| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Botany, University of Florida, Gainesville, Florida 32611
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
5 wk to assess recovery. Salt tolerance was assessed as plant survival under saline conditions and as the ability to retain green leaf tissue under saline conditions. Using either criterion, the rank order of seedling salt tolerance correlated significantly (P 
0.05) with that expected based on species zonation near the coast. Agreement was better, however, using retention of leaf tissue as an index of salt tolerance. Species from forest zones that were frequently exposed to tidal water retained green leaf tissue under saline conditions. Species from zones occasionally subjected to very high tides were shoot killed but resprouted following removal of salt from the root zone. Those restricted to zones exposed only to infrequent storm surges died at salinities 4 g/L. Thus, differential seedling salt tolerance was consistent with tree zonation and, although the ability of young seedlings to resprout following salt removal did not appear to allow tree establishment at the extreme seaward margin of the forest, it appeared important in intermediate zones.
Key Words: Juniperus virginiana var. silicicola; • Quercus virginiana • resprouting • Sabal palmetto • salinity • salt tolerance • sea level • seedling survival
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tree species zonation at the coastal fringe of hydric hammock on the west coast of Florida is correlated with frequency of tidal flooding (Williams et al., unpublished data). Hydric hammocks are seasonally flooded, mixed-hardwood forests of the southeastern United States that are largely restricted to Florida (Vince, Humphrey, and Simons, 1989) and occur most extensively in Florida's Big Bend region on the Gulf of Mexico (Simons, Vince, and Humphrey, 1989). Although hydric hammocks are rich in species, the coastal fringe is dominated almost exclusively by Sabal palmetto (Walt.) Lodd. ex Schultes (cabbage palm), Juniperus virginiana var. silicicola (Small) Silba (southern red cedar), Quercus virginiana Mill. (live oak), and occasionally Pinus taeda L. (loblolly pine). In a transect studied by Vince, Humphrey, and Simons (1989) Sabal palmetto, Juniperus virginiana var. silicicola, and Quercus virginiana dominated the forest between their most coastward sampling point (150 m from the edge of salt marsh) and 2 km inland. Pinus taeda occurred as an occasional tree at the forest edge, Ulmus crassifolia Nutt. (cedar elm) was found 200–300 m from the edge, and Ulmus americana L. var. floridana (Chapm.) Little (Florida elm), Celtis laevigata Willd. (sugarberry), and Acer floridanum (Chapm.) Pax (Florida maple, syn.: Acer barbatum Michx.) were found starting at 1 km from the edge. Liquidambar styraciflua L. (sweetgum), Quercus laurifolia Michx. (swamp laurel oak), and Carpinus caroliniana Walt. (hornbeam) were common 2 km from the coast. Similarly, Thompson (1980) found Liquidambar styraciflua increasing in importance and Juniperus virginiana var. silicicola decreasing in importance with distance from the marsh. Her study, however, did not address zonation at the forest edge (<0.6 km from the marsh). Studies of vegetation change at the coastal margin of hydric hammock (Williams et al., in press) revealed that Sabal palmetto was the only tree species in stands frequently flooded by tidal saltwater. Stands flooded somewhat less frequently supported both Sabal palmetto and Juniperus virginiana var. silicicola, whereas stands flooded only a few times a year supported living Sabal palmetto, Juniperus virginiana var. silicicola, Quercus virginiana, and Celtis laevigata. Stands on higher ground or further inland supported a wide variety of tree species. For several species, mature trees existed in more frequently flooded sites than did their seedlings, suggesting that forest zones were retreating to higher elevations. This pattern also suggested that tree regeneration was more sensitive than mature trees to rising seas and saltwater flooding.
