The large-scale die-back and collapse of forest and woodland communities in recent decades has seen drought-induced tree mortality emerging as a prominent topic in the ecological and physiological literature. Against the spectre of a potential increase in the frequency and intensity of global change-type droughts and associated temperatures, it is argued that better understanding of the mechanisms driving plant mortality provides a basis for predicting how species may respond to future stresses associated with changing climates. Although it is tempting to try and isolate a primary mechanistic driver for tree mortality, the emerging picture is one of the complex interactions between multiple processes and stresses, often including biotic agents. In this issue of Tree Physiology, Gaylord et al. (2015) present an interesting retrospective examination of potential causes of mortality in Pinus–Juniper ecosystems. Their approach is akin to a forest post mortem, where the authors combined established methods in tree-ring analysis, stable isotope chemistry and xylem anatomy to disentangle the proximal (contributing) and ultimate (primary) causes of death among the victims (adult Pinus edulis trees) of an extreme drought ending in 2002. Gaylord et al. (2015) implicate the interactive effects of hydraulic dysfunction, carbon (C) metabolism and pest infestation, a conclusion consistent with a growing body of knowledge emphasizing the role of combined biotic and abiotic stressors in driving forest condition and tree death; see for example, Worrall et al. (2013), Jactel et al. (2012) or Jacquet et al. (2014). Retrospective physiological studies may prove to be important tools in understanding processes of tree death, provided their design and interpretation are not blindsided by confirmatory biases arising from attributing death to factors that are distal to the underlying agent of mortality. The extensive observational and experimental research in the pinyon–juniper woodlands has provided some important insights into how drought impinges on tree survival at a range of scales (McDowell et al. 2013), especially in relation to the relative contributions of hydraulic failure and C starvation. While C starvation has been a front-running hypothesis as a major driver of tree mortality in isohydric plants (McDowell et al. 2008, Mitchell et al. 2013), it could be argued that evidence for exhaustive C starvation remains limited. Several lines of evidence suggest that C depletion may only occur under a limited number of circumstances. First, the majority of studies demonstrating clear reductions in non-structural carbohydrates during water deficit are from studies involving Pinaceae spp. (Guehl et al. 1993, Galiano et al. 2011, Adams et al. 2013, Hartmann et al. 2013, Mitchell et al. 2013, Poyatos et al. 2013, Jacquet et al. 2014). The tendency towards an isohydric-type regulation of gas exchange in Pinaceae species is mediated by their pattern of rising foliar abscisic acid production in response to water deficit (Brodribb and McAdam 2013). This conservative mode of stomatal regulation can promote rapid shifts in plant C balance towards deficit. Secondly, if C depletion is largely dependent on C supply, via reduced stomatal conductance, the C deficit must be protracted enough for C reserves to be consumed. Thirdly, during the period of stomatal closure, for significant C depletion to occur, tree or organ water status must remain above the lethal thresholds associated with hydraulic failure. Hence, in the majority of tree species tending to exhibit anisohydric regulation and in all but the most extreme treatments, such as low CO2 (Hartmann et al. 2013), shading (O’Brien et al. 2014) or stem girdling, the contribution of C relations to tree death is likely to stem from interruptions in C transport and metabolism induced by both abiotic and biotic stressors (Sevanto et al. 2014). Given the inherent inter-dependency of C gain and water transport, the separation between individual stress-inducing processes is somewhat artificial. Recent analyses by Mencuccini et al. (2015) highlight the complexity associated with trying to understand the relative contributions of the tightly coupled mechanisms associated with tree mortality, but provides a useful pathway forward for examining the co-ordination of these underlying mechanisms. The emerging trend in the literature that seeks to provide an empirical basis for this co-ordination (e.g., Mitchell et al. 2014, Sevanto et al. 2014, Gaylord et al. 2015,Duan et al. 2015), will enhance our capacity to better understand this co-ordination. Furthermore, a more process-based integration of the role of pests and pathogens with the underlying physiological processes (see, for example, Oliva et al. 2014) will enhance our understanding of the inherent resistance of plant communities to a range of interacting stresses. Existing analytical approaches can aid in unravelling this complexity and capture the emergent behaviour which potentially makes the problem of predicting impacts at higher scales more tractable (Meir et al. 2015). Clearly greater emphasis needs to be placed on research that addresses both the causes and consequences of drought mortality integrated across various scales. An interesting point of discussion unearthed by the Gaylord et al. (2015) study was the observation that dead trees had smaller pit diameters, and smaller and fewer resin ducts than trees that