Seagrass beds are essential components of coastal ecosystems, providing many valuable ecosystem functions and services (Barbier et al. 2011). These include sediment stabilization, particle trapping, shoreline protection, nutrient cycling, food production, and provision of habitat structure and biodiversity (Orth et al. 2006, Barbier et al. 2011). Rapid declines in global seagrass coverage have been observed over the last century, with at least 29% of global coverage lost since the late 1880s (Waycott et al. 2009). Worldwide declines in seagrass coverage have been caused by anthropogenic impacts, such as degradation of water quality, direct physical disturbances, and as a consequence of global climate change (e.g., increased sea surface temperature, sea level, and storm frequency) (Orth et al. 2006, Waycott et al. 2009). Recognition of extensive seagrass loss has led to enhanced interest in monitoring programs worldwide. These programs provide not only direct measures of seagrass bed coverage and plant health, but also detection of long-term environmental change (Orth et al. 2006).
The development of successful monitoring programs for seagrass beds relies on well-resolved baseline information of the ecological and production dynamics of the beds designated for monitoring. Valuable data include measures of seagrass extent (e.g., bed size), plant characteristics (e.g., shoot density, aboveground and belowground biomass, leaf length), production dynamics (e.g., growth per day, rate of shoot loss), reproductive patterns (e.g., number of reproductive plants, flowers, and seeds), integrative metrics of plant health [e.g., leaf area index (LAI), aboveground-to-belowground biomass ratio, nutrient content], and physical factors that influence seagrass growth (e.g., suspended solids, nutrients, water temperature) (Short and Coles 2001). In temperate regions, it is particularly important to develop baseline datasets that capture the seasonal cycles of plant senescence and growth. Comprehensive baseline datasets can be used to identify representative metrics of ecosystem health, the appropriate temporal and spatial scales for monitoring, efficient sampling methods, and the statistical power required to detect change (Neckles et al. 2012).
Zostera marina L. (eelgrass) beds are characteristic of much of the Atlantic Canadian coastline. Despite their prevalence, few studies have examined seasonal Z. marina growth and production in detail for this region (although see Robertson and Mann 1984). Although there is interest in monitoring the status of these Z. marina beds, rigorous biological data to inform such monitoring is generally unavailable. Z. marina beds in Atlantic Canada face unique environmental pressures, such as winter ice cover and invasive green crab (Carcinus maenas) presence, compared with seagrass beds in other temperate regions that are relatively well studied. The objectives of our study were to develop and analyze a comprehensive baseline dataset of the ecological and production dynamics of Z. marina in three adjacent bays on the Atlantic coast of Nova Scotia, Canada. This information will provide a critical baseline for future comparisons of seagrass condition and estimates of variability in potential monitoring parameters. We developed the baseline dataset by (i) quantifying Z. marina shoot density, aboveground and belowground biomass and biomass ratio, leaf length and production, LAI, spadix density, and plastochrone interval (PI), all measured monthly at each site, and (ii) measuring environmental variables at each site that potentially influence Z. marina growth [i.e., fouling algae, water temperature, salinity, water column nutrients and chlorophyll a, total suspended solids (TSS), and sediment particle size]. The dataset was developed over a 1-year period to capture seasonal cycles in growth and productivity. Statistical analyses quantified variability in these quantities with respect to site and seasonal differences.
Materials and methods
The baseline dataset of Zostera marina dynamics was developed for three adjacent sites on the south shore of Nova Scotia, Canada: Port Joli (PJ), Port l’Hebert (PH), and the Kejimkujik National Park Seaside (Keji) (Figure 1A–C). Each site was visited monthly between June 2009 and the end of May 2010. The exceptions were in January 2010 for all sites and February 2010 at Keji, because winter conditions limited site accessibility. Extensive beds of Z. marina were located at the head of each bay (Figure 1C), and typically experienced low water currents and high sediment deposition.
PJ is a relatively open embayment shoaling from the mouth to the head of the bay, while PH is a more semienclosed bay with a small sill at its mouth. At both of these sites, the monotypic Z. marina beds are located on elevated flats irrigated by a complex system of channels. The beds are continuous with little fragmentation, and are extensive in size (approximately 1–2 km2). Water depths during mean tides are 0.20–1.5 m. The beds are sometimes exposed during spring low tides. Relatively small freshwater bogs are located near both sites, causing moderate tannin concentrations in runoff after large rain events. Both PH and PJ experience relatively low anthropogenic impact due to little coastline development. Both sites are designated as migratory bird sanctuaries.
