Super-performance in a palm species

Merel Jansen

Research output: Thesisinternal PhD, WU

Abstract

The world is changing rapidly due to anthropogenic disturbance. Effects include: global warming, massive pollution, a changed global nitrogen cycle, high rates of land-use change, and exotic species spread. This has a tremendous impact on both natural and agricultural systems. To understand these impacts, good understanding of ecological systems and underlying drivers is necessary. Ecological systems can be studied at different levels of aggregation. Different levels of aggregation influence each other and are also influenced by external drivers like the environment. The population level is of particular interest, because many important ecological processes occur at the population level, like evolution, extinction, and invasion. Ecologists are increasingly recognizing that population processes are strongly influenced by one level of aggregation lower, the individual level. Individual heterogeneity (i.e. differences between individuals in performance), determines many population processes including population growth rate. However, the exact relations between individual heterogeneity, the external drivers of it, and the population level are not always well understood. Furthermore, methods to analyze these relations are not always available.

Individual heterogeneity occurs at different temporal scales, ranging from short- to long-term performance differences between individuals, where short- and long-term refer to the expected lifespan of the species in question. Short-term differences between individuals are relatively easily identifiable and are common in almost all species. But long-term differences are much harder to determine especially for long-lived organisms. Long-term differences between individuals in reproduction have been identified for several animal species, and in growth for several tree species, but less is known about the existence of such differences in other life forms (e.g. palms, lianas or clonal plants). Quantifying the extent to which individuals differ is essential for understanding the influence of individual heterogeneity on population processes. Super-performing individuals (i.e. individuals that persistently grow faster and reproduce more than others), probably contribute more to the growth of the population and therefore to future generations. Future populations will, therefore, have the genetic characteristics of the super-performers. Which characteristics this will be, depends on the genetic and environmental drivers of super-performance. Full understanding of the influence of individual heterogeneity on population processes, therefore, requires knowledge of the underlying causes of individual heterogeneity.

For many species, it is known that spatial variation in environmental conditions can cause short-term performance differences between individuals, but it is often not clear if the same environmental factors that cause short-term performance differences are also the environmental factors that cause long-term performance differences. Furthermore, genetic variation is known to cause performance differences, but to what extent is not well studied in natural long-lived plant populations. Within-population genetic variation can be maintained in habitats that are characterized by strong temporal or spatial heterogeneity in environmental conditions if the performance of a genotype relative to others depends on the environment it experiences.

Super-performing individuals possibly play an important role in the resistance and resilience of populations to disturbance (i.e. maintaining and recovering population growth rate under stress), because super-performers potentially contribute more to the recovery of the population. However, this depends on the relative tolerance to disturbance of super-performers compared to under-performers. A positive relation between performance and tolerance would make super-performers more important, while a negative relation would make them less important. Many types of disturbances entail leaf loss and tolerance to leaf loss is associated with performance being larger than what one would assume based on the amount of leaf area loss. Tolerance can be achieved by compensating for leaf loss in terms of growth rate, which entails either allocating more new assimilates to leaves, allocating new assimilates more efficiently to leaf area (i.e. by increasing specific leaf area), or growing faster with existing leaf area (i.e. by increasing net assimilation rate). Genetic variation in tolerance and compensatory responses would allow populations to adapt to changes in disturbance events that entail leaf loss.

Individual heterogeneity could also have implications for management. Plant and animal populations are managed at many different levels ranging from harvest from natural populations to modern agricultural practices. When harvesting from natural populations, it might be beneficial to spare the individuals that are most important for future production. Individuals could be spared, either because they contribute most to population growth, because they are tolerant to harvesting (which is relevant when only part of a plant is harvested), or when they start producing less or lower quality product. The productivity of natural populations could also be increased by actively promoting those environmental conditions and genotypes that allow for high productivity, which is the basis of agriculture and common practice in forest management. To determine how this can best be done, knowledge of the causes of individual heterogeneity is necessary.

The general aim of this thesis is to identify and quantify the mechanisms that determine individual heterogeneity and to determine how this heterogeneity, in turn, affects population level processes. This aim was divided into four main questions that I addressed: (1) To what extent do individuals differ in performance? (2) What causes individual heterogeneity in performance? (3) What are the demographic consequences of individual heterogeneity? (4) Can individual differences be used to improve the management of populations? To answer these questions, we used the tropical forest understorey palm Chamaedorea elegans as a study system, of which the leaves are an important non-timber forest product that is being used in the floral industry worldwide. We collected demographic data, measured spatial variation in environmental conditions, and applied a defoliation treatment to simulate leaf harvesting, in a natural population in Chiapas, Mexico. Furthermore, we grew seedlings from different mothers from our study population in the greenhouses of Wageningen University, where we also applied a defoliation treatment.

In Chapter 2 we quantified the extent to which individuals differ in long-term growth rate, and analyzed the importance of fast growers for population growth. We reconstructed growth histories from internodes and showed that growth differences between individuals are very large and persistent in our study population. This led to large variation in life growth trajectories, with individuals of the same age varying strongly in size. This shows that not only in canopy trees but also in species in the light limited understorey growth differences can be very large. Past growth rate was found to be a very good predictor of current performance (i.e. growth and reproduction). Using an Integral Projection Model (i.e. a type of demographic model) that was based on size and past growth rate, we showed that fast-growing individuals are much more important for population growth than others: the 50% fastest growing individuals contributed almost two times as much to population growth as the 50% slowest growing individuals.

