Ecosystem Modeling in Cobscook Bay, Maine: A Boreal, Macrotidal Estuary 2004 Northeastern Naturalist 11(Special Issue 2):197–224 Annual Aboveground Biomass and Productivity Estimates for Intertidal Eelgrass (Zostera marina L.) in Cobscook Bay, Maine BRIAN F. BEAL1, *, ROBERT L. VADAS, SR.2, WESLEY A. WRIGHT3, AND STEVE NICKL3,4 Abstract - We estimated the aboveground productivity of eelgrass, Zostera marina, in Cobscook Bay at three soft-bottom lower intertidal locations during 1995–1996. Approximately 30 plants were tagged bi-weekly or monthly at two sites where we used a completely randomized block design to assess spatial variability in leaf initiation and elongation rates. The data showed strong seasonal patterns with highest rates from June–September and lowest rates from November–April. At one site, Bell Farm, plants were tagged at two discrete tidal levels (intertidal and sublittoral fringe). On each sampling date, mean leaf biomass and new leaf production were 2.7 and 1.7 times greater, respectively, for plants in the sublittoral fringe. At another site, Mahar Cove, all tagged plants were restricted to one tidal level. Spatial variation in leaf biomass and new leaf production were examined on the first four sampling dates and significant effects were observed twice. Productivity estimates ranged from 0.095 g dry wt m-2 day-1 (November–January) to 1.215 g dry wt m-2 day-1 (August) at Bell Farm. At Mahar Cove, rates ranged from 0.176 g dry wt m-2 day-1 (March–April) to 1.490 g dry wt m-2 day-1 (August). Average annual productivity at Bell Farm and Mahar Cove was 0.481 ± 0.512 g dry wt m-2 day-1 and 0.784 ± 0.512 g dry wt m-2 day-1, respectively. These estimates correlate directly with seawater temperature, but not with salinity, nitrate, and total phosphorous. The time for plants to fully turn over their leaves at the two sites ranged from 50.5–56.7 days (6.4–7.2 turnovers per year), and are comparable to other locations in the northeast US and the Canadian Maritimes. We estimate that total (interidal + subtidal) eelgrass production in Cobscook Bay ranges from 3.3–5.3 x 108 g C year-1. This is the first appraisal of growth and productivity of Z. marina in eastern Maine. Introduction Seagrass meadows are abundant in protected intertidal and shallow subtidal environments. Zostera marina Linnaeus is one of the more widely distributed seagrasses and occurs throughout the northern hemisphere ranging from warm temperate seas to cold arctic waters (Gaeckle 1Division of Environmental and Biological Sciences, University of Maine at Machias, Machias, ME 04654. 2Department of Biological Sciences and School of Marine Sciences, University of Maine, Orono, ME 04469. 3Department of Biological Sciences, University of Maine, Orono, ME 04469. 4Current address - PO Box 709, Bar Harbor, ME 04609. *Corresponding author - bbeal@maine.edu. 198 Northeastern Naturalist Vol. 11, Special Issue 2 and Short 2003, Phillips et al. 1983a). Its latitudinal distribution ranges from the Bering Sea to the Sea of Cortez in the Pacific Ocean and from Greenland to Georgia in the Atlantic Ocean. The growth morphology of Z. marina is important ecologically because its productivity and perennial nature provides predictable habitat, nursery, and refuge for infaunal and epifaunal organisms (Heck et al. 1989; Thayer et al. 1975a,b). It also forms the basis of important trophic links in many marine food webs (Fenchel 1977). For example, decomposition of both above- and belowground biomass releases inorganic nitrogen and phosphorus which can be captured again for plant and/or macroalgal production (Hemminga and Duarte 2000). Seagrass meadows are among the most productive marine or terrestrial systems, with values ranging from 8 g C m-2 day-1 for Zostera marina in Alaska (McRoy 1966, 1970) to 16 g C m-2 day-1 for Thalassia testudinum Banks ex Konig in Florida (Odum 1963). To begin to understand the contribution of eelgrass meadows to higher trophic levels and the broader community, it is necessary to evaluate the growth dynamics and energetics of the plant, especially changes in biomass and productivity. In the case of seagrass communities, the calculation of biomass and primary productivity usually refers to the aboveground portion (grass blades) of the plant. Eelgrass grows leaves from a basal meristem and continuously produces and sheds leaves, thereby allowing growth and production dynamics to be determined. Individual leaves can be marked, and new growth and leaf survival followed until the leaves die (Zieman 1974). Unmarked small leaves appearing between sampling dates also represent new growth. Conversely, because leaves also die and are lost during an interval, turnover times can be determined and organic input into the system estimated (Burdick 1988). The overall objective of this study was to determine aboveground productivity and turnover rates for eelgrass meadows in Cobscook Bay, ME. Also, we determined growth (elongation) rates of leaves and the rate of initiation of new leaves. This is important because productivity may be the result of one or both of these processes. This is the first appraisal of growth and productivity of Z. marina in eastern Maine, although not the first near this particular latitude (Robertson and Mann 1984). Other studies in Maine were conducted in mid-coast (Burdick 1988) and near the southwestern border with New Hampshire (Gaeckle and Short 2003). Materials and Methods Field sampling and laboratory measurements The energetics of Zostera marina leaves was studied within the low intertidal from July 1995 to July 1996 at two major sites in Cob2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 199 scook Bay (Bell Farm [44o49.37'N, 67o09.34'W] and Mahar Cove [44o 53.16'N, 67o 08.48'W]), and from June through August 1995 at a third site (Weir Cove [44o48.81'N, 67o08.29'W]) (Fig. 1). The Weir Cove site was abandoned after four sampling dates (June to August 1995) due to logistical constraints. The sites reflect two slightly different flow regimes (Brooks et al. 1999). Mahar Cove and Weir Cove represent low-flow (< 0.2 m s-1), fine-grained muddy sediments. Because the Bell Farm site lies at the edge of a constricted channel, water movement was more heterogeneous across tidal cycles, and both low and high-flow (> 0.5 m s-1) regimes were observed, resulting in both muddy and gravelly substrates. The two major sites were within 8 km of each other. Eelgrass beds were largely distributed in the low intertidal zone and/or sublittoral fringe, and the time available for sampling at this tidal level was limited. Figure 1. Map of Cobscook Bay, ME, eelgrass study sites during 1995–96. BF = Bell Farm, MC = Mahar Cove, and WC = Weir Cove. Washington County 200 Northeastern Naturalist Vol. 11, Special Issue 2 We estimated spatial variability in leaf biomass and growth (new leaf production) at Mahar Cove and Bell Farm at the beginning of our study (first four sampling intervals). At each site, we divided the area sampled into six discrete, randomly chosen line transects. Due to differences in the physical structure of the two sites, it was not possible to use an identical sampling design at both. For the first four sampling dates at Mahar Cove, we established six 10-m transects with 2-m spacing between each transect and measured five randomly selected plants per transect. At Bell Farm, we partitioned the site into four discrete transects for the first two sampling dates. Two transects were located near a constricted channel where