Similarity between seed banks and above-ground vegetation along a salinity gradient

Transcription

1 Journal of Vegetation Science 11: , 2000 IAVS; Opulus Press Uppsala. Printed in Sweden - Similarity between seed banks and aboveground vegetation along a salinity gradient Similarity between seed banks and above-ground vegetation along a salinity gradient Egan, Todd P. * & Ungar, Irwin A. Department of Environmental & Plant Biology, Porter Hall 315, Ohio University, Athens, Ohio , USA *Corresponding author: Fax ; egan@uwm.edu Abstract. Zonation of above-ground vegetation often occurs in salt marshes along salinity and moisture gradients. The above-ground vegetation and seed bank in four physiognomically different vegetation zones in a salt marsh were compared to determine their level of similarity using percent similarity as a distance measure. 10-m transects were established along a salinity gradient through four different vegetation zones; a Salicornia zone, a Salicornia-Atriplex zone, an Atriplex zone and an Atriplex-Hordeum zone. A UPGMA cluster analysis demonstrated that the above-ground vegetation was not usually highly correlated with the seed bank composition of zonal communities. Since seeds of these annual salt marsh species occurred in all zones, the levels of salt stress may be the main factor determining which species were found in the aboveground vegetation. Keywords: Atriplex prostrata; Community; Halophyte; Salicornia europaea. Abbreviations: UPGMA = Unweighted pair group method using arithmetic averages. Nomenclature: Gleason & Cronquist (1991). Introduction Although seeds of all species found along a salinity gradient may occur throughout a salt marsh (Haukos & Smith 1994), zonation of plant species often occurs in the above-ground vegetation (McMahon & Ungar 1978; Bertness 1992). Zonational patterns may be determined by a number of biotic and abiotic factors (Ungar 1998). In some instances, seeds of species of low to moderate salt tolerance will be dispersed into a region of high salinity and the salt concentration of the soil will prevent seed germination and establishment of plants (Shumway & Bertness 1992). In other cases, seeds of highly salt tolerant species will be dispersed into less saline areas of the marsh but the less salt tolerant species growing in that zone will be more competitive and prevent halophytes from becoming established (Ungar et al. 1979). Species growing in inland salt marshes of North America are often found in particular portions of the environmental gradient. Salicornia europaea is usually found in the most saline habitats (> 3.5% total salts) of salt marshes (Ungar et al. 1979; Riehl & Ungar 1982), Atriplex prostrata usually occurs in less saline habitats (< 2.0% total salts) (Khan & Ungar 1986) and Hordeum jubatum is found in the lowest salinity habitats (< 1.0% total salts) (Khan & Ungar 1986; Badger & Ungar 1988; Ungar 1998). Soil salinity is a major factor determining seed germination and whether plants can grow to maturity in salt marsh habitats. (Macke & Ungar 1970; Ungar 1996; Egan et al. 1997; Katembe et al. 1998). Salinity can affect germination and growth either osmotically (Ungar & Hogan 1970; Romo & Haferkamp 1987) or through a specific ion effect (Bal & Chattopadhyay 1985; Egan & Ungar 1998). Hypersaline conditions may prevent seeds from germinating and therefore maintain a persistent seed bank (Ungar 1991). Ungar & Woodell (1993) compared species in the seed bank to that of the above-ground vegetation in salt marshes in Norfolk and Wales, UK. They found that there was high correlation between seed bank and aboveground vegetation for annual dominated communities, but low correlation in perennial dominated marsh communities. A similar study by Ungar & Woodell (1996) compared the above-ground vegetation to seed banks in four British salt marshes in which the dominant or codominant above-ground vegetation was Puccinellia maritima. Two of the salt marshes were ungrazed, one was grazed by horses and the other by sheep. The seed bank at all four salt marshes consisted of more than 60% Salicornia europaea and Suaeda maritima but these species comprised less than 15% of the above-ground vegetation of these marsh communities. In the heavily grazed marsh, these two species comprised 100% of the seed bank but only 13.6% of the above-ground vegetation (Ungar & Woodell 1996). Previous investigations of coastal marshes indicated that some zonal salt marsh community seed banks were similar to the above-ground vegetation while others were not (Ungar & Woodell 1993). The purpose of this investigation was to determine whether or not seed

