SEARCH

SEARCH BY CITATION

Keywords:

  • climatic cues;
  • fertilization;
  • mast seeding;
  • Nothofagus solandri;
  • reproductive ecology;
  • resource availability;
  • seed production;
  • synchrony

Summary

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

1. Mast seeding is the intermittent production of large quantities of seed across a perennial plant population. Such seeding events in many plant species are initiated by climatic cues, but whether these cues act solely as triggers or also via alterations to nutrient availability is unclear.

2. Here, we examine the effect of nitrogenous fertilization on the relationship between seed production in Nothofagus solandri var. cliffortioides and two climatic cues (rainfall and temperature) at specific stages in reproductive development from 1999 to 2008.

3. Foliar nitrogen concentrations were positively correlated with rainfall among years, suggesting rainfall was affecting nitrogen availability.

4. Seedfall mass in unfertilized stands was predominantly determined by rainfall during resource priming, while seed production in fertilized stands was more affected by temperature during floral primordia development. Similarly, seedfall mass in older stands, which contain greater internal nutrient reserves, was predominantly determined by temperature.

5.Synthesis. The results of this study demonstrate that the sensitivity of seed production to climatic cues can be altered by manipulation of resource availability and therefore establish that climatic cues involved in the synchronization of mast seeding can influence reproductive effort via an effect on resource availability. These results also indicate that alterations to resource availability have the potential to alter inter-annual patterns of seed production, but further study is required to verify this finding.


Introduction

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

Climatic cues are often related to the synchronization of, and variability in, seed production by perennial plant species, commonly referred to as mast seeding (Kelly & Sork 2002; Schauber et al. 2002). This synchronization of reproductive effort by plant species can occur over great distances (Herrera et al. 1998; Kelly & Sork 2002; Schauber et al. 2002; Selås et al. 2002; Mduma, Sinclair & Turkington 2007). In this paper, we examine the relationships between seed production and selected climatic cues by demonstrating that the influence of rainfall and temperature on seed production can be altered through manipulating resource availability.

Several hypotheses have been proposed to explain the evolution of mast seeding (Kelly 1994). Predator satiation (e.g. Janzen 1971) and increased pollination efficiency (e.g. Nilsson & Wästljung 1987) are currently the best-supported evolutionary explanations, while climate and resource dynamics act as important proximate factors (Kelly & Sork 2002; Koenig & Knops 2005; Crone, Miller & Sala 2009). Temperature and rainfall are two climatic cues often associated with mast seeding. Cool temperatures and increased moisture availability during resource priming c. 2 years prior to seedfall can increase seed production across populations of a plant species (Piovesan & Adams 2001; Richardson et al. 2005), while relatively high temperatures during floral primordia development c. 1 year prior to seedfall cue mast seeding in various tree species (Schauber et al. 2002; Övergaard, Gemmel & Karlsson 2007). However, the process by which these climatic cues influence seed production is not clearly defined. Given the resources required to support substantial increases in seed production, nutrient availability must influence the incidence of mast seeding (Janzen 1971; Kelly & Sork 2002; Rees, Kelly & Bjørnstad 2002). Internal carbohydrate reserves appear to be an important factor in determining the incidence of mast seeding (Isagi et al. 1997; Miyazaki et al. 2002), while nitrogen (N) availability is critical to the timing of mast-seeding events due to the importance of N to photosynthesis and the generation of reproductive tissue (Yasumaru, Hikosaka & Hirose 2006; Han et al. 2008). This is supported by several studies in which the application of nitrogenous fertilizers has increased seed production in various tree species (Le Tacon & Oswald 1977; Fahey, Battles & Wilson 1998; Davis, Allen & Clinton 2004), and it appears that increasing atmospheric N deposition is a factor in both the greater frequency and magnitude of mast seeding in beech (Fagus) forests in southern Sweden (Övergaard, Gemmel & Karlsson 2007). Consequently, it has been suggested that synchronized increases in seed production are a function of the effects of the climatic cues on the availability of resources, such as the role of temperature and soil moisture in facilitating N availability (Walse, Berg & Sverdrup 1998; Paul et al. 2003). However, these cues may also be acting to synchronize reproductive effort by manipulating endogenous plant process such as internal resource allocation (Bazzaz et al. 1987; Koenig & Knops 2005; Richardson et al. 2005) or producing environmental conditions beneficial to other facets of reproduction, such as pollen dispersal (Selås et al. 2002). This position suggests that climatic cues influence the incidence of mast seeding by acting directly on internal reproductive processes to synchronize the incidence of mast seeding (Koenig & Knops 2005; Richardson et al. 2005), but this does not exclude the potential for direct and indirect pathways to operate simultaneously.

