SEARCH

SEARCH BY CITATION

Keywords:

  • autocorrelated growth;
  • Chamaedorea ;
  • elasticity analysis;
  • growth differences;
  • historical effects;
  • individual variation;
  • integral projection model;
  • plant population and community dynamics;
  • shade tolerance;
  • trade-offs

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  1. Persistent variation in growth rate between individual plants can have strong effects on population dynamics as fast growers reach the reproductive size at an earlier age and thus potentially contribute more to population growth than slow growers. In tropical forests, such persistent growth differences have so far been documented for canopy tree species, where they are primarily associated with forest gap dynamics, but not for forest understorey species which are less responsive to gaps. Here, we study persistent growth differences and their demographic consequences for a tropical forest understorey palm, Chamaedorea elegans.
  2. We measured internodes along stems and annual leaf production rates to reconstruct lifetime growth trajectories. Using regression analysis, we determined the relative effect of stem length and past growth rate on vital rates (survival, growth and reproduction). We then simulated population dynamics using integral projection models (IPMs), in which individuals were categorized by both stem length and lifetime past growth rate.
  3. Stem growth differences among individual palms persisted over most of their lifetime. Past growth rate averaged over the palm's lifetime proved to be a very good predictor of growth, reproduction probability and seed production, often much better than stem length or age. The effects of past growth rate were positive, indicating that fast growers maintain high rates of growth and reproduction.
  4. Projected population growth rate (λ) was 1.056, and stable stage distributions closely resembled observed population structures. Separating individuals with above-median and below-median past growth rates in IPMs revealed substantial differences in elasticity values. The 50% fastest growers had a 1.8 times higher elasticity, and thus a 1.8 times higher contribution to population growth, compared to slow growers.
  5. Synthesis. Strong and persistent growth differences that are probably associated with environmental (edaphic) and/or genetic factors govern individual performance and population dynamics of a tropical forest understorey palm. Overall, our study shows that strong inter-individual growth variation is not limited to canopy trees and that it can be generated by other factors than canopy dynamics. It is likely that persistently fast-growing ‘super performers’ govern population growth of many long-lived species.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Not all individuals survive, grow and reproduce at the same rate within a population (Sarukhán, Martínez-Ramos & Piñero 1984). Some grow faster than others, and such growth differences can be maintained for some time or even for a substantial part of the total lifetime. Persistent differences have been documented for some time in animals (e.g. De Leo & Gatto 1995; Pfister & Stevens 2002; Pelletier et al. 2007; Tuljapurkar, Steiner & Orzack 2009), but they have received relatively little attention in plants. Studies in plants that did cover persistent growth differences were mostly restricted to tree species (Kohyama & Hara 1989; Clark & Clark 1992; Terborgh et al. 1997; Brienen, Zuidema & During 2006). Persistent growth variation in trees is likely caused by growth spurts (‘releases’) of juvenile trees that can be sustained over several years or decades (Lusk & Smith 1998; Brienen & Zuidema 2006). These growth spurts occur when juvenile trees growing in the forest understorey suddenly receive more light after the creation of canopy gaps.

While variation in light is an obvious driver for persistent growth differences of canopy trees, many understorey plant species complete their life cycle under low-light conditions, and these tend to exhibit much weaker growth responses to light compared to canopy tree species (e.g. Chazdon 1992; Svenning 2002; Martínez-Ramos, Anten & Ackerly 2009). As a result, differences in growth rate among individuals in understorey species might be smaller than in canopy species. Nevertheless, for forest understorey species, spatial variation in soil fertility, soil texture and water availability as well as genetic variation between individuals may cause growth differences. These factors may be more permanent than light availability (Ceccon, Huante & Rincón 2006) and, as a result, growth differences in understorey species might persist over longer-time spans than those in canopy species. However, until now this issue has been poorly explored.

Temporal autocorrelation of growth and persistent variation in growth rate strongly determine population dynamics (Zuidema & Franco 2001; Pfister & Stevens 2003; Pelletier et al. 2007; Vindenes, Engen & Saether 2008; Tuljapurkar, Steiner & Orzack 2009; Zuidema, Brienen & During 2009). Fast growers can have a disproportionate contribution to population growth because they reach reproductive size at an earlier age and have a higher probability of reaching that size (Zuidema, Brienen & During 2009). This suggests that environmental or genetic factors underlying variation in growth may have a more profound impact on population dynamics than previously assumed. For example, if fast-growing individuals are more sensitive to climate change or harvest regimes, this will have a disproportionately large influence on population growth. The magnitude of this effect will depend on the strength and persistence of growth differences, but also on possible positive or negative trade-offs between growth, survival and reproduction. Negative trade-offs between growth and other vital rates (e.g. Martorell, Vega & Ezcurra 2006) may offset the higher contribution of fast growers to population growth. Positive trade-offs increase this contribution (Van Noordwijk & De Jong 1986). Such trade-offs have hardly been studied so far (Zuidema, Brienen & During 2009) and not at all in forest understorey species.

