Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Zostera marina and Neotrypaea californiensis as indicators of ecosystem integrity in Grice Bay, British… Carty, Sarah Elizabeth 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-0176.pdf [ 7.48MB ]
Metadata
JSON: 831-1.0091050.json
JSON-LD: 831-1.0091050-ld.json
RDF/XML (Pretty): 831-1.0091050-rdf.xml
RDF/JSON: 831-1.0091050-rdf.json
Turtle: 831-1.0091050-turtle.txt
N-Triples: 831-1.0091050-rdf-ntriples.txt
Original Record: 831-1.0091050-source.json
Full Text
831-1.0091050-fulltext.txt
Citation
831-1.0091050.ris

Full Text

ZOSTERA MARINA AND NEOTRYPAEA CALIFORNIENSIS AS INDICATORS OF E C O S Y S T E M INTEGRITY IN G R I C E BAY, BRITISH COLUMBIA by S A R A H ELIZABETH C A R T Y B.Sc, The University of British Columbia, 2001 A THESIS SUBMITTED IN PARTIAL F U L F I L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES Department of Botany, U B C We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A March 2003 © Sarah Elizabeth Carty, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of T3>o"ba. ^ V V J The University of British Columbia Vancouver, Canada Date rjtA.rck ? i / 0 3 DE-6 (2/88) A B S T R A C T Using indicator species as proxies to assess ecosystem integrity has been identified as a necessary means to develop efficient and effective ways to monitor natural ecosystems. Criteria have been developed to aid in the choice of an indicator species and while there are still limitations, improvements can be made by choosing more than one species and having multiple indices to measure. This study established baseline data on the seagrass, Zostera marina and ghost shrimp, Neotrypaea californiensis for their use as health indicators in Grice Bay, Clayoquot Sound, British Columbia. The widespread distribution and ecological importance of seagrasses and their sensitivity to water quality parameters have led to the use of these plants as biological indicators of water quality. In this study, seagrass biomass, density, size structure, reproductive timing & frequency were quantified as indicator parameters. Physiological indices, for use as early warning signals, were also quantified including sugar & chlorophyll concentrations. Ghost shrimp have not previously been used as an indicator species, however their important role in the food web of Grice Bay as the primary detritivore and as a prey species for juvenile gray whales makes them an important organism to monitor. Indicator qualities of ghost shrimp are those sensitive to water quality, such as density, biomass, population size structure, frequency of reproductive females, size of reproductive females and length of reproductive season. TABLE OF CONTENTS Abstract i i Table of Contents i i i List of Tables vi List of Figures vii Acknowledgements x CHAPTER 1 Management of Mudflats and Seagrass Beds 1 1.1 Introduction 1 1.2 Monitoring Programs 1 1.2.1 Utility of monitoring programs 1 1.2.2 Economics of monitoring programs 3 1.2.3 Goal of monitoring programs 4 1.3 The Indicator Species Concept 4 1.3.1 Types of indicators 4 1.3.2 Criteria for choosing health indicators 5 1.3.3 Problems with indicator species 8 1.4 Ecosystem of Interest 8 1.4.1 Study site 8 1.4.2 General biology of soft bottom habitats 11 1.4.3 Seagrass ecology 12 1.4.4 Ghost shrimp ecology 14 1.5 Study Objectives 16 1.6 Outline of Thesis 19 CHAPTER 2 Spatial and Temporal Variability in Zostera marina beds in Clayoquot Sound 21 2.1 Introduction 21 2.2 Methods 24 2.2.1 Study site 24 2.2.2 Field methodology 27 2.2.3 Data analysis 29 iii 2.3 Results 30 2.3.1 Seagrass density 30 2.3.2 Seagrass size 33 2.3.3 Seagrass biomass 38 2.3.4 Seagrass reproduction 41 2.3.5 Physical parameters 43 2.4 Discussion 45 2.4.1 Temporal patterns 45 2.4.2 Spatial patterns 52 2.4.3 Interannual variability 59 2.5 Management Considerations 59 CHAPTER 3 Seagrass Physiology: Chlorophyll and Carbohydrates 63 3.1 Introduction 63 3.2 Methods 68 3.2.1 Carbohydrates 68 3.2.2 Chlorophyll 69 3.2.3 Data analysis 70 3.3 Results 70 3.3.1 Carbohydrates 70 3.3.2 Chlorophyll 73 3.4 Discussion 75 3.4.1 Carbohydrates 75 3.4.2 Chlorophyll 81 3.5 Management Considerations 83 CHAPTER 4 Population Dynamics of the Ghost Shrimp Neotrypaea calif orniensis....%5 AA Introduction 85 4.2 Methods '. 89 4.2.1 Study site 89 4.2.2 Sampling methodology 90 4.2.3 Data analysis 92 4.3 Results 93 4.3.1 Ghost shrimp distribution in Grice Bay 93 4.3.2 Ghost shrimp population structure 95 iv 4.3.3 Ghost shrimp reproductive potential .98 4.3.4 Ghost shrimp and other infauna 100 4.3.5 Burrow counts 101 4.3.6 Physical factors affecting distribution 103 4.4 Discussion 106 4.4.1 Ghost shrimp distribution 106 4.4.2 Ghost shrimp population structure 109 4.4.3 Ghost shrimp reproduction 110 4.4.4 Ghost shrimp and other infauna 112 4.4.5 Ghost shrimp burrows 112 4.4.6 Mudflat sediment characteristics 113 4.5 Management Considerations 115 CHAPTER 5 Management recommendations for mudflats and seagrass beds 118 5.1 Seagrass and ghost shrimp as indicators 118 5.2 Importance of ghost shrimp and seagrass in Grice Bay 120 5.3 Future monitoring in Grice Bay 121 Literature Cited 125 Appendix 1: Seagrass population mean values 137 Appendix 2: Seagrass carbohydrate and chlorophyll mean values 138 Appendix 3: Ghost shrimp mean values 140 Appendix 4: Seagrass population statistics 142 Appendix 5: Seagrass carbohydrate and chlorophyll statistics 152 Appendix 6: Ghost shrimp statistics 158 V LIST OF T A B L E S Table 1.1 Summary of criteria for choosing a health indicator species (adapted from Landres et al. 1988; Caro & O'Doherty 1999) . 7 Table 2.1 Summary of sites with GPS locations, mean elevation (or depth) relative to mean lower low water (Canadian Chart Datum), tidal level, length of transects and type ofbed 27 Table 2.2 Summary of sampling dates for each site 28 Table 2.3 Abiotic differences between strata and sites in elevation, mean sediment grain size and maximum current velocity (Yakimishyn unpublished 2002). Positive numbers in first column (elevations) are intertidal sites, while negative numbers (depths) are subtidal strata, absolute numbers are relative to mean lower low water. Error value for mean sediment grain size is one standard error of the mean 45 V I LIST OF FIGURES Figure 1.1 Map of Grice Bay within Pacific Rim National Park Reserve (dotted line) in Clayoquot Sound on the west coast of Vancouver Island, British Columbia 10 Figure 2.1 Grice Bay and Lemmens Inlet seagrass sampling sites, in Clayoquot Sound, on the west coast of Vancouver Island, British Columbia. Dotted line denotes Pacific Rim National Park Reserve boundary 26 Figure 2.2 Seagrass shoot density (number of shoots per m 2) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean 31 Figure 2.3 Seagrass shoot density (number of shoots per m 2) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean...32 Figure 2.4 Seagrass shoot length (cm) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean 34 Figure 2.5 Seagrass shoot width (cm) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean 35 Figure 2.6 Seagrass shoot length (cm) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean 36 Figure 2.7 Seagrass shoot width (cm) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean 36 Figure 2.8 Seagrass shoot biomass (g dry weight per m 2) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean 39 Figure 2.9 Seagrass shoot biomass (g dry weight per m 2) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean 40 Figure 2.10 Seagrass reproductive frequency (percentage of total shoots which are reproductive) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean 42 Figure 2.11 Seagrass reproductive frequency (percentage of total shoots which are reproductive) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean 43 V l l Figure 2.12 Mean monthly air temperature (a) and total precipitation (b) at Tofmo airport (adjacent to Grice Bay) over the course of the study relative to the 30 year normal 44 Figure 3.1 Soluble sugar (solid lines) and starch (dotted lines) concentrations in Zostera marina shoots (triangles) and rhizomes (squares) in April and June 2002. Error bars are one standard error of the mean 71 Figure 3.2 Soluble sugar concentration in seagrass rhizomes at the five intertidal study sites in April and June 2002. Error bars are one standard error of the mean 73 Figure 3.3 Total chlorophyll (a + b) concentrations (mg/g dry weight) in seagrass shoots at 5 intertidal sites over summer 2002. Error bars are one standard error of the mean....74 Figure 3.4 Ratio of seagrass chlorophyll a:b over summer 2002 at five intertidal beds. Error bars are one standard error of the mean 74 Figure 4.1 Outline of Grice Bay showing the two replicate sites in the mudflat, two transects in each site and one sampling position in each of the six zones on each transect. The transects were 900 m long 90 Figure 4.2 Mean density and biomass of all Neotrypaea californiensis individuals in Site 1 and 2 in each of the sampling dates. Error bars are one standard error of the mean 94 Figure 4.3 Mean density and biomass of all Neotrypaea californiensis individuals in both sites across the six zones. Error bars are one standard error of the mean 94 Figure 4.4 Mean total length of all Neotrypaea californiensis individuals in Site 1 and Site 2 by sampling date (a) and by zone (b). Error bars are one standard error of the mean 95 Figure 4.5 Seasonal trends of adult and juvenile Neotrypaea californiensis population densities (# ind / m 3) at site 1 (a) and site 2 (b) in Grice Bay. Error bars are one standard error of the mean 96 Figure 4.6 Spatial trends of adult and juvenile Neotrypaea californiensis density (# ind / m 3) across the 6 zones at site 1 (a) and site 2 (b) in Grice Bay. Error bars are one standard error of the mean 97 Figure 4.7 The proportion of the Neotrypaea californiensis population that is female at each site and variation over time (a) and space (b). Error bars are one standard error of the mean 98 Figure 4.8 Proportion of Neotrypaea californiensis females carrying eggs at the two study sites throughout the sampling period from May 2001 to August 2002. Error bars are one standard error of the mean 99 viii Figure 4.9 Proportion of Neotrypaea californiensis females carrying eggs in the six zones at the two sites. Error bars are one standard error of the mean 99 Figure 4.10 Cryptomya californica density (# ind / m 3) compared to adult and total ghost shrimp density across the six zones in the two sites. Error bars are one standard error of the mean 100 Figure 4.11 Regression of Cryptomya californica density on Neotrypaea californiensis density [adult (a) and all individuals (b)], in June and August 2002 100 Figure 4.12 Regression of ghost shrimp density and biomass on burrow density using data from all dates combined (a) and summer values only (b) 102 Figure 4.13 Regression of burrow density and ghost shrimp density and biomass using the more precise methodology of couting the burrow holes of each individual core in June (a) and August (b) 2002 103 Figure 4.14 Mean sediment grain size (mm) of the top 10 cm at the midpoints of the zones on two transects per site (dotted line, left axis) and mean sediment depth (cm) of these cores (solid line, right axis) across the size zones in both sites. Sediment depth refers to the depth of ghost shrimp activity, below which sediment is very compact suggesting that ghost shrimp are not burrowing deeper or not able to burrow deeper. Error bars are one standard error of the mean 104 Figure 4.15 Regression of ghost shrimp density and biomass on sediment grain size...104 Figure 4.16 Regression of ghost shrimp density and biomass on core depth (assumed to be the depth penetrable by the ghost shrimp) 105 Figure 4.17 Ghost shrimp density and biomass regressed on mudflat elevation. Ghost shrimp density has a significant negative relationship with mudflat elevation until 1.2 m above M L L W and a significant positive relationship with elevation below 1.2 m above M L L W . Ghost shrimp biomass has a significant negative relationship with elevation until 0.8 m above M L L W and no relationship existed deeper than 0.8 m 106 ix A C K N O W L E D G E M E N T S I could never have done this project alone. Thanks most to Rob DeWreede for getting me started on my Masters, for offering help, support and frequent sarcastic comments along the way and for seeing me through to the end. Thanks to Tom Tomascik for supporting me financially, but most importantly for all the help in the field and for putting up with all my questions and concerns. Thanks to Jenn Yakimishyn for all the seagrass talk, the cold diving and for being a good friend for the past two years. Thanks to everyone else who helped me with my field work including Jenn Yamagishi, Emily McGiffin, Healther Holmes, Pete Clarkson, Rick Holmes and all the Parks Canada staff and wardens at Pacific Rim National Park Reserve and at the Western Canada Service Centre in Vancouver. Thanks to the Botany Department at U B C and especially the people in my lab for this experience. Thanks to my family, who wil l be with me forever and to my friends, who come and go, but keep me sane. This project was funded by Parks Canada, with a micro-grant from Project A W A R E . x C H A P T E R 1 M A N A G E M E N T O F MUDFLATS AND SEAGRASS BEDS 1.1 INTRODUCTION British Columbia supports some of the most diverse marine ecosystems in the temperate zone. However, anthropogenic pressures have changed this environment, sometimes drastically and potentially irreversibly. Due to growing human concern about the health of the environment, many studies have been implemented aimed at assessing the current biological components of coastal ecosystems. These studies are aimed at providing a starting point for monitoring programs, detecting disturbances at an early stage and being able to distinguish such disturbances from natural variation. This project has been initiated and funded by Parks Canada and is part of a larger interdisciplinary study to assess the cumulative human impacts in Grice Bay, Pacific Rim National Park Reserve, British Columbia. This research will enable Parks Canada managers to start implementing adaptive ecosystem-based management for the marine environments encompassed by National Parks and National Marine Conservation Areas. 1.2 MONITORING PROGRAMS 1.2.1 Utility of monitoring programs As humans continue to consume non-renewable natural resources, ecologically sustainable use of these resources through ecosystem-based management becomes necessary. Actions taken (or not taken) in ecosystem, community, population and even species management must be done using science-based knowledge. Often management 1 decisions must be made regardless of scientific uncertainty and in such cases the management plan should be generated with a monitoring program designed to reduce critical gaps in the data (Allison et al. 1998). As monitoring proceeds, knowledge of the system increases as new data are analyzed and management plans are revised accordingly (Johnson 1999). Detailed monitoring programs are also needed to establish whether natural changes are taking place in order for human induced changes to be identified. Long-term monitoring has shown us that natural systems are dynamic and need to be managed as such (Halvorson 1996). In the realm of protected areas, research still needs to be done to determine the success and potential benefits of marine protected areas (Sumaila et al. 2000). When ecosystems have already been degraded due to lack of protection, the knowledge gained from sustained ecosystem level monitoring allows for assessment of causes as well as symptoms (Halvorson 1996). This allows for management plans to consider restoration, which usually involves re-introducing natural biota. In these cases, management should be adaptive and flexible and include a long-term monitoring program that should continue at least until the annual variation in the flora and fauna of the created habitat is similar to nearby natural systems (Dawe et al. 2000). Seagrass restoration projects have supported the use of long-term monitoring programs for the evaluation of recovery, as planted beds may display an initial similarity to natural beds followed by a decline in similarity (Fonseca et al. 1996). Commercial use of land and freshwater resources now requires extensive environmental impact studies that are carefully regulated. Continued monitoring is required and all data are readily accessible to the public (Dayton 1998). Monitoring 2 programs are thus viewed to be critical for protected areas, habitat restoration and environmental impact assessments to assure proper management decisions are implemented. 1.2.2 Economics of monitoring programs While monitoring programs are considered very important, funding sources may not be present to get the desired monitoring done. Monitoring programs need to be designed and periodically assessed by scientists. Research, fieldwork and even the writing of reports can be done by non-scientists when protocols are well established and written as permanent documents. Monitoring programs can also be made efficient by running a proper pilot study to determine the important parameters to measure, the best sampling protocol and the sample size needed to detect significant changes. Monitoring may be more effective at a certain time of year then another or monitoring may have to be conducted seasonally. Data analysis and reporting procedures should be established so that analysing data and writing reports becomes routine and nothing is omitted accidentally. Results should be shared with the public and when possible results should be presented to the scientific community for peer review (Halvorson 1996). Monitoring generally does not give information on cause and effect, but rather on trends and changes. Therefore, it is often necessary to supplement monitoring with experimentally based research or information gathered outside the system (Halvorson 1996). As each area is unique, each monitoring program needs to be designed specifically for the system and goals of study. Managers and scientists must make a commitment to the monitoring program, as an uninterrupted program will lead to the 3 most consistent and accurate procedure, sampling protocol, analysis and reporting (Halvorson 1996). 1.2.3 Goal of monitoring programs In order for monitoring to be useful, the specific goals of the monitoring program need to be clearly defined. Goals range from the successful reestablishment of a given species to the preservation of ecosystems that are unimpaired for future generations. While the latter may be a difficult goal to attain, this is the conservation mandate of Parks Canada and the goal of this research. In this study, Parks Canada is looking for a way to monitor Ecological Integrity. As defined by Parks Canada (Canadian Heritage Parks Canada 2001) "an ecosystem has integrity when it is deemed characteristic for its natural region, including the composition and abundance of native species and biological communities, rates of change and supporting processes". In other words, the term describes ecosystems that are self-sustaining and self-regulating, with complete food webs, a full complement of native species and all natural ecological processes, such as energy flow, nutrient and water cycles (Canadian Heritage Parks Canada 2001). 1.3 THE INDICATOR SPECIES CONCEPT 1.3.1 Types of indicators To be completely aware of all impacts humans have on the environment we would need to monitor every species, population and community on earth. As this is not feasible, the concept of monitoring species that may be affected by environmental changes has gained momentum. In conservation biology, there are different types of 4 indicators, namely, population indicators, biodiversity indicators and health indicators (Caro & O'Doherty 1999). A population indicator is a species that indicates population trends in another species. For example, one would look at a prey population to infer something about a predator population. A biodiversity indicator is the number of species in a well-known taxonomic group used as a surrogate for the total number of species. For example, the number of bivalves could be indicative of the number of species in sympatry. Finally, there are two types of health indicators. The first accumulates a pollutant and by measuring the level of the pollutant in the organism, we get an idea of the level of pollutant in the environment. The second type indicates overall environmental quality, wherein a change in one or more population parameters of the species represent changes in habitat quality. While we can usually measure environmental quality directly, the benefit of using a health indicator is that in one step we are able to detect declines in environmental quality and illustrate how these changes are effecting a biological organism. Therefore, to correctly choose an indicator species, it is necessary that the goals of the monitoring project be clearly defined at the outset. 1.3.2 Criteria for choosing health indicators Health indicators are used to represent change in environmental quality and there are criteria in place for choosing them (Table 1.1). The most important criterion is that the health indicator is sensitive to changes in environmental quality. If deterioration in environmental quality has no effect on an organism, monitoring its population will not be informative. Sensitivity should be related to habitat by cause and effect, not merely by correlation and should be towards controllable habitat attributes. The indicator species 5 should also have low levels of natural variability, at least in terms of the parameters being monitored, so that when these population parameters change, it is detectable. Therefore, while the indicator should be sensitive to environmental quality, the changes should be predictable with low levels of error. Population parameters with predictable seasonal variation may be good indices as they are sensitive to changing temperature, light and other abiotic conditions over the year, however interannual variability and variability within seasons should be small. Indicator species should have a well-known biology, so that it may be hypothesized why changes are occurring in the population. From these hypotheses, the appropriate tests can be conducted to identify and fix the problem. Indicator species should be permanent residents to give information about the specific environment being studied. If indicator species are migratory, they may indicate that something is wrong on a large scale, however it would be difficult to pinpoint the precise problem. Indicator species should also consist of large populations, which are easily sampled. Monitoring rare species may be important to learn about rare species, but it is unlikely to give the population information needed to deduce environmental trends (Landres et al. 1988). There is no clear consensus on all criteria. Ongoing debate concerns whether the indicator species should be physically small or large. Small species may be more sensitive to environmental conditions, due to short generation times and high metabolic rates. However, small species may evolve to become less sensitive. In this case, large species, with stable populations, may be better indicators. Other points of debate include i f indicator species should be specialist or generalist and mobile or sessile. Sessile specialists may be good indicators to identify a specific problem at a specific site, while 6 mobile generalists will indicate large changes and environmental trends. Overall, it is recognized that there are benefits to both sides of each of these dichotomies and the correct choice lies in the objective of the monitoring project and what change in environmental quality is hoped to be detected. This also highlights the likelihood that there will never be one universal indicator and best results may be achieved by using a combination of indicators and approaches. Table 1.1 Summary of criteria for choosing a health indicator species (adapted from Landres et al. 1988; Caro & O'Doherty 1999). A H E A L T H INDICATOR should be: • SENSITIVE to human disturbance • Low V A R I A B I L I T Y in response; predictable natural variability is good • Well known biology • Permanent resident • Easily sampled or observed - abundant, cost effective • Small • Large • More sensitive • Stable population • Short generation times • Monitor long term effects • High metabolic rates • Monitor short term effects • Mobile • Sessile • Provide info on wide area • Pinpoint location of pollutant • Generalist • Specialist • More abundant • More sensitive 7 1.3.3 Problems with indicator species While the benefits of using indicators species have been argued, some caution should be used in their interpretation. Natural variation in the indicator population and conditions such as reproductive success, predator populations, seasonal fluctuations in resources and abiotic conditions need to be considered when changes are found before concluding that the population is subject to human influences. Furthermore, since many external and internal variables may contribute to the measured parameter, the affect of each variable will be obscured (Noss 1990). Incorrectly assuming that other species are receiving protection when an indicator species is protected or falsely believing all is well may be serious problems when using indicator species. Indicator species from one area may not be appropriate for use in another area and it is advisable to monitor both habitat and population variables (Noss 1990). It must also be remembered that no single species can be used as a substitute for all other research and monitoring and that there is a wide gap between indicating something and showing cause and effect (Soule 1988). 