The Roles of Competition and Disturbance in a Marine Invasion

Two hypotheses for the decline of native species are the superior exploitation of disturbance by exotic species and the competitive displacement of native species by their exotic counterparts. Theory predicts that functional similarity will increase the intensity of competition between native and invasive species. Ecologically important “foundation” species, Zostera marina and other seagrasses have globally declined during the past century. This study used transplant and vegetation removal experiments to test the hypotheses that disturbance and competitive interactions with an invasive congener (Z. japonica) are contributing to the decline of native Z. marina in the northeastern Pacific. Interspecific competition reduced Z. marina and Z. japonica above-ground biomass by 44 and 96%, respectively, relative to intraspecific competition. Disturbance substantially enhanced Z. japonica productivity and fitness, and concomitantly decreased Z. marina performance, effects that persisted two years following substratum disturbance. These results demonstrate that disturbance and competitive interactions with Z. japonica reduce Z. marina performance, and suggest that Z. japonica’s success as an invasive species stems dually from its ability to persist in competition with Z. marina and its positive response to disturbance. These results highlight the importance of understanding the interconnected roles of species interactions and disturbance in the decline of seagrass habitats, and provide a rationale for amending conservation policy in Washington State. In the interest of conserving native eelgrass populations, the current policy of protecting both native and invasive Zostera spp. should be refined to differentiate between native and invader, and to rescind the protection of invasive eelgrass.     

Bacterial abundance and aerobic microbial activity across natural and oyster aquaculture habitats during summer conditions in a northeastern Pacific estuary

We measured sediment properties and the abundance and aerobic metabolism of microbes in Willapa Bay, Washington, USA, to test the response of sediment microbes to oyster aquaculture. Sites spanned the estuary gradient (practical salinity units ranged from 24 to 30 under seasonally low river flows) and six different low-intertidal habitat types: eelgrass (Zostera marina), unstructured tideflat, oyster hummocks (reefs of Crassostrea gigas), longline oyster aquaculture, hand-picked on-bottom oyster aquaculture, and dredged on-bottom oyster aquaculture. Aerobic metabolism was assessed by sole-source carbon use (SSCU) of 31 carbon sources on Biolog plates. Sediments generally became siltier and more organically enriched into the estuary, but no consistent differences in sediment properties occurred across habitat types. Bacterial cell density tracked organic content. Across the estuary gradient, overall aerobic SSCU increased less steeply than bacterial cell density, possibly as anaerobic metabolism became more important. Across habitats, aerobic SSCU differed significantly in both overall metabolism and diversity of carbon sources. Aerobic metabolism was generally lower for sediment microbes from intertidal on-bottom oyster aquaculture than from eelgrass. Humans indirectly alter microbial activity through biogenic habitats created during aquaculture, but, as has been shown for bivalves more generally, these changes were relatively small, particularly in comparison to sediment changes along estuarine gradients.     

Effects of salinity on photosynthesis and respiration of the seagrass Zostera japonica: A comparison of two established populations in North America

Photosynthetic responses were quantified for two Zostera japonica Aschers. and Graebn. populations from the northern and southern limits of distribution exposed to a range of salinities along the Pacific Coast of North America. Plants were collected from Padilla Bay, Washington (northern) and Coos Bay, Oregon, USA (southern) and cultured together in experimental tanks at 3 salinities (5, 20 and 35) under saturating irradiance for 3 weeks. Subsequently, photosynthesis–irradiance (P vs. E curves) relationships for leaf segments from the two populations were assessed using an oxygen electrode system. We found no evidence for diel rhythms in either light saturated photosynthesis (P_max) or dark respiration (R_d). For the Padilla Bay population, P_max ranged from 192 to 390 μmol O₂ g DW⁻¹ h⁻¹; for the Coos Bay population P_max ranged from 226 to 774 μmol O₂ g DW⁻¹ h⁻¹. Photosynthetic maxima of the Coos Bay plants occurred at a salinity of 20, whereas salinity had no effect on the photosynthetic maxima of the Padilla Bay plants. There were significant differences in leaf tissue R_d among salinity treatments but the two populations responded similarly to salinity. North American populations of Z. japonica are best adapted to intermediate salinities, displaying minimum R_d rates, lower compensation irradiance, higher saturation irradiance, and greater P_max rates at a salinity of 20. Additionally, the southern population may be better adapted to southward expansion along the Pacific Coast and changes associated with global climate change.     