To partition the role of climate as a trigger on endogenous plant processes from effects on resource availability we examined the effect of N fertilizer addition on the relationship between seed production and selected climatic cues in a New Zealand mountain beech [Nothofagus solandri var. cliffortioides (Hook. f.) Poole: Fagaceae] forest. The relationship between seed production and the climatic cues was assessed at three specific stages of reproductive development – resource priming, floral primordia development and flowering (Richardson et al. 2005). We also assessed the effect of stand development on the relationship between seed production and the climatic cues, as internal nutrient reserves in mountain beech increase substantially with age (Clinton, Allen & Davis 2002), potentially decreasing the dependence of older trees on climatic factors to supply the nutrients required for increased seed production. Given the potential for resource availability to influence the extent of reproductive effort (Isagi et al. 1997; Satake & Iwasa 2000; Rees, Kelly & Bjørnstad 2002; Newbery, Chuyong & Zimmermann 2006), we also explored the possibility that alterations to N supply could influence the pathway by which seed production is determined (see Fig. 1).

image

Figure 1.  Conceptual algorithm illustrating the potential for fertilization to influence the incidence of mast seeding. Temperature in the year prior to flowering is the first critical point, while rainfall 2 years prior to flowering and the presence or absence of fertilization determine if nutrient reserves available to the plant are greater than the required threshold to allow increased reproductive effort.

Download figure to PowerPoint

We investigated the following hypotheses: (i) that increased N availability in mountain beech stands would increase the mass of seed produced in all years, (ii) increased N availability would decrease the dependency of seed production on climatic cues during the resource priming phase of reproductive development and (iii) greater stand age would decrease the dependency of seed production on climatic cues during the resource priming phase of reproductive development. We also hypothesize that any fertilizer-induced variation in the relationship between climatic cues and seed production would alter inter-annual patterns of seed production relative to unfertilized plots.

Examination of the data demonstrated that the synchronization of seed production by rainfall is at least partially related to an effect of rainfall on N availability. We also showed that the relative importance of climatic cues to seed production, during different phases of reproductive development, varies significantly with resource availability and stand maturity. Increased N supply was also observed to partially alter patterns of reproductive effort in 1 year.

Materials and methods

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

Study site

The study site was in the Craigeburn Range of the South Island of New Zealand (43°15′ S, 171°35′ E) within an established developmental sequence of pure mountain beech forest (Allen, Clinton & Davis 1997; Clinton, Allen & Davis 2002; Davis, Allen & Clinton 2003). Extensive seed production research has been carried out in this forest, including the generation of the longest seedfall data set yet produced (Kelly & Sork 2002). The study was carried out in eight 125-year-old (pole) stand plots and eight 25-year-old (sapling) stand plots within the sequence (see Davis, Allen & Clinton 2004). Pole stand plots were 20 × 20 m surrounded by a 10-m buffer zone and sapling stand plots were 10 × 10 m with a 5-m buffer zone. All 16 plots were located within an area of mountain beech covering c. 3.6 km2; plot installation inside this area was stratified according to stand age, with particular attention given to ensuring the integrity of stand age within each plot. Distances between plot buffer zones varied from 40 m to c. 1200 m. In 2007, mean diameter at breast height (d.b.h.) of the pole stand trees was 166 mm (n = 993); mean sapling d.b.h. was 54 mm (n = 1474). All plots were located on side slopes between 1015 and 1208 m a.s.l. with plot slope varying from 7° to 35° and aspect varying from 98° to 170°. Soils were Allophonic Brown (Hewitt 1992) derived from greywacke, loess and colluvium, and were composed of litter (L) and fermentation–humus (FH) layers over a silt loam A horizon and a stony B horizon. Daily air temperatures (mean, minimum and maximum) and precipitation were taken from a climate station located 914 m a.s.l. within 1500 m distance of all plots.

Four sapling stand plots and four pole stand plots (including buffer zones) were randomly selected to receive fertilizer application. Nitrogen was added to the fertilized plots as urea, which was chosen over ammonium nitrate to reduce the risk of nitrate leaching. Two initial applications of fertilizer were made in May 2000 and October 2000, and two further applications were made in November 2004 and October 2006 to maintain a difference in N supply between the fertilized and unfertilized plots, as indicated by annual measurements of foliar N concentrations. Nitrogen addition to the plots was 200 kg N ha−1 at each application. Seed production was examined across the two mountain beech stand ages to address any potential effect of the greater internal nutrient reserves and more efficient nutrient allocation processes in older trees (Miller, Cooper & Miller 1992; Clinton, Allen & Davis 2002).