1.4 E C O S Y S T E M OF I N T E R E S T 1.4.1 Study site The general purpose of this study is to determine the baseline parameters of the dominant flora and fauna of Grice Bay. The two main communities of Grice Bay are the seagrass beds composed ofZostera marina and associated algae, epifauna and infauna and the mudflats supporting primarily the burrowing ghost shrimp, Neotrypaea californiensis. These two habitats are often found adjacent to each other, but are always very distinct, showing very little overlap. The ghost shrimp have difficulty burrowing in 8 between the roots and rhizomes of the seagrass and seagrass seeds and shoots are buried in the mudflat by the constant reworking of sediment by the ghost shrimp. Consequently these two habitats are mutually exclusive (Suchanek 1983). The study site is Grice Bay (49° 06.' N , 125° 46' W), which is contained in the Long Beach Unit of the Pacific Rim National Park Reserve in Clayoquot Sound, on the West Coast of Vancouver Island, British Columbia (Figure 1.1). This Unit is the best known and most accessible part of the Park. It is located between the communities of Ucluelet (on Barkley Sound) and Tofino (on Clayoquot Sound). Before 1970, the only access to Long Beach was via a 110 km gravel road from Port Alberni and was therefore relatively isolated. Presently, Long Beach is visited by 450 000 tourists annually with approximately 80 % arriving between late May and early October (MacFarlane et al. 1996). Grice Bay is at the southeast end of Tofino Inlet and is reached through a 10 km long, narrow inland passage from the open sea at Tofino (Darling et al. 1998). The bay is shallow and dries almost completely at low tide exposing extensive high intertidal mud flats (80 % of the bay) and low intertidal seagrass beds (20 % of the bay). 9 British Columbia C A N A D A Clayoquot Sound & Pacific Rim National Park Reserve 80 km Pacific Ocean 2 km Figure 1.1 Map of Grice Bay within Pacific Rim National Park Reserve (dotted line) in Clayoquot Sound on the west coast of Vancouver Island, British Columbia. 10 Currently Grice Bay remains relatively unimpacted, however there are potential threats in the area, which might drastically affect marine communities in Clayoquot Sound in the future. These potential threats include sedimentation due to deforestation, salmon and oyster aquaculture, fish and crab fisheries and tourism activities such as whale watching. In general, the increase in development due to the large tourism industry in Tofino will increase usage of Grice Bay. This research project was in part initiated in response to a proposal to upgrade the Tofino airport, which is situated in the park adjacent to Grice Bay. The proposed development may open wind channels, which would alter the current regimes of the bay and may increase runoff from airport facilities. It is therefore important to have a good understanding of the population and community processes that are threatened by such human activity in order to manage this ecosystem and maintain its future viability. This research will provide Parks Canada planners and managers with baseline information on the dominant primary produces and detritivore in the bay, which will allow Parks Canada to start implementing ecosystem-based planning and management as defined in the National Marine Conservation Areas Act (Government of Canada 2002). 1.4.2 General biology of soft bottom habitats Grice Bay is a soft bottom habitat where life, as in the water column, is defined in three dimensions. The infauna of such habitats live buried in the sand or mud, emerging only very rarely. For infauna, the ability to move around within the sediment enables then to avoid direct competition with neighbours and escape predation by benthic feeders or other burrowers. These soft shore environments are also home to mobile epifauna, 11 such as crabs and fish, which move in and out with the tide and feed in the shallow-water habitats. Plants such as seagrasses need to root firmly in the sediment, but are always subject to a shifting substrate. Unrooted plants such as algae may survive as drift algae, by colonizing occasional rocks and logs or attaching to the more firmly rooted seagrass. The distribution of organisms in soft bottom environments was originally thought to be controlled by sediment size and stability. Although the picture now appears to be much more complex, the nature of the sediment is still considered an important factor for infaunal organisms (Little 2000). As the tide changes, the sediment acts as a buffer against simultaneous changes in salinity, temperature and pH. Organic material accumulates in the sediments creating a food source. Finer sediments, such as mud, usually retain water at low tide and desiccation is less of a problem than in coarser sediments such as sand. Mudflats are far from static entities and represent areas of changing balance between erosion and deposition. The unstable nature of the sediment requires organisms to be physiologically flexible in this environment. The inhabiting organisms play a role in altering their habitat. Diatoms and seagrasses can be effective in binding mud particles together, while burrowing infauna have the opposite effect and tend to destabilize the sediment with their burrowing and feeding activities (Little 2000). 1.4.3 Seagrass ecology Seagrasses are the only angiosperms adapted for life submerged in the marine environment and consist of more than 50 species. They are found intertidally and in the shallow subtidal on many of the world's temperate and tropical coastlines. The 12 ecological importance of seagrass habitats has long been recognized as they represent sites of high primary productivity (McRoy & McMillan 1977). Seagrass beds also serve as feeding and nursery habitat for many economically and ecologically important fish and invertebrate species and as feeding areas for birds (Thayer et al. 1975, Murphey & Fonseca 1995). Seagrasses regulate water column dissolved oxygen, modify their physical and chemical environments and reduce suspended sediments, chlorophyll and nutrients in the water column. Seagrass roots and rhizome systems bind and stabilize bottom sediments and seagrass leaves baffle currents (Fonseca 1989, Fonseca et al. 1982). Seagrasses may also improve water quality by filtering suspended matter, however excess loading of nutrients or suspended material wil l cause the degradation of seagrass beds (Short & Short 1984). Seagrass beds are considered fragile because of their physical morphology and growth patterns and their occurrence in coastal waters that are subject to frequent environmental disturbances (Frost et al. 1999). Both natural and anthropogenic disturbances that alter water quality and clarity are responsible for the loss and fragmentation of seagrass habitats (Short & Wyllie-Echeverria 1996, Frost et al. 1999). Changes in sea level, salinity, temperature, atmospheric CO2 and U V radiation can alter seagrass distribution, productivity and community composition. Changes in the seagrass distribution and structure may have profound implications for other flora and fauna, nearshore geomorphology and biogeochemical cycles (Short & Neckles 1999). The common seagrasses found in sheltered soft shore environments along the coast of British Columbia, are the eelgrasses, Zostera marina and Zostera japonica (Harbo 1999). Z. marina is found from Alaska to Mexico and is the dominant seagrass 13 species on both coasts of North America, Japan and Europe (Orth et al. 1994). Blades may grow upwards of 2 m long and 10 mm wide in muddy or sandy sediments in the low intertidal to subtidal. Blades have 1-7 longitudinal vein(s) and are dull green in colour. Z. marina is perennial and overwinters as leafy shoots. Z. japonica was introduced to North America from Japan and now grows from British Columbia to Oregon in muddy bays in the mid-intertidal (Harrison 1982b). Its leaves only grow up to 20 cm long and less than 3 mm wide, with a leaf shield composed of a complete tube encircling inner leaves. It is an annual and overwinters as buried seeds, which germinate in the spring (Harbo 1999). While Zostera japonica has invaded Clayoquot Sound, it is not considered a competitor with the native Z. marina, primarily because it typically invades the zone higher in the intertidal that previously lacked permanent macrophyte cover (Posey 1988). In Grice Bay, this higher intertidal region is dominated by ghost shrimp that loosen the sediment such that it is unfavourable for colonization by Z. japonica, which is therefore not found in the bay. This study only focuses on Z. marina population parameters and its use as an indicator species. 1.4.4 Ghost shrimp ecology One of the most abundant invertebrates of the Grice Bay mudflats is the thalassinidean ghost shrimp, Neotrypaea californiensis (formerly Callianassa californiensis; Manning & Felder 1991). Early descriptions of this species by MacGinitie (1934) reported N. californiensis to be one of the most abundant animals in the muddy shores of sheltered marine environments on the West Coast of North America, ranging from Alaska to Baja California (MacGinitie 1934). N. californiensis colour varies 14 between orange, yellow and pink, individuals may reach a maximum total adult length of 7 cm and have a lifespan of approximately 3 to 5 years (Bird 1982). In Northern Pacific estuaries of North America, dense N. californiensis beds are generally restricted to the high and mid intertidal (L.Thompson & Pritchard 1969; Swinbanks & Murray 1981) on mixed sand or mud shores of a sufficient tenacity to allow the construction of burrows (Stevens 1929; MacGinitie 1934). Ghost shrimp rely on temporary burrows for shelter, feeding and reproduction and except for a brief pelagic larval stage, most spend their entire life within the burrows (Griffis & Chavez 1988). Burrows alter soft sediment environments by increasing the surface area of sediments exposed to oxygenated water (Griffis & Suchanek 1991). The burrowing activities of deposit feeders have dramatic effects on the burial and excavation of organic materials, as well as the transport and mixing of sediment grain size (Griffis & Suchanek 1991). Construction and maintenance of burrows by thalassinidean shrimp has specifically been linked to continuous mixing of deep and shallow layers of sediment resulting in substantial sediment resuspension and transport and changes in organic content and grain size of sediment (Jones & Jago 1993). Thayer (1979) argued that biological bulldozing, resulting from ploughing through the sediment while burrowing or feeding, was the most effective form of bioturbation and may cause the evolution towards increased dominance of mobile deposit feeders in soft shore environment. A strong negative correlation between dense seagrass beds and dense ghost shrimp patches has often been reported in the literature, although these communities often lie adjacent to one another. The cause and effect relationship in this case is not so clear, as in some instances it seems that the burrowing shrimp may be excluding seagrass 15 (Suchanek 1983), while in other cases the inverse seems plausible (Harrison 1987). Burrowing ghost shrimp eject sediments, which may reduce light available for photosynthesis or physically bury the seagrass seedling or seed, thereby eliminating it from regions of abundant ghost shrimp. Conversely, a healthy stand of the seagrass, with its mat of rhizomes, could potentially inhibit invasion by ghost shrimp by inhibiting burrowing (Brenchley 1981). 1.5 STUDY OBJECTIVES This project is part of a larger interdisciplinary study within the Pacific Rim National Park Reserve, focused on addressing issues related to cumulative human impacts. The Grice Bay Ecosystem Study was designed to address the following Parks Canada National Marine Conservation Areas policy objectives (Canadian Heritage Parks Canada 1994): 1. Ecosystem management must be based in science. 2. In promoting ecosystem-based management, Parks Canada will work with others in compiling and analyzing baseline information about the physical, oceanographic and biological characteristics, and existing and potential uses of the conservation area. 3. Developing and implementing monitoring programs. 4. Building and strengthening internal science and research capacity. 5. Developing new linkages and strengthening existing science and research partnerships with universities, NGOs and industry. 6. Identification of science and research needs for developing and implementing and ecosystem-based management approach. 16 7. Linking science with policy, planning and management. The objective of this thesis was to obtain baseline data on the seagrass and ghost shrimp populations in Grice Bay and nearby habitats to develop suggestions for future monitoring. Seagrass beds and mudflats are important habitats for all organisms inhabiting or migrating through the sheltered coastal environment of bays and estuaries. These communities must be protected in order to maintain the food web linkages, biotic interactions and population viability, essential to preserving the Ecosystem Integrity of Grice Bay. These baseline data wil l serve as a reference for population parameter means and variance. Data were analyzed on a temporal scale to determine i f seasonal patterns emerge and to identify optimal times for monitoring. Interannual variability was assessed by comparing the summers of 2001 and 2002. In some cases, different sampling protocols were assayed to determine the most effective sampling technique. The specific objectives of this study were to: a. Establish the phenology of seagrass in the area. Obtain a quantitative and qualitative assessment of the structure (biomass, density, depth range) and function (productivity) of intertidal and subtidal seagrass beds. b. Conduct quantitative and qualitative assessments of ghost shrimp population parameters (density, biomass, density to burrow ratio, size class distribution, sex ratio and reproduction). c. Develop a monitoring protocol based on an evaluation of the techniques used above. 17 To fulfill these objectives, the following hypotheses were tested to quantify the spatial and temporal variation in the seagrass and ghost shrimp populations at Grice Bay. a. Seagrass: H i : Zostera marina productivity varies spatially with deeper seagrass strata showing decreased density, size and biomass along with increased total chlorophyll and decreased chlorophyll a:b ratios and carbohydrate stores, relative to shallower beds. H2: Growth and reproduction in Z. marina varies temporally with the majority of growth and seed production occurring in summer. H 3 : Results can be generalized across different beds in Clayoquot Sound. b. Ghost shrimp: H i : Neotrypaea californiensis distribution varies spatially, with adults being most abundant in shallow, near shore strata and juveniles dominate in the deeper strata. H2: N. californiensis reproduction and distribution varies temporally, following seasonal cycles. H 3 : Ghost shrimp density can be expressed as a function of burrow density in order to predict the number of ghost shrimp from the number of burrow holes. Results from both the seagrass and ghost shrimp work were compared with studies in the literature to determine i f the trends in Clayoquot Sound can be generalized when compared to studies on the same species along the rest of the coast and to different species of the same genera around the world. 18 1.6 OUTLINE OF THESIS The first chapter of the thesis was intended to give background information on monitoring and using organisms as indicators of Ecosystem Integrity. The study system and specific study site were presented and information was given as to why the seagrass, Zostera marina and ghost shrimp, Neotrypaea californiensis were chosen as indicator species. The focus of Chapter 2, will be on the distribution of Z. marina. Temporal and spatial patterns in density, length, biomass and reproduction will be presented, as well as correlations with these factors to the abiotic conditions. Future use of Z. marina as an indicator species will be presented with respect to the best parameters to monitor and how to distinguish human-induced and natural variation. Chapter 3 will focus on the physiology of Z. marina in terms of sugar and chlorophyll concentration. Five intertidal sites were sampled to investigate spatial variability and changes over the summer. It wil l be discussed how these physiological parameters relate to primary production in the seagrass beds and their usefulness as early warning signals. Chapter 4 shifts the focus from Z. marina to N. californiensis. Ghost shrimp density and biomass wil l be analysed on a spatial and temporal scale. Reproductive potential of the ghost shrimp population in Grice Bay will be presented. Correlations between these biological parameters and abiotic parameters such as mudflat elevation, sediment depth and sediment grain size will be discussed. The possibility of using burrow counts as a rapid assessment technique of ghost shrimp density will be evaluated and future monitoring protocols wil l be discussed. At the end of each of these three chapters (2 - 4), there is a comprehensive section focusing on management concerns surrounding the indicator species in question and the parameters measured in the given chapter. Finally, Chapter 5 will be an overall summary 19 of the population biology and management considerations for both Z. marina and N. californiensis. This chapter will illustrate how the seagrass and ghost shrimp studies compliment each other, providing two indicator species for future monitoring and management conducted by Parks Canada and other concerned non-governmental organizations. 20 C H A P T E R 2 SPATIAL AND T E M P O R A L VARIABILITY IN ZOSTERA MARINA BEDS IN C L A Y O Q U O T SOUND 2.1 INTRODUCTION Seagrasses have begun to gain recognition by the public as well as the scientific community as important primary producers, habitats and indicator species. As seagrass populations continue to decline worldwide, baseline studies estimating population parameters and monitoring programs have become prevalent. However, such studies are often done by governmental organizations and their findings become part of the gray literature and are not easy to access. These studies are also done by many different stakeholders in Canada, including universities, non-governmental organizations and different government departments, such as the Department of Fisheries and Oceans and Parks Canada. Although research has been published regarding seagrass populations and communities along the Pacific Coast of North America, none of these studies have been conducted on the West Coast of Vancouver Island, British Columbia. Seagrass communities in Clayoquot Sound are potentially affected by sedimentation due to deforestation, salmon and oyster aquaculture, fish and crab fisheries and boating traffic. In general, the increase in development due to the large tourism industry in the area will increase usage of the sound and likely increase eutrophication. Currently the area is thought to have remained in an undisturbed state, however the cumulative impacts of all these activities together may have serious adverse affects on the bay in the future. It is therefore important to obtain a good understanding of the biological processes that may be threatened by such human activity, to manage this 21 ecosystem and maintain its ecological integrity in the future. Seagrass communities are declining worldwide, which has been attributed to widespread deterioration of water quality. Industrialization and increased land use has resulted in increases in nutrient loading, sedimentation, influx of contaminants and toxins and other detrimental effects on these sensitive communities (Orth et al. 1994; Abal & Dennison 1996). The widespread distribution and ecological importance of seagrasses and their sensitivity to water quality parameters have led to the use of these plants as biological indicators of water quality (Abal & Dennison 1996). Monitoring changes in seagrass depth range over time has been proposed as a useful tool for predicting the ecological health of a marine system and detecting non-steady-state conditions (Abal & Dennison 1996). Restoration efforts in Chesapeake Bay have emphasized reversing the declining water quality trend and have focused on using seagrass recolonization and establishment as an indicator of habitat quality (Dennison et al. 1993). In order for a species to be of value as an indicator of ecosystem integrity, it must be sensitive to changes in environmental conditions and knowledge of the natural variability of the parameters to be monitored is necessary. An easy way to assess natural sensitivity is to determine i f the population parameters of interest change seasonally. Human effects are often on environmental factors that vary naturally (temperature, light availability), however human activities may significantly change seasonal patterns, means or maximum and minimum values. Temporal variability is likely important at high latitudes where rates of photosynthesis, growth, biomass turnover and reproduction 1 Ecological integrity was defined in Chapter 1 as ecosystems with intact native components (plants, animals and other organisms) and processes (such as growth and reproduction). 22 may vary seasonally. While predictable variation is important for an indicator species, variability between years or within seasons will make identifying changes due to human disturbances difficult. Inter-annual variation is more difficult to assess without a long-term study, however changes from one summer to the next correlated to environmental factors may help predict this sort of variability. Seagrass beds also differ spatially and connected areas may have areas of high or low density and patchy or non-patchy distribution. The primary reason often cited is tidal height, with intertidal and subtidal beds having very different environmental influences. The effects of tides on seagrass are due primarily to light availability, accounting for the overall light regime of the habitat over the entire tide cycle. Seagrasses are completely dependent on light, therefore increased turbidity in the water column may reduce growth rates with a noticeable impact on density and biomass. Alternatively, intertidal seagrass beds that may look the same at a casual glance may also vary in important ways. Subtle differences in currents, waves and tidal levels between intertidal beds will distinguish these beds. These abiotic factors will in turn affect the substratum composition of the beds, including sediment grain size, organic content of the sediment and water retention as the tide goes out. Temperature and salinity regimes may vary on even the smallest scale between beds and within beds, affecting productivity, growth and the onset and success of sexual reproduction. An important component of a baseline study is to assess the primary productivity of a community. Density and biomass measurements compared between strata at one bed and between different intertidal beds will be discussed in this chapter, along with correlations to physical properties. Physiological parameters of Zostera marina such as 23 sugar and chlorophyll content will be analyzed in Chapter 3. Overall, these measurements will provide an indication of the health of these seagrass beds. The discussion will focus on which parameters of seagrass beds would be best to sample, how often sampling should occur and at what time of year. Expected values and natural variability will be discussed so that changes due human-induced disturbances can be identified. The specific questions addressed in this chapter are as follows: 1. Is there a predictable seasonal pattern of Zostera marina biomass, density, size and reproduction? 2. Do population parameters of Z. marina vary predictably over space in terms of tidal elevation or at different sites around Clayoquot Sound? 3. Does the relationship between physical and biological parameters provide a basis for predictions of Z. marina population structure? 4. Is Z. marina a good indicator species? What are the best Z. marina parameters to monitor? 