The Keji site is located in the innermost part of a tidal lagoon (Figure 1C and D). Water flow at this location is restricted owing to a hydraulic constriction between the inner and outer lagoon regions (Figure 1D). This constriction is caused by a bridge and causeway constructed in the early 1900s and later modified in the 1960s. Median tidal range is thus reduced at the Z. marina bed by 15 cm relative to the outer lagoon, where the tidal range is ∼50 cm (Dowd and Wong 2013). Water depths during mean tides are 0.35–1.0 m. A flushing box model indicates that water turnover time in the outer lagoon is approximately 1 day, while it is approximately 8 days in the inner lagoon (Dowd and Wong 2013). The Z. marina bed is continuous, although less extensive (0.101 km2), than at PH and PJ. The bed is dominated by Z. marina but some Ruppia maritima L. is also present in the upper reaches. The site is surrounded by extensive freshwater bogs, resulting in high tannin concentration in draining waters, particularly after large rain events (M. Wong pers. obs). Keji is a National Park and thus experiences few anthropogenic impacts; however, the bridge and causeway constriction is likely an important long-term disturbance through its effects on water exchange and sediment dynamics. A census of the bed in 1987 (Brylinsky et al. 1987), combined with annual monitoring that began in 2007, suggests a decline in extent of ∼90% over nearly three decades (Ure et al. 2010). The high abundance of invasive green crabs (Carcinus maenas) at Keji may be an important biotic disturbance for the Z. marina (Mersey Tobeatic Research Institute and Parks Canada 2013), although green crabs are also present in similar abundance at PJ and PH where no decline in Z. marina has been observed.
At each site, sampling was conducted within an approximately 1.2×104 m2 area that had negligible elevation changes. Before each field trip, sampling stations were haphazardly chosen using geo-referenced digital maps. Ten different stations were sampled each month, and were located in the field using a hand-held GPS unit (Garmin 60CSx, Garmin International, Inc., Olathe, KS, USA). Sampling was conducted at low tide using a canoe and by snorkeling. At each station, the percent cover of algae was determined visually using a 0.5×0.5 m quadrat. Shoot density of Zostera marina at each station was determined by counting shoots in a 0.25×0.25 m quadrat. It was not possible to estimate the cover of algae or the density of Z. marina shoots using quadrats in some winter months (i.e., December 2009 at PH, January 2010 at all sites, February 2010 at PH and Keji) due to reduced visibility from the high concentration of suspended solids or ice cover. Shoot density for these months was estimated from core samples (see below), while the percent cover of algae could not be determined.
Core samples were collected to obtain estimates of aboveground and belowground Z. marina biomass. At PJ and PH, core samples were collected at six of the 10 stations each month. At Keji, core samples were collected every second month to reduce the adverse impact that might contribute to further decline of this seagrass bed. Core samples were taken using a 0.15-m-diameter hand core that was pushed 12 cm into the sediment and extracted. Samples were placed into plastic bags and refrigerated for 2–4 days until processing. In the laboratory, plant components were separated from the sediment by gently washing the sample with seawater over a 500-μm sieve. The number of Z. marina shoots per core was counted. The relationship between shoot density from core data (SDc) and shoot density from quadrat data (SDq) was determined using linear regression, and was SDq=263.5+0.687SDc (F=39.231,91, p<0.0001, R2=0.40). This relationship was used to determine shoot density from core samples for the winter months (as described above), although we acknowledge the potential high variability in estimates. Plant material in the core samples was separated into aboveground and belowground components, dried at 60°C for 48 h, and weighed to determine dry mass. Ruppia maritima found in the cores from Keji was dried and weighed without separating into aboveground and belowground components. When cores were not taken at Keji, all leaves in the 0.25×0.25 m quadrat used for shoot counts were measured, and aboveground biomass was determined using a leaf length to leaf biomass relationship. This relationship was determined by collecting 37 leaves ranging from 4.9 to 56.8 cm long from Keji in July 2009, measuring each leaf and drying at 60°C for 48 h, and weighing each leaf to determine dry mass. A linear regression on the log10-transformed length and biomass data described the relationship between leaf length (cm) and leaf biomass (dry g) as M=1.65(L)-3.97, where M is the dry mass and L is the leaf length (F=634.21,35, p<0.0001, R2=0.95).
