself a component of fire. ‘Fire interval’ refers to to the time since last fire for individual fires and mean fire
interval for the fire regime. The fire regime arises from repeated patterns (means plus variance) over time of the properties of the
components for each fire. For a certain vegetation structure under a given climate, fire regime is relatively predictable (as indicated
by the circles) and selects for an adapted group of plants, microflora and associated fauna.
of descriptors are possible and that there are correlations variation about the mean (stochasticity) that can have major
among descriptors. A given fire creates a landscape patch, long-term effects on plant responses (Enright et al., 1998).
and this patch has a fire history. Parameters that define the Although rarely formalised, this can be represented by the
temporal component are: (i) time since last fire (fire age), coefficient of variation among fire intervals.
(ii) mean fire interval (mean time between successive fires), Fire ‘intensity’ should not be confused with fire ‘severity’.
and (iii) variability about this mean (Fig. 2). Since it can Severity refers to the impact of the fire event on the associated
only be recognised from the imprint of the last fire on the biota and is often gauged as the fraction of biomass that is
vegetation, a patch is usually defined by the area covered consumed by the fire or a surrogate for that, e.g. the mortality
by that fire and the time since the previous fire occurred. rate of the affected species (Keeley, 2009; Taille et al.,
A patch may also be recognised as an unburnt area within 2018). Consequently, severity may depend on the species
the larger confines of the total area burnt by more recent survival traits (e.g., bark thickness, bud bank) and vegetation
fires. The distribution of different fire-related patches forms characteristics (e.g. density, continuity, flammability), while
a mosaic over the reference area. The reference area usually intensity is a physical component of the fire itself. Severity
refers to a stated community type, but can be extended to is often closely associated with fire intensity but also with
the landscape or regional scales. This mosaic varies through extremes in other fire properties, e.g. unusually frequent fires
time and space as more fires occur in different locations might also be regarded as having severe effects on the viability
and time since fire increases until a new fire occurs. When of particular biota. Alternatively, occasional extremes in
records are available over many years, the mean fire interval fire resilience and intensity justify distinguishing the two
can be inverted to give fire frequency, the number of fire concepts, such as historically fire-protected (non fire-prone)
events per time interval that can include time since last fire vegetation failing to recover from even low-intensity fires
(Fig. 2). However, neither parameter gives any indication of (Cochrane & Barber, 2009; Balch et al., 2015). Of some

Fire history of a given location in a reference area
Fire regime (of reference area)
Fire event F1 F2 F3 F4 Fire patch

Assessment
i1 i2 i3 Time since time
Fire interval
last fire
Fire type T1 T2 T3 T2
Mean fire interval:X/3

Fire over time X
Y Fire frequency: 4/Y
Fig. 2. Illustrations of time-related fire concepts. Recurrent fires occur at each location in a reference area (fire events, F) and
usually vary in their properties – see Fig. 1 for the six components that together describe the fire type, T. Times between successive
fires vary (fire interval, i) that can be summed and their mean taken. The area occupied by a particular fire event and time since
the last fire define a patch. When the occurrence of fires is monitored over an extended period they can be used to estimate fire
frequency. The fire regime is the mean and range of fire types experienced throughout the reference area over a sufficiently long
period to be able to calculate fire frequency.
value in the forest literature is the concept of mixed-severity sense (Keeley et al., 2011). These have major effects on other
fires or just mixed-fire regimes. To be ‘mixed’, the reference components, especially the spread pattern (strata that burn,
area is considered to include a moderate abundance of patchiness) and total area burnt. Often distinguished are
‘stand-replacing patches’ (Agee, 1993; Collins & Stephens, crown and surface fire regimes: the former is a response
2010; Hessburg et al., 2016) although this assumes that all to intense fires at low/moderate frequency that reach the
dominant species do not survive intense fire so it is mainly upper stratum, and the latter to less-intense fires, usually
applicable to certain conifer forests. More generally, ‘mixed’ at high frequency or managed fires under mild weather
simply implies wide variation about the mean of one or conditions, that burn only the ground stratum in shrublands
more fire properties or its effects. Here, adaptive traits can be or the understorey in woodlands/forests. Mixed-severity fire
expected to show intermediate properties (e.g. weak serotiny; regimes require patches of both crown and surface fires.
Enright et al., 1998) or high levels of phenotypic plasticity Fire frequency and intensity are controlled by the
(He et al., 2019). Alternatively, the species pool contains vegetation age and composition as they determine the
species/ecotypes with a wide range of traits on which each available fuel via net primary production (Fig. 3), and
patch type can draw. ‘Mixed-severity’ has some elements in by the weather at the time of each fire event. Both
common with ‘high pyrodiversity’ (Taille et al., 2018) but the are confounded by climate that controls the type of
latter is a broader concept, with stand replacement just one vegetation and sets the bounds within which weather occurs,
of many options for defining patchiness (Section III.4). but also has independent effects such as the incidence
The proposition that fire promotes and maintains and seasonality of lightning and litter dryness (Fig. 1).
biodiversity does not mean that any vegetation type, and the For example, in (sub)tropical biomes with mid-to-high
species within it, are adapted to fire per se (Keeley et al., 2011). net primary productivity, and with much fuel potentially
Rather, species possess a set of functional traits that enable available to be burnt (Fig. 3), the amount of moisture in fuel
them to respond favourably to a certain fire regime. The is the strongest determinant of fire frequency and intensity
concept of fire regime is therefore central to understanding (drought-driven fire regimes; Pausas & Ribeiro, 2013). At
the role of fire as a critical ecological and evolutionary factor. the other end of the spectrum, in areas where the available
Fire regime refers to the typical properties of all fire events fuel is generally meagre, such as deserts, alpine regions and
in a given reference area (community to region) gauged tundras, the availability of fuels produced during a pulse
over an extended time period. It is defined by the repeated of elevated resources (e.g. higher-than-average rainfall) in
patterns of each fire component (Fig. 1). For example, it the growing season has more influence on the fire pattern
may be characterised as experiencing intense crown fires (fuel-driven fire regimes; Pausas & Ribeiro, 2013; Archibald
in summer–autumn at moderate frequency, widespread et al., 2018). In regions with medium productivity, such as
and with little patchiness. The two core components, fire grasslands/savannas and evergreen shrublands, fire activity
frequency and intensity, have upper and lower limits to is much more frequent and predictable (the intermediate
which a species is adapted in an ecological/evolutionary fire-productivity model; Pausas & Ribeiro, 2013).

Fig. 3. Fire intensity and frequency in ecosystems with different levels of net primary production. The intensity of orange
(background color) indicates the level of fire proneness of each vegetation type. The extent to which fire occurrence in each
vegetation type is dependent on critical levels of resources or drought is also shown.
Fire regime has four additional components: the total distinct composition, structure and fire history, that control
size of the burnt area under consideration (fire size), the the level of pyrodiversity (Martin & Sapsis, 1992). Thus,
mean seasonality of the fire, and the spread patterns: pyrodiversity is the level of fire-caused heterogeneity (diversity
horizontal (overall grain of patchiness) and vertical (usual of patches) in a given area at a stated time (Krawchuk &
strata burnt) (Fig. 1). With fire regime, the time component Moritz, 2011). The patch components are not independent,
is given by fire frequency or mean fire interval. These such that the time interval between the most recent fire and
components are influenced by both biotic and abiotic factors, the previous one may be a surrogate for the amount and
including vegetation structure, topographic heterogeneity, distribution of fuel and hence fire intensity, patch size and
local weather during each fire event, and the ignition which strata burn. The temporal pattern of fire histories,
source (Collins et al., 2007; Kane et al., 2015). The fire such as varying historical intensity and fire intervals across
regime, therefore, filters biotas, selecting for adaptations the landscape, has sometimes been termed the ‘invisible
that persist in particular fire-prone environments at the (fire) mosaic’ as usually only the most recent postfire patches
scales of both population (Hernández-Serrano et al., 2013; can be recognised (Bradstock et al., 2005; Parr & Andersen,
Lamont, He & Downes, 2013; Vandvik et al., 2014) and 2006; Bradstock, Tozer & Keith, 2008), and represents the
community (Cavender-Bares et al., 2004; Verdú & Pausas, temporal dimension of pyrodiversity.
2007). Through the feedback cycle, fire shapes the vegetation Mathematically, pyrodiversity may be treated as a
structure that reinforces the tendency for a specific fire regime numerical summation of the scaled range of spatial variation
(with predictable fire frequency and intensity) to prevail in the fire components within a stated area. Different ranges
(Archibald et al., 2018). Where fire heterogeneity patterns of the various components can sum to a similar level of
tend to stabilise over time, i.e. the range of variation in pyrodiversity, just as different species compositions and
the components becomes predictable, a fire regime for the abundance spread may produce similar species diversities,
reference area may be defined. and even biodiversity: both can be low, moderate or high.
Local variations in the distribution of fuel, topography, Thus, it is possible to quantify pyrodiversity using standard
moisture levels, weather conditions and ignition sources cause indices of diversity such as Shannon’s index, classically used
properties of the spatial/seasonal components (Fig. 1) to vary for species diversity, and thus compare them. For further
continuously over the landscape. Together with temporal interpretation, the relative levels of the components need
components (time since previous fire, previous history of to be examined to explain how each diversity estimate was
fire intervals for individual patches), they produce a mosaic reached (Davies et al., 2018). The term pyrodiversity has
of fire-related landscape patches (see Fig. 1 in Agee, 2005 been used increasingly in the literature, as it is seen to be
for an excellent demonstration of fire-related mosaics), with aligned with the ecological hypothesis that ‘diversity begets

diversity’ (Parr & Andersen, 2006; Bowman et al., 2016; Kelly of fire. Here, we review the ecological role of fire in
& Brotons, 2017). In practice only one or a few components creating environmental heterogeneity, and generating and
are assessed on the understanding that they are a surrogate maintaining biodiversity. We focus initially on the effect
for others or that other components do not vary significantly of variation in fire components on biodiversity at a given
(Martin & Sapsis, 1992; Sitters et al., 2014; Ponisio et al., location (temporal scale, Sections III.1–3) then turn to their
2016; Davies et al., 2018). variation over the landscape (spatial scale, Sections III.4–6).
In contrast to pyrodiversity, the concept of fire regime
describes the specific properties of each of the fire-related (1) Individual fire components as drivers
components as gauged over many years, The properties of the of biodiversity
six components of each fire at a given location in the reference
area determine what species will be present in each patch. The fire-biodiversity relationship has been tested for a wide
Fire regime may include reference to variation about the range of taxa worldwide in fire-prone regions (Table 1).
mean where it is pronounced, e.g. mixed-severity fire regime These studies have examined species diversity at various
(Collins & Stephens, 2010), although quantifying/ranking temporal and spatial scales. The parameterisation of fire type
that variation is more the province of pyrodiversity. While the is usually based on variation in one component, such as fire
various fire-related patches have a local effect on population frequency, fire size, strata burnt, fire intensity, or seasonality,
viability, it is the level of pyrodiversity, and its causes, that at different locations or sometimes a combination of two or
dictate overall species presence and abundance at that time, three components (Table 1). The results either demonstrate
and the long-term impact of the fire regime that determines a significant positive relationship (13 studies), negative
what species will remain. Their impacts on biodiversity are (9), a positive relationship dependent on the component
examined in the following sections. considered (7), or no relationship (1) between changes in
With regard to the concept of biodiversity, we usually the fire component and (component of) biodiversity. Species
equate this with species diversity. We recognise that species diversity tends to be positively correlated with increasing fire
diversity has at least two components, number of species severity within limits (Sitters et al., 2014; Tingley et al., 2016)
(richness, S) and evenness (the extent to which total but a plateau or decrease at high severity is also common
abundance is spread among the species, E). S is highlighted (Taille et al., 2018). Time-since-fire may be critical in the
as this is easily assessed and conceptualized and is usually abundance of biota and a humped relationship is the norm
correlated with E (dominance is more likely with fewer (Lazarina et al., 2017; Taille et al., 2018). Decreases/increases
species). Other possible species-diversity components, such in mean fire interval may have a positive/negative (8 studies),
as phylogenetic or genetic disparity, are not considered here. negative/positive (6) or variable/nil (7) effect on species
We also accept that biodiversity has two other components diversity. Temporal aspects are considered further in Section
highly relevant to fire-prone systems: structure and function III.2 as the outcome depends very much on the range of
(Noss, 1990; Lamont, 1995). These are referred to when intervals used and the ecosystem under study.
discussing the effect of fire on vegetation type, successional The diversity of ants, spiders and termites usually responds
processes and fire-related traits that control ecosystem positively to greater fire activity. Birds, reptiles, small animals,
function. and intensity of plant–animal interactions generally benefit
from diverse fire types with varying intensity, while varying
fire size and frequency do not affect their diversity (Table 1).
Among plants, shortened fire intervals often lead to decreased
III. FIRE AS A DRIVER OF BIODIVERSITY species richness and diversity of plant functional traits in
the community, especially among woody species (Bradstock
Fire has been present since terrestrial plants invaded land et al., 1997; Bradstock et al., 1998; Schaffhauser et al., 2011;
420 million years ago (Scott, 2018). It has long been Bowman, O’Brien & Goldammer, 2012; Enright et al.,
recognised that fire is a natural ecological factor in many 2014). Bowman et al. (2016) noted that these mixed results
forest, woodland, shrubland, and grassland systems around could be the outcome of failing to include spatiotemporal
the world (Rundel, 1981; Bond & Keeley, 2005; Bond, heterogeneity among fire types (pyrodiversity; see Section
2008; Pausas & Keeley, 2009), and that it has a major III.4), and assessing study areas over too short a time
role in maintaining biodiversity and ecological processes in scale. Thus, the level of species diversity is overwhelmingly
fire-prone ecosystems (Bond & Parr, 2010; Rundel et al., associated with the fire regime to which the area is subjected.
2016, 2018; Pausas & Keeley, 2019). Fire plays many Are changes in fire-biodiversity relationships always driven
important ecological roles in ecosystems that cannot be by changes in fire properties as the independent variable?
duplicated by any other natural events. Fire affects key Directional variation in the fire regime resulting from
processes in the ecosystem, dictates what species are present climate change has been shown to lead to changes in
in that ecosystem and determines how they are arranged in species richness and associated functional traits (Lamont
space and time. Environmental heterogeneity and resource et al., 2013), i.e. fire drives biodiversity. Conversely, areas
availability are among the most important factors that that are similar climatically but have different fire regimes
determine species-richness gradients (Stein et al., 2014), and, produce different vegetation types that confirm the tight
in many ecosystems, these are controlled by the incidence relationship between the fire regime and biodiversity. Thus,

Table 1. Examples of studies that examined individual fire components as a driver of biodiversity. Time span refers to the period within which fire events were monitored.
BACI: before-after-control-impact design
Fire components Biodiversity Time span Relationship

Taxon group Ecosystem assessed components assessed of study (years) demonstrated Reference
Ants Brazilian Cerrado Fire interval Species composition 16–37 Positive Maravalhas &
Vasconcelos (2014)
Ants Northern Australia Fire history (interval) Species richness 10 Positive Anderson et al. (2014)
Termites African savanna Fire interval and Species diversity 65 Positive Davies et al. (2012)
intensity
Termites Semi-arid Australia Fire age (time since Species diversity 35 No relationship Davis et al. (2018)
Fire as key driver of biodiversity
fire) patchiness
Insects SE Australia Fire history Species richness 75 Positive Brown & York (2017)
heterogeneity
Spiders Western Australia Fire history (interval) Species richness, 20 Positive Langlands, Brennan &
abundance Ward (2012)
birds Montane California, Fire severity Species diversity 10 Positive Tingley et al. (2016)
USA
Birds SE Australia Fire size, intensity Species diversity BACI Positive Sitters et al. (2014)
Birds SE Australia Fire history (interval) Species richness 35 No relationship to Taylor et al. (2012)
positive
Reptiles Central Australia (arid) Fire size Species diversity <1 No relationship to Pastro, Dickman &
positive Letnic (2014)
Reptiles SE Australia Fire history (interval) Species richness 35 Negative Nimmo et al. (2013)
Reptiles Western Desert, Fire size (burnt area) Abundance 20 Positive Bird et al. (2013)
Australia
Reptiles Semi-arid Australia Time since fire α, β, γ diversity 13 No relationship to Farnsworth et al. (2014)
positive
Small mammals Australian tropical Fire interval Persistence 6 Positive Griffith, Garnett &
savanna Brook (2015)
Small mammals SE Australia Fire history (interval) Species richness 35 No relationship to Kelly et al. (2012)
positive
Small mammals Northern Australia Fire size, interval Species diversity 8 Negative Lawes et al. (2015)
Medium-sized Northern Kimberley, Fire history (interval) Species richness, 4 Negative Radford et al. (2015)
mammals Australia abundance
Mammals Central Australia (arid) Fire size (burnt area) Species diversity <1 No relationship to Pastro et al. (2014)
positive
Mammals N Australian savanna Fire frequency Species diversity 9 Positive Davies et al. (2018)
patchiness
Plant–pollinator National parks, Fire interval, age, Bee-plant interactions 10 Positive Ponisio et al. (2016)
communities California, USA extent, severity
Plants SW Australia Fire interval Species richness 32 No relationship to Wittkuhn et al. (2011)
positive
Plants Mediterranean, Fire interval Species richness 13000 No relationship to Leys, Finsinger &
Corsica Island positive Carcaillet (2014)

7
areas burnt frequently to maintain grasslands (savannas) have
Bradstock et al. (1998)
Watson, Bradstock &
Schaffhauser, Curt &
Bowman et al. (2016)

different species composition and diversity than surrounding
Bradstock, Tozer &

Enright et al. (2014)
Bassett et al. (2017)
Foster et al. (2017)

areas that burn rarely (forests) (Bond & Parr, 2010;
Morris (2009)
Tatoni (2011)
Keith (1997)
Dantas, Batalha & Pausas, 2013). These different floras
have different flammability traits that maintain different
Reference
fire regimes and contrasting vegetation although under

a similar climate (stabilising feebacks). Some areas under
different environmental constraints and fire regimes may
nevertheless have similar vegetation structurally, although
differing greatly in terms of species, such as sub-alpine and
savanna grasslands. Thus, certain components of biodiversity
may sometimes be independent of the fire regime, while
demonstrated
species composition may invert the usual relationship by

Relationship
driving the fire regime instead.