Collection and analysis of foliage and seeds

The first foliage collection was conducted immediately before fertilizer was applied in May 2000. All subsequent foliage collections occurred on an annual basis from March 2001. Foliage was taken from the upper crown of five live sample trees in each plot and pooled. Approximately 100 current-year, fully developed leaves were randomly selected from the pooled sample for each plot, then oven-dried and ground for analysis of N after Kjeldahl digestion. A linear mixed-effects model was constructed to assess variations in foliar N concentration over the measured period. Data were log-transformed to meet assumptions of normality. Standard errors were calculated from the mean of the variation around the transformed values, then back-transformed.

Seed production was determined by measuring the mass of seeds falling into three 1 m2 litter traps, installed in each of the 16 plots in early September 1998. Litter traps were placed in a row c. 2 m apart in the centre of each plot. Seeds were collected every 2 months from the traps until late August 1999 when collection was suspended until after the second fertilizer application in October 2000 after which litter collection occurred at three monthly intervals. Collection years were from September to August inclusive, with values reported for the year ending in August (e.g. the values for seedfall mass in 2002 were taken from September 2001 until August 2002). The only year that differed in collection period was 2001 which ran from October 2000 until August 2001. All litter collected from a single plot was mixed, dried at 60 °C, weighed and subsampled. The subsamples were weighed and the mass of seeds determined, allowing the total mass of seeds collected in the litter traps to be estimated. Repeated-measures anova was used to assess the variation in seedfall mass over the entire measurement period, and ancova was used to examine the effect of fertilization in individual years. Mean annual seed production in 1999 was used as a covariate for the analysis in individual years as pre-seedfall mass in the plots randomly selected to receive N applications was less than that in the other plots (7.1 ± 1.9 g m−2 in plots selected for fertilization; 10.6 ± 1.6 g m−2 in plots not selected for fertilization). Where the assumptions of ancova were not met (e.g. nonlinear relationship between the covariate and seedfall mass in a given year), anova was used. Seedfall mass data used to assess variation with fertilization were log-transformed to meet assumptions of normality.

Analysis of seedfall mass in relation to climatic cues

The climatic cues examined in this study were temperature and rainfall during resource priming (December–March 2 years prior to seedfall), floral primordia development (January–April 1 year prior to seedfall) and flowering (December–February immediately prior to seedfall) from 2001 to 2008 inclusive. These intervals were used as they have been identified as important periods during reproductive development in mountain beech (Wardle 1984; Allen & Platt 1990; Richardson et al. 2005). Temperature during these intervals was assessed as the 3- or 4-month mean of daily mean temperatures for each month that fell in the interval; mean daily temperature (Tmean), mean daily minimum temperature (Tmin) and mean daily maximum temperature (Tmax) were calculated for each interval. Rainfall was similarly assessed as the 3- or 4-month mean of daily mean millimetres of precipitation.

To determine if these climatic cues synchronize mast seeding by acting solely on endogenous plant processes, the following rationale was applied: if the synchronizing effect of a climatic cue on seedfall mass is unrelated to nutrient availability in the soil, increased resource availability will have no effect on the correlation between that climatic cue and seedfall mass. Alternatively, if the climatic cue synchronizes seed production at least in part by causing an increase in resource availability, fertilization should decrease the correlation between seedfall mass and that climatic cue. The degree of correlation between seedfall mass and the climatic cues during the three phases of reproductive development was assessed using Pearson’s product moment correlation coefficient. Multiple regression models were constructed and evaluated using Akaike’s Information Criterion (AIC) statistics to determine which combinations of climatic cues explained the greatest degree of variation in seedfall mass in the plots on the basis of fertilization and stand age. To determine if the coefficients describing the effects of the climatic cues on seedfall mass differed significantly with fertilization, a linear model was constructed using dummy variables. This technique enabled the relationships between seedfall mass, rainfall and temperature to be examined with and without of the effect of fertilization, allowing any influence of fertilization on the relationships to be statistically analysed and quantified based on the relative differences. This methodology was also used to determine the significance of any variation in the coefficients with stand age. To corroborate any effects of the climatic cues on nutrient availability, correlations between foliar N concentrations in current-year foliage and climatic cues during shoot expansion in the preceding December–January period (Wardle 1984) were calculated using Pearson’s product moment correlation coefficient. To prevent interference from fertilizer application, this correlation was examined in unfertilized plots only. All statistical analyses were carried out in R Version 2.11.1 (R Development Core Team 2010).