2 190 Egan, T.P. & Ungar, I.A. banks in these annual dominated, inland salt marsh zonal communities have a high similarity to the aboveground vegetation. Cluster analysis will be used to classify both the above-ground vegetation and the seed bank. Material and Methods Study site This investigation was carried out at a salt marsh in Rittman, Wayne County, Ohio, U.S.A. (40 57' 30'' N; 81 47' 30'' W). The location is in northeastern Ohio, ca. 50 km south of Ohio s northern border on Lake Erie of the Laurentian Great Lakes system. This inland salt marsh is located on an abandoned well-field that was formerly used in salt mining. The property is owned by the Morton Salt Company and the local soil has become saline due to brine spills from the operation of these wells. 10-m transects were established in each of four different salinity zones occurring in the salt marsh. The most saline zone was dominated by Salicornia europaea (3.5% NaCl) in the salt pan (Salicornia zone), a moderately high saline zone was dominated by S. europaea and Atriplex prostrata (2% NaCl) (Salicornia-Atriplex zone), a moderately low salinity zone which was dominated by A. prostrata (1% NaCl) (Atriplex zone) and the outermost zone which was dominated by A. prostrata and Hordeum jubatum (0.5% NaCl) (Atriplex-Hordeum zone). The Salicornia zone is in the lowest region of the marsh and S. europaea often grow with their roots and sometimes the lower shoots of the plants inundated in saline water. Vegetation in the outer, less saline zones do not grow inundated in saline water. Seed bank 10 soil cores (6.0 cm diameter 7.5 cm depth) were collected at 1-m intervals in each of the four zones with a bulb planter and placed in situ into 7.7 cm diameter pots on 16 April and 14 June Plants were counted in each of the samples to determine the relative composition of species along each transect. To determine the size of the persistent seed bank, soil cores from the June collection were placed in incubators and sub-irrigated at 5 C night: 25 C day (12 h photoperiod) and allowed to germinate for 8-10 weeks until no more seeds germinated. To ensure that all seeds were accounted for, ungerminated seeds from a soil sub-sample (2.2 cm diameter 2 cm depth, with a no. 14 cork borer) taken from the original 6.0 cm diameter soil core were counted. Preliminary observations indicated that no viable seeds were buried below 2 cm depth in the soil. The number of ungerminated, viable seeds were added to the number of germinated seeds for each core in order to determine the total number of germinable seeds per core to represent the entire seed bank. Vegetation Above-ground vegetation was estimated by counting seedlings in cm 2 randomly placed quadrats along each of the four transects that soil cores were removed from. Seed germination for A. prostrata and S. europaea occurs from February to June and seed dispersal is throughout October and November, with the former species releasing its seeds a few weeks earlier. Collecting soil samples in April allowed us to count initial germination when seeds of these species still possessed cotyledons. Collecting plants in June would have allowed us to determine how many individual plants survived to the middle of the growing season, but a period of heavy rain killed all of the standing vegetation. Voucher specimens of the species collected were deposited in the Bartley Herbarium of Ohio University, Athens, Ohio. Analysis The Multivariate Statistical Package PLUS-Version 2.0 (Kovach 1991) was used to perform a UPGMA cluster analysis between the seed bank and vegetation at each of the vegetation zones. The formula used to