We undertook this study to determine whether the relative salt tolerance of seedlings of the dominant tree species could explain their zonation near the coast. Eight species were selected based on their abundance in existing forest-monitoring plots near Turtle Creek, Waccasassa Bay, Florida (Williams et al., in press) and in the adjacent forest. Sabal palmetto was expected to be the most salt tolerant, based on its complete dominance of the stands most frequently flooded by tidal water. Juniperus virginiana var. silicicola was expected to be the second most salt tolerant, followed by Quercus virginiana and then Celtis laevigata, based on their zonation across a gradient of tidal flooding frequency. Pinus taeda was expected to be slightly less salt tolerant, followed by Ulmus alata Michx. (winged elm) and Acer floridanum, common species in plots 180–400 m from the edge of the marsh where saltwater flooding occurred only during very violent storms and hurricanes. Liquidambar styraciflua, a species common farther inland than any of the existing plots (but still in a zone subject to occasional storm surges), was expected to be least salt tolerant. This ranking is consistent with the distribution of species found by Vince, Humphrey, and Simons (1989) with one exception. In their transect, Celtis laevigata was found 1 km from the coast and inland, whereas in plots established by Williams et al. (in press), some Celtis laevigata were found in areas occasionally flooded by tidal waters.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Seeds and seedlings were collected from sites in Florida on or near the Gulf of Mexico between 27°30' N and 29°15' N. Because intraspecific variation in salt tolerance has been demonstrated for some tree seedlings (e.g., Allen, Chambers, and McKinney, 1994), we included material from a variety of sites for each species to avoid having interspecific differences unduly influenced by any single source of genetic stock. Seeds were planted in potting soil in a greenhouse to germinate. Seedlings were grown in potting soil in the greenhouse prior to commencement of the experiment.
In March 1995, seedlings were transplanted to plastic pots 6.5 cm in diameter x 36 cm tall (Deepots®, Stuewe and Sons, Inc., Corvallis, Oregon, USA). One seedling was planted in each pot. Half the plants were grown in soil, a loamy sand that was collected from a hydric hammock ~0.5 km from the nearest finger of salt marsh vegetation. The other half were grown in sand culture using a very coarse commercially available silica sand (~0.8–2.2 mm grain diameter). Plant performance in soil was expected to more accurately reflect plant performance in the field, where tides and storm surges bring a combination of salt and flooding. However, because chemical reactions in flooded soil could affect plant performance and mask effects of salt, we grew half the plants in well-drained sand culture, where differences in plant performance could more easily be ascribed to effects of salt alone. Plants were watered for 5–8 wk with one-fifth strength Hoagland's solution (Hoagland and Arnon, 1950) prior to being subjected to salinity treatments. Any plant that died within this period was replaced.
Salinity treatments
Plants were subjected to six salinity levels: 0, 2, 4, 8, 15, and 22 g/L dissolved synthetic sea salt. The highest level was selected to reflect the highest seawater salinity measured in tidal creeks near forest stands by Perry and Williams (22 ppt, or 63% full strength seawater salinity; Perry and Williams, 1996), although salinities approaching 30 ppt have been measured in the same tidal creeks during summer months (Williams et al., in press). Continuous drip irrigation to each plant provided a nutrient solution (one-fifth strength Hoagland's solution) with the appropriate salt concentration. Salt was provided by dissolving various amounts of Instant Ocean® synthetic sea salt (Aquarium Systems, Mentor, Ohio, USA) in 40 L of nutrient solution that circulated and provided continual drip irrigation to the plants. This salt was 47% Cl, 26% Na, 6% SO4, 3% Mg, 1% Ca, 1% K, and 16% other, more minor components on a dry mass basis (Pezeshki, 1991).
Plants (N = 14–20) of each species were subjected to each of the six salinity treatments, for a total of 804 plants. At each salinity level, plants were arranged in two groups located in different areas of the greenhouse and associated with different vats of circulating salt solution. Each vat provided nutrient and/or salt solution to both pots containing soil and pots containing sand. In each half of the greenhouse, different salinity treatments (groups of plants receiving water from different vats) were arranged in a stratified random manner such that replicate groups occupied opposite quadrants of the greenhouse. Pairs of plants of each species (one planted in sand and one in soil) were distributed in a randomized manner within each group. Different accessions of plants were distributed as evenly as possible across salinities, rooting media, and replicate vat systems.
