SEARCH

SEARCH BY CITATION

Keywords:

  • disturbance;
  • floodplain;
  • forest regeneration;
  • light;
  • microsites;
  • microtopography

Summary

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

1 Windstorm disturbances create a wide range of microsites which can have complex effects on forest regeneration patterns. We investigated the combined effects of light and microtopography on emergence, mortality and size of seedlings of two bottomland hardwood canopy tree species, Quercus michauxii and Liquidambar styraciflua, over a 2-year period. A split-plot design in experimental tanks represented the range of light levels and the pits and mounds found in a disturbed floodplain forest.

2 Emergence was always higher on mounds than in pits, except for L. styraciflua in full sunlight. For both species, mortality was consistently lower, and seedlings of both species grew better in both years on mounds. Light levels did not affect the two later stages.

3 There were species-specific interactions between the effects of two factors on seedling emergence. Lower emergence of L. styraciflua on mounds in full sunlight suggested that full sunlight at this stage can eliminate the advantage to later stages of being on a mound. The combined stresses of low light and a high water table significantly reduced emergence of Q. michauxii in pits at low light.

4 Microsites optimal for one regeneration component of a species were not always optimal for others as seen for L. styraciflua. The relative significance of environmental factors also varied with regeneration stages, such that neither light nor a light-water interaction influenced regeneration after emergence.

5 Environmental factors may have independent or interacting effects on regeneration, and the nature or presence of these effects can vary among demographic stages. Within each environmental combination, effects may be consistently positive or negative across stages; alternatively, demographic conflicts may develop.


Introduction

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

In forest communities, large-scale natural disturbances modify many resources simultaneously (Bazzaz 1996; Carlton & Bazzaz 1998b) and may cause them to become incongruent over space (Carlton & Bazzaz 1998b). This creates a complex array of microsites and, in highly disturbed areas, many combinations of environmental conditions can be present within a relatively small area (Peterson & Pickett 1995). Given that seedling establishment and growth can be influenced by multiple environmental factors (Bazzaz & Miao 1993; Pacala et al. 1994; Canham et al. 1996; Walters & Reich 1997; Carlton & Bazzaz 1998a), it may be useful to consider post-disturbance microsites as composites of several variables. Studies that separate the individual components and examine their interactions may indicate mechanisms of forest reestablishment (Grubb et al. 1996; Catovsky & Bazzaz 2000).

In some species, plant performance at several life history stages may be related to microsite conditions (Valverde & Silvertown 1998; Clark & Clark 1999; George & Bazzaz 1999a,b). The relative significance of site conditions can vary over time as plants respond to small-scale heterogeneity in ways that change with ontogeny (Parrish & Bazzaz 1985). Slight shifts in the balance between positive and negative effects of microsites may occur, as well as more substantial changes that produce opposing effects at different life history stages. Intraspecific seed-seedling conflicts, whereby seedling growth or survival are poor in sites conducive to seed survival or germination and vice versa (Schupp 1995), may produce seedling distributions that differ markedly from the initial spatial pattern of seeds and recruits following the dispersal and germination stages (Rey & Alcántara 2000). Similar conflicts have been reported between different aspects of seedling demography, as for example in the species-specific patterns of seedling mortality of four tropical tree species which did not always parallel seedling growth across the light availability gradient (Kobe 1999).

In forests regenerating after disturbance, the differential suitability of microsites for different life-history stages may affect seedling abundances and thus ultimately influence the species composition of the canopy. Interspecific differences in response patterns to microsite factors among life-history stages may also provide a subtle mechanism for partitioning of the regeneration niche (Grubb 1977). We therefore argue that studies of forest regeneration should tease apart the effects of individual environmental factors and their interactions on demographic stages.

The floodplain forest community provides a system in which such microsite phenomena can be investigated experimentally. Vegetation patterns are affected primarily by soil moisture and light gradients (Streng et al. 1989; Jones et al. 1994a,b; Hall & Harcombe 1998), and frequent disturbances such as flooding and windthrow create microtopographic and light combinations that affect recruitment patterns (Jones et al. 1994a; Battaglia 1998; Hall & Harcombe 1998; Battaglia et al. 1999).

We created microsites representative of a disturbed floodplain forest by experimentally manipulating light levels and depth to the water table in order to evaluate their joint effects on regeneration of two bottomland hardwood species. We addressed two questions: (i) what are the effects of these two factors on seedling emergence, mortality and size; and (ii) does the suitability of a microsite remain consistent for each component of regeneration?

