| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biological Sciences, University of North Carolina, Wilmington, North Carolina 28403;and Department of Mathematical Sciences, University of North Carolina, Wilmington, North Carolina 28403
Received for publication June 9, 1998. Accepted for publication October 6, 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: biomass estimation • drainage • marsh creation • rhizome • soil organics, Spartina alterniflora, Poaceae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, 1988
; Moy and Levin, 1991
; Matthews and Minello, 1994
; Padgett, Rogerson, and Hackney, 1998
). This interest largely relates to the extensive, subsurface network of roots and rhizomes that cordgrass produces and its role in soil stabilization. Indeed, reports indicate that 60% or more of cordgrass biomass is found below the marsh surface (Good, Good, and Frasco, 1982
; Broome, Seneca, and Woodhouse, 1986
).
To be sure, aerial tissues of this plant are important as habitat and ultimate food source for marsh animals (Teal, 1962
; Odum and de la Cruz, 1967
; Boesch and Turner, 1984
; Posey, Alphin, and Powell, 1997
). Their principal value to soil stabilization, however, is as a carbohydrate source that fuels proliferation of subsurface tissues. It necessarily follows that soil conditions conducive to growth of cordgrass stems and leaves likely would be those that promote rapid soil stabilization.
In light of recent federal policy mandating no net loss of wetlands, the U.S. Army Corps of Engineers routinely requires replacement of an acre of marsh land for every acre impacted by construction projects. Despite this policy, the Corps currently has no regulation governing the type of soil that can be used to construct new marshes (Padgett, Rogerson, and Hackney, 1998
).
Soil organic content and the drainage characteristics it influences are two major factors among a myriad of other abiotic conditions affecting establishment of cordgrass outplants, thus it seems reasonable that they should be optimized during marsh creation efforts. The purpose of this paper is to present results from the first year of a long-term salt marsh mesocosm study whose ultimate objective is to assess the influence of these two important variables on production of aerial and subsurface tissues by S. alternifliora seedlings. Our particular interest is in monitoring rhizome proliferation. We view these horizontal, underground stems as critically important to soil stabilization because they serve as "plant generating" centers analogous to stolons of strawberries. Soil conditions favoring horizontal spread of rhizomes from an initial seedling thus would produce clonal offspring more rapidly, thereby generating more roots to anchor soil against erosion (Broome, Seneca, and Woodhouse, 1986
, 1988
).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
Cordgrass seeds were collected in early October 1996 from tall-form plants in the nearby Shell Island salt marsh at the northern end of Wrightsville Beach and stored in the dark in 20 ppt seawater at 4°C until early March 1997 (Broome, Seneca, and Woodhouse, 1986
). Seeds that germinated during cold storage were transplanted individually into 4 x 4 x 6 cm peat pots filled with commercial potting soil on 10 March and propagated in a greenhouse under natural light regimes for 9 wk. Seedlings were fertilized with MiracidTM commercial plant food every other week during this period. We outplanted seedlings into our mesocosm tanks on 21 May 1997, at which time the average seedling height was 8 cm. Nine seedlings were planted in each tank in a 3 x 3-plant grid design with each 28 cm from any adjacent plant and all plants in outside rows 28 cm from the tank edge. At the time of planting, a bamboo shishkabob skewer was used to anchor each peat pot until subsequent root growth firmly established the seedling in its surrounding soil. These skewers were left in place throughout the growing season and served as points of reference for assessing horizontal rhizome growth (discussed below).
Monitoring plant growth
The emergence of each new aerial shoot was recorded about every 2 wk throughout the growing season by measuring its linear distance to the closest seedling (seedling of presumed origin). These measurements were considered to be conservative indications of the horizontal rate of rhizome growth beneath the soil surface and were compiled for each treatment to assess the effect of soil organic content and drainage depth on rhizome proliferation.