2.2 M E T H O D S 2.2.1 Study site The study was conducted in Grice Bay and Lemmens Inlet, in Clayoquot Sound, on the West Coast of Vancouver Island, British Columbia (Figure 2.1). Grice Bay was chosen as a study site as it is a component of Pacific Rim National Park Reserve and may be protected from direct human impacts in the future. Grice Bay will always be subjected to water quality impacts imposed on the system from outside the park boundary, therefore implementing a monitoring program is important for Parks Canada. Grice Bay is shallow 24 and dries almost completely at low tide exposing large intertidal seagrass beds, which when submerged are habitat for many fish and invertebrate species. The comparison sites are located in Lemmens Inlet, which is another relatively sheltered body of water surrounded by Meares Island. Lemmens Inlet also has large seagrass beds, but does not have the extensive mudflats characteristic of Grice Bay. Lemmens Inlet has more channels and islands and water current patterns are therefore more complicated. In Grice Bay, Zostera marina covers about 160 ha of the lower intertidal and upper subtidal flats (R. Palm pers.comm.). The lower depth limit of Z. marina is approximately 4 m below M L L W , with most of the subtidal seagrass occurring along the edges of the larger channels and on the bottom of the smaller channels. Tides are mixed semidiurnal with a maximum tidal range of 3 m. Surface salinity is highly variable and ranges from 10 to 28 ppt and water temperature ranges from 5 to 15°C. Both salinity and temperature are higher in summer and lower in winter. The specific sites within Grice Bay and Lemmens Inlet were chosen in pairs. LI 2 in Lemmens Inlet (LI) and GB 2 in Grice Bay (GB) are both small beds in channels, not attached to islands (Figure 2.1). They are both exposed to wide wind corridors. It was therefore considered that these sites might have similar water circulation patterns in terms of drainage and flooding. LI 1 and both GB 1 and GB Main are larger beds, somewhat more sheltered or in the case at GB 1 and LI 1, the dominant bed in the area. The sampling that began in July 2001 was only conducted at GB Main due to the time constraints of sampling four strata. The four strata were: the upper edge of the seagrass (2.5 m), the intertidal (200 m), the slope (20 m) and the channel (20 m). The other four intertidal sites were added to the study in April 2002 to compare different beds. Only 25 four more sites could be added due to time constraints when working during a low tide (< 0.5 m). Site comparisons are summarized in Table 2.1 below. Despite the attempt to pair sites, the beds are variable, illustrating the difficulty of finding true replicates in the field. Figure 2.1 Grice Bay and Lemmens Inlet seagrass sampling sites, in Clayoquot Sound, on the west coast of Vancouver Island, British Columbia. Dotted line denotes Pacific Rim National Park Reserve boundary. 26 Table 2.1 Summary of sites with GPS locations, mean elevation (or depth) relative to mean lower low water (Canadian Chart Datum), tidal level, length of transects and type of bed. Latitude & Longitude Elevation or depth (cm) Tidal Level Transect length (m) Type of bed GB Main 49° 06.7'N 125° 46.4' W Attached to Island Edge 68 Intertidal 2.5 Intertidal 68 Intertidal 200 Slope -24 to -370 Subtidal 20 Channel -376 Subtidal 20 GB 1 49° 06.8'N 125° 47.1' W 78 Intertidal 100 Center of GB GB 2 49° 07.2' N 125° 47.2'W 50 Intertidal 100 Channel bed LI 1 49° 10.9'N 125° 53.0'W 177 Intertidal 200 Center of LI LI 2 49° 10.2'N 125° 53.4' W 17 Intertidal 50 Channel bed 2.2.2 Field methodology The GB Main Zostera marina bed in Grice Bay was divided into four strata, the upper edge, intertidal, slope and channel. It has been repetitively sampled in July, September and December 2001 and February, May, June and August 2002. The same sampling protocol was used to compare five intertidal beds, including GB Main. Two beds in Grice Bay, GB 1 and GB 2, were added along with two beds in Lemmens Inlet, LI 2 and LI 1. These beds were sampled four times during the summer of 2002, in April , May, June and August. A summary of the date each bed (or strata) was sampled is shown in Table 2.2 27 Table 2.2 Summary of sampling dates for each site. Month G B Main G B 1 G B 2 LI1 L I 2 Edge Intertidal Slope Channel July 22 July 23 & 2 3 30&31 30&31 < 2001 July July July Sept 15 Sept 15 Sept 14& 15 14 & 15 2001 Sept Sept Dec 8 Dec 8 & 9 7 & 9 7 & 9 2001 Dec Dec Dec Feb 21 Feb 22 Feb 21 & 2 2 21 & 2 2 2002 Feb Feb April 2002 28 Apr 28 Apr 29 Apr 26 Apr 27 Apr 27 Apr May 30 May 2 7 & 3 0 3 June 3 June 27 May 28 May 29 May 2 6 & 2 9 2002 May May June 25 June 25 June 2 6 & 2 7 2 6 & 2 7 23 June 23 June 22 June 24 June 2002 June June Aug 10 Aug 10 Aug 12 Aug 12 Aug 9 Aug 9 Aug 8 Aug 8 Aug 2002 At each site (or strata), three non-permanent transects were sampled. Along each transect, eight randomly positioned 50 x 50 cm quadrats were sampled for density with vegetative and reproductive shoots counted separately. Reproductive shoots were identified due to their different physical appearance. Reproductive shoots are longer than vegetative shoots, consisting of a thickened main stem, many leaves and seed containing sheaths. Sub-samples of above ground biomass were collected along each transect, from 10 x 10 cm quadrats located adjacent to five of the eight density quadrats on each transect. In the lab, the plants were frozen until the maximum length of the shoot and the width directly above the sheath was measured. Dead portions of the shoot were removed, as well as the senescing sheath before the number of blades per shoot was assessed. Shoots were dried for 48 hours at 60°C to obtain a dry weight measurement and ashed for 4 hours at 520°C to estimate ash-free dry weight. 28 The Intertidal and Edge strata at GB Main and the other four intertidal sites were sampled at low tide. The exception was during the winter (Sept, Dec & Feb) when low tides occur during the dark, at these times the Edge and Intertidal strata at GB Main had to be sampled by S C U B A during the day. The subtidal Slope and Channel strata were always sampled by SCUBA. Subtidal macrophytes are typically collected by S C U B A divers because they can see whether plants have been missed, they can avoid collecting plants not rooted within the quadrat area and loss rates are very small (Downing & Anderson 1985). Biomass in this study was estimated by multiplying density counts by the mean dry weight per shoot in order to reduce the amount of destructive sampling as much as possible. A more accurate measure of biomass would be to remove and weigh all seagrass from the 50 x 50 cm quadrats, however this would have resulted in much greater destruction. 2.2.3 Data analysis The majority of the data were found to be highly non-normal due to the occurrence of many zeros (absence of seagrass) and infrequent high numbers, especially in the Edge and Channel strata. These highly non-normal data cannot be forced into normality with statistical transformations. Therefore, most of the analysis was conducted using multiple Mann-Whitney tests. Significance was determined by a p value < 0.05. To correct for the increasing chance of erroneous conclusions, the a value was set to 0.05 / # of pairwise tests (Sokal & Rohlf 1981). Temporal patterns were identified from comparing means of density, biomass, shoot length and width between dates, with each strata or site considered separately. Spatial patterns were examined by comparing the 29 means of the same factors between tidal strata and intertidal sites, with each sampling date considered separately. Reproductive data were considered in terms of the time of the reproductive season and the frequency of flowering shoots. Mean values (Appendix 1) and the results of the statistical analysis (Appendix 4) are found in the appendices at the end of the thesis. 2.3 R E S U L T S 2.3.1 Seagrass density a. Temporal variability Seagrass density in the four zones at GB Main was found to vary seasonally (Figure 2.2; Appendix 4.1). Density values were highest in the summer months and dropped in the winter. The greatest seasonal variation was seen in the Intertidal and Slope strata. Slope seagrass density displayed a 4 fold difference between winter and summer density while density in the Intertidal showed a 1.6 fold difference. Seagrass density in the Edge and Channel strata remained relatively low over the entire year. In the Edge stratum, density dropped continuously throughout the study. In the Channel stratum, the opposite trend was found as density increased over the year. While density was initially higher in the Edge than in the Channel stratum, the opposite was true by August 2002, indicating opposite trends were occurring in the two populations. In both strata, density at the end of the study was significantly different than at the beginning. 30 July Sept Dec Feb M a y June A u g 2001-2002 Figure 2.2 S e a g r a s s s h o o t d e n s i t y ( n u m b e r o f s h o o t s p e r m1) i n f o u r t i d a l s t r a ta o v e r the s a m p l i n g p e r i o d f r o m J u l y 2 0 0 1 t o A u g u s t 2 0 0 2 . E r r o r b a r s r e p r e s e n t o n e s t a n d a r d e r r o r o f t he m e a n . O n a s m a l l e r t e m p o r a l s c a l e , d e n s i t y c h a n g e s o c c u r r e d i n t h e 5 i n t e r t i d a l s i tes o v e r s u m m e r 2 0 0 2 ( F i g u r e 2 . 3 ; A p p e n d i x 4 . 1 ) . T h e r e w a s l i t t l e o r n o c h a n g e i n d e n s i t y b e t w e e n A p r i l a n d M a y , e x c e p t at L I 1 w h e r e d e n s i t y i n c r e a s e d 1.5 f o l d d u r i n g t h i s t i m e . B e t w e e n M a y a n d J u n e , d e n s i t y i n c r e a s e d at a b o u t the s a m e ra te i n f o u r o f t h e f i v e s i tes (1 .2 - 1.5 f o l d ) . T h e e x c e p t i o n a g a i n w a s at L I 1 w h e r e , f o l l o w i n g a l a r g e i n c r e a s e i n d e n s i t y b e t w e e n A p r i l a n d M a y , s h o o t d e n s i t y d e c l i n e d n o n - s i g n i f i c a n t l y i n J u n e . B e t w e e n J u n e a n d A u g u s t , s h o o t d e n s i t y e i t h e r s t a y e d the s a m e ( G B 2 a n d L I 1) d e c l i n e d s l i g h t l y ( G B M a i n a n d G B 1) o r d e c l i n e d s i g n i f i c a n t l y ( L I 2 ) . 31 600 i 100 ' Apr i l M a y June Augus t 2002 Figure 2.3 Seagrass shoot density (number of shoots per m 2) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean. b. Spatial variability The four strata were found to have highly different shoot densities from each other (Figure 2.2; Appendix 4.1). The Intertidal had the highest density ranging from a low of 322 ± 15 shoots / m 2 (mean ± S.E.) in winter to 525 ± 31 shoots / m 2 in summer. Shoot density in the Intertidal was significantly greater than all other strata throughout the whole sampling program. Shoot density in the Slope stratum was the next greatest, although density reached a low of 38 ± 5 shoots / m 2 in May 2001. Shoot density on the Slope was significantly greater than in the Edge and Channel strata in the summer months. The five intertidal seagrass beds were significantly different from each other in terms of shoot density (Figure 2.3; Appendix 4.1). GB Main, LI 2 and LI 1 were the high shoot density sites, while GB 2 and GB 1 were the low shoot density sites. The high 32 density sites were more variable over time and order of density, while density and GB 1 & GB 2 remained similarly low over the summer. In April, highest shoot density was found at GB Main (378 ± 30 shoots / m 2), followed by LI 1 and then LI 2. For the rest of the summer, LI 1 had the highest shoot density followed by GB Main and then LI 2. 2.3.2 Seagrass size a. Temporal variability Shoot length was also found to vary seasonally with greatest shoot length occurring during the summer (Figure 2.4; Appendix 4.2). In all strata, shoot length was greatest in September (overall mean length was 53 ± 4 cm), which is at the end of the growing season and prior to stressful winter conditions, which cause carbohydrate stores to be used up. Strata with the longest shoots in September also experienced the greatest change in length between September and December. Shoot length continued to decrease until February and there was either no change or a slight increase in length between February and May. Overall shoot length in the four strata was lowest in February with a mean.length of 18 ± 2 cm. Differences in shoot length between summer and winter values were significant for the Intertidal, Slope and Channel strata. 33 100 July Sept Dec Feb May June A u g 2001 - 2002 Figure 2.4 Seagrass shoot length (cm) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean. The seasonality pattern in shoot width over the sampling year differed between strata, although all strata did display an autumn/winter decrease followed by a spring/summer increase in mean shoot width (Figure 2.5; Appendix 4.2). As with mean length, the largest change was found in the Slope stratum. The largest mean widths of 0.58 ± 0.05 and 0.61 ± 0.04 cm occurred in July 2001 and in August 2002 respectively, compared to the smallest overall mean width of 0.22 ± 0.01 cm in May 2002. Mean shoot width was the smallest during December in the Intertidal stratum, February in the Edge, May on the Slope and July 2001 and June 2002 in the Channel. 34 0.7 July Sept Dec Feb M a y June A u g 2001 - 2002 Figure 2.5 Seagrass shoot width (cm) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean. At the five intertidal sites there was a trend observed from an overall lower mean shoot length in April (40 ± 2 cm) to a greater mean length in August (83 ± 4 cm) (Figure 2.6; Appendix 4.2). A n increase in shoot length occurred between every summer interval at GB Main, LI 1 and LI 2, although the pattern differed greatly with fluctuations between higher and lower shoot length between months. Mean width did not follow a seasonal pattern, as different trends occurred at the five sites. A significant increase in shoot width between April and August was only seen at GB Main and LI 2 (Figure 2.7; Appendix 4.2). Overall mean length and mean width of all the sampling dates at all sites were found to be correlated by the linear equation: mean width (mm) = 0.0041 * mean length (cm) + 0.2942 (p <0.001; r 2 = 0.643). 35 125 0 Apr i l May June A u g 2002 Figure 2.6 Seagrass shoot length (cm) in five intertidal seagrass beds during the summer of 2 0 0 2 . Error bars represent one standard error of the mean. 36 b. Spatial variability Greatest mean seagrass shoot length was found in the Slope stratum followed by the Intertidal stratum during the summer months (Figure 2.4; Appendix 4.2). During the winter, mean shoot lengths were more similar between the four strata and the order varied. In July, mean shoot length in the Slope stratum (77 +14 cm) was 2 times longer then in the Intertidal and 3 to 4 times larger than in the Edge and Channel strata respectively. In September, differences between the strata decreased. From December to June, differences were usually 2 fold or less, although some of these were statistically significant. Differences between the strata became apparent again in August when the Slope (71 ± 5 cm) and Intertidal (67 ± 9 cm) shoots were 1.7 to 2.2 fold longer then the other two strata. Mean width varied significantly between the strata, but not in a consistent manner (Figure 2.5; Appendix 4.2). In July, the Slope had the greatest mean width. Conversely, in May mean seagrass shoot widths on the Slope were the smallest and significantly smaller then those in the Intertidal and Edge strata. Seagrass in the Edge also had the smallest shoot widths in December and largest widths in May and June. Seagrass in GB Main had the lowest overall mean shoot length (40.8 ± 3.2 cm) out of the five sites. Seagrass in GB 2 had the longest shoots (60.1 ±3 .1 cm), followed by GB 1 (59.6 ± 4.2 cm), LI 2 (57.7 ± 4.8 cm) and then LI 1 (48.1 ± 2.2 cm) (Figure 2.6; Appendix 4.2). However, the trend between months is too variable to be predictable at this scale. When compared to the other intertidal sites, seagrass shoots at GB Main had the smallest mean width at 0.46 ± 0.01, while seagrass at GB 1 and LI 2 had the largest overall mean width (both 0.61 ± 0.02 cm) (Figure 2.7; Appendix 4.2). 37 2.3.3 Seagrass biomass a. Temporal variability The estimates of density and size structure of the Zostera marina population given above have yielded much information, however it is often desirable to consolidate these different measures into one estimate of biomass. The most striking seasonal biomass pattern occurs in the Intertidal and Slope strata (Figure 2.8; Appendix 4.3). Biomass began high in July 2001 at 95 ± 19 and 80 ±25 g of dry weight / m 2 in the Intertidal and Slope strata respectively. On the Slope, biomass dropped by 40 fold to 4 ± 1 and 2 ± 0.1 g / m in December and May respectively. A biomass decrease of 7 fold occurred in the intertidal as biomass declined to 13 ± 2 and 15 ± 2 g / m 2 in December and February respectively. Biomass in both the Intertidal and Slope strata was lower in June and August 2002 then July 2001, however differences were not significant in either stratum. The biomass estimates for the Edge and Channel are similar to those found in the density measures. While biomass remains relatively low throughout the year in both strata, the trend towards increasing abundance in the Channel and decreasing abundance in the Edge stratum holds for biomass as well as density. 38 120 July Sept Dec Feb May June A u g 2001 -2002 Figure 2.8 Seagrass shoot biomass (g dry weight per m2) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean. In the 5 intertidal strata, biomass was seen to increase from April to August (Figure 2.9; Appendix 4.3). Mean biomass increased significantly between April and August at GB Main (23 ± 4 to 78 ± 16 g / m2), GB 1 (23 ± 2 to 88 ± 15 g / m 2) and LI 2 (36 ± 15 to 180 ± 32 g / m 2). This increase was 5 fold at LI 2. At GB 2 the trend persisted but was not significant. Conversely, mean biomass at LI 1 followed a different trend, with a biomass peak in June and no significant differences between any of the sampling dates. 39 E 1 -a on 250 200 150 i o o JS B G B Main * GB 1 — « G B 2 — » - L I 1 — i t —LI 2 50 A p r i l M a y June 2002 Augus t Figure 2.9 Seagrass shoot biomass (g dry weight per m 2) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean. b. Spatial variability Biomass differed greatly between the four strata at GB Main throughout the year (Figure 2.8; Appendix 4.3). In July, biomass in the Intertidal was 95 ± 19 g / m 2 , which was 316 times greater then biomass in the Channel. For the rest of the year, including the following summer, the difference in biomass was between 10 and 57 fold. In both summers sampled, biomass in the Intertidal and Slope did not differ significantly from each other, but were both greater then biomass in the Edge and Channel. Although biomass values in the Edge and Channel are both small, they did differ significantly from one another, with the Edge having greater biomass in summer 2001 and the Channel in summer 2002. Between December and June, the Intertidal has significantly greater biomass than the other three sites. Mean biomass values in the 5 intertidal sites also different spatially (Figure 2.9; 40 Appendix 4.3). Although there were no significant differences in April, by May GB Main had significantly less biomass then at LI 2 and LI 1. The difference between the sites became more pronounced later in the summer. In June, shoot biomass was significantly higher at LI 1 then at both GB Main and GB 1, while shoot biomass at LI 2 was intermediate. In August, biomass at LI 2 increased significantly to 180 ± 32 g / m 2 , which was 2 times greater then GB 1 which had the next highest biomass value and 2.8 times greater then GB 2 which had the lowest biomass. 2.3.4 Seagrass reproduction Sexual reproduction in the seagrass bed was also found to be seasonal (Figure 2.10; Appendix 4.4). Absolutely no reproductive shoots were found in the winter months (Dec & Feb). The reproductive season was longer in the intertidal strata then the subtidal strata. The Edge stratum consistently had the highest percentage of reproductive shoots of the four strata, with more than 8 % of shoots being reproductive in July 2001 and August 2002. The Intertidal and Slope strata had intermediate values, while the Channel site had the lowest values. Reproductive structure were absent throughout most of the year in the Channel site except in July 2001 and August 2002 when less then 1 % of shoots were reproductive. 41 15 July Sept Dec Feb M a y June A u g 2001-2002 Figure 2.10 Seagrass reproductive frequency (percentage of total shoots which are reproductive) in four tidal strata over the sampling period from July 2001 to August 2002. Error bars represent one standard error of the mean. The percentage of shoots that were reproductive was similar over the summer between four of the five intertidal sites (Figure 2.11; Appendix 4.4). Maximum reproduction occurred in August at these 4 sites with 2 to 4 % of shoots becoming reproductive. The exceptional site was LI 1 where these values were as much as 2.9 fold higher than the mean of the next highest site in May and 1.8 fold higher in June. The maximum reproductive output at LI1 was 5 % and occurred in June. 42 Apr i l M a y June 2002 Augus t - S GB Main -A—GB 1 -© GB2 •m - L I I •At - L I 2 Figure 2.11 Seagrass reproductive frequency (percentage of total shoots which are reproductive) in five intertidal seagrass beds during the summer of 2002. Error bars represent one standard error of the mean. 2.3.5 Physical parameters The temporal changes in the seagrass beds predominantly reflect predictable seasonal variation due to climatic conditions. Summer air temperatures are almost 3 times greater than winter temperatures (Figure 2.12a); the same trend occurs in water temperature, but at a smaller magnitude. In winter, light availability is decreased significantly due to shorter daylengths and increased cloud cover as seen in increased precipitation during winter months (Figure 2.12b). 43 a) b) 600 500 e 400 o H '5. 300 -o 1! n. 73 200 -100 -M a y July Sept Nov normal « — 2 0 0 1 - 2 0 0 2 M a y Jul • normal -*—2001-2002 Jan M a r M a y Jul Figure 2.12 Mean monthly air temperature (a) and total precipitation (b) at Tofino airport (adjacent to Grice Bay) over the course of the study relative to the 30 year normal. Spatial differences between seagrass populations in different tidal strata or intertidal beds may also be due to different abiotic conditions between the strata or sites. The most obvious difference between these populations is their establishment at different elevations. Tidal level influences length of exposure to air and water, desiccation and light availability thereby directly affecting photosynthetic efficiency. The sites may also 44 be exposed to different water regimes, influencing temperature, nutrient availability and sediment composition. Some of these differences were measured in this study and are seen in Table 2.3. Table 2.3 Abiotic differences between strata and sites in elevation, mean sediment grain size and maximum current velocity (Yakimishyn unpublished 2002). Positive numbers in first column (elevations) are intertidal sites, while negative numbers (depths) are subtidal strata, absolute numbers are relative to mean lower low water. Error value for mean sediment grain size is one standard error of the mean. Site Elevation or depth (cm) Mean sediment grain size (mm) Maximum current velocity (m/s) GB Main Edge 68 Intertidal 68 0.157 ±0.03 0.271 Slope -24 to -370 Channel -376 GB 1 78 0.118 ±0.001 GB 2 50 0.125 ±0.0004 0.259 LI 1 177 0.116 ±0.0004 LI 2 17 0.122 ±0.0002 0.532 2.4 DISCUSSION 2.4.1 Temporal patterns a. Seasonality in Zostera marina population parameters The results of this study indicate that both temporal and spatial differences exist within and between seagrass beds. Temporal variability may be either predictable or unpredictable. Predictability is usually due to seasonal climatic regimes and results in trends towards a greater biomass (more and larger plants) during the summer months and minimum values in winter. Seasonal trends have been previously reported in all components of Zostera marina growth, from leaf length (Zimmerman et al. 1995a), 45 growth rates (Everett et al. 1995), density, biomass, respiration to photosynthesis (Hansen et al. 2000) and were found in this study. Seasonal cycles in Zostera marina growth have been examined by comparing populations at different latitudes. Duarte (1989) found that more than 70 % of the biomass variability reflected seasonal responses, constrained by latitude. A l l species of macrophytes at lower latitudes sustained a relatively uniform biomass throughout the year, while temperate communities showed seasonal responses. Temperate communities also appear capable of displaying a wider range of variability in responses, for example annual and perennial populations of Z. marina are known to coexist in the same area (Keddy & Patriquin 1978). It has therefore been argued that temperate communities provide a less stable habitat than tropical and subtropical seagrass communities (Duarte 1989). In this study, seagrass shoot length at GB Main were maximal in late summer and minimal from December to May. At the five intertidal sites, shoot length increased only slightly from April to June and then rapidly to maximal lengths in August. Seasonality was most pronounced in the Slope (subtidal) stratum, as this zone had both the longest shoots in summer and shortest shoots in winter, which accounted for a 6 fold difference between maxima and minima values. In the Intertidal, the difference between summer and winter seagrass length was only 3 fold. In general, seasonal variations are thought to be less in the subtidal then the intertidal, where shoots are exposed to the midday sun during the summer and freezing air temperatures during winter nights. However, light intensity does vary seasonally in the shallow water of the slope and increased growth of subtidal shoots increases light availability for further growth.. Furthermore, subtidal 46 shoots are always supported by the water, which offers physical support for growth as well as continued access to water column nutrients. Leaf width did not show a consistent seasonal pattern across the strata sampled. Shoot width was minimal in winter in the Edge and Intertidal strata as found in