Seagrass growth and production
The numbers of spadices on the generative shoots found in each core sample were counted to determine the density of spadices at each site. The lengths of five fully mature third leaves in each core sample were measured at each sampling date if enough leaves were available. LAI was calculated as leaf length×leaf width×number of leaves per shoot×shoot density.
Production of the third leaf in the summer was determined using conventional methods (Dennison 1990). In June, July, August, and September 2009, 30 mature plants at each site were marked in situ by making a small hole at the top of the sheath using a large sewing needle. Marked plants were located approximately 1–2 m apart to avoid marking plants on the same rhizome. Individually marked plants were tagged using plastic cable ties flagged with trail tape, relocated after 1 month, and harvested. Harvested marked plants were transported to the laboratory and refrigerated in the dark until processing. To determine leaf production, plants were rinsed in fresh water and the leaves of each shoot were separated at the meristem. The leaf tissue between the original hole on the sheath and the hole on the third leaf was dried at 60°C for 48 h and then weighed to determine the dry mass (mg) of third leaf growth per day. The PI, the time between the establishment of two successive leaves on a plant, was determined by dividing the growth period by the number of new leaves produced (Gaeckle and Short 2002).
Three replicate water samples were collected at each sampling date and site to measure chlorophyll a concentration, dissolved inorganic nitrogen (DIN) concentration, and TSS. Chlorophyll a concentration in the water column was determined by filtering a known volume of water onto a glass-fiber filter, extracting chlorophyll a overnight in the dark using 90% acetone, and determining the concentration fluorometrically using the acidification technique (Holm-Hansen et al. 1965). TSS in the water column was determined by filtering a known volume of water onto a preweighed glass-fiber filter, drying at 60°C for 24 h, and weighing. DIN was determined using a Segmented Flow Technicon Autoanalyzer and the methods of Armstrong et al. (1967) and Kerouel and Aminot (1997). We focused on DIN as our Zostera marina beds are in non-carbonate sediments, which are usually nitrogen limited, whereas phosphorus limitation usually occurs in biogenic carbonate sediments characteristic of tropical waters (Short 1987, Lee et al. 2007).
Six replicate sediment samples (3 cm diameter×5 cm deep) were collected in August 2009 for analysis of particle size. Samples were frozen until processing. To determine the percent silt content, 25 g of each sediment sample was dried at 60°C for 48 h and weighed. Each sample was mechanically stirred for 15 min with 2 g of sodium hexametaphosphate to disaggregate particles, left overnight, and then stirred for 15 min the following morning. The sample was then wet-sieved on a 64-μm sieve. The sand remaining on top of the sieve was dried at 60°C for 48 h and weighed. The percent silt content of surface sediments was determined as (total dry mass-sand mass)/(total dry mass)×100.
Salinity at each sampling date was measured at each site using a handheld recorder (YSI Inc., Yellow Springs, Ohio, USA). Continuous water temperature records were measured using temperature loggers (TidbiT v2; Onset Computer Corporation, Bourne, MA, USA) from June 2009 to May 2010 at PH and Keji. The logger at PJ was lost during the winter, so data are not available during this period.
Analyses of variance (ANOVAs) with month and site as the independent factors were used to examine patterns in Zostera marina parameters and most physical parameters. Percent silt in sediments was analyzed across sites with a one-way ANOVA. Significant main effects or interactions were examined using Tukey’s test. Residual plots were used to determine if the underlying assumptions of homogeneity of variance and normality were violated. Violations were corrected by weighting the ANOVAs by the inverse of the square root of the replicate variance and/or by transforming data using log(x+1) (Draper and Smith 1998). All ANOVAs were done using R v.2.8.1 statistical software.