Relationships between bed age,bed size,and genetic structure in Chesapeake Bay (Virginia,USA) eelgrass (<Emphasis Type="Italic">Zostera marina</Emphasis> L.)

Genetic structure and diversity can reveal the demographic and selective forces to which populations have been exposed, elucidate genetic connections among populations, and inform conservation strategies. Beds of the clonal marine angiosperm Zostera marinaL. (eelgrass) in Chesapeake Bay (Virginia, USA) display significant morphological and genetic variation; abundance has fluctuated widely in recent decades, and eelgrass conservation is a major concern, raising questions about how genetic diversity is distributed and structured within this metapopulation. This study examined the influence of bed age (<65years versus<6years) and size (>100ha versus<10ha) on morphological and genetic (allozyme) structure and diversity within Chesapeake Bay eelgrass beds. Although both morphology and genetic diversity varied significantly among individual beds (F _ST=0.198), neither varied consistently with bed age or size. The Chesapeake eelgrass beds studied were significantly inbred (mean F _IS=0.680 over all beds), with inbreeding in old, small beds significantly lower than in other bed types. Genetic and geographic distances within and among beds were uncorrelated, providing no clear evidence of isolation by distance at the scale of 10's of km. These results suggest that local environmental conditions have a greater influence on plant morphology than do bed age or size. They support the hypotheses that eelgrass beds are established by multiple founder genotypes but experience little gene flow thereafter, and that beds are maintained with little loss of genetic diversity for up to 65 years. Since phenotypic and genotypic variation is partitioned among beds of multiple ages and sizes, eelgrass conservation efforts should maximize preservation of diversity by minimizing losses of all beds.     

Congener comparison of native (Zostera marina) and introduced (Z. japonica) eelgrass at multiple scales within a Pacific Northwest estuary

A congener comparison of native (Zostera marina) and introduced (Zostera japonica) eelgrasses was conducted in Willapa Bay, Washington, USA. Along intertidal transects, Z. japonica (0.1–1.5 m above mean lower low water [MLLW]) occurred above Z. marina (<0.6 m MLLW). Both species declined in shoot length at higher elevation, but Z. japonica was always shorter (20 vs. 100 cm) and occurred at higher shoot density (>3,800 vs. <360 m^?2 in Z. marina). Z. japonica exhibited greater seasonal variation in biomass, with increases supported by both sustained asexual reproduction (rhizome branching) and recruitment from seeds (30 vs. <5% in Z. marina). Z. japonica’s successful invasion appears related to small size and high reproductive output, allowing it to spread quickly in a variable and stressful intertidal environment where competitive effects are low. Based on interannual changes in abundance, the native eelgrass has also recently increased in Willapa Bay, and one hypothesis involves “engineering” of suitable habitat at higher tidal elevations by Z. japonica.     

Variable responses of native eelgrass Zostera marina to a non-indigenous bivalve Musculista senhousia

The transport and establishment of non-indigenous species in coastal marine environments are increasing worldwide, yet few studies have experimentally addressed the interactions between potentially dominant non-native species and native organisms. We studied the effects of the introduced mussel Musculista senhousia on leaf and rhizome growth and shoot density of eelgrass Zostera marina in San Diego Bay, California. We added M. senhousia over a natural range in biomass (0–1200 g dry mass/m²) to eelgrass in transplanted and established beds. The effects of the non-indigenous mussel varied from facilitation to interference depending on time, the abundance of M. senhousia, and the response variable considered. Consistent results were that mussel additions linearly inhibited eelgrass rhizome elongation rates. With 800 g dry mass/m² of M. senhousia, eelgrass rhizomes grew 40% less than controls in two eelgrass transplantations and in one established eelgrass bed. These results indicate that M. senhousia, could both impair the success of transplantations of eelgrass, which spread vegetatively by rhizomes, and the spread of established Z. marina beds to areas inhabited by M. senhousia. Although effects on leaf growth were not always significant, in August in both eelgrass transplantations and established meadows leaf growth was fertilized by mussels, and showed a saturation-type relationship to sediment ammonium concentrations. Ammonium concentrations and sediment organic content were linear functions of mussel biomass. We found only small, non-consistent effects of M. senhousia on shoot density of eelgrass over 6-month periods. In established eelgrass beds, but not in transplanted eelgrass patches (≈0.8 m in diameter), added mussels suffered large declines. Hence, eelgrass is likely to be affected by M. senhousia primarily where Z. marina beds are patchy and sparse. Our study has management and conservation implications for eelgrass because many beds are already seriously degraded and limited in southern California where the mussel is very abundant. Received: 31 May 1997 / Accepted: 4 September 1997     

