Mechanisms of plant competition for nutrients, water and light

Functional Ecology 2013, 27, 833–840

doi: 10.1111/1365-2435.12081

MECHANISMS OF PLANT COMPETITION

Mechanisms of plant competition for nutrients, water and light Joseph M. Craine*,1 and Ray Dybzinski2 1

Division of Biology, Kansas State University, Manhattan, KA 66506, USA; and 2Ecology and Evolutionary Biology, Princeton University, Princeton, NJ 08544, USA

Summary 1. Competition for resources has long been considered a prevalent force in structuring plant communities and natural selection, yet our understanding of the mechanisms that underlie resource competition is still developing. 2. The complexity of resource competition is derived not only from the variability of resource limitation in space and time and among species, but also from the complexity of the resources themselves. Nutrients, water and light each differ in their properties, which generates unique ways that plants compete for these resources. 3. Here, we discuss the roles of supply pre-emption and availability reduction in competition for the three resources when supplied evenly in space and time. Plants compete for nutrients by pre-empting nutrient supplies from coming into contact with neighbours, which requires maximizing root length. Although water is also a soil resource, competition for water is generally considered to occur by availability reduction, favouring plants that can withstand the lowest water potential. Because light is supplied from above plants, individuals that situate their leaves above those of neighbours benefit directly from increased photosynthetic rates and indirectly by reducing the growth of those neighbours via shade. In communities where juveniles recruit in the shade of adults, traits of the most competitive species are biased towards those that confer greater survivorship and growth at the juvenile stage, even if those traits come at the expense of adult performance. 4. Understanding the mechanisms of competition also reveals how competition has influenced the evolution of plant species. For example, nutrient competition has selected for plants to maintain higher root length and light competition plants that are taller, with deeper, flatter canopies than would be optimal in the absence of competition. 5. In all, while more research is needed on competition for heterogeneous resource supplies as well as for water, understanding the mechanisms of competition increases the predictability of interspecific interactions and reveals how competition has altered the evolution of plants. Key-words: light, nutrients, resource competition, supply pre-emption, water Plants are evidently in general, tolerably impartial as regards soil, if we except certain chemical and physical extremes (abundance of common salt, of lime, or of water), so long as they have not competitors—Eugenius Warming, Oecology of Plants (1909). Competition for resources among plants has long been considered to generate stress for plants and to be important for determining the distribution of species, as well as their evolution. Eugenius Warming (1909) had noted, for example, that many species could be found in botanical *Correspondence author. E-mail: [email protected]

gardens when isolated from interactions with other plants but would not maintain themselves when subjected to competition from other species. Charles Darwin did not discuss competition much, but did write, ‘Not until we reach the extreme confines of life in the arctic regions, or on the borders of an utter desert, will competition cease’ (Darwin 1875, p. 78). Despite its early emphasis, research into the mechanisms by which plants competed developed slowly. For many years, competition between organisms was synonymous with interspecific interaction coefficients in Lotka–Volterra equations. These coefficients relate the phenomenological

© 2013 The Authors. Functional Ecology © 2013 British Ecological Society

834 J. M. Craine & R. Dybzinski net effects of species on each other, but little else. Tilman’s research in the mid-1970s on phytoplankton took a mechanistic approach that could test hypotheses about the causes of observed patterns and thus represented a turning point in our understanding of resource competition (Tilman 1977). If competition for resources has been understood to be of widespread importance for over a century and direct research into how plants compete dates back over a quarter century, how far have we come in understanding the mechanisms by which plants compete, the prevalence and importance of competition for different resources and how competition has altered the evolution of plants? This article seeks to address some of the recent advances in our understanding of the mechanisms that underlie plant competition for nutrients, water and light while also summarizing what has been learned about how competition has altered the evolution of plants. In addressing the mechanisms of competition, we focus on the processes by which individual plants reduce the availability of resources to other individuals. Only through understanding the mechanisms by which plants compete can we understand the adaptations associated with resource competition and how competition plays out across different ecosystems.

The complexity of resources In general, nutrients, water and light are the three main classes of resources that limit plant growth and are considered to be resources for which individual plants compete. The concentration of atmospheric CO2 can also limit plant growth, but because the atmospheric pool of CO2 is so large and so well mixed, plants are not thought to compete for CO2. Investigating resource competition in all of its forms is made complex by the unique characteristics of the different resources that might limit plant growth. Numerous nutrients (here, elements besides C, O and H) can limit plant growth, and each has different properties in soils. For example, individual nutrients vary in their diffusivity in soils (Tinker & Nye 1977), nutrients can be acquired as organic or inorganic forms, with multiple acquirable forms present for nutrients like N (NHþ 4 , NO3 , 3 amino acids) or P (H2 PO4 ,HPO4 , organic P). The properties of the soils also affect the behaviour of nutrients, for example, altering their rate of diffusion. Nutrients can be supplied through abiotic process or through biotic processes like microbial decomposition of organic matter. Beyond their activity in acquiring available nutrients, plant activity can also increase or decrease nutrient availability. As such, nutrient supplies are not necessarily independent of the species present or their dynamics. In comparison with, for example, nitrogen that can be made available to plants as organic N, NHþ 4 or NO3 , water does not differ in form when available. In addition, although plants can ‘harvest’ water from fog or alter the rates at which water is lost in soils, plants are not generally thought to be able to increase the availability of water in a