Time series analyses of the temperature records were conducted. The goal was to characterize the temperature variability at each site on three different time scales: seasonal, subtidal, and tidal. To determine the variations at the annual period (the seasonal cycle), monthly means of temperatures were computed for each of the sites (note that the complete annual cycle was not available at PJ). The subtidal temperature variability (i.e., the part of the temperature record having a longer period than tidal variability) for the summer period was isolated by low-pass filtering the temperature records at each site to remove the higher-frequency tidal variations (Priestley 2004). The frequency cutoff for the low-pass filter corresponded to a period of 2 days. For the study region, these subtidal temperature variations correspond largely to upwelling events associated with wind-driven circulation in adjacent shelf waters (Platt et al. 1972). Finally, the tidal variation in temperature was determined by subtracting the subtidal temperature series from the original summer temperature series. This procedure isolates the summer temperature fluctuations associated with the diurnal and semidiurnal tides independent of any longer period fluctuations.
The continuous temperature data were also used to calculate the total number of hours for which water temperatures were >23°C. This is the mean optimal temperature for photosynthesis for Z. marina, beyond which the net photosynthetic rate slows (Lee et al. 2007).
Zostera marina shoot density showed strong seasonal patterns at PJ and PH, with highest shoot densities between July and September (site×month: F19,256=28.0, p<0.0001; Tukey’s test, p<0.05; Figure 2). Shoot density at Keji was consistently lower than at PH and PJ across all months (Tukey’s test, p<0.05). Mean shoot density throughout the year ranged from 4 to 280 and 221 to 995 shoots m-2 at Keji and PH, respectively. Mean shoot density in all months was generally highest at PJ, ranging from 460 to 1046 shoots m-2. Ruppia maritima was observed at Keji in the spring and summer, and shoot density was 88.8±7.1 shoots m-2 (mean±SD, n=69).
Filamentous algal mats were observed covering Z. marina plants, the sea bottom, and floating on the water surface at Keji, and on the sea bottom at PH, in the spring and summer months. Algal mats at Keji consisted of Ulva linza, U. intestinalis, Pilayella littoralis, Ectocarpus siliculosus, Chaetomorpha linum, and an unidentified cyanobacterium. At PH, algal mats were composed of C. linum only. Algal cover at Keji was highest in May, when it was ∼80% of the sea bottom and Z. marina plants. At PH, algal cover was highest in April to June, with 15–40% cover. At both PH and Keji, percent cover of algae declined to ∼5% by September.
Seasonal patterns in aboveground biomass of Z. marina were evident at all sites (site×month: F19,156=20.6, p<0.0001; Tukey’s test, p<0.05; Figure 3A–C). Aboveground biomass at Keji was highest from June to August, while at PJ and PH it was highest between June and September. When compared across sites within each month, aboveground biomass at Keji was consistently lower than at PH and PJ.
Strong seasonal patterns in belowground biomass within sites were not evident, although belowground biomass was significantly affected by the interaction between site and month (F15,135=4.51, p<0.0001; Figure 3D–F). When compared across sites within each month, belowground biomass was usually lowest at Keji (in months when cores were taken at Keji; Tukey’s test, p<0.05). Belowground biomass was similar throughout the year at PH and PJ.
The aboveground-to-belowground biomass ratios reflected the previously described patterns in aboveground biomass, and were significantly affected by the interaction between site and month (site×month: F15,135=3.52, p<0.0001; Tukey’s test, p<0.05; Figure 3G–I). At all sites, the ratio ranged from close to zero in the winter months, when shoots were sparse or absent, to a maximum of 2.0 in the summer months. Seasonal patterns in the ratio were strongest at PH and PJ. When compared across sites within each month, the ratio at Keji was consistently lower than at PH and PJ, which did not differ from each other (Tukey’s test, p<0.05).
Seasonal patterns in leaf length (third leaf) were evident at all sites (site×month: F19,776=26.2, p<0.0001; Tukey’s test, p<0.05; Figure 4A–C). At Keji, the leaf length was longest in April to July, earlier than that observed at PJ and PH, where leaf lengths were longest between July and October (Tukey’s test, p<0.05).
Seasonal trends in LAI were evident at PH and PJ (site×month: F19,766=10.38, p<0.0001; Tukey’s test, p<0.05; Figure 4D–F), where it was highest in June to August. When compared across sites within each month, LAI was consistently highest at PH and PJ in June to November (Tukey’s test, p<0.05).