Results

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

Variations in foliar N concentration and seedfall mass

Fertilization increased mountain beech foliar N concentrations (F1,12 = 66.33, < 0.001), although the extent of the increase varied over time (Fig. 2). Stand age did not influence foliar N concentrations (F1,12 = 0.00, = 0.96), and no significant interactions between fertilization and stand age were observed. Foliar N concentrations decreased with time from 2001 (F7,84 = 51.64, < 0.001), and the decrease was greater in sapling stands than pole stands (F7,84 = 2.77, < 0.05). Foliar N concentrations in unfertilized plots were positively correlated with rainfall in the preceding December–January (= 0.71, < 0.05). No significant correlations between foliar N concentrations and temperature in this period were observed.

image

Figure 2.  Mean annual foliar nitrogen (N) concentration in fertilized and unfertilized pole and sapling stands. Back-transformed means ± SE are presented, with triangles indicating the timing of fertilizer applications.

Download figure to PowerPoint

Annual seedfall mass was orders of magnitude greater in 1999, 2002, 2004 and 2006 than in other years (Fig. 3). Seedfall mass was also greater in 2008 but by a substantially lesser margin. Analysis with repeated-measures anova determined that fertilization did not significantly influence the mass of seeds produced across all years (F1,12 = 1.22, = 0.29) but did confirm that seed production was significantly greater in pole stand plots than sapling stand plots (F1,12 = 14.75, < 0.01). No interaction between fertilization and stand age was observed. Analysis of mean seedfall mass in individual years (as opposed to across all years) determined that fertilization significantly increased seed masses in five of the eight measured years (Table 1). Significant interactions were observed between the covariate (annual seed production in 1999) and fertilization in 2003 and 2008. Examination of the data determined that these interactions were driven by variation in the response to fertilization in the same four plots in 2003 and 2008, and consequently do not invalidate the use of the covariate.

image

Figure 3.  Mean annual seed production in fertilized and unfertilized (a) pole stand plots and (b) sapling stands. Back-transformed means ± SE are presented, with triangles indicating the relative timing of fertilizer applications. The timing of fertilization events differs from that in Fig. 1 due to differences in the start date of collection periods. Line breaks are used to indicate no measurements were collected for the year ending 2000.

Download figure to PowerPoint

Table 1.   Analysis of variation in mean mountain beech seed production with fertilization in individual years using pre-fertilization seed production as a covariate
YearSeedfall mass (g m−2)F values
Unfertilized (g m−2)Fertilized (g m−2)Fertilized/unfertilizedCovariateFertilizationCovariate × Fertilization
  1. Back-transformed covariate-adjusted means are presented; in years when no significant interaction between the covariate and fertilization was observed the interaction term was removed from the ancova model.

  2. Bolded F values indicate significant P values; *< 0.05, **< 0.01, ***< 0.001.

  3. †Covariate was not significant in this year and was not used; back-transformed mean values for seedfall mass are presented for this year.

20010.040.020.5F1,13 = 6.19*F1,13 = 0.36 
20023.639.112.5F1,13 = 34.56***F1,13 = 7.39* 
20030.020.052.5F1,12 = 20.59***F1,12 = 12.98**F1,12 = 7.82*
20048.4612.201.4F1,13 = 47.33***F1,13 = 2.69 
20050.060.091.5F1,13 = 6.20*F1,13 = 4.76* 
20062.826.382.3F1,13 = 40.21***F1,13 = 17.63** 
2007†0.030.051.7 F1,14 = 1.92 
20080.131.007.7F1,12 = 12.06**F1,12 = 21.54***F1,12 = 16.36**
All      
Mean1.903.611.9   
SD3.024.92    
CV1.591.36    

Effect of climatic factors on seed production

During resource priming, Tmax was strongly negatively correlated with seedfall mass c. 2 years later, and the strength of the correlation did not vary substantially with fertilizer application or stand age (Fig. 4). The correlations between Tmax and seedfall mass were stronger than those with either Tmean, which was weakly positively correlated to seedfall mass, or Tmin, which was weakly negatively correlated to seedfall mass. Rainfall during resource priming was positively correlated with seedfall mass 2 years later, but the correlation was statistically significant only in unfertilized plots and sapling stand plots. During floral primordia development Tmean was the measurement of temperature most strongly correlated with seedfall mass, and the strength of this correlation was greatest in the fertilized and pole stand plots; Tmax was also significantly positively correlated with seedfall mass in the fertilized plots (= 0.77, < 0.05) and sapling stand plots (= 0.73, < 0.05) during floral primordia development.

image

Figure 4.  Correlations between annual seedfall mass from 2001 to 2008 with temperature and rainfall during resource priming (December–March 2 years prior to seedfall), floral primordia development (January–April 1 year prior to seedfall) and flowering (December–February immediately prior to seedfall). Mean daily temperature, mean daily maximum temperature and mean daily minimum temperature were tested for correlations with seedfall mass; in each case the correlation with the highest r values is given: for resource priming and flowering, mean maximum daily temperature; for floral primordia development, mean daily temperature. Correlations were calculated for seedfall mass in all plots, fertilized plots, unfertilized plots, pole stand plots and sapling stand plots. n = 8 for all correlations; *< 0.05.