Salt treatments began 20 April 1995. To allow plants to acclimate to salt, salinity was increased in increments of 1–4 g salt/L, such that it took 3 d to reach 2 g/Land 3 wk to reach 22 g/L. Salt treatments ceased on 2 November 1995, half a year after the introduction of salt.
To minimize salt accumulation in the rooting medium due to surface evaporation, pots with soil were watered with an additional 10–20 mL of the circulating nutrient and/or salt solution each day and pots with sand were watered with an additional 50–60 mL each day to flush the upper soil. (Differences between the sand and the soil in drainage and surface drying dictated that different volumes be used for flushing.) Solutions in the vats were changed once a week. Conductivity of the solutions was monitored with a Model 122 conductivity meter (Orion Research Inc., Cambridge, Massachusetts, USA). The extent to which salts accumulated in the upper soil was assessed by measuring the conductivity of the interstitial soil water just prior to changing solutions in August 1995. Because water tended to pond in pots containing soil, interstitial soil salinity was determined from the conductivity of the surface, ponded water in these pots. In pots containing sand, interstitial soil salinity was determined by coring the sand to 5 cm depth (obtaining ~12 g of sand), measuring water content (mass loss upon drying at 105°C for 62 h), extracting salt with 35 mL of deionized water, measuring the conductivity of the extract, and correcting for the experimental dilution. We sampled soil salinity in six pots containing soil and six pots containing sand at each salinity level.
To determine whether waterlogging caused anaerobic conditions to develop in pots containing soil, we measured the redox potential in eight pots with large healthy plants in the freshwater treatment at the end of the experiment. Large plants were selected to maximize the respiring root mass in each pot and increase our chances of encountering low redox potentials in the soil. We inserted platinum probes to a depth of 10 cm in each pot, taking duplicate measurements in each pot. Probes were read with an Orion model 250A pH/ISE meter (Orion Research Inc., Cambridge, Massachusetts, USA), using a calomel reference electrode (DeLaune, Smith, and Patrick, 1983). The pH of the ponded surface water was measured and redox potential was corrected to pH 7 (Ponnamperuma, Tianco, and Loy, 1967; Megonigal, Patrick, and Faulkner, 1993).
Plants were monitored daily for apparent shoot mortality, assessed as complete loss of live (green) leaf tissue. Because some individuals resprouted, however, it became obvious after week 5 of the study that loss of photosynthetic area was not a good indicator of plant death. We continued to irrigate plants that lost all green tissue with appropriate salt solutions thereafter. After 18 wk of salt treatment, all pots with plants that appeared dead were flushed with freshwater, watered daily, and monitored for resprouting. Plants that appeared to die in the salt treatments after week 18 were immediately removed from salt treatments, flushed with freshwater, and monitored for resprouting. Plants were monitored for resprouting for a minimum of 5 wk.
Mean maximum temperature in the greenhouse during the study was 33°C and mean minimum temperature was 22°C. Temperature ranged from 15° to 40°C. For comparison, air temperature measured near permanent forest monitoring plots on the coast (Williams et al., in press) ranged from 7° to 39°C during this same period (K. Williams, University of Florida, unpublished data). Maximum daily light levels, measured with two gallium arsenide sensors connected to a data logger (21x, Campbell Scientific Inc., Logan, UT, USA), ranged from 79 to 1749 µmol·m·s over the 6-mo study and averaged 465 µmol·m·s (photosynthetic photon flux density, PPFD). Most of the variability resulted from changes in cloudiness, with mean maximum light levels recorded at the two sensors differing by only 20%.