Productivity of the third leaf determined from marked plants was significantly different across sites and summer months (site×month: F6,210=3.75, p=0.001), and was highest in both July and August (Tukey’s test, p<0.05; Figure 5A–C). Productivity of the third leaf ranged from 0.58 to 3.14 dry mg day-1 across all sites and months. When compared across sites within months, productivity was highest in Keji in July and August relative to PH and PJ (Tukey’s test, p<0.05).
The PI at all sites ranged from 4 to 18 days across the main growing season (Figure 5D–F). Different seasonal patterns in PI were observed at each site. Spadices were observed in June, July, and August at all three sites, with density being highest in June (Figure 5G–I).
Several parameters characterizing properties of the water column and sediments were also measured. While chlorophyll a concentration did not show strong seasonal patterns at any site (site×month: F18,278=60.7, p<0.0001; Tukey’s test, p>0.05), it tended to be higher at PH and PJ than at Keji throughout the year (Figure 6A–C). Chlorophyll a values ranged from 1.12 to 15 μg l-1. The TSS was significantly affected by the interaction between site and month (F18,66=16.4, p<0.0001; Figure 6D–F), mainly driven by low values in June at all sites, and evidence of seasonal trends at Keji and PH (Tukey’s test, p<0.05). TSS measurements were relatively high, and were approximately 50 mg l-1 on average at all sites. DIN concentrations from the water column ranged from 0.5 to 4.5 μm at all sites, and followed seasonal patterns typical of open ocean values in the region, where DIN concentrations were highest in the winter months (Figure 6G–I).
Ice cover in the winter months and algal mats in the spring and summer were observed at Keji and PH, while PJ remained ice and algae free (Table 1). Salinity ranged from 20 to 30 ppt across the field sites. The percent silt content in sediments did not differ across sites, and was 52.0±4.2 (mean±SD, n=18).
The seasonal variation in water temperatures showed an annual range in the monthly means from 0°C to 20°C across all three field sites (Figure 7A). From June to October, Keji was slightly warmer than PH and PJ by 1–4°C. Continuous water temperature records were further examined for June to September 2009, the main growing season for Z. marina (Figure 7B). The subtidal variability was coherent between sites, and reflects offshore temperature variations (Figure 7B). PJ and PH exhibited similar temperature variability, although PH tended to be slightly lower in temperature by approximately 1°C. Keji showed a similar pattern with the other sites but had a larger temperature range. The number of hours in which water temperature exceeded 23°C was 348.7, 143.7, and 187.8 h at Keji, PH, and PJ, respectively. Observations of detailed tidal variations across a 10-day period (Figure 7C) indicated that Keji shows a strong asymmetrical diurnal signal, a result of solar heating, with a small semidiurnal tidal influence. The temperature signal at PJ and PH is much more complex, with a daily signal dominated by spikes of high and low temperature, likely due to the phasing of tides and solar heating cycles in these very shallow areas.
Our study provides one of the first comprehensive datasets describing the annual growth and production dynamics of Zostera marina in Atlantic Canada. Similar to other studies of northern temperate Z. marina (e.g., Verhagen and Nienhuis 1983, Robertson and Mann 1984, Olesen and Sand-Jensen 1994, Boström et al. 2004), our study showed clear seasonal patterns in most measured Z. marina parameters at all three sites. Seasonal patterns were most pronounced for aboveground plant components (shoot density, biomass, and leaf length). Only slight differences in seasonal patterns and magnitudes were observed among sites. One such difference was that shoot density in the winter at PH was lower than at PJ, despite it being similar during the summer. This may have resulted from differences in winter conditions. Ice cover was prevalent at PH and Keji throughout the winter, which would have reduced light availability for Z. marina compared with ice-free PJ (Robertson and Mann 1984). Ice cover may have also removed whole plants or only leaves, as has been previously observed in other Nova Scotia seagrass beds (Robertson and Mann 1984).