We studied the magnitude and consequences of persistent growth differences in the tropical forest understorey palm Chamaedorea elegans. We choose this long-lived understorey species as a study system because the growth history of individual palms can be easily reconstructed from internode lengths (Lugo & Rivera Batlle 1987; Pinard 1993) and because demography of a large set of individuals has been monitored over several years (Martínez-Ramos, Anten & Ackerly 2009). This enables us, on the one hand, to investigate functions that relate growth, survival and reproduction to stem length and past growth rate and, on the other hand, to assess the influence of persistent inter-individual growth variation on population dynamics, by adapting an age-size-dependent integral projection model (IPM) (Childs et al. 2003; Ellner & Rees 2006). Specifically, we addressed the following questions: (i) Do growth differences among individuals persist over time and, if so, for how long? (ii) To what extent are historical growth differences related to current individual growth and reproduction? (iii) What are the consequences of persistent growth differences for population dynamics in a long-lived rainforest understorey plant?

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Study site and species

Data were collected at the Chajul Biological Field Station (16°06′ N, 90°56′ W) in the Montes Azules Biosphere Reserve (MABR), Chiapas, Mexico. The dominant vegetation type in the area is lowland evergreen tropical forest (Ibarra-Manríquez & Martínez-Ramos 2002). Annual mean precipitation is around 3000 mm with a dry season (monthly rainfall < 100 mm) from January to April. The mean annual temperature is about 25 °C. Precipitation can vary substantially, among others in relation to El Niño southern oscillation events (Martínez-Ramos, Anten & Ackerly 2009).

Chamaedorea elegans is a dioecious understorey palm found in tropical rain forests in Mexico, Guatemala and Belize (Hodel 1992). In MABR C. elegans occurs on karst-range sites, which are topographically irregular mountain chain areas (300–700 m a.s.l), where the soil is basically composed of a thin layer of organic matter, with masses of limestone rocks exposed over karst topography. Light levels vary between 0.5% and 33% of natural day light (average light level is 4.5%), and soil depth varies between 0 and 40 cm (average soil depth is 21.5 cm; N. P. R. Anten and M. Martínez-Ramos, unpubl. data). Chamaedorea elegans reaches a maximum height of 1.5 m, is unbranched, single-stemmed and produces one single cluster of leaves. Species with these characteristics are very suitable for reconstructing growth histories by measuring internodes (Pinard 1993). The leaves of C. elegans are harvested and sold nationally and internationally (USA and Europe) in the floral industry. As such they are an important non-timber forest product (NTFP) providing income to many people (Oyama 1992). The population of this species is in decline because of habitat loss, and leaf, seed and whole plant extraction (Eccardi 2003).

Data collection

In March 1997, a 100 m × 112 m plot was established in which 353 individuals of C. elegans (including males and females) were mapped and tagged. From then until 2000, plant height, stem length, number of leaves, reproductive status, reproductive activity (inflorescence, infructescence and fruit production), mortality, length of the most fully extended leaf and relative light intensity above each plant were measured annually for all individuals. Additionally, in March 1997, a number of seedling plots were established with a total surface area of 135 m2. In this area, all individuals shorter than 10 cm in height were mapped and tagged. In March 1998, 1999 and 2000, the new seedling cohorts were also identified, and each time the survival and height of all individuals in all cohorts was measured. More details about the study plots, methodology and demographic attributes of the studied population can be found elsewhere (Martínez-Ramos, Anten & Ackerly 2009).

In March 2010, all internodes of each of the individual palms within the plot that were still alive (187 plants in total) were counted and their length was measured. In most palms including C. elegans, this is easy as leaf scars clearly mark individual internodes (Tomlinson 1990). The length of the internode represents the stem growth of the individual palm due to the production of two leaves.

Statistical analysis

To estimate lifetime past growth rates, we calculated the average annual leaf production over the period March 1997 to March 2000 per individual. Combining this information with internode length, we could reconstruct the age of a palm at any stem length. This method assumes that leaf production per individual is approximately constant within a stem. This assumption likely overestimates the differences in stem growth rates among individuals, because leaf production of an individual is unlikely to be constant over time (see 'Results'). To estimate the sensitivity of our results to assumptions on variation in leaf production among individuals, we performed a robustness test. All statistical and modelling analyses were also performed under the assumption that all individuals had equal leaf production. In this case, the differences in lifetime past growth rates between individuals are only based on differences in internode length. This way of calculating certainly underestimates persistent growth differences among individuals, as leaf production rates do vary among individuals (Martínez-Ramos, Anten & Ackerly 2009), but it provides a lower-bound estimate of the importance of past growth on individual performance and population growth.