Data analysis
Because of the large number of treatment x species combinations in which survival was zero, a full factorial analysis of results could not be performed. Therefore, influences of various factors on mortality were analyzed separately, rather than with a factorial model. Initial chi-square tests of homogeneity indicated that neither plant mortality nor leaf loss differed between replicate irrigation systems at any salinity level. Therefore, plants from replicate irrigation systems were pooled for all subsequent analyses. To examine the effect of substrate type (sand or soil) on survival, survival of plants in soil was compared to that of plants in sand for each species at each salinity level using Fisher's exact test. To determine patterns of relative salt tolerance among the species, species were ranked according to (1) their ability to survive 6 mo of salt exposure and (2) their ability to maintain green leaf tissue throughout that period. Rankings were obtained in two ways. Species were ranked based on the highest salinity at which each survived or retained live leaf tissue. Secondly, species were ranked based on graphs of percentage survival or percentage retaining green leaf tissue vs. treatment salt concentration. The latter approach weighted differential performance at lower salinities (salinities allowing some survival) more heavily than the former. These rankings of salt tolerance were compared to rankings that were predicted based on the species' distribution near the coast. Strength of agreement was assessed based on Spearman rank correlation coefficients.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The upper rooting zone in pots containing sand was well aerated. The water content of sand sampled for interstitial salinity determinations revealed that, except for sand directly under drippers, >22% of the upper 5 cm of sand was air-filled pore space. In contrast, water in pots containing soil tended to pond. Measured redox potentials in pots with soil (0 g/L salt) ranged from 315 to 480 mV (mean = 355 mV). Redox potentials measured in two pots with Sabal palmetto, a species with abundant aerenchyma, did not differ from that measured in the six pots with other species (mean ± SD: 352 ± 42 and 344 ± 17 mV, respectively).
Plant mortality
Most species survived well at 0 g/Lsalt, maintaining green leaf tissue throughout the 6-mo study (Fig. 1). Pinus taeda, however, experienced poor survival under control conditions (36% survival at 0 g/L salt) and was not included in the analyses. Some species, notably Quercus virginiana and Celtis laevigata, experienced shoot death during the course of the study, but resprouted after pots were flushed with freshwater (Fig. 1). Fifty percent of the Quercus virginiana seedlings and 22% of the Celtis laevigata seedlings that lost all green tissue resprouted. Juniperus virginiana var. silicicola, Sabal palmetto, and Liquidambar styraciflua did not resprout after being flushed with freshwater. Only one seedling of Ulmus spp. and one of Acer floridanum produced new leaf tissue after freshwater flushing.
|
|
0.05), and no seedlings of Acer floridanum or Ulmus spp. survived at higher salinities (Fig. 2). Contrary to expectation, however, survival of Liquidambar styraciflua under saline conditions was better than that of Acer floridanum (P 0.05 at 2 g/L, Fisher's exact test) and the same or better than that of Ulmus spp. (Fig. 2). Also contrary to expectation, survival of Sabal palmetto, the species with seedlings found closest to salt marsh, tended to be lower than that of Quercus virginiana and Juniperus virginiana var. silicicola at intermediate salinities. This difference was statistically significant (P 0.05, Fisher's exact test) at 8 g/L. Thus, differences in seedling survival among the four most coastward and salt tolerant species did not yield a pattern of relative salt tolerance that corresponded to their zonation.
|
0.05), and none retained green tissue at higher salinities. The number of seedlings retaining green leaf tissue was significantly lower for Quercus virginiana and Celtis laevigata than for either Sabal palmetto or Juniperus virginiana var. silicicola at 4 g/L sea salt (Fisher's exact test, P 0.05), and none retained green leaf leaves above 8 g/L. Only Sabal palmetto and Juniperus virginiana var. silicicola retained live leaves at 15 g/L sea salt, and only Sabal palmetto, the species with seedlings distributed closest to the marsh, retained green leaf tissue at 22 g/L sea salt.