other studies in Washington State (Backmanl991). However, shoot width was at a minimum in May on the Slope and in June and July in the Channel stratum. The greatest variation, as in shoot length, occurred in the Slope stratum with a 3 fold difference between minima in May and maxima in August. A seasonal trend was not consistently found at the five intertidal sites from April to August, as trends in shoot width were highly variable. In Grice Bay, seasonal variation in shoot density, between winter minima and summer maximum was slightly less than the variation in shoot length. In a review article (Olesen & Sand-Jensen 1994b), most eelgrass populations were found to predominantly allocate biomass to increased shoot size and maintaining stable shoot density. Overall, shoot density showed a 4 fold increase in the Slope stratum at GB Main. Seasonal variability in shoot biomass is thus more the result of changing shoot size rather than shoot density. Shoot biomass varied seasonally in the Intertidal and Slope strata, which makes up most of the area of the GB Main seagrass bed. Shoot biomass peaked in July and declined significantly by September in these strata. Biomass began to increase in the Intertidal in May and in June in the other strata. Shoot biomass increased throughout the summer to a maximum in August, before it presumably declined i f following the same trend as the previous year. The largest change in biomass was 40 fold difference between summer and winter on the Slope stratum. However, this was due to the almost negligible 47 biomass found on the Slope in the winter. In the Intertidal, the difference between summer and winter biomass was 7 fold, which is in agreement with previous findings in the literature. In other studies, shoot biomass shows a median 8 fold increase from the winter minimum to the summer maximum (Olesen & Sand-Jensen 1994a). Large seasonal biomass changes (6 fold) have also been found when considering total biomass, which is likely damped by the smaller change in below ground biomass which may make up more than 50 % of the total biomass (Nelson & Waaland 1997; Hansen et al. 2000). Around the world, reproductive efforts in Zostera marina peak at different times in different localities, April in Italy (Sfriso & Ghetti 1998), May in Denmark (Olesen 1999) and May to early June in Chesapeake Bay, U.S.A. (Orth et al. 1994). In the Pacific northwest Z. marina flowering has been found to peak in June (Harrison 1979) or July/August (Harrison 1982a) in Boundary Bay, B.C. In Puget Sound, flowering shoots appear from March to May and are at their greatest abundance in July and August (Moody 1978). Peak flowering intensities usually vary from 0 to 20% in the literature (Phillips et al. 1983a; Sfriso & Ghetti 1998; Olesen 1999). In this study, the timing of peak flowering was consistent with studies at similar latitudes, although peak flowering intensities were in the low range at about 5 %. Flowering occurred for the longest time and at the highest frequency (8 %) in the sheltered Edge stratum at GB Main. Peak flowering intensities in the Intertidal and Slope strata were 5 %. In the Channel stratum, flowering shoots were only found in July and August and comprised less than 1 % of all shoots. Overall, flowering biomass at GB Main was very low (~ 5 %) as the Edge stratum only accounts for a small area of the bed where density is very low. At the five intertidal seagrass beds, flowering had begun in all 48 sites when they were first sampled in April, although at a low frequency, which was maintained in May. By June, flowering frequency had increased and peak flowering at the five sites ranged from only 2 - 4 % in August. Although no reproduction was seen in December and February, it is possible under local climatic conditions. Flowering, in Puget Sound, has been reported to occur in water 7 - 9°C and seeds may mature at 11 - 14°C (Phillips et al. 1983b). Flowering has also been documented to occur at water temperatures of 8 - 9°C in Nova Scotia (Harrison & Mann 1975), 10 - 15°C at Roscoff, France (Jacobs & Pierson 1981) and 7 - 12°C during the winter in Washington State (Gambi 1988). It has been argued that the occurrence and incidence of flowering is not determined by a precise temperature value or range, but is due to temperature increase and the length of time over which the rise in temperature occurs (Gambi 1988). b. Mechanism of seasonality While seasonality has been consistently documented to occur, the mechanism of this seasonality is often debated and is likely different in different locations. Temperature is the most obvious abiotic factor that varies seasonally (Hillman et al. 1989). While the magnitude of change in water temperatures is substantially less than that of air temperature, many biological communities are limited by critical temperatures. Although Zostera marina is able to maintain growth at low water temperatures, exposure to freezing air temperatures during low tide and ice cover in winter may be limiting factors (Phillips et al. 1983b; Duarte et al. 2002). Conversely, temperatures exceeding 30°C lead to negative daily carbon balances of leaves due to increased respiration, which could contribute to mortality or reduced 49 growth of the plants (Marsh et al. 1986). In summer, Zostera marina in the temperate waters of the Pacific Northwest (British Columbia to Oregon) and Northern Europe (Denmark, the Netherlands) are not subjected to an inhibiting temperature and therefore annual growth curves are unimodal, with greatest biomass and production in summer (Robertson & Mann 1984; Dennison 1987; Hillman et al. 1989). At low latitudes, seagrass populations are found to have a bimodal growth curve with two periods of most rapid growth, in May and October and the lowest periods of growth in winter and summer (Felger et al. 1980). Although temperature undoubtedly plays an important role in the life cycle of Zostera marina, it is often difficult to separate the effects of temperature from concurrent changes in daylength and light intensity (Nelson & Waaland 1997). In many field studies, light was the best predictor of eelgrass biomass and productivity and increased daylength predicted the timing of spring growth (Harrison & Mann 1975; Nelson & Waaland 1997; Zharova et al. 2001). In the Pacific northwest, light was more abundant during the El Nino event and lead to an increase in biomass and productivity compared with the control year (Nelson 1997). Aquaria experiments have also suggested that growth is controlled by light (Aioi et al. 1981). Total irradiance received by marine plants is much higher in spring and summer than in winter, due to increased photoperiod and sun angle, decreased cloud cover and low tides occurring during the daylight (Hillman et al. 1989). In British Columbia, the decrease in mean temperature in the winter occurs concurrently with shortened daylengths, less sunlight due to cloud cover and increased respiration during these periods. Interannual variability in these climatic parameters 50 could explain differences in the seagrass population parameters between the two summers sampled. During the study, summer 2001 had lower mean air temperatures and higher total rainfall than the 30 year normal, while summer 2002 had higher mean temperatures and lower total rainfall than normal. Density, biomass and reproduction were all reduced in 2002 compared to 2001 suggesting that temperatures may have been inhibiting that year especially during the long hot low tides that occur midday in summer. Storms in winter may act as serious mechanical disturbance and cause substantial losses of seagrass biomass. In tropical areas, hurricanes may be significant in structuring seagrass ecosystems depending on their intensity and frequency (Poiner et al. 1989). Storms have also been implicated in seasonal seagrass loss due to light limitation during these events (Cabello-Pasini et al. 2002). In San Francisco Bay, winter growth was 50% of the summer rate and pulses of high turbidity that last for days or weeks at a time may be important in regulating the depth distribution of eelgrass. Light limitation during these events may produce negative carbon balances, leading to cessation of growth and even death (Zimmerman et al. 1995a). Month-long pulses in turbidity can account for the loss of transplanted vegetation and potentially explain lack of successful recruitment into formerly vegetated upriver sites (Moore et al. 1997). E l Nino has been argued to be beneficial to seagrass due to increased light availability, due to a significant increase in the number of hours of bright sunshine each day (Nelson 1997). However, increased storm events associated with E l Nino (Nelson 1997) and a warming climate (Short & Neckles 1999) may also have a negative impact on intertidal eelgrass biomass. Overall seasonal changes in density or size of seagrass are natural processes in these populations and indicate the sensitivity of seagrass to the overlying water 51 conditions. It is important to have baseline data of seagrass populations over an entire year to compare these values to future measurements. Deviations from not only expected values, but also a change in the variability within a stratum, site or month may be indicative of stress on this population. 2.4.2 Spatial patterns The spatial distribution and growth of seagrasses are regulated by a variety of water quality factors, such as temperature, salinity, substratum, turbidity, nutrients and light (Dennison 1987; Duarte 1991; Abal & Dennison 1996). Spatial patterns of seagrass parameters in Clayoquot Sound were considered from two approaches, comparisons between connected tidal strata and between unconnected intertidal beds, a. Tidal strata The main site in Grice Bay (GB Main) was sampled in four strata: the high intertidal Edge (intertidal), the Intertidal, the Slope (subtidal) and the Channel (subtidal). Tidal height has historically been cited as the primary factor determining the distribution of marine organisms. It is currently considered the most important factor for seagrass distribution, due to indirect effects such as light penetration and desiccation. In several studies, the shallow edge of the Zostera marina bed was found to be maintained by periodic disturbances, including storms and waves causing physical stress (Dennison & Alberte 1986; Koch & Beer 1996). The upper limit of Z. marina may also be maintained by exposure to air due to desiccation and photoinhibition (Duarte 1991). Under intertidal conditions in summer, growth rates of a Z. marina population in Boundary Bay, BC declined, supporting the argument that a low resistance to desiccation 52 limits the upwards penetration of Z. marina (Harrison 1982b). Other possible stresses affecting intertidal eelgrass include increased substrate compaction and biomass removal by grazing waterfowl (Mather et al. 1998; Ganter 2000). Conversely, subtidal meadows are never photoinhibited, have continuous access to water column nutrients and are less susceptible to wave shock and sediment disruption (Nelson & Waaland 1997). Shoots have been found to have higher rates of leaf elongation when continually submerged than when exposed to a midday low tide (Harrison 1982b). The maximum depth limit of seagrass is believed to be controlled by the critical light level needed to maintain photosynthesis and growth, which is therefore a function of water clarity. Water clarity in turn is primary affected by the concentrations of phytoplankton, suspended particles and dissolved matter in the water (Chambers & Kalff 1985; Dermison & Alberte 1986; Duarte 1991; Abal & Dennison 1996). Changes in water quality, such as increased turbidity and sedimentation, greater concentrations of phytoplankton and enhanced nutrient inputs from the land may cause a reductions in seagrass distribution and depth penetration (Abal & Dennison 1996; Koch & Beer 1996; Zharova et al. 2001). A review of thirty-six studies determined that Zostera marina had a mean maximum depth limit of 3 m (Duarte 1991). However, reports for Z. marina cite extreme depths of 34 m off the coast of Mexico, 25 m off the California coast and 11m along the northeast coast of the U S A (review in Dennison 1987). The depth limit of seagrass meadows is not established as a sharp threshold, but involves the progressive decline in seagrass abundance from the depth where maximum biomass is observed to their depth limit. The reduction in shoot biomass is largely due to a decrease in shoot density, 53 whereas shoot weight does not show any consistent trend with depth. The decline in seagrass biomass with increasing depth is often exponential, suggesting that this decline is due to light attenuation (Duarte 1991). The Zostera marina populations sampled in this study were all perennial, maintaining growing shoots as well as rhizomes all year long. Maximum densities at GB Mam were 47 shoots /m in the Channel and 525 in the Intertidal. Maximum shoot densities in other temperate subtidal beds ranged from 180 (Boundary Bay; Harrison 1982a), 80 (Oregon; Everett et al. 1995), 450 (Massachusetts; Burdick & Short 1999) and 144 (U.K.; Webster et al. 1998) shoots /m 2 . Shoot density in intertidal beds are usually higher with 496 and 441 shoots /m 2 at two locations in Washington State (Backman 1991; Bulthuis 1995). The 9 fold difference between intertidal and subtidal shoot density in Hood Canal, Washington (Backman 1991) is similar to the 11 fold difference found in this study. Maximum shoot length was greatest in the Slope stratum (77 cm), followed by the Intertidal (67 cm), Channel (39 cm) and Edge (33 cm). These results are in agreement to previous findings where subtidal shoots tend to be longer then intertidal shoots. In Washington State, shoot lengths were longer lower in the subtidal (20 to 90 cm) than in the intertidal (9 to 45 cm) (Backman 1991). In Nova Scotia, two forms of Zostera marina occur, in the intertidal, short seagrass forms dense mats, whereas a larger form grows in deeper water (Harrison & Mann 1975). In a transplant experiment along a depth gradient in Massachusetts, it was found that shoot length increased with increasing depth. Shoot length was therefore determined to be a plastic trait in Z. marina (Dennison & Alberte 1986). 54 The differences in seagrass width between strata were similar to that of length. Again, previous studies have found wider shoots in the subtidal compared to the intertidal (Backman 1991). In this study, biomass was significantly greatest in the Intertidal stratum, with only the Slope having comparable biomass. In Oregon, the maximum biomass was 2.8 fold greater in the low intertidal compared to the high intertidal (Kentula & Mclntire 1986). The difference between the main site at Grice Bay and other studies is in terms of shoot size between tidal strata, rather than density. At this specific site, seagrass shoot size may not be a reliable indicator of environmental trends, as the baseline data do not vary in the way predicted by previous studies. The difference between this and other studies may be due to the slope of the bed. The subtidal seagrass in Clayoquot Sound mostly exists on steep slopes, where biomass drops quickly with increasing depth. The Slope stratum in this study measures the biomass range in order to calculate a mean, which may therefore be a relatively low estimate. Reproductive effort in Zostera marina was also found to differ between tidal strata. In general, plants increase their sexual reproductive effort in more stressful habitats or uncertain habitats (Phillips et al. 1983a; Meling-Lopez & Ibarra-Obando 1999). High flowering frequency has been observed in intertidal habitats (Phillips et al. 1983a) and in areas subject to physical disturbance (Silverhorn et al. 1983, van Lent & Verschuure 1994). In Yaquina Bay, Oregon, 91% of intertidal shoots flowered, while only 33% of those between M L W and M L L W flowered and only 17% of subtidal shoots flowered (Bayer 1979). In Izembek Lagoon, Alaska, flowering frequency among intertidal samples ranged from 4 to 36%, whereas flowering in the subtidal averaged 13 % (Phillips et al. 1983a). 55 At GB Main, flowering was likewise higher in the intertidal than the subtidal as found in other studies. Flowering frequency was highest in the Edge stratum, which is at the highest position in the intertidal and the most sheltered. However, the Edge stratum was also adjacent to a small mudflat containing a low density of ghost shrimp, possibly causing stressful conditions. Because the exact location of the edge was not monitored, it may represent a shifting boundary, which relies on sexual reproduction for continual reestablishment. The Intertidal and Slope transect did not differ from each other in terms of flowering frequency. While they were also similar to each other in respect to the other parameters measured, this is surprising as the environmental conditions to which they are exposed would seem to be very different. Shoots on the Slope are always submerged, while shoots in the Intertidal may be exposed for up to several hours in the middle of a summer day. It may be that the different environmental influences at these different tidal elevations tend to balance each other out. The similarity in the magnitude of the error in measurements in the Intertidal and Slope is somewhat surprising as the intertidal is a relatively level area, whereas the Slope changes in elevation by 3 m over a short distance. The stability of the Slope stratum indicates that it is probably a less stressful environment because the shoots are constantly submerged, b. Intertidal sites We have seen so far that seagrass population parameters differ at different tidal elevations. The second spatial sampling approach was to assess five intertidal beds. In previous studies, when replicate sites are sampled, they are often found to differ significantly with respect to size, structure and reproduction. Differences are usually not tested experimentally, but are hypothesised to be due to a combination of factors 56 including light availability, hydrodynamics, water temperature, nutrient concentration and salinity (Meling-Lopez & Ibarra-Obando 1999; van Katwijk et al. 1999). At the five intertidal beds in Clayoquot Sound, the density and biomass trends were similar for the three sites in Grice Bay and were different from the two sites in Lemmens Inlet. Maximum mean density was two fold higher at LI 1 than at GB 1 and GB 2, while maximum biomass was highest at LI2. This indicates that there is variability between these intertidal beds. Significant differences in shoot length and width were found between sites, but were not consistent as the order of sites different between months. Maximum total biomass (above and below ground) values for Zostera marina beds in the Pacific northwest range between 112 to 462 g dw / m 2 along the coast from British Columbia to Oregon (Harrison 1979, 1982a; Kentula & Mclntire 1986; Bulthuis 1995; Nelson & Waaland 1997). In temperate regions in other parts of the world, total biomass reached as high as 1090 g dw / m in Rhode Island USA, while above ground biomass reached 854 g dw / m 2 (Thorne-Miller & Harlin 1984). In Denmark, shoot biomass reached 187 g dw / m in August (Hansen et al. 2000). In warmer climates, total biomass reached 950 g dw / m 2 in Chesapeake Bay (Murray et al. 1992), In a subtidal Gulf of California population, above ground biomass peaks at 550 and 675 g dw / m (Meling-Lopez & Ibarra-Obando 1999). In a review of twenty nine studies on Zostera marina, eighty percent of the maximum leaf biomass values during summer fell between 111 and 391 g dw / m 2 and the median value was 245 g dw / m 2 . These published values are higher then the biomass values found in this study. While biomass did peak in August at LI 2 at 180 g dw / m 2 , this was much higher than all other mean values which 57 were less than 100 g dw / m 2 at the Intertidal sites in and around Grice Bay. While only shoot biomass was analyzed in this study, it is considered to be more sensitive to environmental change than below-ground biomass. In general, Zostera marina is characterised by a particularly high biomass allocation to above-ground relative to below-ground parts compared to other seagrass species (Duarte & Chiscano 1999). In a review of the published literature, average above ground biomass for Z. marina was 298 g dw / m2 (n = 49), while the overall seagrass average was 239 g dw / m 2 (n = 423). Average below ground biomass was 149 g dw / m 2 (n = 29), while the overall seagrass average below ground biomass was 236 g dw / m 2 (n = 250) (Duarte & Chiscano 1999). The proportion of below ground biomass was typically highest in winter and lowest in summer (Nelson & Waaland 1997). In Italy, the biomass of shoots was twice as high as that of roots-rhizomes, with winter ratios being close to one (Sfriso & Ghetti 1998). Flowering timing and frequency were found to vary between intertidal beds. While trends in the three Grice Bay sites were similar, flowering frequency at LI 1 reached its peak in June, while the four other sites peaked in August. Overall flowering frequency was similar between sites, with peaks between 2 and 5 %. On a larger spatial scale, flowering tends to begin earlier in southern locations and display a higher flowering frequency at the extremes of the range, such as 100% and 33 % in the Gulf of California and Southern California respectively. Mid-range frequencies have been 17% in Oregon, 3 - 11% in Washington and 1 - 13% in Southern Alaska 1-13 %. At the Northern end of its range in Alaska flowering frequency again rises to 26 % (Phillips et al. 1983a). 58 2.4.3 Interannual variability In this study, there were no significant differences between the two summers in any of the seagrass parameters investigated. Seagrass populations may be very stable in nature. In a long-term study of an intertidal Zostera marina bed in Izembek Lagoon, Alaska, the overall net change was a 6 % gain between 1978 and 1987 and a < 1 % gain between 1987 and 1995. This lack of significant change in eelgrass cover suggests that the seagrass meadows have been stable over the 17 year period (Ward et al. 1997). This stability is most likely the result of low incidences of disease and anthropogenic disturbances that have been the major sources of change in seagrass distribution in other areas of the world (Ward et al. 1997). Predictably fluctuating populations make better indicators than unpredictably fluctuating populations as the reason for the change is often able to be determined. To be detected, changes in predictable populations need to be less dramatic than changes in unpredictable populations causing them to be identified earlier and benefit management efforts. 2.5 M A N A G E M E N T C O N S I D E R A T I O N S Seagrass populations are declining around the world, which has been attributed to widespread deterioration of water quality. Industrialization and increased land use has resulted in heightened levels of nutrient loading, sedimentation, influx of contaminants and toxins and other detrimental effects on these sensitive communities. The loss of these populations is causing the loss of entire communities of epiphytes, macro- and micro-algae, invertebrates and vertebrates that reside or migrate through seagrass beds (Short & Wyllie-Echeverria 1996). 