Production Ecology of the Non-indigenous Seagrass, Dwarf Eelgrass (Zostera japonica Ascher. & Graeb.), in a Pacific Northwest Estuary, USA

The non-indigenous seagrass Zostera japonica Ascher. & Graeb. (dwarf eelgrass) was first identified in central Oregon (USA) estuaries about 30 years ago. The autecology of this species is poorly described at the southern end of its non-native range although several process oriented studies have been conducted. I examined the production ecology of Z. japonica in the Yaquina Bay estuary. Strong seasonal patterns in light and temperature appeared to control the seasonal variations in biomass and growth. Above- and below-ground biomass ranged between 40–100 and 70–170 gdw m⁻² respectively and seasonal changes in the root:shoot ratio were controlled by above-ground biomass dynamics. Shoot density ranged between 4000 and 11 000 shts m⁻². Areal leaf growth ranged between 0.1 and 1.7 gdw m⁻² d⁻¹ and annual production was about 314 ± 60 gdw m⁻² y⁻¹ (mean ± SD). Nutrients were not limiting in this system as a result of coastal upwelling and watershed inputs. The Z. japonica population studied in Oregon exhibited different patterns of persistence, phenology and flowering intensity relative to other populations along its native and non-native range. These differences suggest that management policies developed for one site may not be appropriate for other sites. The data presented here greatly expands our knowledge base on Z. japonica and provides insight to the processes controlling the dynamics and spread of this non-indigenous seagrass. An erratum to this article is available at .     

Does the recruitment of a non‐native mussel in native eelgrass habitat explain their disjunct adult distributions?

We studied variability in the abundance of small individuals of an invasive mussel (Musculista senhousia) across the depth distribution of a native marine angiosperm, eelgrass (Zostera marina). Adult mussels and eelgrass have a disjunct local distribution, each limiting the other in complex ways. To assess whether eelgrass also influenced the distribution of juvenile mussels, we sampled inside and outside eelgrass beds in one site in Mission Bay and two in San Diego Bay, California, USA. We sampled mussels in size classes 0.26–0.50 mm, 0.51–1.00 mm, 1.10–2.00 mm and > 2.00 mm from September 1997 to April 1999. We also monitored gonad development in larger mussels and in situ growth of mussels ≤ 2 mm tagged with the chemical marker calcein. Spatial and temporal variations in mussel abundances were high but seasonal patterns were roughly similar at San Diego Bay sites; very few mussels were found in Mission Bay. Mussels with full gonads were found year‐round in San Diego Bay, as were mussels in the smallest size class (with a large peak in fall and a smaller secondary one in spring), suggesting that many of the smallest mussels represent recruitment. The observation that most, although not all, tagged mussels increased in size provides further support for recruitment. Some of the highest numbers of mussels in the smallest size class were found inside eelgrass beds, indicating that eelgrass does not restrict and may actually enhance the distribution of very small mussels. The disjunct distribution of adult mussels and eelgrass thus is apparently established primarily postrecruitment. M. senhousia is capable of year‐round reproduction, recruitment and growth, and thus is poised to preempt space from eelgrass following any disturbance that results in eelgrass declines, such as habitat fragmentation, eutrophication, or disease.     

Characterization of microsatellite loci in the dwarf eelgrass Zostera noltii (Zosteraceae) and cross‐reactivity with Z. japonica

Zostera noltii is an important species of eelgrass occurring along European, north African, Mediterranean, Black Sea and Azov Sea coasts. Nine microsatellite loci were developed and no linkage disequilibrium was observed. Cross‐amplification was observed for all loci (polymorphic) in Z. japonica; only four loci amplified (monomorphic) in Z. marina.     

Algicidal and growth-inhibiting bacteria associated with seagrass and macroalgae beds in Puget Sound,WA, USA