given soil profile as with nutrients. Yet, water is supplied heterogeneously in time and is spatially heterogeneous vertically and horizontally. Moreover, plants can redistribute water in the soil profile on diel time-scales. Light would seem less complex – for example, plants cannot increase light supplies – but light competition is still complex. Light is generally supplied directionally at angles that shift daily and seasonally, but light can also be supplied diffusely after scattering through clouds or vegetation. Light varies in its wavelength composition and is temporally variable on a range of scales from seasonal patterns to minute-scale variation associated with sunflecks. In all, a robust set of theories about competition for resources should take into account (i) the forms of resources availability, (ii) the mechanistic role the resource plays in the plant’s physiology, (iii) the temporal variability in its supply – understanding how competition occurs when resources are supplied evenly or heterogeneously in time and (iv) the spatial variation in supplies – being clear about the scale of this variation and how it relates to plant size. That said, in this paper, our focus here is to investigate how plants compete for nutrients, water and light when supplied evenly in space in time without detailing differences in the behaviour of different nutrients or light characteristics. In simultaneously addressing competition for the three types of resources, consistent terminology is important (Craine 2009). Here, the supply of the resource is defined as the production of a resource per unit area or volume that is potentially acquirable by the plant per unit time. In contrast, the availability of a resource is defined as the supply relative to the demand. The most limiting resource is the one that has the lowest supply relative to demand by the plant and thus the lowest availability. Having defined these terms, resource competition has long represented the process by which plants reduce the availability of a limiting resource to other plants. Weaver and Clements (1938) defined competition as occurring ‘where two or more plants make demands for light, nutrients or water in excess of the supply’. Grime stated that resource competition was ‘the tendency of neighbouring plants to utilize the same quantum of light, ion of a mineral nutrient, molecule of water, or volume of space’ (Grime 1973). Craine (2009) improved on past definitions and defined resource competition as ‘the process by which two or more individuals differentially capture a potentially common, limiting resource supply’. The word ‘differentially’ was used to invoke the idea that individual plants were individually acquiring resources from a common supply. A slight modification here would be that resource competition is ‘the process by which two or more individuals acquire resources from a potentially common, limiting supply’.

Nutrients The presence of multiple plants in a given volume of soil can induce nutrient stress in a given plant as neighbours

© 2013 The Authors. Functional Ecology © 2013 British Ecological Society, Functional Ecology, 27, 833–840

Mechanisms of Competition acquire limiting resources. Tilman’s theoretical work led to the prediction that terrestrial plants that can reduce the concentration of nitrogen in soil solution to the lowest level should, all other things equal, displace species that are less able to reduce soil solution N concentrations (Tilman & Wedin 1991a). Experiments with five species of grass grown on soils with low N contents supported this hypothesis (Tilman & Wedin 1991b). The concentration reduction hypothesis, which essentially posited that one species displaced others based on their ability to lower the concentration of resources in the environment, was a great advance over phenomenological approaches and injected much needed mechanism into understanding plant interactions. Yet, the dynamics of nutrients in soils are more complex than the well-mixed algal cultures that generated the concentration reduction hypothesis. Most importantly, nutrients are not well mixed in soil solution, which changes the nature of nutrient competition and elevates the importance of supply pre-emption for nutrients. For plants in soil, nutrient availability is not well represented by average concentrations in soil solution, but instead by the supplies of nutrients to roots (Craine, Fargione & Sugita 2005). For example, one species could reduce a nutrient in soil solution to a lower average concentration than another species simply by taking up less water, but this would not cause it to be a better competitor for the nutrient (Craine, Fargione & Sugita 2005). In contrast to the concentration reduction hypothesis, supply pre-emption hypothesis posited that plants do not out-compete others by reducing the concentration of resources in the environment, but instead by pre-empting the resource supplies from coming in contact with other species. Competition for nutrients when supplied under steady-state conditions is influenced by the rates of diffusion of the nutrients in soil solution. For most nutrients under most soil environments, the diffusion of nutrients to roots is slower than potential uptake rates. Diffusivity of nutrients is determined by their size, but also their charge þ relative to soils. For example, NO 3 and NH4 have similar þ diffusion coefficients, but NH4 diffuses much more slowly in most soils because most SOM and clay are negatively charged (Tinker & Nye 1977). Still, under most conditions experienced by nutrient-limited plants growing in soils, even for the most mobile forms of nutrients, for example, NO 3 in soils with high cation exchange capacity, depletion zones are generated around roots and uptake rates are relatively insensitive to the potential uptake parameters of roots, no less average soil solution concentrations. Diffusion of nutrients to roots are relatively unaffected by changes in minimum concentrations at the root surface, water uptake rate or maximum nutrient uptake rates (Smethurst & Comerford 1993; Craine, Fargione & Sugita 2005; Craine 2006). The partitioning of nutrient supplies is proportional to the root length density of different individuals (Reich et al. 2003; Raynaud & Leadley 2004; Craine, Fargione & Sugita 2005). Plants with higher root length in a given volume of