calculate percent similarity (PS) is as follows: PS = 2W (1) A + B Where W = sum of the lower relative values of compared samples, and A and B are sums of relative values adding up to 100 (Ungar & Woodell 1996). Percent similarity was used as a distance measure in this analysis (Kovach 1991; Barbour et al. 1999). Our measure of abundance was relative density. A mean of the 10 samples taken along each transect was used to plot the cluster analyses to eliminate noise in the data (Gauch 1982). Results Seedling density The Atriplex-Hordeum zone was dominated by 51.6% A. prostrata and also contained 34.0% S. europaea in the above-ground vegetation (Table 1). This zone also had the highest percentage of H. jubatum and Aster subulatus seedlings. The Atriplex zone had the highest proportion (83.5%) of A. prostrata seedlings of all of the

3 - Similarity between seed banks and aboveground vegetation along a salinity gradient Table 1. Density (mean ± S. E. / m 2 ) and relative density (%) of seedlings (13 April 1996) and total seeds (14 June 1996) in each vegetation zone. N = 10 for each of the vegetation zones. Plant community Seed Bank Vegetation Species Mean density S.E. Rel. density Mean density S.E. Rel. density zone (No. of plants /m 2 ) % (No. of seeds/ m 2 ) % Atriplex- A. prostrata Hordeum S. europaea H. jubatum A. subulatus Atriplex A. prostrata S. europaea H. jubatum A. subulatus Salicornia- A. prostrata Atriplex S. europaea H. jubatum A. subulatus S. marina Salicornia A. prostrata S. europaea H. jubatum S. marina A. subulatus P. aviculare zones. The percent of S. europaea seedlings decreased by 46.6% and H. jubatum and A. subulatus relative densities decreased by 64.0% and 2.0%, respectively, in the Atriplex zone. In the Salicornia-Atriplex zone 74.1% of the seedlings were S. europaea, 23.8% were A. prostrata and the remaining species were A. subulatus and Spergularia marina (Table 1). The Salicornia zone was dominated almost exclusively (99.7%) by S. europaea (Table 1). Seed density The seed bank of the Atriplex-Hordeum zone was dominated by A. prostrata (77.1%), with H. jubatum (20.0%) being an important component. Salicornia europaea made up a small portion (2.9%) of the seed bank in this zone. The seed bank of the Atriplex zone consisted entirely of A. prostrata. Atriplex prostrata dominated the seed bank of the Salicornia-Atriplex zone, which contained only 2.7% S. europaea. The Salicornia zone contained four species in the seed bank, including S. europaea (56.9%), A. prostrata (39.2%), A. subulatus (2.0%) and Polygonum aviculare (2.0%) (Table 1). Similarity between vegetation and seed bank Percent similarities between the vegetation and seed bank for each of the zonal communities varied; similarity for the Salicornia zone was 57.1%, for Salicornia- Atriplex 26.5%, 83.4% for the Atriplex zone and 67.9% for the Atriplex-Hordeum zone. Cluster analysis When comparing the above-ground vegetation among the four zones, the basal most region of the dendrogram had a value of 37% similarity (Fig. 1a). The Atriplex- Hordeum zone and the Atriplex zone formed a cluster with 68% similarity and the Salicornia-Atriplex zone and the Salicornia zone formed a cluster with 74% similarity (Fig. 1a). When comparing the seed banks among the four zones, the Salicornia zone clustered with the other three zones with a 41% similarity value (Fig. 1b). The Atriplex- Hordeum zone clustered with the remaining two groups with a value of 78% similarity and the Atriplex zone and the Salicornia-Atriplex zone formed a cluster at 97% similarity (Fig. 1b). When analysing the seed bank and above-ground vegetation data together, the general trend was that these data clustered at the basal most region of the dendrogram with a value of 34% similarity (Fig. 2). Seed banks of some zones had higher similarity with zonal plant communities other than the one they occupied; the seed bank of the Salicornia zone formed a cluster with the vegetation of the Salicornia-Atriplex zone (80%) and the vegetation in the Atriplex zone formed a cluster with seed banks of the Salicornia- Atriplex zone and the Atriplex zone (84%) (Fig. 2).