Materials and methods

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

Species description

Quercus michauxii Nutt. (swamp chestnut oak) and Liquidambar styraciflua L. (sweetgum) are two canopy species common in bottomland hardwood forests (Kellison et al. 1998) of south-eastern North America. Quercus michauxii is a weakly flood-tolerant, moderately shade-intolerant (McKnight et al. 1981) species with large, heavy seeds (average = 5.35 g) (Schopmeyer 1974) that are dispersed by gravity and animals. Liquidambar styraciflua is considered moderately flood-tolerant (Broadfoot 1967) and shade-intolerant (Putnam et al. 1960), except in bottomlands where it is believed to be intermediate in shade tolerance (Fowells 1965). Its seeds are small (average = 5.56 mg) and wind-dispersed (Young & Young 1992).

In the autumn of 1994, prior to seed dispersal, L. styraciflua fruits were collected from two sites on the South Atlantic Coastal Plain. At least 50 fruits were collected from each of seven trees growing in forests on the Savannah River floodplain (N 81°40′, W 33°48′) and from three recently fallen trees in the Congaree Swamp National Monument (N 80°47′, W 33°48′). Fruits were allowed to ripen and dry in a greenhouse (Young & Young 1992), and seeds were then extracted from eight randomly selected fruits per tree. Dissection of approximately 100 L. styraciflua seeds showed that seeds weighing less than 2.5 mg rarely contained viable embryos and these were therefore discarded. The remaining larger seeds were placed individually in tubes and stratified at 5 °C until planting in order to break dormancy (Young & Young 1992).

Quercus michauxii acorns were also collected in the autumn of 1994 from 10 isolated trees in bottomland hardwood forests on the Savannah River floodplain. A minimum of 1000 acorns was collected at the base of each tree and stored in large plastic bags at 5 °C until planting. At 2-week intervals during the storage period, acorns were aerated, rinsed and examined before being returned to cold storage; seeds with damaged seed coats or obvious insect and fungal infections were discarded.

Experimental methods

Microsite treatments were established in 15 circular galvanized steel tanks lined with plastic (2.43 m in diameter and 0.61 m in depth) in a split-plot design. Light quantity was the main plot factor in this experimental design, and tanks were randomly assigned to one of three levels: low (20%), medium (55%), or full sunlight (100%) with five tanks at each light level. Medium and low light treatments were produced by placing tanks inside shade houses with shade cloth of the appropriate mesh size attached to the sides and ceiling. Four 165 L pots (0.58 m in depth and 0.76 m in diameter) containing a Rembert sandy loam soil, collected from the Savannah River floodplain, were placed in each tank, and the tanks were filled with water.

Depth to the water table was used as the subplot factor and two levels were created in each tank to simulate pit and mound microsites in bottomlands. Pots were either placed on submerged concrete blocks to obtain a depth to water table of 50 cm from the soil surface (mounds) or on the bottom of the tank (pits), in which case water level was maintained at or within 1 cm of the soil surface. Pots were perforated to facilitate waterlogging of soil to the appropriate depth and to maintain hydraulic connectivity between the water in the soil and in the tank. Tanks were topped up at least once a week to maintain the set depth to the water table. Soil surfaces were not watered, although they received natural rainfall, so that differences in moisture levels of the upper soil layers could develop, as expected in natural forests.

Treatment combinations were chosen to represent variation in microtopography and light following hurricane disturbance. Microtopography treatments were based on data from a field study initiated 5 years after the passage of Hurricane Hugo through the Congaree Swamp, an old-growth bottomland hardwood forest (Battaglia 1998). Within plots arrayed across the gradient of hurricane disturbance, the maximum difference in relative elevation between adjacent pits and mounds ranged from 26 to 98 cm, with a median of 64 cm, similar to our imposed difference of 50 cm. We acknowledge, however, that natural mounds and pits may differ in other ways such as in soil particle size distribution, litter accretion (Facelli & Pickett 1991), nutrient availability, erosion (Peterson et al. 1990) and temporal variability in hydrology that are not simulated in our experimental microsites.

Large blowdowns due to Hurricane Hugo resulted in losses of overstorey approaching 100% in the most severely disturbed sites (Pederson et al. 1997) and we therefore used full sunlight to mimic the highest level of canopy openness. Field observations of partially shaded microsites in areas recently subjected to hurricane disturbance suggested that 20% was an appropriate lower level. We did not attempt to simulate sunflecks, which may have an important effect on carbon gain (Chazdon 1988). Although we are examining the effects of experimental microsites, for brevity we will refer to microtopographic treatments as mounds and pits, and light levels as full (100%), medium (55%) and low (20%) sunlight.