835

soil acquire more of the nutrient supply. For example, Hodge et al. (1999) grew two grass species alone and in mixture and found that the amount of nitrogen acquired from patches of N was proportional to their relative root length in the patch, explaining why plants proliferate roots in patches of high nutrient availability (Robinson et al. 1999). Raynaud & Leadley (2004) showed that what held for small patches also applied to larger soil volumes. They created a fine-scale model of roots in soil that included the supply and diffusion of nutrients along concentration gradients generated by root uptake of nutrients. In their model, partitioning of nutrient supplies by two competing plants was proportional to the relative amounts of root length in soil. Craine, Fargione and Sugita (Craine, Fargione & Sugita 2005) also used a fine-scale processbased model of soil dynamics to explicitly compare the ability of concentration reduction and supply pre-emption hypotheses to predict competitive outcomes. Their work showed that the plant that produced and maintained higher root length density displaced competitors. However, high root length density also generated lower soil solution nutrient concentrations, suggesting that concentration reduction and supply pre-emption hypotheses would lead to similar predictions of competitive outcomes, all else equal.

SL as an analogue for R* While empirical work and simulations of nutrient dynamics in soils have supported the role of supply preemption for nutrients, supply pre-emption has never been investigated analytically. According to the concentration reduction hypothesis, R* is the minimum concentration to which a plant species can reduce a soil nutrient in monoculture, and the species with the lowest R* for a particular nutrient is predicted to win in competition for that nutrient. Tilman (1990) considered a range of concentration reduction models that included various aspects of plant allocation and physiology, relating them to R* values. Although it has not been developed as fully, the pre-emption analogue for R* would be SL , the equilibrial nutrient supply per unit root length. Parallel to R*, the species with the lowest SL is predicted to win in competition. Depending on the scale of nutrient supply, like R*, SL can be assessed at the individual or population level. Similar to Tilman’s (1990) effort, the factors that affect the pre-emption of nutrients and the growth and loss of biomass can be analysed to determine the factors that alter competitive success. As discussed earlier, the nutrient supply per unit length (SL ) determines uptake per unit root length when supplies of a nutrient are limiting to growth. SL is determined by the total nutrient supply and the fraction of the total root length a plant possesses in a given soil volume, as well as the position of roots relative to one another. As such, being able to maintain biomass at a low


836 J. M. Craine & R. Dybzinski supply per unit root length is the key to maintaining a high root length per unit volume of soil (LV) and therefore to being competitive for nutrients. Just as for R*, one can find the factors that determine the steady-state size of a population of plants, except that instead of expressing growth as a function of average soil solution concentrations, it is expressed as a function of SL , and instead of examining changes in population biomass, the analysis examines changes in root length. Let dL ¼ fðSL Þ m Ldt

eqn 1

where L is root length, f expresses growth in root length as a function of SL , and m is a density-independent loss rate. Each of these can take on species-specific values. Setting the relative growth rate of the root system equal to zero, the equation can be solved for the SL at which the system’s root length is at steady state: SL ¼ f1 ðmÞ:

eqn 2

When root length is at its steady-state value, if one knows the loss rate and the relationship between growth and supply per unit root length in the focal volume of soil, one can determine the factors that affect SL . These species-specific SL values could then be compared among species grown at the same nutrient supplies to predict competitive outcomes when plants are competing for the same limiting nutrient. If we assume that uptake is a linear function of the supply per unit root length over low nutrient supplies (Tinker & Nye 2000), then over this range fðSL Þ ¼

SL AllocR SRL ½ NR

eqn 3

where AllocR is the relative allocation of acquired nutrients to root biomass, [NR] is the nutrient concentration of roots, and SRL is the specific root length of root biomass. Again, all of these can take on species-specific values. In plain terms, the amount of root length produced is dependent on how much nutrient is acquired, how much of the nutrient is allocated to root production, how much root biomass is produced per unit N allocated to root production and how much root length is produced per unit biomass. Solving for SL provides the relationship between factors that affect root length production and the supply per unit length at which the relative growth rate of the root system is zero: SL ¼

m ½ NR : AllocR SRL

eqn 4

Plants that have a low loss rate, low root nutrient concentration, high allocation rate to roots or a high SRL will have a low SL and should therefore out-compete plants with the opposite traits. These are the very predictions supported by Craine, Fargione & Sugita (2005) using mechanistic models of nutrient transport and uptake.