59 These declines are seen as a decrease in shoot density and biomass of seagrass populations. This study has provided baseline values for density and biomass parameters of seagrass populations in several beds in Clayoquot Sound, specifically in Grice Bay and Lemmens Inlet. Declines from these values should be taken seriously and action taken to rectify pollutants or causes of environmental decline. While seagrass loss can often be drastic, a goal of conservation is to detect subtle declines early in order to reverse the trend before the seagrass bed is lost. To do so, it is important to monitor telltale parameters at the right intensity to identify a biologically meaningful decline in seagrass. In this study, the four measurements (density, length, width and biomass) gave different results. While these measures of seagrass parameters may all be useful alone, it is necessary to use a combination of density and size measures to get an idea of bed structure. Beds may have similar density, but very different biomass or have similar biomass with very different structure due to different densities. For example, a bed with many short shoots would be physically different then a bed with fewer long shoots, although overall biomass is the same. These beds may be subject to different water regimes, would affect the water current differently, display different sediment composition and possibly support different faunal assemblages. When monitoring this bed in the future the two measures that showed the most consistent trends are density and biomass. Biomass estimates give an overall estimate of the integrity of the bed by assessing the amount of above ground material available for photosynthesis. Density measures will supplement this information with an idea of structural differences between beds or strata. These two measures had relatively small error values and their interannual variation was low, with non-significant differences in 60 estimates between the two summers. Both of these indices showed similar trends for the three intertidal beds found in Grice Bay, which are subject to similar water column effects. The order of sites or strata from high density/ biomass to low density/ biomass remained relatively consistent. Overall, these parameters also showed the least amount of unexplained variation. Sexual reproduction is another important parameter to monitor, as water quality may strongly influence flowering and reproduction of seagrasses. Seagrass reproduction is primarily influenced by temperature and secondarily by daylength and salinity (McMillan 1976). These environmental factors may effect the duration of the reproductive season and frequency of reproductive shoots. Reproduction in these Zostera marina populations can be monitored easily, by focusing the density and biomass measurement during the summer when the seagrass are actively growing and reproducing. In order for monitoring to be effective, natural changes need to be distinguishable from human induced changes. As this study only encompassed two growing seasons, it does not give a full range of the potential variation. However, the data presented here provide a starting point for monitoring. We have also learnt that areas of seagrass beds do differ, both between connected tidal strata and unconnected beds. It is important to keep these differences in mind when protecting a seagrass bed. It is not enough to consider that all seagrass beds are protected by protecting one bed, as different beds do have different properties and may support different faunal assemblages. Furthermore, seagrass reproduces asexually and it is therefore unknown how much genetic variability exists in a seagrass bed. In order to protect the greatest amount of genetic variability, as 61 for any other species, it would be prudent to conserve multiple beds across large geographic ranges. At present, the destruction of seagrass beds is not permitted in Canada. Seagrass beds are an important fish habitat and as such, they are offered protection with a no-net loss policy by the Canadian Department of Fisheries and Oceans. If a bed needs to be lost due to development, the parties in charge have the responsibility to replace the lost seagrass habitat by planting in another suitable location. In order for such restoration projects to have any chance of success, having baseline data on factors such as density and biomass of natural seagrass beds present in areas with similar abiotic influences is important. 62 C H A P T E R 3 SEAGRASS PHYSIOLOGY: C H L O R O P H Y L L AND C A R B O H Y D R A T E S 3.1 INTRODUCTION Seagrass populations are declining around the world, which has been attributed to widespread deterioration of water quality. Industrialization and increased land use has resulted in heightened levels of nutrient loading, sedimentation, influx of contaminants and toxins and other detrimental effects on these sensitive communities (Orth et al. 1994; Abal & Dennison 1996). The loss of these populations is causing the loss of entire communities of fauna that use seagrass beds for shelter or as nursery habitat. The spatial distribution and growth of seagrasses are regulated by a variety of water quality factors such as temperature, salinity, substratum characteristics, turbidity, nutrient availability, and submarine irradiance (Dennison 1987; Duarte 1991; Abal & Dennison 1996). Of these parameters, light availability seems to be most important as it controls the maximum depth (Chambers & Kalff 1985; Duarte 1991) and biomass accumulation of seagrasses (Murray et al. 1992; Abal et al. 1994; Abal & Dennison 1996). The presence of seagrasses is therefore a function of water clarity (Chambers & Kalff 1985), which is in turn affected by the concentration of phytoplankton and other suspended particles in the water (Abal & Dennison 1996). Changes in water quality, such as increased turbidity and sedimentation, greater concentrations of phytoplankton and enhanced nutrient inputs from the land due to natural or anthropogenic impacts may cause a reduction in seagrass cover and maximum depth (Abal & Dennison 1996). 63 Seagrasses live in a low and variable light environment compared to the majority of angiosperms. However, as their roots and rhizomes are generally contained within anoxic sediments, the requirement to supply oxygen to root tissue places high demand on the plant for photosynthetically produced oxygen (Gallegos & Kenworthy 1996). Therefore, seagrasses typically have higher minimum light requirements than those of phytoplankton, macroalgae and terrestrial plants (Dennison 1987). While minimum light requirements vary consistently for each species, Zostera marina has been shown to require from 18 to 28 % of the surface irradiance and approximately 6 hours of light-saturation per day (Dennison et al. 1993). Photosynthesis is an important physiological function of plants, in which the plant uses solar energy to oxidize water, thereby releasing oxygen, and to reduce carbon dioxide into organic compounds, primarily sugars. In higher plants, photosynthesis occurs within the chloroplasts of leaves, which contain the specialized light-absorbing green pigments, chlorophylls (Taiz & Zeiger 1998). Both the quality and quantity of light is important to plants. While the light spectrum extends from ultraviolet to infrared, only light in the visible spectrum (400 - 760 run) is photosynthetically active radiation and available for photosynthesis. As light enters the water it is absorbed by water molecules and other suspended particles and dissolved substances, with some wavelengths passing further through the water than others. Because of this, the quality and intensity of the light changes rapidly as the light penetrates to greater depths. The quantity of light also changes daily and seasonally corresponding hours of daylight in the area. Other factors that affect the access of seagrass to light include the cloudiness of the local climate, the roughness of the sea and the concentration of soluble materials in the 64 water (Milne 1995). Due to higher loads of particulates and dissolved organic substances, light transmission in coastal waters is much lower than in clear ocean waters. Seagrasses respond to changes in light availability with a variety of morphological and physiological mechanisms. While loss of physical structure (declines in density or biomass) is often a sign of stresses being imposed on the seagrass, physiological changes often occur earlier and may be early warning signs of impending losses of shoot density and biomass. In the previous chapter, it was shown that biomass, density and size of the seagrass plant varies on a spatial and temporal scale. In this chapter, the physiological differences in carbohydrate and chlorophyll concentration that occur between sites and over time will be examined in order to obtain baseline measurements of these parameters. Carbohydrates are the primary products of photosynthesis and the measurement of carbohydrates gives the nutritional status of the plant. Carbohydrates are needed for growth, maintenance and survival of the whole plant. Carbohydrates are synthesised in the shoots and are transported in the phloem to the roots and rhizomes. New shoots are carbohydrate sinks, and older shoots are carbohydrate sources (Zimmerman et al. 1995b). Sucrose is the most abundance carbohydrate in Zostera marina, comprising on average 96% of the soluble sugars. The other soluble sugars present include glucose, fructose and wyo-inositol (Drew 1978, 1983). A small portion of the carbohydrates produced will also be converted to starch, which is a storage product. The rhizome is the main storage organ, although it also contains mostly sucrose. The primary storage compound can be influence by growth status; plants undergoing rapid growth tend to have higher levels of sucrose relative to starch (Taiz & Zeiger 1998). The advantage of 65 storing sucrose is that it is readily available in case of a disturbance and energy is saved by not converting it to starch and then back again (Burke et al. 1996). Seagrass in the Pacific Northwest grow primarily during the summer months when days are longest and temperatures are optimal. During this period of long day-lengths, seagrasses are able to photosynthesize optimally and may accumulate carbohydrates. These reserves are then used for respiration, growth and maintenance of the plant in winter and early spring. As reserve carbohydrates are used during these stressful periods, decreased rhizome carbon reserves could be a reliable indicator of future decline in seagrass distribution and biomass. Because some of the sucrose is converted to starch, it is important to measure both as only depletion of both of these compounds would indicate an overall decrease of carbohydrates in the plant. While light availability is responsible for the survival of Zostera marina, survival is also dependent on the plant being able to store and allocate carbohydrates successfully during periods of light limitation that occur nightly, seasonally or unpredictably (Zimmerman et al. 1995a; Alcoverro et al. 1999; Cabello-Pasini et al. 2002). Another consequence of photosynthesis and thus light is the ability of Z. marina to create a favourable aerobic environment for its roots, by leaking oxygen produced in photosynthesis into the surrounding sediment. When photosynthesis is reduced due to darkness or high turbidity, the roots must maintain themselves in an anoxic environment. Z. marina roots can tolerate oxygen deprivation for 3 days and recover fully (Smith & Alberte 1989); tissue death will occur i f anoxia is prolonged. Protein synthesis is required to maintain cellular integrity and can continue as long as there are adequate carbohydrate stores providing the energy source for these reactions (Kraemer & Alberte 66 1995). Periods of light limitation cause reduced carbohydrate content in seagrass and eventually lead to a decrease in shoot density (Kraemer & Alberte 1995; Zimmerman et al. 1995b, Lee & Dunton 1997; Burke et al. 1996; Cabello-Pasini et al. 2002). In a study of a subtidal Z. marina population in Baja California, eel grass shoots died after sugar and starch content in the leaves decreased by approximately 85% after 3 weeks. Survival and leaf carbohydrate content decreased by more than 90% when plants were incubated for 3 weeks in darkness in the lab (Cabello-Pasini et al. 2002). Seagrasses, like all other higher plants, contain chlorophyll a and b. An analysis of chlorophyll content is important to identify relative potential for photosynthesis between beds and within beds (Dennison 1990). Because the spatial distribution and growth of seagrass is regulated by light availability, the presence of seagrass is therefore a function of water clarity. Changes in water quality may induce photoadaptive responses of seagrass including an increase in total chlorophyll content and a decrease in the chlorophyll a:b ratio as the plant tries to compensate for the reduced light levels (Taiz & Zeiger 1998). Studies have shown that shading of seagrasses or natural increase in depth is correlated with a photoadaptive response involving an increase in seagrass chlorophyll content and a decrease in the chlorophyll a:b ratio along with decreases in shoot density, biomass, shoot canopy and growth rates (Abal et al. 1994). The specific questions addressed in this chapter are: 1. What are the baseline carbohydrate and chlorophyll concentrations during the summer at five intertidal seagrass beds? 2. Do carbohydrate and chlorophyll concentrations follow a predictable trend over time and between beds, correlated to abiotic parameters (light, depth, temperature)? 67 3. Do carbohydrate and chlorophyll concentrations hold potential to be good early warning signs of declining environmental quality? 3.2 METHODS Zostera marina plants between 30 and 50 cm tall were collected at five intertidal sites, three in Grice Bay and two in Lemmens Inlet, to assess carbohydrate and chlorophyll concentrations in shoots of the same age. Healthy, green plants were collected haphazardly along the three transects also sampled for density and biomass. Individual shoots were sampled at least 1 m apart to minimize the chance of sampling clones. Samples were collected in ziplock bags and placed on ice. When possible shoots were analyzed that afternoon, however it was sometime necessary to freeze samples for a maximum time of 48 hours. Samples were cleaned of salt and epiphytes, by rinsing under fresh water and scraping epiphytes by hand. 3.2.1 Carbohydrates Both the shoot and rhizome of the seagrass were analysed for carbohydrate and starch concentrations. Carbohydrates were extracted from a macerated 0.2 g plant sample by three successive extractions in hot 80 % ethanol. The alcohol solution was poured through Whatman 1 filter paper between extractions and the final volume of the filtrate was made up to 20 mL. Residue was dried and frozen for later starch determination. Starch was extracted by incubating the residue in 6 mL of cold (0°C) 30 % perchloric acid for 20 minutes, followed by centrifugation and decantation (AOAC 1995). 68 The concentration of sugar and starch was determined using the phenol-sulfuric acid colourimetric method (Hodge & Hefreiter 1962). The test involved combining 1 mL of the aqueous sugar solution with 1 mL of 5 % phenol and 5 mL of 96 % sulfuric acid. Test tubes were shaken and then incubated in a hot water bath at 25 - 30°C for 20 minutes. The absorbance of the yellow-orange colour produced was measured at 490 nm. The average absorbance of the blanks was subtracted from each sample reading and the amount of sugar was determined by reference to a standard curve prepared for solutions containing known concentrations of glucose and starch. Carbohydrate concentrations are expressed in mg/g dry weight. A 0.2 g sample (wet weight) of shoot and rhizome segments was dried to obtain wet weight to dry weight conversion factors. 3.2.2 Chlorophyll To analyse for chlorophyll concentration, 0.02 g (wet weight) of the top portion of healthy green leaves was macerated with a mortar and pestel. Ground samples were incubated in 6 mL of 80 % acetone in a dark refrigerator for 24 hours. After 24 hours, the volume of acetone in the test tubes was adjusted to 8 mL. Test tubes were centrifuged at 1100 rpm for ten minutes. Acetone was decanted into spectrophotometer tubes and absorbances were recorded at 645 (chl b absorbs maximally), 663 (chl a absorbs maximally) and a turbidity correction factor at 725 nm. To determine the concentration of chlorophyll a and b, the equations of Arnon (1949) were used. A wet weight to dry weight conversion factor was obtained from a subsample of these small tissue samples so that the results could be calculated as concentration of the compound per g of dry weight. 69 3.2.3 Data analysis A l l values are reported as means ± standard error of the mean. Carbohydrate data were analysed to determine i f there were different levels of glucose and starch between the shoots and rhizomes, in both April and June. Since paired shoot and rhizome samples were obtained from the same plant, a non-parametric Wilcoxon paired rank sum test was used to test for these differences. Differences of soluble sugar and starch concentrations in the shoots and rhizomes between the two sampling dates were analysed using a non-parametric Mann-Whitney analysis of variance. To examine differences in chlorophyll levels between study sites and sample dates a non-parametric Kruskal-Wallis non-parametric analysis of variance was conducted to determine i f any statistically significant differences existed. Post-hoc like tests were conducted using multiple Mann-Whitney tests with a correction for the increased chance of error by setting a equal to 0.05/# of tests. These statistical tests were conducted for chlorophyll concentration and chlorophyll a:b ratio. Mean values (Appendix 2) and the results of the statistical analysis (Appendix 5) are found in the appendices at the end of the thesis. 3.3 R E S U L T S 3.3.1 Carbohydrates a. Within plant Both soluble sugar and starch concentrations were significantly greater in rhizomes then in shoots in both April and June as seen in Figure 3.1 (Appendix 5.1). Mean sugar concentration in April was 11.7 ± 0.5 mg/g in the shoots and 37.2 ± 3.9 mg/g 70 in the rhizomes. In June sugar concentration was 32.9 ± 2.8 mg/g in shoots and 140.2 ± 7.4 mg/g in the rhizomes. Mean starch concentrations in April were 16.0 ± 0.7 and 21.6 ± 0.6 mg/g in the shoots and rhizomes respectively. The difference in starch concentration in the shoots and rhizomes in June was less pronounced with 16.4 ± 1.0 and 20.8 ± 1.0 mg/g respectively. In this study, soluble sugars were 4 times more concentrated in the rhizome then in the shoots. Starch concentration consistently exhibits a trend towards a slightly higher concentration in the rhizomes then in the shoots, but not significantly. 160 73 50 |f 120 -a o § 80 c o cj D 1 J> 40 J3 o -O a 0 — * — S h o o t [Sugar] —• Rhizome [Sugar] - A - - Shoot [Starch] -r> - Rhizome [Starch] A p r i l June 2002 Figure 3.1 Soluble sugar (solid lines) and starch (dotted lines) concentrations in Zostera marina shoots (triangles) and rhizomes (squares) in April and June 2002. Error bars are one standard error of the mean. b. Temporal patterns An increase in carbohydrate concentration between April and June 2002 can also be seen in Figure 3.1 (Appendix 5.1). Soluble sugar concentrations significantly increased in seagrass shoots and rhizomes between April and June 2002 at all sites, 71 except in shoots at GB Main. Overall shoot sugar increased from 11.7 + 0.5 to 32.9 ± 2.8 mg/g and rhizome sugar increased from 37.2 ± 3.9 to 140.2 ± 7.4 mg/g. Conversely, starch concentrations did not differ between April and June. In the shoot, the overall mean starch concentration was 16.0 ± 0.7 and 16.4 ±1 .0 mg/g in April and June respectively and in the rhizome the mean starch concentration was 21.6 ± 0.6 and 20.8 ± 1.0 mg/g. c. Spatial patterns Differences in soluble sugar concentrations were found between the five intertidal sites although concentrations were highly variable between April and June (Appendix 5.1). Starch concentrations showed no significant differences between sites and no apparent trend. From the previous graph, we see that rhizome soluble sugar concentrations are the most sensitive to changing environmental conditions between April and June. While the carbohydrates are being produced in the shoots, increased carbohydrate production in June may result in both increased usage in the shoots and increased transport out of the shoots to the rhizomes. In Figure 3.2, the difference in rhizome sugar concentration between sites is shown. In April , seagrass rhizomes at both sites in LI had more sugar than those at GB 2. In June, there were no significant differences in rhizome sugar concentrations between sites, however the trend indicated that carbohydrate concentrations in April are not predictive of levels in June. While the two sites in LI still had the most sugar, rhizome sugar concentration at GB 1 & 2 was now comparable and GB Main had the lowest sugar concentration in seagrass rhizomes. 72 ^ 200 -o 60 — G B M a i n — G B 1 - G B 2 - L I 1 - L I 2 Apr i l June 2002 Figure 3.2 Soluble sugar concentration in seagrass rhizomes at the five intertidal study sites in April and June 2002. Error bars are one standard error of the mean. 3.3.2 Chlorophyll a. Temporal patterns When considering the five sites together, there is evidence of seasonality in seagrass shoot chlorophyll concentration. Total chlorophyll concentration in the shoots changed significantly every month beginning with the highest concentration in April, dropping to lowest concentrations in May and June and then increasing in concentration again in July and August (Figure 3.3; Appendix 5.2). The difference in the temporal patterns of chlorophyll a and b concentration is reflected in the change in the chlorophyll a:b ratio from April to August. The chlorophyll a:b ratio decreased significantly each sampling date throughout the course of the summer (Figure 3.4). The only dates that were not significantly different from each other were June and July, which were only 2 weeks apart, whereas the other dates were all four to six weeks apart. 73 Is Apr i l M a y - B - • G B Main • G B 1 •GB 2 •m -ui •h — L I 2 June 2002 July Augus t Figure 3.3 Total chlorophyll (a + b) concentrations (mg/g dry weight) in seagrass shoots at 5 intertidal sites over summer 2002. Error bars are one standard error of the mean. 2.5 n _o JS U 0.5 A p r i l M a y June July Augus t 2002 Figure 3.4 Ratio of seagrass chlorophyll a:b over summer 2002 at five intertidal beds. Error bars are one standard error of the mean. 74 When the sites are considered separately, the same seasonal trends are still apparent at each site however, there is considerable variability in these fluctuations over the summer at the different sites. Two of the sites, GB Main and LI 2 had the lowest total chlorophyll concentrations in June, while GB 2 and LI 1 have lowest chlorophyll concentrations in May. Chlorophyll concentration was also highest in April at GB 2, LI 1 and LI 2, but higher concentrations were reached later in the summer at GB Main. Chlorophyll concentrations at GB 1 remained relatively constant throughout the summer. Seasonality in chlorophyll a:b ratio was likewise variable between sites, b. Spatial patterns Differences in chlorophyll concentrations between the five sites appear to be significant, however in the first four sampling dates, shoots were collected from the different sites on different days due to logistics. In August, all five sites were sampled on the same day to minimize sampling error. The only significant difference between sites in August was in chlorophyll a:b ratio, indicating that daily fluctuations in chlorophyll content was likely obscuring between site differences (Appendix 5.2). 