sediments were gravelly. One of these transects was intertidal, and the other was at the sublittoral fringe where plants appeared more robust. The other two transects were located in a shallower, muddy intertidal area. Measurements were made on six randomly chosen plants per transect. Growth measurements were made biweekly during summer, monthly during spring and fall, and bimonthly during winter. This sampling schedule reflected seasonal growth of the plants (see below), and longer sampling intervals were necessitated during winter because of extensive ice cover over the eelgrass meadows. At all sites, 20 to 30 plants were marked with a numbered tag (30 cm long metal rod of 14 gauge wire) inserted into the sediments next to the plant (shoot). To avoid tagging the same plant or rhizome system, each tagged plant was separated by 1 to 1.5 m. A small hole was punched with a dissecting needle through the middle of the sheath (through all inner and outer leaves) near the sediment surface. This is a slightly modified version of Zieman’s technique (Ibarro-Obando and Boudouresque 1994). New plants were successively tagged and harvested (taken to the laboratory) at biweekly, monthly, or bimonthly intervals depending on growth during that period. Plants were sampled at the end of the interval by cutting the rhizome below the second node and placing them in numbered individual plastic bags. Shoot densities were determined in August 1995 when densities of perennial shoots were highest (see also Burdick 1988). This results in a small overestimate of productivity for the winter months, but has no effect on leaf initiation or elongation rates (the other factors contributing to production) for that period. In the laboratory, plants were washed to remove sediments and cleaned of epiphytes. The length of the longest leaf per plant was measured from the second node to the distal end of the leaf. Elongation of each leaf was determined by measuring the distance from the hole in the sheath to the hole in the blade. It is generally recognized that the outermost leaf of seagrasses is the oldest and will not grow (elongate) 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 201 during a test interval (Ibarro-Obando and Boudouresque 1994). Therefore, the hole on the outermost leaf is considered an accurate reference point. Leaves were separated into two classes: (1) the outer, older leaves that have a hole, and (2) the inner, young leaves that were produced after the hole was punched and do not have a hole. Total growth was the sum of the length of the new leaves plus the new growth of the older leaves. Other measurements included the width of the widest leaf (above the sheath) and the distance from the reference point to the second node. Following these measurements, each shoot was dried at 60 ºC to a constant weight and weighed, with weight recorded as grams dry weight. We measured several water column, or environmental, variables at Bell Farm and Mahar Cove during 1995–1996: temperature, salinity, nitrate, and total phosphorus. Water samples for nutrient analysis were taken in 1-liter brown plastic bottles 25 cm below the seawater surface. Samples were immediately placed on ice, maintained in the dark, and placed in a low-temperature (4 oC) box overnight. Samples were prepared for analysis within 24 hours and were filtered through a 0.45 μ membrane filter. Nitrate was analyzed with a Lachat flow injection analyzer by the cadmium reduction method. Phosphorus was analyzed by inductively coupled plasma analysis with a T.J.A. Atomcomp Model 975 analyzer. Aboveground biomass estimates Biomass is usually defined in seagrasses as the weight of all living material in a unit area at a given time, including the root structure (Zieman and Wetzel 1980). In this study, only aboveground biomass was measured. Aboveground biomass is synonymous with shoot biomass. It is important to note that estimates based on aboveground shoots can greatly underestimate total biomass, especially in seagrasses. The aboveground biomass at each site per sampling period was estimated by multiplying the mean weight per shoot from each site by the summer density of shoots at that site. We calculated a weighted mean annual aboveground biomass estimate for each site using Equation 1: Σ (# days per interval x mean biomass for interval) Σ (# days of all intervals) Although more samples were taken at Mahar Cove, only samples from 10 August 1995 to 30 July 1996 were used to estimate average annual biomass for comparison with Bell Farm samples. We made no attempt to sample the belowground biomass (root and rhizome system) due to the difficulty of excavation and the destructive nature of sampling below ground. However, the ratio of above- to belowground biomass in seagrasses is nearly 1:1 (Duarte and Chiscano 1999). 202 Northeastern Naturalist Vol. 11, Special Issue 2 Leaf elongation and initiation rates Two other measurements used in this study are leaf elongation rates (LER) and the initiation rate (IR) of new leaves. LER per shoot was calculated by summing all new growth per shoot and dividing by the number of days in the sampling period. The IR of new leaves during an interval was calculated by summing the number of new leaves produced per plant, dividing by the number of plants sampled, dividing that number by the number of days in the sampling period (IR x 100% = percent .shoot-1 .day-1). A weighted mean annual leaf elongation and initiation rate was calculated using Equation 1 by substituting LER and IR for mean biomass, respectively. Aboveground productivity estimates Productivity is defined as the weight of new organic material added over a period of time (Zieman and Wetzel 1980). Our estimate of productivity is equivalent to net productivity because it does not include energy lost due to respiration, injury, grazing, or death. The following steps were utilized in calculating productivity from our measurements. We converted length and width measurements (see above in Field Sampling and laboratory measurements subsection, paragraph 4) into weight on each sampling date by developing a multiple regression model of dry weight (dependent variable) versus leaf length and width (independent variables). To estimate new growth during a sampling interval (expressed as g dry wt shoot-1 day-1), new growth (elongation) for each shoot was converted into dry weight using the regression model. The average dry weight per shoot for each site was divided by the number of days in the sampling interval. This rate was then multiplied by the summer density of shoots at that site to provide an estimate of productivity. Finally, the productivity estimate expressed in g dry wt m-2 day-1 was then transformed to g C m-2 day-1 using a conversion factor of 0.38 (Thom 1990, Westlake 1963). Site-specific annual estimates of primary productivity were calculated by multiplying the weighted mean productivity rate (substituting productivity rate for biomass in Equation (1) by 365 days. Turnover estimates Turnover rates were calculated for the individual sampling periods by dividing the productivity for that period ([g dry wt m-2 day-1] x number of days per period) by biomass for the same period (P/B). Turnover time for leaves or shoots is the reciprocal of turnover rate day-1. To estimate the average annual turnover rate, we calculated a weighted mean turnover per day (substituting mean turnover for biomass in Equation 1) and multiplied this by 365 days. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 203 Statistics To provide visual inspection of differences in seasonal patterns, graphical data for the individual sampling periods are presented as untransformed means ± 95% confidence intervals. One- and two-way analysis of variance (ANOVA) were used to compare mean estimates of biomass, productivity, leaf elongation rate, and leaf initiation rates between Weir Cove, Bell Farm, and Mahar Cove sites (α = 0.05). Separation of means was performed using the Student-Neumann-Keuls (SNK) test. In addition, we tested for spatial effects (low intertidal vs. sublittoral fringe transects) for leaf biomass and amount of new leaf production for the first two sampling dates in 1995 at Bell Farm and for the first four sampling dates in 1995 at Mahar Cove. ANOVA was performed on the untransformed leaf biomass and new leaf production data by comparing lower intertidal (n = 3) to sublittoral transects (n = 1) at Bell Farm. In addition, we used regression analysis to determine whether any of the individual environmental variables were correlated significantly with productivity (g dry wt m-2 day-1). Results Spatial variability in leaf biomass and new leaf production The population of eelgrass at Bell Farm exhibited significant heterogeneity in plant morphology along the tidal gradient. For example, during the first sampling interval (31 July to 10 August 1995), mean leaf biomass was 2.7 times greater for plants in the sublittoral fringe than those in the low intertidal (1.15 ± 0.46, n = 6 vs. 0.43 ± 0.13 gm dry wt, n = 18; F = 13.07; df = 3, 20; P < 0.0001). An a priori comparison of sublittoral to lower intertidal areas accounted for 79% of the total variation due to spatial differences. New leaf production from July 31 to August 10 followed a similar pattern. Production of sublittoral plants was 1.7 times that of lower intertidal plants (102.6 ± 22.52, n = 6 vs. 60.09 ± 9.82 cm, n = 18; F = 9.12; df = 3, 20; P < 0.0001). The a priori comparison explained 82% of the total variation due to spatial effects. Similar results were obtained during the second sampling interval (10 to 29 August 1995). Leaf biomass of plants in the sublittoral fringe was 2.3 times greater than those in the lower intertidal areas (0.60 ± 0.46, n = 5 vs. 0.26 ± 0.06 gm dry wt, n = 14; F = 3.48; df = 3, 15; P = 0.043). New leaf production was 1.6 times greater for plants in the sublittoral fringe compared with those in the lower intertidal (133.35 ± 43.71, n = 4 vs. 83.11 ± 8.89 cm, n = 14; F = 7.29; df = 3, 14; P = 0.004). For both tests, 91% and 99% of the variation, respectively, was due to the comparison between the sublittoral fringe and intertidal transects. 204 Northeastern Naturalist Vol. 11, Special Issue 2 At Mahar Cove, we observed significant spatial effects during two of the four sampling intervals (1 July to 12 July and 31 July to 10 August 1995). The effects were not consistent, however, and involved only one of the two variables examined during those two intervals. In the first interval, significant differences between transects were noted for leaf biomass (P = 0.006) which ranged from 0.49 ± 1.27 (n = 2) to 0.06 ± 0.07 gm dry wt (n = 2), but not for new leaf production (mean = 38.04 ± 9.47 cm, n = 11; P = 0.679). From 12 to 31 July, no discernible spatial patterns for either variable were observed (mean leaf biomass = 0.34 ± 0.09 gm dry wt, n = 23; P = 0.524; mean new leaf production = 61.10 ± 10.26 cm, n = 23; P = 0.192). Significant differences between transects were observed for new leaf production from 31 July to 10 August (P = 0.049), but not for leaf biomass (P = 0.093). During that interval, mean leaf biomass was 0.16 ± 0.06 gm dry wt (n = 14) and mean new leaf production ranged from 73.23 ± 43.14 (n = 3) to 31.58 ± 21.30 cm (n = 5). Spatial effects for the two variables were not observed at this site from 10 to 29 August 1995 (mean leaf biomass = 0.24 ± 0.05, n = 23, P = 0.071; mean new leaf production = 66.50 ± 12.55 cm, P = 0.075). Biomass Shoot densities taken during the maximum period of growth were approximately two times greater at Mahar Cove (168 plants m-2) than at Bell Farm (88 plants m-2) or Weir Cove (75 plants m-2). Aboveground Biomass (g dry wt m-2) ± 95% Cl 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 205 estimates of biomass for Zostera marina showed large seasonal changes at both of the major sites (Fig. 2), and ANOVA confirmed these temporal differences (P < 0.0001). No temporal pattern in mean biomass was observed at Weir Cove over the four sampling dates between early July to early September 1995 (Fig. 2; `). A two-factor ANOVA on mean biomass using data common to all three sites and sampling dates (10 and 29 August 1995) revealed a significant interaction (P = 0.0038). A single-factor ANOVA for the 10 August sampling date demonstrated significant differences between sites (P = 0.0063), and a SNK test indicated that mean biomass at Bell Farm (53.4 ± 16.5 g dry wt m-2, n = 24) was nearly double the value at Mahar Cove and Weir Cove, which were equal (28.0 ± 5.1 g dry wt m-2, n = 31). No significant differences in mean biomass were observed between the three sites on 29 August (mean = 34.2 ± 5.5 g dry wt m-2, n = 60; P = 0.2021). For Bell Farm and Mahar Cove, highest mean biomass estimates occurred during the summer months, while the lowest occurred during late fall, winter, and early spring, but the pattern at each site differed significantly (P < 0.0001). At Bell Farm, highest mean biomass occurred on 10 August (see above), whereas biomass estimates on all other dates were statistically equivalent according to the a posteriori SNK test (mean = 25.7 ± 2.8 g dry wt m-2, n = 130). Biomass estimates were consistently higher at Mahar Cove (6 of 8 common sampling dates MC > BF), the low flow site, and ranged from a high of 77.0 ± 9.7 g dry wt m-2 (n = 24) on 31 July 1996 to a low of 18.3 ± 3.7 g dry wt m-2 (n = 23) on 17 April 1996. According to the SNK test, the Zostera population at Mahar Cove had two peaks of biomass, October 1995 and July 1996, and a drastic decline in mid-August 1995 (Fig. 2). The two distinct peaks were not evident in the eelgrass population at Bell Farm. However, plants at Bell Farm did have a complete die-back of aboveground biomass in April 1996. Leaf initiation The initiation rate of new leaves at Mahar Cove and Bell Farm showed a strong seasonal trend with highest rates in the summer and fall and lowest rates in winter and spring, although patterns through time differed between sites (P = 0.0014; Fig. 3). New leaf production at Mahar Cove peaked in August and November 1995, declined until April 1996, and then rose again through July 1996. At Bell Farm, leaf initiation peaked during both summers and was lowest from November 1995 through April 1996. Mean values for new leaf initiation at Mahar Cove ranged from 3.3 ± 1.6 to 18.6 ± 6.0 percent shoot-1 day-1, whereas at Bell Farm, means ranged from 6.0 ± 3.9 to 23.4 ± 7.5 percent shoot -1 day -1. 206 Northeastern Naturalist Vol. 11, Special Issue 2 Leaf elongation and morphology The patterns of leaf elongation exhibited a strong seasonal trend (Fig. 4). Highest rates occurred during both summers, whereas low- Figure 3. Rate of initiation of new leaves of Zostera marina at two sites in Cobscook Bay during 1995–96. Figure 4. Leaf elongation rate of Zostera marina at three sites in Cobscook Bay during 1995–96. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 207 est, but surprisingly positive, values were recorded during winter. On the sampling dates common to all three sites (see above), the pattern of leaf elongation differed significantly between sites (P = 0.0062). At Mahar Cove and Bell Farm, rates declined significantly (P < 0.05) between 10 and 29 August (e.g., 26–30% decline). No similar decline in leaf elongation rate was observed between sampling dates at Weir Cove (P = 0.1407), however the average rate (2.6 ± 0.4 cm shoot-1 day- 1) was 65–130% less than the rates observed at Mahar Cove and Bell Farm, respectively. Highest leaf elongation rate peaked first at Weir Cove in late July 1995 (9.0 ± 1.5 cm shoot-1 day-1; n = 18) followed by the other two sites approximately one month later. Lowest rates at Bell Farm and Mahar Cove occurred from January through April 1996 and then peaked once again in early summer. Rates ranged from 0.6 ± 0.8 cm shoot-1 day-1 (n = 4) during winter to 7.1 ± 1.2 cm shoot-1 day-1 (n = 24) in summer at Bell Farm and from 0.65 ± 0.1 cm shoot-1 day-1 (n = 11) during winter to 4.9 ± 1.3 cm shoot-1 day-1 (n = 14) in summer at Mahar Cove. Leaf width varied seasonally at Mahar Cove, Wier Cove, and Bell Farm (P < 0.001). At Mahar Cove and Bell Farm, leaf width was significantly narrower in plants sampled in the fall and winter compared to those in the summer. At Weir Cove, leaf width increased significantly from 13 to 20 July 1995 and then progressively decreased until the fi- nal sampling date on 1 September 1995. In addition, plant morphology differed greatly from the intertidal to sublittoral transects at Bell Farm. Leaves from the sublittoral fringe transect were approximately 30% wider (P = 0.001) than those in the intertidal. Similarly, the length of the longest leaf per plant was nearly 70% greater for shoots in the fringe transect (P = 0.005). Productivity estimates The linear relationship for converting length and width measurements of leaves into dry weight is given by the following equation: Dry weight = 0.037 + 0.002 (length) + 0.008 (width) The regression was highly significant (F = 450.70, P < 0.0001, r2 = 0.94), however the addition of the variable “leaf width” to a model already containing “leaf length” was not significant (P = 0.06). Productivity patterns for eelgrass at the three sites on the two common sampling dates (see above) were similar (P = 0.0877), with significant differences occurring between each site (P < 0.0001), but not among dates (P = 0.0511). Mean productivity estimates were highest at Mahar Cove, lowest at Weir Cove, and intermediate at Bell Farm (1.302 ± 0.22, n = 37; 0.369 ± 0.068, n = 36; and 1.057 ± 0.143 g dry wt m-2 day-1, n = 42; respectively). 208 Northeastern Naturalist Vol. 11, Special Issue 2 Productivity patterns for eelgrass differed significantly at the two major sites (P = 0.0124; Table 1 and Fig. 5). ANOVA and the SNK test detected two peaks in productivity at Mahar Cove in 1995: late July (1.490 ± 0.461 g dry wt m-2 day-1, n = 14) and October (1.081 ± 0.217 g dry wt m-2 day-1, n = 22). Lowest values occurred in January and April (meanpooled = 0.201 ± 0.056 g dry wt m-2 day-1, n = 34). ANOVA indicated no significant differences between July 1995 and 1996 peaks (P > 0.05). Although there appeared to be a single peak for productivity in 1995 at Bell Farm (8 August; Fig. 5), SNK indicated no significant difference in mean productivity between the first two (August 1995) sampling dates. By early September, productivity at Bell Farm declined by 60% and remained below 0.75 g dry wt m-2 day-1 until June Table 1. Mean Productivity (g dry wt) of Zostera marina at two sites in Cobscook Bay during 1995–96. Productivity (g dry wt m-2 day-1) Date Days # plants Mean Std. dev. Bell Farm 08/10/1995 10 24 1.215 0.511 08/29/1995 19 18 0.846 0.275 09/08/1995 10 24 0.449 0.351 11/04/1995 28 6 0.172 0.097 01/16/1996 61 4 0.095 0.089 06/18/1996 13 21 0.744 0.382 07/01/1996 13 23 1.021 0.435 07/17/1996 16 22 0.626 0.248 07/31/1996 14 18 0.874 0.438 Mean 20.4 17.8 0.481* 0.383 Mahar Cove 07/12/1995 12 11 0.965 0.465 07/31/1995 19 23 1.031 0.456 08/10/1995 10 14 1.490 0.798 08/28/1995 18 23 1.188 0.563 09/08/1995 11 21 0.657 0.424 10/09/1995 29 22 1.081 0.490 11/03/1995 25 18 0.863 0.369 01/16/1996 61 11 0.185 0.076 04/17/1996 29 23 0.176 0.099 06/18/1996 14 20 0.929 0.484 07/01/1996 13 24 1.476 0.612 07/17/1996 16 24 1.450 0.539 07/30/1996 14 24 1.416 0.451 Mean 20.9 19.9 0.784* 0.512 *Weighted mean For comparative purposes with Bell Farm, only samples from 8/10/95 to 7/30/96 were included in the weighted mean for Mahar Cove.The long intervals between January and April (Mahar) and January and June (Bell) was due, in part, to ice cover (Mahar and Bell) and disturbance by geese (Bell). 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 209 1996 (Fig. 5). Peaks occurred at both sites in July 1996, but means for that month were significantly higher at Mahar Cove than Bell Farm (P < 0.0001; 1.448 ± 0.125, n = 72, vs. 0.841 ± 0.104 g dry wt m-2 day-1, n = 63). Specific estimates of productivity were highly seasonal and ranged from 0.095 g dry wt m-2 day-1 from November to January to 1.215 g dry wt m-2 day-1 in August at Bell Farm. Average annual productivity was 0.784 ± 0.512 g dry wt m-2 day-1 at Mahar Cove and 0.481 ± 0.512 g dry wt m-2 day-1 at Bell Farm. To allow for comparison with previous studies, productivity was also converted to carbon and expressed as g C m-2 day-1 (Fig. 5). The converted productivity estimates at Bell Farm ranged from 0.036 to 0.462 g C m-2 day-1 with weighted average annual mean of 0.183 ± 0.146 g C m-2 day-1. The carbon based productivity estimates for Mahar Cove ranged from 0.067 to 0.566 g C m-2 day-1 with weighted average annual mean of 0.298 ± 0.195 g C m-2 day-1). Turnover The turnover rates of eelgrass also showed strong seasonal patterns, being highest during the warmer months (Table 2). The weighted average length of time for plants to fully turn over their leaves at the two sites ranged from 50.5 to 56.7 days (7 to 8 weeks; Table 2). The estimated number of turnovers of eelgrass plants at Bell Farm and Mahar Cove averaged 7.2 and 6.4 turnovers per year, respectively. Figure 5. Productivity of Zostera marina at three sites in Cobscook Bay during 1995–96. 210 Northeastern Naturalist Vol. 11, Special Issue 2 Relationship between productivity and environmental variables Several environmental factors were monitored during the tagging studies to determine if productivity could be correlated with one or more of these variables. These data are presented graphically for each variable at the two sites (Figs. 6–9). Salinity. At Mahar Cove, salinity was relatively constant and ranged from 31 to 34 ppt, suggesting well-mixed conditions at that site. There was no significant relationship between productivity and salinity at this site (r2 = 0.012; P = 0.7806; df = 1, 7; Fig. 6). Salinity values at Bell Farm were more variable and exhibited a dramatic reduction during April 1996 due to snow melt and run-off from a pond headwater stream. The die-off and reduced biomass of eelgrass at Table 2. Mean turnover of Zostera marina at two sites in Cobscook Bay during 1995–96. Turnover # of days Date Days # plants per period for T/O Bell Farm 08/10/1995 10 24 0.227 44.05 08/29/1995 19 18 0.543 34.99 09/08/1995 10 24 0.227 44.05 11/04/1995 28 6 0.556 50.36 01/16/1996 61 4 0.416 146.63 06/18/1996 13 21 0.468 27.78 07/01/1996 13 23 0.455 28.57 07/17/1996 16 22 0.404 39.60 07/31/1996 14 18 0.348 40.23 Mean 20.4 17.8 0.404* 50.50 Mahar Cove 07/12/1995 12 11 0.361 33.24 07/31/1995 19 23 0.354 53.67 08/10/1995 10 14 0.559 17.89 08/28/1995 18 23 0.529 34.03 09/08/1995 11 21 0.153 71.89 10/09/1995 29 22 0.504 57.54 11/03/1995 25 18 0.571 43.78 01/16/1996 61 11 0.399 152.88 04/17/1996 29 23 0.279 103.94 06/18/1996 14 20 0.275 50.91 07/01/1996 13 24 0.361 36.01 07/17/1996 16 24 0.345 46.38 07/30/1996 14 24 0.257 54.48 Mean 20.9 19.9 0.369* 56.71 *Weighted mean. Average annual turnover = (weighted mean / average # days) x 365 days Bell Farm: (0.404 / 20.4) x 365 = 7.23 turnovers per year Mahar Cove: (0.369 / 20.9) x 365 = 6.44 turnovers per year For comparative purposes with Bell Farm, only samples from 8/10/95 to 7/30/96 were included in the weighted mean for Mahar Cove. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 211 Figure 6. Relationship between eelgrass productivity and salinity at two sites in Cobscook Bay during 1995–96. that time may be related to the influx of freshwater (Fig. 6); however, there was no significant correlation between productivity and salinity (r2 = 0.481; P = 0.0839; df = 1, 5). Water Temperature. Productivity was positively correlated with the seasonal changes in temperature at Mahar Cove (r2 = 0.570; P = 0.0468; df = 12, 5; Fig. 7). At Bell Farm, the r2 value between temperature and productivity was 0.828; however, this was not significantly different from zero because the number of data points was low (n = 4). Nitrate. Nitrate exhibited a highly variable seasonal pattern and did not appear to influence growth responses in eelgrass at either Mahar Cove or Bell Farm (Mahar: r2 = 0.254; P = 0.1373; df = 1, 8; Bell: r2 = 0.261; P = 0.3000; df = 1, 4). A small increase in nitrate 212 Northeastern Naturalist Vol. 11, Special Issue 2 may have influenced the spike in production at the former site in late summer–fall (Fig. 8). However, the general increase in nitrate lags behind growth. Similarly, the decline in production in November precedes the decline in nitrate levels. Phosphate. The patterns of phosphate concentration were highly variable at the two sites (Fig. 9). There was no significant correlation between phosphate levels and seasonal changes in productivity at Mahar Cove (r2 = 0.0001; P = 0.9801; df = 1, 8). During the spring of both years, growth appeared to follow the increase in phosphate levels. After this time, however, productivity was independent of phosphate levels. Similar results occurred at Bell Farm (r2 = 0.248; P = 0.2089; df = 1, 6). The low phosphate values at Bell Farm during spring suggest that snow run-off may have diluted phosphate concentrations at the site (Fig. 9). Figure 7. Relationship between eelgrass productivity and temperature at two sites in Cobscook Bay during 1995–96. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 213 Discussion The overall objective of this study was to estimate the aboveground productivity and energetic contributions of eelgrass, Zostera marina, to Cobscook Bay. The results of the tagging studies show that growth, productivity, and turnover are variable both within and between sites, and are closely related to season. Because the goal of the study was to provide an overall estimate of seagrass productivity at selected sites within Cobscook Bay, we did not incorporate the within-site variability described here into our general analyses of biomass, productivity, turnover, and leaf elongation rates. However, these data show that even within intertidal sites that appear homogeneous, there can be substantial variability in the parameters used to assess productivity. Figure 8. Relationship between eelgrass productivity and nitrate concentration at two sites in Cobscook Bay during 1995–96. 214 Northeastern Naturalist Vol. 11, Special Issue 2 As expected, there are distinct seasonal patterns of growth and productivity, with higher values during warmer months and lower values during the remaining months of the year (Dennison 1987, Nelson and Waaland 1997, Olessen and Sand-Jensen 1994). Not surprisingly, there was a significant positive correlation between rising water temperature and productivity. The relationship, however, may be spurious; the absence of experimental data and the fact that temperature is a function of solar irradiance (Thom and Albright 1990) makes it impossible to separate the effects of irradiance and temperature. Salinity, nitrate, and phosphate levels were not closely correlated with productivity. The nutrient patterns were out of phase with seagrass growth, and are more likely the result of phytoplankton Figure 9. Relationship between eelgrass productivity and PO4 concentration at two sites in Cobscook Bay during 1995–96. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 215 blooms and nutrient regeneration by zooplankton and other invertebrates (Garside and Garside 2004). Maximum productivity estimates for eelgrass in Cobscook Bay occurred in July and August, and were 1.22 (Bell Farm) to 1.49 (Mahar Cove) g dry wt m-2 day-1 (Table 1). This variation in seasonal growth is similar to that described for Z. marina and Z. japonica Aschers & Graebn. in Padilla Bay, WA (Thom 1990). There are several assumptions and/or limitations in this study. First, density was estimated quantitatively only once during the summer of 1995. Densities of eelgrass plants at all three sites were noticeably lower from fall to late spring. The consequence of this is that productivity is overestimated during that time. However, this potential bias has no effect on leaf elongation and leaf initiation rates, which were greater than zero during the colder months at both sites, because these estimates do not include measures of eelgrass density. Second, productivity was measured using both field growth measurements and a date-specific regression model based on leaf length and width to estimate dry weight. This estimate was then converted to grams carbon using a conversion factor that was obtained from the literature (Thom 1990), not a Cobscook Bay-specific factor. Third, fewer plants were sampled in winter and early spring due to ice cover and lower population densities, resulting in larger confidence intervals around specific means. Fourth, we do not know if the biomass and productivity estimates from our three sites are representative of others in the larger Bay system. Nonetheless, these two estimates show good concordance and provide a reasonable estimate of the aboveground energetics of eelgrass populations in Cobscook Bay. Fifth, the short overlap of sampling dates between years does not permit an adequate assessment of interannual variability. Nevertheless, we evaluated within-site variability for leaf biomass and new leaf production at the two major sites during the first two months of this study and found this component to be important in explaining differences in those two variables, especially between intertidal and sublittoral populations at Bell Farm. Leaf biomass and new leaf production exhibited different patterns at Mahar Cove and Bell Farm. At Mahar Cove, there was no consistent pattern for either variable, however, spatial effects were observed on two of four dates. From 1 to 12 July 1995, leaf biomass varied by an order of magnitude across the six transects (0.06 to 0.49 gm dry wt). From 31 July to 10 August 1995, new leaf production varied by 2.3 times between transects (31.6 to 73.2 cm). At Bell Farm, we observed a consistent spatial pattern with significantly higher leaf biomass and new leaf production for plants located in the sublittoral fringe compared with those 216 Northeastern