Tilman’s similar analysis (model #3, Tilman (1990)) found the same qualitative relationships between the first three of these four traits and R*. Although model #3 of Tilman (1990) does not include SRL, it does include maximal rate of nutrient uptake per unit biomass, which should be positively correlated with SRL. If so, the qualitative relationships expressed by the two models are again similar, as increasing the maximal rate of nutrient uptake decreases R*, just as increasing SRL decreases SL . However, there is a predictive distinction between the two frameworks because factors that affect either the uptake capacity of roots or diffusion of nutrients to roots alter R* but have little effect on SL . Development of the supply pre-emption hypothesis with more detailed growth and loss equations deserves more attention than is provided here, but it is clear that the approach originally taken by Tilman (1990) furthers the supply pre-emption hypothesis and our understanding of competition for nutrients.

Applying pre-emption theory Fargione & Tilman (2006) tested the relative power of metrics derived from concentration reduction hypotheses (soil inorganic nutrient concentrations) and supply pre-emption hypotheses (soil root length density) to explain the relative abundance of different grassland species in experimental communities. They found that species that had high relative yield in mixtures (relative to their biomass in monocultures) produced both high root length density in monoculture and reduced soil solution N concentrations to low levels. The two metrics explained a similar proportion of variation in relative yield among species. There were individual species that reduced soil solution concentrations to low levels in monoculture that appeared to be reduced in abundance by competition, but there were also species that had high root length density in monoculture that also performed poorly in mixtures. Supply pre-emption also generated unique hypotheses to explain unique aspects of the morphology of plants that would have evolved when competition for nutrients was a strong selective force. Craine (2006) used the fine-scale model of soils and roots to calculate optimal rooting densities under competitive and noncompetitive scenarios. In short, in the absence of competitors, optimal root densities are a few per cent of optimal root densities in the presence of competitors. Introducing competition increased the optimal allocation of resources below-ground and would likely select for thinner, more long-lived roots depending on contrasting constraints. Therefore, it is possible that competition has selected for species that maintain higher root length densities than would be optimal in the absence of competition.

Water Despite the well-known ecological effects of shortages of water to plants, competition for water is less studied than


Mechanisms of Competition nutrients (or light) (also see Schwinning, this issue). In part, this can be ascribed to the fact that reduction in water availability can occur through both abiotic and biotic means, which obscures the effects of competition. Although evaporation becomes the dominant mechanism of soil drying as plant critical water potentials are approached, transpiration often dries soils faster than evaporation, which leads to evaporation increasing in relative importance. As such, it is not hard to understand why the traits that might be associated with water competition have never been isolated from those associated with environmentally induced drought despite a long history of considering competition for water (Cable 1969). Putting competition for water on equal footing with nutrients and light requires applying competition-related questions generated for other resources to water. For example, one goal of exploring competition for water is to understand the functional traits that are favoured when water is limiting. Are competitive plants selected to use water faster, either by having low water use efficiency or transpiration at night? Are adaptations for water competition similar to those of nutrient competition, such that water supplies can be pre-empted on a small scale by individuals with relatively high root length density? Or does competition follow concentration reduction hypotheses and favour plants that can maintain function at lower water potentials? Already, it is well known that plants can withstand immense tensions on their water columns, not necessarily to move water from great depths or to great heights, but instead to withstand dry soils. Only 1 MPa of tension is required to move water 100 m (Zimmermann 1983), but many woody species and grasses can withstand more than 10 MPa of tension (Pratt et al. 2008; Craine et al. 2013). Such physiological drought tolerance allows plants to function in dry environments, but it might also allow plants to reduce water availability to levels low enough that competitors are shut down or killed. Although the physiological drought tolerance of many species could be considered a product of environmental conditions, competition for water could have influenced the low water availability requirements of many species, as was postulated for concentration reduction hypotheses that have been applied for nutrients and light.