In April 1995, seeds were removed from cold storage and weighed individually. Each species was randomly assigned to one mound and one pit microsite treatment in each tank. We randomly selected 80 seeds (eight from each of 10 trees) of each species, which were sown at regular intervals, approximately 10 cm apart in each assigned pot. Seeds were consistently covered by 2 cm of soil. Although both species would have been exposed to mycorrhizal fungi present in the soil used, we ensured mycorrhizal inoculation, as recommended for L. styraciflua seedlings (Kormanik et al. 1981), by adding a small amount of soil collected from beneath mature L. styraciflua individuals in the Savannah River floodplain to each pot containing seeds of that species. This procedure was not prescribed and therefore not repeated for Q. michauxii. The location of each propagule was marked, and emergence, mortality and height of each individual were monitored throughout the 1995 and 1996 growing seasons. Periodic weeding was necessary to eliminate competition from seedbank recruits of other species; individuals of the study species were not observed during the study except directly adjacent to our marked planting sites. Seedling emergence was monitored twice each week during the first 3 months (April–June) of the 1995 growing season and weekly thereafter. New seedlings were also censused weekly in April and May of 1996. Survival and growth were assessed in October 1995, April 1996 and October 1996. The height of each seedling was measured as distance from ground level to the tip of the apical bud or, if the apical bud had died, to the tip of the highest bud.

Survival and growth rates of Q. michauxii seedlings in the first year of the study were higher than expected on all mound microsites. We wished to exclude intraspecific competition and therefore chose to thin individuals of this species in this treatment to a constant density of 30 per pot immediately following leaf-out in 1996 (early April), thus eliminating potential internal shading effects in the second growing season. Seedlings were randomly selected, clipped at ground level and removed; stumps were monitored for subsequent sprouting, and sprouts were clipped weekly.

Data analyses

Analyses of emergence included only individuals emerging in 1995 because there were too few 1996 recruits for conventional statistical analyses. We examined cumulative mortality of 1995 recruits over the 2-year study period and weighted cumulative mortality by number of seedlings that emerged in 1995 for each pot. In addition, cumulative mortality of Q. michauxii seedlings on all mound microsites was adjusted to reflect thinning at the beginning of the 1996 growing season.

We used split-plot analyses of variance (proc MIXED, SAS 1996) (Littell et al. 1996) to test for differences in seedling emergence, mortality and height across light and depth-to-water table treatments, using the pot as our observational unit. To equalize variances for emergence and mortality, which are proportion data (Manly 1994; Underwood 1997), these variables were transformed to the logit scale (Hosmer & Lemeshow 1989), which may be more interpretable than the traditional arc-sine square-root transformation (P.M. Dixon, personal communication). The analysis is equivalent to a generalized linear mixed model because sample sizes in the split plots were equal in the case of emergence (n = 80) and were weighted with respect to number of individuals in the case of mortality. Percentages of zero or 100 are undefined when using the logit transformation, and we adjusted these values using (100/2n) and [100 − (100/2n)], respectively (P.M. Dixon, personal communication). Least-squares means and their 95% confidence intervals were calculated for individual treatment combinations when interactions were significant (P < 0.05) or for significant main effects when interactions were absent. All pairwise comparisons used adjusted Bonferroni t-tests.

Spearman rank correlations (proc CORR, SAS 1996) were used to examine the relationship between seed weight and first- and second-year seedling heights.

Results

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

Seedling emergence

Across all treatments, emergence was much higher in the first year than in the second year. Percent emergence for Q. michauxii and L. styraciflua in 1995 was 44.3% and 17.2%, respectively, compared to 1.2% and 6.0% in 1996. A significant light-microtopography interaction occurred for both species (Table 1). Quercus michauxii emergence was significantly higher on mounds than in pits (Fig. 1a); light levels had no effect on emergence on mounds and only low light, where no seedlings occurred, produced a significant difference in pits.

Table 1.  Summary of split-plot analysis of variance on 1995 seedling emergence of Quercus michauxii and Liquidambar styraciflua. Terms tested in the model include light level, distance to the water table (depth) and their interaction term. NDF, numerator degrees of freedom; DDF, denominator degrees of freedom
Fixed effectsNDFDDFFP
Q. michauxii
Light2128.690.0046
Depth112120.080.0001
Light × depth2127.570.0075
L. styraciflua
Light2120.790.4775
Depth11256.780.0001
Light × depth21212.920.0010
image

Figure 1. Emergence of (a) Quercus michauxii and (b) Liquidambar styraciflua seeds across light and microtopography treatment combinations. Means are shown with their 95% confidence intervals; those with the same letter are not significantly different from each other. ● = pit, ○ = mound.