Download figure to PowerPoint

The multiple regression models explained 84–92% of the variation in seedfall mass (Table 2). Model selection using AIC statistics determined that optimum models retained only rainfall during resource priming and Tmean during floral primordia development. The relative degree of variation in seedfall mass explained by the rainfall during resource priming and Tmean during floral primordia development varied substantially with fertilization and stand age, but both were consistently positively correlated with seedfall mass. Although Tmax during resource priming was significantly negatively correlated with seedfall mass regardless of fertilization treatment or stand age (Fig. 4), the manual inclusion of this term degraded model performance in all cases, as did the addition of any other term describing rainfall or temperature during flowering.

Table 2.   Relative contribution of components in optimum generalized least-squares models describing seedfall mass as a function of mean monthly rainfall and mean daily temperature (Tmean) during resource priming (P) and floral primordia development (D)
Plot typeEffectCoefficientFPr2
  1. Bolded P and r2 values indicate figures for overall model significance and degree of variance explained for each plot type. F value numerator and denominator degrees of freedom for each component were 1 and 5, respectively.

AllRainfall (P)0.05426.900.0040.454
Tmean (D)8.12322.260.0050.454
   0.0030.908
UnfertilizedRainfall (P)0.05623.750.0050.600
Tmean (D)5.3938.520.0330.266
   0.0070.866
FertilizedRainfall (P)0.05215.010.0120.322
Tmean (D)10.85233.450.0020.585
   0.0030.907
PoleRainfall (P)0.08224.400.0040.417
Tmean (D)13.72135.900.0020.506
   0.0020.923
SaplingRainfall (P)0.02619.260.0070.580
Tmean (D)2.5236.980.0460.260
   0.0100.840

Analysis of the effect of fertilization on the climatic cues in the multiple regression models given in Table 2 determined that the value of the partial regression slope describing the relationship between rainfall and seedfall mass did not vary significantly with fertilizer addition (F1,10 = 0.00, = 0.96). Fertilization did significantly alter the partial regression slope for temperature (F1,10 = 5.34, < 0.05), causing seed production to increase with temperature at a proportionally greater rate in fertilized plots compared with unfertilized plots. The values of the partial regression slopes for rainfall (F1,10 =12.60, < 0.01) and temperature (F1,10 = 17.58, < 0.01) against seedfall mass varied significantly with stand age.

Discussion

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

Resource availability and allocation

The effects of nitrogenous fertilization on foliar N concentration and increases in seedfall mass with fertilization agree with several previous studies (e.g. Crane & Banks 1992; Raison et al. 1992; Davis, Allen & Clinton 2004) and are critical to confirming the effects of N application on resource availability in fertilized stands. Increased N resources can increase seed production directly by allowing more N to be allocated to the generation of reproductive tissue (Yasumaru, Hikosaka & Hirose 2006; Han et al. 2008) or by enhancing the photosynthetic capacity of the trees (Evans 1989), thereby increasing carbohydrate supply to reproductive tissue (Isagi et al. 1997; Piovesan & Adams 2001). Consequently, it was concluded that the increased supply of N to reproductive tissue and increased photosynthetic capacity, either separately or in combination, were the most probable causes of increased seedfall mass following N application.

Variation in climatic dependencies

Variation in the significance of the correlations between seedfall mass and rainfall during resource priming indicated that rainfall does not synchronize seed production solely by influencing endogenous plant processes. Rainfall influences soil moisture, which is an important driver of N mineralization (Walse, Berg & Sverdrup 1998; Paul et al. 2003). In this study, this relationship was clearly evident in the positive correlation between rainfall and foliar N concentrations among years. As N supply is increased in fertilized plots, the relative importance of rainfall to N availability should be diminished relative to the unfertilized plots during the resource priming phase. This manifests itself in the lower strength of the correlation of rainfall with seedfall mass in the fertilized plots (Fig. 4). Variation in the correlation between rainfall during resource priming and seedfall mass with stand age is also probably due to the role of N availability. Pole stands of mountain beech contain substantially more N in stem wood biomass than sapling stands (Clinton, Allen & Davis 2002), allowing greater masses of in planta N reserves to be allocated to meet nutritional demands of reproduction (Miller, Cooper & Miller 1992; Yasumaru, Hikosaka & Hirose 2006); consequently, rainfall is less important to promoting seed production.