Supply pre-emption for water? Under steady-state supplies, the key to understanding whether competition for water should be conceptualized as supply pre-emption or concentration reduction is whether there are water potential gradients around roots. If there are no water potential gradients around roots, then soils within the rooting zone would all be considered a similar water potential and competition for water would be associated with the plant that can withstand the lowest water potentials, just as with an R* model. Here, the critical water potential (Ψcrit) at which photosynthesis or stomatal

837

conductance ceases (Tucker, Craine & Nippert 2011) would represent the lowest level to which plants could reduce water availability in the soil, assuming they explore the soil relatively thoroughly. With gradients of water availability around roots, supplies of water have the potential to be pre-empted through root length dominance, just as with nutrients. Single root models have long showed that water potential gradients should exist around roots (Hillel 1998). Each day, as transpirational demand increases, plant water potentials decline by up to 1–2 MPa, while soil water potential declines minimally (Woodruff et al. 2007). With the lowering of the water potential of the plant transpiration stream, zones of water depletion are generated around roots that can extend out multiple mm. Recent empirical work supports this theory. Herron, Gage & Cardon (2010) recently used bacteria that were engineered with a reporter system based on osmotic potential to test for water potential gradients around roots. By observing the spatial patterns of fluorescence around roots, they demonstrated a gradient of water potential around roots. Hence, although spatially explicit models are required, plants might be able to pre-empt water supplies from contacting the roots of neighbours in a manner analogous to nutrients. When supplies of water are directional, roots might be preferentially placed in the soil to pre-empt the supply from competitors as occurs with light. A simulation model of forests in the Netherlands showed that the optimal distribution of roots in the absence of competition was deeper than in the presence of competition (van Wijk & Bouten 2001). Empirical root distributions most closely matched theoretical root distributions of plants competing for water. Here, maintaining shallower roots than optimum pre-empts water from plants with deeper roots, but comes at a cost. Theoretically, 10-year transpiration was reduced 10–20% in studied forests as a result of plants holding their roots shallower than optimal, which in dry years prevents them from accessing deeper water. In summary, the consequences for competition for water for the evolution of plants and the functioning of ecosystems are poorly explored. Theoretically, competition for water likely involves reducing soil water potential to low levels, but might require supply pre-emption in some cases or concentration reduction in others. Supply pre-emption for water might select for supra-optimal root length density, greater resistance to cavitation and alteration of root placement in soil in response to directional water supplies. Likely, soils dry out faster as a consequence of competition for water, although the magnitude of this effect is poorly quantified.

Light The empirical signature of light limitation is an increase in growth in response to an increase in light availability, which will only occur if plant demand for carbon exceeds the supply of carbon from the canopy. A necessary, but


838 J. M. Craine & R. Dybzinski not sufficient condition for light limitation at the wholeplant level is light limitation at the leaf level, which occurs whenever the photosynthetic capacity of a leaf is in excess of the light available for photosynthesis. Numerous lines of evidence suggest that light limitation should be common. However, because few experiments have increased light availability to ecosystems (Wilson & Tilman 1991; Hautier, Niklaus & Hector 2009), we have little direct knowledge regarding both the quantitative extent of light limitation and the importance of light competition relative to other resources among ecosystems. Because a leaf that is chronically light limited (i.e. in possession of excessive photosynthetic machinery) incurs respiratory and maintenance costs for that unused potential, as well as missed opportunity costs for the resources tied up in that unused potential, evolution has selected leaf traits that economically coordinate photosynthetic capacity with light levels typical of a species’ life history (Wright et al. 2004). There is often tremendous variation in the photosynthetic capacity between leaves of different species, between leaves of different individuals of the same species and, indeed, between different leaves in the same individual (Bassow & Bazzaz 1997; Kattge et al. 2011). Thus, at any given light level, some plants may be light limited and others not. For example, an early-successional colonist may have a high photosynthetic capacity consistent with the open conditions for which its life history is coordinated. In contrast, a forest understorey herb may have a low photosynthetic capacity consistent with the dark conditions for which its life history is coordinated. If exposed to the same, intermediate light level, the early-successional colonist may be light limited and the understorey herb light saturated. Species that must situate leaves across a wide range of mean light availabilities (e.g. late-successional trees) have evolved the ability to plastically build leaves of differing photosynthetic capacities (Ellsworth & Reich 1993), for example, sun and shade leaves. The directional nature of light leads to size-asymmetric competitive dynamics that are qualitatively different from the size-symmetric competitive dynamics of nutrients or water (Weiner 1990). From the maximum light availability at the top of a vegetative canopy, light levels are reduced exponentially by each successive layer of leaves. Critically, even though a given leaf reduces the availability of light to the leaves below it, it does not suffer that reduction in availability itself. As such, competition for light can be cast in terms of pre-emption of supplies, and there is a premium placed on situating leaves above those of a competitor, both for the direct benefit – maximal photosynthetic rates – and for the indirect benefit – slowing the growth (via reduced photosynthetic rates) of a competitor (Falster & Westoby 2003).