Download figure to PowerPoint

Emergence of L. styraciflua seedlings was also significantly higher on mounds than in pits except in full sunlight (Fig. 1b), but light had no significant effect in either topographic treatment.

Seedling mortality

Seedlings of the two species exhibited similar patterns of cumulative mortality. There were no significant differences in mortality across light levels (Table 2), but depth to the water table was significant, with mortality higher in pits than on mounds (Fig. 2). In both species, mortality was related to seedling size with a disproportionate number of smaller individuals dying, regardless of cohort membership (Table 3).

Table 2.  Summary of anova on cumulative mortality of seedlings of Quercus michauxii and Liquidambar styraciflua that emerged in 1995. We weighted the values with respect to sample size due to unequal sample sizes across treatment combinations. Abbreviations as in Table 1.
Fixed effectsNDFDDFFP
Q. michauxii
Light2120.840.4575
Depth1715.540.0056
Light × depth272.140.1881
L. styraciflua
Light2121.890.1932
Depth1106.440.0295
Light × depth2103.890.0563
image

Figure 2. Cumulative mortality of (a) Quercus michauxii and (b) Liquidambar styraciflua seedlings across experimental microsites. Means are shown with their 95% confidence intervals; those with the same letter are not significantly different from each other. Light-microtopography interaction terms were not significant for either species, and means for pit and mound microsites include individuals pooled across light levels.

Download figure to PowerPoint

Table 3.  Percentage of seedling mortality of Quercus michauxii and Liquidambar styraciflua by size class calculated for first cohort and second cohort seedlings. Overwinter mortality estimates (first cohort) are based on deaths occurring between October 1995 and April 1996, and second growing season mortality is based on deaths occurring between April 1996 and October 1996. A dash (–) indicates no seedlings present in that cohort and size class
 Q. michauxiiL. styraciflua
Height class (cm)10/95–4/9614/96–10/9614/96–10/96210/95–4/9614/96–10/9614/96–10/962
  • 1

    Based on first cohort individuals only.

  • 2

    Based on second cohort individuals only.

  • 3

    This percentage is based on one individual out of 14 in the size class.

0–548293373156
5.1–101685003
10.1–152203
15.1–200100
20.1–2500073
25.1–300100

Seedling height

Patterns of seedling height in response to light and microtopography were similar for the two species. For both species in both years, final seedling height was significantly greater on experimental mounds than in pits (Fig. 3). Light had no effect on height (Table 4).

image

Figure 3. Seedling heights of (a) Quercus michauxii and (b) Liquidambar styraciflua across experimental microsites at the end of the 1995 and 1996 growing seasons. Height means are shown with their 95% confidence intervals; those with the same letter are not significantly different from each other. Light-microtopography interaction terms were not significant for either species, and means for pit and mound microsites include individuals pooled across light levels. ● = pit, ○ = mound.

Download figure to PowerPoint

Table 4.  Summary of anova on seedling height at the end of the 1995 and 1996 growing seasons. Abbreviations as in Table 1
 19951996
Fixed effectsNDFDDFFPNDFDDFFP
Q. michauxii
Light2121.810.20632121.240.3235
Depth1665.150.00021646.430.0005
Light × depth160.300.6036160.140.7229
L. styraciflua
Light2122.890.09432122.790.1010
Depth1819.620.002218104.110.0001
Light × depth280.230.8015283.100.1009

Seed weights of Q. michauxii were weakly but positively correlated with seedling heights recorded at the end of the first (n = 1035, r = 0.17, P = 0.0001) and second (n = 528, r = 0.13, P = 0.0039) growing seasons. At the end of the first growing season, L. styraciflua seedling height was also weakly correlated with seed weight (n = 370, r = 0.18, P = 0.0008), but this relationship had disappeared by the end of the second growing season (n = 395, r = 0.08, P = 0.11).

Discussion

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

Responses to microsite components

Our results suggest that uncoupling the effects of light and water is a useful approach for studying the post-disturbance reestablishment of woody seedlings in these wetland forests. We were able to detect interactions between these factors and to isolate differences in responses between species and among regeneration components. For both species studied, one component of regeneration was affected by the interaction between the two factors, while the others were influenced only by water.

Interactions between the two factors had negative effects on seedling emergence of both species. Emergence of Q. michauxii was consistently higher on mounds than in pits across all light levels, but emergence in pits was also significantly lower at the lowest light level (no emergence) relative to pits with more light. The combination of low light and a high water table (or flooding) therefore constitutes a very restrictive environmental filter for some floodplain species (Menges & Waller 1983; Hall & Harcombe 1998).