We quantified total production of aerial and subsurface tissues in each tank during the first growing season in early October 1997. Aerial mass was harvested by cutting all plants at the soil surface and separating live from dead leaves as well as stems from leaves. Prior to harvesting living leaves, we measured their width and total length. These data were used to derive a model to predict the mass of standing-live leaves based on width/length measurments alone (rationale discussed below). Harvested tissues were washed free of superficial sediment, oven dried at 90°C to constant mass, and weighed.
Subsurface biomass was estimated by taking duplicate soil cores to a depth of 30 cm using a sharpened 8.3-cm internal-diameter stainless steel cylinder. Coring was done midway between adjacent original plants, thus including only roots and rhizomes that had extended out a minimum distance >9.85 cm from each seedling. Each core was partitioned horizontally into six, 5-cm thick slabs using a hacksaw. The subsurface tissues in each were washed, separated into live vs. dead root and root vs. rhizome categories, oven dried, and weighed.
Leaf biomass prediction model
Our rationale for developing a model to predict leaf biomass without actually cutting a plant was driven by the fact that, at this latitude, many plants overwinter in the standing-green condition and resume growth the following spring. We recognize, therefore, that yearly, destructive harvests would have an impact upon the production not only of aerial, but possibly also subsurface biomass in subsequent growing seasons. Our ultimate objective, pending verification of our predictive model, was to assess leaf biomass at the end of year 2 through the termination of the project by coupling standard harvests of dead and dying leaves with predictions of standing-live leaf mass in each tank without cutting them. Procedures used to develop and test this model are discussed below.
Preliminary observations indicated that the bases of cordgrass leaves tightly ensheathed the stem and overlapped those of younger leaves produced successively closer to the plant's longitudinal axis (centripetal leaf origin). We refer to these basal, overlapping leaf segments hereafter as leaf sheaths (LS). The blade (BL), however, represented the majority of the total length of each leaf and diverged from the stem such that both upper and lower epidermal surfaces were freely exposed to the atmosphere. The BLs of all leaves were uniformly dark green, but LS regions contained progressively less chlorophyll as a function of how deeply they were "buried" by the series of older leaves to their exterior. We inferred from this that BLs represented the most photosynthetically significant aerial tissue and should serve as the focal point for our model.
Observations of a large number of leaves revealed that BLs closely conformed to the shape of a gently tapering triangle (lanceolate) of uniform thickenss. We, therefore, hypothesized that a model could be developed to predict the total area of photosynthetically significant leaf surface using basal width of each BL (determined at the point where the LS transitioned into the BL) and the BL length (from basal-width measurement point to the BL apex) alone.
To test this, we cut the BLs from a number of living cordgrass leaves of various sizes from three natural salt marshes in the area. Each (now truncated lanceloate in shape) was measured to determine basal width and total length. These were then flattened and photocopied before drying could cause any distortion. The actual area of each photocopied BL was determined by scanning the surface three times with a digital planimeter (Lasico Inc., Los Angeles, CA; Model 42P) and calculating the mean value.
Upon closer examination we noted that BLs change shape slightly along their entire length as they mature. Margins of young leaves exhibit a smoothly convex taper from base to apex, older ones taper linearly, and fully mature leaves are slightly concave. We adjusted our model to accommodate this morphological variation by including an exponential factor that alters the equation as a function of leaf length (i.e., age). Our derived equation to predict area thus was
 |
We tested the accuracy of our model by comparing the predicted area of a cohort of randomly collected BLs to their known areas as determined by planimetry. Regression analysis of this comparison yielded a coefficient of determination of 0.977.