Modelling light competition The size-asymmetric nature of light competition, though easy to understand, is difficult to model in a way that is

simple enough to yield analytical insights analogous to those of nutrient models (such as the one presented above). Consequently, simulations have traditionally been used to model height-structured light competition (e.g. Tilman 1988; Pacala et al. 1996). However, the recently proposed perfect plasticity approximation (Strigul et al. 2008; Dybzinski et al. 2011) provides, in the special case of old growth, equilibrial forests, a closed form expression for the fitness, W, of an individual in the context of height-structured light competition: W Fa

GhC DlGC e C: lhþ1 C

eqn 5

where F is fecundity; GC and GU are stem diameter growth rates in the canopy and understorey life-history stages, respectively; lC and lU are mortality rates in the canopy and understorey life-history stages, respectively; a and h are constants that relate stem diameter to crown area; and D is the stem diameter (related allometrically to crown height, not shown) at which trees transition from the understorey to the canopy life-history stage. Depending on the question, these parameters can be treated as constants, variables or functions of other phenomena. For example, Dybzinski et al. (2011) modelled growth rates as a function of nitrogen and light availability to make predictions of carbon allocation across gradients of resource availability. It is important to note that this equation is not assumed, but is rather the mathematical approximation to an integral that quantitatively characterizes fecundity, growth and survival in both the understorey and canopy stages (Adams, Purves & Pacala 2007). Thus, its inspection reveals many of the critical components of height-structured competition for light, as well as their interrelationships and relative importance (Adams, Purves & Pacala 2007). In general, fitness is more sensitive to the understorey vital rates (which are exponentiated) than to the canopy vital rates (which are not exponentiated) as a result of the understorey’s role in providing recruits to the canopy stage and, less intuitively but just as importantly, in setting the mean canopy height (Z*, see below). Moreover, that understorey sensitivity increases as the average time spent in the understorey stage increases (via increased height of the canopy, /D; increased understorey mortality rate; or decreased understorey growth rate). With respect to traits for which species lack sufficient plasticity, this suggests that the trait requirements of the understorey stage will weigh more heavily in the species mean than will those of the canopy stage as demonstrated empirically by Poorter (2007). Moreover, the PPA modelling framework yields a predictive index, termed Z*, that in many cases allows species to be ranked according to their competitive ability for light (Adams, Purves & Pacala 2007), analogous to R* or SL discussed above. Z* is the mean canopy height, that is, the mean height at which a surviving understorey individual


Mechanisms of Competition transitions to the canopy stage, and again, it is a solution, not an assumption. Unlike R* or SL however, the species with the highest index, not the lowest index, is predicted to win in competition, all else equal: " #!q GU GhC Z b ln Fa hþ1 eqn 6 lU lC where b and q are constants that relate stem diameter to height and all other parameters are as above. In general, species with faster growth rates, greater fecundity, greater crown area and lower mortality will be more competitive, and again, inspection reveals that Z* is more sensitive to understorey parameters than to canopy parameters. A variant of this PPA parameterized for common temperate forest species yielded good quantitative predictions of measured forest structure and dynamics, and Z* was largely successful in predicting the observed outcome of competition over nearly a century of succession (Purves et al. 2008). An index such as Z*, which integrates the whole life history of a species within a rigorous height-structured framework, is preferable to ranking species according to the light remaining at the soil surface in monoculture, an index usually labelled I*. Although I* has shown some qualified success in predicting the outcome of competition (e.g. Dybzinski & Tilman 2007; Vojtech, Turnbull & Hector 2007), even advocates of the concentration reduction hypothesis (e.g. Tilman 1988) observe that both the asymmetry of height and the independence of light consumption from the light environment that consumption creates make I* an approximation at best. Adams, Purves & Pacala (2007) used the PPA to demonstrate that interspecific differences in I* due only to interspecific differences in crown light transmissivities (i.e. holding all other traits constant) led to neutral, not competitively hierarchical, dynamics. If and when I* works, it does so because species traits in the juvenile stage, such as shade tolerance, are coordinated with traits at the adult stage, such as leaf area index. However, Z* can incorporate these traits more directly and more mechanistically than can I*.

Evolutionary consequences of light competition Even though the community as a whole would be no less productive without it, evolution has favoured height growth as an unavoidable, though inefficient arms race in many plant communities. Potential problems of engineering aside (e.g. Ryan & Yoder 1997), it is only because the costs of height growth (e.g. increased risk of wind-throw or cavitation) eventually outweigh the benefits (in ways that are unique to different ecosystems) that the evolutionary arms race for ever greater height reaches stasis (Falster & Westoby 2003). Beyond selecting for taller plants, competition for light has also selected for species to maintain higher leaf area and to hold leaves more horizontally than is optimal in the

839

absence of competition (parallel to the effects of competition on optimal root length discussed above). Modelling of light acquisition for plants grown in the absence and presence of neighbours shows that some species maintain twice the leaf area than the leaf area that maximizes canopy carbon gain in the absence of competition (Anten 2005). Holding greater leaf area than is optimal reduces net carbon gain for the plant when growing in the absence of competition, but reduces the growth of competitors enough to provide an unassailable competitive advantage (i.e. it is evolutionarily stable). Similarly, holding leaves more horizontally creates shallower penetration of light into the canopy, which reduces canopy-level carbon gain for a plant, but again also restricts the growth of competitors enough to make tall plants with a high area of flatly held leaves evolutionarily stable.