Negative
Negative
Negative
Negative
Negative
Negative
Positive
Positive
(2) Diversity through time: fire resets community
dynamics
By consuming biomass, fire creates vegetation gaps and
200 (simulation)
of study (years)
Time span
initiates population and community recovery. In fire-prone

3–20
ecosystems, most plant species present before the fire either

20
50
20
55
3
recover vegetatively and/or germinate from stored seeds

(Keeley et al., 2012). Seeds released by serotinous fruits/cones
in response to fire spread through and across adjacent burnt
communities and immediately add to the species available for
vegetation structure
recruitment (Merwin et al., 2012). Short-lived or geophytic

components assessed
richness, fire trait
species that had long completed their aboveground life cycle

functional traits
functional traits
functional traits
Heterogeneity in
Heterogeneity in
Species richness,
Species richness,
Species richness,
Species richness,
Biodiversity
Species richness
become visible as their seeds or belowground bud banks

Woody species
abundance
break dormancy and their foliage appears above the soil

(Pausas et al., 2018b). In fire-prone ecosystems, almost all
plant species in the postfire community recorded prefire are
evident within 12 months of a fire (Herath et al., 2009) – this
phenomenon conforms with the ‘autosuccession’ (Hanes,
1971) or ‘initial floristic composition’ (Noble & Slatyer, 1980;
Fire interval patchiness
since fire patchiness
Enright et al., 2007) models of community succession. The

subsequent rapid decline in population size is accompanied
Fire severity, time
Fire components
by a gradual increase in vegetative growth of survivors,

assessed
a period of escalating fecundity, then declining foliage

Fire interval
Fire interval
Fire interval
Fire interval
Fire interval
Fire interval
and seed production, and eventual death or return to

belowground dormancy. However, the key to understanding
community dynamics is that individual species undertake
their bell-shaped response curve at different rates (Fig. 4A).
Some annuals are only evident the year following fire
(pyroendemics; Keeley & Pausas, 2018), although most linger
Mediterranean, SE
in the increasingly scarce bare patches. Other herbaceous

species complete their (aboveground) life cycle within a
SW Australia
SE Australia
SE Australia
SE Australia
SE Australia
SE Australia
few years (Hinman & Brewer, 2007; Lamont & Downes,

Ecosystem
Tasmania
France
2011), weakly woody shrubs may survive for about a decade,

strongly woody shrubs lacking persistent budbanks may
survive for several decades, strongly woody plants with
persistent budbanks may survive for hundreds of years, and
Table 1. Continued
clonal species with belowground budbanks may survive for

thousands of years (Lamont & Wiens, 2003; Merwin et al.,
2012; Pausas et al., 2018b).
Taxon group
As particular plant species and growth forms take

Vegetation
Vegetation
turns to dominate the community, and the structure and

Plants
Plants
Plants
Plants
Plants
Plants
availability of habitat types changes progressively, the

apparent successional stages are due to alternating maximum

Fig. 4. (A) Postfire abundance/performance of individual species relative to the highest value in each data set. Curve fits are based
on the best-fit lines using at most two variables (shown with R2 ), or are lines joining the raw data points. Time-since-fire effects are
given for four species in SW Australia: Acacia 1 (fire-killed) is for annual seed production; population size decreased by 20-fold over
the period shown (Monk, Pate & Loneragan, 1981). Banksia 1 (fire-killed) is for on-plant seed storage (serotiny) with up to 200 released
seeds per plant after fire reducing one individual by 15 years, and with mean longevity estimated at 45 years (Lamont & Enright,
2000). Eucalyptus 1 and 2 represent the stature of a mixture of resprouting eucalypts over time (Gosper et al., 2012). Time-since-fire
effects are also given for the abundance of four mammals in savanna grasslands of Brazil (Briani et al., 2004). Note that all plant
species are present from year 1, but some animals enter the system later. (B) Trend lines for species richness (S, * indicates values
are plotted relative to highest values) in a given community based on the processes shown in A. Beetles in N Australian savanna
(Orgeos & Anderson, 2001, birds 1 (functional richness in heathland) and birds 2 (wet forest) in SE Australia (Sitters et al., 2016),
small mammals in Brazil (Briani et al., 2004), boreal forest trees in Canada [both S and evenness (E)], for data over 50 years since
fire (Yeboah, Chen & Kingston, 2016), mallee and mallee-heath in SW Australia (Gosper et al., 2012), and annuals in pine forest,
Greece (Kazanis & Arianoutsou, 2004). (C) Relative S for different growth forms and total species, and relative species heterogeneity
(1 – mean similarity index between quadrats) in relation to mean fire interval (inverse of fire frequency) for a forest–grassland
ecotone in central Northern USA (Peterson & Reich, 2008) (quadratic fit for all lines, P < 0.01).

performance by persistent species rather than continual the seed bank is depleted and becomes insufficient to
colonisation and exodus as expected in non-fire-prone maintain population viability (Enright et al., 2014). Declines
systems and among some animal groups (Pausas & Parr, in abundance, and even local extinction or exclusion,
2018). Most animals are recorded surprisingly early in with increased fire frequencies have been documented
postfire communities, with some surviving in cryptic refuges among many plants and animals (Evans, 1988; Watson
(Kiss & Magnin, 2006), or attracted by enhanced food & Wardell-Johnson, 2004; Kiss & Magnin, 2006; Moretti
sources (Lamont, Ralph & Christensen, 1985; Curtis, 1998; et al., 2006; Griffith et al., 2015). With the decline in supply
Pausas & Parr, 2018) or by predator reduction (Pausas et al., of propagules, multiple extinctions are possible (Enright
2018a), then decline as the habitat continues to change et al., 2015). Bowman et al. (2012) observed conversion of
(Raphael, Morrison & Yoder-Williams, 1987). Others must serotinous Eucalyptus regnans forests in SE Australia to grassy
wait for the return of more mature stands to meet their shrublands after sites burnt in 2003 and 2007 were reburnt
niche requirements (Bradstock et al., 2005; Smucker, Hutto in 2009 and 2013 while trees were still immature (lacking
& Steele, 2005; Moretti, Duelli & Obrist, 2006; Jacquet & seeds), explaining complete regeneration failure. Resprouters
Prodon, 2009; Santos, Badiane & Matos, 2016; Fig. 4A). are likely to be more resilient to frequent fires, especially
Some fire-killed epiphytes, such as mistletoes, enter the among grasses, but there is evidence of a reduction in woody
community from unburnt stands also after some delay, resprouters with anthropogenic increases in fire frequency
depending on the availability of suitable aerial substrates. through invasion by annual species (Pausas & Keeley, 2014a).
However, among both plants and animals, there is a general This may be due to insufficient time for starch reserves to
decline in species richness (S) with time since fire (TSF, build up or because of the exhaustion of resprouting buds.
Fig. 4B), confirming that succession is essentially a function At the other extreme, consider a stand unburnt for much
of the species present initially. As a rule of thumb (including longer periods, e.g. several decades. By then many geophytes,
certain functional groups following their sharp initial decline evergreen herbs and short-lived shrubs (‘fireweeds’) have
in S), the data collated here indicate a typical slope for both retreated to dormant belowground bud banks or completed
plants and animals of ∼0.2, i.e. a decline of about 10% in S their life cycle (Fig. 4C). Dead serotinous plants will no
every 50-year interval since the last fire. But TSF is not the longer represent a source of seeds should a fire occur (as any
only temporal variable that affects S: fire frequency (range of fruits/cones on dead plants are incinerated; Lamont, 1995),
fire intervals) must also be taken into account. while the longevity of some soil-stored seeds may only span
In addition, both plants and animals can recolonise a few years. Palmquist, Peet & Weakley (2014) recorded a
burned habitats when they suffer local extinction. Species 30–40% decline in S in Florida pine savannas when fire
co-existence is achieved because species are differentially was withheld for 30 years, especially among herbs, that for
adapted to different successional stages (i.e. different habitat many species was permanent when fire was re-introduced
openness) and these stages (TSF) vary across the landscape, (the propagules had presumably perished). Dee & Menges
while their long-term persistence is achieved through (2014) noted the importance of postfire bare patches (gaps)
metapopulation dynamics, as species migrate to suitable for recruitment of herbs and shrubs lacking persistent bud
habitats (He et al., 2004, 2009). Frequent fires retard
banks and that these became unavailable within 26 years in
the over-dominance of the most competitive species, and
Florida shrublands due to the accumulation of impenetrable
allow less-competitive species to persist (Grubb, 1977).
litter, with a consequent decline in species diversity. Animals
Similarly, vegetation patches that were burnt in severe
dependent on these species for their survival decline in
fires may lead directly to increased gamma biodiversity,
parallel. This reduction is simply the long tail of the negative
as intense fires create habitats for pioneer species (Tingley
S versus TSF slopes described in Fig. 4B and applies to both
et al., 2016). Boucher et al. (2012) showed that successional
plants and animals. By contrast, rare species are finally
changes in species composition were more pronounced in
eliminated or excluded and species evenness (E) increases
burned sites than in those with a girdling treatment, which
(Fig. 4B).
indicates that the contribution of fire-generated habitats to
regional diversity includes promoting successional processes. If the potential presence of species in the community
Consequently, these mechanisms allow species climaxing at is curtailed by both frequent short-interval and rare
different successional stages to coexist in space and time, long-interval fires, it follows that there must be an optimum
facilitating the maintenance of diverse communities over moderate-interval fire regime for promoting species diversity
landscape and regional scales. (Fig. 5). Somewhere between a mean fire interval of a few
years and some decades is a range of intervals where most
species are present – they are producing or storing sufficient
(3) The intermediate disturbance hypothesis seeds for self-replacement should a fire occur (Enright et al.,
Consider the outcome of a range of successive fire intervals. 1998). This principle is known as the intermediate disturbance
Imagine a stand is intensively burnt again after a few hypothesis (IDH) (Connell, 1978; Huston, 2014). This model
years: many juveniles of fire-killed woody species have has been applied to total S, but also to the Shannon diversity
not yet reached maturity (immaturity risk; Zedler, 1995), index (H = S × evenness), mean S of n patches, and species
Fire-intolerant species that regenerate via seeds (especially heterogeneity (mean pairwise dissimilarity of n patches based
if stored on the plant), can potentially be eliminated, as on S or H) (Peterson & Reich, 2008). For the latter index,

Fig. 5. Conceptualised relationship between species richness (S) of a reference area (community, landscape, region) and mean fire
interval (fire regime). For a given fire regime, there is a mosaic of patches defined by different times since the last fire (represented
by black circles) about the mean time interval. For convenience, these two measures of time are plotted against a common time axis
but in reality time-since-fire forms a third dimension, with, for example, the furthest left dot at each of the three intervals possibly
representing the same time-since-fire. Each patch is associated with a different S whose mean S at the different mean intervals
describe a bell-shaped curve (broken line). One (or a combination) of the patches (here taken as the one at shortest time-since-fire on
the grounds that this usually has the highest S) accounts for the highest S under that regime (total S) and collectively they describe a
bell-shaped curve (solid line). The difference between total and mean S is an approximate index of species heterogeneity (biodiversity)
in the study area. Note that the curves are asymmetric with there being a sharp decline in S as intervals become shorter and a much
more gradual reduction in S at long intervals. This model is consistent with the intermediate disturbance hypothesis (see Fig. 4C).
heterogeneity will be a function of (i) the scale of patchiness of Another problem is uncertainty that the full range of
species distributions, and (ii) the range of TSF patches (and mean fire intervals has been covered – there is the risk that
their fire history) present in the community – the mosaic only one side of the ‘bell’ has been included because of
effect. As S declines with each repeated short- or long-interval logistic limitations. While species composition/structure at
fire, variation about the mean due to TSF patchiness, gauged the frequent and infrequent ends of the fire-interval spectrum
as (total S – mean S) or mean dissimilarity, will also reduce may be quite different, whether these values increase at
(Fig. 4C). This means that total S, mean S and heterogeneity moderate frequencies as expected (Gordijn et al., 2018) may
will peak under moderate mean fire intervals but highest not be evident at the community scale but will apply at the
S can be attributed to just one (recently burnt) patch, landscape scale (Keith & Tozer, 2012). Other claims of lack
whereas mean S and heterogeneity are a function of the net of support for the hypothesis have been due to failure to
values for all patches. If the initial composition successional appreciate that the theory only applies to closed sysems, i.e.
pathway holds for plants, then total S may prodivde an with a fixed species pool – gamma diversity might continue
adequate diversity measure for botanists; if the replacement to rise with increasing disturbance in landscapes subject
successional pathway holds for animals, then mean S and to invasion by exotic species (Kuneš, Abraham & Herben,
heterogeneity, with their emphasis on mosaics, will be of 2019) although diversity might decrease at the local (alpha)
more relevance to zoologists (Sitters et al., 2016). scale if the alien species start to outcompete the indigenous
The IDH has been well supported for fire-prone vegetation flora (Milberg & Lamont, 1995; Fisher et al., 2009). This is
(Dee & Menges, 2014; Gordijn, Everson & O’Connor, 2018; elaborated further in Section III.5.
Fig. 4C) but there has been little relevant work done on One potentially confounding issue is the difficulty of
animals (Lazarina et al., 2019). If the species trajectories subjecting vegetation at different fire intervals to the same
we have outlined above hold generally, it follows that fire intensity. Thus, Maikano, Cohn & Di Stefano (2018)
the model must apply to fire-prone systems generally. In obtained higher species diversity at greater fire intensities
that case, failure to support the model (Schwilk, Keeley & due to enhanced germination in SE Australian woodland
Bond, 1997) could be due to limitations of the test rather and forest. Among birds, burn intensity may greatly affect
than inapplicability of the model. Beckage & Stout (2000) the shape of the response curve to time-since-fire (Taille et al.,
showed a P = 0.08 quadratic fit to their Florida pine forest 2018). Individual species of birds in North American forests
diversity data, but if we compare the two most diverse sites are favoured by either absence of fire, or low, moderate or
at moderate fire frequency with the four sites having lesser intense fires (Smucker et al., 2005; Taille et al., 2018). These
and greater fire frequency, a one-tailed t-test with equal variations could distort their relative positions on the graph.
variances yields P = 0.006. This indicates a problem of In conclusion, fire resets community dynamics and increases
inadequate sample size. species richness and diversity from the outset. Repeated fires