We investigated the persistence of growth differences between individuals over ages using Spearman rank correlations. We correlated growth rates at each age with growth rate at subsequent ages.

We analysed the combined effects of stem length, age and past growth rate on vital rates (growth, survival, reproduction probability and seed production) over the period March 1997 to March 2000. We included only female palms in the statistical analyses of vital rates. In the case of survival, we could not identify a relation to past growth or age, as the internodes of plants that had already died could no longer be measured. Therefore, in the analyses of survival, we used all female palms (142 individuals), and in the analysis of growth rate, reproduction chance and seed production, we used all female palms for which we were able to collect internode data in 2010 (68 individuals). Determinants of the mean annual growth rate and of the mean annual seed production were evaluated using multiple (stepwise backward) linear regressions. Determinants of the 3-year probability of survival and the probability of reproduction (per year) were evaluated using multiple (stepwise backward) logistic regressions. In all regression analyses, we tested for the effect of size, age and lifetime past growth rate (i.e. size/age) on vital rates (age was also added as age2). Regression analyses were performed in R and were also performed for the robustness test.

Construction of a size-past growth model

To evaluate the effect of persistently fast growers on population growth, an integral projection model (IPM) was constructed that included past growth. The basis for this model was an age- and size-dependent IPM (Childs et al. 2003; Ellner & Rees 2006), which was adapted such that population dynamics depended on size (stem length) and past growth rate (in stem length). In an age-size IPM, population dynamics are described as (Ellner & Rees 2006):

  • display math(eqn 1a)
  • display math(eqn 1b)

in which x is size at time t, y is size at time + 1 and Ω the set of all possible sizes. The probability density function n a (y,t) describes the state of the population of individuals of age a. F a (x,y) and P a (x,y) are the fecundity and survival-growth function, respectively, and m is the maximum age. Applying the midpoint rule (Easterling, Ellner & Dixon 2000), this model can be transformed into a set of large transition matrixes (one transition matrix per age), where Ω is now divided into very narrow size classes. In an age-size IPM, the functions F a (x,y) and P a (x,y) are based on continuous functions that relate vital rates (growth, survival and reproduction) to both size and age. Lifetime past growth rate (p) can be expressed as a function of size (x) and age (a) as inline image. Therefore, in a linear example case, vital rate (v) can be related to size and past growth as:

  • display math(eqn 2)

where α, β and γ are regression coefficients. Note that an age-term (δ*a) can be added to (eqn 2) in case age per se explains (additional) variation in vital rate v. As (eqn 2) is a function of size and age, incorporating such a function in F a (x,y) and P a (x,y) allows applying the analyses outlined by Ellner & Rees (2006). A detailed explanation and R code are included in Appendix S1 in the Supporting Information.

We used regression equations for vital rates to construct F a (x,y) and P a (x,y). We assumed no pollen limitation, and therefore, we based the model on female palms only. As we lacked data on the influence of past growth rate on the performance of seedlings and individuals <10 cm stem length, these size classes were not included. New stemmed individuals entered the model with a size distribution based on the growth rate distribution of individuals smaller than 10 cm, which was determined from the internode data, see Appendix S1. To construct F a (x,y), we averaged values of the three annual reproduction probability functions and multiplied this by the seed production function and by the average number of seedlings per seed. We applied a normal distribution in P a (x,y) to describe the variation in growth rate. As growth variation was independent of stem length or past growth rate, we used mean variation. As we did not find a significant contribution of size to survival (see 'Results'), we used the average adult survival in P a (x,y). Maximum size in the model was 1.1 times the maximum observed stem length, minimum size 0.9 times the minimum observed stem length. The maximum age was taken to be 30 years, as very few individuals exceed this age. All individuals smaller than 11 cm were considered to be non-reproductive, as we did not observe any smaller individual with flowers or fruits. Two hundred points were used when applying the midpoint rule to construct the transition matrix. To verify whether including persistent growth differences in IPMs changes population growth rate, we also constructed IPMs based on regression equations with only size as an explanatory variable.