Relating BL area to biomass involved determining the average density per unit area of randomly collected cordgrass leaves and using these data to derive the following mass-prediction equation:
 |
|
have shown that vegetational changes in created marsh soils are reflected in recruitment patterns of small invertebrates that live within the soil matrix (infauna). Therefore, we thought it would be useful to conduct yearly sampling to document changes in the infaunal community through time that might correlate with plant growth. We collected duplicate soil cores for analysis from comparable locations in each tank on 19 September 1997 after first draining all tanks to a depth of 20 cm. Our coring tube was a 23.7-mm internal diameter plastic pipe, which was pushed vertically into the sediment to a depth of 84.2 mm. Recovered soil samples were fixed in 10% formalin containing rose bengal stain and infauna extracted by filtration through a 75-µm mesh sieve. Samples then were preserved in 70% ethanol and refrigerated until counts could be performed.
Statistical analyses
Raw data for each monitored parameter (discussed above) were grouped by treatment (N = 3 tanks per treatment) and analyzed by two-way ANOVA (SAS Institute, Research Triangle Park, North Carolina) where soil organic content and drainage depth of the treatment were main effects.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In terms of predicting the mass of standing-live mesocosm leaf BLs (Table 2), our model proved accurate within 6.2% for the 5-cm drain/sand, 4.8% for 5-cm/sand + peat, and 5.0% for 15-cm/sand treatment. The mean prediction error for the 15-cm/sand + peat treatment was artificially high (10.3%) due to a 25.2% error for one replicate tank that was measured by a student assistant. The remaining two replicates of this treatment were measured by DEP (who measured 11 of the 12 tanks) and were 2.0 and 3.7%, respectively.
|
Soil coring to assess subsurface plant growth (Table 3) showed that limited root tissue extended outward farther than 9.85 cm from the original seedlings and that none had penetrated deeper than 20 cm below the surface. It is interesting to note that there were no dead roots at any soil depth in any of the 24 cores. Only one core contained a rhizome fragment (sand substrate only, 15-cm drainage depth; fragment contained in core slab that extended from 5 to 10 cm soil depth). Statistical analysis revealed no significant effect of soil drainage depth or substrate organic content either on total root biomass accumulation, on the maximum soil depth to which roots penetrated, or on root "preference" for a particular depth stratum within the 30 cm long core profile.
|
|
.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Most long-term field experiments require periodic, destructive harvests of aerial plant tissues to quantify production. These involve cutting live as well as dead plant tissue at the soil surface for drying and weighing (e.g., Broome, Seneca, and Woodhouse, 1986
). This is a justifiable practice without measurable, long-term consequences when an experimental plot is sufficiently large to permit sampling in areas not previously impacted. Our mesocosms, however, presented only 1.25 m2 of actual marsh surface per tank; thus the impact of removing standing-live leaves, which could overwinter and resume growth the following year, would increase as plants grow larger and denser. Although validation of our model necessitated a destructive harvest of aerial tissue at the end of year 1, we conclude that the accuracy of our BL modeling will obviate this in subsequent years.
We should note that our harvest of aerial mass included stems as well as leaves. We do not, at this time, have an adequate way of quantifying total production either of LS tissue or the standing-live stem mass that it surrounds without destructive harvest, but propose to develop a model to predict these during the second year of this project. Even though LS tissue and stems are much less significant than BLs as sources of photosynthate (because of shading that results from the overlapping growth habit described above), they represent significant carbohydrate sinks that could not be quantified except by destructive harvest, which would also kill standing-live BLs they subtend.
All 108 cordgrass seedlings outplanted into our mesocosm tanks became successfully established regardless of treatment. Prior to outplanting, these seedlings were maintained in a greenhouse and watered only with tap water. Their surviving the osmotic shock associated with instantaneous exposure to full-strength seawater indicates a hardiness that underscores the value of this species for marsh restoration projects (Woodhouse, 1979
).