|
0.05). However, at high salinities (15 and 22 g/L salt) Sabal palmetto appeared to outperform Juniperus virginiana var. silicicola. Sabal palmetto was the only species to survive and retain green leaf tissue at the highest salinity (although percentage survival was not statistically different from zero), and it retained green leaf tissue much longer than Juniperus virginiana var. silicicola at 15 and 22 g/L salt (Fig. 1). Survival was significantly greater for seedlings of Juniperus virginiana var. silicicola, Sabal palmetto, and Celtis laevigata, which were growing in sand, compared to those growing in soil, at several salinity levels (Fig. 4). These differences did not affect the rank order of salt tolerance among species. Analyzing the survival of seedlings in sand and soil separately yielded rank orders of seedling salt tolerance that were not different from those shown in Figs. 2 and 3 (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To determine whether seedling tolerance of root-zone salinity adequately explains where forest zones occur, rather than simply explaining the relative positions of species across such zones, an extensive record of soil salt concentrations is required. In coastal areas, the salt concentration in soil solution is extremely dynamic (e.g., Kurz and Wagner, 1957; Martin and Young, 1997), being controlled by factors that affect soil salt content (i.e., the dynamics of salt spray, tidal flooding, and heavy rain events that flush salt from soil) and factors that affect soil water content (e.g., patterns of rainfall, evapotranspiration, and drought).
Soil salinity dynamics have not been described in detail across the forest zones described here. However, preliminary data on soil chloride concentrations in two forest stands (D1 and H1; Williams et al., in press) yielded levels of soil salinity that were fairly consistent with those predicted based on differences in species composition between the two stands and differences in seedling salt tolerance determined in this study. In November 1991, the chloride concentration in soil from a stand supporting seedlings of Sabal palmetto, Juniperus virginiana var. silicicola, Quercus virginiana, and Celtis laevigata, but no other tree species, was 0.7 g chloride/Lsoil solution (K. Williams, University of Florida, unpublished data), equivalent to that in a synthetic sea salt solution of 1.5 g/L sea salt. Based on our greenhouse results, such a concentration would have been adequate to exclude Ulmus spp. and cause very high mortality in Acer floridanum, but allow regeneration of Liquidambar styraciflua and the four species found regenerating in the stand. At the same time, a stand of mature Sabal palmetto and Juniperus virginiana var. silicicola that lacked all tree regeneration had soil chloride concentrations of 6 g chloride/L soil solution (K. Williams, University of Florida, unpublished data), equivalent to that in a synthetic sea salt solution of 13 g/L sea salt. Based on our greenhouse results, such a concentration, if persistent, should have allowed regeneration of Sabal palmetto and Juniperus virginiana var. silicicola, while excluding all other species. The failure of regeneration of Sabal palmetto and Juniperus virginiana var. silicicola in this stand was not due to a high sensitivity of seed germination to salt. The salinity range over which germination fails in these two species is similar to that over which seedling survival failed: between 20 and 30 g/L NaCl for Sabal palmetto (Brown, 1976), and between 20 and 30 g/L synthetic sea salt for Juniperus virginiana, a close relative of Juniperus virginiana var. silicicola (Martin and Young, 1997). Furthermore, transplant studies have revealed that neither seedlings of Sabal palmetto (Perry and Williams, 1996) nor those of Juniperus virginiana var. silicicola (K. Williams, University of Florida, unpublished data) can survive at this site. Soil salinity at this site was sampled one day after it was flooded with a high tide, partly or fully reversing any evaporative concentration of salts that would have occurred after the previous flooding tide 2 wk earlier, a 2-wk period with very little rain (NOAA tide gauge data, Cedar Key, Florida). Therefore, whereas measured soil chloride concentrations in the infrequently flooded stand (H1; Williams et al., in press) adequately explained the exclusion of Ulmus spp. and Acer floridanum from that stand, measured chloride concentrations in the more frequently flooded stand failed to explain regeneration failure of Sabal palmetto and Juniperus virginiana var. silicicola seedlings there. However, the measured soil chlorinity was unlikely to be the highest achieved in that stand and was sufficiently high to prevent seedling survival in all but the two most salt tolerant of the species studied.