The uniform negative correlations between seedfall mass and temperature during resource priming (Fig. 4) agree with both Piovesan & Adams (2001) and Richardson et al. (2005). As temperature was not correlated to foliar N concentrations, this suggests that synchronicity in mast seeding initiated by temperature during resource priming is unrelated to any effect on N availability. Cooler temperatures during resource priming may act to synchronize reproductive effort in plants by influencing endogenous processes (Richardson et al. 2005), while it is also possible that cooler temperatures are important because they limit reproductive effort in the following year (refer Fig. 1), allowing more resources to be available 2 years later (Norton & Kelly 1988).

Increased temperatures during floral primordia development are associated with heavy flowering and enhanced seed production in various species (e.g. Schauber et al. 2002; Kon et al. 2005; Kelly et al. 2008). However, significant correlations between temperature during floral primordia development and seedfall mass in this study were only observed in fertilized and pole stands where rainfall during resource priming was not correlated with seedfall mass (Fig. 4). This suggests that variations in the importance of climatic cues during resource priming influence the strength of correlations between seedfall mass and climatic cues during floral primordia development. Consequently, N availability during resource priming may also indirectly affect the synchronization of seed production by climatic cues during later stages of reproductive development.

Further evidence that the cueing effect of rainfall on seedfall mass in mountain beech is associated with an effect on N availability is provided by comparisons of the optimized multiple regression models (Table 2). As the multiple regression models all contained only two terms, the order in which the terms were added to the model does not influence the relative explanatory power of each term. Therefore, the high partial r2 values presented in Table 2 confirm that the decreases in the relative importance of rainfall during resource priming have been induced by increased N availability in the fertilized stands and most likely a lesser reliance on external N supply in the pole stands. The concomitant increases in the relative importance of temperature during floral primordia development in the fertilized and pole stand plots agree with the individual correlations given in Fig. 4. These variations in the relative degree of explanation associated with rainfall and temperature also lend further support to the proposition that alterations to the strength of a climatic cue during one stage of reproductive development affect the correlation between seed production and climatic cues during subsequent stages of reproductive development. Consequently, as well as demonstrating that the climatic cues may be influencing mast seeding by altering nutrient availability, tangible evidence is provided that sensitivity to climatic cues can be altered by changes to external nutrient supply. This has previously been simulated in models of flowering patterns in other mast-seeding species (Kelly & Sork 2002) and demonstrated in an assessment of the differences in reproductive effort in Chionochloa tussocks associated with natural variations in soil nutrient status (Hay, Kelly & Holdaway 2008).

It was also noted that there was no consistent interaction between the effect of fertilization on seedfall mass and the actual mass of seedfall from year to year, demonstrated by the lack of any trend in the ratios presented in Table 1. Various studies have concluded that increased productivity (facilitated through greater resource availability) causes relatively greater increases in seedfall mass in low seed production years relative to mast years (e.g. Richardson et al. 2005; Kelly et al. 2008), contrary to the results of this study. Examination of two previous studies (Fahey, Battles & Wilson 1998; Davis, Allen & Clinton 2004) suggests that the magnitude of any alterations to seed production does not always relate to the total mass of seed production in tree species, as observed in our study, but the data sets in these two studies are relatively limited in terms of collection years and do not support comprehensive analysis. Therefore, the mechanism responsible for this departure from the behaviour observed by Richardson et al. (2005) and Kelly et al. (2008) cannot be explained at this time.

Stimulated variations in reproductive effort

This study provides some evidence that altering the relative importance of climatic cues by manipulating nutrient supply can influence the frequency of mast events. Various authors (e.g. Satake & Iwasa 2000; Rees, Kelly & Bjørnstad 2002; Newbery, Chuyong & Zimmermann 2006) have postulated that mast seeding is precipitated first by exceeding a resource threshold, then by favourable climatic conditions during flowering. As illustrated in Fig. 1, sub-optimal climatic conditions during resource priming may prevent the initiation of increased seed production in unfertilized stands, whereas the same conditions in fertilized stands may be sufficient to pass the threshold for N supply and precipitate mast-seeding behaviour. If climatic cues (e.g. temperature) during subsequent phases of reproductive development are favourable, this may create conditions inducing greatly increased seed production in resource-rich sites but not others. We suggest this pathway may be related to the greatly increased seed production observed in fertilized plots but not in unfertilized plots in 2008. This scenario in Fig. 1 also illustrates how fertilization can shift more explanatory power to temperature during floral primordia development (see Table 2) as this climatic cue acquires an increased likelihood of determining the extent of reproductive effort in fertilized plots.

Finally, the incidence of mast seeding was not affected by stand age, despite the differences in the relationship between rainfall and seed production, and temperature and seed production, being much greater than observed for the fertilization treatment (Table 2). This also suggests that other factors, such as pollen limitation (Crone, Miller & Sala 2009), are influencing the processes that initiate mast seeding, and are acting to maintain synchronization in mast seeding across different stands ages despite any differences in the effect of rainfall and temperature on seed production.