Emergence of L. styraciflua was significantly higher on mounds than in pits, except in full sunlight where L. styraciflua emergence was always low. This poor performance is probably due to stress in both microsites albeit of different types. Light and moisture do not necessarily vary independently, and soil moisture retention may therefore be lower under high light conditions due to greater evaporation, while other factors such as humidity and temperature may not change concomitantly. Seeds on mounds in full sunlight may have experienced both heat and moisture stress (Becker et al. 1988) in the upper layers of the soil; at other light levels, the much higher emergence on mounds suggests that light was the pivotal influence. Such stresses, along with potential propagule limitation, may contribute to low seedling densities in the centres of some large natural gaps where light levels are high (Popma et al. 1988). Emergence in pits was low, presumably because the stress imposed by a high water table could not be mitigated by elevated light levels.

In both species, cumulative mortality and both first- and second-year seedling size were affected by water table depth, but not by light levels, or at least those included in our study. These aspects of plant performance generally showed a similar response to topography as did emergence. Mortality of seedlings was high in pits and low on mounds, and seedling size was greater on mounds particularly in the second year. Interestingly, mortality and seedling size were related. Smaller seedlings, particularly ones with delayed emergence, had lower survival, perhaps because of competition with ones that established in the first year of the study. Early colonizers may capture space and usurp resources, resulting in a decline in survival with successive cohorts (Peterson & Pickett 1995) or within a cohort over time (Jones & Sharitz 1989; Streng et al. 1989; Jones et al. 1994b). Seedlings that establish later probably experience reduced photosynthesis under a canopy of earlier germinants or a more completely leafed-out overstorey.

The overall poorer performance of both species in experimentally created pits indicates that areas with a high water table are likely to be very poor regeneration microsites, even for floodplain species, such as L. styraciflua and Q. michauxii. These findings support reports that pits and other low elevation areas are strong filters on recruitment in bottomland hardwood forests (Huenneke & Sharitz 1986; Streng et al. 1989; Titus 1990; Battaglia 1998; Jones & Sharitz 1998) and suggest that these microsites may function as propagule sinks. In floodplains, elevated sites such as mounds generally have a higher number of surviving seedlings compared to lower elevation microsites, especially in wet years when flooding may be extensive (Huenneke & Sharitz 1986; Streng et al. 1989). Although the hydrologic regime is known to be the major environmental influence shaping structure and composition of bottomland communities (Wharton et al. 1982; Sharitz & Mitsch 1993; Hall & Harcombe 1998), we still lack basic information about the effects of fine-scale hydrologic variation on the demography of floodplain species, as well as on the spatial distribution and composition of the seedling layer.

Differential resource investment in propagules between and within these species may explain some of the patterns we observed. The greater amounts of stored resources in Q. michauxii may have contributed to the overall high seedling emergence and the relationship between seedling size and seed weight in this species in both years of our study. By comparison, emergence of the light-seeded L. styraciflua was low and although first year seedling size was again related to seed weight, there was no relationship in the second year. Propagule size may have a more prolonged effect on seedlings of heavy-seeded species, and initial growth of their seedlings may depend more on stored reserves (seed weight) than on light availability. Heavy-seeded species may therefore have an early advantage, but a reversal can occur as light-seeded species become established and grow.

Microsite complexity and suitability

The relative significance of environmental factors can vary with respect to regeneration components in at least two ways. We found limited evidence for changes in both the number of factors significant to plant performance and the suitability of microsites, which may indicate demographic conflicts. However, many regeneration patterns are possible (Fig. 4). A scenario in which the same two interacting factors consistently affect all stages is shown in Fig. 4(a), whereas both species studied here behave as in Fig. 4(b), with factor 2 (light) affecting seedling emergence through its interaction with factor 1 (water) but having no influence on first or second year seedling size. Environmental combinations optimal for one aspect of regeneration may be sub-optimal for others (Kobe 1999) as shown in Fig. 4(c,d) (with and without interactions). Such a seed-seedling conflict may develop between components of regeneration.

image

Figure 4. Hypothetical regeneration patterns along two environmental gradients. Each panel shows performance trends for three regeneration measures, representing different life stages (e.g. seedling emergence, mortality and growth rate): (a) there is a significant interaction between the two factors, which is consistent across all measures; (b) the number of significant factors differs among measures—both factors affect measure 1, but factor 2 has no effect on measures 2 and 3; (c) interactions between factors affect all measures, and environmental combinations sub-optimal for measure 1 are optimal for measures 2 and 3; (d) factors have additive effects on all measures, and environmental combinations sub-optimal for measure 1 are optimal for measures 2 and 3.