4 192 Egan, T.P. & Ungar, I.A. Fig. 1. A UPGMA cluster analysis of above-ground plant vegetation (a) and for seed bank (b) for each vegetation zone, using percent similarity as a distance measure. Fig. 2. A UPGMA cluster analysis combining the aboveground vegetation and the seed bank for each vegetation zone, using percent similarity as a distance measure. Names in italics represent seed banks in each community type. Discussion Seed and seedling density In some coastal marshes, the dominant Spartina species do not form a persistent seed bank (Hopkins & Parker 1984; Hartman 1988). Seed banks in coastal marshes are often dominated by clonal plants (Bertness 1992), but in the inland salt marsh we studied there are clonal plants as well as annual plants (Ungar et al. 1979). Seed banks were found to over-represent some species and under-represent others in salt marsh communities (Milton 1939; Ungar & Woodell 1993; Jutila 1998; Marañón 1998). Peco et al. (1998) compared the vegetation and seed banks in five pastures along an altitudinal gradient in Spanish Mediterranean grasslands. They found that the seed bank of most species is affected by altitude, topography and grazing. They also found that to predict the composition of future seed banks the perennial/annual ratio in the above-ground vegetation and gaps for germination must be taken into account (Peco et al. 1998). The Rittman salt marsh is different from the Spanish grasslands in that the perennial species (e.g. Phragmites australis) occur in the less saline regions at the borders of the marsh. When comparing plant density to the seed bank in the Atriplex-Hordeum and Atriplex zones, A. prostrata was the dominant species in both the seed bank and the above-ground vegetation. The presence of high densities of A. prostrata in the above-ground vegetation may be because of reduced soil salinity and the alleviation of the inhibitory affect of the salt stress (McMahon & Ungar 1978). Salicornia europaea made up 34% of the plant density in the Atriplex-Hordeum zone and 15.8% in the Atriplex zone. However, the seed bank contained less than 3% S. europaea in each of these zones after the spring germination period. This indicates that seed dispersal of S. europaea in this marsh may be low, or that all of the seeds germinated early in the growing season, and plants did not reach reproductive maturity in these zones. 74% of the seedlings in the Salicornia-Atriplex zone were S. europaea and 23.8% were A. prostrata. The seed bank of this zone was 97.3% A. prostrata and 2.7% S. europaea, this is in accordance with the findings of Jefferies et al. (1981) and Davy & Smith (1988) who determined that the annual S. europaea does not form a persistent seed bank in some salt marsh communities. The absence of persistent seed banks is also common for many perennial species in salt marshes (Hutchings & Russell 1989). The size of the S. europaea seed bank may be related to the level of soil salinity, with habitats of hypersaline soils having greater numbers of seeds in the seed bank than those with reduced salinity (Ungar & Woodell 1993). The above-ground vegetation from the Salicornia zone was dominated by 99.7% S. europaea but the seed bank contained 39.2% A. prostrata and 56.9% S. europaea, indicating that the A. prostrata seeds are more widely distributed than any other species in the seed bank that persist after the spring germination period. Seed bank and vegetation When comparing above-ground vegetation, the cluster analysis demonstrates that the Atriplex and Atriplex- Hordeum zones grouped separately from the Salicornia and Salicornia-Atriplex zones along the salinity gradient,

5 - Similarity between seed banks and aboveground vegetation along a salinity gradient which corresponds to the zonational pattern of communities along this salinity gradient (Ungar et al. 1979). The cluster analysis of the seed bank indicates that the cluster including the Atriplex-Hordeum, Atriplex, and Salicornia-Atriplex zones formed a closely related group. The differences between the above-ground vegetation and the seed bank concurred with other studies which indicated that the above-ground vegetation does not necessarily reflect seed bank composition (Ungar & Woodell 1996). Ungar (1995) reported that when soil salinity is above the germination limit for a given species of halophyte, their seeds can remain dormant in the seed bank until salt stress is alleviated. The role of the seed bank is especially important in inland salt marshes where hypersaline conditions often occur, and for annual species that are limited to a single chance for reproduction (Ungar 1995). This is significant in the