Naturalist Vol. 11, Special Issue 2 in the intertidal. This single-degree-of-freedom comparison accounted for > 79% of the overall variability due to between-transect differences at this site. It is possible that these differences are due to disparities in the physical environment at Bell Farm (e.g., water depth, flow regime, disturbance from ice, light, etc.) that could cause differential stresses (Middelboe et al. 2003). Plant morphology (leaf width) differed greatly from the intertidal to sublittoral transects at Bell Farm and seasonally at all three sites. Abal et al. (1994) observed similar differences in plant morphology due to depth (light) in Zostera capricorni Aschers. These morphological differences may have played a role in the variation observed in leaf biomass and new leaf production and suggest that annual production among sublittoral plants is significantly (≈ 2 times) greater than those within the intertidal (Dennison 1987, Kentula and McIntire 1986). Comparison of the two major sites in Cobscook Bay reveals both similarities and differences in population and energetic dynamics. Direct comparison of the two sites reveals that biomass is greater at Mahar Cove than Bell Farm (P = 0.002), but that productivity was similar between the two sites (P = 0.074). At an individual level, leaf elongation rates and new leaf initiation rates appeared higher at Bell Farm than Mahar Cove, but were not statistically different (P = 0.336 and P = 0.186, respectively). Thus, the differences in biomass between the sites may be attributed to the higher density of eelgrass at Mahar Cove, where within-site differences in water depth occurred (see Middleboe et al. 2003). The higher water flow at Bell Farm and other within-site differences may have resulted in higher mean individual growth rates. Higher individual growth rate usually results in larger plants that shade each other, similar to what has been observed subtidally (Phillips et al. 1983b). Eelgrass at Mahar Point had an interesting double peak in both biomass and productivity during August 1995. Harlin and Thorne- Miller (1981) also observed double peaks in summer in a Rhode Island lagoon. They suggested that a build-up of green algae suffocated and shaded the eelgrass plants. Once the green algae died or eroded, the eelgrass resumed growth and produced another peak. This may explain partially the die-back at Mahar Cove. Although there was no widespread bloom of green algae (cf., Vadas and Beal 1987) on the flats at Mahar Cove, there was an accumulation of filamentous and other green algae on the wire marker flags used to locate tagged plants (B. Beal, pers. observ.). Another possibility for the observed difference may be that sediments frequently are resuspended on mudflats, thereby reducing light and photosynthesis. Light availability was considered 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 217 the principal factor governing growth in Zostera marina in Chesapeake Bay (Moore and Wetzel 1999) and Waquoit Bay, MA (Hauxwell et al. 2003). Thus, it is possible that our experimental plants could have been shaded and received lower irradiance levels during the period of reduced growth. Seasonal patterns of temperature, salinity, nitrogen, and phosphorus do not appear to explain the cause of the double peaks. Eelgrass at Bell Farm exhibited unimodal growth throughout the 1995 season. However, the complete absence of aboveground bio-mass in April 1996 is anomalous. The apparent negative correlation with salinity (Fig. 6) suggests that the spring ice melt may have stressed the plants and caused the leaves to defoliate. Damage to plants as a result of ice rafting may also have occurred. In addition, a flock of migrating Canadian geese rested at the site and foraged on the upright shoots (B. Beal, pers. observ.). Similarly, Brent Geese Table 3. Comparison of productivity for eelgrass, Zostera marina. Location TL1 H2 Est. DW3 Est. gC4 Reference Izembeck Lagoon, AK I a 8.7–21.1 3.3–8.0 McRoy 1966 and 1970 Netarts Bay, OR I a 4.7–13.6 1.79–5.17 Kentula & McIntire 1986 Cobscook Bay, ME I a 0.11–1.5 0.04–0.57 This study Tills Cove, ME I a 1.35 0.52 Burdick 1988 San Quintin Bay, MX5 I a 12.9 4.9 Ibarra-Obando and Huerta- Tamayo 1987 Zandkreek, Netherlands I a 1.64 0.62 Vermaat et al. 19876 Roscoff, France I a 3.04 1.16 Jacobs 1979 Roscoff, France I b 1.37 0.52 Jacobs 1979 Puget Sound, WA S a 1.8–10.5 0.7–4.0 Phillips 1972 Padilla Bay, WA S a 0.55 0.21 Thom 1990 Friday Harbor, WA S a 4.1–5.2 1.56–1.98 Nelson 1997 Friday Harbor, WA S t 4.8 1.84 Nelson and Waaland 1997 Point Judith, RI S a 1.1–7.6 0.4–2.9 Conover 1968 Point Judith, RI S a 5.2 1.98 Hamburg and Homann 1986 Point Judith, RI S b 1.2–1.5 0.46–0.57 Hamburg and Homann 1986 Nauset Marsh, MA S a 2.7 1.03 Roman and Able 1988 Town Cove, MA S a 1.2 0.46 Roman and Able 1988 Chezzetcook Inlet, NS S a 1.46 0.56 Robertson and Mann 1984 Chezzetcook Inlet, NS S b 1.08 0.41 Robertson and Mann 1984 Oeresund, Denmark S a 8.54 3.24 Wium-Andersen and Borum 1984 Vellerup Vig, Denmark S a 2.4–8.4 0.9–3.2 Sand-Jensen 1975 Limfjorden, Denmark S t 1.8–2.7 0.64–1.03 Olesen and Sand-Jensen 1994 1Tide level: I = intertidal, S = subtidal 2Habitat: a = above-ground, b = below-ground, t = total (above- and below-ground) 3DW = g dry weight m-2 day-1 4gC = g C m-2 day-1 (calculated by multiplying g dry weight m-2 day-1 by 0.38) 5Baja California, Mexico 6Zostera noltii 218 Northeastern Naturalist Vol. 11, Special Issue 2 (Branta bernicla (L.)) and several duck species reduced approximately one-half of the aboveground biomass of intertidal eelgrass, Zostera noltii Hornem, during the fall and early winter in the Dutch Wadden Sea (Jacobs et al. 1981). It is likely that the combination of these factors contributed to the drastic decline in shoot biomass. The seasonal estimates of aboveground eelgrass productivity in Cobscook Bay (0.04 to 0.57 g C m-2 day-1; Fig. 5) are within the range of other estimates from this geographic area. Earlier productivity estimates in Maine from south of Cobscook Bay (i.e., Penobscot River) during summer months (0.52 g C m-2 day-1; Burdick 1988) are similar to estimates observed in Cobscook Bay (Table 3). In addition, our estimates overlap within the range of productivity estimates (0.09 to 0.94 g C m-2 day-1 ; Table 3) for Nova Scotia (Robertson and Mann 1984). Duarte and Chiscano (1999) determined that the average ratio of above- to below-ground production in seagrasses was 16.4:1. Using this ratio, our range of estimates of eelgrass productivity in Cobscook Bay increases slightly from 0.042 to 0.61 g C m-2 day-1. When compared to studies outside this geographic area, our estimates of eelgrass productivity are low (Table 3). These lower estimates may reflect different life history strategies. The studies in Nova Scotia by Robertson and Mann (1984) may shed light on why our estimates of eelgrass productivity are relatively low. The eelgrass meadows there had similar but slightly higher aboveground productivity levels; however, their study was more extensive because it included below-ground productivity estimates. They found that perennial eelgrass plants allocated 43% of the total production to below-ground biomass and suggested that the distribution of biomass to the rhizome system was a mechanism to protect the genet from ice scouring. Also, the allocation of energy to the rhizome system may provide access to nutrients in the sediments (Abal et al. 1994). Intertidal eelgrass in Cobscook Bay may be allocating