modify the responses and set up new patterns but the basic mole cricket (Gryllotalpa major) in Oklahoma, and hypothesised
bell-shaped response curve to increasing mean fire intervals that burnt grassland sites may be more efficient for acoustic
may still be expected. communication than more complex unburnt sites with dense
surface vegetation, with responding females spending less
(4) The pyrodiversity–biodiversity hypothesis effort discovering mates, therefore improving reproductive
efficiency. Thus, the presence of burnt patches fosters the
Environmental heterogeneity is considered one of the most presence of this mole cricket in grasslands.
important factors that determine species-richness gradients Fire-generated landscape mosaics also enhance environ-
(Stein et al., 2014) and fire is one of the major natural sources mental heterogeneity at the regional scale, creating differen-
of such heterogeneity. Every fire has unique outcomes, as the tiated species selection and promoting regional biodiversity
burning of each stand of vegetation is the combined effect (Talluto & Benkman, 2013; Castellanos et al., 2015). For
of local weather conditions, available fuel, and topography example, in tropical ecosystems, Dantas et al. (2013); Dan-
of the reference area. Thus, the spatiotemporal variation tas et al. (2016) showed that different types of fire have
in fire types, and time since each fire occurred, may driven the alternative states of forest and savanna under a
create a diverse fire-caused mosaic over the landscape
given climate, with both containing a different set of species
that is quantified as pyrodiversity. Because different plants
with markedly different traits (Table 2). Therefore, such
and animals may depend on different resources and
fire-driven savanna–forest mosaics contribute to enhancing
habitat conditions that are created by fire, it has been
regional plant diversity, which in turn, contributes to the
argued that high pyrodiversity produces a wider range of
generation of heterogeneous habitats that provide diverse
habitats that can support more species in the total area.
niches and resources for animals (Bond & Parr, 2010; Bow-
Fire-generated environmental heterogeneity is regarded as
man et al., 2016; Ponisio et al., 2016). Landscape mosaics as a
a useful framework for understanding species co-existence
result of diverse fire patterns may also drive phenotypic and
at local and regional scales (Dantas et al., 2013; Bowman
genetic divergence within species (Gómez-González et al.,
et al., 2016), population divergences (He, 2013; Castellanos,
González-Martínez & Pausas, 2015; He et al., 2019), and the 2011; Pausas et al., 2012; Hernández-Serrano et al., 2013;
presence of specific fire-related traits in fire-prone ecosystems Vandvik et al., 2014; Pausas, 2015; He et al., 2019). In addi-
(Simon et al., 2009; Lamont & He, 2012; Talluto & Benkman, tion, increased patchiness of fires increases habitat edge, and
2013; Pausas, 2015). Diverse plant communities create attracts birds and other animals since both shelter and new
different niches for fauna (Stevens & Tello, 2011) and more food sources become available (Longhurst, 1978; Parkins,
opportunities for synergistic species interactions (Janz, Nylin York & Di Stefano, 2018).
& Wahlberg, 2006; Ponisio et al., 2016; García, Castellanos Mathematical constraints also contribute to explaining
& Pausas, 2018; Carbone et al., 2019). the level of pyrodiversity because, as the reference area
Martin & Sapsis (1992) were the first to explore is divided up into more (fire-related) patches, the number
the consequences on biodiversity of variation in the of patches increases exponentially and the size of patches
burning pattern in fire-defined habitat mosaics. The declines linearly (Fig. 6). Given that each patch has a
pyrodiversity–biodiversity hypothesis recognises that fires different fire history, species composition will differ among
create landscape heterogeneity (patch mosaics with different patches, and the species richness in each patch will depend
vegetation structure and openness) that cause resource on its species–area relationship, which in turn depends on
partitioning over the landscape, thus promoting and the growth and life form of the species present, growing
maintaining multiple plant functional types and their conditions and historical factors (Keeley & Fotheringham,
associated fauna. There is a range of possible ecological 2003; Magadzire et al., 2019). This relationship is asymptotic
mechanisms under which pyrodiversity may promote and suggests that as the area (x-axis in Fig. 6A) is reduced,
biodiversity (Table 2). Species have finely partitioned niches the reduction in the number of species (y-axis) is small at
that match these spatially heterogeneous habitats. For first until a point is reached where a small reduction in
example, fire generates spatially or temporally localised area generates a large decrease in species richness (rare
release of mineral nutrients for plant growth (Stock & species are lost). Thus, with increasing pyrodiversity (i.e. an
Allsopp, 1992) and pockets of food for fauna. Diverse habitat increasing number of patches of decreasing size), the number
types are associated with more niches that are favoured by of species is likely to increase until a point is reached where
different species (Parr & Brockett, 1999; Bond & Keeley, it drops because the patches are so small that the number
2005; Bird et al., 2008; Kelly et al., 2012; Cohn et al., 2015). of species in each patch decreases at a faster rate than the
Different species are favoured at various times after the fire area. Thus, the pyrodiversity-diversity relationship is likely
disturbance (see Section III.2). Variations in fire intensity to be humped, with the peak towards the upper end of
create niches varying greatly in resource availability (light, the pyrodiversity gradient (left-skewed; Fig. 6). In addition,
moisture, nutrients, space) that have different effects on with increasing subdivision, the vegetation and fire history
survival of individual biota, seed germination and plant become less different among patches (spatial autocorrelation),
growth rates (Pausas et al., 2003), and finally, selection, so that it takes more patches to give an equal increase in
divergence and speciation. For example, Howard & Hill pyrodiversity. Overall, there is a theoretical limit to support
(2007) reported fire-stimulated reproduction in the Prairie for the ‘pyrodiversity begets diversity’ hypothesis.

Table 2. Possible mechanisms underpinning the hypothesised pyrodiversity-biodiversity relationship, with examples
Mechanism Interpretation with examples Scale of diversity References

Resource partitioning Fire patchiness promotes coexistence of multiple Local Martin & Sapsis (1992)
promoted at fine plant functional types
scales
Intermediate disturbance maintains species Local Bond & Keeley (2005); Bird et al.
co-existence as species with different resource (2008); Kelly et al. (2012); Cohn
requirements are not excluded et al. (2015); Beale et al. (2018)
Cross-trophic Fire stimulates flowering, leading to more species Local Campbell, Alfaro & Hawkes (2007);
interactions promoted interactions, and species-rich pollinator Grundel et al. (2010); Van Nuland
communities et al. (2013); García et al. (2018)
Regular fires may promote species with canopy Local Benkman & Siepielski (2004)
seed banks, and cues simultaneous seed
release – providing abundant food resources
for granivorous animals
Black-backed woodpeckers are attracted to Regional Hutto (1995); Dixon & Saab (2000);
post-wildfire conditions because of the Hoyt & Hannon (2002); Rota
abundance of wood-boring beetle species et al. (2014)
attracted to the fire-killed trees
Environmental Savannas and forests harbour different species, Regional Dantas et al. (2013); Dantas et al.
heterogeneity and different fire regimes generate (2016); Pausas (2015); Bond &
enhanced at the savanna–forest mosaics with high regional Parr (2010)
landscape scale diversity
Landscape mosaics are a suitable setting for Regional Talluto & Benkman (2013);
population divergences and selective processes Castellanos et al. (2015); Pausas
in fire-prone ecosystems et al. (2012); Hernández-Serrano
et al. (2013); Vandvik et al. (2014);
Pausas (2015); Davies et al. (2018);
He et al. (2019); Gordijn et al.
(2018)
Animal diversity is enhanced by landscape Regional Law & Dickman (1998); Bond &
mosaics through their association with Parr (2010); Bowman et al. (2016);
particular plant communities Ponisio et al. (2016)
New niches for Some moss and lichen species are relatively Regional Ahlgren & Ahlgren (1960)
fire-dependent species restricted to severely burnt forest conditions
created through
elevated food supply
The mushroom Morchella elata, and the shrub Regional Heinselman (1981); Pilz et al. (2004)
Geranium bicknellii, are promoted by immediate
postfire conditions
Fungi differ in their adaptations to recently burnt Regional McMullan-Fisher et al. (2011)
and long-unburnt conditions
Several beetle species are relatively restricted to Regional Saint-Germain, Drapeau & Hebert
recently burnt forests (2004); Boucher et al. (2012);
Pausas et al. (2018a)
White-headed woodpeckers (Leuconotopicus Regional Hollenbeck, Saab & Frenzel (2011);
albolarvatus) are more abundant in burnt Lorenz et al. (2015)
(wildfire) than unburnt forest
In aquatic systems, severe fire events can Regional Benda et al. (2003); Burton (2005);
rejuvenate stream habitats by causing large Malison & Baxter (2010); Jackson
amounts of gravel, cobble, woody debris, and & Sullivan (2015)
nutrients to be imported, resulting in increased
productivity and aquatic-insect emergence
rates
Community-dynamic Succession processes allow species with different Local See Section III.4
processes initiated postfire stage preferences to replace each other,
thus promoting species-rich communities, with
their associated fauna and soil biota, when
different stages co-exist among a mosaic of fire
intervals
Fires buffer stress, and Fire mitigates drought by reducing transpiration Local and regional Alfaro-Sánchez et al. (2016); Pausas
maintain local of a site; the resprouting plants and newly et al. (2003)
diversity emerged seedlings postfire should benefit from
higher resource availability

Fig. 6. (A) The upper two diagrams show the conceptual relationship between different fire-caused patches (A, B, C) within two
reference areas (1, 2) of the same size and species diversity (S for the patch, S for the reference area) assessed at the same time
(Assessment time 1). The mosaic of patches in each area can be combined to give an estimate of the range of variation in fire events,
pyrodiversity (P), of that area. Here, we use Shannon’s diversity index to show that the area with three fire-related patches has more
than twice the P of that with two, with each patch in the reference area making an equal but different contribution to P. Using
the same index for species diversity (lower two graphs) shows that the relationship between S and patch size depends on the slope
and asymptote of the species-area curve and where the patch area fits along the area axis. S AB and S ABC represent the number
of species in common between the various patches that has to be subtracted from the number of species in each patch, reducing
the certainty that S 2 > S 1 . (B) Model outcomes showing possible relationships between P and biodiversity (here given by S). The
number of patches (N ) rises exponentially as the mean area per patch declines linearly. The mean number of species per patch (S ) is
based on Fig. 6A, with a mean slope (b) modelled here that is relatively high (approaches linear, e.g. b = 0.9, for the power function)
or low (strongly curvilinear, e.g. b = 0.6). Multiplying N by S at a given P gives the total number of species (S) present. The two
extreme scenarios given here show a marked peak in S at moderately high and high P values. All parameters were scaled between 0
and 100 to facilitate comparison.

Despite increasing adoption of the mosaic-burning moderately low P. The greatest range of patch heterogeneity
approach in fire management (Kelly & Brotons, 2017), corresponds to the highest P but biodiversity falls as the
there have been few tests of the validity of the mean size of patches is now so small that each patch
pyrodiversity-biodiversity hypothesis. Davies et al. (2018) contains few rare species and population viability of those
modelled the effect of different-sized patches varying in fire present is threatened (the Allee effect; Lamont, Klinkhamer
frequency on its known relationship with the presence and & Witkowski, 1993). With the IDH, species most favoured
abundance of seven mammal taxa in northern Australian by low or high fire frequency are usually still present but
savanna. Pyrodiversity (P) was based on Shannon’s index rare, as well as the majority of species that optimise at
and, since the type of patches was fixed, P increased as moderate levels of disturbance. Patch-size constraints are
the size of the patches (E) became more equal. Mammal not relevant to the IDH provided that the optimal patch
diversity peaked at intermediate P. Had the deleterious types are present and are large enough to support rare
effect of small patch size on population viability been taken species. MSRs also incorporate the mean and frequency
into account (the area was dimensionless), this may have range of the three possible fire types but patch size variation
pushed the optimum further to the right, i.e. generating a is limited (moderate) otherwise the required mean would not
more left-tailed distribution not unlike our model (Fig. 6), be reached (Agee, 1998).
even though our data are based only on S rather than
S × E. Gordijn et al. (2018) also noted that species diversity (6) Fire and biodiversity at broad scales
should be greatest with a mixture of patches at different
fire frequencies as the extremes were quite different in Fire is almost ubiquitous in global terrestrial ecosystems
species composition (although any effect of patch size on (Archibald et al., 2018). Simulation studies show that biome
S was not considered). Others have observed no effect distributions in the absence of fire would be very different
(Davis et al., 2018) or a negative effect (Bassett et al., 2017; from their current patterns (Bond et al., 2005). However,
Foster et al., 2017). The extent to which this fails to support different environments, and different geographical regions
intermediate P as maximizing biodiversity due to (i) testing within the same environment, may have quite different fire
an inadequate range of P, thus observing only part of histories and pyrodiversity levels (Rogers, Soja & Randerson,
the ‘hump’, (ii) not using a fixed reference area, thus not 2015; Archibald et al., 2018). The IDH is often proposed
appreciating that increasing patch type is at the expense of as the most likely pattern for maximising species diversity
patch area and thus species—area limitations come into (Huston, 2014; Section III.3). However, this is not the
play, or (iii) using different indices of P and diversity, case when studying fire-diversity patterns at broader scales;
requires further investigation. In practice, the specific instead, these show a monotonic positive relationship with
pyrodiversity—biodiversity relationship will depend on the increasing fire activity for plants (Pausas & Ribeiro, 2017)
ecosystem under study, the taxonomic group examined (as and vertebrates (Beale et al., 2018). Fire frequency gradients
these will have different species—area relationships), the at the biogeographical scale cannot be equated with diferent
range of patchiness that can be applied, and the indices of fire frequencies at given sites. This is because an increase in a
P and diversity used. However, theory and empirical studies disturbance at the local scale takes place within a given species
that are area-constrained indicate that the relationship will pool, and thus is likely to include fire regimes beyond the
be essentially humped and left-skewed. historical range (which could decrease diversity). By contrast,
biogeographical studies include gradients of disturbance
across different species pools, and in each species pool,
(5) Mixed fire regime and intermediate disturbance
the fire regime may be within the historical range (e.g. from
hypothesis
fire-sensitive ecosystems with low fire activity to fire-prone
Since a mixed-severity fire regime (MSR) promotes ecosystems with high fire activity). In addition, fires in many
maximum biodiversity, a parallel can be drawn with the IDH. ecosystems are frequent and predictable enough to have
Arno, Parsons & Keane (2000) and Perry et al. (2011) seem to selected species with adaptive traits for persisting under the
have been among the few to recognise that MSRs conform historical fire regime (Keeley et al., 2012; Charles-Dominique
with the IDH at the landscape scale, i.e. both represent et al., 2015; Pausas, 2015). Consequently, frequent fires do not
the peak of the hump along a fire component gradient always imply species loss, and thus the decrease in diversity
(Fig. 4). Our contribution is to demonstrate that species at the high end of the disturbance gradient (as in the IDH)
diversity is also humped when plotted against P (Fig. 6). is unlikely for broad areas that cover numerous fire-prone
This integrates the three concepts (IDH, P and MSR) to ecosystems. The overall trend is therefore that at broad scales
show that biodiversity is maximal at moderate levels of patch diversity increases along the fire activity gradient (Pausas &
diversity, disturbance and fire severity over the landscape. Ribeiro, 2017).
But the optimal fire-regime conditions represented by each Broad-scale species diversity is also strongly controlled by
term are not identical. Biodiversity is greatest when there environmental factors (Kreft & Jetz, 2007) and variations in
is a wide range of fire types corresponding to moderately productivity (Gillman & Wright, 2006) that modulate the
high P and fire severity. In terms of the IDH, biodiversity fire regime to some extent. So, the question that arises is to
is greatest when the fire types range around the midpoint what extent can the increasing species diversity be attributed
between the extremes (Fig. 5): this would be equivalent to a to fire, or to the environmental factors that mediate it? The

current evidence indicates that both environmental factors ‘fire-mimicking’ disturbances, such as severe drought and hot
and fire are important and complementary dimensions summers, but were selected for by the constraints imposed
in explaining biogeographical patterns of biodiversity. For by the prevailing fire regime. Excellent reviews already exist
instance, once productivity, environmental variables and that conclude that fire has been a significant agent of natural
their heterogeneity are included in the model, fire activity selection in driving the evolution of terrestrial biota for
explains a major fraction of the variability in the diversity 420 million years (Bond & Scott, 2010; Keeley et al., 2011;
of plants at the global scale (Pausas & Ribeiro, 2017) and Rundel et al., 2016; Archibald et al., 2018; He & Lamont,
mammals and birds at the continental scale (Africa; Beale 2018a; Pausas et al., 2018b).
et al., 2018). In fact, for global plant diversity, fire activity is The roles of fire in the evolution and adaptation of ani-
the best single variable that accounts for the level of plant mals have received little attention (reviewed by Pausas & Parr,
diversity. 2018). The long co-existence over evolutionary time between
organisms and fire suggests that these organisms must have
evolved to adapt to, even rely on, recurring wildfires to
IV. FIRE AS AN EVOLUTIONARY DRIVER OF persist in fire-prone environments (i.e. ‘pyrophilous’ species;
BIODIVERSITY Table S2). Organisms that depend on fire-disturbance for
their long-term survival may be regarded as adapted to
fire-prone habitats (Hutto et al., 2016; Pausas & Parr, 2018),
Evidence is mounting that fire has been a significant force
although the evidence for unique fire adaptations in ani-
in the selection and evolution of terrestrial biotas through
mals has rarely been explored (Nimmo et al., 2019). Recent
geological time and has shaped the structure and function
studies (Stawski et al., 2015, 2017; Matthews et al., 2017;
of global biomes throughout the 420-million-year history of
Nowack et al., 2018) indicate that torpor (a state of decreased
terrestrial life on Earth (Pausas & Keeley, 2009; Belcher,
physiological activity) is used extensively among terrestrial
Collinson & Scott, 2013; He & Lamont, 2018a). The species
mammals in Australia to deal with fires, or the scorched
richness of current fire-prone ecosystems can be explained
postfire environment, suggesting that fire may act as a sig-
by high diversification rates associated with fire (Cowling &
nal, and leads to adaptive changes in animal behaviour and
Pressey, 2001; Barraclough, 2006; Pausas & Lamont, 2018).
physiology. There are also many pyrophilous fungi species
Thus, fire has been a strong driver of diversification through
promoting mutation, selection, divergence, and speciation (Robinson, Mellican & Smith, 2008), i.e. fungi that are depen-
(Lamont et al., 2013; He & Lamont, 2018b). In this section, dent on fire to stimulate spore germination and mycelial
we focus on how fire has driven the evolution and adaptations growth, analogous to fire-stimulated flowering in plants.
of terrestrial floras but also refer to some recent advances in Fires generate a wider range of habitat types than
other taxon groups. We further compare the net speciation possible in their absence, that in turn leads to increased
rate of plant groups with sister lineages in fire-prone and selection pressure to occupy such environments. From the
non-fire-prone habitats. Finally, we propose that fire itself is perspective of evolutionary processes that promote fitness in
a source of mutagens that can generate genetic novelties in fire-prone habitats, two main response types exist in plants:
plants. species with fire-stimulated resprouting, and those killed
by fire (non-resprouters) but coupled with fire-stimulated
germination of fire-surviving seeds (obligate seeders). These
(1) Fire-driven evolution and adaptation two major persistence syndromes both have special genetic
The diversity of life histories and morphologies uniquely mechanisms that promote rapid diversification (Pausas &
related to proliferation and persistence in fire-prone Keeley, 2014b). Obligate seeders have short generation
environments (Table 3; Table S1) challenges the long-held times that track the mean fire intervals of the region. For
belief that diversity patterns can be explained solely by example, the longevity of many fire-killed banksias in SW
climate, topography and soil (Pausas & Bond, 2019). Australian heathlands is about 45 years in the absence of fires
Abundant evidence shows that much of the world’s (Burgman & Lamont, 1992; Enright, Lamont & Marsula,
vegetation has been strongly fire prone since the 1996) but they are likely to be burned at 10–30-year
Mid-Cretaceous, taking precedence over Cenozoic drought intervals (Miller et al., 2007; Enright & Thomas, 2008).
as the critical agent of selection in the evolution of adaptive Species with shorter generation times evolve faster as the
traits (Brown et al., 2012; He et al., 2012; He et al., 2016; result of the accumulation of more DNA replication errors
Lamont, He & Yan, 2019b). There has been debate over per unit time, as observed in both animals (Bromham et al.,
whether adaptations to fire or drought (exaptations to fire) 2015) and plants (Soria-Hernanz et al., 2008; Yue et al., 2010;
came first among plants in fire-prone ecosystems (Keeley Buschiazzo et al., 2012), and the effect is stronger when there
et al., 2011; Bowman et al., 2016). Recent research indicates is no overlap between generations as occurs among obligate
that the presence of fire-related traits among plants has seeders. Thus, frequent fires further shorten the generation
been a direct response to the unique characteristics of time of obligate-seeder species, creating the potential for
the local fire regime as fire was already present before more rapid molecular evolution and therefore contributing
or synchronously with the advent of these traits (Lamont to rapid diversification in a landscape with diverse fire-driven
& He, 2017). Almost no fire-related traits emerged through opportunities for the selection of novel genotypes.

