, Volume 146, Issue 2, pp 167-181

Hydrochory, seed banks, and regeneration dynamics along the landscape boundaries of a forested wetland

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


Following the environmental sieve concept, the setting in which the recruitment of Taxodium distichum occurs in, becomes increasingly restrictive from the seed to seedling stage in an impounded forested wetland. Although a wide elevational band of dispersing seed moves across the boundary of a swamp-field in the water sheet, the zone of germination is relegated to that portion of the forested wetland that draws down during the growing season. Seedling recruitment is further restricted to the uppermost zone of the winter water sheet. These patterns are likely applicable to other species of dominant swamp species, e.g., Cephalanthus occidentalis crossed the boundary of a forested wetland and abandonded field in winter flooding (November–December and November–March, respectively) in Buttonland Swamp. The elevation of the boundary was 101.3 m NGVD. While the seeds of at least 40 swamp species were dispersed across the boundary, few viable seeds were dispersed after the winter season. Kriged maps showed seeds of T. distichum and C. occidentalis dispersed in patches in the water depending on the position of the water sheet. Most species of both water- and gravity-dispersed species had a localized pattern of seed distribution (either spherical or exponential) and this indicated that seeds may not be dispersed for great distances in the swamp. Water-dispersed T. distichum and C. occidentalis had larger dispersal ranges (A 0=225 and 195 m, respectively) than Bidens frondosa and B. discoidea (A 0=14 and 16 m, respectively). Seed dispersal varied with season depending on the availability of seeds. In Buttonland Swamp, viable seeds typically were dispersed for T. distichum in November–June, and for C. occidentalis in November-July. Low water occurred in August 1993 and high in February 1994 (99.8 and 101.6 m NGVD, respectively). The seed banks along the landscape boundary varied in species composition according to elevation (r 2 = 0.996). While the similarity of species richness between water-dispersed seeds and the seed bank at elevations that flooded (during June 1993 through May 1995) was high (10–17%), it was low between water-dispersed seeds and the seed bank at elevations that did not flood (5%). T. distichum seeds had a short germination window in that seeds germinated within a year following their production in zones that were flooded in the winter followed by drawdown during the next growing season. After 1 year, less than 5% of the T. distichum seeds remained viable on the surface of the soil. Germination of T. distichum was confined to specific elevations (above 99.3 but below 101.6 m NGVD) during this study with 4.1% of the seedlings surviving for more than 2 years at a mean of 101.4 m NGVD. All seedlings below this elevation died. To maximize natural regeneration along the boundaries of swamps in abandoned farm fields targeted for restoration, this study suggests a flood pulse regime consisting of high water in the winter to maximize dispersal of live seeds followed by low water in the summer to facilitate seed germination and seedling recruitment. Hydrologic restoration could assist in the natural recovery of damaged wetlands if a seed source exists nearby.