Download figure to PowerPoint

Although there were no statistical differences in L. styraciflua or Q. michauxii seedling heights across light levels in 1996, we were interested in examining the fate of recruits at each combination of light and microtopographic conditions, given that interaction of these factors had significantly affected initial seedling establishment. Microsites had consistently positive or negative effects on all aspects of plant performance, except on mounds in full sunlight where L. styraciflua emergence was low but both seedling survival and final seedling size were high (mean survival = 73.6% (23 seedlings), mean height = 33.5 ± 3.1 cm (mean ± 1 SE), five replicates). This suggests a seed-seedling conflict as described by Schupp (1995), whereby full sunlight mounds are not conducive to seedling emergence, but the seedlings that do establish there tend to thrive. Enhanced seedling size might, however, be the result of lower densities and therefore greater resource availability, and such possibilities should be investigated where apparent demographic conflicts are indicated by the suitability of a given microsite seeming to improve with successive regeneration stages. With larger sample sizes, our study may have produced clearer evidence to resolve this issue.

Our results suggest that microsites in these forests should be defined by both light and water because changes in the levels of one or both factors can affect plant performance during regeneration. Pits, mounds and other microtopographic features in the understorey of bottomland forests may provide very distinct regeneration opportunities relative to those found at other points along the light availability gradient. Propagules and seedlings are exposed to many environmental combinations, particularly in naturally disturbed forests where light and soil moisture can be incongruent at a small scale (Carlton & Bazzaz 1998b), and identifying suitable microsites for each regeneration component can be very difficult for some species. The collective effects of these factors and potential variation in microsite suitability over time may influence population processes and vegetation patterns in spatially complex and temporally dynamic floodplain forests (Streng et al. 1989; Hall & Harcombe 1998).

To understand the effects of disturbance on forest composition and diversity, we must understand better the mechanisms of regeneration (Grubb 1977) and potential effects of interacting environmental factors (Bazzaz & Wayne 1994). Although small-scale abiotic heterogeneity following disturbance is ephemeral (Denslow & Hartshorn 1994), the effects of that initial physical template on germination patterns and the seedling pool may have a residual impact on forest composition and dynamics.