The algicidal and growth-inhibiting bacteria associated with seagrasses and macroalgae were characterized during the summer of 2012 and 2013 throughout Puget Sound, WA, USA. In 2012, Heterosigma akashiwo-killing bacteria were observed in concentrations of 2.8 × 10⁶ CFU g⁻¹ wet in the outer organic layer (biofilm) on the common eelgrass (Zostera marina) in north Padilla Bay. Bacteria that inhibited the growth of Alexandrium tamarense were detected within the biofilm formed on the eelgrass canopy at Dumas Bay and North Bay at densities of ∼10⁸ CFU g⁻¹ wet weight. Additionally, up to 4100 CFU mL⁻¹ of algicidal and growth-inhibiting bacteria affecting both A. tamarense and H. akashiwo were detected in seawater adjacent to seven different eelgrass beds. In 2013, H. akashiwo-killing bacteria were found on Z. marina and Ulva lactuca with the highest densities of ∼10⁸ CFU g⁻¹ wet weight at Shallow Bay, Sucia Island. Bacteria that inhibited the growth of H. akashiwo and A. tamarense were also detected on Z. marina and Z. japonica at central Padilla Bay. Heterosigma akashiwo cysts were detected at a concentration of 3400 cysts g⁻¹ wet weight in the sediment from Westcott Bay (northern San Juan Island), a location where eelgrass disappeared in 2002. These findings provide new insights on the ecology of algicidal and growth-inhibiting bacteria, and suggest that seagrass and macroalgae provide an environment that may influence the abundance of harmful algae in this region. This work highlights the importance of protection and restoration of native seagrasses and macroalgae in nearshore environments, in particular those regions where shellfish restoration initiatives are in place to satisfy a growing demand for seafood.     

Recovery of the seagrass <Emphasis Type="Italic">Zostera marina</Emphasis> in a disturbed Mediterranean lagoon (Etang de Berre,Bouches-du-Rhône,Southern France)

Until 1966, Berre lagoon (Provence, southern France) was a salty lagoon with a highly diversified marine fauna and flora. Increased inputs of freshwater from a hydroelectric plant led to the desalination and stratification of the lagoon. Following these changes, a wide diversity of species disappeared, including the eelgrass Zostera marina, a keystone species which previously constituted extensive meadows in the lagoon. After almost three decades of absence, patches of eelgrass were observed in 2001 and 2002 but the success of future recolonization by this community is uncertain.     

Large‐Scale Zostera marina (eelgrass) Restoration in Chesapeake Bay,Maryland, USA. Part I: A Comparison of Techniques and Associated Costs

The Chesapeake Bay, like many other temperate estuaries, has exhibited dramatic declines in the abundance of submerged aquatic vegetation (SAV) during the later half of the twentieth century. Because of the functions SAV serve in maintaining a healthy estuarine ecosystem, SAV restoration has become an important component of Chesapeake Bay restoration. Specifically, recent water quality improvements in areas from which populations of Zostera marina (eelgrass) have been extirpated have suggested that Z. marina restoration could succeed. Early restoration efforts involved transplanting Z. marina plants from healthy source beds to restoration locations, but this was labor intensive, time consuming, expensive, and potentially detrimental to donor beds. This multi‐year project investigated new techniques for large‐scale Z. marina seed collection and processing and compared two seed dispersal methods to evaluate cost effectiveness. Tens of millions of mature Z. marina seeds were collected through snorkeling, SCUBA, or with a mechanical harvester. Seed storage conditions and processing techniques were manipulated in order to maximize seed yield. Seeds were dispersed using two methods: spring seed buoys and fall seed broadcasts. Our costs for planting 1 ha of bottom with Z. marina seeds ranged from $6,674 to $165,699 depending on seeding density and dispersal method used. The average cost per Z. marina seed was $0.17. Interannual variations in seed collection yield and seed viability after summer storage had great impact on final costs. Our results suggest that the use of seeds for large‐scale Z. marina restoration offers a competitive advantage to more traditional transplanting methods.     

Lack of epifaunal response to the application of salt for managing the noxious green alga <Emphasis Type="Italic">Caulerpa taxifolia</Emphasis> in a coastal lake

Since the discovery of the green alga Caulerpa taxifolia in Lake Macquarie (New South Wales, Australia) in 2001, the New South Wales Department of Primary Industries (Fisheries) has attempted various control methods, including covering the alga with granulated sea salt to induce osmotic shock and cell lysis. In Lake Macquarie, C. taxifolia often occurs in patches within beds of the native seagrass Zostera capricorni. Although the effects of the salt treatment on blades of Z. capricorni and infauna have been shown to be minimal, there have been no tests of any effects on other native biota, including seagrass epifauna. In this study, we tested the general hypothesis that the abundance and diversity of epifauna would be reduced by salting. We used a ‘Beyond BACI’ experimental design whereby epifaunal invertebrates were sampled 3 months, 6 weeks and 6 days before and then again after salting. Epifaunal abundances at the putatively impacted (salted) location were compared to those at 4 control locations (where no salt was applied). Abundances of most organisms varied significantly among times and locations with no evidence of the consistent effect of salting on diversity or abundance of epifauna. The study represents an example of the use of large-scale managerial action as a scientific experiment. Electronic supplementary material Supplementary material is available for this article at <> and accessible for authorised users     