Demographic analyses

An age-size-dependent IPM has a dominant eigenvalue, which represents the population growth rate (λ) and a right and left dominant eigenvector, which represent the stable size-age distribution and the reproductive value, respectively. To determine the relative contribution to population growth of fast growers, we conducted elasticity analysis (Ellner & Rees 2006). Elasticity values quantify the effect of a proportional change in a given transition probability or a certain size or age category, on population growth rate (λ; see Childs et al. 2003; Ellner & Rees 2006). The proportion of elasticity values accounted for by persistent fast growers therefore provides information on their contribution to λ. We distinguished fast and slow growers using quantiles in the age distribution per size class (where 50% is the median age): fast growers are the individuals below quantile age. Quantile ages were obtained from the stable age distribution of the corresponding size class. The total contribution of the fast growers was then calculated as the sum of the elasticity values per size class over all ages below quantile age (the fast growers), summed over all size classes. The same analysis was performed for the robustness test. The R-script is included in Appendix S1.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Persistent growth differences

Lifetime growth trajectories showed strong and persistent growth differences between individuals, which was first indicated by the width of the ‘fan shape’ and relatively few crossings of the growth trajectories (Fig. 1a). A formal proof of persistent growth differences was obtained from rank correlations (Fig. 1b). The rank order of individuals with regard to growth rate at a particular age was almost always significantly correlated to that of the next and further age classes. These rank correlations were high (>0.8) if growth rates were compared with the next age class and gradually decreased when comparing to further age classes (Fig. 1b). The correlations were significant up to a 26-year age difference (Fig. 1b). The rank correlation results indicate that growth rank at one age is a good predictor for growth rank at older ages, and therefore, differences in growth rate are persistent and long lasting.

image

Figure 1. (a) Reconstructed lifetime growth trajectories of 187 individuals of Chamaedorea elegans in a Mexican tropical rain forest. Each line represents one individual. The width of the fan shape and few ‘crossings’ of lines indicate the magnitude of the persistent growth differences among individuals. (b) Spearman rank correlations of growth rate between subsequent ages. Significant rank correlations (< 0.05) are shown and indicate that the rank order of individuals based on their stem growth rates at one age is maintained in the next or following ages. The dashed line delimits the ages for which correlations were conducted.

Download figure to PowerPoint

Leaf production rates were autocorrelated in time. For example, leaf production rates during the second and third year were significantly correlated (= 0.45, < 0.001). This suggests that variation in leaf production among individuals persists over time, at least to some extent. Furthermore, mean leaf production rates were also correlated with average lifetime internode length (= 0.32, < 0.001). This relation indicates that fast-growing individuals in terms of stem length (per leaf) also tend to produce more leaves per year. These results suggest that assuming constant leaf production rate over the lifetime of individuals is probably quite realistic. Nevertheless, it is useful to test how sensitive results are to changing leaf production rates to an equal value for all individuals. The results of this robustness test are included in Appendix S2.

Effect of persistent growth differences on individual fitness

The results of the multiple regression analyses for vital rates are shown in Table 1, and functions are illustrated in Fig. 2. Age (or age2) did not enter in any of the regression models and therefore was not a good predictor of vital rates. Stem growth rate was positively related to both stem length and past growth rate. Survival probability was not significantly related to stem length. For reproduction probability and seed production, past growth rate (i.e. size/age) explained most variation (positive relation) while current stem length did not appear in the regression equations. In all, these results suggest that past growth rate is a better predictor of these vital rates than stem length or age. Furthermore, the relationships between growth, probability of reproduction and seed production with past growth were strongly positive. Thus, fast growers kept on growing fast (and therefore reached the reproductive size at an earlier age) had a higher probability of reproducing and produced more seeds.

image

Figure 2. Results of multiple regression analyses in which vital rates growth (a), reproduction probability (b) and seed production (c), were related to stem length and past growth rate of Chamaedorea elegans in a Mexican tropical rain forest. Past growth rate strongly influenced current growth, reproduction probability and seed production. Regression results are included in Table 1. Note that sample sizes in 2b are larger as most individuals are represented by three dots, one for each year.

Download figure to PowerPoint

Table 1. Results of regression analyses to explain variation in vital rates of Chamaedorea elegans in a Mexican forest. Estimated coefficients, significance and amount of variation explained R 2 (Nagelkerke for logistic, Nagelkerke 1991) are shown. Age did not appear in any of the regression equations and is therefore not shown. Sample sizes are 142 individuals for survival chance and 68 individuals for growth, reproduction probability in year 1, 2 and 3 and seed production
InterceptStem length (cm)Past growth rate (cm year−1) R 2
  1. Significance levels are *< 0.05, **< 0.01, ***< 0.001; ns, non-significant; na, the variable was not included in the analysis.