Acknowledgements

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

We are grateful to C. Grimsley for her invaluable help in setting up and maintaining this study. We thank B. Allen, C. Benenhaley, H. Brown, E. Cantonwine, C. King, R. Lide, N. McArthur, B. Moyer, J. Singer, P. Stankus and B. Summer for technical support. We thank B. S. Collins, J. S. Denslow, L. Haddon, K. W. McLeod, P. R. Minchin, C. J. Peterson and two anonymous referees whose suggestions improved this manuscript. The authors acknowledge P. M. Dixon for his statistical advice and help with experimental design. We acknowledge J. S. Denslow for providing computer support during the latter stages of manuscript preparation. This research was funded by Financial Assistance Award Number DE-FC09-96SR18546 between the University of Georgia's Savannah River Ecology Laboratory and the Department of Energy and a grant from the Society of Wetland Scientists to the senior author.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Battaglia, L.L. (1998) Microsite heterogeneity and regeneration patterns along a hurricane disturbance gradient in an old-growth bottomland hardwood forest. PhD Thesis. University of Georgia, Athens, Georgia, USA.
  • Battaglia, L.L., Sharitz, R.R. & Minchin, P.R. (1999) Patterns of seedling and overstory composition along a gradient of hurricane disturbance in an old-growth bottomland hardwood community. Canadian Journal of Forest Research, 29, 144156.
  • Bazzaz, F.A. (1996). Plants in Changing Environments: Linking Physiological, Population, and Community Ecology. Cambridge Press, Cambridge, UK.
  • Bazzaz, F.A. & Miao, S.L. (1993) Successional status, seed size, and responses of tree seedlings to CO2, light, and nutrients. Ecology, 74, 104112.
  • Bazzaz, F.A. & Wayne, P.M. (1994) Coping with environmental heterogeneity: the physiological ecology of tree seedling regeneration across the gap–understory continuum. Exploitation of Environmental Heterogeneity by Plants: Ecophysiological Processes Above- and Below-ground (eds M.M.Caldwell & R.W.Pearcy), pp. 349390. Academic Press, San Diego.
  • Becker, P., Rabenold, P.E., Idol, J.R. & Smith, A.P. (1988) Water potential gradients for gaps and slopes in a Panamanian tropical moist forest's dry season. Journal of Tropical Ecology, 4, 173184.
  • Broadfoot, W.M. (1967) Shallow-water impoundment increases soil moisture and growth of hardwoods. Soil Science Society of America Proceedings, 31, 562564.
  • Canham, C.D., Berkowitz, A.R., Kelly, V.R., Lovett, G.M., Ollinger, S.V. & Schnurr, J. (1996) Biomass allocation and multiple resource limitation in tree seedlings. Canadian Journal of Forest Research, 26, 15211530.
  • Carlton, G.C. & Bazzaz, F.A. (1998a) Regeneration of three sympatric birch species on experimental hurricane blowdown microsites. Ecological Monographs, 68, 99120.
  • Carlton, G.C. & Bazzaz, F.A. (1998b) Resource congruence and forest regeneration following an experimental hurricane blowdown. Ecology, 79, 13051319.
  • Catovsky, S. & Bazzaz, F.A. (2000) The role of resource interactions and seedling regeneration in maintaining a positive feedback in hemlock stands. Journal of Ecology, 88, 100112.DOI: 10.1046/j.1365-2745.2000.00428.x
  • Chazdon, R.L. (1988) Sunflecks and their importance to forest understorey plants. Advances in Ecological Research, 18, 163.
  • Clark, D.A. & Clark, D.B. (1999) Assessing the growth of tropical rain forest trees: issues for forest modeling and management. Ecological Applications, 9, 981997.
  • Denslow, J.S. & Hartshorn, G.S. (1994) Tree-fall gap environments and forest dynamic processes. La Selva: Ecology and Natural History of a Neotropical Rain Forest (eds L.McDade, K.Bawa, H.A.Hespenheide & G.S.Hartshorn.), pp. 120127. University of Chicago Press, Chicago.
  • Facelli, J.M. & Pickett, S.T.A. (1991) Plant litter: dynamics and effects on plant community structure and dynamics. Botanical Review, 57, 132.
  • Fowells, H.A. (ed. ) (1965) Silvics of forest trees of the United States. Agriculture Handbook 271. US Department of Agriculture, Washington, D.C.
  • George, L.O. & Bazzaz, F.A. (1999a) The fern understory as an ecological filter: emergence and establishment of canopy-tree seedlings. Ecology, 80, 833845.
  • George, L.O. & Bazzaz, F.A. (1999b) The fern understory as an ecological filter: growth and survival of canopy-tree seedlings. Ecology, 80, 846856.
  • Grubb, P.J. (1977) The maintenance of species richness in plant communities: the importance of the regeneration niche. Biological Reviews, 52, 107145.
  • Grubb, P.J., Lee, W.G., Kollmann, J. & Wilson, J.B. (1996) Interaction of irradiance and soil nutrient supply on growth of seedlings of ten European tall-shrub species and Fagus sylvatica. Journal of Ecology, 84, 827840.
  • Hall, R.B.W. & Harcombe, P.A. (1998) Flooding alters apparent position of floodplain saplings on a light gradient. Ecology, 79, 847855.
  • Hosmer, D.W. & Lemeshow, S. (1989) Applied Logistic Regression. Wiley Series In Probability and Mathematical Statistics. John Wiley & Sons, New York.
  • Huenneke, L.F. & Sharitz, R.R. (1986) Microsite abundance and distribution of woody seedlings in a South Carolina cypress-tupelo swamp. American Midland Naturalist, 115, 328335.