Conclusions

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

These results confirm that increased N availability can increase seed production, and demonstrate that the relationship between seed production and some climatic cues is related to an effect on resource supply. That the manipulation of resource availability can alter the synchronization of reproductive effort is important to improving understanding of how mast-seeding behaviour is instigated and synchronized, and further research is required to integrate the effects of fluxes in resource availability, either natural or anthropogenic in origin, with other processes known to influence the incidence and extent of mast seeding.

Acknowledgements

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

The authors thank Graeme Rogers, Dave Henley and Chris Morse for providing technical assistance and Rowland Burdon, Tracy Williams and Colleen Carlson for comments and advice during the preparation of the manuscript. We also wish to acknowledge the many useful comments and suggestions of the anonymous referees and editorial staff. This research was funded by the New Zealand Foundation for Research, Science and Technology Ecosystem Resilience Outcome Based Initiative (Contract No. C09X0502).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  • Allen, R.B., Clinton, P.W. & Davis, M.R. (1997) Cation storage and availability along a Nothofagus forest development sequence in New Zealand. Canadian Journal of Forest Research, 27, 323330.
  • Allen, R.B. & Platt, K.H. (1990) Annual seedfall variation in Nothofagus solandri (Fagaceae), Canterbury, New Zealand. Oikos, 57, 199206.
  • Bazzaz, F.A., Chiariello, N.R., Coley, P.D. & Pitelka, L.F. (1987) Allocating resources to reproduction and defense. BioScience, 37, 5867.
  • Clinton, P.W., Allen, R.B. & Davis, M.R. (2002) Nitrogen storage and availability during stand development in a New Zealand Nothofagus forest. Canadian Journal of Forest Research, 32, 344352.
  • Crane, W.J.B. & Banks, J.C.G. (1992) Accumulation and retranslocation of foliar nitrogen in fertilised and irrigated Pinus radiata. Forest Ecology and Management, 52, 201223.
  • Crone, E.E., Miller, E. & Sala, A. (2009) How do plants know when other plants are flowering? Resource depletion, pollen limitation and mast-seeding in a perennial wildflower. Ecology Letters, 12, 11191126.
  • Davis, M.R., Allen, R.B. & Clinton, P.W. (2003) Carbon storage along a stand development sequence in a New Zealand Nothofagus forest. Forest Ecology and Management, 177, 313321.
  • Davis, M.R., Allen, R.B. & Clinton, P.W. (2004) The influence of N addition on nutrient content, leaf carbon isotope ratio, and productivity in a Nothofagus forest during stand development. Canadian Journal of Forest Research, 34, 20372048.
  • Evans, J.R. (1989) Photosynthesis and nitrogen relationships in leaves of C3 plants. Oecologia, 78, 919.
  • Fahey, T.J., Battles, J.J. & Wilson, G.F. (1998) Responses of early successional northern hardwood forests to changes in nutrient availability. Ecological Monographs, 68, 183212.
  • Han, Q., Kabeya, D., Iio, A. & Kakubari, Y. (2008) Masting in Fagus crenata and its influence on the nitrogen content and dry mass of winter buds. Tree Physiology, 28, 12691276.
  • Hay, J., Kelly, D. & Holdaway, R.J. (2008) Causes and consequences of frequent flowering on edges in the mast-seeding genus Chionochloa (Poaceae). New Zealand Journal of Ecology, 32, 8091.
  • Herrera, C.M., Jordano, P., Guitian, J. & Traveset, A. (1998) Annual variability in seed production by woody plants and the masting concept: reassessment of principles and relationships to pollination and seed dispersal. American Naturalist, 152, 576594.
  • Hewitt, A.E. (1992) New Zealand Soil Classification. Department of Scientific and Industrial Research Land Resources, Lower Hutt, New Zealand.
  • Isagi, Y., Sugimura, K., Sumida, A. & Ito, H. (1997) How does masting happen and synchronize? Journal of Theoretical Biology, 187, 231239.
  • Janzen, D.H. (1971) Seed predation by animals. Annual Review of Ecology and Systematics, 2, 465492.
  • Kelly, D. (1994) The evolutionary ecology of mast seeding. Trends in Ecology and Evolution, 9, 465470.