a large portion of energy to below-ground biomass, since estuaries in this region are usually covered by ice for several months (Gordon and Desplanque 1983). If we recalculate productivity by adding 43% to our above-ground estimates, the range becomes 0.06 to 0.82 g C.m-2 day-1, which is still lower than regions outside the northwest Atlantic (Table 3). Although eelgrass beds at the two sites in Cobscook Bay are not as productive as populations in warmer waters, these plants (shoots) are adding substantial amounts of carbon to the system as they die and decay and are therefore contributing to the productivity and organic input into the Bay. For example, we observed for Mahar Cove and Bell Farm, respectively, mean turnover rates between 6.4 and 7.2 plants per year, minimum turnover time for leaves between 18 and 28 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 219 days, and annual primary production estimates of 67 to 109 g C m-2 .year-1. These turnover rates are comparable to eelgrass populations in Cape Cod, MA (Roman and Able 1988), and Netarts Bay, OR (Kentula and McIntire 1986). The estimated area of subtidal eelgrass in Cobscook Bay was 466 hectares based on areal estimates of habitat types by Larsen et al. (2004) and Campbell (2004). From our observations of Cobscook Bay over the past two decades, we estimate the extent of intertidal populations to be about 5% (23 hectares) of subtidal values. Therefore, total (intertidal + subtidal) eelgrass production in Cobscook Bay is estimated to range from 10.9 to 17.5 x 108 g dry year-1, or 3.3 to 5.3 x 108 g C .year-1. Dense beds of macrophytes, such as eelgrass, can have profound effects on water movements and provide physical structure to an otherwise structurally homogeneous environment, such as sand and mudflat habitats (Fonseca et al. 1983). Macrophyte canopies reduce water flow, increase sedimentation, and reduce light intensities (Eckman et al. 1989). Zostera marina is clearly an important source of trophic energy in Cobscook Bay that contributes directly to benthic herbivores (Robertson and Mann 1982) that may feed on the leaves or epiphytes (e.g., Littorina littorea L., Idotea balthica (Pallas)) and to grazing wildfowl (e.g., Branta canadensis (L.)). In addition, eelgrass contributes indirectly to detritus-based food webs (Vähätalo and Søndergaard 2002). Cebrian and Duarte (2001) showed that herbivores consume < 15% of the production of seagrasses such as Posidonia oceanica (L.). Both intertidal and subtidal seagrass meadows increase habitat complexity and are areas of increased species richness and abundance (Lazzari and Tupper 2002). Seagrass meadows serve as seasonal habitat, refuge, and nursery for many invertebrates such as mussels, (Mytilus edulis L.; Newell et al. 1991, Short et al. 1991), sand shrimp (Crangon septemspinosa Say; B. Beal and R.L.Vadas, pers. observ.), green crabs Carcinus maenas (L.) (Beal 1994), and fish such as sticklebacks (Gasterosteus aculeatus L.), mummichogs (Fundulus heteroclitus (L.)), and Atlantic silversides (Menidia menidia (L.)) (B. Beal and R.L. Vadas, pers. observ.; Lazzari et al. 2003; Mattila et al. 1999). In addition, beds provide refuge for burrowing infaunal bivalves (Orth et al. 1984) and serve as spawning grounds and nursery sites for a number of commercial and recreational marine fish species (Matic-Skoko et al. 2004, Thayer et al. 1975a). Acknowledgments We thank the following people for helping in the field and the laboratory: Dennis Anderson, Sherrie Emerson, Jill Fegley, Jeff Rodzen, and Ken Vencile. 220 Northeastern Naturalist Vol. 11, Special Issue 2 We thank property owners at Bell Farm (Bob and Terry Bell) for allowing us access to their shoreland. We acknowledge support from the Maine Agricultural and Forestry Experiment Station and the Maine Sea Grant Program. The work was conducted as part of a research program entitled “Developing an Ecological Model of a Boreal, Macrotidal Estuary: Cobscook Bay, Maine,” funded by a grant from the A.W. Mellon Foundation to the Maine Chapter of the Nature Conservancy, with matching funds from the University of Maine at Orono and the University of Maine at Machias. We appreciate the comments of two anonymous reviewers on an earlier draft. Literature Cited Abal, E.G., N. Loneragan, P. Bowen, C.J. Perry, J.W. Udy, and W.C. Dennison. 1994. Physiological and morphological responses of the seagrass, Zostera capricorni Aschers, to light intensity. Journal of Experimental Marine Biology and Ecology 178:113–129. Beal, B.F. 1994. Biotic and abiotic factors influencing growth and survival of wild and cultured individuals of the soft-shell clam, Mya arenaria L., in eastern Maine. Ph.D. Dissertation, University of Maine, Orono, ME. 499 pp. Brooks, D.A., M.W. Baca, and Y.-T. Lo. 1999. Tidal circulation and residence time in a macrotidal estuary: Cobscook Bay, Maine. Estuarine, Coastal, and Shelf Science 49:647–665. Burdick, D.B. 1988. The ecology and productivity of eelgrass (Zostera marina, Zosteraceae) in a Maine estuary. M.Sc. Thesis, University of Maine, Orono, ME. 76 pp. Campbell, D.E. 2004. Evaluation and emergy analysis of the Cobscook Bay ecosystem. Northeast Naturalist 11(Special Issue 2):355–424. Cebrian, J., and C.M. Duarte. 2001. Detrital stocks and dynamics of the seagrass Posidonia oceanica (L.) Delile in the Spanish Mediterranean. Aquatic Botany 70:295–309. Conover, J.T. 1968. The importance of natural diffusion gradients and transport of substances related to benthic plant metabolism. Botanica Marina 11:1–9. Dennison, W.C. 1987. Effects of light on seagrass photosynthesis, growth, and depth distribution. Aquatic Botany 27:15–26. Duarte, C.M., and C.L. Chiscano. 1999. Seagrass biomass and production: A reassessment. Aquatic Botany 65:159–174. Eckman, J.E., D.O. Duggins, and A.T. Sewell. 1989. Ecology of understory kelp environments. I. Effects of kelps on flow and particle transport near the bottom. Journal of Experimental Marine Biology and Ecology 129:173–187. Fenchel, T. 1977. Aspects of the decomposition of seagrasses. Pp. 123–146, In C.P. McRoy and C. Helfferich (Eds.). Seagrass Ecosystems: A Scientific Perspective. Dekker Publishing, New York, NY. 314 pp. Fonseca, M.S., J.C. Zieman, G.W. Thayer, and J.S. Fisher. 1983. The role of current velocity in structuring eelgrass (Zostera marina L.) meadows. Estuarine, Coastal, and Shelf Science 17:367–380. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 221 Gaeckle, J.L., and F.T. Short 2003. A plastochrone method for measuring leaf growth in eelgrass, Zostera marina L. Bulletin of Marine Science 71:1237–1246. Garside, C., and J.C. Garside. 2004. Nutrient sources and distributions in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):75–86. Gordon, Jr., D.C., and C. Desplanque. 1983. Dynamics and environmental effects of ice in the Cumberland Basin of the Bay of Fundy. Canadian Journal of Fisheries and Aquatic Sciences 40:1331–1342. Hamburg, S.P., and P.S. Homann. 1986. Utilization of growth parameters of eelgrass, Zostera marina, for productivity estimation under laboratory and in situ conditions. Marine Biology 93:299–303. Harlin, M.M., and B. Thorne-Miller. 1981. Nutrient enrichment of sea grass beds in a Rhode Island coastal lagoon. Marine Biology 65:221–229. Hauxwell J., J. Cebrian, and I. Valiela. 2003. Eelgrass Zostera marina loss in temperate estuaries: Relationship to land-derived nitrogen loads and effect of light limitation imposed by algae. Marine Ecology Progress Series 247:59–73. Heck, K.L., Jr., K.W. Able, N.P. Fahy, and C.T. Roman. 