Table 3. Plant morphological traits whose evolution and diversification are promoted in fire-prone ecosystems, and their contribution
to fitness
Trait type General category Specific example Contribution to fitness

Protection Aerial bud bank Epicormic bud Survival
(heat-insulated
regenerative structure)
Apical bud Survival
Belowground bud Lignotuberous bud Survival
bank – woody
Rhizomatous bud
Root sucker Survival/recruitment (ramets)
Belowground bud Bulbous, tuberous, cormous buds Survival/recruitment (ramets)
bank – non-woody
Heat-insulated seed store Canopy storage Survival/recruitment
Soil storage Survival/recruitment
Heat-insulated meristems Thick bark Survival
Reproduction (fire Fire-stimulated flowering Some geophytes, grasses, shrubs Recruitment
initiates reproduc-
tive/recruitment
cycle)
Fire-stimulated seed Woody follicles, cones Recruitment
release
Fire-stimulated Smoke stimulated Recruitment
germination
Heat stimulated Recruitment
Flammability (trait Finely divided foliage Small leaves, highly branched twigs Indirect
enhances ignition and
combustion)
Retention of dead biomass Flowers, leaves, branches Indirect
Flammable secondary Essential oils, resins Indirect
compounds
Flammable tissues High cellulose, lignin; low moisture content Indirect
Fire suppression (trait Fire deterrent Crown and litter discontinuity Indirect
reduces flammability)
Low combustibility High moisture content Indirect
Poor flame transmitter Branch shedding Indirect
Fire-stimulated budburst during postfire resprouting can apply equally across all habitats, then lineages that occur
promote the expression of somatic mutations through in fire-prone habitats should be better represented here
mechanisms described in Section IV.3. Somatic mutations than in non-fire-prone habitats. To test this hypothesis,
accumulate during plant growth. Gametes produced late we searched the literature for studies that contained dated
in a plant’s life cycle are likely to possess such somatic phylogenies of seed plants with sister lineages in fire-prone
mutations and these can be transmitted to the next generation and non-fire-prone habitats. We estimated net speciation
(Bobiwash, Schultz & Schoen, 2013; Watson et al., 2016). rate of each lineage as (log10 N )/t, where N is the number
In regions with short fire intervals (e.g. 10–30 years in of extant species in the lineage, and t is the stem age of the
heathlands of SW Australia and the Cape, 1–5 years in lineage (Magallón & Sanderson, 2001). We performed 39
sub-tropical savannas), resprouting species usually live for comparions (in 20 plant families) containing sister clades with
hundreds to thousands of years, e.g. 300–1200 years for >90% of their species in either fire-prone or non-fire-prone
some Banksia species (Enright et al., 1996; Merwin et al., habitats (Fig. 7; Table S1). For 26 comparisons the number
2012), or 3800 years for Jacaranda decurrens (Alves et al., 2013), of species in fire-prone lineages was higher than that in
passing through tens to hundreds of fire cycles in that time and the non-fire-prone lineages, so contributing significantly to
creating abundant opportunities for the expression of somatic biodiversity in their respective ecosystems (Fig. 7; Table S1).
mutations as a source of novel traits among resprouting Overall, speciation rates are significantly higher in fire-prone
plants. habitats, and the number of pairs in each direction (26 and
13) is also significant with more lineages containing higher
species numbers in fire-prone habitats than in non-fire-prone
(2) Fire and diversification of plant lineages
habitats (Fig. 7). The number of families (12) containing
If fire is a major driver of ecological and evolutionary lineage pairs with higher speciation rates in fire-prone
processes, and other agents of selection are minor or lineages was not significantly greater than the number of

Fig. 7. Logarithm of net speciation rate for sister lineages in fire-prone (fp) and non-fire-prone (nfp) habitats. Data derived from
Table S1. — separates fp lineages on the left from nfp on the right. Lineages are listed from highest to lowest speciation rate for the
fp lineage when it exceeds the nfp lineage then lowest to highest speciation rate for the nfp lineage when it exceeds the fp lineage.
Speciation rates are significantly higher in fp than in nfp habitats (P = 0.008; paired, one-tailed t-test); and there are significantly
more lineages containing higher species numbers in fp habitats than in nfp habitats (26 and 13, respectively; P < 0.05, chi-squared
test with Yate’s correction).
families (11) containing pairs with higher speciation rates in Some extremely species-rich groups, such as Acacia
non-fire-prone lineages (P > 0.05); thus, the differences in (Fabaceae) with 1065 species and hard, arillate seeds, and
diversification rates are likely operating at below the family Grevillea (ant-dispersed) + Hakea (serotinous) with over 510
level. While these results are decisive, we note that the sample species, are found overwhelmingly in fire-prone habitats,
size is limited, that some clades are related and the taxon compared with only a few species in their non-fire-prone
groups available may not be representative of seed plant taxa sister lineages (Fig. 7, Table S1). Other speciose fire-prone
in general. taxon groups that were not included here (due to a lack of

clear, non-fire-prone sister lineages) are Eucalyptus and Erica. material for trait innovation and speciation (He & Lamont,
Eucalyptus (Myrtaceae, > 850 species, including Angophora and 2018b). There are two main components of a natural fire:
Corymbia) has fire-stimulated resprouting with the majority heat (from sub-lethal to >1000◦ C; Wittkuhn, Lamont &
present in fire-prone Australia (Crisp et al., 2011), and He, 2017) and products of combustion (∼4000 chemicals in
only nine species occuring in non-fire-prone Indonesia smoke; Flematti et al., 2004). Abundant evidence exists that
and New Guinea. In Erica (Ericaceae, fire-stimulated both components can induce mutations.
germination, some species with lignotubers) 690 species Fire heat causes ignition and death of unprotected
(out of 860) occur in the fire-prone Cape of South combustible tissues (Pausas et al., 2018b). A ground-breaking
Africa and Mediterranean Basin, with the remainder in discovery of the effect of heat on plant mutations was
relatively non-fire-prone parts of Africa/Madagascar and made by Pettersson (1961), who used a magnifying lens
Europe. These two cases suggest that our comparison in to concentrate sunlight onto the flower buds of plant
Fig. 7 is conservative. Among the 26 cases with higher species and subsequently germinated the seeds set from
speciation rates in fire-prone habitats, the lineages migrated the heated flowers. Many plants possessed extra leaves per
from non-fire-prone ancestral habitats (e.g. rainforest) into node, bud proliferation, and odd leaf shapes and sizes.
fire-prone environments in 24 cases. The two exceptions Surprisingly, many of these mutants resembled traits of
are Scaevola (Goodeniaceae) where limited dispersal occurred (untreated) congeneric species. Pécrix et al. (2011) showed
more recently from fire-prone Australia to the non-fire-prone that mild heat may cause meiotic dysfunctioning and
Pacific rim (Jabaily et al., 2014), and a few species of lead to the formation of diploid pollen. The formation
Disa (Orchidaceae) dispersing from fire-prone fynbos to of diploid gametes may have a significant role in
rarely burnt riverbanks (Bytebier et al., 2011). On acquiring speciation and diversification, as fertilisation by the diploid
fire-adapted traits, vast new habitats become available pollen and eggs produce novel polyploids. It is also a
that create new speciation opportunities in terms of the mechanism that overcomes the reproductive failure of
altered physical environment (climate, soil, topography) and triploids when a diploid species crosses with a polyploid
different plant–animal interactions (pollinators, herbivores, species (Ramanna & Jacobsen, 2003). Polyploidy, sometimes
granivores) become possible (Lamont et al., 2013; Pausas & as result of whole-genome duplication, is considered an
Lamont, 2018). important mechanism for the rapid diversification of
For 13 comparisons, the number of species in flowering plants across the Cretaceous–Cenozoic boundary
non-fire-prone habitats was higher than in fire-prone (Vanneste et al., 2014). In support of this, the Myrtaceae
habitats, with eight of these cases in California (Lancaster & and Proteaceae, two prominent plant families in the
Kay, 2013). Higher species numbers were also recorded fire-prone floras of Australia and South Africa, ancient
in another seven fire-prone lineages of California. The whole-genome duplication was estimated to have occurred
non-fire-prone Elaeocarpus alliance has produced six times in the Mid-Cretaceous–Paleocene (Johnson & Briggs, 1975;
more species than the fire-prone Tremandraceae (Fig. 7) Myburg et al., 2014) when fire activity is estimated to have
since the two groups diverged in the Paleogene. The high been at the highest level during the last 350 million years
speciation rate in Elaeocarpus followed migration events (Belcher et al., 2010).
from Australian rainforest to continental Asia, Madagascar,
It is also well known that smoke from numerous sources
New Zealand and the Pacific (Phoon, 2015). It therefore
carries mutagens and carcinogens (Simoneit et al., 2000;
seems that, while fire may often increase speciation of
Nolte et al., 2001; Vu et al., 2012). Polycyclic aromatic
lineages in fire-prone habitats, it is not the only, nor
hydrocarbons (PAHs) are well-known components of the
always the ‘best’, environment to promote diversity. Fire-free
gas and particulate components of smoke from burning
areas (e.g. rainforest) may also present novel agents of
biomass (Schmidl et al., 2008). Research indicates that PAHs
selection in association with extensive new habitats and
can be absorbed by plants and cause mutations (Nisbet &
foster speciation. Nevertheless, migration to fire-free habitats
LaGoy, 1992). PAH extracts from burning the Australian
among essentially fire-prone lineages is relatively recent and
tree, Eucalyptus globulus, were more mutagenic than those
the comparatively low levels of speciation observed might
from trees and shrubs native to Spain (Sharma & Tripathi,
simply be attributed to their restricted numbers (Bytebier
et al., 2011; Jabaily et al., 2014; Lamont et al., 2019b). Overall 2009; Sojinu, Sonibare & Zeng, 2011). Studies that have
these results suggest that fire is as good, or an even more used plant species from fire-prone vegetation as a source of
powerful, driver of diversification than the traditionally fire-related mutagens showed that these cause among the
considered agents of selection (climate, soils, biogeography highest levels of mutagenesis (White et al., 1983; de Oliveira
and species interactions; Pausas & Lamont, 2018). Alves et al., 2011; Vu et al., 2012; Orecchio et al., 2016).
Flower buds rarely survive fire but insulated vegetative
buds (either via thick bark or soil) usually do (Pausas et al.,
(3) Mutagenic effects of heat and chemicals 2018b). Sublethal heating of stored seeds and protected
released by combustion
meristems during fire might act as a mutagenic agent, and
Mutations are the ultimate genetic basis for adaptation and these buds could be a potent source of somatic mutations for
diversification. There is increasing evidence pointing to the promoting speciation (Lamont & Wiens, 2003). Fowler et al.
potential for fire to act as a mutagen, generating novel genetic (2018) reported somatic mutations in resprouting Banksia

Fig. 8. Possible fire-based mutagenic pathways in fire-prone ecosystems (small white ovals are buds). Modified from He & Lamont
(2018b).
attenuata in SW Australia. Litter and foliage consumption by and maintains biodiversity at local, regional and global scales
fire opens the vegetation and exposes stored seeds and buds which is manifested through relatively stable and predictable
to heat and fire-related chemicals. Burning plants and lit- fire regimes that set upper and lower limits to trait selection.
ter release aerosols and gases contaning mutagens including Not only is fire a mechanism that consumes biomass (fuel) and
PAHs and terpenoids that spread through the plant commu- creates environmental heterogeneity that drives biodiversity,
nity via the atmosphere, and most of these mutagens reach the it also promotes and maintains biodiversity by serving as an
soil surface or are deposited on unburnt leaves, buds or other agent of natural selection in evolution and speciation, and
surviving plant parts (Fig. 8). Mutagens then enter seeds or by regulating nutrient cycles and biotic interactions.
seedlings through the growing medium. The critical step is Global fire-dependent ecosystems are not an artefact
mutagen uptake by the root system (Zach & Sheppard, 1991). of recent anthropogenic burning as commonly believed.
Although no work has been done specifically in the Instead, they have existed for so long that they have
context of wildfire as a mutagenic agent that might lead to enabled the evolution of distinctive fire-adapted species and
trait innovation and speciation, current research collectively communities (Bond et al., 2005; Pausas & Keeley, 2009;
supports the view that fire (and its byproducts) can induce Bond, 2015; Archibald et al., 2018; Karp, Behrensmeyer &
mutations in meristems and seeds (He & Lamont, 2018b). Freeman, 2018; He & Lamont, 2018a). Doerr & Santín
In turn, this provides a mechanism that can explain how (2016) reported an overall decline in fire activity over
fire may have a direct role in stimulating plant speciation in the past century, and argued that much of the world
fire-prone environments. To what extent smoke may be a suffers from the harmful consequences of fire-management
source of mutations among animals remains unknown, but strategies centred on ‘fire suppression’. As fire is a natural and
insects with immature stages (eggs, larvae, or pupae) in the integral part of many terrestrial ecosystems, fire-management
soil are likely to receive mutagenic agents from wildfires. strategies that attempt to remove fire from these areas
to protect human life and property may have negative
consequences on biodiversity conservation (Kitchen, 2012).
Active fire suppression has resulted in less-frequent but
V. IMPLICATIONS FOR BIODIVERSITY higher intensity fires, causing fundamental changes in stand
CONSERVATION AND FIRE MANAGEMENT structure and composition, and greater risk to life and
property (Schoennagel et al., 2017). Management strategies
The bigger picture that emerges from this review is that fire such as intensive logging and clearfelling might be able to
is a major ecological and evolutionary force that promotes reduce fire risks to some extent but they will be detrimental to

maintaining ecosystem biodiversity, structure and function. lead to species decline and forest degradation (Driscoll
Fire itself is the most important and naturally available tool et al., 2010). An understanding of how lightning-caused
used in landscape/vegetation management and provides a wildfires structure landscapes is essential when considering
range of services to humankind (Pausas & Keeley, 2019), prescribed burning (anthropogenic ignition) to manage and
but current sociocultural, political and economic factors, restore fire-dependent ecosystems (Ghazoul & Chazdon,
together with historical biases (Pausas & Bond, 2019), greatly 2017; Gordijn et al., 2018). Although the ecological and
influence acceptable fire-management practices (Buizer & evolutionary roles of fires are increasingly understood
Kurz, 2016). by fire ecologists and fire-management authorities, their
Prescribed burning (vegetation burnt under mild weather implementation remains controversial among the public and
conditions) has long been recognised as an effective strategy decision-makers as they continue to view fire as essentially
for reducing fire risk. This revolutionary technique with destructive (Ingalsbee, 2017). This could lead to undesirable
its underlying philosophy of using fire to ‘fight’ fire was fire-management outcomes for plant communities and
developed in Western Australia in the early 1960s (Shea, the environment. Science-based fire management of
1994) although we note that ‘wildfires’, to which most fire-dependent ecosystems requires an understanding of the
species will be adapted, are not part of this protocol. It was feedbacks between ecosystem structure, biodiversity values,
later adopted by fire-management authorities throughout and societal perceptions (Moritz et al., 2014; Pausas & Keeley,
Australia and the USA. The hypothesis that pyrodiversity 2019). A paradigm shift to restore natural fire regimes
begets biodiversity has become increasingly influential in across broad landscapes is required among fire research
guiding land management and biodiversity conservation over and management communities, and requires fire ecologists
recent years (Kelly & Brotons, 2017). The practice of patch and other professionals to spread fire-ecology knowledge
(mosaic) burning, creating spatial variation in components of beyond scientific publications to reach the broader public,
the fire regime, has recently been promoted as the preferred politicians and the media. Failure to appreciate that floras
fire-management protocol to maintain biodiversity based in fire-prone regions are fire-adapted can only undermine
on the pyrodiversity-biodiversity hypothesis (Bowman et al., our ability to use fire effectively, and undermine public
2016; Kelly & Brotons, 2017). Evidence in support of this confidence in fire-management practices.
proposal is currently mixed, while our modelling indicates
that there is a ceiling to its benefits.
Similarly, the IDH, that there is an optimum fire activity VI. FUTURE RESEARCH DIRECTIONS AND
for maximizing biodiversity, is supported by ecological OUTCOMES
principles when variations in fire frequency are considered
but the theory does not recognise that species composition at To advance our understanding of ecological processes
the extremes of the fire gradient may be quite different from and biodiversity patterns, we must move away from the
those at moderate levels (Davies et al., 2018; Gordijn et al., traditional view that biodiversity is controlled solely by
2018). When and where the fire occurs in a landscape are climate and soils, and accept fire as a major ecological
determined by the vegetation, season, local climate/weather, factor and evolutionary force that shapes the distribution
habitat topography and ignition source. The interaction and abundance of biota (Keeley et al., 2011, 2012; Bond,
between fire and biodiversity depends on context, and its 2015; Pausas & Lamont, 2018; He & Lamont, 2018a; Pausas
properties vary among species, ecosystems and across spatial & Bond, 2019). The relationships between fire properties and
scales. High pyrodiversity may have theoretical support but biodiversity are likely to vary among species and ecosystems.
variation in its components must be within the bounds of the It is essential that policy-makers recognise that natural
natural fire regime that has assembled the vegetation in the ecosystems contain different species with different fire-regime
region over millions of years to maintain the current species requirements. Fire management should be guided by
composition and diversity. Further, fires may open up the demonstrated knowledge on the fire ecology of plants and
vegetation and promote invasion by exotic species, that, once animals in the landscape that they inhabit. Research focused
established, may outcompete indigenous species. Thus, an on the population dynamics of target species (Groeneveld
unintended consequence of patches of more frequent fires et al., 2013) is essential for understanding how different species
(higher pyrodiversity) may in fact be a decrease in biodiversity respond to different fire regimes.
(Fisher et al., 2009). This can never be an objective of The fundamental question at the centre of the relationships
conservation management. between fire and biota is how does fire stimulate the evolution
In the absence of a complete understanding of of new species? To answer this question, we must move
how fire regimes are maintained, it is likely that from a descriptive and process-based narrative on the role
random production of fire-caused patches will not always of fire in speciation to a mechanism-based approach. For
generate optimal fire mosaics for maintaining biodiversity. example, research examining direct evidence that fire (and its
Fire-management practices that recognise historical fire byproducts) act as a source of mutagens that induce mutations
regimes have positive nature conservation outcomes in many upon which selection can act, leading to divergence and
systems. Evidence exists that inappropriate fire regimes, adaptation, will advance our knowledge on the fundamental
as result of ill-planned prescribed burning, could also role of fire in the evolution of terrestrial biota (He & Lamont,