Statistical comparisons of aerial biomass as a function of treatment indicated no significant effect of soil drainage depth on accumulation of leaves (live or dead) or stem tissue. Padgett, Rogerson, and Hackney (1998)
also studied effects of soil drainage on cordgrass production but noted a direct correlation with aerial tissue accumulation during the first growing season after transplantation. By the end of the second season, however, they found no differences regardless of soil drainage depth. It may be that their first-year results differed from the present study due to use of buckets that were much smaller than our mesocosm tanks and outplanting of more mature plants directly from a natural marsh rather than greenhouse-grown seedlings. Mendelssohn and Seneca (1980)
similarly compared effects of soil drainage on cordgrass growth in field and greenhouse experiments and noted exactly opposite effects on aerial mass accumulation. They found that greater soil saturation reduced overall plant height, biomass, and stand density in the field but caused increases in the smaller scale greenhouse trials. These perplexing differences coupled with those cited above suggest that the overall scale of nature-simulation experiments likely has a pronounced effect on their correlation with plant growth in natural marshes.
Unlike the effect of drainage depth, we noted a statistically significant increase in stem and dead leaf mass in treatments that contained peat moss. Conversely, there was no significant difference in living-leaf mass when peat tanks were compared to sand. Increased total aerial tissue production (stem and leaf masses combined) could result from greater adsorption of soil nutrients onto the smaller particles present in organic soils as compared to sand (Lindau and Hossner, 1981
; Craft, Seneca, and Broome, 1991
). It seems counterintuitive, however, that increased organics would stimulate overall plant production yet increase leaf mortality. We can neither explain this nor find other literature that indicates why increased organic content resulted in more dead leaf mass compared to sand treatments. In general, we conclude that increased production of aerial tissue in peat-containing soils suggests that soil organic content could be a significant factor in early establishment and aerial growth of cordgrass in created salt marshes.
Results of soil coring (Table 3) revealed that neither soil drainage nor organic content had a significant effect on accumulation of subsurface tissue or on its vertical distribution. Our drainage depth effects are in agreement with findings of Padgett, Rogerson, and Hackney (1998)
. It was interesting to note that there was very little tissue in any of our soil cores, and only one contained a segment of rhizome. Furthermore, we found that all roots and rhizomes were alive. Judging from the steady increase in size of stems and leaves over the growing season, we infer that there was corresponding production of new root mass. The fact that our cores were taken midway between adjacent plants means that only roots and rhizomes extending out (horizontally) farther than 9.85 cm from a seedling were included in the samples. This, in turn, suggests that the majority of root tissue must have been clumped within 9.85 cm of its associated seedling. We confirmed this by carefully excavating (then replanting) several seedlings and found the great majority of root mass within a 5-cm radius.
It is well documented that cordgrass roots are fundamentally important in marsh soil stabilization (see review by Matthews and Minello, 1994
). Accordingly, the limited horizontal root proliferation noted herein might suggest that rapid soil stabilization in created marshes would mandate planting propagules very close together regardless of soil organic content. Indeed this would be desirable, but Woodhouse (1979)
points out that close plant spacing adds dramatically both to labor and cost. Broome, Seneca, and Woodhouse (1986)
experimented with cordgrass spacing intervals of 45, 60 and 90 cm in a coastal North Carolina project and found 60 cm to be the best compromise between cost efficiency and soil stabilization.
It seems likely that every marsh creation project would attempt to achieve a balance between cost efficiency and rapid soil stabilization. In this connection, Broome, Seneca, and Woodhouse (1986)
noted the importance of maximizing rhizome proliferation as an alternative to close propagule spacing. They implied that rhizome proliferation from original transplants can be an efficient means of vegetating barren soil between transplants without adding to cost. The present experiments showed a highly significant increase in linear rhizome growth when the sand substrate was amended with peat moss (Table 5, compare 13 September data). Drainage depth, on the other hand, had no effect.
Only one of our 24 soil cores contained even a single rhizome fragment regardless of substrate organic content. This could suggest that tanks containing peat produced longer, but still few rhizomes per original seedling. If rhizomes are important in soil stabilization, we suggest that the true measure of their worth is not their length, but rather the total number of aerial shoots (each with a subtending root network) they produce. We infer, therefore, that the peat-containing tanks encouraged greater soil stabilization in that their rhizomes generated 126 aerial shoots compared to a total of 58 shoots in sand tanks.