In general, the greatest differences in measured Z. marina parameters were observed at Keji compared with PH and PJ. Shoot density and aboveground biomass of Z. marina were substantially lower at Keji throughout the year than at PH and PJ, sometimes by an order of magnitude. The restricted water exchange at Keji, caused by the bridge and causeway constriction, may have contributed to the low shoot density of Z. marina at this site. The constriction caused slower water turnover and reduced flushing in the inner lagoon compared with the outer lagoon (Dowd and Wong 2013), resulting in higher water temperatures relative to PH and PJ. Most important, the number of hours that water temperatures exceeded 23°C, the mean optimal temperature for photosynthesis (Lee et al. 2007), was substantially higher at Keji than at PH and PJ. These higher water temperatures may have caused a rapid increase in the rate of respiration relative to photosynthesis (Marsh et al. 1986, Staehr and Borum 2011), resulting in a lowered production to respiration ratio (P/R ratio) and subsequently reduced net production and growth (Marsh et al. 1986). Higher water temperatures may have also been detrimental for Z. marina by causing increased periods of sediment and water column hypoxia or anoxia, resulting in accumulation of toxic metabolites in plant tissues (Crawford and Braendle 1996) and exposure to high sulfide levels (Goodman et al. 1995, Pulido and Borum 2010). Also, temperature stress may have changed reproductive output and strategies (De Cock 1981, Jarvis et al. 2012). Thus, Z. marina growth and production at Keji may have been at least partly limited by seasonally high water temperatures in the summer months through a variety of mechanisms. Temperature-dependent growth limitation has been observed in other studies of Z. marina, particularly those at lower latitudes where summer temperatures exceed the optimal temperatures for growth and photosynthesis (e.g., Lee et al. 2005, Moore and Jarvis 2008, Moore et al. 2012). In addition to increased water temperature, other aspects resulting from the restricted water exchange, such as nutrient delivery, turbidity, and sediment properties, may have also been important. In general, the physical regime at Keji was very different compared with PJ and PH, and likely contributed to the observed differences in Z. marina dynamics.
The high water temperatures and low flushing rates at Keji also provided ideal growing conditions for fouling algal mats, and may have contributed to the loss of Z. marina from this site. Filamentous brown and green algal mats were observed fouling Z. marina plants, the sea bottom, and the water surface in the spring and summer. Many studies have shown that fouling of Z. marina by filamentous algal mats reduces light availability, with negative consequences for growth and production (e.g., Hauxwell et al. 2003). Although algal mats were also observed at PH, the mats were a different species, and tended to cover only the sea bottom. These characteristics meant that algal mats at PH had less effect on Z. marina growth and production than the mats at Keji.
Reduced shoot density and aboveground biomass at Keji relative to PH and PJ may also have resulted from light limitation caused by high concentration of suspended solids and colored dissolved organic matter (CDOM) in the water column. Our measurements of TSS concentrations were consistent throughout most of the year at all of our field sites, and were typical for shallow, low current areas where sediment resuspension by tides and winds enhance TSS concentrations in the innermost parts of bays (Dowd 2003). Low flushing rates at Keji may have resulted in high TSS concentrations for longer periods relative to PH and PJ, limiting light availability to the Z. marina bed. Light limitation at Keji may have been further compounded by high CDOM from the surrounding bogs; although we were unable to obtain direct CDOM measurements, higher tannin concentrations were often observed at Keji than at PJ and PH, particularly after heavy rain events. Clearly, the role of TSS concentration and CDOM in light limitation of Z. marina growth at our field sites requires further investigation. A full understanding would be achieved by obtaining more highly resolved spatial and temporal measurements of TSS concentrations, direct measurements of CDOM, and data on light attenuation through the water column at varying tidal phases and times of the year.
Our comprehensive baseline data of Z. marina growth and production in Atlantic Canada can be used as preliminary data to inform certain aspects of monitoring program development. Our data are best suited to provide insight into program objectives related to measurements of plant parameters and site selection. Our results show that the sites we sampled are not identical replicates, despite their geographical proximity. PJ and PH should not be used as reference sites against which Keji is evaluated, due to different water exchange properties among sites. Measurement techniques of plant parameters should be adjusted to be less destructive at Keji, which appears sensitive to disturbance. We found that seasonal sampling of plant parameters is important, and suggest inclusion of multiple sampling times within each season to allow meaningful assessment of annual (seasonal) variability. Similar to many other studies, we found that measurements of physical properties are invaluable to understanding the dynamics of Z. marina beds (Koch 2001). A particularly valuable aspect of our data is that they provide essential baseline measurements of Z. marina condition that can be used for future comparisons. These measurements include estimates of variability for potential monitoring parameters, information that can be used to determine the appropriate spatial and temporal scales for sampling. Our study suggests that future work to inform development of a monitoring program focused on assessing status of plant parameters should include data of annual variability for multiple years, light limitation, and nutrient status of plants and in sediment pore waters. Combined with these additional data, our study will potentially provide valuable insight into the development of certain aspects of Z. marina monitoring programs in Atlantic Canada.