2018b). Current genomic advances are contributing to an the biodiversity relationship with P is hump-shaped, with the
understanding of the genetic architecture of plant adaptive peak at relatively high P (left-skewed).
traits, including those related to fire (Castellanos et al., (7) Mixed-severity (fire) regimes (MSRs) are a particular
2015; He et al., 2019). Ultimately, we need research into type of high P involving stand replacement. The relationships
the underlying genetic basis of ecological adaptations of of biodiversity with increasing fire frequency/severity are
fire-prone floras. hump-shaped for both the IDH and MSR.
The notion that ‘pyrodiversity begets biodiversity’ is (8) Species associated with recurrent fire have evolved
increasingly advocated in fire management, where fires are many traits that increase their fitness in the presence of fire.
manipulated to create a mosaic of patches with different fire For 39 comparisons, the number of species in 26 fire-prone
histories across space and time (Kelly & Brotons, 2017). Our lineages is much higher than that in their non-fire-prone
preliminary modelling indicates that there are limits to the sister lineages.
benefits of pyrodiversity, but more sophisticated approaches (9) Fire itself may be a mutagenic agent and can produce
are required, such as calculating the effect of varying novel traits, although its significance in driving biodiversity
patch size. To what extent are the various components of has yet to be explored.
pyrodiversity correlated and how does it increase as the size (10) A paradigm shift aimed at restoring
of patches decreases? The logistic constraints in research and biodiversity-maintaining fire regimes across broad
implementation in management should also be evaluated. landscapes is required from both research and management
Finally, fire has not featured strongly in the development perspectives.
of current ecological and evolutionary theories in general
because of the prevailing cultural bias against viewing fire
as a significant ecological factor and fundamental agent of VIII. ACKNOWLEDGEMENTS
natural selection (Pausas & Bond, 2019; Lamont et al., 2019a).
Fire ecologists should communicate current understanding
of fire–biodiversity relationships to fellow biologists, the The research of T.H. and B.B.L. on fire-driven evolu-
general public, the media, and policy-makers, so that the key tion was supported by the Australian Research Council
role of fire on ecological and evolutionary processes in much (DP120103389). J.G.P. acknowledges support from the
of the Earth’s terrestrial biota is incorporated into nature Spanish and Valencia Regional Governments (projects
conservation and management. CGL2015-64086-P, PGC2018-096569-B-I00, PROME-
TEO/2016/021).
VII. CONCLUSIONS
IX. REFERENCES
(1) Many biodiverse ecosystems are fire prone and, by *References marked with an asterisk are cited only within the supporting information.
examining mechanisms and fire-related gradients operating Agee, J. K. (1993). Fire ecology of Pacific Northwest Forests. Island Press, Washington, D.C.
at ecological and evolutionary scales, we show that the Agee, J. K. (1998). The landscape ecology of western forest fire regimes. Northwest
Science 72, 24–34.
relationship is usually causal. Agee, J. K. (2005). The complex nature of mixed severity fire regimes. In Mixed Severity
(2) Ecological responses are controlled by six fire Fire Regimes: Ecology and Management (eds L. Taylor, J. Zelnik, S. Cadwallader
components whose properties vary greatly through space and B. Hughes). Association for Fire Ecology, Spokane.
Ahlgren, I. F. & Ahlgren, C. E. (1960). Ecological effects of forest fires. The Botanical
and time and define the type of fire during a fire event, and Review 26, 483–533.
the fire regime over extended periods. Alfaro-Sánchez, R., Camarero, J. J., Sánchez-Salguero, R.,
(3) Fire intensity (vertical spread) and frequency are Sangüesa-Barreda, G. & De Las Heras, J. (2016). Post-fire Aleppo
pine growth, C and N isotope composition depend on site dryness. Trees 30,
the major components controlling species composition, 581–595.
vegetation structure, and plant functional traits. In turn, Alves, R. J. V., Silva, N. G., Fernandes Junior, A. J. & Guimaraes, A. R.
species composition may feed back to control the fire regime. (2013). Longevity of the Brazilian underground tree Jacaranda decurrens Cham. Anais
da Academia Brasileira de Ciências 85, 671–678.
(4) Species diversity is greatest immediately postfire and * Aparicio, A., Martín-Hernanz, S., Parejo-Farnés, C., Arroyo, J., Lavergne,
declines gradually, although the abundance of individual S., Yeşilyurt, E. B., Zhang, M. L., Rubio, E. & Albaladejo, R. G. (2017).
Phylogenetic reconstruction of the genus Helianthemum (Cistaceae) using plastid
plant and animal species peaks at different times; and nuclear DNA-sequences: systematic and evolutionary inferences. Taxon 66,
postfire immigration/emigration are rare among plants 868–885.
(autosuccession) but common among animals. Archibald, S., Lehmann, C., Belcher, C., Bond, W., Bradstock, R., Daniau,
A., Dexter, K., Forrestel, E., Greve, M., He, T., Higgins, S., Lamont,
(5) The intermediate disturbance hypothesis (IDH) B. B., Mcglinn, D., Moncreiff, G., Osborne, C., et al. (2018). Biological and
regarding conditions that favour maximum species diversity geophysical feedbacks with fire in the Earth system. Environmental Research Letters 13,
can be applied to fire-prone landscapes and is supported, 3.
Arno, S. F., Parsons, D. J. & Keane, R. E. (2000). Mixed-severity fire regimes in the
provided there is a fixed species pool. northern Rocky Mountains: consequences of fire exclusion and options for the future.
(6) Pyrodiversity (P) emphasises patchiness of the landscape In Wilderness Science in a Time of Change Conference-Volume 5: Wilderness Ecosystems, Threats,
caused by varying fire types and time since fire across the and Management; 1999 May 23–27; Missoula, MT. Proceedings RMRS-P-15-VOL-5 (eds
D. N. Cole, S. F. McCool, W. T. Borri and J. O’Loughlin), pp. 225–232.
landscape. Species—area constraints mean that there is a US Department of Agriculture, Forest Service, Rocky Mountain Research Station
limit to P as a promoter of biodiversity within a defined area: Ogden, Ogden.

Balch, J. K., Brando, P. M., Nepstad, D. C., Coe, M. T., Silvério, D., Massad, * Brown, G. K., Murphy, D. J. & Ladiges, P. Y. (2011). Relationships of the
T. J., Davidson, E. A., Lefebvre, P., Oliveira-Santos, C., Rocha, W. & Australo-Malesian genus Paraserianthes (Mimosoideae: Leguminosae) identifies the
Cury, R. T. (2015). The susceptibility of southeastern Amazon forests to fire: sister group of Acacia sensu stricto and two biogeographical tracks. Cladistics 27,
insights from a large-scale burn experiment. Bioscience 65, 893–905. 380–390.
Barraclough, T. G. (2006). What can phylogenetics tell us about speciation in the Brown, J. & York, A. (2017). Fly and wasp diversity responds to elements of both the
Cape flora? Diversity and Distributions 12, 21–26. visible and invisible fire mosaic. International Journal of Wildland Fire 26, 434–443.
* Barton, A. M. (2005). Response of Arbutus arizonica (Arizona Madrone) to fire in
Brown, S. A., Scott, A. C., Glasspool, I. J. & Collinson, M. E. (2012).
southeastern Arizona. The Southwestern Naturalist 50, 7–11. Cretaceous wildfires and their impact on the Earth system. Cretaceous Research 36,
Bassett, M., Leonard, S. W., Chia, E. K., Clarke, M. F. & Bennett, A. F. 162–190.
(2017). Interacting effects of fire severity, time since fire and topography on vegetation Buizer, M. & Kurz, T. (2016). Too hot to handle: depoliticisation and the discourse
structure after wildfire. Forest Ecology and Management 396, 26–34. of ecological modernisation in fire management debates. Geoforum 68, 48–56.
Beale, C. M., Courtney Mustaphi, C. J., Morrison, T. A., Archibald, S., Burgman, M. A. & Lamont, B. B. (1992). A stochastic model for the viability of
Anderson, T. M., Dobson, A. P., Donaldson, J. E., Hempson, G. P., Probert, Banksia cuneata populations: environmental, demographic and genetic effects. Journal
J. & Parr, C. L. (2018). Pyrodiversity interacts with rainfall to increase bird and of Applied Ecology 29, 719–727.
mammal richness in African savannas. Ecology Letters 21, 557–567. Burkle, L. A., Myers, J. A. & Belote, R. T. (2015). Wildfire disturbance and
Beckage, B. & Stout, J. (2000). Effects of repeated burning on species richness in productivity as drivers of plant species diversity across spatial scales. Ecosphere 6, 1–4.
a Florida pine savanna: a test of the intermediate disturbance hypothesis. Journal of Burton, T. A. (2005). Fish and stream habitat risks from uncharacteristic wild re:
Vegetation Science 11, 113–122. observations from 17 years of re-related disturbances on the Boise National Forest,
Belcher, C. M., Collinson, M. E. & Scott, A. C. (2013). A 450-million-year Idaho. Forest Ecology and Management 211, 140–149.
history of fire. In Fire Phenomena and the Earth System: An Interdisciplinary Guide to Fire Buschiazzo, E., Ritland, C., Bohlmann, J. & Ritland, K. (2012). Slow but not
Science (ed. C. M. Belcher), pp. 229–229. Wiley, New Jersey. low: genomic comparisons reveal slower evolutionary rate and higher dN/dS in
Belcher, C. M., Yearsley, J. M., Hadden, R. M., McElwain, J. C. & Rein, conifers compared to angiosperms. BMC Evolutionary Biology 12, 8.
G. (2010). Baseline intrinsic flammability of Earth’s ecosystems estimated from Bytebier, B., Antonelli, A., Bellstedt, D. U. & Linder, H. P. (2011). Estimating
paleoatmospheric oxygen over the past 350 million years. Proceedings of the National the age of fire in the Cape flora of South Africa from an orchid phylogeny. Proceedings
Academy of Sciences of the United States of America 107, 22448–22453. of the Royal Society B 278, 188–195.
Benda, L., Miller, D., Bigelow, P. & Andras, K. (2003). Effects of post-wildfire Campbell, E. M., Alfaro, R. I. & Hawkes, B. (2007). Spatial distribution of
erosion on channel envi ronments, Boise River, Idaho. Forest Ecology and Management mountain pine beetle outbreaks in relation to climate and stand characteristics: a
178, 105–119. dendroecological analysis. Journal of Integrative Plant Biology 49, 168–178.
Benkman, C. W. & Siepielski, A. M. (2004). A keystone selective agent? Pine Carbone, L. M., Tavella, J., Pausas, J. G. & Aguilar, R. (2019). A global synthesis
squirrels and the frequency of serotiny in lodgepole pine. Ecology 85, 2082–2087. of fire effects on pollinators. Global Ecology and Biogeography. https://doi.org/10.1111/
Bird, R. B., Bird, D. W., Codding, B. F., Parker, C. H. & Jones, J. H. (2008). The geb.12939.
‘‘fire stick farming’’ hypothesis: Australian aboriginal foraging strategies, biodiversity, Castellanos, M. C., González-Martínez, S. & Pausas, J. G. (2015). Field
and anthropogenic fire mosaics. Proceedings of the National Academy of Sciences of the United heritability of a plant adaptation to fire in heterogeneous landscapes. Molecular
States of America 105, 14796–14801. Ecology 24, 5633–5642.
Bird, R. B., Tayor, N., Codding, B. F. & Bird, D. W. (2013). Niche construction Cavender-Bares, J., Ackerly, D. D., Baum, D. A. & Bazzaz, F. A. (2004).
and dreaming logic: aboriginal patch mosaic burning and varanid lizards (Varanus Phylogenetic overdispersion in Floridian oak communities. American Naturalist 163,
gouldii). in Australia. Proceedings of Royal Society B 280, 20132297. 823–843.
Bobiwash, K., Schultz, S. T. & Schoen, D. J. (2013). Somatic deleterious mutation Charles-Dominique, T., Beckett, H., Midgley, G. F. & Bond, W. J. (2015). Bud
rate in a woody plant: estimation from phenotypic data. Heredity 111, 338–344. protection: a key trait for species sorting in a forest–savanna mosaic. New Phytologist
Bond, W. J. (2008). What limits trees in C4 grasslands and savannas? Annual Review of 207, 1052–1060.
Ecology, Evolution, and Systematics 39, 641–659. Cochrane, M. A. & Barber, C. P. (2009). Climate change, human land use and
Bond, W. J. (2015). Ancient grasslands at risk. Science 351, 120–122. future fires in the Amazon. Global Change Biology 15, 601–612.
Bond, W. J. & Keeley, J. E. (2005). Fire as a global ’herbivore’: the ecology and Cohn, J. S., DiStefano, J., Christie, F., Cheers, G. & York, A. (2015). How
evolution of flammable ecosystems. Trends in Ecology & Evolution 20, 387–394. do heterogeneity in vegetation types and post- re age-classes contribute to plant
Bond, W. J. & Parr, C. L. (2010). Beyond the forest edge: ecology, diversity and diversity at the landscape scale? Forest Ecology and Management 346, 22–30.
conservation of the grassy biomes. Biological Conservation 143, 2395–2404. Collins, B. M., Kelly, N. M., van Wagtendonk, J. W. & Stephens, S. L. (2007).
Bond, W. J. & Scott, A. C. (2010). Fire and the spread of flowering plants in the Spatial patterns of large natural fires in Sierra Nevada wilderness area. Landscape
cretaceous. New Phytologist 188, 1137–1150. Ecology 22, 545–557.
Bond, W. J., Woodward, F. I. & Midgley, G. F. (2005). The global distribution of Collins, B. M. & Stephens, S. L. (2010). Stand-replacing patches within a
ecosystems in a world without fire. New Phytologist 165, 525–538. ‘mixed severity’fire regime: quantitative characterization using recent fires in a
Boucher, J., Azeria, E. T., Ibarzabal, J. & Hébert, C. (2012). Saproxylic beetles long-established natural fire area. Landscape Ecology 25, 927–939.
in disturbed boreal forests: temporal dynamics, habitat associations, and community Connell, J. H. (1978). Diversity in tropical rainforests and coral reefs. Science 199,
structure. Ecoscience 19, 328–343. 1302–1310.
Bowman, D. M. J. S., O’Brien, J. A. & Goldammer, J. G. (2012). Pyrogeography Cowling, R. M. (1987). Fire and its role in coexistence and speciation in Gondwanan
and the global quest for sustainable fire management. Annual Review of Environment shrublands. South African Journal of Science 83, 106–112.
and Resources 38, 57–80. Cowling, R. M. & Pressey, R. M. (2001). Rapid plant diversification: planning for
Bowman, D. M. J. S., Perry, G. L., Higgins, S. I., Johnson, C. N., Fuhlendorf, an evolutionary future. Proceedings of the National Academy of Sciences of the United States of
S. D. & Murphy, B. P. (2016). Pyrodiversity is the coupling of biodiversity and fire America 98, 5452–5457.
regimes in food webs. Philosophical Transactions of the Royal Society B 371, 20150169. Cramer, M. D. & Verboom, G. A. (2017). Measures of biologically relevant
Bradstock, R. A. (2008). Effects of large fires on biodiversity in south-eastern environmental heterogeneity improve prediction of regional plant species richness.
Australia: disaster or template for diversity? International Journal of Wildland Fire 17, Journal of Biogeography 44, 579–591.
809–822. Crisp, M. D., Burrows, G. E., Cook, L. G., Thornhill, A. H. & Bowman,
Bradstock, R. A., Bedward, M., Gill, A. M. & Cohn, J. S. (2005). Which mosaic? D. M. J. S. (2011). Flammable biomes dominated by eucalypts originated at the
A landscape ecological approach for evaluating interactions between fire regimes, Cretaceous-Palaeogene boundary. Nature Communications 2, 193.
habitat and animals. Wildlife Reseearch 32, 409–423. Curtis, N. P. (1998). A postfire ecological study of Xanthorrhoea australis following
Bradstock, R. A., Bedward, M., Kenny, B. J. & Scott, J. (1998). Spatially-explicit prescribed burning in the Warby Range State Park, north-eastern Victoria, Australia.
simulation of the effect of prescribed burning on fire regimes and plant extinctions Australian Journal of Botany 46, 253–272.
in shrublands typical of south-eastern Australia. Biological Conservation 86, 83–95. Dantas, V. L., Batalha, M. A. & Pausas, J. G. (2013). Fire drives functional
Bradstock, R. A., Tozer, M. G. & Keith, D. A. (1997). Effects of high frequency thresholds on the savanna–forest transition. Ecology 94, 2454–2463.
fire on floristic composition and abundance in a fire prone heathland near Sydney. Dantas, V. L., Hirota, M., Oliveira, R. S. & Pausas, J. G. (2016). Disturbance
Australian Journal of Botany 45, 641–655. maintains alternative biome states. Ecology Letters 9, 12–19.
Briani, D. C., Palma, A. R. T., Viera, E. M. & Henriques, R. P. B. (2004). Davies, A. B., Eggleton, P., van Rensburg, B. J. & Parr, C. L. (2012). The
Postfire succession of small mammals in the Cerrado of central Brazil. Biodiversity and pyrodiversity–biodiversity hypothesis: a test with savanna termite assemblages.
Conservation 13, 1023–1037. Journal of Applied Ecology 49, 422–430.
Bromham, L., Hua, X., Lanfear, R. & Cowman, P. F. (2015). Exploring the Davies, H. F., McCarthy, M. A., Rioloi, W., Puruntatameri, J., Roberts,
relationships between mutation rates, life history, genome size, environment, and W., Kerinaiua, C., Kerinauia, V., Womatakimi, K. B., Andersen, A. N. &
species richness in flowering plants. The American Naturalist 185, 507–524. Murphy, B. P. (2018). An experimental test of whether pyrodiversity promotes