Rittman salt marsh because two of the dominant species, S. europaea (Gleason & Cronquist 1991) and A. prostrata (McMahon & Ungar 1978), are annuals. Marañón (1998) studied the seed bank of an annual grassland of the upper salt marsh in southwestern Spain during a wet and dry year. A total of 29 species were found in the seed bank, but during the course of the two year study only 17 species were recruited. It was determined that seeds indicative of ephemeral wet environments and species that resist flooding emerged from the seed bank in an unusually wet year. The following year, rainfall decreased to its usual levels and flooding intolerant species germinated. In addition, it was found that salinity can alter the balance of two otherwise co-dominant species. Environmental factors are, therefore, important in determining species recruitment from the seed bank into the above-ground vegetation, since species may have specific germination requirements (Marañón 1998). Noor & Khan (1995) found that temperature and salinity could interact to inhibit seed germination. They determined that Halopyrum mucronatum seeds produced in the summer were not affected by high salinity at high temperatures, but low temperatures and high salinity prevented summer seeds from germination. All temperatures included in the experiment, i.e. 25/10, 30/20, 35/20, 35/25, inhibited germination of winter-produced seeds at high salinity (Noor & Khan 1995). If soil salinity levels or other habitat conditions were beyond the maximum germination limits of species, such as A. prostrata, this would explain its relatively high abundance in the seed bank of all community types on the marsh. Jutila (1998) studied the seed bank of grazed and ungrazed meadows along a seashore in western Finland and found a total of 75 species among the seed bank and above-ground vegetation. However, there was little similarity between them. The seed bank of ungrazed plots was larger and more diverse than grazed plots. Grazed sites were less similar, when comparing their seed bank and the above-ground vegetation, than the non-grazed sites. Grazing may also alter micro-environmental conditions and maintain dormancy in some species, while creating habitats suitable for other species (Bakker & de Vries 1992). Van der Valk & Davis (1978) indicated that seed banks in wetland communities contain species from all zones and may not resemble the above-ground vegetation in a zone, leading to chance establishment of species based on water level. An exception to this may be the dominance of Atriplex prostrata seeds and absence of seeds of other species in the Atriplex zone, but this may be because seeds of other species germinated under the more moderate conditions in this zone. The absence of H. jubatum seeds in the Atriplex zone seed bank may be due to the fact that all of its seeds germinated in the autumn of the previous year (Badger & Ungar 1988). Apparently, H. jubatum has a transient seed bank in the Atriplex zone. Cluster analyses indicated that the four above-ground transects could be grouped into high and low salinity zonal communities. The seed bank cluster analysis did not indicate any specific zonal pattern. Our data demonstrate that the relative density of different species in the above-ground vegetation does not occur in the same proportion for each of the species in the seed bank. Some species such as S. europaea are under-represented in the seed bank while others, such as A. prostrata, are over-represented. This study demonstrated that the proportion of different species in the above-ground vegetation does not always reflect the same ratios of seeds for each species in the seed bank. An example of this is in the high saline Salicornia zone, although the above-ground vegetation is over 99% S. europaea, the seed bank contains almost 40% A. prostrata seeds. Also, seed banks are important in determining the makeup of communities because different species will be recruited in the above-ground vegetation from year to year depending on the level of soil salinity. Acknowledgments. The authors thank Dr. M. Ajmal Khan, Mark Tabor, Li-Wen Wang, Dave Keough, Todd Bean, Meagan Backus and J. Forrest Meekins for their many hours of help in collecting field specimens and counting seeds. Their help was paramount in allowing this study to be completed in a timely manner.

