Title: Physiological responses to fire that drive tree mortality
Author: Sharon M. Hood
Contact information:
sharon.hood@usda.gov
ORCID 0000-0002-9544-8208
US Department of Agriculture, Forest Service, Rocky Mountain Research
Station, Fire Sciences Laboratory, 5775 Highway 10 W, Missoula, MT,
59808
Running head: Tree physiological fire effects
Abstract
This article comments on: ‘Short- and long-term effects of fire on
stem hydraulics in Pinus ponderosa saplings ’ by Partelli-Feltrinet al. (2020 ), https://doi.org/10.1111/pce.13881.
Keywords: hydraulics, hormesis, fire effects, drought, disturbance
interactions
A summer storm passes, flashing lightning but little rain and a fire
starts to grow. Flames lick the bases of trees, charring stems and
strong winds fan the fire at times, causing the flames to scorch
foliage, as it burns through a forest. The fate of these burned trees
after the flames are doused is one of intense interest. Will the forest
become a carbon source? Will there be suitable wildlife habitat? Will
enough trees survive to perpetuate and regenerate the forest? Are the
trees now more vulnerable to bark beetle attacks or drought? To answer
these questions, ecologists are reliant on accurate predictive models of
fire-caused tree mortality. Yet, we still know remarkably little about
the actual physiological impacts of fire on trees, limiting our ability
to build mechanistic mortality models (Bär, Michaletz, & Mayr, 2019;
Hood, Varner, van Mantgem, & Cansler, 2018).
Depending on the intensity and duration of heat that affects crown,
stem, and root tissues, some trees are killed immediately, others may
die over the next several years, while others fully recover and survive
future fires (Hood et al., 2018). Death of foliage and branch meristems
in the crown during a high intensity fire is the most well-described
cause of tree death or top-kill in species capable of resprouting. There
is far less research on the impact of lower intensity fires, where crown
meristems are not as affected, causing injuries to the stem and
ultimately tree death. Delayed mortality is inherently difficult to
predict, as numerous factors influence physiological responses, such as
plant water stress, that can increase post-fire tree mortality over time
(Partelli-Feltrin, Johnson, et al., 2020; van Mantgem et al., 2013). The
article by Partelli et al. (2020) tested the two leading hypotheses
explaining stem injury from fire that can cause both immediate and
delayed tree mortality, cambium necrosis and xylem dysfunction (Bär et
al., 2019). The cambium necrosis hypothesis predicts that when fire
kills stem phloem and cambium tissue it severs the connection between
foliage and roots. Over time, without photosynthates to replenish carbon
reserves, roots die, increasing xylem tension until tree mortality
occurs due to hydraulic failure. The xylem dysfunction hypothesis
predicts that fire-caused heating of xylem tissue causes irreversible
deformation of cell walls, reducing hydraulic conductivity, that
increases xylem tension and stomatal closure, eventually leading to tree
death from a combination of depleted carbon stores and hydraulic
failure.
Partelli-Feltrin, Smith, Adams, Kolden, and Johnson (2020) examined
immediate (1-day) and delayed (21 month) impacts of fire on tree
hydraulics through a series of small-scale, experimental fires using
1-year old Pinus ponderosa saplings. Importantly, the authors
standardized the “dose” of the fire using Fire Radiative Energy (FRE),
a measure of the intensity or radiative energy released from fuel during
the fire, that each tree received to ensure that the treatments were
applied consistently. Previous work by some of the authors has
established the range of FRE dosages that are typically observed during
surface fires in Pinus ponderosa forests (A. M. S. Smith et al.,
2017) and corresponding mortality dose-response curves (Steady et al.,
2019). They examined plant stem hydraulic responses to two fire
intensity treatments: a high dose (1.4 MJm-2) and a
lower, sub-lethal dose (0.7 MJm-2) through two
experiments. Plants were well-watered both before and after the
experiments.
The first experiment examined short-term impacts on plant hydraulics.
One day after receiving a high dose of FRE, no differences were observed
in maximum xylem hydraulic conductivity (k max),
native percentage loss of conductivity (nPLC), or vulnerability to
cavitation between the burned and unburned trees.
The second experiment examined plant hydraulics 21 months post-fire
between unburned saplings and those receiving a low dose FRE. The
original design also included a high dose treatment, but all saplings
receiving this treatment died by the 21-month postfire measurement. As
in the short-term experiment, no differences kmax or
nPLC were found between burned and unburned saplings. Despite this, the
burned saplings were more vulnerable to cavitation than the unburned
saplings. When the stem xylem tissue was examined, no deformation of the
conduits was found in the xylem that had formed before the fire.
However, several of the burned saplings had traumatic xylem tissue
forming post-fire in response to the fire killing cambium and phloem
tissue. In addition, in the areas of these fire scars, resin was
impregnating the pre-fire xylem. The authors attribute the increased
vulnerability to water stress in the long-term to the post-fire wound
tissue that formed and resin soaking and clogging preexisting xylem.