  • Kelly, D. & Sork, V.L. (2002) Mast seeding in perennial plants: why, how, where? Annual Review of Ecology and Systematics, 33, 427447.
  • Kelly, D., Turnbull, M.H., Pharis, R.P. & Sarfati, M.S. (2008) Mast seeding, predator satiation, and temperature cues in Chionochloa (Poaceae). Population Ecology, 50, 343355.
  • Koenig, W.D. & Knops, J.M.H. (2005) The mystery of masting in trees. American Scientist, 93, 340347.
  • Kon, H., Noda, T., Terazawa, K., Koyama, H. & Yasaka, M. (2005) Proximate factors causing mast seeding in Fagus crenata: thee effect of resource level and weather cues. Canadian Journal of Botany, 83, 14021409.
  • Le Tacon, F. & Oswald, H. (1977) Influence de la fertilisation minérale sur la fructification du Hêtre (Fagus sylvatica). Annals of Forest Science, 34, 89109.
  • Mduma, S.A.R., Sinclair, A.R.E. & Turkington, R.O.Y. (2007) The role of rainfall and predators in determining synchrony in reproduction of savanna trees in Serengeti National Park, Tanzania. Journal of Ecology, 95, 184196.
  • Miller, H.G., Cooper, J.M. & Miller, J.D. (1992) Response of pole-stage Sitka spruce to application of fertilizer nitrogen, phosphorous and potassium in upland Britain. Forestry, 65, 1533.
  • Miyazaki, Y., Hiura, T., Kato, E. & Funada, R. (2002) Allocation of resources to reproduction in Styrax obassia in a masting year. Annals of Botany, 89, 767772.
  • Newbery, D.M., Chuyong, G.B. & Zimmermann, L. (2006) Mast fruiting of large ectomycorrhizal African rain forest trees: importance of dry season intensity, and the resource-limitation hypothesis. New Phytologist, 170, 561579.
  • Nilsson, S.G. & Wästljung, U. (1987) Seed predation and cross-pollination in mast-seeding beech (Fagus sylvatica) patches. Ecology, 68, 260265.
  • Norton, D.A. & Kelly, D. (1988) Mast seeding over 33 years by Dacrydium cupressinum Lamb. (rimu) (Podocarpaceae) in New Zealand: the importance of economies of scale. Functional Ecology, 2, 399408.
  • Övergaard, R., Gemmel, P. & Karlsson, M. (2007) Effects of weather conditions on mast years in beech (Fagus sylvatica L.) in Sweden. Forestry, 80, 555565.
  • Paul, K.I., Polglase, P.J., O’Connell, A.M., Carlyle, J.C., Smethurst, P.J. & Khanna, P.K. (2003) Defining the relation between soil water content and net nitrogen mineralization. European Journal of Soil Science, 54, 3947.
  • Piovesan, G. & Adams, J.M. (2001) Masting behaviour in beech: linking reproduction and climatic variation. Canadian Journal of Botany, 79, 223240.
  • R Development Core Team (2010) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. Available at: http://www.R-project.org.
  • Raison, R.J., Connell, M.J., Khanna, P.K. & Falkiner, R.A. (1992) Effects of irrigation and nitrogen fertilisation on fluxes of soil mineral nitrogen in a stand of Pinus radiata. Forest Ecology and Management, 52, 4364.
  • Rees, M., Kelly, D. & Bjørnstad, O.N. (2002) Snow tussocks, chaos and the evolution of mast seeding. American Naturalist, 160, 4459.
  • Richardson, S.J., Allen, R.B., Whitehead, D., Carswell, F.E., Ruscoe, W.A. & Platt, K.H. (2005) Climate and net carbon availability determine temporal patterns of seed production by Nothofagus. Ecology, 86, 972981.
  • Satake, A. & Iwasa, Y. (2000) Pollen coupling of forest trees: forming synchronized and periodic reproduction out of chaos. Journal of Theoretical Biology, 203, 6384.
  • Schauber, E.M., Kelly, D., Turchin, P., Simon, C., Lee, W.G., Allen, R.B., Payton, I.J., Wilson, P.R., Cowan, P.E. & Brockie, R.E. (2002) Masting by eighteen New Zealand plant species: the role of temperature as a synchronizing cue. Ecology, 83, 12141225.
  • Selås, V., Piovesan, G., Adams, J.M. & Bernabei, M. (2002) Climatic factors controlling reproduction and growth of Norway spruce in southern Norway. Canadian Journal of Forest Research, 32, 217225.
  • Walse, C., Berg, B. & Sverdrup, H. (1998) Review and synthesis of experimental data on organic matter decomposition with respect to the effect of temperature, moisture and acidity. Environmental Reviews, 6, 2540.
  • Wardle, J.A. (1984) The New Zealand Beeches. New Zealand Forest Service, Christchurch.
  • Yasumaru, Y., Hikosaka, K. & Hirose, T. (2006) Resource allocation to vegetative and reproductive growth in relation to mast seeding in Fagus crenata. Forest Ecology and Management, 229, 228233.