In Chapter 3 we analyzed the extent to which observed long-term growth differences can be caused by environmental heterogeneity. Short-term variation in performance was mainly driven by light availability, while soil variables and leaf damage had smaller effects, and spatial heterogeneity in light availability and soil pH were autocorrelated over time. Using individual-based simulation models, we analyzed the extent to which spatial environmental heterogeneity could explain observed long-term variation in growth, and showed that this could largely be explained if the temporal persistence of light availability and soil pH was taken into account. We also estimated long-term inter-individual variation in reproduction to be very large. We further analyzed the importance of temporal persistence in environmental variation for long-term performance differences, by analyzing the whole range of values of environmental persistence, and the strength of the effect of the environmental heterogeneity on short-term performance. We showed that long-term performance differences become large when either the strength of the effect of the environmental factor on short-term performance is large, or when the spatial variation in the environmental factor is persistent over time. This shows that an environmental factor that in a short-term study might have been dismissed as unimportant for long-term performance variation, might, in reality, contribute strongly.

In Chapter 4 we tested for genetic variation in growth potential, tolerance to leaf loss, compensatory growth responses, and if growth potential and tolerance were genetically correlated in our study population. We quantified compensatory responses with an iterative growth model that takes into account the timing of leaf loss. Genetic variation in growth potential was large, and plants compensated strongly for leaf loss, but genetic variation in tolerance and compensatory growth responses was very limited. Growth performances in defoliated and undefoliated conditions were positively genetically correlated (i.e. the same genotypes perform relatively well compared to others, both with and without the stress of leaf loss). The high genetic variation in growth potential and the positive correlation between treatments suggests that the existence of super-performing individuals in our study population likely has (at least in part) a genetic basis. These super-performing individuals, that grow fast even under the stress of leaf loss, possibly contribute disproportionately to population resistance and resilience to disturbance. The low genetic variation in tolerance and compensatory responses, however, suggests that populations might have limited ability to adapt to changes in disturbance regimes that entail increases in leaf loss. Furthermore, the high genetic variation in growth potential could potentially be used in management practices like enrichment planting.

In Chapter 5 we explore the potential of using individual heterogeneity to design smarter harvest schemes, by sparing individuals that contribute most to future productivity. We tested if fast and slow growers, and small and large individuals, responded differently to leaf loss in terms of vital rates, but found only very limited evidence for this. Using Integral Projection Models that were based on stem length and past growth rate, we simulated leaf harvest over a period of 20 years, in several scenarios of sparing individuals, which we compared to “Business as usual” (i.e. no individuals being spared, BAU). Sparing individuals that are most important for population growth,  was beneficial for population size (and could, therefore, reduce extinction risk), increased annual leaf harvest at the end of the simulation period, but cumulated leaf harvest over 20 years was much lower compared to BAU.  Sparing individuals that produced leaves of non-commercial size (i.e. <25cm), therefore allowing them to recover, also resulted in a lower total leaf harvest over 20 years. However, a much higher harvest (a three-fold increase) was found when only leaves of commercial size were considered. These results show that it is possible to increase yield quality and sustainability (in terms of population size) of harvesting practices, by making use of individual heterogeneity. The analytical and modeling methods that we present are applicable to any natural system from which either whole individuals, or parts of individuals, are harvested, and provide an extra tool that could be considered by managers and harvest practitioners to optimize harvest practices.

In conclusion, in this thesis, I showed that in a long-lived understorey palm growth differences are very large and persistent (Chapter 2) and that it is likely that long-term differences in reproduction are also very large (Chapter 3).  I also showed that spatial heterogeneity in environmental conditions can to a large extent explain these differences and that when evaluating the environmental drivers of individual heterogeneity, it is important to take the persistence of spatial variation into account (Chapter 3). Individual heterogeneity also is partly genetically determined. I showed that genetic variation in growth potential to be large (Chapter 4), and that fast growers keep on growing fast under the stress of leaf loss (Chapters 4,5). Therefore it is likely that genetic variation contributes to long-term differences between individuals. Genetic variation for tolerance and compensatory responses was estimated to be low (Chapter 4), suggesting that the adaptive potential of our study population to changes in disturbance events that entail leaf loss might be low.  I also showed that super-performing individuals are much more important for the growth of the population (Chapter 2) and that individuals that are important for future production could be used to improve the management of natural populations (Chapter 5).  

This study provides improved insight into the extent of individual heterogeneity in a long-lived plant species and its environmental and genetic drivers, and clearly shows the importance of individual heterogeneity and its drivers for population processes and management practices. It also presents methods on how persistent performance differences between individuals can be incorporated into demographic tools, how these can be used to analyze individual contributions to population dynamics, to extrapolate short-term to long–term environmental effects, and to analyze smart harvesting scenarios that take differences between individuals into account. These results indicate that individual heterogeneity, underlying environmental and genetic drivers, and population processes are all related. Therefore, when evaluating the effect of environmental change on population processes, and in the design of management schemes, it is important to keep these relations in mind. The methodological tools that we presented provide a means of doing this. 

Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Wageningen University
Supervisors/Advisors
  • Anten, Niels, Promotor
  • Zuidema, Pieter, Promotor
  • Bongers, Frans, Co-promotor
  • Martínez-Ramos, M., Co-promotor, External person
Award date20 Dec 2016
Place of PublicationWageningen
Publisher
Print ISBNs9789462579996
DOIs
Publication statusPublished - 2016

Fingerprint

leaves
population growth
genetic variation
environmental factors
spatial variation
leaf area
demographic statistics
understory
growers
compensatory growth
defoliation
Chamaedorea elegans
soil pH
genotype
growth performance
population size
extinction
nontimber forest products
lianas
net assimilation rate

Keywords

  • chamaedorea elegans
  • understorey
  • tropical forests
  • spatial variation
  • leaves
  • growth
  • population ecology
  • defoliation
  • genetic variation

Cite this

Jansen, M. (2016). Super-performance in a palm species. Wageningen: Wageningen University. https://doi.org/10.18174/394386
Jansen, Merel. / Super-performance in a palm species. Wageningen : Wageningen University, 2016. 193 p.
@phdthesis{04de3209dd6e42988545b1121436e757,
title = "Super-performance in a palm species",
abstract = "The world is changing rapidly due to anthropogenic disturbance. Effects include: global warming, massive pollution, a changed global nitrogen cycle, high rates of land-use change, and exotic species spread. This has a tremendous impact on both natural and agricultural systems. To understand these impacts, good understanding of ecological systems and underlying drivers is necessary. Ecological systems can be studied at different levels of aggregation. Different levels of aggregation influence each other and are also influenced by external drivers like the environment. The population level is of particular interest, because many important ecological processes occur at the population level, like evolution, extinction, and invasion. Ecologists are increasingly recognizing that population processes are strongly influenced by one level of aggregation lower, the individual level. Individual heterogeneity (i.e. differences between individuals in performance), determines many population processes including population growth rate. However, the exact relations between individual heterogeneity, the external drivers of it, and the population level are not always well understood. Furthermore, methods to analyze these relations are not always available. Individual heterogeneity occurs at different temporal scales, ranging from short- to long-term performance differences between individuals, where short- and long-term refer to the expected lifespan of the species in question. Short-term differences between individuals are relatively easily identifiable and are common in almost all species. But long-term differences are much harder to determine especially for long-lived organisms. Long-term differences between individuals in reproduction have been identified for several animal species, and in growth for several tree species, but less is known about the existence of such differences in other life forms (e.g. palms, lianas or clonal plants). Quantifying the extent to which individuals differ is essential for understanding the influence of individual heterogeneity on population processes. Super-performing individuals (i.e. individuals that persistently grow faster and reproduce more than others), probably contribute more to the growth of the population and therefore to future generations. Future populations will, therefore, have the genetic characteristics of the super-performers. Which characteristics this will be, depends on the genetic and environmental drivers of super-performance. Full understanding of the influence of individual heterogeneity on population processes, therefore, requires knowledge of the underlying causes of individual heterogeneity. For many species, it is known that spatial variation in environmental conditions can cause short-term performance differences between individuals, but it is often not clear if the same environmental factors that cause short-term performance differences are also the environmental factors that cause long-term performance differences. Furthermore, genetic variation is known to cause performance differences, but to what extent is not well studied in natural long-lived plant populations. Within-population genetic variation can be maintained in habitats that are characterized by strong temporal or spatial heterogeneity in environmental conditions if the performance of a genotype relative to others depends on the environment it experiences. Super-performing individuals possibly play an important role in the resistance and resilience of populations to disturbance (i.e. maintaining and recovering population growth rate under stress), because super-performers potentially contribute more to the recovery of the population. However, this depends on the relative tolerance to disturbance of super-performers compared to under-performers. A positive relation between performance and tolerance would make super-performers more important, while a negative relation would make them less important. Many types of disturbances entail leaf loss and tolerance to leaf loss is associated with performance being larger than what one would assume based on the amount of leaf area loss. Tolerance can be achieved by compensating for leaf loss in terms of growth rate, which entails either allocating more new assimilates to leaves, allocating new assimilates more efficiently to leaf area (i.e. by increasing specific leaf area), or growing faster with existing leaf area (i.e. by increasing net assimilation rate). Genetic variation in tolerance and compensatory responses would allow populations to adapt to changes in disturbance events that entail leaf loss. Individual heterogeneity could also have implications for management. Plant and animal populations are managed at many different levels ranging from harvest from natural populations to modern agricultural practices. When harvesting from natural populations, it might be beneficial to spare the individuals that are most important for future production. Individuals could be spared, either because they contribute most to population growth, because they are tolerant to harvesting (which is relevant when only part of a plant is harvested), or when they start producing less or lower quality product. The productivity of natural populations could also be increased by actively promoting those environmental conditions and genotypes that allow for high productivity, which is the basis of agriculture and common practice in forest management. To determine how this can best be done, knowledge of the causes of individual heterogeneity is necessary. The general aim of this thesis is to identify and quantify the mechanisms that determine individual heterogeneity and to determine how this heterogeneity, in turn, affects population level processes. This aim was divided into four main questions that I addressed: (1) To what extent do individuals differ in performance? (2) What causes individual heterogeneity in performance? (3) What are the demographic consequences of individual heterogeneity? (4) Can individual differences be used to improve the management of populations? To answer these questions, we used the tropical forest understorey palm Chamaedorea elegans as a study system, of which the leaves are an important non-timber forest product that is being used in the floral industry worldwide. We collected demographic data, measured spatial variation in environmental conditions, and applied a defoliation treatment to simulate leaf harvesting, in a natural population in Chiapas, Mexico. Furthermore, we grew seedlings from different mothers from our study population in the greenhouses of Wageningen University, where we also applied a defoliation treatment. In Chapter 2 we quantified the extent to which individuals differ in long-term growth rate, and analyzed the importance of fast growers for population growth. We reconstructed growth histories from internodes and showed that growth differences between individuals are very large and persistent in our study population. This led to large variation in life growth trajectories, with individuals of the same age varying strongly in size. This shows that not only in canopy trees but also in species in the light limited understorey growth differences can be very large. Past growth rate was found to be a very good predictor of current performance (i.e. growth and reproduction). Using an Integral Projection Model (i.e. a type of demographic model) that was based on size and past growth rate, we showed that fast-growing individuals are much more important for population growth than others: the 50{\%} fastest growing individuals contributed almost two times as much to population growth as the 50{\%} slowest growing individuals. In Chapter 3 we analyzed the extent to which observed long-term growth differences can be caused by environmental heterogeneity. Short-term variation in performance was mainly driven by light availability, while soil variables and leaf damage had smaller effects, and spatial heterogeneity in light availability and soil pH were autocorrelated over time. Using individual-based simulation models, we analyzed the extent to which spatial environmental heterogeneity could explain observed long-term variation in growth, and showed that this could largely be explained if the temporal persistence of light availability and soil pH was taken into account. We also estimated long-term inter-individual variation in reproduction to be very large. We further analyzed the importance of temporal persistence in environmental variation for long-term performance differences, by analyzing the whole range of values of environmental persistence, and the strength of the effect of the environmental heterogeneity on short-term performance. We showed that long-term performance differences become large when either the strength of the effect of the environmental factor on short-term performance is large, or when the spatial variation in the environmental factor is persistent over time. This shows that an environmental factor that in a short-term study might have been dismissed as unimportant for long-term performance variation, might, in reality, contribute strongly. In Chapter 4 we tested for genetic variation in growth potential, tolerance to leaf loss, compensatory growth responses, and if growth potential and tolerance were genetically correlated in our study population. We quantified compensatory responses with an iterative growth model that takes into account the timing of leaf loss. Genetic variation in growth potential was large, and plants compensated strongly for leaf loss, but genetic variation in tolerance and compensatory growth responses was very limited. Growth performances in defoliated and undefoliated conditions were positively genetically correlated (i.e. the same genotypes perform relatively well compared to others, both with and without the stress of leaf loss). The high genetic variation in growth potential and the positive correlation between treatments suggests that the existence of super-performing individuals in our study population likely has (at least in part) a genetic basis. These super-performing individuals, that grow fast even under the stress of leaf loss, possibly contribute disproportionately to population resistance and resilience to disturbance. The low genetic variation in tolerance and compensatory responses, however, suggests that populations might have limited ability to adapt to changes in disturbance regimes that entail increases in leaf loss. Furthermore, the high genetic variation in growth potential could potentially be used in management practices like enrichment planting. In Chapter 5 we explore the potential of using individual heterogeneity to design smarter harvest schemes, by sparing individuals that contribute most to future productivity. We tested if fast and slow growers, and small and large individuals, responded differently to leaf loss in terms of vital rates, but found only very limited evidence for this. Using Integral Projection Models that were based on stem length and past growth rate, we simulated leaf harvest over a period of 20 years, in several scenarios of sparing individuals, which we compared to “Business as usual” (i.e. no individuals being spared, BAU). Sparing individuals that are most important for population growth,  was beneficial for population size (and could, therefore, reduce extinction risk), increased annual leaf harvest at the end of the simulation period, but cumulated leaf harvest over 20 years was much lower compared to BAU.  Sparing individuals that produced leaves of non-commercial size (i.e. <25cm), therefore allowing them to recover, also resulted in a lower total leaf harvest over 20 years. However, a much higher harvest (a three-fold increase) was found when only leaves of commercial size were considered. These results show that it is possible to increase yield quality and sustainability (in terms of population size) of harvesting practices, by making use of individual heterogeneity. The analytical and modeling methods that we present are applicable to any natural system from which either whole individuals, or parts of individuals, are harvested, and provide an extra tool that could be considered by managers and harvest practitioners to optimize harvest practices. In conclusion, in this thesis, I showed that in a long-lived understorey palm growth differences are very large and persistent (Chapter 2) and that it is likely that long-term differences in reproduction are also very large (Chapter 3).  I also showed that spatial heterogeneity in environmental conditions can to a large extent explain these differences and that when evaluating the environmental drivers of individual heterogeneity, it is important to take the persistence of spatial variation into account (Chapter 3). Individual heterogeneity also is partly genetically determined. I showed that genetic variation in growth potential to be large (Chapter 4), and that fast growers keep on growing fast under the stress of leaf loss (Chapters 4,5). Therefore it is likely that genetic variation contributes to long-term differences between individuals. Genetic variation for tolerance and compensatory responses was estimated to be low (Chapter 4), suggesting that the adaptive potential of our study population to changes in disturbance events that entail leaf loss might be low.  I also showed that super-performing individuals are much more important for the growth of the population (Chapter 2) and that individuals that are important for future production could be used to improve the management of natural populations (Chapter 5).   This study provides improved insight into the extent of individual heterogeneity in a long-lived plant species and its environmental and genetic drivers, and clearly shows the importance of individual heterogeneity and its drivers for population processes and management practices. It also presents methods on how persistent performance differences between individuals can be incorporated into demographic tools, how these can be used to analyze individual contributions to population dynamics, to extrapolate short-term to long–term environmental effects, and to analyze smart harvesting scenarios that take differences between individuals into account. These results indicate that individual heterogeneity, underlying environmental and genetic drivers, and population processes are all related. Therefore, when evaluating the effect of environmental change on population processes, and in the design of management schemes, it is important to keep these relations in mind. The methodological tools that we presented provide a means of doing this. ",
keywords = "chamaedorea elegans, understorey, tropical forests, spatial variation, leaves, growth, population ecology, defoliation, genetic variation, chamaedorea elegans, onderlaag, tropische bossen, ruimtelijke variatie, bladeren, groei, populatie-ecologie, ontbladering, genetische variatie",
author = "Merel Jansen",
note = "WU thesis 6542 Includes bibliographic references. - With summary in English",
year = "2016",
doi = "10.18174/394386",
language = "English",
isbn = "9789462579996",
publisher = "Wageningen University",
school = "Wageningen University",

}

Jansen, M 2016, 'Super-performance in a palm species', Doctor of Philosophy, Wageningen University, Wageningen. https://doi.org/10.18174/394386

Super-performance in a palm species. / Jansen, Merel.

Wageningen : Wageningen University, 2016. 193 p.

Research output: Thesisinternal PhD, WU

TY - THES

T1 - Super-performance in a palm species

AU - Jansen, Merel

N1 - WU thesis 6542 Includes bibliographic references. - With summary in English

PY - 2016

Y1 - 2016

N2 - The world is changing rapidly due to anthropogenic disturbance. Effects include: global warming, massive pollution, a changed global nitrogen cycle, high rates of land-use change, and exotic species spread. This has a tremendous impact on both natural and agricultural systems. To understand these impacts, good understanding of ecological systems and underlying drivers is necessary. Ecological systems can be studied at different levels of aggregation. Different levels of aggregation influence each other and are also influenced by external drivers like the environment. The population level is of particular interest, because many important ecological processes occur at the population level, like evolution, extinction, and invasion. Ecologists are increasingly recognizing that population processes are strongly influenced by one level of aggregation lower, the individual level. Individual heterogeneity (i.e. differences between individuals in performance), determines many population processes including population growth rate. However, the exact relations between individual heterogeneity, the external drivers of it, and the population level are not always well understood. Furthermore, methods to analyze these relations are not always available. Individual heterogeneity occurs at different temporal scales, ranging from short- to long-term performance differences between individuals, where short- and long-term refer to the expected lifespan of the species in question. Short-term differences between individuals are relatively easily identifiable and are common in almost all species. But long-term differences are much harder to determine especially for long-lived organisms. Long-term differences between individuals in reproduction have been identified for several animal species, and in growth for several tree species, but less is known about the existence of such differences in other life forms (e.g. palms, lianas or clonal plants). Quantifying the extent to which individuals differ is essential for understanding the influence of individual heterogeneity on population processes. Super-performing individuals (i.e. individuals that persistently grow faster and reproduce more than others), probably contribute more to the growth of the population and therefore to future generations. Future populations will, therefore, have the genetic characteristics of the super-performers. Which characteristics this will be, depends on the genetic and environmental drivers of super-performance. Full understanding of the influence of individual heterogeneity on population processes, therefore, requires knowledge of the underlying causes of individual heterogeneity. For many species, it is known that spatial variation in environmental conditions can cause short-term performance differences between individuals, but it is often not clear if the same environmental factors that cause short-term performance differences are also the environmental factors that cause long-term performance differences. Furthermore, genetic variation is known to cause performance differences, but to what extent is not well studied in natural long-lived plant populations. Within-population genetic variation can be maintained in habitats that are characterized by strong temporal or spatial heterogeneity in environmental conditions if the performance of a genotype relative to others depends on the environment it experiences. Super-performing individuals possibly play an important role in the resistance and resilience of populations to disturbance (i.e. maintaining and recovering population growth rate under stress), because super-performers potentially contribute more to the recovery of the population. However, this depends on the relative tolerance to disturbance of super-performers compared to under-performers. A positive relation between performance and tolerance would make super-performers more important, while a negative relation would make them less important. Many types of disturbances entail leaf loss and tolerance to leaf loss is associated with performance being larger than what one would assume based on the amount of leaf area loss. Tolerance can be achieved by compensating for leaf loss in terms of growth rate, which entails either allocating more new assimilates to leaves, allocating new assimilates more efficiently to leaf area (i.e. by increasing specific leaf area), or growing faster with existing leaf area (i.e. by increasing net assimilation rate). Genetic variation in tolerance and compensatory responses would allow populations to adapt to changes in disturbance events that entail leaf loss. Individual heterogeneity could also have implications for management. Plant and animal populations are managed at many different levels ranging from harvest from natural populations to modern agricultural practices. When harvesting from natural populations, it might be beneficial to spare the individuals that are most important for future production. Individuals could be spared, either because they contribute most to population growth, because they are tolerant to harvesting (which is relevant when only part of a plant is harvested), or when they start producing less or lower quality product. The productivity of natural populations could also be increased by actively promoting those environmental conditions and genotypes that allow for high productivity, which is the basis of agriculture and common practice in forest management. To determine how this can best be done, knowledge of the causes of individual heterogeneity is necessary. The general aim of this thesis is to identify and quantify the mechanisms that determine individual heterogeneity and to determine how this heterogeneity, in turn, affects population level processes. This aim was divided into four main questions that I addressed: (1) To what extent do individuals differ in performance? (2) What causes individual heterogeneity in performance? (3) What are the demographic consequences of individual heterogeneity? (4) Can individual differences be used to improve the management of populations? To answer these questions, we used the tropical forest understorey palm Chamaedorea elegans as a study system, of which the leaves are an important non-timber forest product that is being used in the floral industry worldwide. We collected demographic data, measured spatial variation in environmental conditions, and applied a defoliation treatment to simulate leaf harvesting, in a natural population in Chiapas, Mexico. Furthermore, we grew seedlings from different mothers from our study population in the greenhouses of Wageningen University, where we also applied a defoliation treatment. In Chapter 2 we quantified the extent to which individuals differ in long-term growth rate, and analyzed the importance of fast growers for population growth. We reconstructed growth histories from internodes and showed that growth differences between individuals are very large and persistent in our study population. This led to large variation in life growth trajectories, with individuals of the same age varying strongly in size. This shows that not only in canopy trees but also in species in the light limited understorey growth differences can be very large. Past growth rate was found to be a very good predictor of current performance (i.e. growth and reproduction). Using an Integral Projection Model (i.e. a type of demographic model) that was based on size and past growth rate, we showed that fast-growing individuals are much more important for population growth than others: the 50% fastest growing individuals contributed almost two times as much to population growth as the 50% slowest growing individuals. In Chapter 3 we analyzed the extent to which observed long-term growth differences can be caused by environmental heterogeneity. Short-term variation in performance was mainly driven by light availability, while soil variables and leaf damage had smaller effects, and spatial heterogeneity in light availability and soil pH were autocorrelated over time. Using individual-based simulation models, we analyzed the extent to which spatial environmental heterogeneity could explain observed long-term variation in growth, and showed that this could largely be explained if the temporal persistence of light availability and soil pH was taken into account. We also estimated long-term inter-individual variation in reproduction to be very large. We further analyzed the importance of temporal persistence in environmental variation for long-term performance differences, by analyzing the whole range of values of environmental persistence, and the strength of the effect of the environmental heterogeneity on short-term performance. We showed that long-term performance differences become large when either the strength of the effect of the environmental factor on short-term performance is large, or when the spatial variation in the environmental factor is persistent over time. This shows that an environmental factor that in a short-term study might have been dismissed as unimportant for long-term performance variation, might, in reality, contribute strongly. In Chapter 4 we tested for genetic variation in growth potential, tolerance to leaf loss, compensatory growth responses, and if growth potential and tolerance were genetically correlated in our study population. We quantified compensatory responses with an iterative growth model that takes into account the timing of leaf loss. Genetic variation in growth potential was large, and plants compensated strongly for leaf loss, but genetic variation in tolerance and compensatory growth responses was very limited. Growth performances in defoliated and undefoliated conditions were positively genetically correlated (i.e. the same genotypes perform relatively well compared to others, both with and without the stress of leaf loss). The high genetic variation in growth potential and the positive correlation between treatments suggests that the existence of super-performing individuals in our study population likely has (at least in part) a genetic basis. These super-performing individuals, that grow fast even under the stress of leaf loss, possibly contribute disproportionately to population resistance and resilience to disturbance. The low genetic variation in tolerance and compensatory responses, however, suggests that populations might have limited ability to adapt to changes in disturbance regimes that entail increases in leaf loss. Furthermore, the high genetic variation in growth potential could potentially be used in management practices like enrichment planting. In Chapter 5 we explore the potential of using individual heterogeneity to design smarter harvest schemes, by sparing individuals that contribute most to future productivity. We tested if fast and slow growers, and small and large individuals, responded differently to leaf loss in terms of vital rates, but found only very limited evidence for this. Using Integral Projection Models that were based on stem length and past growth rate, we simulated leaf harvest over a period of 20 years, in several scenarios of sparing individuals, which we compared to “Business as usual” (i.e. no individuals being spared, BAU). Sparing individuals that are most important for population growth,  was beneficial for population size (and could, therefore, reduce extinction risk), increased annual leaf harvest at the end of the simulation period, but cumulated leaf harvest over 20 years was much lower compared to BAU.  Sparing individuals that produced leaves of non-commercial size (i.e. <25cm), therefore allowing them to recover, also resulted in a lower total leaf harvest over 20 years. However, a much higher harvest (a three-fold increase) was found when only leaves of commercial size were considered. These results show that it is possible to increase yield quality and sustainability (in terms of population size) of harvesting practices, by making use of individual heterogeneity. The analytical and modeling methods that we present are applicable to any natural system from which either whole individuals, or parts of individuals, are harvested, and provide an extra tool that could be considered by managers and harvest practitioners to optimize harvest practices. In conclusion, in this thesis, I showed that in a long-lived understorey palm growth differences are very large and persistent (Chapter 2) and that it is likely that long-term differences in reproduction are also very large (Chapter 3).  I also showed that spatial heterogeneity in environmental conditions can to a large extent explain these differences and that when evaluating the environmental drivers of individual heterogeneity, it is important to take the persistence of spatial variation into account (Chapter 3). Individual heterogeneity also is partly genetically determined. I showed that genetic variation in growth potential to be large (Chapter 4), and that fast growers keep on growing fast under the stress of leaf loss (Chapters 4,5). Therefore it is likely that genetic variation contributes to long-term differences between individuals. Genetic variation for tolerance and compensatory responses was estimated to be low (Chapter 4), suggesting that the adaptive potential of our study population to changes in disturbance events that entail leaf loss might be low.  I also showed that super-performing individuals are much more important for the growth of the population (Chapter 2) and that individuals that are important for future production could be used to improve the management of natural populations (Chapter 5).   This study provides improved insight into the extent of individual heterogeneity in a long-lived plant species and its environmental and genetic drivers, and clearly shows the importance of individual heterogeneity and its drivers for population processes and management practices. It also presents methods on how persistent performance differences between individuals can be incorporated into demographic tools, how these can be used to analyze individual contributions to population dynamics, to extrapolate short-term to long–term environmental effects, and to analyze smart harvesting scenarios that take differences between individuals into account. These results indicate that individual heterogeneity, underlying environmental and genetic drivers, and population processes are all related. Therefore, when evaluating the effect of environmental change on population processes, and in the design of management schemes, it is important to keep these relations in mind. The methodological tools that we presented provide a means of doing this. 

AB - The world is changing rapidly due to anthropogenic disturbance. Effects include: global warming, massive pollution, a changed global nitrogen cycle, high rates of land-use change, and exotic species spread. This has a tremendous impact on both natural and agricultural systems. To understand these impacts, good understanding of ecological systems and underlying drivers is necessary. Ecological systems can be studied at different levels of aggregation. Different levels of aggregation influence each other and are also influenced by external drivers like the environment. The population level is of particular interest, because many important ecological processes occur at the population level, like evolution, extinction, and invasion. Ecologists are increasingly recognizing that population processes are strongly influenced by one level of aggregation lower, the individual level. Individual heterogeneity (i.e. differences between individuals in performance), determines many population processes including population growth rate. However, the exact relations between individual heterogeneity, the external drivers of it, and the population level are not always well understood. Furthermore, methods to analyze these relations are not always available. Individual heterogeneity occurs at different temporal scales, ranging from short- to long-term performance differences between individuals, where short- and long-term refer to the expected lifespan of the species in question. Short-term differences between individuals are relatively easily identifiable and are common in almost all species. But long-term differences are much harder to determine especially for long-lived organisms. Long-term differences between individuals in reproduction have been identified for several animal species, and in growth for several tree species, but less is known about the existence of such differences in other life forms (e.g. palms, lianas or clonal plants). Quantifying the extent to which individuals differ is essential for understanding the influence of individual heterogeneity on population processes. Super-performing individuals (i.e. individuals that persistently grow faster and reproduce more than others), probably contribute more to the growth of the population and therefore to future generations. Future populations will, therefore, have the genetic characteristics of the super-performers. Which characteristics this will be, depends on the genetic and environmental drivers of super-performance. Full understanding of the influence of individual heterogeneity on population processes, therefore, requires knowledge of the underlying causes of individual heterogeneity. For many species, it is known that spatial variation in environmental conditions can cause short-term performance differences between individuals, but it is often not clear if the same environmental factors that cause short-term performance differences are also the environmental factors that cause long-term performance differences. Furthermore, genetic variation is known to cause performance differences, but to what extent is not well studied in natural long-lived plant populations. Within-population genetic variation can be maintained in habitats that are characterized by strong temporal or spatial heterogeneity in environmental conditions if the performance of a genotype relative to others depends on the environment it experiences. Super-performing individuals possibly play an important role in the resistance and resilience of populations to disturbance (i.e. maintaining and recovering population growth rate under stress), because super-performers potentially contribute more to the recovery of the population. However, this depends on the relative tolerance to disturbance of super-performers compared to under-performers. A positive relation between performance and tolerance would make super-performers more important, while a negative relation would make them less important. Many types of disturbances entail leaf loss and tolerance to leaf loss is associated with performance being larger than what one would assume based on the amount of leaf area loss. Tolerance can be achieved by compensating for leaf loss in terms of growth rate, which entails either allocating more new assimilates to leaves, allocating new assimilates more efficiently to leaf area (i.e. by increasing specific leaf area), or growing faster with existing leaf area (i.e. by increasing net assimilation rate). Genetic variation in tolerance and compensatory responses would allow populations to adapt to changes in disturbance events that entail leaf loss. Individual heterogeneity could also have implications for management. Plant and animal populations are managed at many different levels ranging from harvest from natural populations to modern agricultural practices. When harvesting from natural populations, it might be beneficial to spare the individuals that are most important for future production. Individuals could be spared, either because they contribute most to population growth, because they are tolerant to harvesting (which is relevant when only part of a plant is harvested), or when they start producing less or lower quality product. The productivity of natural populations could also be increased by actively promoting those environmental conditions and genotypes that allow for high productivity, which is the basis of agriculture and common practice in forest management. To determine how this can best be done, knowledge of the causes of individual heterogeneity is necessary. The general aim of this thesis is to identify and quantify the mechanisms that determine individual heterogeneity and to determine how this heterogeneity, in turn, affects population level processes. This aim was divided into four main questions that I addressed: (1) To what extent do individuals differ in performance? (2) What causes individual heterogeneity in performance? (3) What are the demographic consequences of individual heterogeneity? (4) Can individual differences be used to improve the management of populations? To answer these questions, we used the tropical forest understorey palm Chamaedorea elegans as a study system, of which the leaves are an important non-timber forest product that is being used in the floral industry worldwide. We collected demographic data, measured spatial variation in environmental conditions, and applied a defoliation treatment to simulate leaf harvesting, in a natural population in Chiapas, Mexico. Furthermore, we grew seedlings from different mothers from our study population in the greenhouses of Wageningen University, where we also applied a defoliation treatment. In Chapter 2 we quantified the extent to which individuals differ in long-term growth rate, and analyzed the importance of fast growers for population growth. We reconstructed growth histories from internodes and showed that growth differences between individuals are very large and persistent in our study population. This led to large variation in life growth trajectories, with individuals of the same age varying strongly in size. This shows that not only in canopy trees but also in species in the light limited understorey growth differences can be very large. Past growth rate was found to be a very good predictor of current performance (i.e. growth and reproduction). Using an Integral Projection Model (i.e. a type of demographic model) that was based on size and past growth rate, we showed that fast-growing individuals are much more important for population growth than others: the 50% fastest growing individuals contributed almost two times as much to population growth as the 50% slowest growing individuals. In Chapter 3 we analyzed the extent to which observed long-term growth differences can be caused by environmental heterogeneity. Short-term variation in performance was mainly driven by light availability, while soil variables and leaf damage had smaller effects, and spatial heterogeneity in light availability and soil pH were autocorrelated over time. Using individual-based simulation models, we analyzed the extent to which spatial environmental heterogeneity could explain observed long-term variation in growth, and showed that this could largely be explained if the temporal persistence of light availability and soil pH was taken into account. We also estimated long-term inter-individual variation in reproduction to be very large. We further analyzed the importance of temporal persistence in environmental variation for long-term performance differences, by analyzing the whole range of values of environmental persistence, and the strength of the effect of the environmental heterogeneity on short-term performance. We showed that long-term performance differences become large when either the strength of the effect of the environmental factor on short-term performance is large, or when the spatial variation in the environmental factor is persistent over time. This shows that an environmental factor that in a short-term study might have been dismissed as unimportant for long-term performance variation, might, in reality, contribute strongly. In Chapter 4 we tested for genetic variation in growth potential, tolerance to leaf loss, compensatory growth responses, and if growth potential and tolerance were genetically correlated in our study population. We quantified compensatory responses with an iterative growth model that takes into account the timing of leaf loss. Genetic variation in growth potential was large, and plants compensated strongly for leaf loss, but genetic variation in tolerance and compensatory growth responses was very limited. Growth performances in defoliated and undefoliated conditions were positively genetically correlated (i.e. the same genotypes perform relatively well compared to others, both with and without the stress of leaf loss). The high genetic variation in growth potential and the positive correlation between treatments suggests that the existence of super-performing individuals in our study population likely has (at least in part) a genetic basis. These super-performing individuals, that grow fast even under the stress of leaf loss, possibly contribute disproportionately to population resistance and resilience to disturbance. The low genetic variation in tolerance and compensatory responses, however, suggests that populations might have limited ability to adapt to changes in disturbance regimes that entail increases in leaf loss. Furthermore, the high genetic variation in growth potential could potentially be used in management practices like enrichment planting. In Chapter 5 we explore the potential of using individual heterogeneity to design smarter harvest schemes, by sparing individuals that contribute most to future productivity. We tested if fast and slow growers, and small and large individuals, responded differently to leaf loss in terms of vital rates, but found only very limited evidence for this. Using Integral Projection Models that were based on stem length and past growth rate, we simulated leaf harvest over a period of 20 years, in several scenarios of sparing individuals, which we compared to “Business as usual” (i.e. no individuals being spared, BAU). Sparing individuals that are most important for population growth,  was beneficial for population size (and could, therefore, reduce extinction risk), increased annual leaf harvest at the end of the simulation period, but cumulated leaf harvest over 20 years was much lower compared to BAU.  Sparing individuals that produced leaves of non-commercial size (i.e. <25cm), therefore allowing them to recover, also resulted in a lower total leaf harvest over 20 years. However, a much higher harvest (a three-fold increase) was found when only leaves of commercial size were considered. These results show that it is possible to increase yield quality and sustainability (in terms of population size) of harvesting practices, by making use of individual heterogeneity. The analytical and modeling methods that we present are applicable to any natural system from which either whole individuals, or parts of individuals, are harvested, and provide an extra tool that could be considered by managers and harvest practitioners to optimize harvest practices. In conclusion, in this thesis, I showed that in a long-lived understorey palm growth differences are very large and persistent (Chapter 2) and that it is likely that long-term differences in reproduction are also very large (Chapter 3).  I also showed that spatial heterogeneity in environmental conditions can to a large extent explain these differences and that when evaluating the environmental drivers of individual heterogeneity, it is important to take the persistence of spatial variation into account (Chapter 3). Individual heterogeneity also is partly genetically determined. I showed that genetic variation in growth potential to be large (Chapter 4), and that fast growers keep on growing fast under the stress of leaf loss (Chapters 4,5). Therefore it is likely that genetic variation contributes to long-term differences between individuals. Genetic variation for tolerance and compensatory responses was estimated to be low (Chapter 4), suggesting that the adaptive potential of our study population to changes in disturbance events that entail leaf loss might be low.  I also showed that super-performing individuals are much more important for the growth of the population (Chapter 2) and that individuals that are important for future production could be used to improve the management of natural populations (Chapter 5).   This study provides improved insight into the extent of individual heterogeneity in a long-lived plant species and its environmental and genetic drivers, and clearly shows the importance of individual heterogeneity and its drivers for population processes and management practices. It also presents methods on how persistent performance differences between individuals can be incorporated into demographic tools, how these can be used to analyze individual contributions to population dynamics, to extrapolate short-term to long–term environmental effects, and to analyze smart harvesting scenarios that take differences between individuals into account. These results indicate that individual heterogeneity, underlying environmental and genetic drivers, and population processes are all related. Therefore, when evaluating the effect of environmental change on population processes, and in the design of management schemes, it is important to keep these relations in mind. The methodological tools that we presented provide a means of doing this. 

KW - chamaedorea elegans

KW - understorey

KW - tropical forests

KW - spatial variation

KW - leaves

KW - growth

KW - population ecology

KW - defoliation

KW - genetic variation

KW - chamaedorea elegans

KW - onderlaag

KW - tropische bossen

KW - ruimtelijke variatie

KW - bladeren

KW - groei

KW - populatie-ecologie

KW - ontbladering

KW - genetische variatie

U2 - 10.18174/394386

DO - 10.18174/394386

M3 - internal PhD, WU

SN - 9789462579996

PB - Wageningen University

CY - Wageningen

ER -

Jansen M. Super-performance in a palm species. Wageningen: Wageningen University, 2016. 193 p. https://doi.org/10.18174/394386