We thank L. Hartman, M. Kenny, and A. MacKenzie for field and laboratory assistance. C. Anstey processed the nutrient samples, and J. Anning and colleagues assisted with the chlorophyll a samples. Comments from H. Vandermeulen and A. Locke helped improve the initial manuscript. We are particularly grateful to two anonymous reviewers whose comments and suggestions resulted in a stronger and more comprehensive manuscript. Funding was provided by Fisheries and Oceans Canada. We also thank C. McCarthy and Parks Canada for in-kind support of sampling at Kejimkujik Seaside.
Armstrong, F.A.J., C.R. Sterns and J.D.H. Strickland. 1967. The measurement of upwelling and subsequent biological processes by means of the Technicon Autoanalyzer and associated equipment. Deep-Sea Res. 14: 381–389.Google Scholar
Boström, C., C. Roos and O. Rönnberg. 2004. Shoot morphometry and production dynamics of eelgrass in the northern Baltic Sea. Aquat. Bot. 79: 145–161.Google Scholar
Brylinsky, M., K.P. Crawford and G.R. Daborn. 1987. Marine resource inventory of the Seaside Adjunct, Kejimkujik National Park. Acadia Centre for Estuarine Research, Acadia University, Wolfville, NS. pp. 133.Google Scholar
Crawford, R.M.M. and R. Braendle 1996. Oxygen deprivation stress in a changing environment. J. Exp. Biol. 47: 145–159.Google Scholar
De Cock, A.W.A.M. 1981. Influence of temperature and variations in temperature on flowering in Zostera marina L. under laboratory conditions. Aquat. Bot. 10: 125–131.Google Scholar
Dennison, W.C. 1990. Leaf production. In: (R.C. Phillips and C.P. McRoy, eds.) Seagrass research methods. UNESCO, Paris. pp. 77–79.Google Scholar
Dowd, M. 2003. Seston dynamics in a tidal inlet with shellfish aquaculture: a model study using tracer equations. Est. Coast. Shelf Sci. 57: 523–537.Google Scholar
Dowd, M. and M.C. Wong. 2013. Tidal analysis in Little Port Joli Basin, Kejimkujik Seaside, NS: before and after bridge/causeway removal. Report for Parks Canada. pp. 17.Google Scholar
Draper, N.R. and H. Smith. 1998. Applied regression analysis. John Wiley and Sons, New York. pp. 706.Google Scholar
Gaeckle J.L. and F.T. Short. 2002. A plastochrone method for measuring leaf growth in eelgrass, Zostera marina L. Bull. Mar. Sci. 71: 1237–1246.Google Scholar
Goodman, J.L., K.A. Moore and W.C. Dennison. 1995. Photosynthetic responses of eelgrass (Zostera marina L.) to light and sediment sulfide in a shallow barrier island lagoon. Aquat. Bot. 50: 37–47.Google Scholar
Hauxwell, J., J. Cebrián 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. Mar. Ecol. Prog. Ser. 247: 59–73.Google Scholar
Holm-Hansen, O., C.J. Lorenzen, R.W. Holmes and J.D.H. Strickland. 1965. Fluorometric determination of chlorophyll. J. Cons. Int. Explor. Mer. 30: 3–15.Google Scholar
Jarvis, J.C., K.A. Moore and W.J. Kenworthy. 2012. Characterization and ecological implication of eelgrass life history strategies near the species’ southern limit in the western North Atlantic. Mar. Ecol. Prog. Ser. 444: 43–56.Web of ScienceGoogle Scholar
Kerouel, R. and A. Aminot. 1997. Fluorometric determination of ammonia in sea and estuarine waters by direct segmented flow analysis. Mar. Chem. 57: 265–275.Google Scholar