Stem growth rate (cm year−1)0.413** 1.033*** 0.443*** 0.842
Survival chance (individual individual−1 3-year−1)2.382*** −0.006 nsna0.002
Reproduction probability year 1 (individual individual−1 year−1)−2.146*** ns0.480* 0.125
Reproduction probability year 2 (individual individual−1 year−1)−2.830*** ns0.687** 0.220
Reproduction probability year 3 (individual individual−1 year−1)−0.249 nsns1.094* 0.209
Seed production (seeds individual−1 year−1)5.651 nsns7.675*** 0.285

Effect of persistent growth differences on population dynamics

There was good correspondence between the stable population size structure predicted by the IPM and the structure observed in the field (Fig. 3a). This indicates that the dynamics simulated by the kernel are representative of the past dynamics of the population. The population growth rate (λ) projected by the model was 1.056, suggesting modest population growth. This value was very close to the value (1.059) obtained with a model that was based on size only.

image

Figure 3. (a) A comparison of observed population structure (symbols; category width = 2.23 cm) and predicted stable size distribution (line) of the integral projection model for Chamaedorea elegans in a Mexican tropical rain forest. (b) Contour plot showing the distribution of elasticities for survival and growth (representing 82% of total elasticity) over size and age classes. High values indicate a large contribution to population growth. The dotted line represents median age per size class. For most size classes, elasticity values tend to be highest for below-median ages, that is, persistent fast growers. (c) Idem for fecundity elasticities (18% of total elasticity).

Download figure to PowerPoint

Elasticity values calculated for our IPM could be grouped by size or by past growth rates. The latter option allowed us to separate contributions of those individuals that have reached high lifetime growth rates vs. those that were slow growers. This separation showed that fast growers contributed considerably more to population dynamics than slow growers (Figs 3b, c and 4). For example, when we summed the elasticity values of the 10% fastest-growing individuals over all size classes, we found that these accounted for 17% of the elasticity. Similarly, the 20% fastest growers contributed 29% to the elasticity (Fig. 4). When the population was divided in half, with fast and slow growers each representing 50% of the population, fast growers accounted for 64% of the elasticity. Thus, the fastest-growing 50% of the population contributed 1.8-fold more to population growth compared to the slow growers. This is also illustrated by the size-age distributions of elasticity values, showing the highest values for individuals in age classes of below-median age (Fig. 3b and c).

image

Figure 4. Relative contribution of fast growers to population growth for the understorey palm species Chamaedorea elegans in a Mexican tropical rain forest. The y = x line represents the situation in which fast growers would proportionally contribute to population growth. The left-most dot indicates that 10% of the fastest growers (i.e. 10% youngest individuals per narrow size class) account for 19% of the elasticity and therefore contribute disproportionately to population growth.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Lifetime persistence of growth differences

We measured internodes along stems and combined this information with annual leaf production rates to reconstruct lifetime growth trajectories for individuals of the tropical understorey palm C. elegans. We then showed that differences in growth rate between individuals persisted for long periods of time (up to 26 years). This strong persistent growth variation was also maintained in a robustness test, under the assumption of equal leaf production for all individuals (see Appendix S2). As only few individuals live longer than 30 years, our results imply that growth differences among individuals of understorey palm C. elegans persist over a substantial part of its lifespan.

Intriguingly, we found higher rank correlations for our study species compared to those obtained for large-stature tree species (Brienen, Zuidema & During 2006; Rozendaal & Zuidema 2011; Zuidema, Vlam & Chien 2011), and we found such correlations for a larger part of the life of individuals of our study species. The stronger persistence of growth differences in this understorey palm compared to canopy tree species studied before may reflect different causes of inter-individual variation in these groups of species. It has been proposed that growth variation in juvenile individuals of canopy tree species is determined by gap dynamics, which leads to strong spatial and temporal variation in light availability (Pfister & Stevens 2002; Brienen & Zuidema 2006). Growth of understorey palm species, on the other hand, is relatively insensitive to light variation (Chazdon 1992), as was also documented for C. elegans (Anten, Martinez-Ramos & Ackerly 2003; Martínez-Ramos, Anten & Ackerly 2009). Inter-individual growth differences in these species could be more closely associated with edaphic condition or genetic variation, which are more permanent than gap-induced shifts in light availability (Ceccon, Huante & Rincón 2006), potentially leading to more persistent growth differences. In accordance with this idea, growth differences among Cedrela odorata trees were found to be more persistent in a dry forest than in a wet forest (Brienen, Zuidema & Martínez-Ramos 2010), likely because they were caused by long-term spatial variation in water availability in the dry forest and by shorter-term spatial variation in light conditions in the wet forest. It is likely that other factors than light availability drive strong and persistent inter-individual growth differences in C. elegans and related understorey (palm) species. Therefore, overall, we would expect more persistent growth differences if these are mainly caused by relatively permanent conditions like soil water and/or nutrient availability or by genetic differences among individuals.