mammal diversity in a northern Australian savanna. Journal of Applied Ecology 55, Gómez-González, S., Torres-Díaz, C., Bustos-Schindler, C. & Gianoli, E.
2124–2134. (2011). Anthropogenic fire drives the evolution of seed traits. Proceedings of the National
Davis, H., Ritchie, E. G., Avitable, S., Doherty, T. & Nimmo, D. G. (2018). Academy of Sciences of the United States of America 108, 18743–18747.
Testing the assumptions of the pyrodiversity begets biodiversity hypothesis for Gordijn, P. J., Everson, T. M. & O’Connor, T. G. (2018). Resistance of
termites in semi-arid Australia. Royal Society Open Science 5, 172055. Drakensberg grasslands to compositional change depends on the influence of
de Oliveira Alves, N., Matos Loureiro, A. L., Dos Santos, F. C., Nascimento, fire-return interval and grassland structure on richness and spatial turnover.
K. H., Dallacort, R., de Castro Vasconcellos, P., de Souza Hacon, S., Perspectives in Plant Ecology, Evolution and Systematics 34, 26–36.
Artaxo, P. & de Medeiros, S. R. (2011). Genotoxicity and composition of Gosper, C. R., Yates, C. J., Prober, S. M. & Parsons, B. C. (2012). Contrasting
particulate matter from biomass burning in the eastern Brazilian Amazon region. changes in vegetation structure and diversity with time since fire in two Australian
Ecotoxicology and Environmental Safety 74, 1427–1433. Mediterranean-climate plant communities. Austral Ecology 37, 164–174.
Dee, J. R. & Menges, E. S. (2014). Gap ecology in the Florida scrubby flatwoods: Griffith, A. D., Garnett, S. T. & Brook, B. W. (2015). Fire frequency matters
effects of time-since-fire, gap area, gap aggregation and microhabitat on gap species more than fire size: testing the pyrodiversity–biodiversity paradigm for at-risk small
diversity. Journal of Vegetation Science 25, 1235–1246. mammals in an Australian tropical savanna. Biological Conservation 186, 337–346.
Dixon, R. D. & Saab, V. A. (2000). Black-backed woodpecker (Picoides arcticus). In The Groeneveld, J., Enright, N. J., Lamont, B. B., Reineking, B., Frank, K. &
Birds of North America (eds A. Poole and F. Gill ). Academy of National Science and Perry, G. L. W. (2013). Species-specific traits plus stabilizing processes best explain
American Ornithologists’ Union, Philadelphia.Number 509. coexistence in biodiverse fire-prone plant communities. PLoS One 8, e65084.
Doerr, S. H. & Santín, C. (2016). Global trends in wildfire and its impacts: Grubb, P. J. (1977). The maintenance of species-richness in plant communities: the
perceptions versus realities in a changing world. Philosophical Transactions of the Royal importance of the regeneration niche. Biological Reviews 52, 107–145.
Society B 371, 20150345. Grundel, R., Jean, R. P., Frohnapple, K. J., Glowacki, G. A., Scott, P. E. &
Driscoll, D. A., Lindenmayer, D. B., Bennett, A. F., Bode, M., Bradstock, Pavlovic, N. B. (2010). Floral and nesting resources, habitat structure, and fire
R. A., Cary, G. J., Clarke, M. F., Dexter, N., Fensham, R., Friend, G., Gill, influence bee distribution across an open-forest gradient. Ecological Applications 20,
M., James, S., Kay, G., Keith, D. A., Macgregor, C., Russell-Smith, J., Salt, 1678–1692.
D., Watson, J. E. M., Williams, R. J. & York, A. (2010). Fire management for Hanes, T. L. (1971). Succession after fire in the chaparral of southern California.
biodiversity conservation: key research questions and our capacity to answer them. Ecological Monographs 41, 27–52.
Biological Conservation 143, 1928–1939. Hawkins, B. A., Field, R., Cornell, H. V., Currie, D. J., Guégan, J. F.,
Enright, N. J., Fontaine, J., Bowman, D., Bradstock, R. & Williams, R. Kaufman, D. M., Kerr, J. T., Mittelbach, G. G., Oberdorff, T., O’Brien,
(2015). Interval squeeze: altered fire regimes and demographic responses interact E. M. & Porter, E. E. (2003). Energy, water, and broad-scale geographic patterns
to threaten woody species persistence as climate changes. Frontiers in Ecology and the of species richness. Ecology 84, 3105–3117.
Environment 13, 265–272. He, T. (2013). Structural equation modelling analysis of evolutionary and ecological
Enright, N. J., Fontaine, J., Lamont, B., Miller, B. & Westcott, V. (2014). patterns in Australian Banksia. Population Ecology 55, 461–467.
Resistance and resilience to changing climate and fire regime depend on plant He, T., Krauss, S. L., Lamont, B. B., Miller, B. P. & Enright, N. J. (2004). Long
functional traits. Journal of Ecology 102, 1572–1581. distance dispersal in a metapopulation of Banksia hookeriana inferred by population
allocation from AFLP data. Molecular Ecology 13, 1099–1109.
Enright, N. J., Lamont, B. B. & Marsula, R. (1996). Canopy seed bank dynamics
He, T. & Lamont, B. B. (2018a). Baptism by fire: the pivotal role of ancient
and optimum fire regime for the highly serotinous shrub, Banksia hookeriana. Journal
conflagrations in evolution of the Earth’s flora. National Science Review 5, 237–254.
of Ecology 84, 9–17.
He, T. & Lamont, B. B. (2018b). Fire as a potent mutagen driving the evolution of
Enright, N. J., Marsula, R., Lamont, B. B. & Wissel, C. (1998). The ecological
terrestrial plants. Critical Reviews in Plant Sciences 37, 1–14.
significance of canopy seed storage in fire-prone environments: a model for
He, T., Lamont, B. B. & Downes, K. S. (2011). Banksia born to burn. New Phytologist
nonsprouting shrubs. Journal of Ecology 86, 946–959.
191, 184–196.
Enright, N. J., Mosner, E., Miller, B. P., Johnson, N. & Lamont, B. B.
He, T., Lamont, B. B., Enright, N., D’Agui, H. M. & Stock, W. D.
(2007). Soil versus canopy seed storage and plant species coexistence in species-rich
(2019). Environmental drivers and genomic architecture of trait differentiation in
shrublands of southwestern Australia. Ecology 88, 2292–2304.
fire-adapted Banksia attenuata ecotypes. Journal of Integrative Plant Biology 61, 417–432.
Enright, N. J. & Thomas, I. (2008). Pre-European fire regimes in Australian
He, T., Lamont, B. B., Krauss, S. L., Enright, N. J., Miller, B. P. & Gove, A. D.
ecosystems. Geography Compass 2, 979–1011.
(2009). Ants cannot account for inter-population dispersal of the arillate pea Daviesia
Evans, E. W. (1988). Grasshopper (Insecta: Orthoptera: Acrididae) assemblages of
triflora. New Phytologist 181, 725–733.
tallgrass prairie: influences of fire frequency, topography, and vegetation. Canadian
He, T., Lamont, B. B. & Manning, J. A. (2016). A Cretaceous origin for fire
Journal of Zoology 66, 1495–1501. adaptations in the Cape flora. Scientific Reports 6, 34880.
Farnsworth, L. M., Nimmo, D. G., Kelly, L. T., Bennett, A. F. & Clarke, He, T., Pausas, J. G., Belcher, C. M., Schwilk, D. W. & Lamont, B. B. (2012).
M. F. (2014). Does pyrodiversity beget alpha, beta or gamma diversity? A case study Fire-adapted traits of Pinus arose in the fiery cretaceous. New Phytologist 194, 751–759.
using reptiles from semi-arid Australia. Diversity and Distributions 20, 663–673. Heinselman, M. L. (1981). Fire and succession in the conifer forests of northern
Fernandes, G. W., Barbosa, N. P. U., Alberton, B., Barbieri, A., Dirzo, R., North America. In Forest Succession (eds. D. C. West., H. H. Shugart. and D. B.
Goulart, F., Guerra, T. J., Morellato, L. P. C. & Solar, R. R. C. (2018). The Botkin), pp. 374–405. Springer, New York.
deadly route to collapse and the uncertain fate of Brazilian rupestrian grasslands. Herath, D. N., Lamont, B. B., Enright, N. J. & Miller, B. P. (2009). Impact
Biodiversity and Conservation 27, 2587–2603. of fire on plant-species persistence in post-mine restored and natural shrubland
Fisher, J. L., Loneragan, W. A., Dixon, K., Delaney, J. & Veneklaas, E. J. communities in southwestern Australia. Biological Conservation 142, 2175–2180.
(2009). Altered vegetation structure and composition linked to fire frequency and Hernández-Serrano, A., Verdú, M., González-Martínez, S. C. & Pausas,
plant invasion in a biodiverse woodland. Biological Conservation 142, 2270–2281. J. G. (2013). Fire structures pine serotiny at different scales. American Journal of Botany
Flematti, G. R., Ghisalberti, E. L., Dixon, K. W. & Trengove, R. D. (2004). A 100, 2349–2356.
compound from smoke that promotes seed germination. Science 305, 977. Hessburg, P. F., Spies, T. A., Perry, D. A., Skinner, C. N., Taylor, A. H.,
Foster, C. N., Barton, P. S., Robinson, N. M., MacGregor, C. I. & Brown, P. M., Stephens, S. L., Larson, A. J., Churchill, D. J., Povak, N. A.
Lindenmayer, D. B. (2017). Effects of a large wildfire on vegetation structure & Singleton, P. H. (2016). Tamm review: management of mixed-severity fire
in a variable fire mosaic. Ecological Applications 27, 2369–2381. regime forests in Oregon, Washington, and Northern California. Forest Ecology and
Fowler, W., Deng, X., Lamont, B. B. & He, T. (2018). Resprouters, assisted by Management 366, 221–250.
somatic mutations, are as genetically diverse as nonsprouters in the world’s fire-prone * Hill, S. R. & Halse, R. R. (2014). A new subspecies of Sidalcea hickmanii (Malvaceae)
ecosystems. Acta Oecologica 92, 1–6. from Oregon. Phyton 113, 1–8.
Francis, A. P. & Currie, D. J. (2003). A globally consistent richness– climate Hinman, S. H. & Brewer, J. S. (2007). Responses of two frequently-burned wet pine
relationship for angiosperms. American Naturalist 161, 523–536. savannas to an extended period without fire. The Journal of the Torrey Botanical Society
García, Y., Castellanos, M. C. & Pausas, J. G. (2018). Differential pollinator 134, 512–526.
response underlies plant reproductive resilience after fires. Annals of Botany 122, * Hjältén, J., Johansson, T., Alinvi, O., Danell, K., Ball, J. P., Pettersson,
961–971. R. & Hilszczański, G. J. (2007). The importance of substrate type, shading and
Ghazoul, J. & Chazdon, R. (2017). Degradation and recovery in changing forest scorching for the attractiveness of dead wood to saproxylic beetles. Basic and Applied
landscapes: a multiscale conceptual framework. Annual Review of Environment and Ecology 8, 364–376.
Resources 42, 161–188. Hollenbeck, J. P., Saab, V. A. & Frenzel, R. W. (2011). Habitat suitability and nest
Gillman, L. N. & Wright, S. D. (2006). The influence of productivity on the species survival of white-headed woodpeckers in unburned forests of Oregon. The Journal of
richness of plants: a critical assessment. Ecology 87, 1234–1243. Wildlife Management 75, 1061–1071.
Goldblatt, P. & Manning, J. C. (2000). Cape Plants: A Conspectus of the Cape Flora. Howard, D. R. & Hill, P. S. M. (2007). The effect of fire on spatial distributions of
National Botanical Institute, RSA & Missouri Botanical Garden, St. Louis. male mating aggregations in Gryllotalpa major Saussure (Orthoptera: Gryllotalpidae)