The relative abilities of Sabal palmetto and Juniperus virginiana var. silicicola to tolerate salt depended on the salinity level to which the seedlings were exposed. At moderate salinity levels, Juniperus virginiana var. silicicola appeared more salt tolerant. At high salinity levels, Sabal palmetto appeared more salt tolerant. It is likely that performance at high salinities is more important in determining seedling survival at the marshward fringe of the coastal forest than performance at lower salinities. The salinity levels used in this study were chosen to approximate the highest salinity observed by Perry and Williams (1996) in a tidal creek at Waccasassa Bay (22 g/L salt), freshwater, and an intermediate range. Due to evaporative concentration of salts, however, soil salinity may frequently be much higher. The rapid death of seedlings of Juniperus virginiana var. silicicola during periods of high soil salinity and the ability of Sabal palmetto to tolerate high salinity for extended periods (Fig. 1) may suffice to cause zonation between these two species near the forest fringe.
In this study, mortality was higher in soil than in sand for several species at several salinities. The soil had lower interstitial salinity than the sand, but remained much more saturated than sand. Some seedlings in soil exhibited shallow rooting, characteristic of rooting in flooded soils. Redox potentials of 315–488 mV suggested that conditions were only weakly reducing in the soil. At these redox potentials, free oxygen may have been absent in some pots and manganese reduction may have been occurring (Marschner, 1986). These redox potentials are high for flooded soils, but only slightly higher than those found in declining, tidally flooded coastal forest stands (Williams et al., in press). The reduced survival of Sabal palmetto under flooded conditions is somewhat surprising, given the presence of well-developed aerenchyma in its roots, and contrasts with previous work in which Sabal palmetto seedlings grown in flooded sand culture had higher photosynthetic rates than seedlings grown in sand that was flooded briefly twice a day (Perry and Williams, 1996). Whereas continuous flooding in that study probably prevented accumulation of salt in the sand, it probably did not result in the level of hypoxia achieved in soil in this study or in the field. Because the pattern of apparent salt tolerance in this study was not altered by growth in soil vs. sand, we conclude that the ability to tolerate flooded soils was not an important factor in determining the relative positions of the species we studied in forest zonation near the coast. Soil flooding may, however, interact with soil salinity to determine exactly where the forest zones occur on the landscape.
All of the species used in this study experience occasional storm surges. In three storm surges during 1993–1995, we measured conductivity indicating surge-water salt concentrations of 3.5, 14.5 and 19.3 g/L. Thus, the highest measured salinity in storm surge water was similar to the highest salinity used in this study. Surge-water salinity may be lower or only temporarily high, however, when heavy rainfall accompanies the surge or precedes the surge, limiting infiltration of saltwater (Gardner et al., 1992). Storm surges or exceptionally high tides may cause transiently high soil salinities, however, when followed by periods without rain (Kurz and Wagner, 1957).
Shoot death and resprouting following such temporary exposure to saltwater have been found in several tree species (e.g., Conner and Askew, 1992, 1993). Results of this study indicate that some individuals of Quercus virginiana and Celtis laevigata can withstand months of high salt exposure in a leafless condition, resprouting upon flushing with freshwater. Periods of temporarily high soil salinity have been noted in stands occupied by Quercus virginiana. Kurz and Wagner (1957) found that, although soil chloride concentrations in coastal sites occupied by Quercus virginiana were typically 0–5 g chloride/L soil solution (equivalent to synthetic sea salt solutions of 0–11 g/L sea salt), they could reach 43.2 g chloride/L soil solution (equivalent to 92 g synthetic sea salt/L, or almost three times full-strength seawater) after storm tides and prolonged drought. The ability of seedlings to resprout following salt-kill of the shoot may allow survival in these areas where occasional storm surges or high tides are not immediately followed by sufficient rain to flush the salt from the soil.