The main finding of this work is the increased vulnerability to water
stress that developed over the 21 months after the fire. Research on
fire scar formation corroborates the finding of resin soaking into
pre-existing xylem around the area killed by fire (K. T. Smith,
Arbellay, Falk, & Sutherland, 2016). Mundo, González, Stoffel,
Ballesteros-Cánovas, and Villalba (2019) found reduced xylem
conductivity near fire scars and increased vulnerability to water
stress, but there was a slow recovery in the years afterwards. Results
are also consistent with experiments using water baths as a surrogate
for fire-caused heating showing increased vulnerability to water stress
(Bär, Nardini, & Mayr, 2018; Lodge, Dickinson, & Kavanagh, 2018;
Michaletz, Johnson, & Tyree, 2012; West, Nel, Bond, & Midgley, 2016).
Field-based studies of burned tree branches showed mixed support of fire
impacting hydraulic safety (Bär et al., 2018; Battipaglia et al., 2016).
The finding that fire did not cause xylem deformation conflicts with
other research (Michaletz et al., 2012; West et al., 2016). These
studies used water baths as a surrogate for fire, suggesting that water
baths are not reliable substitutes for examining physiological effects
of fire on plants. More research is needed to develop standardized
protocols for pyro-ecophysiological studies. Differences in results may
also be due to differences in tree sizes that range from the one-year
old saplings to 40+ cm diameter trees.
Fire is a natural ecological disturbance in many ecosystems around the
world, and tree species have numerous adaptations to survive fire. Yet
climate change is also increasing drought stress and driving changes in
fire regimes that can alter tree susceptibility to fire. The
discrepancies in existing studies on the physiological effects of fire
suggest that responses are species dependent and driven by suites of
“pyrohydraulic traits” (West et al., 2016). Quantification of and
accounting for traits known to affect tree responses to fire and drought
are needed, such as has been documented for Mediterranean species (Paula
et al., 2009). For example, bark thickness is perhaps the single most
important trait protecting cambium from heating during fire (Pellegrini
et al., 2017), but other traits such as meristem size and protection
(Charles‐Dominique, Beckett, Midgley, & Bond, 2015), xylem anatomy
(West et al., 2016), and nonstructural carbohydrate storage pools
(Varner et al., 2009) almost certainly affect a species tolerance of
fire.
During a fire, heat can affect both the stem and crown. The results of
Partelli-Feltrin, Smith, et al. (2020) hint that some meristems were
killed and this may have contributed to mortality, but crown scorch and
bud kill were not assessed. Carbon acquisition and hydraulic function
are linked in ways still not fully understood, as evidenced by a study
of experimental manual defoliation that caused an increased
vulnerability to embolism (Hillabrand, Hacke, & Lieffers, 2019). Fire
may act in a similar way as defoliation to kill foliage, also causing
changes to xylem anatomy. Quantifying impacts to foliage and meristems
and tracking the time-to-death of saplings would have complemented the
work and future experiments should report both injuries to the crown and
stem tissues if possible.
The authors quantified the fire dose each sapling received using FRE, a
measure of the radiative heat released during a fire. FRE has the
advantages of directly relating to the level of fuel consumed, can be
used to estimate the convective heat released, and can be assessed with
remote sensing techniques over large spatial scales (Kremens, Dickinson,
& Bova, 2012; A. M. Smith et al., 2016). A limitation, however, is the
inability to estimate conductive heat. In fires where long-term
smoldering occurs (Varner et al., 2009), this is a fatal limitation of
using FRE to measure physiological responses. Advances in quantifying
total heat flux from all heat transfer processes are needed to link the
energy released during a fire to physiological and ecological effects
(O’Brien et al., 2018).
Increased vulnerability to embolism makes trees more prone to death if a
severe drought occurs within a few years of fire. As Partelli-Feltrin,
Smith, et al. (2020) point out, how long this response persists is
unknown. Studying a range of physiological responses to fire over time
is required to ultimately integrate the responses into improved models
of fire-induced tree mortality (Figure 1). For example, Partelli’s
experiments caused an adverse effect, as shown in Figure 1a, over the
21-month time period, but it is unknown which curve the saplings given
the low dose of fire would have ultimately followed. Would they fully or
partially grow out of the increased xylem vulnerability or would a
drought interact to kill them during this phase of increased
vulnerability? Other studies have documented induced effects of enhanced
physiological activity after fire to resin defenses (Hood, Sala,
Heyerdahl, & Boutin, 2015) and increased stomatal conductance and
photosynthesis (Gričar et al., 2020; Valor et al., 2018; Wallin, Kolb,
Skov, & Wagner, 2003). How physiological responses interact with
species traits and exogenous factors will determine long-term,
fire-induced tree mortality (Figure 1B) and allow a better understanding
of the full range of a species’s biological plasticity and tolerance to
fire (Figure 1C).
The idea of fire having a dose-dependent effect on plant physiological
responses is promising and needs much more research. Integrating the
concept of hormesis, or non-linear responses and adaptative conditioning
to the environment (Agathokleous, Kitao, Harayama, & Calabrese, 2019),
could provide a platform to develop additional hypotheses of plant
responses to fire for future experiments. As droughts become
increasingly frequent and intense, fire-affected trees may be more
vulnerable to water stress, thereby increasing the chance a tree dies in
the years after fire. The complex interactions of climate and fire on
tree physiological responses and mortality underscores the importance of
improving our understanding of how climate change and fire will impact
terrestrial ecosystems. Advances in quantifying and predicting
physiological effects of fire on trees will require a multi-disciplinary
approach of plant ecophysiologists, ecologists, physical scientists, and
modelers.