1989. Fishes and decapod crustaceans of Cape Cod eelgrass meadows: Species composition, seasonal abundance patterns, and comparison with unvegetated substrates. Estuaries 12:59–65. Hemminga, M.A., and C.M. Duarte. 2000. Seagrass Ecology. Cambridge University Press. Cambridge, UK. 298 pp. Ibarra-Obando, S.E., and R. Huerta-Tamayo. 1987. Blade production of Zostera marina L. during the summer–autumn period on the Pacific coast of Mexico. Aquatic Botany 28:301–315. Ibarra-Obando, S.E., and C.F. Boudouresque. 1994. An improvement of the Zieman leaf marking technique for Zostera marina growth and production assessment. Aquatic Botany 47:293–302. Jacobs, R.P.W.M. 1979. Distribution and aspects of the production and biomass of eelgrass, Zostera marina L., at Roscoff, France. Aquatic Botany 7:151–172. Jacobs, R.P.W.M., C. Den Hartog, B.F. Braster, and F.C. Carriere. 1981. Grazing of the sea grass, Zostera noltii, by birds at Terschelling (Dutch Wadden Sea). Aquatic Botany 10:241–259. Kentula, M.E., and C.D. McIntire. 1986. The autecology of production dynamics of eelgrass (Zostera marina L.) in Netarts Bay, Oregon. Estuaries 9:188–199. Larsen, P.F., S. Barker, J. Wright, and C.B. Erickson. 2004. Use of remote sensing to map and measure marine intertidal habitats in support of ecosystem modeling efforts: Cobscook Bay, Maine. Northeast Naturalist 11(Special Issue 2):225–242. Lazzari, M.A., and B. Tupper. 2002. Importance of shallow water habitats for demersal fishes and decapod crustaceans in Penobscot Bay, Maine. Environmental Biology of Fishes 63:57–66. 222 Northeastern Naturalist Vol. 11, Special Issue 2 Lazzari, M.A., S. Sherman, and J.K. Kanwit. 2003. Nursery use of shallow habitats by epibenthic fishes in Maine nearshore waters. Estuarine, Coastal, and Shelf Science 56:73–84. Matic-Skoko, S., B. Antolic, and M. Kraljevic. 2004. Ontogenic and seasonal feeding habits of the annular seabream (Diplodus annularis L.) in Zostera sp. beds, eastern Adriatic Sea. Journal of Applied Ichthyology. 20:376–381. Mattila, J., G. Chaplin, M.R. Eilers, K.L. Heck, Jr., J.P. O’Neil, and J.F. Valentine. 1999. Spatial and diurnal distribution of invertebrate and fish fauna of a Zostera marina bed and nearby unvegetated sediments in Damariscotta River, Maine (USA). Journal of Sea Research 42:321–332. McRoy, C.P. 1966. The standing stock and ecology of eelgrass, Zostera marina, in Izembeck Lagoon, Alaska. M.Sc. Thesis, University of Washington. Seattle, WA. 138 pp. McRoy, C.P. 1970. Standing stocks and other features of eelgrass (Zostera marina) populations on the coast of Alaska. Journal of the Fisheries Research Board of Canada 27:1811–1821. Middleboe, A.L., K. Sand-Jensen, and D. Krause-Jensen. 2003. Spatial and interannual variations with depth in eelgrass populations. Journal of Experimental Marine Biology and Ecology 291:1–15. Moore, K.A., and R.L. Wetzel. 1999. Seasonal variations in eelgrass (Zostera marina L.) responses to nutrient enrichment and reduced light availability in experimental ecosystems. Journal of Experimental Marine Biology and Ecology 244:1–28. Nelson, T.A. 1997. Interannual variance in a subtidal eelgrass community. Aquatic Botany 56:245–252. Nelson, T.A., and J.R. Waaland. 1997. Seasonality of eelgrass, epiphyte, and grazer biomass and productivity in subtidal eelgrass meadows subjected to moderate tidal amplitude. Aquatic Botany 56:51–74. Newell, C.R., H. Hidu, B.J. McAlice, G. Podniesinski, F. Short, and L. Kindblom. 1991. Recruitment and commercial seed procurement of the blue mussel Mytilus edulis in Maine. Journal of the World Aquaculture Society 22:134–152. Odum, H.T. 1963. Productivity measurements in Texas turtle grass and the effects of dredging an intercoastal channel. Publication of the Institute of Marine Science, University of Texas, Austin, TX, 6:45–58. Olessen, B., and K.A.J. Sand-Jensen. 1994. Demography of shallow eelgrass (Zostera marina) populations: Shoot dynamics and biomass development. Journal of Ecology 82: 379–390. Orth, R.J., K.L. Heck, Jr., and J. van Montfrans. 1984. Faunal communities in seagrass beds: A review of the influence of plant structure and prey characteristics on predator-prey relationships. Estuaries 7:339–350. Phillips, R.C. 1972. Ecological life history of Zostera marina L. (eelgrass) in Puget Sound, Washington. Ph.D. Dissertation, University of Washington, Seattle, WA. 154 pp. 2004 B. Beal, R.L. Vadas, Sr., W.A. Wright, and S. Nickl 223 Phillips, R.C., W.A. Grant, and P.C. McRoy. 1983a. Reproductive strategies of eelgrass (Zostera marina L.). Aquatic Botany 16:1–20. Phillips, R.C., C. McMillan, and K.W. Bridges. 1983b. Phenology of eelgrass, Zostera marina L., along latitudinal gradients in North America. Aquatic Botany 15:145–156. Robertson, A.I., and K.H. Mann. 1982. Population dynamics and life history adaptations of Littorina neglecta Bean in an eelgrass meadow (Zostera marina L.) in Nova Scotia. Journal of Experimental Marine Biology and Ecology 63:151–171. Robertson, A.I., and K.H. Mann. 1984. Disturbance by ice and life-history adaptations of the seagrass Zostera marina. Marine Biology 80:131–141. Roman, C.T., and K.W. Able. 1988. Production ecology of eelgrass (Zostera marina L.) in a Cape Cod salt marsh-estuarine system, Massachusetts. Aquatic Botany 32:353–363. Sand-Jensen, K. 1975. Biomass, net production, and growth dynamics in an eelgrass (Zostera marina L.) population in Vellerup Vig, Denmark. Ophelia 14:185–201. Short, F., H. Hidu, H. Hoven, L. Kindlum, C. Newell, K. Verny, and A. Mathieson. 1991. The role of eelgrass (Zostera marina) in the recruitment of the blue mussel (Mytilus edulis) in Maine. Journal of Shellfish Research 10:300–301. Thayer, G.W., S.M. Adams, and M.W. LaCroix. 1975a. Structural and functional aspects of a recently established Zostera marina community. Estuarine Research 1:517–540. Thayer, G.W., D.A. Wolfe, and R.B. Williams. 1975b. The impact of man on seagrass systems. American Scientist 63:288–296. Thom, R.M. 1990. Spatial and temporal patterns in plant standing stock and primary production in a temperate seagrass system. Botanica Marina 33:497–510. Thom, R.M., and R.G. Albright. 1990. Dynamics of benthic vegetation standing- stock, irradiance, and water properties in central Puget Sound. Marine Biology 104:129–141. Vadas, R.L., and B. Beal. 1987. Green algal ropes: A novel estuarine phenomenon in the Gulf of Maine. Estuaries 10:171–176. Vähätalo, A.V., and M. Søndergaard. 2002. Carbon transfer detrital leaves of eelgrass (Zostera marina) to bacteria. Aquatic Botany 73:265–273. Vermaat, J.E., M.G.M. Hootsmans, and P.H. Nienhuis. 1987. Seasonal dynamics and leaf growth of Zostera noltii Hornem., a perennial intertidal seagrass. Aquatic Botany 28:287–299. Westlake, D.F. 1963. Comparisons of plant productivity. Biological Reviews 38:385–425. Wium-Andersen, S., and J. Borum. 1984. Biomass variation and autotrophic production of an epiphyte-macrophyte community in a coastal Danish area. I. Eelgrass (Zostera marina L.) biomass and net production. Ophelia 23:33–46. 224 Northeastern Naturalist Vol. 11, Special Issue 2 Zieman, J.C. 1974. Methods for the study of the growth and production of turtle grass, Thalassia testudinum Konig. Aquaculture 4:139–143. Zieman, J.C., and R.G. Wetzel. 1980. Productivity in seagrasses: Methods and rates. Pp. 87–116, In R.C. Phillips, and C.P. McRoy (Eds.). Handbook of Seagrass Biology: An Ecosystem Perspective. Garland STPM Press, New York, NY. 353 pp.