3.4 DISCUSSION 3.4.1 Carbohydrates Carbohydrates are the main energy source of plants and are required to keep the plant in a healthy state. Carbohydrates are the main products of photosynthesis and when deprived of light, shoots die after sugar and starch content in the leaves has decreased past a critical point. The accumulation and mobilization of carbon reserves appear to play an essential role in the dynamics of eelgrass survival in temporally variable 75 environments (Cabello-Pasini et al. 2002). Therefore, the measurement of decreased rhizome carbon reserves could be a reliable indicator of impending decline in seagrass distribution and biomass (Lee & Dunton 1997). a. Within plant patterns The allocation pattern of carbohydrates found in this study is consistent with results found in other studies (Drew 1983; Zimmerman et al. 1995b & 1997; Burke et al. 1996; Kikuchi et al. 2001). Research on a range of seagrass species from the temperate and tropical regions of the world have always found higher concentrations of soluble carbohydrates in rhizomes than in shoots (Touchette & Burkholder 2000). In this study, soluble sugars were almost four times more concentrated in the rhizome then in the shoots. Decrease in rhizome stored carbohydrates as a result of defoliation by blade clipping have supported the hypothesis that rhizome carbohydrates are transported from rhizome to leaf tissue and used for leaf regeneration (Dawes and Guiry 1992). Soluble carbohydrate content in rhizome tissues likely serves as an energy reserve for the plant during the winter (Lee & Dunton 1997). b. Temporal patterns As sugars are the product of photosynthesis, they are produced in greater quantities when environmental conditions support increased levels of photosynthesis. Typically, sugar concentrations display seasonal fluctuations with accumulation in spring/summer followed by a decline throughout the winter, when respiration rates exceed photosynthetic rates. In the present study, sugars in the shoots and rhizomes of Zostera marina accumulated between April and June at the five intertidal beds sampled. This indicates that the supply of carbohydrates through photosynthesis was exceeding the 76 demand for growth and respiration during this time. In Clayoquot Sound, it would be expected that sugar accumulation would continue throughout the summer since light and temperature continue to be favourable, but not inhibiting throughout the summer. Seagrass in this study would begin to deplete their carbohydrate stores in autumn and throughout the winter. Carbohydrate concentrations would then be lowest in the spring, before the longer days. A decline in carbohydrate stores from summer to winter is a common pattern for many angiosperms as well as non-vascular marine macrophytes. This suggests that carbohydrates accumulated during summer are likely critical for support of metabolic activity and growth during the winter and early spring (Zimmerman et al. 1995a). However, photosynthesis is likely still important in supporting maintenance throughout the winter in this Z. marina population. Vermaat & Verhagen (1996) estimated that the storage carbohydrates of a Z. noltii population in the Netherlands could cover 28% of the respiratory needs during winter, which would necessitate a substantial photosynthesis to meet the remaining 72%. Light is usually implicated as the main factor responsible for carbohydrate levels from field studies and laboratory work. These studies have demonstrated the effect light and carbohydrate have on density, biomass (Burke et al. 1996; Lee & Dunton 1997) and survival (Cabello-Pasini et al. 2002). When insufficient light is available rhizome carbohydrate stores are depleted so that the plant is not capable of providing carbon to meet their daily metabolic energy requirements (Lee & Dunton 1997). Survival of plants in Baja California dropped to zero when carbohydrate levels dropped by 85% from peak summer amounts, while sugar levels dropped to almost undetectable levels in the lab. These results showed that subtidal eelgrass from Baja California had enough 77 carbohydrate reserves to maintain metabolic activity under light limiting conditions for only 3 weeks (Cabello-Pasini et al. 2002). In an aquaria study, Zostera marina responded to negative carbon balances imposed by low light treatments by suppressing the production of new roots, depleting sucrose reserves and effecting a gradual decrease in growth rate. Plants under extreme light limitation died although one-third of their carbon reserves remained immobilized in the rhizome. Thus, extreme light limitation can prevent full mobilization of carbon reserves probably through the effects of anoxia on translocation (Alcoverro et al. 1999). Seasonal trends of carbohydrate concentrations are apparent and are more pronounced in soluble sugars then in starch, as found in this study and others (Burke et al. 1996; Kikuchi et al. 2001). In this study, there was a 2.8 and 3.8 fold change in shoot and rhizome sugar concentration between April and June compared with the negligible change in starch concentration between the same two sampling dates in this study. Also seasonal changes in soluble sugar concentration was also found to be of a greater magnitude in the rhizome compared to the shoots in this study and in Zostera marina in Japan (Kikuchi et al. 2001). A similar trend was found in Z. noltii (Vermaat & Verhagen 1996) and Thalassia testudinum (Lee & Dunton 1999). Soluble sugar concentration in the seagrass rhizome appears to be the most useful indicator parameter as this parameter changed significantly between April and June indicating that it is sensitive to the environmental changes such as increased light and temperature that occur between this period, while sugar concentrations in the shoots and starch concentrations were not. 78 c. Spatial patterns In this study, significant differences in sugar concentration were found between five intertidal sites, but relative concentrations differed between April and June indicating that we do not have ability to predict such differences between beds. Starch concentrations were not found to differ between sites. In San Francisco Bay, transplanted seagrass also exhibited no significant between-site differences in shoot soluble sugar concentration. However, while starch represented less than 5 % of the extractable carbohydrate, a significant difference in shoot starch concentration did appear between the two sites (Zimmerman et al. 1995a). This is contradictory to the results found in this study in British Columbia, where soluble sugar concentrations were more sensitive to environmental changes than to starch levels. d. Comparing absolute values Because carbohydrate concentration was only measured twice during the early growing season, the maximum and minimum values were likely not measured. Soluble sugar concentrations found in this study were lower than those found in other studies of Zostera marina populations in Chesapeake Bay (Burke et al. 1996), San Francisco Bay (Zimmerman et al. 1995a) and Odawa Bay, Japan (Kikuchi et al. 2001). Minimum values of shoot sugar content were 2.5 to 5 times larger in the other studies, while minimum rhizome sugar content was similar to that at Odawa Bay, but 3 times lower than at Chesapeake Bay. Starch concentrations were found to be less variable than those of sugar. The results in the present study agreed with those found in San Francisco Bay and Odawa Bay, but minimum starch concentrations in both shoots and rhizomes were approximately 2 times lower than minimum values in Chesapeake Bay. Mean maximum 79 sugar and starch values were probably not measured in Clayoquot Sound as carbohydrate concentration would presumably increase further until August or September. The differences in soluble sugar concentrations between studies suggest a latitudinal gradient, most likely controlled by temperature and light. The small temperature range of 12 to 17°C is near optimal for the California Zostera marina population. In North Carolina, temperatures may fluctuate up to 30°C, which limits growth in the summer. At the current study site in British Columbia water temperatures range from 5 to 15°C, and short day-lengths may be limiting in the winter. In the summer, low tides occur during the day causing the plants to spend long daylight hours exposed to higher air temperatures and lying flat on the ground, preventing optimum photosynthesis. The lower mean values in this study suggest that seagrasses at high latitudes in the Pacific Northwest must experience lower levels of sugars at certain times of the year than those at other locations, e. Types of sugars Soluble sugars were found as the largest proportion of the carbohydrate reserves in Zostera marina, consistent with other studies (Burke et al. 1996, Drew 1980, Zimmerman et al. 1995a). Z. marina has large soluble sugar reserves relative to other plants, which is thought to be so that sugar is immediately available for respiration and may be beneficial for recovery after a disturbance. Energy is also conserved by not changing sugar to starch and back again (Burke et al. 1996). In Z. marina plants from British populations, sucrose was found to make up 83% of the soluble sugars and in order of decreasing concentrations fructose, /nyo-inositol and glucose each accounted for 9, 8 and <1 % respectively (Drew 1980). Kikuchi et al. (2001) also determined that sucrose 80 was the main type of carbohydrate stored. In Clayoquot Sound, starch accounted for 33 to 58% in shoots and 13 to 37 % in rhizomes. Low starch content in Z. marina was also found in others studies on the Pacific Coast of North America (Zimmerman et al. 1995a). Burke et al. (1996) reported that starch could form more than 65 % (up to 140 mg/g dry weight ) of the total nonstructural carbohydrate content in Z. marina in Chesapeake Bay. 3.4.2 Chlorophyll Changes in chlorophyll content of plants are usually attributed to factors associated with light. Changes in light include 1) changes in weather, influencing hours per day of light or clear days per year; 2) changes in light attenuation due to increased turbidity, depth, tidal cycles or epiphyte cover; 3) changes in light quality due to plankton community abundance and composition and 4) self-shading, especially during low tides when the seagrass is lying flat (Abal et al. 1994). Seagrasses can respond to light reduction by increasing their total chlorophyll content and by decreasing their chlorophyll a:b ratio, before measurable reductions in shoot density, biomass and growth rates occur (Dennison & Alberte 1982; Lee & Dunton 1997; Abal et al. 1994), a. Temporal patterns As chlorophyll levels are directly affected by light levels, it is expected that differences in total chlorophyll and chlorophyll a:b ratios wil l be found between summer and winter (Zimmerman et al. 1995a) or between shallow and deep habitats (Dennison & Alberte 1986; Lee & Dunton 1997). hi this study, chlorophyll levels were only measured in summer. Total chlorophyll was found to be highest in April , with subsequent decreases in May and June and then slight increases again in June and July. While many 81 of these differences were significant, the maximum change in chlorophyll level was only 64 %. These values predict that the highest light levels occurred in May and June. The chlorophyll a:b ratio followed a slightly different trend, again with highest values in April (1.85), but a consistent decrease until August (1.29), suggesting that light levels continued to increase throughout the summer. Other studies have also only been able to describe general trends as fluctuations in total chlorophyll and chlorophyll a:b ratios occur regularly (Lee & Dunton 1997). b. Spatial patterns Although changes in chlorophyll concentrations with changing depth were not measured in this study, trends in Zostera marina populations have been found in other studies that could be applicable to monitoring programs in Grice Bay. An increase in depth is typically accompanied by a decrease in chlorophyll a:b ratio (Dennison & Alberte 1986), as the light wavelengths absorbed by chlorophyll a decrease more rapidly than the wavelengths absorbed by chlorophyll b with increasing water depth (Lee & Dunton 1997). Thus while increasing total chlorophyll concentration to increase light absorption is useful, a rapid increase in chlorophyll b relative to chlorophyll a would allow more efficient use of wavelengths at depth (Lee & Dunton 1997). Differences in chlorophyll level and chlorophyll a:b ratios were found between the five intertidal sites investigated in this study. However, as with all the other parameters investigated, these differences were not consistent, exhibiting no clear trend between beds. Consequently, the data obtained are useful to show the natural variability that exists in Clayoquot Sound and to demonstrate the ability to generalize the results found at any one site. 82 c. Comparing absolute values The level of total chlorophyll in this study is higher and the chlorophyll a:b ratio is less than that reported in other studies. However, the literature values include those from different species and from Zostera marina in lower latitudes (Dennison & Alberte 1986; Lee & Dunton 1997; Cabello-Pasini et al. 2002). Chlorophyll levels are often expressed in many different ways, including on a fresh weight and shoot surface area basis. Tests have also not been done to compare the different techniques of chlorophyll extraction and measurement, therefore most conclusions have been formulated by within study comparisons. For the purposes of this study, the important consideration is to measure the chlorophyll levels of this population using the same protocol to compare to these baseline values. 3.5 MANAGEMENT CONSIDERATIONS Management efforts in several areas have emphasized improving environmental quality and have focused on using seagrass establishment and success as an indicator of habitat quality (Dennison et al. 1993). The widespread distribution and ecological importance of seagrasses and their sensitivity to water quality parameters have led to the use of these plants as biological indicators of water quality (Abal & Dennison 1996). The establishment of environmental criteria for the protection and restoration of seagrass requires an understanding of the temporal variation in environmental quality as well as physiological performance of the species of interest (Zimmerman et al. 1995a). Measurement of seagrass carbohydrate concentration is an informative way to assess the photosynthetic output of the seagrass The carbohydrate concentrations reported here are 83 presented as baseline measurements for April and June 2002. These values can be correlated with climatic conditions and will be elevated in summers with more sunlight and decrease in summers with less sunlight. However, these values fall at the low end of the range reported in the literature and therefore any significant decreases in carbohydrate concentration may be detrimental to plant survival throughout the winter. This parameter may be indicative of water quality as significant results were found with a low number of samples and variance was low. However, the extraction of sugars and determination of their concentration is a lab intensive procedure. It would be useful to measure this parameter periodically in the future, although it is unlikely to become part of a regular monitoring routine. While a monitoring program would probably not test chlorophyll and sugar regularly, i f it is suspected that the seagrass distribution is changing, measuring changes in these physiological parameters may indicate problems due to light limitation. 84 C H A P T E R 4 POPULATION DYNAMICS OF T H E GHOST SHRIMP NEOTRYPAEA CALIFORNIENSIS 4.1 INTRODUCTION Intertidal mudflats in the temperate areas of the world are increasingly well studied due to their importance in supporting large populations of migrating shore birds. While they are surrounded by productive salt marsh and seagrass vegetation, mudflats themselves support a surprisingly large diversity of invertebrates and microscopic plants (Little 2000). Grice Bay has an extensive mudflat, which is almost completely exposed at low tide. The most abundant organism in this habitat is the thalassinidean ghost shrimp, Neotrypaea californiensis (formerly Callianassa californiensis; Manning & Felder 1991). Early descriptions of this species by MacGinitie (1934) reported N californiensis to be one of the most abundant animals in the muddy shores of sheltered marine environments on the west coast of North America, ranging from Alaska to Baja California (MacGinitie 1934). N californiensis tends to be orange, yellow, or pink in colour, may reach a total adult length of 70 mm and have a lifespan of approximately 3 to 5 years (Bird 1982). In the Pacific northwest, dense N. californiensis beds are generally restricted to the high and mid intertidal (L. Thompson & Pritchard 1969; Swinbanks & Murray 1981) on mixed sand or mud shores (Stevens 1929; MacGinitie 1934). N. californiensis are continuously in motion, either feeding, cleaning or keeping water circulating through the burrow for respiratory purposes (MacGinitie 1934). Neotrypaea californiensis are ovigerous from April through August (Feldman et 85 al. 2000) . The eggs are carried by the female for 5 to 6 weeks (Bird 1982) until the embryos have reached the zoea stage, when hatching takes place (MacGinitie 1934). Females may become sexually mature at 3 0 mm total length (Bird 1982) and produce on average 4 0 0 0 eggs each, investing 3 0 - 4 0 % of their body weight in egg production (Dumbauld et al. 1996). The eggs begin to hatch in June and zoeae are released primarily during the night ebbs of neap tide series and exported to nearshore coastal waters during an 8 week planktonic period (McCrow 1972; Johnson & Gonor 1982). Thalassinidean shrimp rely on temporary burrows for shelter, feeding and reproduction and, except for the pelagic larval stage, spend most of their life within their burrows (Griffis & Chavez 1988). The burrows of Neotrypaea californiensis are dichotomously branched with two or more openings per burrow. The unlined burrows may be up to 9 0 cm deep, but are usually 6 0 cm deep (Dumbauld et al. 1996) and may branch horizontally for distances up to 1 m (Swinbanks & Murray 1981). Burrows of different individuals may be connected causing intraspecific interactions to be common (MacGinitie 1934), although other researches have found few connections between burrows (Griffis & Chavez 1988). Although burrows are thought to be extensive, at low tide they readily collapse. As N californiensis cannot osmoregulate, the collapsed walls protect the ghost shrimp against changes in water salinity during low tide (L. Thompson & Pritchard 1969). Furthermore, the ability of jV. californiensis individuals to slow their heart rate and possibly use anaerobic metabolism leads to high resilience to hypoxic conditions found in the collapsed burrows (R.Thompson & Pritchard 1969) . Neotrypaea californiensis has a high sediment processing rate and, i f left on the surface, will quickly re-establish itself in the sediment, in a existing burrow or wil l begin 86 digging a new burrow. Swinbanks & Luternauer (1987) determined that N. californiensis process 9 to 30 ml of wet sediment per shrimp per day. Reduced burrowing activity in winter is correlated with reduced activity at low salinities (Posey 1987) and in cold water (Swinbanks & Luternauer 1987). The presence of the mounds is a clear indication that the organisms are actively processing sediments, which may occur during normal burrowing or as part of wall grazing (MacGinitie 1934; Griffis & Suchanek 1991). Counting burrow holes may be used as a fast and non-destructive means of assessing shrimp density. A l l thalassinidean burrows alter soft sediment environments by increasing the surface area of sediments exposed to oxygenated water (Griffis & Suchanek 1991). Construction and maintenance of burrows by thalassinidean shrimp has specifically been linked to continuous mixing of deep and shallow layers of sediment resulting in substantial sediment resuspension and transport and changes in organic content and grain size of sediment (Griffis & Suchanek 1991, Jones & Jago 1993). Because of these physical effects, thalassinideans have been demonstrated to influence the community composition of their associated infauna significantly (Posey 1986b). Thalassinidean density is negatively correlated with the survival and growth of various sedentary suspension feeders (Stevens 1929; Peterson 1977; Bird 1982), surface deposit feeders (Brenchley 1981), corals (Aller & Dodge 1974), and seagrasses (Suchanek 1983). Deposit feeding thalassinideans have also been shown to reduce populations of meiofauna, bury microalgae (diatoms) and increase sediment bacteria numbers (Branch & Pringle 1987). Positive effects of thalassinideans on other species have also been recorded. The 87 benefits of sediment reworking, burrow ventilation, and shrimp metabolism include greater oxygen penetration into the sediments adjacent to the burrow wall and nutrient flux of ammonium and phosphate to the sediment surface and water column (Murphy & Kremer 1992). Thalassinideans may have a positive effect on colonization by mobile taxa, by stimulating the growth of microalgae and bacteria through irrigation or fertilization of the sediment or by loosening up the sediment to allow access to other burrowers into the sediment (Branch & Pringle 1987, Tamaki 1988, Murphy & Kremer 1992). When irrigating its burrow, Neotrypaea californiensis may create a favourable habitat for other organisms (MacGinitie & MacGinitie 1949). In temperate estuaries, the goby, Clevelandia ios, and the crab, Hemigrapsus oregonensis, occupy burrows temporarily to avoid predators and the effects of low tide. The clam, Cryptomya californica, and the crab, Scleroplax granulata, are permanent residents of the burrows, while two species of copepods, Clausidium vancouverenses and Hemicyclops thysanotus, are found on N. californiensis individuals (Bird 1982). The association seems to be obligatory for C californica.. When N. californiensis were completely removed from a study site in southern California, all the commensal clams died within a year (Peterson 1977). In California N. californiensis was found to share its burrow with five macroscopic species at once, most of which move freely in the burrow and are ignored by N. californiensis. The commensal organisms presumably derive shelter from the burrow and some may steal food from the host (MacGinitie & MacGinitie 1949). Although Neotrypaea californiensis has not previously been used as an indicator species, its important role in the food web of Grice Bay as the primary detritivore and as 88 a prey species for juvenile Grey whales (Eschrichtius robustus) makes it an important organism to monitor. The specific questions addressed in this chapter are: 1. Does reproduction and distribution in Neotrypaea californiensis vary temporally, following a predictable seasonal cycle? 2. Does the spatial distribution of N. californiensis vary predictably across the mudflat correlated to abiotic factors (tidal height, sediment grain size)? 3. Can N. californiensis density be expressed as a function of burrow density in order to predicted the number of ghost shrimp from the number of burrow holes? 4. Is N. californiensis a good indicator species? What are the most informative population parameters to monitor changes in environmental quality? 4.2 METHODS 4.2.1. Study site Core samples were taken from the mudflat to assess the Neotrypaea californiensis population in Grice Bay, British Columbia. Grice Bay is part of Clayoquot Sound at the southeast end of Tofino Inlet and is approximately 10 km from the open ocean. The bay is shallow and dries almost completely at low tide exposing two dominant habitats, the seagrass beds composed of Zostera marina and associated algae and fauna and the mudflats dominated by the burrowing ghost shrimp, N. californiensis. This site was chosen as it is part of Pacific Rim National Park Reserve of Canada and there is therefore interest in preserving this habitat through monitoring and management. Ghost shrimp were sampled at two replicate sites within Grice Bay (Figure 4.1). Both sites are located 89 on the southwestern shore of Grice Bay. Site 1 is located in a small bay with salt marsh vegetation marking the beginning and seagrass marking the end of the mud flat. Site 2 is not in a noticeable bay and has less total area, with the 900 m transects often extending further into the seagrass at the deeper end of the mudflat. Site 2 is also located further towards the end of Grice Bay. Figure 4.1 Outline of Grice Bay showing the two replicate sites in the mudflat, two transects in each site and one sampling position in each of the six zones on each transect. The transects were 900 m long. 4.2.2 Sampling methodology The ghost shrimp were sampled approximately every two months, from July 2001 to August 2002. Sampling occurred from the boat with a K C hand operated sediment corer (KC Denmark) of 13 cm diameter and a maximum depth of 50 cm. Since the actual 90 depth of the core was often limited by the presence of clay, the actual depth of each core was recorded. Three permanent transects were established at the two sites in Grice Bay. Each transect extends from the shore into the seagrass bed (900 m). They were divided into six zones designated by distance from shore (each zone is 150 m long). On each sampling trip, two transects in each site were randomly chosen for sampling. One station per zone per transect was also randomly chosen using GPS co-ordinates. At each of these stations, seven cores are taken haphazardly around the boat. These seven cores are considered independent from each other and are therefore used as replicates in the analysis. The sediment from each core was sieved through mesh with 2.5 mm hole size and the number of ghost shrimp was recorded. Ghost shrimp were measured to the nearest mm total length and the sex of all individuals was recorded. Individuals were classified as females i f pleopods were present on the first abdominal somite (Ruppert & Barnes 1994). After preliminary analysis it was observed that individuals less than 21 mm were predominantly recorded as male and it was therefore uncertain whether these animals were male or had not yet developed visible pleopods. Individuals less than 21 mm long were considered juveniles in subsequent analysis. The reproductive condition of females (whether or not they were carrying eggs on the outside of their body) and the development stage of eggs (red, orange or orange with eyespots) was recorded. A subsample of 200 Neotrypaea californiensis individuals were measured and dried for 24 hours at 70°C. Dry weight was found to be a power function of total length [wet weight = 5x10~7 (length)3'5] and this function was used to estimate biomass. This was less accurate than drying all the ghost shrimp we caught, but less destructive. Photos of 10 x 91 10 cm quadrats were taken to determine i f a linear density-independent relationship between burrow density and ghost shrimp density and biomass could be found. These relationships were determined by regressing the mean number of burrow holes with the mean of the ghost shrimp variable at each station. In June and August 2002, burrow counts were taken from individual cores to find a more precise relationship between ghost shrimp density and biomass and the number of burrow holes in the mudflat. Sediment was sampled from the midpoints of the zones along four transects in April 2002. Sediment was extracted to 50 cm and analysed in 10 cm portions. This is a one time only measurement as it is assumed that this sediment size remained constant over the time of this study. A l l sediment was initially treated in 10% HC1 for 15 minutes to remove carbonate material before drying. The size fraction of the mudflat sediment was determined by sieving the sediment through sieves # 10, 35, 60 and 230. Replicate subsamples were analysed to minimise error. Percent combustibles in each sample was determined by ashing the sediment at 500°C for 5 hours. 4.2.3 Data analysis Statistical analyses were performed to identify differences in density, size structure, sex ratios and reproductive potential of Neotrypaea californiensis between the six zones and eight sampling periods. As in the seagrass analysis, the data were highly non-normal due to the occurrence of many zeros and could not be forced into normality. Therefore, most of the analysis was conducted using multiple Mann-Whitney tests. To correct for the increasing error, the a value was set to 0.05 / # of pairwise tests. Ghost shrimp density was also regressed on burrow density to determine i f ghost 92 shrimp density can be predicted from a known burrow density. The standard regression test was used, as it is robust for non-normal data. A Chi-squared test was used to determine i f the ghost shrimp sex ratio was significantly different from the predicted 1:1 ratio. Mean values (Appendix 3) and the results of the statistical analysis (Appendix 6) are found in the appendices at the end of the thesis. 