The failure of Acer floridanum seedlings to resprout in this study is somewhat surprising. Observations of seedlings in the field suggest that they develop thickened stems and regularly die back and resprout near the coast (K. Williams, personal observation). Seedlings of red maple have been shown to resprout following temporary exposure to saltwater (Conner and Askew, 1993). It is possible that seedlings of Acer floridanum require time to build up carbohydrate reserves and otherwise develop the ability to resprout. Conner and Askew (1992) found that young (6 mo old) baldcypress seedlings were less likely to resprout and survive temporary saltwater flooding than older (18 mo old) seedlings. Although the very young Acer floridanum seedlings used in this study did not resprout upon freshwater flushing, the ability to resprout may develop with age.
The fact that leaf retention under saline conditions, rather than seedling survival, more closely reflected plant zonation suggests that the importance of resprouting as a mechanism for coping with salt exposure varies with distance from the salt marsh. The two species that occupied the most frequently flooded zones at the forest fringe (Sabal palmetto and Juniperus virginiana var. silicicola) retained photosynthetic tissue under saline conditions. The two species that became established at slightly higher, less frequently flooded sites (Quercus virginiana and Celtis laevigata) lost photosynthetic tissue under saline conditions, but resprouted well following removal of salt. The species restricted to yet higher or more inland sites neither retained photosynthetic tissue under high salinities, nor resprouted to any great extent. The intermediate position of Quercus virginiana and Celtis laevigata in this zonation suggests that resprouting following temporary salt exposure does not ensure survival at the most frequently flooded margins of the coastal forest. We hypothesize that either salt is never sufficiently flushed from the soil to allow resprouting or repeated episodes of shoot loss and resprouting in response to frequent cycles of saltwater exposure and rainfall flushing lead to exhaustion of seedling reserves and mortality. At slightly higher elevations, where saltwater exposure is less frequent, the ability to resprout appears to be a viable method of coping with episodic salt exposure. Patterns observed in this study further suggest that developmental changes in the ability to resprout following salt exposure may play a role in determining forest zonation at the coast. Species with seedlings that could resprout at a very young stage of development (Quercus virginiana and Celtis laevigata) occupy forest zones that are more regularly flooded by tidal waters than those occupied by species that are known to resprout but did not do so as very young seedlings in this study (i.e., Acer floridanum). We suggest that the ability to resprout at a very young stage of development may allow a species to occupy forest zones that are flooded by saltwater almost every year. In contrast, species that require some time to develop salt tolerance, including the ability to resprout following salt exposure, are excluded from these zones but may survive in areas where saltwater flooding is less frequent and eliminates only occasional cohorts of young seedlings.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Brown, K. E.1976Ecological studies of the cabbage palm, Sabal palmetto. III. Seed germination and seedling establishment. Principes 20: 98–115.
Clark, J. S.1986Coastal forest tree populations in a changing environment, southeastern Long Island, New York. Ecological Monographs 56: 259–277. [CrossRef]
Conner, W. H., and G. R. Askew.1992Response of baldcypress and loblolly pine seedlings to short-term saltwater flooding. Wetlands 12: 230–233. [Web of Science]
———, and ———.1993Impact of saltwater flooding on red maple, redbay, and Chinese tallow seedlings. Castanea 58: 214–219.
DeLaune, R. D., C, J. Smith, and W. H. Patrick, Jr.1983Relationship of marsh elevation, redox potential and sulfide to Spartina alterniflora productivity. Soil Science Society of America Journal 47: 930–935.
Gardner, L. R., W. K. Michener, T. M. Williams, E. R. Blood, B. Kjerve, L. A. Smock, D. J. Lipscomb, and C. Gresham.1992Disturbance effects of Hurricane Hugo on a pristine coastal landscape: North Inlet, South Carolina, USA. Netherlands Journal of Sea Research 30: 249–263.
Gresham, C. A., T. M. Williams, and D. J. Lipscomb.1991Hurricane Hugo wind damage to southeastern U.S. coastal forest tree species. Biotropica 23: 420–426. [CrossRef][Web of Science]
Hoagland, D. R., and D. I. Arnon.1950The water culture method of growing plants without soil. California Agriculture Experiment Station Circular 347.
Kurz, H., and K. Wagner.1957Tidal marshes of the Gulf and Atlantic coasts of northern Florida and Charleston, South Carolina. Florida State University Studies Number 24. The Florida State University, Tallahassee, FL.
Marschner, H.1986The mineral nutrition of higher plants. Academic Press Limited, London.
Martin, D. W., and D. R. Young.1997Small-scale distribution and salinity response of Juniperus virginiana on an Atlantic Coast barrier island. Canadian Journal of Botany 75: 77–85.
Megonigal, J. P., W. H. Patrick, Jr., and S. P. Faulkner.1993Wetland identification in seasonally flooded forest soils: soil morphology and redox dynamics. Soil Science Society of America Journal 57: 140–149. [Web of Science]
Perry, L., and K. Williams.1996Effects of salinity and flooding on seedlings of cabbage palm, (Sabal palmetto). Oecologia 105: 428–434. [CrossRef][Web of Science]
Pezeshki, S. R.1991Population differentiation in Spartina patens: gas-exchange responses to salinity. Marine Ecology Progress Series 72: 125–130. [CrossRef][Web of Science]
Ponnamperuma, F. N., E. M. Tianco, and T. A. Loy.1967Redox equilibria in soils. Soil Science 103: 374–382. [Web of Science]
Ross, M. S., J. J. O'Brien, and L. daS. L. Sternberg.1994Sea-level rise and the reduction in pine forests in the Florida Keys. Ecological Applications 4: 144–156. [CrossRef][Web of Science]
Sherman, S. L., and D. E. Giannasi.1988Foliar flavonoids of Ulmus in eastern North America. Biochemical Systematics and Ecology 16: 51–56.
Simons, R. W., S. W. Vince, and S. R. Humphrey.1989Hydric hammock: a guide to management. United States Fish and Wildlife Service Biological Report 85 (7.26 supplement).
Thompson, S. K. S. 1980Hammock vegetation in the northern Gulf Hammock region of Florida. Master's thesis, The Florida State University, Tallahassee, FL.
Vince, S. W., S. R. Humphrey, and R. W. Simons.1989The ecology of hydric hammocks: a community profile. United States Fish and Wildlife Service Biological Report 85 (7.26).
White, D. A.1983Plant communities of the lower Pearl River basin, Louisiana. American Midland Naturalist 110: 381–396. [CrossRef][Web of Science]
Wiegrefe, S. J., K. J. Sytsma, and R. P. Guries.1994Phylogeny of elms (Ulmus, Ulmaceae): molecular evidence for a sectional classification. Systematic Botany 19: 590–612. [CrossRef][Web of Science]
Williams, K., F. E. Putz, K. C. Ewel, T. W. Workman, and R. P. Stumpf.In pressSea-level rise and coastal forest retreat on the west coast of Florida, USA. Ecology.
Wolfe, S. H. [ed.]. 1990An ecological characterization of the Florida Springs Coast: Pithlachascotee to Waccasassa Rivers. United States Fish and Wildlife Service Biological Report 90(21).
This article has been cited by other articles:
|
|
 |
|
  A. H. Hussein Modeling of Sea-Level Rise and Deforestation in Submerging Coastal Ultisols of Chesapeake Bay Soil Sci. Soc. Am. J., January 21, 2009; 73(1): 185 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Houle, L. Morel, C. E. Reynolds, and J. Siégel The effect of salinity on different developmental stages of an endemic annual plant, Aster laurentianus (Asteraceae) Am. J. Botany, January 1, 2001; 88(1): 62 - 67. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