Conclusions Despite the need for more research, our understanding of competition has come a long way over the past 100 years, verifying the initial observations and impressions of ecologists. Recent investigations of competition have revealed some of the mechanisms of how plants interact when limited by the same resource and how resource competition has altered the evolution of species. That said, research into resource competition is still developing. It is clear that more spatially explicit models of water and nutrient dynamics are necessary to further understand how plants compete for nutrients. There are no models that explore mechanistically how plants compete for water, no less how water and nutrient competition might interact. How the mechanisms of competition might be altered with heterogeneity of resource supplies is still poorly understood. For example, water supplies to plants are pulsed and many species are able to store water to different degrees, but water storage strategies have poorly been incorporated into a competitive framework.

Acknowledgements The authors thank David Robinson for the opportunity to contribute to this special issue. Deborah Goldberg and an anonymous referee contributed valuable discussion.

References Adams, T.P., Purves, D.W. & Pacala, S.W. (2007) Understanding heightstructured competition in forests: is there an R* for light? Proceedings of the Royal Society B-Biological Sciences, 274, 3039–3047. Anten, N.P.R. (2005) Optimal photosynthetic characteristics of individual plants in vegetation stands and implications for species coexistence. Annals of Botany, 95, 495–506. Bassow, S.L. & Bazzaz, F.A. (1997) Intra- and inter-specific variation in canopy photosynthesis in a mixed deciduous forest. Oecologia, 109, 507–515. Cable, D.R. (1969) Competition in the semidesert grass-shrub type as influneced by root systems, growth habits, and soil moisture extraction. Ecology, 50, 27–38. Craine, J.M. (2006) Competition for nutrients and optimal root allocation. Plant and Soil, 285, 171–185.


840 J. M. Craine & R. Dybzinski Craine, J. (2009) Resource Strategies of Wild Plants. Princeton University Press. Craine, J.M., Fargione, J. & Sugita, S. (2005) Supply pre-emption, not concentration reduction, is the mechanism of competition for nutrients. New Phytologist, 166, 933–940. Craine, J.M., Ocheltree, T.W., Nippert, J.B., Towne, E.G., Skibbe, A.M., Kembel, S.W. & Fargione, J.E. (2013) Global diversity of drought tolerance and grassland climate-change resilience. Nature Climate Change, 3, 63–67. Darwin, C. (1875) Origin of the Species. John Murray, London. Dybzinski, R. & Tilman, D. (2007) Resource use patterns predict long-term outcomes of plant competition for nutrients and light. American Naturalist, 170, 305–318. Dybzinski, R., Farrior, C., Wolf, A., Reich, P.B. & Pacala, S.W. (2011) Evolutionarily stable strategy carbon allocation to foliage, wood, and fine roots in trees competing for light and nitrogen: an analytically tractable, individual-based model and quantitative comparisons to data. American Naturalist, 177, 153–166. Ellsworth, D.S. & Reich, P.B. (1993) Canopy structure and vertical patterns of photosynthesis and related leaf traits in a deciduous forest. Oecologia, 96, 169–178. Falster, D.S. & Westoby, M. (2003) Plant height and evolutionary games. Trends in Ecology & Evolution, 18, 337–343. Fargione, J. & Tilman, D. (2006) Plant species traits and capacity for resource reduction predict yield and abundance under competition in nitrogen-limited grassland. Functional Ecology, 20, 533–540. Grime, J.P. (1973) Competitive exclusion in herbaceous vegetation. Nature, 242, 344–347. Hautier, Y., Niklaus, P.A. & Hector, A. (2009) Competition for light causes plant biodiversity loss after eutrophication. Science, 324, 636–638. Herron, P.M., Gage, D.J. & Cardon, Z.G. (2010) Micro-scale water potential gradients visualized in soil around plant root tips using microbiosensors. Plant Cell and Environment, 33, 199–210. Hillel, D. (1998) Environmental Soil Physics: Fundamentals, Applications, and Environmental Considerations. Academic Press, San Diego. Hodge, A., Robinson, D., Griffiths, B.S. & Fitter, A.H. (1999) Why plants bother: root proliferation results in increased nitrogen capture from an organic patch when two grasses compete. Plant, Cell and Environment, 22, 811–820. Kattge, J., Diaz, S., Lavorel, S., Prentice, C., Leadley, P. & Bonisch, G. et al. (2011) TRY - a global database of plant traits. Global Change Biology, 17, 2905–2935. Pacala, S.W., Canham, C.D., Saponara, J., Silander, J.A., Kobe, R.K. & Ribbens, E. (1996) Forest models defined by field measurements: estimation, error analysis and dynamics. Ecological Monographs, 66, 1–43. Poorter, L. (2007) Are species adapted to their regeneration niche, adult niche, or both? American Naturalist, 169, 433–442. Pratt, R.B., Jacobsen, A.L., Mohla, R., Ewers, F.W. & Davis, S.D. (2008) Linkage between water stress tolerance and life history type in seedlings of nine chaparral species (Rhamnaceae). Journal of Ecology, 96, 1252– 1265. Purves, D.W., Lichstein, J.W., Strigul, N. & Pacala, S.W. (2008) Predicting and understanding forest dynamics using a simple tractable model. Proceedings of the National Academy of Sciences, 105, 17018–17022. Raynaud, X. & Leadley, P.W. (2004) Soil characteristics play a key role in modeling nutrient competition in plant communities. Ecology, 85, 2200– 2214. Reich, P., Wright, I., Cavender-Bares, J., Craine, J., Oleksyn, J., Westoby, M. & Walters, M. (2003) The evolution of plant functional variation: traits, spectra, and strategies. International Journal of Plant Sciences, 164, S143–S164.