4.3 RESULTS 4.3.1 Ghost shrimp distribution in Grice Bay a. Temporal patterns Ghost shrimp density and biomass increased from May to November, decreased . until April and then increased again until August (Figure 4.2; Appendix 6.1). The seasonal trend was more similar between Site 1 and 2 for density then it was for biomass. When the zones were considered separately, no consistent seasonal trend was apparent for either ghost shrimp density or biomass. This suggests that seasonal effects only occur as a general averaging out of variable patterns occurring within each zone. There are more significant differences between sampling dates when considering ghost shrimp density compared to ghost shrimp biomass. 93 M a y Sep Feb 2001 - 2002 Jun M a y Sep Feb 2001 - 2002 Jun Figure 4.2 M e a n d e n s i t y a n d b i o m a s s o f a l l Neotrypaea californiensis i n d i v i d u a l s i n S i t e 1 a n d 2 i n e a c h o f t he s a m p l i n g d a t e s . E r r o r b a r s a re o n e s t a n d a r d e r r o r o f t he m e a n . b. Spatial patterns G h o s t s h r i m p d e n s i t y a n d b i o m a s s w a s g rea tes t i n t he m i d d l e o f the b e d ( z o n e 3 & 4) as o p p o s e d to n e a r t he s h o r e ( z o n e 1) o r t he s e a g r a s s b e d ( z o n e 6) at b o t h s i t es i n G r i c e B a y ( F i g u r e 4.3; A p p e n d i x 6.1). O v e r a l l , m o s t o f t he p a i r w i s e c o m p a r i s o n s b e t w e e n z o n e s w e r e s i g n i f i c a n t l y d i f f e r e n t . T h e o n l y n o n - s i g n i f i c a n t d i f f e r e n c e s l a y b e t w e e n a d j a c e n t z o n e s a n d the o u t e r m o s t z o n e s . 2 3 4 Zone Figure 4.3 M e a n d e n s i t y a n d b i o m a s s o f a l l Neotrypaea californiensis i n d i v i d u a l s i n b o t h s i tes a c r o s s t h e s i x z o n e s . E r r o r b a r s a re o n e s t a n d a r d e r r o r o f t h e m e a n . 94 4.3.2 Ghost shrimp population structure a. Size Using size as a proxy for the age structure of the population, there are some differences over time and between zones on the mudflat. Overall, mean shrimp length at all zones combined was between 21 and 27 mm. Shrimp length at both sites was found to decrease between May and July 2001, increase from July to September 2001, decrease from September to November and increase between June and August 2002 (Figure 4.4a; Appendix 6.2). Trends between November and June were not consistent between sites. Spatially, mean total length was greater in zone 1 then 2 and then increased again in zones 3 and 4 at both sites (Figure 4.4b; Appendix 6.2). In site 1, mean length was greater in zone 5 then in zone 4, and mean length in zone 6 was significantly less than zone 5. Conversely, at Site 2 mean shrimp length decreased significantly in zones 5 and 6 to 13.8 and 12.8 mm respectively from a high of 24.2 mm in zone 4. This is consistent with the fact that at Site 1, stations in zone 6 were sometimes found in the seagrass bed but not often. Conversely, at Site 2, stations in zones 5 and 6 were often located in the seagrass bed. a) Date b) Zone — . — Site 1 Site 2 ~a e2 15 M a y July Sept N o v Feb A p r June A u g 1 2 3 4 5 6 2001 - 2002 Zone Figure 4.4 Mean total length of all Neotrypaea californiensis individuals in Site 1 and Site 2 by sampling date (a) and by zone (b). Error bars are one standard error of the mean. 95 When the density of the adult and juvenile populations in sympatry are compared, the seasonal cycles of these two life history stages follow the same general trend of increased density and biomass in spring and summer and decreased biomass in autumn and winter (Figure 4.5; Appendix 6.2). However, the absolute values and times when the population increases and decreases is different between the juvenile and adult populations. The low ability to extrapolate trends between the two replicate sites suggests that the variability over the entire mudflat is not predictable. a) Site 1 b) Site 2 1200 •o 800 I 400 H ID Q >— Juvenile >— Adul t M a y July Sept N o v Feb A p r June A u g M a y July Sept N o v Feb A p r June A u g 2001 - 2002 Figure 4.5 Seasonal trends of adult and juvenile Neotrypaea californiensis population densities (# ind / m 3) at site 1 (a) and site 2 (b) in Grice Bay. Error bars are one standard error of the mean. Adult and juvenile density follows similar spatial patterns, with greater densities in the central zones (3 and 4) and lower densities in the outer zones (Figure 4.6; Appendix 6.2). Juveniles were defined as ghost shrimp < 21 mm because ghost shrimp smaller than this size could not be consistently separated into sexes. At site 1, adult and juvenile densities follow relatively parallel trajectories with a major exception in zone 6 where adult density drops significantly below juvenile density. At site 2, this phenomenon is much more apparent when adult densities effectively drop to zero in 96 zones 5 and 6, while juvenile densities remain elevated in zones 5 and 6 with 426 and 215 juveniles /m respectively. The other significant difference between juveniles and adults is a 1.6 fold difference between juvenile and adult density in zone 2, compared to the much smaller differences in the other zones. a) Site 1 b) Site 2 1500 1 2 3 4 5 6 Zone 1 2 3 4 5 6 Figure 4.6 Spatial trends of adult and juvenile Neotrypaea californiensis density (# ind / m ) across the 6 zones at site 1 (a) and site 2 (b) in Grice Bay. Error bars are one standard error of the mean. b. Sex ratios There were always more than 50 % females, across all sampling dates and at all sites, however this trend was not significant (Appendix 6.2). This difference in abundance between the two sexes was greater during both summer periods when compared to the winter (Figure 4.7a; Appendix 6.2), which coincides with the period of reproduction for Neotrypaea californiensis. The proportion of females was greater in zones 2 and 3 then all other zones (Figure 4.7b; Appendix 6.2), which are the zones of highest ghost shrimp density. 97 a) Date b) Zone M a y July Sept N o v Feb A p r June A u g 2001 -2002 3 4 Zone Figure 4.7 The proportion of the Neotrypaea californiensis population that is female at each site and variation over time (a) and space (b). Error bars are one standard error of the mean. 4.3.3 Ghost shrimp reproductive potential Reproduction in the Neotrypaea californiensis population in Grice Bay occurred primarily in May and June in the two years studied (Figure 4.8; Appendix 6.3). Reproduction dropped significantly from May to July and only one ovigerous female was found in August and two in September. Reproduction ceased completely in winter, with no reproductive females being found in November and February. There was high interannual variability in ghost shrimp reproduction between the two summers studied, as 65 % of females were ovigerous at one time in 2001 and only 21 % in 2002. 98 M a y July Sept Nov Feb A p r June A u g 2001 - 2002 Figure 4.8 Proportion of Neotrypaea californiensis females carrying eggs at the two study sites throughout the sampling period from May 2001 to August 2002. Error bars are one standard error of the mean. A greater proportion of females was ovigerous in zone 4 (and 5 in site one) than other zones in both sites (Figure 4.9; Appendix 6.3). This spatial trend is not consistent every sampling date however, indicating that this zonation may change over the reproductive season, or that there is just too much variability to identify a pattern. These were also the zones with the greatest overall biomass. Site 1 Site 2 ft 0 4— —r— —r— —i 1 2 3 4 5 6 Zone Figure 4.9 Proportion of Neotrypaea californiensis females carrying eggs in the six zones at the two sites. Error bars are one standard error of the mean. 99 4.3.4 Ghost shrimp and other infauna Apart from Neotrypaea californiensis the other most abundant organism in the mudflat was the commensal clam Cryptomya californica. C. californica density was found to regress significantly on both adult and total N. californiensis density (p < 0.0005) (Figure 4.10; Appendix 6.4). However, the small r 2 values indicate that N. californiensis density explains less than half of the variation in C. californica density (Figure 4.11). a) Site 1 b) Site 2 10000 -o .S c a 7500 5000 2500 -O— Cryptomya • Adult Neotrypaea A All Neotrypaea Figure 4.10 Cryptomya californica density (# ind / m 3) compared to adult and total ghost shrimp density across the six zones in the two sites. Error bars are one standard error of the mean. a) C I 6000 4000 5, 2000 y = 3.40x+474 • • K. = R = 0.47 500 1000 Adul t Neotrypaea density (perm 2 ) b) 6000 I 4000 2000 y = 2.44x+ 134 R = 0.49 500 1000 1500 Neotrypaea density (perm ) Figure 4.11 Regression of Cryptomya californica density on Neotrypaea californiensis density [adult (a) and all individuals (b)], in June and August 2002. 100 The other organisms found in the mudflat to varying degrees include the cockle, Clinocardium nuttalli, polycheates including Arenicola claparedii, Nephthys sp. and tube worms of the Sedentaria order, the mudflat crab Hemigrapsus oregonensis and several unidentified copepods. Sampling for ghost shrimp closer or in the seagrass beds revealed different organisms, such as amphipods including the skeleton shrimp, Caprellidae and the eelgrass isopod, Idotea resecata. 4.3.5. Burrow counts Ghost shrimp density was found to regress significantly on the number of burrow holes in the mudflat in July, September and November 2001 (p<0.016), when combining the two summer dates (July 2001 & June 2002; p=0.001) and all dates together (p<0.0005) (Figure 4.12; Appendix 6.5). In these cases, the number of burrow holes explained between 32 and 80 % of the variability in the ghost shrimp density. Overall there are 2.52 ± 0.20 ghost shrimp individuals per burrow, however this number varies between summer (Apr to Aug), 1.75 ± 0.15 (mean ± S.E.) ghost shrimp per burrow and autumn (Sept & Nov), 2.97 ± 0.29 ghost shrimp per burrow. Ghost shrimp biomass was also found to regress significantly on the number of burrow holes for all dates sampled (p<0.05), with r values ranging from 0.30 to 0.92 (Figure 4.12). Overall, there was 0.13 ± 0.01 g dw of ghost shrimp per burrow. 101 a) A l l dates y=0.92x+916 y=0.11x+18 0 1000 2000 3000 0 1 0 0 0 2 0 0 0 3 0 0 o Burrow density (per m ) Burrow density (per m ) b) Summer only (June, July & Aug) 2000 Burrow density (per rn ) B u r r o w d e n s i t y <P e r m > Figure 4.12 Regression of ghost shrimp density and biomass on burrow density using data from all dates combined (a) and summer values only (b). In June and August 2002, a more precise method of estimating the relationship between burrow holes and ghost shrimp biomass and density was used. Ghost shrimp density regressed significantly on the density of burrow holes in June (p=0.006) and August (p=0.016), however ghost shrimp biomass only regressed significantly on burrow hole density in June (p=0.045), but not August (p=0.64) (Figure 4.13; Appendix 6.5). Using this method, there were 1.37 ± 0.10 ghost shrimp individuals per burrow and 0.13 ± 0.01 g dw of ghost shrimp per burrow in June and August combined. 102 a) June 2002 •S 3000 c u a 2000 -S a IOOO i 0 y=0.7584x+430.23 R 2 = 0.4302 400 300 H y =0.1293x-29.659 2 O Q. P E Z. 200 H « ^ 100 0 500 1000 1500 2000 2500 2 Burrow density (per m ) 6 o 0 500 1000 1500 2000 2500 2 Burrow density (per m ) b) August 2002 e T J 3000 2000 •f= & iooo 6 y =0.621 l x + 967.29 R 2 = 0.1442 * 0 500 1000 1500 2000 2500 2 Burrow holes (perm ) 600 400 200 y =-0.0304x+ 226.48 • R = 0.0059 0 500 1000 1500 2000 2500 2 Burrow holes (per m ) Figure 4.13 Regression of ghost shrimp density and biomass on burrow density using the more precise methodology of counting the burrow holes of each individual core in June (a) and August (b) 2002. 4.3.6 Physical factors affecting distribution Mean sediment grain size varied across the mudflat, being larger further from shore (Figure 4.14 dotted line). Ghost shrimp biomass regressed significantly upon mean sediment grain size (p<0.0005), but ghost shrimp density did not (p=0.1). However, sediment grain size only explained 8 % of the variability in ghost shrimp biomass (Figure 4.15; Appendix 6.6). The sediment depth that was penetrable with the core was assumed to be the sediment depth that was available to the ghost shrimp. This depth also varied across the mudflat, being greatest in zones 2, 3 and 4 (Figure 4.11 solid line). Both ghost shrimp density and biomass regressed significantly on sediment depth (p<0.0005), but 103 sediment depth only explained 16 and 28 % of the variability in ghost shrimp density and biomass respectively (Figure 4.16; Appendix 6.6). 0.2 0.15 .s 2 60 B T3 0.1 0.05 a) Site 1 0.75 0.5 0.25 CL -a -o C/2 - • - - - Sediment grain size Sediment depth 1 2 3 4 5 6 1 2 3 4 5 6 Zone Figure 4.14 Mean sediment grain size (mm) of the top 10 cm at the midpoints of the zones on two transects per site (dotted line, left axis) and mean sediment depth (cm) of these cores (solid line, right axis) across the size zones in both sites. Sediment depth refers to the depth of ghost shrimp activity, below which sediment is very compact suggesting that ghost shrimp are not burrowing deeper or not able to burrow deeper. Error bars are one standard error of the mean. Sediment grain size Sediment grain size Figure 4.15 Regression of ghost shrimp density and biomass on sediment grain size. 104 .« 6000 n CO -I ^ 5 0 0 0 £-"9 4000 S j= c 3000 ~ W 2000 1000 0 o 6 y=3363x+364 « 0.173 0.2 0.4 Core depth (m) 600 500 400 300 H 200 100 0 y =370x-34 R 2 =0.281 0.6 0 0.2 0.4 Core depth (m) 0.6 Figure 4.16 Regression of ghost shrimp density and biomass on core depth (assumed to be the depth penetrable by the ghost shrimp). The elevation of the mudflat also varies predictably over the study sites, being greatest at the upper edge of the mudflat and lowest at the lower edge. Ghost shrimp density increased significantly with decreasing elevation from the upper edge of the mudflat to 1.2 m above M L L W (p<0.0005) and then decreased significantly with further decreasing elevation to the lower edge (p<0.0005) (Figure 4.17; Appendix 6.6). Ghost shrimp biomass increased significantly with decreasing elevation only until 0.8 m above M L L W (p<0.0005) and then no relationship was found between ghost shrimp biomass and mudflat elevation. 105 o 6 3 2 1 0 -1 - C 400 -i 00 3 2 1 0 - 1 shallow deep Mudflat elevation (m) Figure 4.17 Ghost shrimp density and biomass regressed on mudflat elevation. Ghost shrimp density has a significant negative relationship with mudflat elevation until 1.2 m above M L L W and a significant positive relationship with elevation below 1.2 m above M L L W . Ghost shrimp biomass has a significant negative relationship with elevation until 0.8 m above M L L W and no relationship existed deeper than 0.8 m. 4.4 DISCUSSION 4.4.1 Ghost shrimp distribution a. Variability in distribution across space Concentrations of Neotrypaea californiensis are known to reach high values in mudflats on the west coast of North America. Due to their bioturbation impact in the sediment, these shrimp create distinct habitats in which only certain other infauna can survive and which are typically devoid of plants and resident epifauna. The lower 106 boundary of the ghost shrimp bed often presents a marked decline in density from hundreds of ghost shrimp to almost none over a distance of a couple metres (Posey 1986b). Density is often highest immediately above the upper limit of the eelgrass zone, with density declining abruptly in the seagrass bed (Swinbanks & Luternauer 1987). In Grice Bay, density was highest in the zones at intermediate distances between the salt marsh and seagrass, but biomass values were greatest adjacent to the seagrass beds, indicating that large individuals inhabited this area preferentially. The significant differences in ghost shrimp distribution between the zones in this study likely reflects habitat quality, with the contributing factors possibly including substrate size, depth and food availability. Because juvenile and adult distribution follows similar trends, these factors seem to be important to the entire life cycle of ghost shrimp. In an Oregon estuary, Neotrypaea californiensis density and biomass have been found to be highest close to the mouths of estuaries and decrease with distance upstream, reflecting the distribution of the benthic food supply (Bird 1982). Predators may also play an important role in the structuring of the ghost shrimp population. Through the use of predator exclusion cages, Posey (1986a) determined that predators, such as the staghorn sculpin, Leptocottus armatus, were limiting the lower intertidal distribution of N. californiensis on an Oregon mudflat from spring to early autumn (Posey 1986a). Although predators are absent in winter, the reduced activity of ghost shrimp during this time was thought to limit the expansion of its distribution into the lower intertidal (Posey 1986a). Other predators of burrowing shrimp include cutthroat trout, Salmo clarkii, Dungeness crabs, Cancer magister and Western gulls, Larus occidentalis (Posey 1986a). In Grice Bay, dungeness crabs and fish may prey on 107 ghost shrimp in the lower intertidal, while gulls and other shore birds may have a strong influence on the ghost shrimp population in the higher reaches of the mudflat. The grey whale, Eschrichtius robustus, has also been observed to feed extensively on ghost shrimp in Grice Bay, although this is unusual, with only one other reported ghost shrimp feeding area in Puget Sound, Washington (Weitcamp et al. 1992; Darling et al. 1998). b. Seasonality in ghost shrimp distribution In this study, a reduction in ghost shrimp density and biomass was apparent during the winter. Temperature likely reduces Neotrypaea californiensis activity as burrowing and feeding have been shown to be reduced in winter and greatest during summer. During winter, the number of holes per shrimp declines, as evidenced by both lower hole-to-shrimp ratios and lower hole densities in dense beds (Posey 1986a). Ghost shrimp also move deeper into their burrow (McCrow 1972) and produce fewer eggs between December and March (Bird 1982). In this study, no reproductive females were found in winter. Lower salinities, which may be present during the winter due to greater precipitation in British Columbia, also reduce ghost shrimp activity under laboratory conditions (Posey 1987). In this study, density and biomass did not differ significantly between the two consecutive summers. Studies in Oregon have also found that Neotrypaea californiensis populations show little variation in distribution and body size over the years. Alternatively, others have suggested ghost shrimp populations may be cyclic in abundance and that these cycles correlate with changes in the physical characteristics of the mudflat and may take as long at 10 years (Bird 1982). 108 4.4.2 Ghost shrimp population structure a. Ghost shrimp size structure Significantly larger ghost shrimp are found in the centre of the Grice Bay mudflat, while only very small ghost shrimp were found in the seagrass bed closest to the mudflat. Therefore, the optimal habitat, likely in terms of food availability, is at the centre of the mudflat. Either juveniles are excluded to the deep margin due to competition with larger individuals or their occurrence in the seagrass bed is random, due to passive settlement processes. It is also unknown i f the ghost shrimp are able to migrate from this sub-optimal habitat as they grow or i f ghost shrimp simply do not survive in these zones after a certain size. Bird (1982) argued that the large disparities in size between individuals from different cohorts could force smaller animals into less preferred habitats to explain the apparent displacement of juveniles from the area near the mouth of the estuary (Bird 1982). b. Ghost shrimp sex ratio Neotrypaea californiensis is sexually dimorphic in terms of claw size and overall size. The master claw is exceedingly large in males and may approach 25 % of total body weight in mature individuals, but rarely exceed 10 % in females. The shape of the gape of the master claw and the presence of fine teeth about its periphery, strongly suggests it functions in a highly stereotyped form of grappling during aggressive encounters between similar-sized conspecifics fighting over females (Labadie & Palmer 1996). Due to this aggressive behaviour and the chance males take in exposing themselves to predators when exiting their burrows in search for females (Wenner 1972), 109 males have higher mortality rates than females, as seen in the unequal sex ratios observed in the field (Bird 1982; Dumbauld et al. 1996). Results in this study indicate that the population always consisted of greater than 50% females throughout the year and in all areas surveyed. These results agree with those found other field studies of Neotrypaea californiensis in Washington and Oregon (Bird 1982; Dunbauld et al. 1996). In general, a skewed sex ratio among mature animals in favour of females is the norm in marine crustacean populations. 4.4.3 Ghost shrimp reproduction As the mating behaviour in thalassinid shrimps remains undescribed (Dumbauld et al. 1996), studies on reproduction of Neotrypaea californiensis have focused on ovigerous females, larval and postlarval stages. N. californiensis females are ovigerous from April through August (Feldman et al. 2000). The eggs are carried by the female for 5 to 6 weeks (Bird 1982) until the embryos have reached the zoea stage, when hatching takes place (MacGinitie 1934). Eggs begin to hatch in June and planktonic zoeae are released primarily during the night ebbs of neap tide series and exported to nearshore coastal waters (Johnson & Gonor 1982). In general, northern ghost shrimp populations have a relatively late breeding season compared to populations further south (Berkenbusch & Rowden 2000). It appears that initiation and length of the breeding season in Neotrypaea californiensis and other thalassinideans is not necessarily determined by a relative increase in temperature as . generally believed to be the case for marine invertebrates, but may be linked to food availability (Bird 1982; Berkenbusch & Rowden 2000). Low temperatures apparently 110 did not preclude breeding, since many sites in Oregon had low frequencies of ovigerous females during the winter (up to 12 %) (Bird 1982). In this study, reproduction peaked in May and June and had almost ceased completely by August. There were no ovigerous females found in November and February. Differences in the timing of the reproductive cycle between locations within estuaries have been observed in Neotrypaea californiensis, with populations closer to the estuary mouth extruding eggs earlier than those within the estuary (Bird 1982; Dumbauld et al. 1996). Callianassa filholi, from New Zealand, has also shown differences in the reproductive timing between populations at different locations. In this study, no differences in timing of reproduction were found between zones. This may be because the population was not sampled intensely enough in this study to detect subtle differences or because differences only exist on larger geographical scales, such as within Clayoquot Sound or between populations along the coast of North America When the eggs hatch the larvae are retained in the water column for over a month (McCrow 1972; Johnson & Gonor 1982). Because Grice Bay dries almost completely at low tide, exposing the mudflat completely and the seagrass bed below partially, it is assumed that larvae must be washed out of the bay completely. Johnson & Gonor (1982) found a net export of larvae from their study estuary, with no retention in the estuary. While larvae continually re-enter the estuary, only megalopae entering the estuary a few days before settling would be retained. Genetic exchange between Neotrypaea californiensis populations is therefore both an extensive and annual event. Since the mudflats in Grice Bay are the largest in the area, this population may act as a source of N californiensis for nearby populations in sub-optimal habitats. I l l 4.4.4 Ghost shrimp and other infauna Very few other species of macro-invertebrates were found in the mudflats of Grice Bay. In this study, the most abundant commensal is the small suspension-feeding clam, Cryptomya californica may reach densities of 1200 individuals/m2. In other studies, these densities reflect those of the host (Bird 1982). In several estuaries in Oregon, 35 species total were found within the Neotrypaea californiensis habitat, ranging from 17 to 32 species per site. Mobile deposit feeders dominated at all sites, although all feeding types (sedentary, mobile, commensal, suspension feeders, and deposit feeders) were present in each colony (Bird 1982). In this study, ghost shrimp density only explained half of the variability in the density of its small commensal clam, C californica. However, in an experimental manipulation the association between with ghost shrimp was obligatory for C. californica (Peterson 1977). The commensal organisms presumably derive shelter from the burrow and some may steal food from the host (MacGinitie & MacGinitie 1949). Individuals that rely on other species to maintain the connection at the surface-water interface have more energy available for growth and reproduction, thus strengthening the commensal relationship (Bird 1982). Burrows may also provide shelter from predators and harsh physical conditions such as high surface temperature, low tides and low salinity (Bird 1982). 