at the Nature Conservancy’s Tallgrass Prairie Preserve in Oklahoma: evidence of a * Klocke, D., Schmitz, A., Soltner, H., Bousack, H. & Schmitz, H. (2011).
fire-dependent species. Journal of the Kansas Entomological Society 80, 51–64. Infrared receptors in pyrophilous (‘fire loving’) insects as model for new un-cooled
Hoyt, J. S. & Hannon, S. J. (2002). Habitat associations of black- backed and infrared sensors. Beilstein Journal of Nanotechnology 2, 186–197.
three-toed woodpeckers in the boreal forest of Alberta. Canadian Journal of Forest Krawchuk, M. & Moritz, M. (2011). Constraints on global fire activity vary across
Research 32, 1881–1888. a resource gradient. Ecology 92, 121–132.
Hu, M., Liu, Y., Sun, Z., Zhang, K., Liu, Y., Miao, R. & Wan, S. (2018). Fire Krawchuk, M. A., Moritz, M. A., Parisien, M. A., Van Dorn, J. & Hayhoe, K.
rather than nitrogen addition affects understory plant communities in the short term (2009). Global pyrogeography: the current and future distribution of wildfire. PLoS
in a coniferous-broadleaf mixed forest. Ecology and Evolution 8, 8135–8148. One 4, 5102.
Huston, M. A. (2014). Disturbance, productivity, and species diversity: empiricism Kreft, H. & Jetz, W. (2007). Global patterns and determinants of vascular plant
vs. logic in ecological theory. Ecology 95, 2382–2396. diversity. Proceedings of the National Academy of Sciences of the United States of America 104,
Hutto, R. L. (1995). Composition of bird communities following stand replacement 5925–5930.
fires in northern Rocky Mountain (USA) conifer forests. Conservation Biology 9, Kuneš, P., Abraham, V. & Herben, T. (2019). Changing disturbance-diversity
1041–1058. relationships in temperate ecosystems over the past 12000 years. Journal of Ecology
* Hutto, R. L., Bond, M. L. & DellaSala, D. A. (2015). Using bird ecology to learn 107, 1678–1688. https://doi.org/10.1111/1365-2745.13136.
about the benefits of severe fire. In The Ecological Importance of Mixed-Severity Fires: Lamont, B. B. (1995). Testing the effect of ecosystem composition/structure on its
Nature’s Phoenix (eds D. A. DellaSala and C. T. Hanson), pp. 55–88. New York, functioning. Oikos 74, 283–295.
Elsevier. Lamont, B. B. & Downes, K. S. (2011). Fire-stimulated flowering among resprouters
* Hutto, R. L., Conway, C. J., Saab, V. A. & Walters, J. R. (2008). What constitutes and geophytes in Australia and South Africa. Plant Ecology 212, 2111–2125.
a natural fire regime? Insight from the ecology and distribution of coniferous forest Lamont, B. B. & Enright, N. J. (2000). Adaptive advantages of aerial seed banks.
birds in North America. Fire Ecology 4, 115–132. Plant Species Biology 15, 157–166.
Hutto, R. L., Keane, R. E., Sherriff, R. L., Rota, C. T., Eby, L. A. & Saab, V. A. Lamont, B. B. & He, T. (2012). Fire-adapted Gondwanan Angiosperm floras arose
(2016). Toward a more ecologically informed view of severe forest fires. Ecosphere 7, in the Cretaceous. BMC Evolutionary Biology 12, 223.
e01255. Lamont, B. B. & He, T. (2017). Fire-proneness as a prerequisite for the evolution of
Ingalsbee, T. (2017). Whither the paradigm shift? Large wildland fires and the wildfire fire-adapted traits. Trends in Plant Science 22, 278–288.
paradox offer opportunities for a new paradigm of ecological fire management. Lamont, B. B., He, T. & Downes, K. S. (2013). Adaptive responses to directional trait
International Journal of Wildland Fire 26, 557–561. selection in the Miocene enabled Cape proteas to colonize the savanna grasslands.
Jabaily, R. S., Shepherd, K. A., Gardner, A. G., Gustafsson, M. H. G., Evolutionary Ecology 27, 1099–1115.
Howarth, D. G. & Motley, T. J. (2014). Historical biogeography of the Lamont, B. B., He, T. & Yan, Z. (2019a). Evolutionary history of fire-stimulated
predominantly Australian plant family Goodeniaceae. Journal of Biogeography 41, resprouting, flowering, and seed release and germination. Biological Reviews 94,
2057–2067. 903–928.
Jackson, B. K. & Sullivan, S. M. P. (2015). Responses of riparian tetragnathid Lamont, B. B., He, T. & Yan, Z. (2019b). Fire as a pre-emptive evolutionary trigger
spiders to wildfire in forested ecosystems of the California Mediterranean climate among seed plants. Perspectives in Plant Ecology, Evolution and Systematics 36, 13–23.
region, USA. Freshwater Science 34, 1542–1557. Lamont, B. B., Klinkhamer, P. G. & Witkowski, E. T. F. (1993). Population
Jacquet, K. & Prodon, R. (2009). Measuring the postfire resilience of a fragmentation may reduce fertility to zero in Banksia goodii – a demonstration of the
bird–vegetation system: a 28-year study in a Mediterranean oak woodland. Oecologia Allee effect. Oecologia 94, 446–450.
161, 801–811. Lamont, B. B., Ralph, C. S. & Christensen, P. E. S. (1985). Mycophagous
Janz, N., Nylin, S. & Wahlberg, N. (2006). Diversity begets diversity: host marsupials as dispersal agents for ectomycorrhizal fungi on Eucalyptus calophylla and
expansions and the diversification of plant-feeding insects. BMC Evolutionary Biology Gastrolobium bilobum. New Phytologist 101, 651–656.
6, 4. Lamont, B. B. & Wiens, D. (2003). Are seed set and speciation always low among
species that resprout after fire, and why? Evolutionary Ecology 17, 277–292.
Johnson, L. A. S. & Briggs, B. G. (1975). On the Proteaceae—the evolution and
Lancaster, L. T. & Kay, K. M. (2013). Origin and diversification of the California
classification of a southern family. Botanic Journal of the Linnean Society 70, 83–182.
Flora: re-examining classic hypotheses with molecular phylogenies. Evolution 67,
Kane, V. R., Lutz, J. A., Cansler, C. A., Povakc, N. A., Churchill, D. J., Smith,
1041–1054.
D. F., Kanea, J. T. & North, M. P. (2015). Water balance and topography predict
Langlands, P. R., Brennan, K. E. C. & Ward, B. (2012). Is the reassembly of an
fire and forest structure patterns. Forest Ecology and Management 338, 1–13.
arid spider assemblage following fire deterministic? Austral Ecology 37, 429–439.
Karp, A. T., Behrensmeyer, A. K. & Freeman, K. H. (2018). Grassland fire ecology
Law, B. S. & Dickman, C. R. (1998). The use of habitat mosaics by terrestrial
has roots in the late Miocene. Proceedings of the National Academy of Sciences of the United
vertebrate fauna: implications for conservation and management. Biodiversity and
States of America 115, 12130–12135.
Conservation 7, 323–333.
Kazanis, D. & Arianoutsou, M. (2004). Long-term post-fire vegetation dynamics in
Lawes, M. J., Fisher, D. O., Johnson, C. N., Blomberg, S. P., Frank, A. S.,
Pinus halepensis forests of Central Greece: A functional group approach. Plant Ecology
Fritz, S. A., McCallum, H., VanDerWal, J., Abbott, B. N., Legge, S.,
171, 101–121.
Letnic, M., Thomas, C. R., Thurgate, N., Fisher, N., et al. (2015). Correlates
Keeley, J. E. (2009). Fire intensity, fire severity and burn severity: a brief review and of recent declines of rodents in northern and southern Australia: habitat structure is
suggested usage. International Journal of Wildland Fire 18, 116–126. critical. PLoS One 10, e0130626.
Keeley, J. E., Bond, W. J., Bradstock, R. A., Pausas, J. G. & Rundel, P. W. Lazarina, M., Devalez, J., Neokosmidis, L., Sgardelis, S. P., Kallimanis,
(2012). Fire in Mediterranean Ecosystems: Ecology, Evolution and Management. Cambridge A. S., Tscheulin, T., Tsalkatis, P., Kourtidou, M., Mizerakis, V., Nakas,
University Press, Cambridge. G., Palaiologou, P., Kalabokidis, K., Vujic, A. & Petanidou, T. (2019).
Keeley, J. E. & Fotheringham, C. J. (2003). Species–area relationships in Moderate fire severity is best for the diversity of most of the pollinator guilds in
Mediterranean-climate plant communities. Journal of Biogeography 30, 1629–1657. Mediterranean pine forests. Ecology 100, e02615.
Keeley, J. E. & Pausas, J. G. (2018). Evolution of ’smoke’ induced seed germination Lazarina, M., Sgardelis, S. P., Tscheulin, T., Devalez, J., Mizerakis, V.,
in pyroendemic plants. South African Journal of Botany 115, 251–255. Kallimanis, A. S., Papakonstantinou, S., Kyriazis, T. & Petanidou, T.
Keeley, J. E., Pausas, J. G., Rundel, P. W., Bond, W. J. & Bradstock, R. A. (2017). The effect of fire history in shaping diversity patterns of flower-visiting insects
(2011). Fire as an evolutionary pressure shaping plant traits. Trends in Plant Science 16, in post-fire Mediterranean pine forests. Biodiversity and Conservation 26, 115–131.
406–411. Leys, B., Finsinger, W. & Carcaillet, C. (2014). Historical range of fire frequency
* Keeley, S. C., Keeley, J. E., Huthinson, S. M. & Johnson, A. W. (1981). Postfire
is not the Achilles’ heel of the Corsican black pine ecosystem. Journal of Ecology 102,
succession of the herbaceous flora in Southern California chaparral. Ecology 62, 381–395.
1608–1621. Linder, H. P. (2005). Evolutionary history of Poales. Annual Review of Ecology and
Keith, D. A. & Tozer, M. G. (2012). Vegetation dynamics in coastal heathlands of Systematics 36, 107–124.
the Sydney Basin. Proceedings of the Linnean Society of New South Wales 134, B181–B197. Longhurst, W. (1978). Responses of bird and mammal populations to fire in
Kelly, L. T. & Brotons, L. (2017). Using fire to promote biodiversity. Science 355, chaparral. California Agriculture 32, 9–12.
1264–1265. Lorenz, T. J., Vierling, K. T., Kozma, J. M. & Millard, J. E. (2015). Space use
Kelly, L. T., Nimmo, D. G., Spence-Bailey, L. M., Taylor, R. S., Watson, S. J., by white- headed woodpeckers and selection for recent forest disturbances. Journal
Clarke, M. F. & Bennett, A. F. (2012). Managing fire mosaics for small mammal of Wildlife Management 79, 1286–1297.
conservation: a landscape perspective. Journal of Applied Ecology 49, 412–421. Magadzire, N., De Klerk, H. M., Esler, K. J. & Slingsby, J. A. (2019). Fire and
Kiss, L. & Magnin, F. (2006). High resilience of Mediterranean land snail communities life history affect the distribution of plant species in a biodiversity hotspot. Diversity
to wildfires. Biodiversity and Conservation 15, 2925–2944. and Distributions 25, 1015–1023. https://doi.org/10.1111/ddi.12921.
Kitchen, S. G. (2012). Historical fire regime and forest variability on two eastern Magallón, S. & Sanderson, M. J. (2001). Absolute diversification rates in
Great Basin fire-sheds (USA). Forest Ecology and Management 285, 53–66. angiosperm clades. Evolution 55, 1762–1780.

Maikano, G. N., Cohn, J. & Di Stefano, J. (2018). Are germination cues for Orgeos, J. & Anderson, A. N. (2001). Fire and biodiversity: responses of grass-layer
soil-stored seed banks different in structurally different fire-prone communities. beetles to experimental fire regimes in an Australian tropical savanna. Journal of
Austral Ecology 43, 89–101. Applied Ecology 38, 49–62.
Malison, R. L. & Baxter, C. V. (2010). Effects of wildfire of varying severity on Palmquist, K. A., Peet, R. K. & Weakley, A. S. (2014). Changes in plant species
benthic stream insect assemblages and emergence. Freshwater Science 29, 1324–1338. richness following reduced fire frequency and drought in one of the most species-rich
Maravalhas, J. & Vasconcelos, H. L. (2014). Revisiting the savannas in North America. Journal of Vegetation Science 25, 1426–1437.
pyrodiversity–biodiversity hypothesis: long-term fire regimes and the Parkins, K., York, A. & Di Stefano, J. (2018). Edge effects in fire-prone
structure of ant communities in a Neotropical savanna hotspot. Journal of Applied landscapes: ecological importance and implications for fauna. Ecology and Evolution 8,
Ecology 51, 1661–1668. 5937–5948.
Martin, R. E. & Sapsis, D. B. (1992). Fires as agents of biodiversity: pyrodiversity Parr, C. L. & Andersen, A. N. (2006). Patch mosaic burning for biodiversity
promotes biodiversity. In Proceedings of the Symposium on Biodiversity in Northwestern conservation: a critique of the pyrodiversity paradigm. Conservation Biology 20,
California. Wildland Resources Centre, University of California, Berkeley, California. 1610–1619.
Matthews, J. K., Stawski, C., Körtner, G., Parker, C. A. & Geiser, F. (2017). Parr, C. L. & Brockett, B. H. (1999). Patch-mosaic burning: a new paradigm for
Torpor and basking after a severe wildfire: mammalian survival strategies in a savanna fire management in protected areas? Koedoe 42, 117–130.
scorched landscape. Journal of Comparative Physiology B 187, 385–393. Pastro, L. A., Dickman, C. R. & Letnic, M. (2014). Fire type and hemisphere
McMullan-Fisher, S. J. M., May, T., Robinson, R., Bell, T., Lebel, T., determine the effects of fire on the alpha and beta diversity of vertebrates: a global
Catcheside, P. & York, A. (2011). Fungi and fire in Australian ecosystems: meta-analysis. Global Ecology and Biogeography 23, 1146–1156.
a review of current knowledge, management implications and future directions. Pausas, J. G. (2015). Bark thickness and fire regime. Functional Ecology 29, 315–327.
Australian Journal of Botany 59, 70–90. Pausas, J. G., Alessio, G. A., Moreira, B. & Corcobado, G. (2012). Fires enhance
Merwin, L., He, T., Lamont, B. B., Enright, N. J. & Krauss, S. L. (2012). flammability in Ulex parviflorus. New Phytologist 193, 18–23.
Low rate of between-population seed dispersal restricts genetic connectivity and Pausas, J. G., Belliure, J., Mínguez, E. & Montagud, S. (2018a). Fire benefits
metapopulation dynamics in a clonal shrub. PLoS One 7, e50974. flower beetles in a mediterranean ecosystem. PLoS One 13, e0198951.
Milberg, P. & Lamont, B. B. (1995). Fire enhances weed invasion of roadside Pausas, J. G. & Bond, W. J. (2019). Humboldt and the reinvention of nature. Journal
vegetation in southwestern Australia. Biological Conservation 2, 45–49. of Ecology 107, 1031–1037.
Miller, B. P., Walshe, T., Enright, N. J. & Lamont, B. B. (2007). Error in the Pausas, J. G. & Keeley, J. E. (2009). A burning story: the role of fire in the history of
inference of fire history from grasstrees. Austral Ecology 32, 908–916. life. Bioscience 59, 593–601.
* Milne, L. A. (1988). Palynology of a late Eocene lignitic sequence from the western Pausas, J. G. & Keeley, J. E. (2014a). Abrupt climate-independent fire regime
margin of the Eucla Basin, Western Australia. Memoirs of the Association of Australasian changes. Ecosystems 17, 1109–1120.
Palaeontologists 5, 285–310. Pausas, J. G. & Keeley, J. E. (2014b). Evolutionary ecology of resprouting and seeding
Monk, D., Pate, J. S. & Loneragan, W. A. (1981). Biology of Acacia pulchella R.br. in fire-prone ecosystems. New Phytologist 204, 55–65.
with special reference to symbiotic nitrogen fixation. Australian Journal of Botany 27, Pausas, J. G. & Keeley, J. E. (2019). Wildfires as an ecosystem service. Frontiers in
579–592. Ecology and the Environment 17, 289–295.
Moretti, M., Duelli, P. & Obrist, M. K. (2006). Biodiversity and resilience of Pausas, J. G. & Lamont, B. B. (2018). Ecology and biogeography in 3D: the case of
arthropod communities after fire disturbance in temperate forests. Oecologia 149, the Australian Proteaceae. Journal of Biogeography 45, 1469–1477.
312–317. Pausas, J. G., Lamont, B. B., Paula, S., Appezzato-da-Glória, B. & Fidelis, A.
Moritz, M. A., Batllori, E., Bradstock, R. A., Gill, A. M., Handmer, J., (2018b). Unearthing belowground bud banks in fire-prone ecosystems. New Phytologist
Hessburg, P. F., Leonard, J., McCaffrey, S., Odion, D. C., Schoennagel, 217, 1435–1448.
T. & Syphard, A. D. (2014). Learning to coexist with wildfire. Nature 515, Pausas, J. G., Ouadah, N., Ferran, A., Gimeno, T. & Vallejo, R. (2003). Fire
58–66. severity and seedling establishment in Pinus halepensis woodlands, eastern Iberian
Mutch, R. W. (1970). Wildland fires and ecosystems - A hypothesis. Ecology 51, Peninsula. Plant Ecology 169, 205–213.
1046–1051. Pausas, J. G. & Parr, C. L. (2018). Towards an understanding of the evolutionary
Myburg, A. A., Grattapaglia, D., Tuskan, G. A., Hellsten, U., Hayes, R. D., role of fire in animals. Evolutionary Ecology 32, 1–13.
Grimwood, J., Jenkins, J., Lindquist, E., Tice, H., Bauer, D. & Goodstein, Pausas, J. G. & Ribeiro, E. (2013). The global fire-productivity relationship. Global
D. M. (2014). The genome of Eucalyptus grandis. Nature 510, 356–362. Ecology and Biogeography 22, 728–736.
Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A. B. Pausas, J. G. & Ribeiro, E. (2017). Fire and plant diversity at the global scale. Global
& Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature 403, Ecology and Biogeography 26, 889–897.
853–858. Pécrix, Y., Rallo, G., Folzer, H., Cigna, M., Gudin, S. & Le Bris, M. (2011).
Nimmo, D. G., Avitabile, S., Banks, S. C., Bird, B. R., Callister, K., Clarke, Polyploidization mechanisms: temperature environment can induce diploid gamete
M. F., Dickman, C. R., Doherty, T. S., Driscoll, D. A., Greenville, A. C., formation in Rosa sp. Journal of Experimental Botany 62, 3587–3597.
Haslem, A., Kelly, L. T., Kenny, S. A., Lahoz-Monfort, J. J., Lee, C., Perry, D. A., Hessburg, P. F., Skinner, C. N., Spies, T. A., Stephens, S. L.,
et al. (2019). Animal movements in fire-prone landscapes. Biological Reviews 94, Taylor, A. H., Franklin, J. F., McComb, B. & Riegel, G. (2011). The ecology of
981–998. mixed severity fire regimes in Washington, Oregon, and northern California. Forest
Nimmo, D. G., Kelly, L. T., Spence-Bailey, L. M., Watson, S. J., Taylor, R. S., Ecology and Management 262, 703–717.
Clarke, M. F. & Bennett, A. F. (2013). Fire mosaics and reptile conservation in Peterson, D. W. & Reich, P. B. (2008). Fire frequency and tree canopy structure
a fire-prone region. Conservation Biology 27, 345–353. influence plant species diversity in a forest- grassland ecotone. Plant Ecology 194,
Nisbet, C. & LaGoy, P. (1992). Toxic equivalency factors (TEFs) for polycyclic 5–16.
aromatic hydrocarbons (PAHs). Regulatory Toxicology and Pharmacology 16, 290–300. Pettersson, B. (1961). Mutagenic effect of radiant heat shocks on phanerogamous
Noble, I. R. & Slatyer, R. O. (1980). The use of vital attributes to predict successional plants. Nature 191, 1167–1169.
changes in plant communities subject to recurrent disturbances. Vegetatio 43, Phoon, S-N. (2015). Systematics and biogeography of Elaeocarpus (Elaeocarpaceae). PhD
5–21. Thesis: James Cook University, Townsville.
Nolte, C., Kotto-Same, J., Moukam, A., Thenkabail, P. S., Weise, S. F., Pilz, D., Weber, N. S., Carter, M. C., Parks, C. G. & Molina, R.
Woomer, P. L. & Zapfack, L. (2001). Land-Use Characterization and Estimation of (2004). Productivity and diversity of morel mushrooms in healthy, burned, and
Carbon Stocks in the Alternatives to Slash-and-Burn Benchmark Area in Cameroon. Resource insect-damaged forests of northeastern Oregon. Forest Ecology and Management 198,
and Crop Management Research Monograph 28. IITA, Ibadan. 367–386.
Noss, R. F. (1990). Indicators for monitoring biodiversity: a hierarchical approach. Ponisio, L. C., Wilkin, K., M’gonigle, L. K., Kulhanek, K., Cook, L., Thorp,
Conservation Biology 4, 355–364. R., Griswold, T. G. & Kremen, C. (2016). Pyrodiversity begets plant–pollinator
Nowack, J., Stawski, C., Körtner, G. & Geiser, F. (2018). Physiological and community diversity. Global Change Biology 22, 1794–1808.
behavioral responses of an arboreal mammal to smoke and charcoal-ash substrate. Radford, I. J., Gibson, L. A., Corey, B., Carnes, K. & Fairman, R. (2015).
Physiology and Behavior 184, 116–121. Influence of fire mosaics, habitat characteristics and cattle disturbance on mammals
O’Brien, E. (1998). Water–energy dynamics, climate, and prediction of woody in fire-prone savanna landscapes of the northern Kimberley. PLoS One 10,
plant species richness: an interim general model. Journal of Biogeography 25, e0130721.
379–398. Ramanna, M. S. & Jacobsen, E. (2003). Relevance of sexual polyploidization for
Orecchio, S., Amorello, D., Barreca, S. & Valenti, A. (2016). Wood crop improvement: a review. Euphytica 133, 3–18.
pellets for home heating can be considered environmentally friendly fuels? Raphael, M. G., Morrison, M. L. & Yoder-Williams, M. P. (1987). Breeding bird
Polycyclic aromatic hydrocarbons (PAHs). In their ashes. Microchemical Journal 124, populations during twenty-five years of postfire succession in the Sierra Nevada. The
267–271. Condor 89, 614–626.