6 194 Egan, T.P. & Ungar, I.A. References Badger, K.S. & Ungar, I.A The effect of salinity and temperature on the germination of the inland halophyte Hordeum jubatum. Can. J. Bot. 67: Bakker, J.P. & de Vries, Y Germination and early establishment of lower salt-marsh species in grazed and mown salt marsh. J. Veg. Sci. 3: Bal, A.R. & Chattopadhyay, N.C Effect of NaCl and PEG 6000 on germination and seedling growth of rice (Oryza sativa L.). Biol. Plant. 27: Barbour, M.G., Burk, J.H., Pitts, W.D., Gilliam, F.S., & Schwartz, M.W Terrestrial plant ecology, 3d ed., p Benjamin Cummings, Menlo Park, CA. Bertness, M.D The ecology of a New England salt marsh. Am. Sci. 80: Davy, A.J. & Smith, H I. Life-history variation and environment. In: Davy, A.J., Hutchings, M.J. & Watkinson, A.R. (eds.) Plant population ecology, pp Blackwell Scientific Publishers, Oxford. Egan, T.P. & Ungar, I.A The effect of different salts of sodium and potassium on the growth of Atriplex prostrata (Chenopodiaceae). J. Plant Nutr. 21: Egan, T.P., Ungar, I.A. & Meekins, J.F The effect of different salts of sodium and potassium on the germination of Atriplex prostrata (Chenopodiaceae). J. Plant Nutr. 20: Gauch, H.G Multivariate analysis in community ecology. Cambridge University Press, Cambridge. Gleason, H.A. & Cronquist, A.C Manual of vascular plants of northeastern United States and adjacent Canada. The New York Botanical Garden, Bronx, NY. Hartman, J.M Recolonization of small disturbance patches in a New England salt marsh. Am. J. Bot. 75: Haukos, D.A. & Smith, L.M Composition of seed banks along an elevational gradient in playa wetlands. Wetlands 14: Hopkins, D.R. & Parker, V.T A study of the seed bank of a salt marsh in northern San Francisco Bay. Am. J. Bot. 71: Hutchings, M.J. & Russell, P.J The seed regeneration dynamics of an emergent salt marsh. J. Ecol. 77: Jefferies, R.L., Davy, A.J. & Rudmick, J Population biology of the salt-marsh annual Salicornia europaea agg. J. Ecol. 69: Jutila, H Seed banks of grazed and ungrazed Baltic seashore meadows. J. Veg. Sci. 9: Katembe, W.J., Ungar, I.A. & Mitchell, J.P Effect of salinity on germination and seedling growth of two Atriplex species (Chenopodiaceae). Ann. Bot. 82: Khan, M.A. & Ungar, I.A Life history and population dynamics of Atriplex triangularis. Vegetatio 66: Kovach, L.W Multivariate Statistical Package PLUS- Version 2.0, Users Manual. Kovach Computing Services, Wales, UK. Macke, A.J. & Ungar, I.A The effects of salinity on germination and early growth of Puccinellia nuttalliana. Can. J. Bot. 49: Marañón, T Soil seed bank and community dynamics in an annual-dominated Mediterranean salt-marsh. J. Veg. Sci. 9: McMahon, K.C. & Ungar, I.A Phenology, distribution and survival of Atriplex triangularis Willd. in an Ohio Salt pan. Am. Mid. Nat. 100: Milton, W.E.J The occurrence of buried viable seeds in soils at different elevations and on a salt marsh. J. Ecol. 27: Noor, M., & Khan, M.A Factors affecting the germination of summer and winter seeds of Halopyrum mucronatum under salt stress. In: Khan, M.A & Ungar, I.A (eds.) Biology of salt tolerant plants, pp University of Karachi, Karachi. Peco, B., Ortega, M. & Levassor, C Similarity between seed bank and vegetation in Mediterranean grassland: a predictive model. J. Veg. Sci. 9: Riehl, T.E. & Ungar, I.A Growth and ion accumulation in Salicornia europaea under saline field conditions. Oecologia (Berl.) 54: Romo, J.T. & Haferkamp, M.R Effects of osmotic potential, potassium chloride, and sodium chloride on germination of greasewood (Sarcobatus vermiculatus). Great Basin Nat. 47: Shumway, S.W. & Bertness, M.D Salt stress limitation of seedling recruitment in a salt marsh plant community. Oecologia (Berl.) 92: Ungar, I.A Ecophysiology of vascular halophytes. CRC Press Inc., Boca Raton, FL. Ungar, I.A Seed bank ecology of halophytes. In: Khan, M.A & Ungar, I.A. (eds.) Biology of salt tolerant plants, pp University of Karachi, Karachi. Ungar, I.A Effect of salinity on seed germination, growth, and ion accumulation of Atriplex patula (Chenopodiaceae). Am. J. Bot. 83: Ungar, I.A Are biotic factors significant in influencing the distribution of halophytes in saline habitats? Bot. Rev. 64: Ungar, I.A. & Hogan, W.C Seed germination in Iva annua L. Ecology 51: Ungar, I.A. & Woodell, S.R.J The relationship between the seed bank and species composition of plant communities in two British salt marshes. J. Veg. Sci. 4: Ungar, I.A. & Woodell, S.R.J Similarity of seed banks to aboveground vegetation in grazed and ungrazed salt marsh communities on the Gower Peninsula, South Wales. Int. J. Plant Sci. 157: Ungar, I.A., Benner, D.K., & McGraw, D.C The distribution and growth of Salicornia europaea on an inland salt pan. Ecology 60: van der Valk, A.G. & Davis, C.B The role of seed banks in the vegetation dynamics of prairie glacial marshes. Ecology 59: Received 17 December 1998; Revision received 6 July 1999; Accepted 23 July Coordinating Editor J.P. Bakker.