Lee, K.-S., S.R. Park and J.-B. Kim. 2005. Production dynamics of the eelgrass, Zostera marina in two bay systems on the south coast of the Korean peninsula. Mar. Biol. 147: 1091–1108.Google Scholar
Lee, K.-S., S.R. Park and Y.K. Kim. 2007. Effects of irradiance, temperature, and nutrients on growth dynamics of seagrasses: a review. J. Exp. Mar. Biol. Ecol. 350: 144–175.Web of ScienceGoogle Scholar
Marsh, J.A., W.C. Dennison and R. S. Alberte. 1986. Effects of temperature on photosynthesis and respiration in eelgrass (Zostera marina L.). J. Exp. Mar. Biol. Ecol. 101: 257–267.Google Scholar
Mersey Tobeatic Research Institute and Parks Canada. 2013. Annual report of research and monitoring in the Greater Kejimkujik Ecosystem 2012. Kempt, Nova Scotia. pp. 110.Google Scholar
Moore, K.A. and J.C. Jarvis. 2008. Environmental factors affecting recent summertime eelgrass diebacks in the Lower Chesapeake Bay: implications for long-term persistence. J. Coast. Res. 55: 135–147.Web of ScienceGoogle Scholar
Moore, K.A., E.C. Shields, D.B. Parrish and R.J. Orth. 2012. Eelgrass survival in two contrasting systems: role of turbidity and summer water temperatures. Mar. Ecol. Prog. Ser. 448: 247–258.Web of ScienceGoogle Scholar
Olesen, B. and K. Sand-Jensen. 1994. Demography of shallow eelgrass (Zostera marina) populations – shoot dynamics and biomass development. J. Ecol. 82: 379–390.Google Scholar
Orth, R.J., T.J.B. Carruthers, W.C. Dennison, C.M. Duarte, J.W. Fourqurean, K.L. Heck, A.R. Hughes, G.A. Kendrick, W.J. Kenworthy, S. Olyarnik, F.T. Short, M. Waycott and S.L. Williams. 2006. A global crisis for seagrass ecosystems. BioScience 56: 987–996.CrossrefGoogle Scholar
Platt, T., A. Prakash and B. Irwin. 1972. Phytoplankton, nutrients and flushing of inlets on the coast of Nova Scotia. Nat Can 99: 253–261.Google Scholar
Priestley, M.B. 2004. Spectral analysis and time series. Academic Press, London. pp. 877.Google Scholar
Pulido, C. and J. Borum. 2010. Eelgrass (Zostera marina) tolerance to anoxia. J. Exp. Mar. Biol. Ecol. 385: 8–13.Google Scholar
Robertson, A.I. and K.H. Mann. 1984. Disturbance by ice and life-history adaptations of the seagrass Zostera marina. Mar. Biol. 80: 131–141.Google Scholar
Short, F.T. 1987. Effects of sediment nutrients on seagrasses: literature review and mesocosm experiment. Aquat. Bot. 27: 41–57.Google Scholar
Short, F.T. and R.G. Coles. 2001. Global seagrass research methods. Elsevier Science, Amsterdam, Netherlands. pp. 473.Google Scholar
Ure, D., S. Chisholm and D. Kehler. 2010. Kejimkujik National Park and National Historic Site Technical Compendium to the 2010 State of the Park Report. Parks Canada, Maitland Bridge, Nova Scotia. pp. 224.Google Scholar
Verhagen, J.H.G. and P.H. Nienhuis. 1983. A simulation model of production, seasonal changes in biomass and distribution of eelgrass (Zostera marina) in Lake Grevelingen. Mar. Ecol. Prog. Ser. 10: 187–195.Google Scholar
Waycott, M., C.M. Duarte, T.J.B. Carruthers, R.J. Orth, W.C. Dennison, S. Olyarnik, A., Calladine, J.W. Fourqurean, K.L. Heck, A.R. Hughes, G.A. Kendrick, W.J. Kenworthy, F.T. Short and S.L. Williams. 2009. Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proc. Nat. Acad. Sci. 106: 12377–12381.Google Scholar
About the article
Published Online: 2013-10-29
Published in Print: 2013-12-01