4.4.5 Ghost shrimp burrows Counting burrow holes has frequently been used as a fast and non-destructive means of assessing shrimp density. Dumbauld et al. (1996) found an average of 1.2 burrow openings per individual Neotrypaea californiensis in Willapa Bay, Washington, 112 which was lower than the 2.5 burrow openings per individual found in Boundary Bay, British Columbia by Swinbanks & Murray (1981). The relationship between burrow count and shrimp density in Willapa Bay was extremely variable and no correlation was found in the winter when shrimp activity is reduced and increased wave exposure caused burrow openings to collapse (Dumbauld et al. 1996). In this study, there were between 1.37 and 1.75 individual shrimp per burrow in the summer months, depending on the method used to count burrows. In the winter, there was still a significant regression, however the number of shrimp per burrow holes increased to 3.2. 4.4.6 Mudflat sediment characteristics The distribution of organisms in soft bottom environments was originally thought to be controlled by sediment size and stability. Although the picture now appears to be much more complex, the nature of the sediment is still considered an important factor for infaunal organisms (Little 2000). As the tide changes, the sediment acts as a buffer against simultaneous changes in salinity, temperature and pH. Organic material accumulates in the sediments creating a food source. Finer sediments, such as mud, usually retain water at low tide and desiccation is less of a problem than in coarser sediments such as sand (Little 2000). In sand, water drains quickly and at low tide, desiccation becomes a possibility unless the burrow extends deep enough. Conversely, most of the water is retained in mud even at low tide and consequently, the burrows are not required to reach as great a depth (Griffis & Chavez 1988). The high-intertidal coarse-sand habitats occupied by Neotrypaea californiensis drain quickly and retain little water during low tide. The water 113 level in sands during an average low-low tide drops an average of 34 cm below the surface of the sediment, while there was little or no drop in water level in mud sediments (1.33 cm) (Griffis & Chavez 1988). Changes in burrow morphology with different sediment types have also been observed for several species (Griffis & Suchanek 1991). In a laboratory experiment, Griffis & Chavez (1988) showed that the burrows of N. californiensis were of greater volume when produced in fine-grained, muddy sediments than in coarser sands. The different architecture due to sediment type was thought to be due to the water holding capacities of the different sediments. Several studies have observed changes in burrow characteristics with changes in tidal height. The rapid sediment reworking by Neotrypaea californiensis results in a more homogeneous profile of sediment organic matter with depth. N. californiensis inhabits a broad spectrum of habitats with respect to particle size distribution, organic content and depth (Bird 1982). Swimbanks & Luternauer (1987) did not find that burrow opening density was correlated to grain size parameters, as densities > 50 /m2 were observed in sediments that ranged from 5 - 5 0 % mud content and 0.063 - 0.165 mm in median grain size. The highest densities of burrows occurred in the dominantly sand sediments (~ 10% silt/clay). While there were significant differences in sediment size in areas occupied by ghost shrimp, the shrimp are able to inhabit a wide range of sediment types including granular sediment found in smaller bays in the area. Consequently, it may be the depth of the sediment that controls biomass accumulation, with higher success of shrimp that are able to penetrate deeper into the sediment, to escape predation or due to behavioural characteristics. 114 4.5 M A N A G E M E N T C O N S I D E R A T I O N S Mudflats are far from static entities and represent areas of changing balance between erosion and deposition. The unstable nature of the sediment requires organisms to be physiologically flexible in this environment. The mudflats of Grice Bay, along with many of those along the west coast of North America, support primarily the burrow ghost shrimp, Neotrypaea californiensis along with a select few other species that can survive the constant reworking of the sediment. Ghost shrimp are important organisms due to their important role in structuring mudflats. Their bioturbation activity aerates the sediment and reduces sediment compaction. Ghost shrimp are instrumental in the food web of Grice Bay. They are the primary detritivores in the bay and therefore breakdown much of the organic matter. Ghost shrimp are also prey items to fish, dungeness crabs, migrating water birds and juvenile Grey whales. Ghost shrimp habitat has been lost in the past due to development, including landfill and dredging. Early research on Neotrypaea californiensis focused on how this species is harmful to the oyster industry in Washington State. The young commercial oysters (Crassostrea gigas) are smothered by the material thrown up in the process of burrow excavation (Stevens 1929). As shrimps burrow through the sediment, compaction is reduced to the point that oysters growing directly on the benthos sink into the unconsolidated mud (Stevens 1929; Feldman et al. 2000). Since the 1960s, ghost shrimp populations in Washington have been controlled with the chemical Sevin (WA Dept Fisheries 1970). While the burrowing thalassinidean shrimp are killed by the pesticides, so are the subadult Dungeness crabs (Cancer magister), juvenile English sole 115 (Parophrys vetulus) and other non-target species present on the mudflats at the time of application. Commercial crabbers and other groups who have economic, recreational and environmental interests in the bays have generally opposed use of the chemical, however oyster growers maintain it is essential to sustain production levels (Simenstad & Fresh 1995;Feldmanet al. 2000). While there is no published literature on the use of ghost shrimp as indicator species, the use of crustaceans in beach habitats has received some attention because these organisms are sensitive to declines in water quality. A reduction in water quality would likely affect the survival, growth and reproductive capabilities of a ghost shrimp population. Neotrypaea californiensis density and biomass would likely decline under conditions of stress. A n analysis of shrimp size can provide comparative information about a species living in an impacted compared to non-impacted habitat (Wenner 1988). Reduced survival and/or reduced growth would likely be seen in a change of the size frequency of the population, with large individuals absent and overall density lower. Reproduction in crustacean populations is also known to be highly sensitive to environmental quality in terms of timing, frequency and success. In poor quality habitats, the reproductive season might be shorter than in a non-impacted area. The food supply may be affected by a pollutant and energy could be diverted to maintenance metabolism rather than reproduction. Alternatively, females under stress might not be able to convert food energy into eggs. Females living in a stressed habitat could be smaller when first capable of producing eggs than those females living in non-impacted areas. Due to a reduction in survival, older females may not be present leading to egg production only by younger females when they first become reproductive (Wenner 1988). 116 Since females generally produce more eggs per clutch as they grow, the overall reproduction in a degraded habitat will be reduced. As eggs are carried externally during incubation, egg quality can also be affected directly by adverse environmental factors (Wenner 1988). The loss of eggs by females can be estimated by a quantification of the percentage of mature eggs (orange with eyespots) compared to immature eggs (red to orange) during the reproductive season. Finally, crustaceans are often used as biomonitors of accumulating toxins in the environment. As ghost shrimp live surrounded by the sediment, any accumulation of toxic chemicals, such as organics or heavy metals, in the sediments will likely accumulate in the shrimp. If these toxins cause adverse developments in the ghost shrimp, this would be revealed by population studies and signal a serious adverse effect of the pollutant. However, crustaceans are often able to accumulate pollutants at high levels without showing i l l effects, while these toxins are causing other organisms, such as shore birds, to suffer. While this study focuses on population parameters, i f pollutants are suspected to be present in the environment, due to industry run-off or a point source pollutant, ghost shrimp can be used as biomonitors for toxic chemicals by tissue analysis (Wenner 1988). 117 C H A P T E R 5 M A N A G E M E N T RECOMMENDATIONS F O R M U D F L A T S AND SEAGRASS BEDS The baseline study conducted in Grice Bay and presented in this thesis has been effective in documenting seasonal fluctuations and spatial differences in population parameters of the seagrass Zostera marina and the ghost shrimp Neotrypaea californiensis. This has allowed preliminary analysis of the spatial and temporal patterns of these populations, the assessment of monitoring techniques and baseline data against which to compare future trends. Ghost shrimp and seagrass populations in Grice Bay may be useful as indicators of the water quality in the bay and may be used in future monitoring programs for assessing the long-term maintenance of Ecological Integrity. 5.1 Seagrass and ghost shrimp as indicators In Chapter 1, the criteria for choosing a health indicator species was discussed (Table 1.1). The most important criterion is sensitivity to environmental quality and changes in environmental variables. Both seagrass and ghost shrimp are organisms that are sensitive to water quality. Seagrass populations are in decline all over the world due to water quality deterioration. These declines mean decreased shoot density and biomass of seagrass populations. Human impacts may affect sexual reproduction in terms of duration of the reproductive season and frequency of reproductive shoots. These changes are often drastic and may be best detected early by looking at physiological parameters, which are 118 the first response of a plant to stress. When water quality declines, seagrass plants will use up their sugar reserves, increase total chlorophyll concentration and decrease chlorophyll a:b ratios before large changes in population structure are apparent. A reduction in water quality would likely affect the survival, growth and reproductive capabilities of a ghost shrimp population. Neotrypaea californiensis density and biomass would likely decline under conditions of stress imposed upon the population (Wenner 1988). Reduced survival and/or reduced growth would likely be seen in a change of the size frequency of the population, with large individuals absent and a lower overall density. Reproduction in crustacean populations is also known to be highly sensitive to environmental quality in terms of timing, frequency and success. In poor quality habitats, the reproductive season might be shorter than in a non-impacted area. The food supply may be affected by a pollutant and energy could be diverted to growth and survival, rather than reproduction. Females living in a stressed habitat could be smaller when first able to produce eggs and mean size of ovigerous females could be reduced. As eggs are carried externally during incubation, egg quality can also be affected directly by adverse environmental factors (Wenner 1988). The second criterion is low levels of random variability. Variability in all the sensitive parameters was measured in this study and their means and standard errors are given in this thesis. This is now the baseline data and i f change beyond these expected values are detected then it would need to be determined why these changes have occurred and how to manage the situation. Both of these species have a well-known biology, so i f changes in these parameters are detected, we will be able to come up with hypothesis to deduce why based on past research. Experimental protocol is also established for both 119 species, making experimental work needed to determine cause and effect relationships highly possible. Finally, both of these species are abundant in Grice Bay, making them relatively easy to sample. While traditionally these populations have been sampled destructively, it is possible to sample quickly, efficiently and non-destructively. Seagrass density, size and reproductive status can be easily counted and measured in the field by researchers at low tide. Ghost shrimp populations can be enumerated based on burrow holes. Furthermore, the seagrass/mudflat interface can be monitored for shifts in this boundary to detect changing environmental conditions and provide further information on the relationship between these two communities. 5.2 Importance of ghost shrimp and seagrass in Grice Bay This project is part of a larger interdisciplinary project aimed at assessing the cumulative human impacts in Grice Bay. These studies include other projects in the seagrass beds, including a study on fish community assemblages in seagrass beds and another on seagrass epiphytes. Parks Canada is also developing a Biological Index for stream communities in the area, remote sensing water column organisms for ground truthing satellite photos and undertaking a regular monitoring program of water quality. Seagrass declines around the world have been attributed to widespread deterioration of water quality. Industrialization and increased land use has resulted in heightened levels of nutrient loading, sedimentation, influx of contaminants and toxins and other detrimental effects on these sensitive communities. The loss of these populations is causing the loss of entire communities of epiphytes, macro- and micro-algae, invertebrates and vertebrates that reside or migrate through seagrass beds. 120 However, seagrass beds are an important fish habitat and as such they are offered protection with a No-net loss policy by the Canadian Department of Fisheries and Oceans. However, they are always subject to water quality deterioration that originates outside their boundaries. Ghost shrimp habitat has been lost in the past due to development, including landfill and dredging and since ghost shrimp are not considered a valuable species economically, they are not protected. The ghost shrimp population in Grice Bay is important due to the large area it occupies and their importance in the food web. They are the primary detritivores in the bay and therefore breakdown much of the organic matter. Ghost shrimp are also prey items to fish, dungeness crabs, migrating water birds and juvenile Grey whales. Ghost shrimp are important organisms due to their important role in structure mudflats. Their bioturbation activity aerates the sediment to a great depth and reduces compaction of the sediment. Ghost shrimp may be used in the future as biomonitors of accumulating toxins in the environment i f pollutants are suspected to be present in the environment, due to a point source pollutant or general changes. An analysis of the tissue of ghost shrimp wil l show increases in organics and heavy metals in both the water column and the sediment as an indication of habitat degradation (Wenner 1988). 5.3 Future monitoring in Grice Bay In this study we have identified a suite of population parameters of both seagrass and ghost shrimp that have been shown to be sensitive to water quality in other studies. We have measured these parameters to give baseline information including the natural 121 level of variance and the protocol for efficient sampling in order to develop a regular monitoring program for the seagrass beds of Clayoquot Sound. This research will provide planners and managers from Parks Canada, the Department of Fisheries and Oceans and concerned non-governmental environmental organizations with information on these important primary producers. This study will allow scientists to understand data collected in future monitoring programs and to determine i f changes are occurring in these populations in Grice Bay. It will also be used to determine whether these changes depict an overall declining trend or are spatially occurring, possibly reflecting a point source pollutant or disturbance. Overall this research will augment our knowledge of this important marine habitat and allow for better monitoring and future management. The baseline data that are presented in this thesis and any monitoring program that will develop out of it do not investigate causal mechanisms for the observed spatial and temporal patterns. While theories and correlations have been presented in this thesis, it is important to consider all options when making a management decision and in some cases experimentally based research may be conducted to supplement the long-term population monitoring. Manipulative research of this kind may be lab or field based and can be designed to test hypotheses regarding the causal factors generating the observed patterns and the effects that human activities may have on these biological communities. The future of the seagrass and ghost shrimp monitoring program in Grice Bay depends on continued and regular collection of data by the protocol given here. Sampling should be undertaken with sufficient replicates, based on the variance found in this study. For both seagrass and ghost shrimp populations, monitoring may be carried out in the spring and summer, as this is most efficient from both a resource and biological 122 standpoint. In the summer, there are more Parks Canada staff employed in Pacific Rim National Park Reserve and favourable working conditions will increase accuracy. The summer is also the peak growing season in both populations and conditions during this time may be critical for population survival over the winter, as seagrasses need to accumulate carbohydrate stores and ghost shrimp may need to reach a critical size. Reproduction also occurs exclusively in the spring and summer in both species. Since reproduction is typically very sensitive to abiotic factors, changes in environmental quality will often be apparent in changes in timing or frequency in reproduction. Therefore, a more intense monitoring program during the summer will be able to pick up changes in reproductive parameters. Concurrent measurements of water quality which are already being conducted on a monthly basis in Grice Bay will help in developing hypotheses to predict or explain biological changes. Monitoring will also be integrated with an adaptive management plan of the bay and will be used to collect more data to fill in gaps in the knowledge. During monitoring, experimental projects may be used to answer important questions that arise. Monitoring can be conducted by trained non-biologists. However, at least for the course of one summer, monitoring should be conducted by as few different people as possible to reduce sampling error. Monitoring needs to be overseen by a marine biologist in order to periodically assess the protocol, assumptions and techniques being used. Data should also be analysed regularly by a qualified person and monitoring or management should be adapted as required. Parks Canada policy makers and managers must consider the results from this study and the future monitoring program to implement management and user regulations based on science for the conservation of this marine habitat. 123 In general, as almost every estuary and sheltered inlet is unique, it is important to establish baseline data for the physical and biological components within the area of interest. Without baseline data, it will never be able to be shown conclusively that changes are occurring and in the case of a large environmental disaster, such as an oil spill, the assessment of recovery may not be possible. It is therefore important to monitor changes in this bay and develop ecosystem based management in sheltered marine habitats, especially those encompassed by Parks Canada. 124 L I T E R A T U R E C I T E D Abal, E.G. & Dennison, W.C. 1996. Seagrass depth range and water quality in Southern Moreton Bay, Queensland, Australia. Marine and Freshwater Research, 47: 763-771. Abal, E.G., Loneragan, N . , Bowen, P., Perry, C.J., Udy, J.W., & Dennison, W.C. 1994. Physiological and morphological responses of seagrass Zostera capricorni Aschers. to light intensity. Journal of Experimental Marine Biology and Ecology, 178: 113-129. Aioi , K . , Mukai, H . , Koike, I., Ohtsu, M . & Hattori, A . 1981. Growth and organic production of eelgrass (Zostera marina L.) in temperate waters of the Pacific coast of Japan. II. Growth analysis in winter. Aquatic Botany, 10: 175-182. Alcoverro, T., Zimmerman, R.C., Kohrs, D.G. & Alberte, R.S. 1999. Resource allocation and sucrose mobilization in light-limited eelgrass Zostera marina. Marine Ecology Progress Series, 187: 121-131. Aller, R.C. & Dodge, R.E. 1974. Animal-sediment relations in a tropical lagoon, Discovery Bay, Jamaica. Journal of Marine Research, 32: 209-232. Allison, G.W., Lubchenco, J. & Carr, M . H . 1998. Marine reserves are necessary but not sufficient for marine conservation. Ecological Applications, 8 (Suppl.): S79-S92. Arnon, D.I. 1949. Copper enzymes in isolated chloroplasts. Polyphenoloxidase in Beta vulgaris. Plant Physiology, 24: 1-15 Association of Official Analytical Chemists 1995. Official Methods of Analysis of A O A C International, 16 t h edition. Virginia, A O A C International, chap. 3 & 44. Backman, T.W.H. 1991. Genotypic and phenotypic variability of Zostera marina on the west coast of North America. Canadian Journal of Botany, 69: 1361-1371. Bayer, R.D. 1979. Intertidal zonation of Zostera marina in the Yaquina estuary, Oregon. Svesis, 12: 147-154. Berkenbusch, K & Rowden, A . A . 2000. Latitudinal variation in the reproductive biology of the burrowing ghost shrimp Callianassa filholi (Decapoda: Thalassinidea). Marine Biology, 136: 497-504. Bird, E . M . 1982. Population dynamics of the thalassinidean shrimps and their community effects through sediment modification. PhD. Dissertation, University of Maryland, College Park, Maryland, 140 pp. 125 Branch, G . M . & Pringle, A . 1987. The impact of the sand prawn Callianassa kraussi Stebbing on sediment turnover and on bacteria, meiofauna, and benthic microflora. Journal of Experimental Marine Biology and Ecology, 107: 219-235. Brenchley, G.A. 1981. Disturbance and community structure: an experimental study of bioturbation in marine soft bottom environments. Journal of Marine Research, 39: 767-790. Bulthuis, D.A. 1995. Distribution of seagrasses in a North Puget sound estuary: Padilla Bay, Washington, USA. Aquatic Botany. 50: 99-105. Burdick, D . M . & Short, F.T. 1999. The effects of boat docks on eelgrass beds in coastal waters of Massachusetts. Environmental Management, 23: 231-240. Burke, M . K . , Dennison, W.C. & Moore, K . A . 1996. Non-structural carbohydrate reserves of eelgrass Zostera marina. Marine Ecology Progress Series, 137: 195-201. Cabello-Pasini, A . , Lara-Turrent, C. & Zimmerman, R.C. 2002. Effect of storms on photosynthesis, carbohydrate content and survival of eelgrass populations from a coastal lagoon and the adjacent open ocean. Aquatic Botany, 74: 149-164. Canadian Heritage Parks Canada. 1994. Guiding Principles and Operational Policies. Minister of Supply and Services Canada, Ottawa. 125 pp. Canadian Heritage Parks Canada. 2001. Definition of ecological integrity found on the Parks Canada website: www.parkscanada. gc.ca; Search: ecological integrity. Caro, T .M. & O'Doherty, G. 1999. On the use of surrogate species in conservation biology. Conservation Biology, 13: 805-814. Chambers, P. A . & Kalff, J. 1985. Depth distribution and biomass of submersed aquatic macrophyte communities in relation to secchi depth. Canadian Journal of Fisheries and Aquatic Science, 42: 701-709. Darling, J.D., Keogh, K . E . & Steeves, T.E. 1998. Gray whale (Eschrichtius robustus) habitat utilization and prey species off Vancouver Island, B.C. Marine Mammal Science, 14: 692-720. Dawe, N .K . , Bradfield, G.E., Boyd, W.S., Trethewey, D.E.C. & Zolbrod, A . N . 2000. Marsh creation in a Northern Pacific estuary: is thirteen years of monitoring vegetation dynamics enough? Conservation Ecology, 4: 12. http://www.consecol.org/vol4/iss2/artl2 126 Dawes, C.J. & Guiry, M.D. 1992. Proximate constituents in the seagrasses Zostera marina and Z. noltii in Ireland: seasonal changes and the effect of blade removal. Marine Ecology, 13: 307-315. Dayton, P.K. 1998. Reversal of the burden of proof in fisheries management. Science, 279: 821-822. Dennison, W.C. 1987. Effects of light on seagrass photosynthesis, growth and depth distribution. Aquatic Botany, 27; 15-26. Dennison, W.C. 1990. Chlorophyll content. In: Phillips, R.C. & McRoy. C P . (Eds.) Seagrass Research Methods. UNESCO, France, pp. 83-85. Dennison, W.C. & Alberte, R.S. 1982. Photosynthetic responses of Zostera marina L. (eelgrass) to in situ manipulations of light intensity. Oecologia, 55: 137-144. Dennison, W.C. & Alberte, R.S. 1986. Photoadaptation and growth of Zostera marina L. (eelgrass) transplants along a depth gradient. Journal of Experimental Marine Biology and Ecology, 98: 265-282. Dennison, W . C , Orth, R.J., Moore, R.A. , Stevenson, J .C , Carter, V . , Kollar, S., