The Effects of Phragmites Removal on Nutrient Pools in a Freshwater Tidal Marsh Ecosystem

Selected nitrogen and phosphorus pools in two freshwater tidal marsh ecosystems on the lower Connecticut River were measured relative to Phragmites, Typha and mixed native wetland plant cover types. For both the Chapman Pond Preserve and Chester Creek Marsh, significant differences were found between plant cover types in porewater ammonium and phosphate for some months during the 1998 growing season; porewater nitrate was always below the detection limit. At Chapman Pond, no significant differences were detected between Phragmites and Typha for plant tissue N concentrations. The standing stock of nitrogen was higher in Phragmites stands, however, owing to its greater aboveground biomass. No significant difference was found between plant cover types for total soil N at Chapman Pond, but KCl extractable ammonium was higher in the mixed cover type than Phragmites or Typha. The results of this study suggest that Phragmites is affecting nutrient pools in freshwater tidal marshes, a result that should be considered in future management design.     

丹顶鹤多尺度生境选择机制——以黄河三角洲自然保护区为例

生境在鸟类生活史中发挥着重要的作用,关系到鸟类的生存和繁衍。由于鸟类对环境变化的响应发生在等级序列空间尺度上,基于多尺度的研究更能深入刻画鸟类-环境之间关系。以丹顶鹤(Grus japonensis)为研究对象,以其迁徙和越冬的重要地区-黄河三角洲自然保护区为研究区域,应用等级方差分解法和等级划分法,分析丹顶鹤与微生境、斑块、景观尺度因子之间的关系,探求丹顶鹤生境选择的主要影响因素和尺度。等级方差分解结果表明,在第1等级水平,景观尺度因子与微生境、斑块尺度因子之间的联合效应大于独立效应,景观尺度因子的独立效应大于微生境和斑块尺度因子;在第2等级水平,景观尺度上的景观组成因子重要性大于景观结构因子,微生境尺度上的植被和水分因子为重要影响因素。等级划分结果表明,景观尺度上,翅碱蓬滩涂、水体面积大小是主要影响因素;微生境尺度上,植被盖度和水深为主要限制因子;在斑块尺度上,斑块类型对丹顶鹤生境选择最为重要。研究认为,在黄河三角洲自然保护区,景观尺度是影响丹顶鹤生境选择的主要尺度,景观尺度因子通过与微生境和斑块尺度因子的独立和联合作用制约着丹顶鹤在保护区的生境选择和空间分布格局。建议加强对翅碱蓬滩涂、芦苇沼泽、水体等湿地生境的保护和管理,规范和控制保护区内人类活动强度。     

近33年白洋淀景观动态变化

湿地具有重要的生态功能,由于各种因素的影响改变了湿地水文条件,并且导致湿地退化。利用1974、1987、1996和2007年遥感影像,使用ERDAS 9.1、ARCGIS 9.2和FRAGSTATS 3.3,分析了白洋淀景观动态变化过程及其驱动力。研究结果表明:从1974至2007年,白洋淀湿地面积从249.4km²下降到182.6km²,农田和居民地面积分别从70.0、2.1km²增加到126.4、12.5km²;在景观水平上,斑块密度、景观形状指数和香农多样性指数增加,蔓延度减小,景观破碎化程度增加,优势景观类型湿地对整个景观的控制作用减小;在类型水平上,明水面和沼泽的斑块密度和周长-面积分维数增加,最大斑块指数、平均斑块面积和连接度减小,居民地的斑块密度、最大斑块指数、平均斑块面积、周长-面积分维数和连接度都增加,农田的斑块密度、最大斑块指数和连接度增加,平均斑块面积和周长-面积分维数减小;水位变化是影响白洋淀景观变化的主要因素,水位升高会使湿地面积增加,反之,使湿地面积减小;流域人类活动和降水量变化是白洋淀水位和景观变化的重要驱动力。引水补淀只能暂时缓解白洋淀缺水问题,从流域角度考虑,实施有效的水资源管理方式,才能从根本上解决白洋淀湿地退化的问题。     

Chemical oxygen demand,nitrogen and phosphorus removal by subsurface wetlands with Phragmites vegetation in different models

To improve the removal efficiency of subsurface wetlands vegetated mainly by Phragmites, pilot‐scale gravel‐based wetlands were used to treat sewage characterized by chemical oxygen demand (COD), total nitrogen (TN) and total phosphorus (TP) pollution. For Phragmites vegetation, COD, TP and TN removal loads of wetland vegetation with Phragmites australis–Typha angustata–Scirpus validus as main species reached 0.517 g m^?2 d^?1, 0.277 g P m^?2 d^?1 and 0.023 g N m^?2 d^?1. The COD removal loads in pilot‐scale and medium‐scale (260 m² in area) wetlands with Phragmites‐monoculture vegetation were 0.62–0.64 g m^?2 d^?1, while that of P. australis–T. angustata–S. validus wetland reached 0.974 g m^?2 d^?1. Thus, the preferable poly‐culture model for Phragmites wetland vegetation was P. australis, T. angustata, S. validus and Zizania latifolia with stem density ratio of 8:1:5:1. After harvest, nitrogen and phosphorus standing stocks of wetland vegetations ranged only 2.2–9.93 g N m^?2 and 5.39–13.5 g P m^?2, respectively, as both the above ground biomass and the nitrogen and phosphorus contents of the wetland vegetation harvested in late autumn were low.     

Aquatic grazers reduce the establishment and growth of riparian plants along an environmental gradient

相似文献   

Alternation between clear,high-vegetation and turbid,low-vegetation states in a shallow lake: the role of birds

External phosphorus load to a wetland with two shallow lakes in the Botshol Nature Reserve, the Netherlands, was reduced since 1989, resulting in a rapid reduction of phosphorus levels, phytoplankton biomass and turbidity, and after 4 years, explosive growth of Characeae. The clear water state was unstable, however, and the ecosystem subsequently alternated between clear, high-vegetation and turbid, low-vegetation states. The switch from clear to turbid states occurred at a higher phosphorus level than the switch from turbid to clear states and area covered by macrophytes was significantly correlated to Secchi disc depth (r = 0.86, p < 0.001). Beginning in 1997, phosphorus input from droppings of greylag geese and black-headed gulls increased. The hypothesis that grazing prevented the return of Characeae populations was tested during 1996–1997 using large-scale exclosure experiments. The biomass and cover of Characeae were monitored in 17 exclosures for 2 years. The experiments showed no significant increase in Characeae in the absence of grazers compared with the controls. During the course of the experiments, the entire lake switched from turbid water to clear water and high Chara biomass. These ecosystem developments suggest that light limitation was the main factor controlling the collapse and return of Characeae in Botshol.     

黑河中游湿地景观破碎化过程及其驱动力分析

在遥感和GIS技术支持下,基于1975-2010年长时间序列遥感影像,选取斑块密度指数(PD)、景观内部生境面积指数(IA)、斑块平均面积指数(MPS)、斑块形状破碎化指数(FS1、FS2)等具有典型生态意义的景观指数模型,系统分析了黑河中游湿地景观的破碎化过程,并结合灰色关联分析、主成分分析等方法,探讨了影响研究区湿地景观破碎化过程的各驱动因子.结果表明:近35年来,研究区湿地景观破碎化主要表现为斑块平均面积的萎缩,斑块密度的上升以及斑块形状破碎化指数的增大.整个研究时段内,研究区湿地斑块平均面积减少了48.95hm2,斑块密度的上升0.006个/hm2;导致黑河中游湿地景观破碎化发生和发展的驱动力包含自然和人文两个方面.自然因子对湿地景观破碎化进程的影响则主要体现在气温和降水上,而且气温对湿地景观破碎化进程的影响程度明显大于降水.但在1975-2010年间的这一较小时间尺度上,人类活动对湿地景观破碎化的贡献率明显高于自然因子,人类活动能力的增强以及影响范围的不断扩大是引发黑河中游湿地景观破碎化的主因.     

PhragNet: crowdsourcing to investigate ecology and management of invasive <Emphasis Type="Italic">Phragmites australis</Emphasis> (common reed) in North America

Invasion biology research, often performed by scientists at relatively small spatial scales, provides experimental precision but may be limited in generalizability. Conversely, large-scale invasive species management represents a largely untapped wealth of information on invasion ecology and management, but such data are difficult to capture and synthesize. We developed a network (“PhragNet”) of individuals managing wetlands occupied by native and non-native lineages of the invasive wetland grass Phragmites australis (common reed). This network collected environmental and genetic samples, habitat data, and management information to identify environmental and plant community associations of Phragmites invasion and patterns of management responses. Fifty managers overseeing 209 Phragmites stands in 16 US states and ON, Canada participated. Participants represented federal agencies (26%), municipalities (20%), NGOs (20%), academia (14%), state agencies (12%), and private landowners (8%). Relative to the native lineage, non-native Phragmites occurred in areas with higher nitrate/nitrite and ammonium than non-native Phragmites. Stand interiors had higher soil electrical conductivity than nearby uninvaded areas, consistent with use of road salt promoting spread of Phragmites. Non-native Phragmites co-occurred with fewer plant species than native Phragmites and was actively targeted for management. Herbicide was applied to 51% of non-native stands; surprisingly, 11% of native stands were also treated with herbicide. This project demonstrates the utility of crowdsourcing standardized data from resource managers. We conclude by describing how this approach could be expanded into an adaptive management framework, strengthening connections between wetland management and research.     

Constructed wetlands in Queensland: Performance efficiency and nutrient bioaccumulation

Nine pilot wetlands (eight free water surface and one subsurface flow) have been constructed in Queensland as joint projects between the State and Local Governments, to treat municipal wastewater. The wetlands are in several geographical locations which include tropical, subtropical and arid climates. Each wetland is a different configuration and contains a variety of macrophyte types and species. Most species are native and were collected in the locality or self colonised. This paper examines the performance efficiency of the wetlands and nutrient bioaccumulation in wetland plants. Biochemical oxygen demand concentrations were reduced by 17–89% and suspended solids concentrations by 14–77% to produce wetland effluent with BOD less than 12 mg l⁻¹ and suspended solids less than 22 mg l⁻¹. Reduction in total nitrogen concentrations ranged from 18 to 86%, ammonia nitrogen from 8 to 95% and oxidised nitrogen from 55 to 98%, producing effluent with total nitrogen between 1.6 and 18 mg l⁻¹. Reduction in reactive phosphorus concentration was less than 13% in the free water surface systems with concentration in the effluent exceeding the influent in many of the systems over long term operation. In contrast reduction through the single household subsurface system was 65%. Nutrient bioaccumulation was investigated in 60 species. Submerged (Ceratophyllum) and free floating species (duckweed) had the highest tissue nutrient concentrations, followed by the waterlily (Nymphoides indica), aquatic vines (Ipomoea spp., Ludwigia peploides), and waterferns (Ceratopteris, Marsilea). All these species remove nutrients from the water column. Emergent species had lower nutrient concentrations with the highest nutrients occurring in the exotic sedge Cyperus involucratus. Aquatic grasses including Phragmites had higher nutrient content than the sedges. Nitrogen concentrations were higher in leaf/stem tissue compared to the root/rhizome, whereas phosphorus was higher in root/rhizome tissue. Emergent species had a greater biomass than submerged or free floating species and were therefore able to store more nutrients per unit area of wetland. Cropping the shoots of emergent species increased nutrient content in new shoot growth.     

Characteristic community structure of Florida’s subtropical wetlands: the Florida wetland condition index for depressional marshes, depressional forested, and flowing water forested wetlands

Land use and land cover change has a marked affect on wetland condition, and different wetland types are affected differentially depending on many abiotic and biotic variables. To assess wetland condition, we have developed a Florida wetland condition index (FWCI) composed of indicators of community structure in the diatom, macrophyte, and macroinvertebrate assemblages for 216 wetlands (n = 74 depressional marsh, n = 118 depressional forested, n = 24 flowing water forested wetlands). Depressional wetlands located along a human disturbance gradient throughout Florida were sampled for each assemblage. Forested flowing water wetlands were sampled for macrophytes only. The landscape development intensity index (LDI) was used to quantify the human disturbance gradient. In general, human disturbance in adjacent areas had the greatest impact on depressional herbaceous wetlands, followed by depressional forested wetlands. Forested flowing water wetlands (i.e., forested strands and floodplain wetlands) were less affected by local conditions, with most of their changes in wetland condition correlated with alterations at the larger watershed scale. Strong correlations between the FWCIs and LDI index scores suggest that changes in community structure can be detected along a gradient of human land use activities adjacent to wetland ecosystems.     

Disjunct performance and distribution in the sedge Carex prasina

Efforts to understand species distributions and predict responses to environmental changes depend on specifying how the abiotic environment determines distributions. At landscape scales, it is critical to distinguish effects of environmental factors from other mechanisms such as competition and dispersal limitation. We examined how environmental factors affect the distribution and performance of the sedge Carex prasina across a 10-km² old-growth forest in southern Québec. We isolated the effects of soil characteristics by conducting a greenhouse experiment that assessed the performance of C. prasina on soils from a range of wetland habitats where it could potentially occur. This allowed us to compare how the species’ performance and its distribution across the landscape relate to the same soil characteristics. In the experiment, the biomass and leaf chlorophyll content of C. prasina increased with increasing soil organic matter (OM). Across the landscape, however, the species’ probability of occurrence and abundance decreased with increasing soil OM. C. prasina had similar biomass on soils from sites where it did and did not occur, but it had higher leaf chlorophyll content on soils from sites where it did not occur. We found no evidence that differential performance across environments determines the distribution of this species, as C. prasina tended to occur on soils where it showed reduced performance. Rather, other processes such as competition or dispersal limitation likely override any direct effects of the soil environment on distribution. Our results caution against the common assumption that the environments where a species tends to occur or be most abundant are the environments where it performs best. C. prasina presents a clear example of a species whose performance, at least along edaphic gradients, cannot explain its distribution. This example highlights the importance of distinguishing the relative roles of biotic and abiotic factors that shape species distributions across landscapes.     

Dominant environmental factors in wetland plant communities invaded by Phragmites australis in East Harbor, Ohio, USA

Elevation, standing crop, disturbance and soil fertility often emerge from studies of freshwater plant communities as the dominant environmental factors determining both species richness and species composition. Few studies in North America have investigated the relationship between these factors and species abundance (standing crop) and species composition in the context of invasion by Phragmites australis. This study explores the influence of key abiotic and biotic variables on species abundance and composition across three Lake Erie wetlands differing in hydrology and Phragmites abundance in East Harbor, Ohio, USA. Standing crop for 92 species was related to standard sediment analyses, wave exposure, distance to shoreline, elevation, light interference, species density, and Phragmites standing crop in each of 95 1 × 1 m quadrats by using canonical correspondence analysis (CCA). Elevation (Axis I) and Phragmites standing␣crop-soil fertility (Axis II) explained 35.7 and 26.2%, respectively, of the variation in the species–environment relationships. Wave exposure was not a primary component of the first four canonical axes. Axis I was instrumental in describing species composition, separating wet meadow species from marsh species. Axis II was inversely related to species density for both wet meadow and marsh species. These findings generally support prevailing models describing the distribution of wetland plants along environmental gradients. Two discrepancies were noted, however: (1) species density was highest in the most sheltered sites and (2) wave exposure was directly associated with Phragmites standing crop-soil fertility gradient. The structural integrity of Phragmites stems, topographic heterogeneity and differential responses to anthropogenic disturbance may contribute to departure from prevailing multivariate models. This information has direct implications for local and regional wetland managers.     

Assessing invasive plant infestation and disturbance gradients in a freshwater wetland using a GIScience approach

The assessment of aquatic ecosystems requires information on biological and disturbance gradients in order to evaluate quality. As a result decision makers need improved monitoring tools for characterizing relationships between invasive species infestation and disturbance to make informed choices regarding wetland condition and management plans. The overarching goal of this research was to assess invasive plant infestation and disturbance gradients using a GIScience approach. The study was conducted in a fresh-water, coastal wetland in the Muskegon River watershed, Michigan, USA. Airborne hyperspectral imagery (20 bands, 440–880 nm) was classified for Phragmites australis distribution using the Spectral Angle Mapper algorithm. Indicator semivariograms were utilized to define landscape structure and associated spatial scales, and assist in creating a transect scheme to generate landscape pattern metrics quantifying valued ecosystem attributes. Hydrological modifications, as measured by an area-weighted fractal dimension index, served as a proxy for human disturbance and was found to moderately influence Phragmites percent cover (R ² = 0.4, n = 40), mean patch size (R ² = 0.5), and patch shape (R ² = 0.5). A general conclusion was that increased hydrological disturbances were correlated with increased infestation magnitude. The systematic approach executed in this study outlined how geospatial monitoring tools can be used as an assessment framework to provide more meaningful information that lends itself to comprehensive wetlands assessment.     

Life on the edge: reproductive mode and rate of invasive <Emphasis Type="Italic">Phragmites australis</Emphasis> patch expansion

The dynamics of plant invasions from initial colonization through patch expansion are driven in part by mode of reproduction, i.e., sexual (seed) and asexual (clonal fragments and expansion) means. Expansion of existing patches—both rate and mode of spread into a matrix of varying conditions—is important for predicting potential invader impacts. In this study, we used fine-scale genetic assessments and remote sensing to describe both the rate and mode of expansion for 20 Phragmites australis patches in flooded and unflooded wetland units on the Great Salt Lake, UT. We found that the majority of Phragmites patch expansion occurred via clonal spread but we also documented instances of (potentially episodic) seedling recruitment. The mode of patch expansion, inferred from patch edge genet richness, was unrelated to flooding in the wetland unit in the preceding growing season. The rate of Phragmites patch expansion varied from 0.09 to 0.35 year^?1 and was unrelated to the mode of spread. In six patches monitored across two years, monoclonal patches stayed monoclonal, whereas patches with higher genet richness had a marked increase in diversity in the second year. The findings of the present study suggest how this partially clonal species can exploit the benefits of both sexual (i.e., genetic recombination, widespread dispersal, colonization of new areas) and asexual reproduction (i.e., stability of established clones suited to local environmental conditions) to become one of the most successful wetland plant invaders. To control this species, both forms of reproduction need to be fully addressed through targeted management actions.     

Preference of wet dune species for waterlogged conditions can be explained by adaptations and specific recruitment requirements

The preference of wetland angiosperms for waterlogged soils has been explained by several hypotheses: (1) wetland species are adapted to waterlogging and sensitive to drought; (2) wetland species are tolerant to drought, but inferior competitors at drier conditions; (3) wetland species have narrow moisture optima for recruitment. We tested these hypotheses by the application of permanent and fluctuating water levels to experimental wet dune plant communities with four species that frequently occur in wet dune slacks (Carex flacca, Molinia caerulea, Samolus valerandi and Schoenus nigricans) and one competitive species of drier conditions (Calamagrostis epigejos). After 3 years, fluctuating water levels had led to lower total biomass production than permanent water levels, indicating that switching from aerated to anoxic soil conditions involved physiological costs. The collective biomass of wetland species was highest at permanently waterlogged conditions with interspecific variation in the biomass optima. At the nutrient-poor growth conditions of this long-term experiment, biomass of Calamagrostis was independent of water level treatment, thus the hypothesized competitive superiority of this species at drier conditions could not explain the biomass responses of the wetland species in this study. Instead, this is evidence for an ecological preference of adult wet dune plants for waterlogged to moist conditions. Recruitment of most wet dune slack species occurred in a narrower range of water levels than adult growth, indicating that recruitment requirements also pose a limitation to the distribution of these wet dune species. Incorporation of recruitment into nature management support models may improve their predictions.     

盐城海滨湿地景观演变关键土壤生态因子与阈值研究

文章以盐城海滨湿地典型区域为案例,以2011年4月和2012年4月对海滨湿地土壤数据为基础,结合2011年ETM 遥感影像,运用灰色关联分析、线性回归模拟和地统计学方法,辨识海滨湿地景观演变的关键生态因子,并确定其生态阈值。得出基本结论如下：(1)海滨湿地土壤理化性质海陆差异明显：从米草沼泽—碱蓬沼泽—芦苇沼泽,土壤水分和盐度表现出递减的趋势;土壤有机质、营养盐总体上从米草沼泽—碱蓬沼泽—芦苇沼泽,表现出两头高中间低的特征;湿润年份土壤水分高于干旱年份,湿润年份土壤盐度低于干旱年份。(2)灰色关联分析表明:干旱年份,水分＞盐度＞铵态氮＞速效钾＞有机质＞有效磷;湿润年份为：盐度＞水分＞有效磷＞铵态氮＞速效钾＞有机质,因此把土壤水分和盐度确定为海滨湿地景观演变的关键生态因子。(3)土壤水分和盐度在东西海陆方向上的变异大于南北海岸延伸方向上的变异。(4)将景观类型图和海滨湿地土壤水分与盐度分异图叠加分析,得出：芦苇滩土壤水分阈值＜42.332%,盐度阈值＜0.745%;碱蓬滩土壤水分阈值为38.836%～46.593%,盐度阈值为0.403%～1.314%;米草滩土壤水分阈值＞39.475%,盐度阈值＞0.403%;光滩的土壤阈值＞41.550%,盐度阈值＞0.656%。     

Bioturbation by the sand dollar <Emphasis Type="Italic">Encope emarginata</Emphasis> (Echinoidea,Mellitidae) changes the composition and size structure of microphytobenthic assemblages

The goal of this study is to examine the suitability of three plants, Typha spp., Phragmites spp. and Iris pseudacorus, in a free-water surface constructed wetland created to treat eutrophic water from Lake Albufera (Valencia, Spain), a wetland of international importance. The growth, coverage and nutrient content of the three plants were studied, and chemical analyses were performed according to standard methods. The maximum standing crops measured for each plant were 1.9, 18.2 and 3.3 kg m^?2, respectively, and their average nutrient concentrations were 2.1, 1.2 and 1.7 g P kg^?1 and 12.1, 11.7 and 10.1 g N kg^?1, respectively. A multiple harvest of Iris pseudacorus revealed that the removal of nutrients could be increased up to 50% for N and 100% for P compared with a single harvest. Biomass decomposition assays showed high values for five-day biochemical oxygen demand (115–207 mg O₂ g^?1, depending on the plant and its age) and a substantial release of phosphorus, up to 100% of that contained in the biomass, highlighting the need to remove the litter fall. This study provides key aspects for vegetation selection and management (planting and harvesting) in a novel application of constructed wetlands to enhance water quality and biodiversity.     

长江流域景观格局与生态系统水质净化服务的关系

流域景观格局通过影响生态过程,改变进入河流污染物的数量,进而对水质净化服务产生重要影响,探讨流域景观格局对水质净化服务的影响对于流域景观规划、生态系统保护和生态系统服务提升具有重要意义。以长江流域为研究区域,在分析45个子流域景观格局特征、应用In VEST模型评估流域水质净化服务基础上,探讨了两者的关系,结果表明:①景观组成上,长江流域农田和城镇面积比例分别与生态系统水质净化服务存在显著对数关系(P0.01),森林面积比例则与之呈极显著正相关关系(P0.01);②流域景观配置上,斑块密度和景观破碎度与水质净化服务呈显著负相关(P0.01),而平均斑块面积和形状规律相反(P0.01);③斑块类型水平上,森林平均斑块面积、灌丛/湿地平均斑块形状与生态系统水质净化服务呈显著正相关(P0.01),而农田平均斑块面积/边缘密度、城镇斑块密度则与之呈显著负相关(P0.01);④森林主导景观的子流域,仅有景观破碎度与生态系统水质净化服务呈显著负相关,而农田主导景观的子流域,景观蔓延度、香农多样性与其分别呈正相关(P0.01)。研究结果可以为长江流域生态系统水质净化服务的提升提供多途径的管理信息:流域景观尺度上,增加森林面积比例、控制农田与城镇面积比例,并减少景观破碎度而增加平均斑块形状复杂性;斑块类型水平上,可增加灌丛/湿地斑块形状复杂性,减小农田边缘密度和城镇斑块密度;森林主导景观的子流域应降低景观破碎度,以农田主导景观的子流域则应该增加斑块类型丰富度和团聚程度。研究也可为其他流域水质净化服务管理提供参考。     

A rhizotron to study root growth under flooded conditions tested with two wetland Cyperaceae

A new type of rhizotron is described that facilitates the study of root growth of inundated wetland plants. This rhizotron provides: (1) non-destructive root morphometric measurements during an experiment, (2) rhizotron-specific independent experimental units, (3) a water level of up to 45 cm above the soil surface, (4) easy access for planting, and (5) the ability to harvest at multiple soil depths. A removable back allows easy access for planting and harvesting at multiple soil depths. A gasket ensures a watertight seal. Detailed methods of construction allow researchers to construct and use similar rhizotrons for experimental research. Advantages and disadvantages of the new rhizotron type are discussed.Using the new rhizotron construction, growth response of the Cyperaceae species Eleocharis cellulosa and Rhynchospora tracyi from the Florida Everglades under different water levels and phosphate availabilities were investigated. Destructive and non-destructive analysis of development of above- and belowground biomass for the two species demonstrates that the two species differ in their response to alterations in water level, but show similarities in their response to phosphorus availability. While E. cellulosa showed larger increases in root density, root biomass, total shoot length, and shoot biomass with rising water level, the opposite was the case in R. tracyi. Both species produced more aboveground biomass with increasing phosphorus availability. Only E. cellulosa showed an increase in root density and a root biomass increase with high phosphorus availability. In R. tracyi, there was no significant increase in root biomass and root density with high phosphorus availability.     

Seed banks of Phragmites australis-dominated brackish wetlands: Relationships to seed viability, inundation, and land cover

In tidal wetlands of the eastern United States, buried seeds of the non-native haplotype of Phragmites australis may be a source of propagules for re-establishment after eradication efforts but factors controlling the development and expression of seed banks in non-native Phragmites stands have not been examined. We sampled surface soil at four Chesapeake Bay brackish tidal wetlands dominated by the non-native (European) haplotype M of Phragmites and used the seedling emergence method to quantity species of seedlings emerging under flooded and non-flooded soil conditions. Within each subestuary, one site was dominated by Phragmites that produced viable seeds (high viability) and the other by Phragmites that did not (low viability). We also described standing vegetation in plots, measured soil salinity, analyzed soil characteristics, and described surrounding land cover. Based on number of emerging seedlings, we found that 284 and 698 Phragmites seeds m⁻² occurred at the two high-viability sites, which was significantly higher than seed densities at the low-viability sites (10 seeds m⁻²), and greater than densities reported elsewhere. We also found that emergence of Phragmites seedlings from soil samples was prevented by continuous flooding of 3.5 cm of standing water, suggesting that colonization of deep water areas is due to vegetative clonal expansion from Phragmites in adjacent higher elevations. The density of Phragmites seeds was not related to soil salinity or abundance of other species in the seed bank or vegetation, but instead was positively related to greater wave energy disturbance (much longer fetch and more open water) and lower area of wetlands nearby. The seed bank was more species-rich (15-22 species observed) than standing vegetation (3-15 species) at all sites, meaning that the dominance of Phragmites in vegetation does not prevent the development of a diverse seed bank and implying that a species-rich community may establish rapidly following control efforts. Based on these results and our findings in related studies, we postulate that wave energy disturbance generates repeated opportunities for seedling recruitment by Phragmites, which creates stands of Phragmites with higher genotypic diversity. In turn, genetically diverse stands favor greater cross-pollination and production of viable seed. These findings suggest that, in North America, targeting control efforts on non-native Phragmites patches in areas of higher exposure to wave energy may be more effective in reducing source populations than efforts in more protected locations.