Robinson, R. M., Mellican, A. E. & Smith, R. H. (2008). Epigeous macrofungal Soria-Hernanz, D. F., Fiz-Palacios, O., Braverman, J. M. & Hamilton, M. B.
succession in the forest five years following a wild fire in karri (Eucalyptus diversicolor) (2008). Reconsidering the generation time hypothesis based on nuclear ribosomal
regrowth forest in Western Australia. Austral Ecology 33, 807–820. ITS sequence comparisons in annual and perennial angiosperms. BMC Evolutionary
* Roche, S., Dixon, K. W. & Pate, J. S. (1997). Seed ageing and smoke: partner cues Biology 8, 344.
in the amelioration of seed dormancy in selected Australian native species. Australian Stawski, C., Körtner, G., Nowack, J. & Geiser, F. (2015). The importance
Journal of Botany 45, 783–815. of mammalian torpor for survival in a post-fire landscape. Biology Letters 11,
Rogers, B. M., Soja, A. J., Goulden, M. L. & Randerson, J. T. (2015). Differences 20150134.
in boreal fires and climate feedbacks. Nature Geoscience 8, 228–234. Stawski, C., Nowack, J., Körtner, G. & Geiser, F. (2017). A new cue for
Rota, C. T., Rumble, M. A., Millspaugh, J. J., Lehman, C. P. & Kesler, D. C. torpor induction: charcoal, ash and smoke. Journal of Experimental Biology 220,
(2014). Space-use and habitat associations of black-backed woodpeckers (Picoides 220–226.
arcticus) occupying recently disturbed forests in the Black Hills, South Dakota. Forest Stein, A., Gerstner, K. & Kreft, H. (2014). Environmental heterogeneity as a
Ecology and Management 313, 161–168. universal driver of species richness across taxa, biomes and spatial scales. Ecology
Rundel, P. W. (1981). Structural and chemical components of flammability. In Letters 17, 866–880.
The Role of Past and Present Frequency and Intensity of Fire on Ecosystem Development and Stevens, R. D. & Tello, J. S. (2011). Diversity begets diversity: relative roles of
Management. USDA Forest Service General Technical Report WO-26. (ed. H. A. structural and resource heterogeneity in determining rodent community structure.
Mooney), pp. 183–207. USDA Forest Service, Washington D.C. Journal of Mammalogy 92, 387–395.
Rundel, P. W., Arroyo, M. T. K., Cowling, R. M., Keeley, J. E., Lamont, Stock, W. D. & Allsopp, N. (1992). Functional perspective of ecosystems. In The
B. B., Pausas, J. G. & Vargas, P. (2018). Fire and plant diversification in Ecology of Fynbos: Nutrients, Fire and Diversity (ed. R. M. Cowling), pp. 241–259.
Mediterranean-climate regions. Frontiers in Plant Science 9, 851. Oxford University Press, Cape Town.
Rundel, P. W., Arroyo, M. T. K., Cowling, R. M., Keeley, J. E., Taille, P.J., Burnett, R.D., Roberts, L.J. & Campos, B.R. (2018). Interacting and
Lamont, B. B. & Vargas, P. (2016). Mediterranean biomes: evolution of their non-linear avian responses to mixed-severity wildfire and time since fire. Ecosphere 9,
vegetation, floras and climate. Annual Review of Ecology, Evolution and Systematics 47, e02291.https://doi.org/10.1002/ecs2.2291.
383–407. Talluto, M. V. & Benkman, C. W. (2013). Landscape-scale eco-evolutionary
Saint-Germain, M., Drapeau, P. & Hebert, C. (2004). Landscape-scale dynamics: selection by seed predators and fire determine a major reproductive
habitat selection patterns of Monochamus scutellatus(Coleoptera: Cerambycidae). strategy. Ecology 94, 1307–1316.
In a recently burned black spruce forest. Environmental Entomology 33, Taylor, R. S., Watson, S. J., Nimmo, D. G., Kelly, L. T., Bennett, A. F. &
1703–1710. Clarke, M. F. (2012). Landscape-scale effects of fire on bird assemblages: does
Santos, X., Badiane, A. & Matos, C. (2016). Contrasts in short- and long-term pyrodiversity beget biodiversity? Diversity and Distributions 18, 519–529.
responses of Mediterranean reptile species to fire and habitat structure. Oecologia Tingley, M. W., Ruiz-Gutiérrez, V., Wilkerson, R. L., Howell, C. A. &
180, 205–216. Siegel, R. B. (2016). Pyrodiversity promotes avian diversity over the decade
* Sauquet, H., Weston, P. H., Anderson, C. L., Barker, N. P., Cantrill, D. J.,
following forest fire. Proceedings of the Royal Society B 283, 20161703.
Mast, A. R. & Savolainen, V. (2009). Contrasted patterns of hyperdiversification Van der Niet, T., Peakall, R. & Johnson, S. D. (2014). Pollinator-driven ecological
in Mediterranean hotspots. Proceedings of the National Academy of Sciences of the United
speciation in plants: new evidence and future perspectives. Annals of Botany 113,
States of America 106, 221–225.
199–212.
Schaffhauser, A., Curt, T. & Tatoni, T. (2011). Fire-vegetation interplay in
Van Nuland, M. E., Haag, E. N., Bryant, J. A., Read, Q. D., Klein, R. N.,
a mosaic structure of Quercus suber woodlands and Mediterranean maquis under
Douglas, M. J., Greenwell, T. D., Busby, M. W., Collins, J., Leroy, J. T.,
recurrent fires. Forest Ecology and Management 262, 730–738.
Schuchmann, G., Schweitzer, J. A. & Bailey, J. K. (2013). Fire promotes
Schmidl, C., Bauer, H., Dattler, A., Hitzenberger, R., Weissenboeck, G.,
pollinator visitation: implications for ameliorating declines of pollination services.
Marr, I. L. & Puxbaum, H. (2008). Chemical characterisation of particle emissions
PLoS One 8, e79853.
from burning leaves. Atmospheric Environment 42, 9070–9079.
Vandvik, V., Töpper, J. P., Cook, Z., Daws, M. I., Heegaard, E., Måren, I. E.
Schnitzler, J., Barraclough, T. G., Boatwright, J. S., Goldblatt, P.,
& Velle, L. G. (2014). Management-driven evolution in a domesticated ecosystem.
Manning, J. C., Powell, M. P., Rebelo, T. & Savolainen, V. (2011). Causes
Biology Letters 10, 20131082.
of plant diversification in the Cape biodiversity hotspot of South Africa. Systematic
Vanneste, K., Baele, G., Maere, S. & Van de Peer, Y. (2014). Analysis of 41 plant
Biology 60, 343–357.
genomes supports a wave of successful genome duplications in association with the
Schoennagel, T., Balch, J. K., Brenkert-Smith, H., Dennison, P. E., Harvey,
Cretaceous-Paleogene boundary. Genome Research 24, 1334–1347.
B. J., Krawchuk, M. A., Mietkiewicz, N., Morgan, P., Moritz, M. A., Rasker,
R. & Turner, M. G. (2017). Adapt to more wildfire in western North American Verdú, M. & Pausas, J. G. (2007). Fire drives phylogenetic clustering in
forests as climate changes. Proceedings of the National Academy of Sciences of the United States Mediterranean Basin woody plant communities. Journal of Ecology 95, 1316–1323.
of America 114, 4582–4590. Vu, B., Alves, C. A., Gonçalves, C., Pio, C., Gonçalves, F. & Pereira, R.
Schwilk, D. W., Keeley, J. E. & Bond, W. J. (1997). The intermediate disturbance (2012). Mutagenicity assessment of aerosols in emissions from wood combustion in
hypothesis does not explain fire and diversity pattern in fynbos. Plant Ecology 132, Portugal. Environmental Pollution 166, 172–181.
77–84. Watson, J. M., Platzer, A., Kazda, A., Akimcheva, S., Valuchova, S.,
Scott, A. C. (2018). Burning Planet: The Story of Fire through Time. Oxford University Nizhynska, V., Nordborg, M. & Riha, K. (2016). Germline replications
Press, Oxford. and somatic mutation accumulation are independent of vegetative life span in
Sharma, A. P. & Tripathi, B. D. (2009). Assessment of atmospheric PAHs profile Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 113,
through Calotropis gigantea R.Br. leaves in the vicinity of an Indian coal-fired power 12226–12231.
plant. Environmental Monitoring and Assessment 149, 477–482. Watson, P. J., Bradstock, R. A. & Morris, E. C. (2009). Fire frequency influences
Shea, S. (1994). Fighting fire with fire. Nature 372, 399. composition and structure of the shrub layer in an Australian subcoastal temperate
Simon, M. F., Grether, R., de Queiroz, L. P., Skema, C., Pennington, R. T. & grassy woodland. Austral Ecology 34, 218–232.
Hughes, C. E. (2009). Recent assembly of the Cerrado, a neotropical plant diversity Watson, P. J. & Wardell-Johnson, G. (2004). Fire frequency and time-since-fire
hotspot, by in situ evolution of adaptations to fire. Proceedings of National Academy of effects on the open-forest and woodland flora of Girraween National Park, south-east
Science of the United States of America 106, 20359–20364. Queensland, Australia. Austral Ecology 29, 225–236.
Simoneit, B. R. T., Rogge, W. F., Lange, Q. & Jaffé, R. (2000). Molecular Wells, P. V. (1969). The relation between mode of reproduction and extent
characterization of smoke from campfire burning of pine wood (Pinus elliottii). of specialization in woody genera of the California chaparral. Evolution 23,
Chemosphere 2, 107–122. 264–267.
Sitters, H., Christie, F. J., Di Stefano, J., Swan, M., Penman, T., Collins, White, R. D., Krumperman, P. K., Cheeke, P. R. & Buhler, D. R. (1983).
P. C. & York, A. (2014). Avian responses to the diversity and configuration of fire age An evaluation of acetone extracts from six plants in the Ames mutagenicity test.
classes and vegetation types across a rainfall gradient. Forest Ecology and Management Toxicology Letters 15, 25–31.
* Wikars, L.O. (1997). Effect of forest fire and the ecology of fire adapted species. PhD Thesis:
318, 13–20.
Sitters, H., Di Stefano, J., Christie, F., Swan, M. & York, A. (2016). Bird University of Uppsala, Uppsala.
functional diversity decreases with time since disturbance: does patchy prescribed Wittkuhn, R. S., Lamont, B. B. & He, T. (2017). Combustion temperatures
fire enhance ecosystem function? Ecological Applications 26, 115–127. and nutrient transfers when grasstrees burn. Forest Ecology and Management 399,
Smucker, K. M., Hutto, R. L. & Steele, B. M. (2005). Changes in bird abundance 179–187.
after wildfire: importance of fire severity and time since fire. Ecological Applications 15, Wittkuhn, R. S., McCaw, L., Wills, A. J., Robinson, R., Anderson, A.,
1535–1549. Heurck, P. V., Farr, J., Liddelow, G. & Cranfield, R. (2011). Variation in
Sojinu, O. S., Sonibare, O. O. & Zeng, E. Y. (2011). Concentrations of polycyclic fire interval sequences has minimal effects on species richness and composition in
aromatic hydrocarbons in soils of a mangrove forest affected by forest fire. Toxicological fire-prone landscapes of southwest Western Australia. Forest Ecology and Management
& Environmental Chemistry 93, 450–461. 261, 965–978.

Yeboah, D., Chen, Y. H. & Kingston, S. (2016). Tree species richness decreases on the left from non-fire-prone lineages on the right. Number
while species evenness increases with disturbance frequency in a natural boreal
forest landscape. Ecology and Evolution 6, 842–850. of species was estimated as the species that occur exclusively
Yue, J.-X., Li, J., Wang, D., Araki, H., Tian, D. & Yang, S. (2010). Genome-wide in the specified habitat. Whenever possible, all species in
investigation reveals high evolutionary rates in annual model plants. BMC Plant the lineage were evaluated, and the number of species is
Biology 10, 242.
Zach, R. & Sheppard, S. C. (1991). Food-chain and dose model, CALDOS, for given; otherwise, the number of species in the published
assessing Canada’s Nuclear Fuel Waste Management concept. Health Physics 60, phylogeny is given, assuming that a representative sample
643–656. was in the original study. When there are multiple migrations
Zedler, P. H. (1995). Fire frequency in southern California shrublands: biological
effects and management options. In Brushfires in California: Ecology and Management between fire-prone and non-fire prone habitats within the
(eds J. E. Keeley and T. Scott). International Association of Wildland Fire, taxon group, the number of species and divergent time for
Fairfield.
each migration are given. My, million years.
Table S2. List of the currently known non-plant pyrophilous
species (fungi, insects, amphibians and birds). *, documented
attraction to ongoing fire (open fire, hot ash or smoke);?
X. SUPPORTING INFORMATION indicates the absence of confirmative literature. Region:
NA: North America; SA: South America; EA: Eurasia; AU:
Additional supporting information may be found online in Australia; AF: Africa. Data compiled from Wikars (1997),
the Supporting Information section at the end of the article. Hjältén et al. (2007), Hutto et al. (2008); Hutto, Bond &
Table S1. Species richness of sister lineages in fire-prone DellaSala (2015), McMullan-Fisher et al. (2011) and Klocke
and non-fire-prone habitats. – separates fire-prone lineages et al. (2011).
(Received 3 April 2019; revised 19 June 2019; accepted 21 June 2019 )

He&Lamont 2019 Fire As A Key

Uploaded by

Copyright:

Available Formats

He&Lamont 2019 Fire As A Key

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

He&Lamont 2019 Fire As A Key

Uploaded by

Copyright:

Available Formats

Biol. Rev. (2019), pp. 000–000.

Fire as a key driver of Earth’s biodiversity

* Author for correspondence (Tel.: +61 8 9457 7164; E-mail: [email protected])

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

(3) Mutagenic effects of heat and chemicals released by combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

I. INTRODUCTION Lamont & Downes, 2011; He et al., 2012; He et al., 2016),

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

fuel distribution, ignition source

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Fire history of a given location in a reference area

Fire regime (of reference area)

Fire event F1 F2 F3 F4 Fire patch

Mean fire interval:X/3

Y Fire frequency: 4/Y

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Fire components Biodiversity Time span Relationship

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

areas burnt frequently to maintain grasslands (savannas) have

Bradstock et al. (1998)

Watson, Bradstock &

Schaffhauser, Curt &

Bowman et al. (2016)

Bradstock, Tozer &

Bassett et al. (2017)

Foster et al. (2017)

fire regimes and contrasting vegetation although under

species composition may invert the usual relationship by

driving the fire regime instead.

initiates population and community recovery. In fire-prone

ecosystems, most plant species present before the fire either

recover vegetatively and/or germinate from stored seeds

recruitment (Merwin et al., 2012). Short-lived or geophytic

richness, fire trait

species that had long completed their aboveground life cycle

become visible as their seeds or belowground bud banks

break dormancy and their foliage appears above the soil

Enright et al., 2007) models of community succession. The

by a gradual increase in vegetative growth of survivors,

a period of escalating fecundity, then declining foliage

and seed production, and eventual death or return to

in the increasingly scarce bare patches. Other herbaceous

few years (Hinman & Brewer, 2007; Lamont & Downes,

2011), weakly woody shrubs may survive for about a decade,

clonal species with belowground budbanks may survive for

As particular plant species and growth forms take

turns to dominate the community, and the structure and

availability of habitat types changes progressively, the

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Mechanism Interpretation with examples Scale of diversity References

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society

Biological Reviews (2019) 000–000 © 2019 Cambridge Philosophical Society