Importance of Shallow Water Habitats for Demersal Fishes and Decapod Crustaceans in Penobscot Bay, Maine

We characterized demersal fish and decapod crustacean habitats in 14 shallow (< 12m) areas in Penobscot Bay, Maine, by mapping the distribution of subtidal substrata with an acoustic sea bottom profiler. We identified the aquatic vegetation and the fishes and decapods associated with these habitats. Samples of fishes and decapods in each habitat were collected with a small beam trawl. The seabed at most of the stations sampled was composed of a mosaic of substrata. More species of fishes tow^-1 and higher abundances of fishes and decapods were found in stations with vegetation present, particularly eelgrass, Zostera marina. Decapod species richness tow^-1 was similar between vegetated and unvegetated habitats. Shallow habitats in Penobscot Bay were dominated by juvenile fishes and function as nursery areas. The greater species diversity and higher abundances of epibenthic fishes and decapod crustaceans observed in vegetated habitats, particularly beds of Zostera marina, compared with unvegetated areas in Penobscot Bay conform to the hypothesis that increased habitat complexity results in increased species richness and abundance.The first author is also senior author     

Causes of the eelgrass wasting disease: Van der Werff's changing theories

The 1930's wasting disease among the North Atlantic population of eelgrass,Zostera marina, is still an ecological and historical enigma, despite several attractive theories. Van der Werff investigated the die-back of eelgrass in the thirties in the Dutch Wadden Sea, and he considered the micro-organismLabyrinthula as the possible cause of the disease. In 1980, Grevelingen lagoon, harbouring an extensive population ofZostera marina, experienced a major decline of the area covered by the submerged macrophyte. Speculations about the cause of this dramatic decline induced us to think that the wasting disease had struck again. Van der Werff investigated the Grevelingen population and found bothLabyrinthula and a Chaetophoracean endophytic alga to be presumably responsible for the decline. During the quest for the ultimate cause of the wasting disease the question remains whether both micro-organisms are the cause of the disease or simply an effect of decomposition processes triggered by other factors.     

Differential anaerobic decomposition of seagrass (Zostera noltii) and macroalgal (Monostroma obscurum) biomass from Arcachon Bay (France)

Arcachon Bay is characterized by extensive meadows of the seagrass Zostera noltii. Moreover, as a consequence of eutrophication, massive proliferations of the macroalga (Monostroma obscurum) have occurred since the beginning of 1990s.This paper describes the anaerobic decomposition of biomass of both species under experimental conditions by two methods. Firstly, the dynamics of decomposition were studied in situ using litter bags. The remaining biomass and the elemental composition of the decomposing macrophytes were monitored. Secondly, degradation was studied in experimental containers under anoxic conditions in which the release of inorganic nutrients and the development of fermentative and sulfate-reducing bacterial populations were followed.The decomposition rate of total biomass was faster for macroalgae than for the vascular plants, thus corroborating previous observations. However, both in situ and laboratory experiments showed that the anaerobic decomposition of the seagrass Z. noltii resulted in rapid release of inorganic N and P, and increasing C/N and C/P ratios of the residual biomass. As a result, the recycling of inorganic nitrogen and phosphorus compounds was slightly more efficient for Z. noltii than for M. obscurum. Recycling of inorganic nutrients appears to be of a great importance to the whole ecosystem, because of the extensive spreading of Z. noltii in the bay.     

The role of seedlings and seed bank viability in the recovery of Chesapeake Bay,USA, <Emphasis Type="Italic">Zostera marina</Emphasis> populations following a large-scale decline

The objective of this study was to quantify the spatial and temporal recolonization characteristics of Zostera marina beds in the lower Chesapeake Bay following large scale declines in the late summer of 2005. Transects were established and monitored monthly for changes in eelgrass abundance at three sites (two downriver, one upriver) in the York River from April–October 2006 and 2007. Measurements included percent bottom cover, above ground biomass, shoot density, shoot origin (seedling or vegetative), seed bank abundance and seed viability. During 2006, the eelgrass beds at all sites recovered with seedlings providing the largest proportion of the total shoot abundance. This trend shifted in 2007 and surviving vegetative shoots were the dominant component of shoot standing crop. A second consecutive decline related to low light conditions occurred during the summer of 2006 in the upriver site and recovery there was minimal in 2007. These results highlight that after a single die off event, seed germination with subsequent seedling growth is the principal method for revegetation in lower Chesapeake Bay Z. marina beds. However, no viable seeds remain in the seed bank during this first year of recovery and shoots produced by the seedling growth do not flower and produce seeds until their second year of growth. Therefore the seed-bank density is low and is not immediately replenished. This suggests that the resiliency of perennial Chesapeake Bay Z. marina populations to repeated disturbances is restricted and repeated annual stress may result in much longer term bed loss.     