Past growth drives current palm performance

Past growth rate and stem length importantly govern current growth in C. elegans. Several studies on canopy tree species found similarly large contributions of past growth rates on current growth rates (Brienen, Zuidema & During 2006; Rozendaal & Zuidema 2011; Zuidema, Vlam & Chien 2011). Intriguingly, in C. elegans, seed production and reproduction probability are chiefly determined by past growth rate, without a significant effect of stem length. The strong positive correlation of stem growth and reproduction suggests that both are constrained by a third factor or combination of factors that allow the individual palm to both grow faster and reproduce more than others for long periods of time (Van Noordwijk & De Jong 1986). Palm age did not contribute significantly to explaining variation in any of the vital rates in our study species, in contradiction to what has been found for many plant species (e.g. Van Dijk 2009). This suggests that individual performance of C. elegans is not governed by age.

Super performers within populations: what causes inter-individual differences in vital rates?

We showed that inter-individual growth differences among individuals can persist for long periods of time and that these fast growers reproduce more than slow growers. This suggests the existence of super performers in our study species. Although the causes of these persistent differences in growth and reproduction are unknown, we suspect that they are associated with a combination of environmental factors (soil and topographic heterogeneity) and/or genetic differences. Most differences among individuals in animals are attributable to genetic differences (e.g. Coltman, Pilkington & Pemberton 2003), although examples of variation attributed to neutral processes are also found (Steiner & Tuljapurkar 2012). However, in plants this is likely to be different. As animals are generally able to move between (micro-)sites and search for optimal resource conditions, local variation in habitat does not necessarily lead to variation in vital rates. In contrast, sessile plants depend on the resources available at the micro-site where they grow; therefore, spatial environmental variation is more important in creating variation in individual performance for plants compared to animals (Harper 1977). Nevertheless, there are examples of genetic differences that clearly influence individual performance. For instance, in the tropical rain forest understorey palm (Astrocaryum mexicanum), strong variation in growth rates could be explained by genetic differences among individuals, with heterozygous individuals growing faster than homozygous ones (Eguiarte, Pérez-Naser & Piñero 1992). Furthermore, for the same species, it was found that palms with higher reproductive rates were spatially aggregated in spots of the forest with more light (Piñero & Sarukhán 1982), which indicates that both genetic variation and environmental factors may contribute to variation in vital rates. The role of genetic differences can relatively easily be tested using quantitative genetic experiments, in which (seeds of) fast and slow growers are grown under similar conditions. However, there is reason to assume that genetic differences are not very important in explaining persistent growth differences, as heritable traits that would allow plants to make more efficient use of the available resources would be readily selected for and lead to faster (average) growth and likely reduced variation.

In our study species, variation in light availability and soil depth explain 8% of the variation in stem growth, but there is no relation between these variables and reproductive output (Martínez-Ramos, Anten & Ackerly 2009). Thus, other factors, including soil nutrient or water availability, are probably responsible for the observed persistent growth differences. One potential factor is the presence of karst-range sites in the study area: growth and survival of canopy trees growing at these sites is much more sensitive to changes in annual rainfall than those growing on alluvial soils where soil nutrients and water availability is higher (M. Martinez-Ramos, unpubl. data). If the differences are indeed caused by environmental factors, the spatial heterogeneity in growth conditions, possibly in association with dispersal patterns, may play an important role in regulating population dynamics (Svenning 2002). This in turn implies that changes in this spatial pattern may strongly affect population growth. For example, if super performers require certain growing conditions, they may be relatively vulnerable to changes in those conditions because of habitat loss or climate change.

The importance of fast growers for population growth

When some individuals persistently grow faster than others, they may have a disproportional contribution to population growth. Compared to slow growers, fast growers reach the reproductive size at a younger age and have a higher probability of doing so. This fast-growth effect can be larger if fast-growing individuals also produce more seeds – which was the case in our study species. For C. elegans, we estimated that the contribution of fast growers (individuals below-median age of a certain size) to population growth is 1.8 times higher than that of slow growers (individuals above-median age of a certain size). A similar difference in the importance of fast and slow growers was found for the tropical forest canopy tree species Cedrela odorata (Zuidema, Brienen & During 2009). Using loop analyses in an age-size classified matrix model, they found that fast-growing juvenile individuals contributed two times more to population growth compared to slow growers.

The fast-growth effect we presented for Celegans may have been slightly overestimated as we assumed that observed differences in leaf production rates among individuals measured during a couple of years are representative of lifetime differences in leaf production. However, our main conclusion was maintained in a rigorous robustness analysis that showed, even under the most conservative assumption of all plants having equal leaf production rates, faster growers still contributed 30% more to population growth than slow growers (Appendix S2). On the other hand, the fast-growth effect of the original model may also have been an underestimate, if past growth rate is positively related to survival probability. We lacked data to test this, but such relations were recently found for many tropical tree species (Rüger et al. 2011). In all, our results show that population growth is disproportionately governed by fast-growing individuals that attain high rates of reproduction.