Robinson, D., Hodge, A., Griffiths, B.S. & Fitter, A.H. (1999) Plant root proliferation in nitrogen-rich patches confers competitive advantage. Proceedings of the Royal Society of London - Series B: Biological Sciences, 266, 431–435. Ryan, M.G. & Yoder, B.J. (1997) Hydraulic limits to tree height and tree growth. BioScience, 47, 235–242. Schwinning, S. & Kelly, C.K (2013) Plant competition, temporal niches and implications for productivity and adaptability to climate change in water-limited environments. Functional Ecology, 27, 886–897. Smethurst, P.J. & Comerford, N.B. (1993) Simulating nutrient uptake by single or competing and contrasting root systems. Soil Science Society of America Journal, 57, 1361–1367. Strigul, N., Pristinski, D., Purves, D., Dushoff, J. & Pacala, S. (2008) Scaling from trees to forests: tractable macroscopic equations for forest dynamics. Ecological Monographs, 78, 523–545. Tilman, D. (1977) Resource competition between planktonic algae - experimental and theoretical approach. Ecology, 58, 338–348. Tilman, D. (1988) Plant Strategies and the Dynamics and Structure of Plant Communities. Princeton University Press, Princeton. Tilman, D. (1990) Mechanisms of plant competition for nutrients the elements of a predictive theory of competition. Perspectives on Plant Competition (eds J.B. Grace & D. Tilman), pp. 484. Academic Press, New York. Tilman, D. & Wedin, D. (1991a) Dynamics of nitrogen competition between successional grasses. Ecology, 72, 1038–1049. Tilman, D. & Wedin, D. (1991b) Plant traits and resource reduction for five grasses growing on a nitrogen gradient. Ecology, 73, 685–700. Tinker, P.B. & Nye, P.H. (1977) Solute Movement in the Soil-root System. University of California Press, Berkeley. Tinker, P.B. & Nye, P.H. (2000) Solute Movement in the Rhizosphere. Oxford University Press, Oxford. Tucker, S.S., Craine, J.M. & Nippert, J.B. (2011) Physiological drought tolerance and the structuring of tallgrass assemblages. Ecosphere, 2, 48. Vojtech, E., Turnbull, L.A. & Hector, A. (2007) Differences in light interception in grass monocultures predict short-term competitive outcomes under productive conditions. PLoS ONE, 2, e499. Weiner, J. (1990) Asymmetric competition in plant populations. Trends in Ecology & Evolution, 5, 360–364. van Wijk, M.T. & Bouten, W. (2001) Towards understanding tree root profiles: simulating hydrologically optimal strategies for root distribution. Hydrology and Earth System Sciences, 5, 629–644. Warming, E. (1909) The Oecology of Plants. Clarendon Press, Oxford. Weaver, J.E. & Clements, F. (1938) Plant Ecology, 2nd edn. McGraw-Hill, New York. Wilson, S.D. & Tilman, D. (1991) Components of plant competition along an experimental gradient of nitrogen availability. Ecology, 72, 1050– 1065. Woodruff, D.R., McCulloh, K.A., Warren, J.M., Meinzer, F.C. & Lachenbruch, B. (2007) Impacts of tree height on leaf hydraulic architecture and stomatal control in Douglas-fir. Plant Cell and Environment, 30, 559–569. Wright, I.J., Reich, P.B., Westoby, M., Ackerly, D.D., Baruch, Z., Bongers, F. et al. (2004) The worldwide leaf economics spectrum. Nature, 428, 821– 827. Zimmermann, M. (1983) Xylem Structure and the Ascent of Sap. SpringerVerlag, New York. Received 1 April 2012; accepted 1 February 2013 Handling Editor: David Robinson


Mechanisms of plant competition for nutrients, water and light

Recommend Documents