  • Jones, R.H. & Sharitz, R.R. (1989) Potential advantages and disadvantages of germinating early for trees in floodplain forests. Oecologia, 81, 443449.
  • Jones, R.H. & Sharitz, R.R. (1998) Survival and growth of woody plant seedlings in the understorey of floodplain forests in South Carolina. Journal of Ecology, 86, 574587.
  • Jones, R.H., Sharitz, R.R., Dixon, P.M., Segal, D.S. & Schneider, R.L. (1994b) Woody plant regeneration in four floodplain forests. Ecological Monographs, 64, 345367.
  • Jones, R.H., Sharitz, R.R., James, S.M. & Dixon, P.M. (1994a) Tree population dynamics in seven South Carolina mixed-species forests. Bulletin of the Torrey Botanical Club, 121, 360368.
  • Kellison, R.C., Young, M.J., Braham, R.R. & Jones, E.J. (1998) Major alluvial floodplains. Southern Forested Wetlands: Ecology and Management (eds M.G.Messina & W.H.Conner), pp. 291341. CRC Press LLC, Boca Raton, Florida.
  • Kobe, R.K. (1999) Light gradient partitioning among tropical tree species through differential seedling mortality and growth. Ecology, 80, 187201.
  • Kormanik, P.P., Bryan, W.C. & Schultz, R.C. (1981) Effects of three vesicular-arbuscular mycorrhizal fungi on sweetgum seedlings from nine mother trees. Forest Science, 27, 327335.
  • Littell, R.C., Milliken, G.A., Stroup, W.W. & Wolfinger, R.D. (1996) SAS System for Mixed Models. SAS Institute Inc., Cary, NC.
  • Manly, B.F.J. (1994) The Design and Analysis of Research Studies. Cambridge University Press, Cambridge, UK.
  • McKnight, J.S., Hook, D.D., Langdon, O.G. & Johnson, R.L. (1981) Food tolerance and related characteristics of trees of the bottomland forests of the southern United States. Wetlands of Bottomland Hardwood Forests (eds J.R.Clark & J.Benforado), pp. 2969. Elsevier, Amsterdam.
  • Menges, E.S. & Waller, D.M. (1983) Plant strategies in relation to elevation and light in floodplain herbs. American Naturalist, 122, 454473.
  • Pacala, S.W., Canham, C.D., Silander, J.A. & Kobe, R.K. (1994) Sapling growth as a function of resources in a north temperate forest. Canadian Journal of Forest Research, 24, 21722183.
  • Parrish, J.A.A. & Bazzaz, F.A. (1985) Ontogenetic niche shifts in old-field annuals. Ecology, 66, 12961302.
  • Pederson, N.A., Jones, R.H. & Sharitz, R.R. (1997) Age structure and possible origins of old Pinus taeda stands in a floodplain forest. Journal of the Torrey Botanical Society, 124, 111123.
  • Peterson, C.J., Carson, W.P., McCarthy, B.C. & Pickett, S.T.A. (1990) Microsite variation and soil dynamics within newly created treefall pits and mounds. Oikos, 58, 3946.
  • Peterson, C.J. & Pickett. S.T.A. (1995) Forest reorganization: a case study in an old-growth forest catastrophic blowdown. Ecology, 76, 763774.
  • Popma, J., Bongers, F., Martínez-Ramos, M. & Veneklaas, E. (1988) Pioneer species distribution in treefall gaps in Neotropical rain forest: a gap definition and its consequences. Journal of Tropical Ecology, 4, 7788.
  • Putnam, J.A., Furnival, G.M. & McKnight, J.S. (1960) Management and inventory of southern hardwoods. Agriculture Handbook 181. US Department of Agriculture, Washington, D.C.
  • Rey, P.J. & Alcántara, J.M. (2000) Recruitment dynamics of a fleshy-fruited plant (Olea europaea): connecting patterns of seed dispersal to seedling establishment. Journal of Ecology, 88, 622633.DOI: 10.1046/j.1365-2745.2000.00472.x
  • SAS Institute Inc. (1996) SAS/STAT® Software: Changes and Enhancements Through Release 6.12. SAS Institute Inc., Cary, NC.
  • Schopmeyer, C.S. (1974) Seeds of woody plants in the United States. Agriculture Handbook 450. US Department of Agriculture, Washington, D.C.
  • Schupp, E.W. (1995) Seed-seedling conflicts, habitat choice, and patterns of early recruitment. American Journal of Botany, 82, 399409.
  • Sharitz, R.R. & Mitsch, W.J. (1993) Southern floodplain forests. Biodiversity of the Southeastern United States/ Lowland Terrestrial Communities (eds W.H.Martin, S.G.Boyce & A.C.Echternacht), pp. 311372. J. Wiley & Sons, Inc., New York.
  • Streng, D.R., Glitzenstein, J.S. & Harcombe, P.A. (1989) Woody seedling dynamics in an east Texas floodplain forest. Ecological Monographs, 59, 177204.
  • Titus, J.H. (1990) Microtopography and woody plant regeneration in a hardwood floodplain swamp in Florida. Bulletin of the Torrey Botanical Club, 117, 429437.
  • Underwood, A.J. (1997) Experiments in Ecology: their Logical Design and Interpretation Using Analysis of Variance. Cambridge University Press, Cambridge, UK.
  • Valverde, T. & Silvertown, J. (1998) Variation in the demography of a woodland understorey herb (Primula vulgaris) along the forest regeneration cycle: projection matrix analysis. Journal of Ecology, 86, 545562.
  • Walters, M.B. & Reich, P.B. (1997) Growth of Acer saccharum seedlings in deeply shaded understories of northern Wisconsin: effects of nitrogen and water availability. Canadian Journal of Forest Research, 27, 237247.
  • Wharton, C.H., Kitchens, W.M., Pendleton, E.C. & Sipe, T.W. (1982) The ecology of bottomland hardwood swamps of the southeast: a community profile. U.S. Fish and Wildlife Service, Biological Services Program PL, Washington, D.C.
  • Young, J.A. & Young, C.G. (1992) Seeds of Woody Plants in North America. Dioscorides Press, Portland, Oregon.

Received 1 April 1999 revision accepted 19 June 2000