Implications for conservation and management

If fast growers contribute more to population growth than slow growers, it would be effective to target conservation and management efforts at this group (Zuidema, Brienen & During 2009). Differentially exploiting (leaves, fruits, entire individual) of fast and slow growers may be beneficial for the survival and recovery of harvested populations and may increase yields in the long run. Clearly, an important condition to put such recommendations in practice is the identification of these fast growers. The internodes of understorey palms allow for relatively straightforward recognition of fast growers. Thus, understorey palms are a very suitable group to test the effects of protecting fast growers to improve sustainability of harvesting practices.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Jorge Rodriguez-Velázquez and Gilberto Jamangape for technical support and logistic help. Particular gratitude goes to Etienne Langouche for assistance with data collection and data entry. Heinjo During is thanked for his valuable comments on the manuscript and Marco Visser for his help with part of the graphics. This work was supported by the National Autonomous University of Mexico PAPIIT-DGAPA (grant IN-2297507) to M.M.R. M.J. was financially supported by Stichting dr Hendrik Muller's Vaderlandsch Fonds, the K.F Hein fund, Alberta Mennega Stichting and the Miquel fund. P.A.Z. was supported by ERC grant 242955. M.M.R. acknowledges sabbatical fellowships from PASPA-DGAPA (UNAM) and CONACyT.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Anten, N.P.R. , Martinez-Ramos, M. & Ackerly, D. (2003) Compensatory growth in a tropical understory palm subjected to repeated defoliation. Ecology, 84, 29052918.
  • Brienen, R.J.W. & Zuidema, P.A. (2006) Lifetime growth patterns and ages of Bolivian rain forest trees obtained by tree ring analysis. Journal of Ecology, 94, 481493.
  • Brienen, R.J.W. , Zuidema, P.A. & During, H.J. (2006) Autocorrelated growth of tropical forest trees: unraveling patterns and quantifying consequences. Forest Ecology and Management, 237, 179190.
  • Brienen, R.J.W. , Zuidema, P.A. & Martínez-Ramos, M. (2010) Attaining the canopy in dry and moist tropical forests: strong differences in tree growth trajectories reflect variation in growing conditions. Oecologia, 163, 485496.
  • Ceccon, E. , Huante, P. & Rincón, E. (2006) Abiotic factors influencing tropical dry forests regeneration. Brazilian Archives of Biology and Technology, 49, 305312.
  • Chazdon, R.L. (1992) Photosynthetic plasticity of two rain forest shrubs across natural gap transects. Oecologia, 92, 586595.
  • Childs, D.Z. , Rees, M. , Rose, K.E. , Grubb, P.J. & Ellner, S.P. (2003) Evolution of complex flowering strategies: an age- and size-structured integral projection model. Proceedings of the Royal Society of London. Series B, Biological sciences, 270, 18291838.
  • Clark, D.A. & Clark, D.B. (1992) Life history diversity of canopy and emergent trees in a neotropical rain forest. Ecological Monographs, 62, 315344.
  • Coltman, D.W. , Pilkington, J.G. & Pemberton, J.M. (2003) Fine-scale genetic structure in a free-living ungulate population. Molecular Ecology, 12, 733742.
  • De Leo, G.A. & Gatto, M. (1995) A size and age-structured model of the European eel (Anguilla anguilla L.). Canadian Journal of Fisheries and Aquatic Sciences, 52, 13511367.
  • Easterling, M.R. , Ellner, S.P. & Dixon, P.M. (2000) Size-specific sensitivity: applying a new structured population model. Ecology, 81, 694708.
  • Eccardi, F. (2003) La palma camedora. Biodiversitas, 50, 17.
  • Eguiarte, L. , Pérez-Naser, N. & Piñero, D. (1992) Genetic structure, outcrossing rate and heterosis in Astrocaryum mexicanum (tropical palm): Implications for evolution and conservation. Heredity, 69, 217228.
  • Ellner, S.P. & Rees, M. (2006) Integral projection models for species with complex demography. American Naturalist, 167, 410419.
  • Harper, J.L. (1977) Plant Population Biology. Academic Press, London.
  • Hodel, D.R. (1992) Chamaedorea palms, the species and their cultivation. Allen press, Lawrence, Kansas, USA.
  • Ibarra-Manríquez, G. & Martínez-Ramos, M. (2002) Landscape variation of liana communities in a neotropical rain forest. Plant Ecology, 160, 91112.
  • Kohyama, T. & Hara, T. (1989) Frequency-distribution of tree growth-rate in natural forest stands. Annuals of Botany, 64, 4757.