The dieback ofZostera marina in the 1930's in the Waddensea; An eye-witness account by A. van der Werff

In the archives of the State Institute of Fisheries Research a handwritten report of A. van der Werff on the wasting disease ofZostera marina L. in the Waddensea in 1932 and 1933 is kept. It contains an eye-witness account of the decline of the eelgrass beds. In 1932 the first attack of the disease took place in May and caused heavy devastation. Regeneration started soon, but a second attack at the end of August seemed to destroy the beds definitively. The next year was a year of struggle for the few poorly developed surviving plants. In 1938 the last surviving specimens were seen in the sublittoral.     

Experimental test of biodeposition and ammonium excretion from blue mussels (Mytilus edulis) on eelgrass (Zostera marina) performance

Blue mussels and eelgrass have been found to coexist in many locations. However, knowledge of the interactions between these species is limited. Two experiments were conducted in the laboratory, a “Deposit” and an “Epiphyte” experiment. The Deposit experiment examined possible effects of increasing load of blue mussel (Mytilus edulis) biodeposits on sediment biogeochemistry and eelgrass (Zostera marina) performance. Z. marina mesocosms received normal or high loads of mussel biodeposits (Normal and High), while no biodeposits were added to the Control. High dosage had overall negative effects on Z. marina, which was reflected as lower leaf numbers and biomass and accumulation of elemental sulphur in rhizomes. The sediment biogeochemical conditions were altered, as the mussel biodeposits enhanced sulphate reduction rates and increased sulphide concentrations in the porewater, which may result in sulphide invasion and reduced growth of Z. marina.In the Epiphyte experiment effects of mussel excretion, with particular emphasis on ammonium, on the growth of Z. marina and their epiphytes were examined. A thick cover of epiphytes developed on Z. marina growing together with M. edulis, and the relative growth rate was reduced with 20% compared to plants from control without mussels. Overall the experiments showed negative effects on Z. marina growing together with M. edulis, thereby supporting a preceding field study by Vinther et al. [Vinther, H.F., Laursen, J.S., Holmer, M. 2008. Negative effects of blue mussel (Mytilus edulis) presence in eelgrass (Zostera marina) beds in Flensborg fjord, Denmark. Est. Coast Shelf. Sci. 77, 91-103.].     

Large‐Scale Zostera marina (eelgrass) Restoration in Chesapeake Bay,Maryland, USA. Part II: A Comparison of Restoration Methods in the Patuxent and Potomac Rivers

In response to systemic losses of submerged aquatic vegetation (SAV) in the Chesapeake Bay (east coast of North America), the U.S. Environmental Protection Agency's (EPA) Chesapeake Bay Program (CBP) and Maryland Department of Natural Resources (MD DNR) have considered SAV restoration a critical component in Bay restoration programs. In 2003, the CBP created the “Strategy to Accelerate the Protection and Restoration of Submerged Aquatic Vegetation in the Chesapeake Bay” in an effort to increase SAV area. As part of this strategy, large‐scale eelgrass (Zostera marina) restoration efforts were initiated in the Patuxent and Potomac Rivers in Maryland. From 2004 to 2007, nearly 4 million Z. marina seeds were dispersed over 10 ha on the Patuxent River and almost 9 million seeds over 16 ha on the Potomac River. Z. marina seedling establishment was consistent throughout the project (<4%); however, restored eelgrass survival was highly dependent on restoration site. Restoration locations on the Patuxent River experienced initial Z. marina seedling germination, but no long‐term plant survival. Restored Z. marina on the Potomac River has persisted and expanded, both vegetatively and sexually, beyond initial seeding areas. Healthy Z. marina beds now cover approximately five acres of the Potomac River bottom for the first time in decades. The differential success of Z. marina restoration efforts in the two rivers is evidence for the necessity of carefully considering site‐specific characteristics when using large‐scale seeding methods to achieve successful SAV restoration.