survived. Apart from challenging a pre-existing paradigm, that trees with larger pit diameters should be more vulnerable to cavitation (Sperry and Tyree 1988), the study also raises the question: is it possible to assess the vulnerability of forest stands to extreme stress events based on an understanding of the intraspecific trait variability? To date, a significant proportion of the global effort on understanding species vulnerability to drought has been focused on understanding interspecific variation in drought tolerance traits; see for example, Maherali et al. (2006), Blackman et al. (2012) and Choat et al. (2012). More recent studies have capitalized on this variation to explore relationships between key drought tolerance traits and climate (Blackman et al. 2014). Surprisingly, much less effort has been focused on understanding variability in drought tolerance within species (Maherali et al. 2009, Gaspar et al. 2013) and the role that intraspecific variability plays in the resistance of these communities to environmental stresses. The Gaylord et al. (2015) study highlighted the regional nature of the drought and widespread exposure to drought and associated biotic agents, but also highlights significant differences in traits of the species that survived compared with those that died. Rosner and Hannrup (2004) noted that defence traits such as resin duct structure are likely to be under strong genetic control. The extent of genetic control on many drought traits remains poorly quantified (Anderegg 2015), possibly due to the significant phenotyping investment required to quantify intraspecific variation in physiological traits. Genotypic and phenotypic variations in drought tolerance across the landscape may prove to be critical in controlling how populations are reconfigured in response to global change-type droughts. Recent observational studies in north-eastern Australian savannas suggest that severe drought tends to alter population dynamics from the ‘inside out’ as opposed to generating more predictable expansion or contraction of a species’ geographical range (Fensham et al. 2015). The ecological consequences of widespread tree mortality remain largely unresolved. While tree mortality is in itself an end for the individual trees, the outcome for the ecosystems and risks to the ecosystem services they provide are less clear. If the end game of the science is to predict how communities will change and adapt to future stresses, then a fruitful future pathway of inquiry may be better understanding the inherent adaptive capacity within the existing communities. Furthermore, the frequency and timing of disturbances may be as important as the inherent capacity to absorb the stress associated with any individual drought event. An insightful study from Australian alpine ash forests (Eucalyptus delagatensis) highlights the critical role that changes in frequency of large-scale disturbances can play (Bowman et al. 2014). Fire plays a major role in alpine ash forests, with infrequent yet intense fires killing mature trees and essentially resetting the ‘successional clock’. However, these systems are resilient to these fire regimes in so far as fire provides the ecological conditions necessary for rapid regeneration. Increased frequency of wildfires over recent years has meant that these regenerating forests have been burnt repeatedly before the ash forests had reached reproductive maturity and produced the seed for the next generation. These stand-replacing fires have the potential to fundamentally alter the structure and function of these forests in important water supply catchments in Victoria, Australia. Similarly, drought is a cyclical phenomenon. Historically, drought-related tree mortality has always sparked interest, as indicated by this quote from an Australian regional newspaper in 1889: ‘1888 was the driest year known since the settlement of the country. A fact in striking contrast is also noted—that the previous year was the wettest on record. So intense was the drought that the native trees on the hills were all in a dying state and over large areas absolutely dead, a state of matters which it was evident from the age of the trees killed could not have been experienced within the last 50 years’ (Bathurst Free Press and Mining Journal 17 September 1889). Placing individual drought events and their impacts within the broader context of historical disturbance regimes provides greater clarity on the potential severity of current and future global change. Retrospective analyses have an important role to play in achieving this. Tree-ring reconstructions of Palmer Drought Severity Indices highlight the historical occurrence of mega droughts, the intensity of which has not been seen in the twentieth century across much of the south western USA (Stahle et al. 2012). Gray et al. (2006) argue that this historical multi-decadal climate variability played a key role in the structuring and functioning of the pinyon–juniper systems. While it can be argued that modern ecosystems face challenges that are unique historically, for example, the highly fragmented nature of modern landscapes, understanding this historical variability in both the climate and related vegetation responses can also provide insights into the resilience of these communities to future climate stresses.
No takes yet. Share an insight, caveat, or question.
O’Grady et al. (2015) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: