SEARCH

SEARCH BY CITATION

Keywords:

  • Chamaedorea palms;
  • drought;
  • leaf harvesting;
  • Mexico;
  • non-timber forest products;
  • plant population and community dynamics

Summary

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

1. Perennial plants often endure chronic loss of leaf area due to recurrent physical damage, herbivory and, for species used as non-timber forest products, due to leaf harvesting. However, little is known about functional and demographic resilience (extent and speed of recovery) of plants subjected to varying levels of chronic defoliation.

2. We used a dioecious, understorey palm (Chamaedorea elegans) to evaluate temporal trajectories and rates of recovery of leaf functional traits and vital rates (survival, growth and reproduction) after being subjected to experimental chronic defoliation regimes.

3. Pristine populations of mature C. elegans, categorized by gender (male and female), were subjected to five defoliation levels (0%, 33%, 50%, 66% or 100% of newly produced leaves) every 6 months over a period of 3 years (1997–2000). To evaluate recovery from defoliation, surviving palms were monitored for 3 years after the cessation of the defoliation treatment (2000–2003). We recorded leaf functional traits (leaf persistence, leaf production rate, leaf size and leaf area) and annual rates of mortality, growth and reproduction.

4. Cumulative effects of chronic defoliation concomitantly reduced leaf traits, survival, growth and reproduction, and this effect was stronger in female than in male palms, independent of plant size. Recovery from defoliation was faster in males than in females, but proceeded gradually overall. Survival increased first, followed by growth, while reproductive traits showed the slowest recovery rate. Recovery was independent of plant size. Notably, 3 years after defoliation treatment, the standing leaf area and probability of reproduction had not recovered to pre-defoliation levels. Additionally, we found that the occurrence of a severe drought in the first year (2000) after defoliation ceased led to decreased survival, growth and reproduction and the ability of plants to recover from defoliation.

5.Synthesis. Chronic defoliation reduces fitness components of C. elegans palms differentially between genders. Recovery is gradual and is slower and less complete in females compared with males. The lower level of resilience to chronic defoliation shown by female plants may have profound consequences for the dynamics and genetic variability of populations of tropical understorey plants undergoing prolonged defoliation. Such effects may be aggravated by severe drought episodes that are expected to increase in frequency according to global climate change predictions.


Introduction

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

During their life, long-lived plants usually undergo several defoliation events due to recurrent biotic (e.g. due to herbivores and pathogens) or physical damage (e.g. due to wind, fire, lightning and mechanical traumas). Leaves of many wild plant species, particularly palms, are also repeatedly harvested (Endress, Gorchov & Noble 2004). Depending on the severity and frequency of leaf loss, negative effects on fitness components may occur (Sampaio & Scariot 2010). While the recovery of perennial plants from single defoliation events has been clearly documented, much less is known about the extent and rate at which functional and demographic attributes can recover from multiple defoliation events occurring at different magnitudes (Belsky et al. 1993).

Plants can compensate for leaf area losses by adjusting the balance between photosynthesis and respiration, reducing resource allocation to reproduction and/or mobilizing stored reserves from roots or other storage organs to the production of new leaves (McNaughton 1983; Belsky et al. 1993; Cunningham 1997; Anten, Martinez-Ramos & Ackerly 2003). This compensatory ability enables plants to mitigate the negative impact of defoliation on survival and growth. However, under repeated defoliation, the capacity for compensation can be reduced or lost as stored reserves are depleted, which may result in reductions in both growth and survival (Canham et al. 1999; Martínez-Ramos, Anten & Ackerly 2009). The degree and pattern of recovery in functional and demographic rates once defoliation has ended is still poorly understood. Assuming that carbon limitation would be the main limiting factor on recovery, the following general pattern could be expected: (i) allocation of remaining resources and new photosynthetic products (carbon, energy and nutrients) to increase leaf area growth, (ii) allocational shifts should lead to enhanced plant growth and (iii) reproductive activity resumes once the plant attains enough photosynthetic reserves.

The time involved in this recovery process has not been studied, but we can expect that the recovery rate will depend on the plant function under consideration and the magnitude of defoliation to which plants were exposed. Some studies have shown that recovery rates from single defoliation events decrease with the magnitude of leaf area loss and that larger plants may recover faster than smaller ones, possibly due to greater pools of stored reserves (Marquis 1984; Boege 2005). Also, recovery rates may depend on resource availability at the growing site (Hawkes & Sullivan 2001), species habit, plant phenology, ontogenetic stage or the presence of specialized storage organs (Belsky et al. 1993; Hawkes & Sullivan 2001; Boege 2005; Lapointe et al. 2010). In the case of dioecious species, chronic defoliation might have gender-specific effects on fitness, due to gender differences in resource allocation to reproduction, growth patterns and defence against herbivores (Cepeda-Cornejo & Dirzo 2010). For example, in understorey palms, it has been found that females have larger carbon investment to reproduction (Oyama & Dirzo 1988), grow more slowly and are often better defended than males (Cepeda-Cornejo & Dirzo 2010). Due to these differences, it can be expected that recovery rates would differ between female and male plants, with females taking more time to recover their reproductive activity than males due to the higher energetic costs involved in the female reproductive function (Obeso 2002).

The occurrence of stochastic events, such as those imposed by climatic factors (e.g. severe drought events as those caused by El Niño Southern Oscillation (ENSO), in Mesoamerica), may aggravate the effects of defoliation (Martínez-Ramos, Anten & Ackerly 2009) and likely the ability to recover from chronic leaf area losses. Overall, while there are several potential intrinsic and extrinsic factors that may affect the rate of recovery from chronic defoliation, there is still little knowledge on how these factors operate and on the long-term functional and demographic consequences of such disturbances. For example, it remains unclear whether recovery of vital rates (survival, growth and reproduction) occurs along different temporal trajectories and whether there are defoliation thresholds beyond which the plants can no longer recover (i.e. loss of resilience). Understanding these issues is important from an evolutionary and ecological perspective, as well as to design sustainable management programmes for species whose leaves are harvested as non-timber forest products.

Many palm species are important as non-timber forest products in tropical regions. Specifically, several are commercially valued for their leaves and are subjected to intensive repeated leaf harvesting (Bridgewater et al. 2006; Zuidema, De Kroon & Werger 2007; Lopez-Toledo, Horn & Endress 2011). The simple growth form of palms (one or few monopodic stems with a single cluster of leaves) makes them excellent model species to assess demographic consequences and resilience to chronic defoliation in plant populations (Endress, Gorchov & Noble 2004). In this study, we used the understorey, dioecious, neotropical palm Chamaedorea elegans to evaluate temporal trajectories and rates of recovery of functional (leaf persistence, leaf production rate, leaf size and plant leaf area) and demographic traits (survival, growth and reproduction) after being subjected to different levels of experimental chronic defoliation regimes. To our knowledge, this is the first study describing the cumulative effects of chronic defoliation on the recovery process following such defoliation. In the previous studies, we have evaluated mechanisms of compensation in response to defoliation (Anten & Ackerly 2001; Anten, Martinez-Ramos & Ackerly 2003) and examined the demographic behaviour of C. elegans under contrasting chronic defoliation regimes (Martínez-Ramos, Anten & Ackerly 2009). Here, we assess the recovery capacity of the palms, as a function of the intensity of defoliation (percentage of leaf area loss), gender and stem size over 3 years following the end of defoliation. Specifically, we tested the following hypotheses: (i) the rate of recovery declines with the magnitude (percentage of leaf area loss) of chronic defoliation, (ii) males recover faster than females, (iii) plant size determines responses to chronic defoliation and recovery, (iv) recovery is a gradual, stepwise process with leaf traits being the first to recover to pre-defoliation levels, followed by growth, and finally reproductive performance, with females showing a more pronounced delay in reproductive recovery. Finally, we explored the effect of a severe drought on the recovery process.

Material and methods

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

Study site

This study was conducted at the Chajul Biological Field Station, within the Montes Azules Biosphere Reserve in Chiapas, Mexico (l6°06′ N, 90°56′ W). Dominant vegetation in this region is lowland evergreen tropical forest. Precipitation and temperature were quite variable during the study period (1999–2003; Fig. S1); a dry season, with considerably less rainfall than the long-term average, occurred in 2000 (January–May). This rainfall shortage was similar to that in 1998, when a severe ENSO event occurred (Martínez-Ramos, Anten & Ackerly 2009).

Study species

Chamaedorea elegans Willd (Arecaceae) is a solitary and dioecious understorey palm species native to south-east Mexico, Guatemala and Belize (Anderson, Galeano & Bernal 1997). In Chajul, C. elegans grows up to 1.5 m in height and is restricted to the karst-range areas (300–700 m asl). Chamaedorea elegans represents one of the most commercially important NTFP’s in SE Mexico, Guatemala and Belize, as their leaves and fruits are collected and exported for the European and US floral industry (Hodel 1992). However, due to overexploitation, this species has become scarce in some areas (Sánchez-Carrillo & Valtierra-Pacheco 2003). For a detailed description of the species, see Martínez-Ramos, Anten & Ackerly (2009).

Experimental system

Pristine natural populations of C. elegans within Montes Azules reserve were selected. These populations were subjected to experimental harvesting every 6 months over 3 years (1997–2000). Plants were selected and stratified by initial height and gender categories (male and female) and then randomly assigned to one of the five following defoliation treatments: 0%, 33% (one from every three leaves), 50% (two of every four leaves), 66% (two of every three leaves) and 100% (all leaves). There were 190–196 replicates per treatment for the 0%, 33%, 50 and 66% defoliation levels. The first defoliation was conducted in March 1997, and the final event was in March 2000 after which the palms were left untouched (details in Anten, Martinez-Ramos & Ackerly 2003 and Martínez-Ramos, Anten & Ackerly 2009).

Previously published work described the immediate effects of defoliation from 1997 to 1999, while in this study, we evaluate recovery of the surviving palms, over six subsequent censuses, from October 2000 to March 2003. Data for 2000 represent the final year under defoliation and are presented as ‘Cumulative effects of chronic defoliation’, while data from 2001 (Recovery-Yr1) to 2003 (Recovery-Yr3) are presented as ‘Recovery from defoliation’.

At each census date, we recorded mortality and, for surviving individuals, gender, stem length, number of standing leaves, number of persistent leaves (live leaves produced in previous years), leaf production rate, length of the most recent fully expanded leaf and reproductive activity (inflorescence, infructescence and fruit production). As another measure of reproductive performance for females, we quantified the proportion of inflorescences that matured into infructescences. For data analysis, a palm’s reproductive status (female and male) was based on cumulative observations made throughout the 6 years of study.

Data analysis

We carried out analyses focusing on two issues: (i) to evaluate the cumulative effects of chronic defoliation regimes (i.e. 3 years after the first defoliation event) and (ii) to assess the ability of palms to recover from the negative effects of such regimes (over the course of 3 years since the last defoliation event).

We included 11 different functional and demographic traits as response variables for analyses. Functional traits included five leaf attributes: (i) leaf persistence (number of persistent leaves produced in previous year), (ii) leaf production rate (newly produced leaves per palm year−1), (iii) total standing number of leaves, (iv) leaf size (length) and (v) total leaf area per palm. Leaf length refers to the distance from the base to the tip of the lamina of the newest leaf of each palm, and total leaf area was estimated from an allometric equation previously obtained, based on the number of leaves and the lamina length of each leaf (Anten & Ackerly 2001). Demographic traits included mortality, growth and reproduction. Mortality was analysed as a binary response variable and expressed as an annual mortality rate (ind ind−1 year−1). Growth was quantified as the annual extension rate of the stem (cm ind−1 year−1). Reproduction was analysed through four traits: (i) the probability of reproduction, that is, the proportion of total female or male palms reproducing (flowers production) each year (ind ind−1 year−1), (ii) annual inflorescence production per female or male palm (inflorescences ind−1 year−1), (iii) the proportion of inflorescences that mature into infructescence per female palm and (iv) fruit production per reproductive female (fruits reproductive ind−1 year−1).

The extent to which defoliated plants were able to recover was analysed by comparing the mean trait response for a given defoliation treatment to that of the control palms of the same gender. We considered a trait fully recovered from defoliation, when the mean value for the defoliated treatment was not significantly different from that of the controls (based on analysis below). The speed of recovery was quantified by determining the number of years it took for full recovery, and the possible outcomes were 1, 2 and 3 years after defoliation stopped.

We conducted two separate statistical analyses to assess the cumulative effects of the defoliation treatments on the response variables and to assess the recovery from the treatments. For the first analysis, we tested effects due to gender (‘G’ with two levels: females and males), defoliation treatment (‘DT’ with four levels: 0%, 33%, 50% and 66%), initial stem length (‘SL97’), which represent plant size at the beginning of the defoliation experiment (in the year 1997), and all the interactions among these variables, considered as fixed factors. The analyses were conducted using linear models (LM) for continuous response variables (e.g. stem growth, leaf length and total leaf area) and generalized linear models (GLM) for count (e.g. leaf persistence, inflorescence and fruit production) and binomial (e.g. probability of mortality, proportion of infrutescences/inflorescences, probability of reproduction) response variables (Crawley 2007).

For the second analysis, we also tested the effects of gender and defoliation treatment, but we used stem length at the year 2000 (SL00), which represents plant size at the beginning of the recovery period. We also included the factor year (Yr) as another variable to assess differences among years (2001, Recovery-Yr1 to 2003, Recovery-Yr3), and all interactions among these four independent variables were considered as fixed factors. In this case, due to the repeated measures design, we conducted linear mixed models (LMM) for the continuous variables and generalized LMM for the count and binomial variables. The repeated measures of every year on individuals were included as a random factor (Pinheiro & Bates 2001).

For LM and LMM analyses, when required, response variables were log(x) or log(x + 1) transformed to meet normality criteria. For count variables, a Poisson error and a logarithmic link function were used, while for binomial variables, a binomial error and a logistic link function were applied. For GLMs, when overdispersion problems occurred, a proper rescaled model was used (Crawley 2007; Everitt & Hothorn 2010). All analyses were completed in R 2.11.1 (R Development Core Team 2010).

Results

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

Results for leaf traits are presented first, followed by mortality, growth and reproduction. For each set of characteristics, we first show the cumulative effects of defoliation as reflected in the 2000 final harvest data (Table 1) and subsequently present the results for recovery (Table 2). The charts presented in the results section highlight the main effects of defoliation treatment, gender and years. Additional charts showing the relationship between the attributes analysed and stem length are presented in the Supporting information.

Table 1.   Results of linear models (LM) and generalized linear models (GLM) used to assess the cumulative effects of chronic defoliation on Chamaedorea elegans in south-eastern Mexico. The terms tested in the models were gender (G), defoliation treatment (DT), initial stem length (SL97) and the interaction among these terms
 GDTG*DT
  1. GLM (leaf persistency, leaf production rate, total leaves and mortality) and LM (leaf length, total leaf area and growth rate) statistics are provided (F values for LM and χ2 for GLM, degrees of freedom in brackets); significance level: *≤ 0.05; **< 0.01; ***< 0.001; ns, non-significant. Only significant terms were included in the minimal adequate model, – indicates that this term was removed from the model. For analysis of female-exclusive variables (infructescences/inflorescences ratio and fruit production), the term gender was not tested.

Functional traits
 Leaf persistency3.6* (1)50.4*** (3)ns
 Leaf product rate5.3* (1)7.8* (3)ns
 Total leaves3.6 (3)28.3*** (3)ns
 Leaf length6.5* (1,609)24.4*** (1,609)3.1* (1,609)
 Total leaf area6.5* (1,609)24.1*** (1,609)3.1* (3,609)
Demographic traits
 Mortality6.4** (1)8.1* (3)ns
 Growth rate4.1* (1,595)3.9* (3,595)ns
Functional traits
 Probability of reproductionns39.2*** (3)ns
 Inflorescences production4.5* (1)7.9* (3)ns
 Infrutescences/inflorescences ratio 8.9* (3) 
 Fruit production 23.6** (3) 
Table 2.   Results of linear mixed models (LMM) and generalized linear mixed models (GLMM) used to assess the recovery from chronic defoliation of functional and demographic traits of Chamaedorea elegans in south-eastern Mexico. The terms tested in the models were gender (G), defoliation treatment (DT), year (Yr), stem length after defoliation (SL00) and the interactions among these terms. Individual within year was included in the model as random factor to account for the repeated measurements
 Factors
GDTSLYrG:DTDT:YrG:YrDT:SLYr:SL
  1. GLMMs (leaf persistency, leaf production rate, total leaves and mortality) and LMMs (leaf length, total leaf area and growth rate) statistics are provided (F values for LM and χ2 values for GLMM; degrees of freedom in brackets); significance level: *≤ 0.05; **< 0.01; ***< 0.001; ns, non-significant). The table presents only significant terms. The infrutescences/inflorescences ratio and fruit production were only tested in females.

Functional traits
 Leaf persistency3.6* (1)37.9*** (3)4.1* (1)nsnsnsns14.5** (3)ns
 Leaf product rate10.2*** (1)7.8* (3)ns19.4*** (2)nsnsnsnsns
 Total leaves19.6*** (1)47.2*** (3)4.3* (1)nsnsnsnsnsns
 Leaf lengthns13.9*** (3,1604)ns11.8*** (2,1604)12.1*** (3,1604)2.5* (6,1604)4.7* (2,1604)6.9* (3)ns
 Total leaf area7.5** (1,1604)19.1** (3,1604)ns4.1* (2,1604)2.7* (3,1604)4.6*** (6,1604)nsnsns
Demographic traits
 Mortalityns11.2*** (3)5.2* (3)nsns11.4*** (6)ns11.9*** (3)35.1*** (2)
 Growth rate7.8* (3,1600)3.6* (3,1600)39.1*** (2,1600)3.6* (3,1600)nsnsnsns
Reproductive traits
 Probability of reproductionns23.8*** (3)ns11.5** (3)nsns7.6** (2)nsns
 Inflorescence production14.7*** (1)45.3*** (3)7.1** (2)27.7*** (2)nsns11.4** (2)nsns
 Infrutescences/inflorescences ratio 7.5** (3)ns5.8* (2) 16.1** (6) nsns
 Fruit production 19.2** (3)ns21.7** (2) ns 12.6** (3)ns

Leaf functional traits

Cumulative effects of chronic defoliation

After 3 years of sustained defoliation, values of all leaf traits analysed had declined with respect to that of control palms. Overall, across all defoliation treatments, male palms had significantly higher leaf trait values than females (Table 1; Fig. 1). The negative effect of defoliation on leaf persistence and leaf length was stronger among females than among males (Fig. 1a–d; Table 1). For the other leaf traits, defoliation effects did not differ significantly between genders. Stem length did not affect any of the leaf traits analysed (Table 1).

image

Figure 1.  Recovery patterns of leaf functional traits of Chamaedorea elegans palms, as a function of gender and chronic defoliation levels in south-eastern Mexico. Leaf persistence, leaf production rate, total leaves, leaf length and total leaf area for females (a, c, e, g and i) and males (b, d, f, h and j). The indicated defoliation treatments correspond to the percentage of standing leaves removed every 6 months over 3 years. Last defoliation event was applied in 2000. Bars indicate mean value, and vertical lines represent ± 1 standard error.

Download figure to PowerPoint

Recovery from defoliation

Recovery of leaf traits in defoliated palms compared with the control group varied depending on gender and the level of defoliation. The magnitude of such effects varied among years with higher values in 2001 for most of the leaf traits. However, differences among years were only significant for leaf production rate, leaf length and total leaf area (Table 2, Fig. 1). Overall, male palms from all defoliation treatments recovered more quickly than female palms, and after 2 or 3 years, all leaf trait values of defoliated palms had values that were not significantly different from those observed in the control male palms (Fig. 1). By contrast, over the same time span, female palms in the 50% and 66% defoliation treatments only recovered leaf production rate (Fig. 1c) and the standing number of leaves (Fig. 1e), whereas values of leaf persistence (Fig. 1a), leaf length (Fig. 1b) and total leaf area (Fig. 1i) were still lower than those of non-defoliated female palms. For each of the three recovery years, on average, male palms exhibited significantly higher leaf trait values than female palms, except leaf length in the third year of recovery (Table 2, Fig. 1).

It is interesting to note that in Recovery Yr-1 (2001), female palms, and to a lesser extent male palms, from most treatments (included control palms) increased their mean total leaf area (Fig. 1i) due to an increase in leaf production rate (Fig. 1c) and the production of larger leaves (Fig. 1g). We only found significant interactive effects of defoliation treatment with stem length (SL00) on leaf persistence and leaf length, independent of gender (Table 2). Overall, bigger palms from the 50% and 66% defoliation treatment had less persistent leaves compared with control palms (Fig. S2). While leaf length increased with stem length in defoliated palms, this was not the case for the control plants. As a result of these trends, smaller defoliated palms had smaller leaves than control plants, while the contrary was true for bigger palms (Fig. S3). Finally, we detected a positive single effect of stem length on the number of leaves, which was independent of the defoliation treatment, gender and year; however, stem length explained only 9% of the inter-individual variation.

Mortality

Cumulative effects of chronic defoliation

Overall, in the final year of defoliation treatments, the effect of defoliation on mortality rates differed between genders, with higher mortality rates in females (Table 1, Fig. 2). Among female palms, mortality increased with defoliation level, with 66% defoliated females having a threefold higher mean (±SE) mortality rate (9.1% ± 0.1) than control palms (Fig. 2a). By contrast, mortality rates did not differ significantly between defoliated and control male palms (Fig. 2a,b).

image

Figure 2.  Temporal changes in mortality (ind ind−1 ha−1) in Chamaedorea elegans palms after being subjected to different chronic leaf defoliation treatments in Chajul, south-eastern Mexico. 2000 represents the last year of defoliation treatment. Differences among genders are illustrated in the different charts: (a) females and (b) males. Bars indicate mean value, and vertical lines represent ± 1 standard error.

Download figure to PowerPoint

Recovery from defoliation

The ability to survive after chronic defoliation depended on defoliation treatment, and there were no significant differences among genders (Table 2, Fig. 2). In the first year after defoliation (2001), mortality rate increased for both genders and for all defoliation treatments including the control (0% defoliation). This amplified differences among defoliation treatments, where mortality rate rose over 15% in the 66% defoliation treatment (Fig. 2). Two years after last defoliation event (2002), mortality rates of male and female defoliated palms were not significantly different from those of control palms.

Mortality rate varied depending on the interaction between defoliation treatment and stem length (SL00) and between defoliation treatment and year (Table 2). Effects of plant size were mainly observed in the Recovery-Yr1, where bigger palms (independent of gender) subjected to 66% defoliation exhibited higher mortality rates than palms from the other defoliation treatments, and this difference disappeared in later years (Fig. S4).

Growth

Cumulative effects of chronic defoliation

Three years of repeated defoliation lead to a reduction in stem growth rate (Table 1), principally in palms from the 66% defoliation treatment (Fig. 3). Defoliation impacted growth of males and females in a similar way (Table 1). Across all treatments, females grew significantly slower than males with a mean (±SE) of 4.01 (±0.19) vs. 4.56 (±0.22) cm year−1, respectively. Initial stem size did not affect the response on stem growth.

image

Figure 3.  Temporal changes in stem growth (cm year−1) in Chamaedorea elegans palms after being subjected to different chronic leaf defoliation treatments in Chajul, south-eastern Mexico. 2000 represents the last year of defoliation treatment. Differences among genders are illustrated in the different charts: (a) females and (b) males. Bars indicate mean value, and vertical lines represent ± 1 standard error.

Download figure to PowerPoint

Recovery from defoliation

One year after the last defoliation event (2001), growth rates of palms from all defoliation treatments, including the control, decreased and were slower than in subsequent years. In this year, greater impacts of defoliation were observed on female compared with male palms (Table 2); female palms from the 50% and 66% defoliation treatments had growth rates that were 28% and 49% slower than those of control palms, respectively. Defoliated females took more time to recover; it took 2 years until growth rates were similar to growth rates of the control female palms compared with only 1 year for the males (Table 2, Fig. 3a,b). However, 3 years after the last defoliation event, differences in growth between males and females were no longer significant (Table 2, Fig. 3). Stem size did not affect the recovery process of stem growth (Table 2).

Reproductive traits

Cumulative effects of chronic defoliation

All defoliation treatments reduced the probability of reproduction, and to a lesser extent inflorescence production, in female palms, while in male ones, only the 66% treatment produced a significant reduction (Table 2; Fig. 4). In female palms, infructescence/inflorescence ratio (Fig. 5a) and fruit production per reproductive palm (Fig. 5b) strongly decreased as defoliation level increased. Stem length did not affect the response of any of the reproductive traits analysed (Table 1).

image

Figure 4.  Temporal changes in reproductive traits of Chamaedorea elegans palms after being subjected to different chronic leaf defoliation treatments in Chajul, south-eastern Mexico. 2000 represents the last year under defoliation. The top charts show the probability of reproduction, and the bottom charts show inflorescence production for: females (a, c) and males (b, d).

Download figure to PowerPoint

image

Figure 5.  Temporal changes in reproductive traits of female palms of Chamaedorea elegans after being subjected to different chronic leaf defoliation treatments in Chajul, south-eastern Mexico. 2000 represents the last year of defoliation treatment. (a) Infrutescences/inflorescences ratio and (b) fruit production per reproductive female (ind.ind repr−1 year−1). Bars indicate mean value, and vertical lines represent ± 1 standard error.

Download figure to PowerPoint

Recovery from defoliation

Recovery depended on the reproductive trait considered and on the defoliation treatment and gender (Table 2). In the first year after defoliation (2001), most reproductive trait values of palms from all defoliation treatments (including control palms) decreased with respect to those of previous and subsequent years, especially in female palms (Figs 4 and 5). In this and the following 2 years, female palms from all defoliation treatments had a much lower probability of reproduction than control female palms (Fig. 4a, Table 2). By contrast, male palms from all defoliation treatments exhibited probabilities of reproduction similar to control male palms 2 years after defoliation was suspended (Table 2, Fig. 4b).

In contrast to the probability of reproduction, inflorescence production of female palms from all defoliation treatments had risen to values similar to control palms 2 years after the last defoliation event (Fig. 4c). Inflorescence production in male palms was not affected by defoliation, and individuals from all defoliation treatments recovered inflorescence production in just 1 year (Fig. 4d).

Recovery of other reproductive components of female palms varied depending on the trait considered. While the infructescence/inflorescence ratio of female palms from all defoliation treatments reached values similar to that of control palms 2 years after defoliation (Fig. 5a), rates of fruit production per reproductive palm recovered only after 3 years (Fig. 5c) did not recover at all (Fig. S2). The only significant interactive effect between stem length and defoliation was observed on fruit production (Table 2). Overall, while in the control palms, fruit production increased with stem length, in the defoliated palms, such relationship was negative, especially for palms subjected to the 50% and 66% defoliation treatments (Fig. S5). Finally, inflorescence production increased with stem length, independent of defoliation treatment and gender, but the explained variance was only 10% (Table 2).

Discussion

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

Leaf area losses, which are caused by physical or biotic agents, can strongly influence functional and demographic behaviour of plants (Marquis 1984; Anten & Ackerly 2001; Endress, Gorchov & Noble 2004). Our results show that the cumulative effects of repeated defoliation on leaf functional traits and demographic attributes of C. elegans palms depend on both the defoliation level and gender, but not on plant size. Specifically, our findings support the hypotheses that: (i) female palms suffer stronger negative demographic effects due to defoliation than males, (ii) the magnitude of the negative effects increased with the level of defoliation, (iii) the effects on reproduction were stronger than effects on leaf traits, growth and survival. Regarding the ability of palms to recover from defoliation (resilience), we found that: (i) leaf length and leaf production rate recover faster than total leaf area, (ii) survival rates recover more quickly than growth and reproduction, (iii) male palms recover faster than females and (iv) the recovery time increases with higher levels of defoliation (Table 3). Plant size did not have a consistent effect on the recovery of defoliated palms.

Table 3.   Rates of functional and demographic recovery of Chamaedorea elegans palms with different levels of chronic defoliation. Speed of recovery is measured as the number of years (1–3 years) required for a palm to reach statistically same trait values than those of non-defoliated (control) palms. ‘>3’ means that this specific trait did not recover 3 years after the last defoliation event. The – symbol indicates those factors are restricted to females. Percentages correspond to three chronic defoliation levels (per cent of standing leaves removed) applied every 6 months over 3 years (1997–2000)
TraitFemalesMales
33%50%66%33%50%66%
Functional leaf traits
 Leaf persistence1>3>3233
 Leaf production rate222122
 Total leaves233233
 Leaf length13>3112
 Total leaf area2>3>3233
Demographic traits
 Mortality112112
 Growth rate223111
Reproductive traits
 Probability of reproduction>3>3>3122
 Inflorescence production222111
 Infrutescences/inflorescences ratio223
 Fruit production per reproductive female3>3>3

For C. elegans, leaf harvesting is probably the main cause of leaf area loss, and our defoliation treatments bracketed the levels at which this harvesting occurs. Although such defoliation levels in Chamaedorea species may also happen due to natural causes (N. P. R. Anten, personal observations), they typically tend to be lower (1–23% of leaf area; Martínez-Ramos, Anten & Ackerly 2009; Cepeda-Cornejo & Dirzo 2010). However, for many other species, stochastic events such as insect outbreaks, branch fall, strong storms, browsing and fire may damage higher proportion of leaf area and cause higher functional damages and affect population patterns (Coley, Bryant & Chapin 1985; McPherson & Williams 1998; Lopez-Toledo, Horn & Endress 2011).

Effects of chronic defoliation

Earlier defoliation studies conducted on palm species found that single defoliation events generally had no negative functional or demographic effects. In some cases, overcompensation, that is, faster growth and reproduction in defoliated plants, was even observed (Mendoza, Piñero & Sarukhán 1987; Oyama & Mendoza 1990; Chazdon 1991). In contrast, our results strongly suggest that cumulative effects of chronic defoliation may lead to strong reductions in demographic rates. Palms tend to maintain large carbohydrate stores, and these stores play a crucial role in survival and regrowth after defoliation (Kobe 1997). For C. elegans, the cumulative effects of chronic defoliation were independent of stem size, which may be indicative of the importance of storage organs other than stems, such as roots. In chronically defoliated palms, these stores may be depleted by the continuous resource demand to support regrowth of photosynthetic tissues and maintenance of vegetative organs (Belsky et al. 1993; McPherson & Williams 1998). For example, non-structural carbohydrates in below-ground organs of Sabal palmetto seedlings were depleted after repeated events of defoliation, which then resulted in reduced growth rates and increased mortality rates (McPherson & Williams 1998).

The effects of leaf area losses also depend on the level of defoliation. Even a single, but severe (75–100% of leaves removed), defoliation event may result in significant reductions in functional or demographic traits (Mendoza, Piñero & Sarukhán 1987; Oyama & Mendoza 1990; Boege 2005). Our results agree with these studies, as our lower defoliation treatment (33%) had only minor negative effects in some reproductive traits (probability of reproduction and fruit production), while the more intensive defoliation treatments (50–66%) had strong negative effects on all analysed demographic rates. This indicates that chronic defoliation at high intensities may cause strong negative effects on plant fitness (Endress, Gorchov & Berry 2006; Martínez-Ramos, Anten & Ackerly 2009).

Gender effects and recovery from chronic defoliation

In dioecious species, resource allocation to reproduction tends to be asymmetrical between genders (Case & Ashman 2007). It has been documented that female plants invest more resources in reproductive functions than males and that their growth rates are consequently slower than those of males (Ataroff & Schwarzkopf 1992). This has also been documented in four species of Chamaedorea palms (Oyama & Dirzo 1988; Cepeda-Cornejo & Dirzo 2010). Our study on C. elegans is the first showing differences between genders both in the cumulative effects of chronic defoliation and in the resilience to such disturbance in perennial plants. Differential gender-specific resource allocation could explain why, after chronic defoliation, female palms produced lower photosynthetic total area and, concomitantly, lower survival and growth than males. Also, the recovery of most functions, and particularly reproduction, following the cessation of chronic defoliation was both slower and less complete in females. In the case of fruit production, there seemed to be no recovery at all even 3 years after defoliation had stopped, that is, the relative difference between defoliated and non-defoliated females did not decrease. The latter effect was probably caused by the slower recovery in leaf area and associated carbon reserves in females.

The detected interactive effect between plant size and defoliation on leaf length and leaf persistence could be indicative of different strategies associated with leaf area production. Smaller defoliated palms maintained more but smaller leaves, while larger defoliated palms maintained fewer bigger leaves. Such different strategies could be associated with the fact that more but smaller leaves entail lower support costs at the leaf level, but larger costs of stem growth and self shading (Anten & Ackerly 2001). On the other hand, the higher mortality and lower fruit production exhibited by bigger recovering palms may have two possible complementary explanations. In general, as plant size increases, the leaf area ratio (leaf area per biomass unit) and thus the balance between photosynthetically active and non-active tissue decreases (Poorter 1999). A reduction in leaf area by defoliation could therefore be more likely to result in a negative carbon balance in large plants, relative to smaller ones. It is also possible that reproductive costs, expressed as future reductions in survival and or reproduction, are higher for larger plants (e.g. Piñero, Sarukhán & Alberdi 1982) and that such costs are more strongly expressed in defoliated plants.

Overall, our results suggest that the recovery process from chronic defoliation in C. elegans proceeds in a series of gradual steps. As discussed previously, single or few defoliation events have limited or no perceptible negative consequences on functional or demographic traits, and plants easily recover their pre-defoliation status (Oyama & Mendoza 1990; Chazdon 1991; Endress, Gorchov & Noble 2004). If defoliation is recurrent, functions of relatively low energetic and demographic costs, such as reproduction, are the first traits negatively affected. In long-lived understorey palms, it has been documented that individuals allocate more resources to somatic structures (roots, stems and leaves) important for survival and growth than to reproductive organs (Piñero, Sarukhán & Alberdi 1982; Oyama & Dirzo 1988). In long-lived plants, reproductive rates tend to be much less important for population growth and fitness (i.e. they have much lower elasticity) than growth and survival parameters (Franco & Silvertown 2004). As defoliation persists through time, a decrease in functional components of higher energetic and demographic costs, such as stem growth and leaf area production, occurs. Finally, if defoliation continues, plants are not able to maintain a positive carbon balance and die.

The recovery process from chronic defoliation also follows a gradual pattern. Leaf area recovered first, mediated by recovery of other leaf traits such as leaf persistence, leaf production and leaf size. This fast recovery of leaf area and associated light capture is probably vital for survival in the forest understorey. In males, all leaf traits had recovered to the levels of non-defoliated plants within 3 years following the cessation of defoliation. By contrast, leaf persistence and leaf length, and by consequence total standing leaf area, did not fully recover in females during this period. Once leaf area has recovered, survival recovers faster than growth and growth recovers faster than reproduction.

Our results indicate that chronic and high levels of defoliation strongly reduce seed production, increase mortality and reduce growth. All this together may shrink populations and skew sex ratios to a higher proportion of males, which may consequently influence long-term population growth as found in other dioecious palm species (Holm, Miller & Cropper 2008). This may also affect genetic variation (Eguiarte et al. 1993) well beyond the period over which defoliation takes place. It is likely that chronic defoliation may be responsible for the reduction or local extinction of Chamaedorea populations in some tropical regions in Mexico and Central America with high leaf extraction rates (Sánchez-Carrillo & Valtierra-Pacheco 2003; Bridgewater et al. 2006; Endress, Gorchov & Berry 2006). Our results thus strongly indicate that the impact of stress events on plant performance and population dynamics should include estimates of recovery time, and in long-lived species, this should be extended several years beyond the period over which the stress events occurred. Another important implication of our results is that studies on gender differences in demographic rates should have long-term approaches and recognize historical or severe stress events to identify possible effects of stress events that occurred in the past. These issues have not been addressed in natural populations and will be very important in future research.

Stochastic effects

Defoliation effects may interact with other factors such as climatic stochasticity. Previous studies have shown that the impact of such external factors may even be stronger than those of defoliation. For example, our previous study with C. elegans included a severe drought related to an ENSO event in 1998, which caused strong effects in mortality and reproduction independently of defoliation level (Martínez-Ramos, Anten & Ackerly 2009). Surprisingly, during the year period 2000–2001, a similar anomaly in precipitation and temperature to that of El Niño-1998 was recorded in our study locality (Lacantum, Chiapas, CNA-Mexico). During the 2000 dry season, rainfall from February to April was 66% lower than the long-term average (Fig. S1). As our results show, this drought event had clear impacts on the ability of palms to recover, as it intensified the effects of defoliation and slowed recovery rates. As droughts have been predicted to become more frequent in tropical forest areas in the future (Easterling et al. 2000), their effects on plant recovery from defoliation should be considered when quantifying the impact of chronic defoliation in understorey palms.

Conclusion

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

In this study, we have shown that chronic defoliation reduces fitness components (survival, growth and reproduction), and this reduction is higher as defoliation levels increase, particularly for female plants. Plants appear to survive chronic defoliation by strongly reducing reproduction and subsequently growth. Once defoliation stops, males recover faster than females, and the recovery appears to be stepwise, with survival and growth increasing first and reproductive functions recovering more slowly. The slow recovery of females, in particular, may have significant consequences for population dynamics and genetic variation and should be considered in demographic studies as well as in studies that determine the impact of leaf harvesting. Climatic factors, such as severe drought, affect the ability of plants to recover from chronic defoliation. Thus, the incidence of severe episodic disturbances, such as drought, may play a dominant role in the population dynamics of tropical rain forest understorey plants when combined with chronic defoliation.

Acknowledgements

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

This study was supported by the U.S. National Science Foundation (grant IBN 9604030) and the Packard Foundation (No. 1999-4903). CIECO-UNAM provided additional financial support. LLT was supported by a Master Scholarships (CONACYT: 163218) and the Postdoctoral Program of the San Diego Zoo Global. We acknowledge logistic support from CIECO-UNAM and Chajul Field Station. We thank Jorge Rodríguez, Gilberto Jamangape and Praxedis Sinaca for help with fieldwork. MMR thanks DGAPA-UNAM and CONACYT for sabbatical fellowships at the University of California at Berkeley. Finally, we thank the Editors and anonymous reviewers for their valuable suggestions.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • Anderson, H., Galeano, G. & Bernal, R. (1997) Field Guide to the Palms of the Americas. Princeton University Press, New Jersey, USA.
  • Anten, N.P.R. & Ackerly, D.D. (2001) Canopy-level photosynthetic compensation after defoliation in a tropical understorey palm. Functional Ecology, 15, 252262.
  • Anten, N.P.R., Martinez-Ramos, M. & Ackerly, D.D. (2003) Defoliation and growth in an understory palm: quantifying the contributions of compensatory responses. Ecology, 84, 29052918.
  • Ataroff, M. & Schwarzkopf, T. (1992) Leaf production, reproductive patterns, field germination and seedling survival in Chamaedorea bartlingiana, a dioecious understorey palm. Oecologia, 92, 250256.
  • Belsky, A., Carson, W., Jensen, C. & Fox, G. (1993) Overcompensation by plants – herbivore optimization or red herring. Evolutionary Ecology, 7, 109121.
  • Boege, K. (2005) Influence of plant ontogeny on compensation to leaf damage. American Journal of Botany, 92, 16321640.
  • Bridgewater, S.G.M., Pickles, P., Garwood, N.C., Penn, M., Bateman, R.M., Morgan, H.P., Wicks, N. & Bol, N. (2006) Chamaedorea (Xate) in the Greater Maya Mountains and the Chiquibul Forest Reserve, Belize: an economic assessment of a non-timber forest product. Economic Botany, 60, 265283.
  • Canham, C.D., Kobe, R.K., Latty, E.F. & Chazdon, R.L. (1999) Interspecific and intraspecific variation in tree seedling survival: effects of allocation to roots versus carbohydrate reserves. Oecologia, 121, 111.
  • Case, A.L. & Ashman, T.L. (2007) An experimental test of the effects of resources and sex ratio on maternal fitness and phenotypic selection in gynodioecious Fragaria virginiana. Evolution, 61, 19001911.
  • Cepeda-Cornejo, V. & Dirzo, R. (2010) Sex-related differences in reproductive allocation, growth, defense and herbivory in three dioecious neotropical palms. PLoS ONE, 5, 3.
  • Chazdon, R. (1991) Effects of leaf and ramet removal on growth and reproduction of Geonoma congesta, a clonal understorey palm. Journal of Ecology, 79, 11371146.
  • Coley, P.D., Bryant, J.P. & Chapin III, S. (1985) Resource availability and plant antiherbivore defense. Science, 230, 895899.
  • Crawley, M. (2007) The R Book. John Wiley & Sons Ltd, Chichester, UK.
  • Cunningham, S.A. (1997) The effect of light environment, leaf area, and stored carbohydrates on inflorescence production by a rain forest understory palm. Oecologia, 111, 3644.
  • Easterling, D.R., Meehl, G.A., Parmesan, C., Changnon, S.A., Karl, T.R. & Mearns, L.O. (2000) Climate extremes: observations, modeling, and impacts. Science, 289, 20682074.
  • Eguiarte, L., Búrquez, A., Rodríguez, J., Martínez-Ramos, M., Sarukhán, J. & Piñero, D. (1993) Direct and indirect estimates of neighborhood and effective population size in a tropical palm, Astrocaryum mexicanum. Evolution, 47, 7587.
  • Endress, B.A., Gorchov, D.L. & Berry, E.J. (2006) Sustainability of a non-timber forest product: effects of alternative leaf harvest practices over six years on yield and demography of the palm Chamaedorea radicalis. Forest Ecology and Management, 234, 181191.
  • Endress, B.A., Gorchov, D.L. & Noble, R.B. (2004) Non-timber forest product extraction: effects of harvest and browsing on an understory palm. Ecological Applications, 14, 11391153.
  • Everitt, B.S. & Hothorn, T. (2010) A Handbook of Statistical Analyses Using R. Chapman and Hall/CRC, London, UK.
  • Franco, M. & Silvertown, J.J. (2004) A comparative demography of plants based upon elasticities of vital rates. Ecology, 85, 531538.
  • Hawkes, C.V. & Sullivan, J.J. (2001) The impact of herbivory on plants in different resource conditions: a meta-analysis. Ecology, 82, 20452058.
  • Hodel, D.R. (1992) Chamaedorea Palms. The Species and Their Cultivation. Allen Press, Kansas.
  • Holm, J.A., Miller, C.J. & Cropper, W.P. (2008) Population dynamics of the dioecious Amazonian palm Mauritia flexuosa : simulation analysis of sustainable harvesting. Biotropica, 40, 550558.
  • Kobe, R.K. (1997) Carbohydrate allocation to storage as a basis of interspecific variation in sapling survivorship and growth. Oikos, 80, 226233.
  • Lapointe, L., Bussieres, J., Crete, M. & Ouellet, J.P. (2010) Impact of growth form and carbohydrate reserve on tolerance to simulated deer herbivory and subsequent recovery in Liliaceae. American Journal of Botany, 97, 913924.
  • Lopez-Toledo, L., Horn, C. & Endress, B.A. (2011) Distribution and population patterns of the threatened palm Brahea aculeata in a tropical dry forest in Sonora, Mexico. Forest Ecology and Management, 261, 9011910.
  • Marquis, R. (1984) Leaf herbivores decrease fitness of a tropical plant. Science, 226, 537539.
  • 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.
  • McNaughton, S. (1983) Compensatory plant-growth as a response to herbivory. Oikos, 40, 329336.
  • McPherson, K. & Williams, K. (1998) The role of carbohydrate reserves in the growth, resilience, and persistence of cabbage palm seedlings (Sabal palmetto). Oecologia, 117, 460468.
  • Mendoza, A., Piñero, D. & Sarukhán, J. (1987) Effects of experimental defoliation on growt, reproduction and survival of Astrocaryum mexicanum. Journal of Ecology, 75, 545554.
  • Obeso, J.R. (2002) The costs of reproduction in plants. New Phytologist, 155, 321348.
  • Oyama, K. & Dirzo, R. (1988) Biomass allocation in the dioecious tropical palm Chamaedorea tepejilote and its life history consequences. Plant Species Biology, 3, 2333.
  • Oyama, K. & Mendoza, A. (1990) Effects of defoliation on growth, reproduction and survival of a netropical dioecious palm, Chamaedorea tepejilote. Biotropica, 22, 119123.
  • Piñero, D., Sarukhán, J. & Alberdi, P. (1982) The costs of reproduction in a tropical palm Astrocaryum mexicanum. Journal of Ecology, 70, 473481.
  • Pinheiro, J. & Bates, D.M. (2000) Mixed-effects Models in S and S-Plus. Springer-Verlag, New York, USA.
  • Poorter, L. (1999) Growth responses of 15 rain-forest tree species to a light gradient: the relative importance of morphological and physiological traits. Functional Ecology, 13, 396410.
  • R Development Core Team (2010) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org .
  • Sampaio, M.B. & Scariot, A. (2010) Effects of stochastic herbivory events on population maintenance of an understorey palm species (Geonoma schottiana) in riparian tropical forest. Journal of Tropical Ecology, 26, 151161.
  • Sánchez-Carrillo, D. & Valtierra-Pacheco, E. (2003) Social organization for the exploitation of camedor palms (Chamaedorea spp.) in the Lacandona rainforest, Chiapas. Agrociencia, 37, 545552.
  • Zuidema, P.A., De Kroon, H. & Werger, M.J.A. (2007) Testing sustainability by prospective and retrospective demographic analyses: evaluation for palm leaf harvest. Ecological Applications, 17, 118128.

Supporting Information

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

Figure S1.Climate inter-annual (1994?2003) variation at the Chajul Biological Station, southeastern Mexico.

Figure S2.Relationships between leaf persistency and stem length for recovering Chamaedorea elegans palms subjected to different defoliation levels.

Figure S3.Relationships between leaf length and stemlength for recovering Chamaedorea elegans palms subjected to different defoliation levels.

Figure S4.Temporal changes inmortality rate of recovering Chamaedorea elegans palms as a function of defoliation treatments and stem length.

Figure S5. Relationships between mean annual fruit production and stem length for recovering Chamaedorea elegans female palms subjected to different defoliation treatments in south-eastMexico.

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
jec1992_sm_FigS1-S5.doc599KSupporting info item

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.