The microinvertebrate fauna recruited into mesocosm tanks exhibited a great deal of spatial heterogeneity similar to that reported by Posey, Alphin, and Powell (1997)
in a North Carolina created salt marsh. We, therefore, feel that these data, whose value will become apparent as a baseline for evaluating subsequent annual samples, should be viewed as preliminary. We note simply that oligochaetes were vastly predominant in all treatments and that the ranked order of predominance for all treatments was oligochaetes >> nematodes > copepods > polychaetes.
Our major inference from the present study is that soil organic content is fundamentally important to early establishment and proliferation of smooth cordgrass seedlings and may well determine the ultimate success of a created marsh. Accordingly, we strongly recommend that regulatory agencies consider enacting rules mandating use of fill material with appropriate organic content for all salt marsh creation projects. Although this view might be considered unduely burdensome to developers, the likelihood that it may enhance chances for a successful project could save money in the long run.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Broome, S. W., E. D. Seneca, and W. W. Woodhouse, Jr. 1986 Long-term growth and development of transplants of the salt-marsh grass Spartina alterniflora. Estuaries 9: 63–74.
———, ———, and ———. 1988 Tidal salt marsh restoration. Aquatic Botany 32: 1–22.[CrossRef]
Craft, C. B., E. D. Seneca, and S. W. Broome. 1991 Porewater chemistry of natural and created marsh soils. Journal of Experimental Marine Biology and Ecology 152: 187–200.[CrossRef]
Good, R. E., N. F. Good, and B. R. Frasco. 1982 A review of primary production and decomposition dynamics of the belowground component. In V. S. Kennedy [ed.], Estuarine comparisons, 139–158. Academic Press, New York, NY.
Lindau, C. W., and L. R. Hossner. 1981 Substrate characterization of an experimental marsh and three natural marshes. Soil Science Society of America Journal 45: 1171–1176.
Matthews, G. A., and T. J. Minello. 1994 Technology and success in restoration, creation, and enhancement of Spartina alterniflora marshes in the United States, vol. 1, Executive summary and annotated bibliography. NOAA Coastal Ocean Program Decision Analysis Series, Number 2. NOAA Coastal Ocean Office, Silver Spring, MD.
Mendelssohn, I. A., and E. D. Seneca. 1980 The influence of soil drainage on the growth of salt marsh cordgrass Spartina alterniflora in North Carolina. Estuarine and Coastal Marine Science 11: 27–40.[CrossRef][ISI]
Moy, L. D., and L. A. Levin. 1991 Are Spartina marshes a replaceable resource? A functional approach to evaluation of marsh creation efforts. Estuaries 14: 1–16.
Odum, E. P., and A. A. de la Cruz. 1967 Particulate organic detritus in a Georgia salt marsh-estuarine ecosystem. In G. A. Lauff [ed.], Estuaries, 383–388. American Association for the Advancement of Science Publication Number 83. Washington, DC.
Padgett, D. E., C. B. Rogerson, and C. T. Hackney. 1998 Effects of soil drainage on vertical distribution of subsurface tissues in the salt marsh macrophyte Spartina alterniflora Lois. Wetlands 18: 35–41.[ISI]
Posey, M. H., T. D. Alphin, and C. M. Powell. 1997 Plant and infaunal communities associated with a created marsh. Estuaries 20: 42–47.[CrossRef][ISI]
Teal, J. M. 1962 Energy flow in the salt marsh ecosystem of Georgia. Ecology 43: 614–624.[CrossRef][ISI]
Woodhouse, W. W. 1979 Building salt marshes along the coasts of the continental United States. U.S. Army, Corps of Engineers Coastal Engineering Research Center Special Report Number 4. U.S. Army, Corps of Engineers, Fort Belvoir, VA.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