  • Lugo, A.E. & Rivera Batlle, C.T. (1987) Leaf production, growth rate, and age of the palm Prestoea montatna in the Luquillo experimental forest, Puerto Rico. Journal of Tropical Ecology, 3, 151161.
  • Lusk, C.H. & Smith, B. (1998) Life history differences and tree species coexistence in an old-growth New Zealand rain forest. Ecology, 79, 795806.
  • Martínez-Ramos, M. , Anten, N.P.R. & Ackerly, D.D. (2009) Defoliation and ENSO effects on vital rates of an understorey tropical rain forest palm. Journal of Ecology, 97, 10501061.
  • Martorell, C. , Vega, E. & Ezcurra, E. (2006) Morphological concequences of the trade-off between growth and reproduction in a columnar cactus (Lophocereus schottii). Plant Ecology, 183, 125131.
  • Nagelkerke, N.J.D. (1991) A note on a general definition of the coefficient of determination. Biometrika, 78, 691692.
  • Oyama, K. (1992) Conservation and exploitation of tropical resources- the case of Chamaedorea palms. Evolutionary Trends in Plants, 6, 1720.
  • Pelletier, F. , Clutton-Brock, T. , Pemberton, J. , Tuljapurkar, S. & Coulson, T. (2007) The evolutionary demography of ecological change: linking trait variation and population growth. Science, 315, 15711574.
  • Pfister, C.A. & Stevens, F.R. (2002) The genesis of size variability in plants and animals. Ecology, 83, 5972.
  • Pfister, C.A. & Stevens, F.R. (2003) Individual variation and environmental stochasticity: implications for matrix model predictions. Ecology, 84, 496510.
  • Pinard, M. (1993) Impacts of stem harvesting on populations of Iriartea deltoidea (palmae) in an extractive reserve in Acre, Brazil. Biotropica, 25, 214.
  • Piñero, D. & Sarukhán, J. (1982) Reproductive behaviour and its individual variability in a tropical palm, Astrocaryum mexicanum . Journal of Ecology, 70, 461472.
  • Rozendaal, D.M.A. & Zuidema, P.A. (2011) Dendroecology in the tropics: a review. Trees, 25, 316.
  • Rüger, N. , Huth, A. , Hubbell, S.P. & Condit, R. (2011) Determinants of mortality across a tropical lowland rainforest community. Oikos, 120, 10471056.
  • Sarukhán, J. , Martínez-Ramos, M. & Piñero, D. (1984). The analysis of demographic variability at the individual level and its population consequences. Perspectives on Plant Population Ecology (eds R. Dirzo & J. Sarukhán ), pp. 83106. Sinauer Associates Inc., Sunderland.
  • Steiner, U.K. & Tuljapurkar, S. (2012) Neutral theory for life histories and individual variability in fitness components. Proceedings of the National Academy of Sciences, 109, 46844689.
  • Svenning, J.C. (2002) Crown illumination limits the population growth rate of a neotropical understorey palm (Geonoma macrostachys, arecaceae). Plant Ecology, 159, 185199.
  • Terborgh, J. , Flores, C. , Mueller, P. & Davenport, L. (1997) Estimating the ages of successional stands of tropical trees from growth increments. Journal of Tropical Ecology, 13, 833856.
  • Tomlinson, P.B. (1990) The structural biology of palms. Clarendon Press, Oxford.
  • Tuljapurkar, S. , Steiner, K.U. & Orzack, S.H. (2009) Dynamic heterogeinities in life histories. Ecology letters, 12, 93106.
  • Van Dijk, H. (2009) Ageing effects in an iteroparous plant species with a variable life span. Annals of botany, 104, 115124.
  • Van Noordwijk, A.J. & De Jong, G. (1986) Acquisition and allocation of resources: their influence on variation in life history tactics. American Naturalist, 128, 137142.
  • Vindenes, Y.S. , Engen, S. & Saether, B.E. (2008) Individual heterogenity in vital parameters and demographic stochasticity. American Naturalist, 171, 455467.
  • Zuidema, P.A. , Brienen, R.J.W. & During, H.J. (2009) Do persistently fast-growing juveniles contribute disproportionately to population growth? A new analysis tool for matrix models and its application to rainforest trees American Naturalist, 174, 709719.
  • Zuidema, P.A. & Franco, M. (2001) Integrating vital rate variability into perturbation analysis: an evaluation for matrix population models of six plant species. Journal of Ecology, 89, 9961006.
  • Zuidema, P.A. , Vlam, M. & Chien, P.D. (2011) Ages and long-term growth patterns of four threatened Vietnamese tree species. Trees, 25, 2938.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
jec2001-sup-0001-AppendixS1.docxWord document18K Appendix S1. Methods and R script for a size and past growth rate–dependent IPM and for the analysis of the relative importance of fast growers.
jec2001-sup-0002-AppendixS2.docxWord document174K Appendix S2. Results of a robustness test to evaluate the effect of leaf production rates on model output.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.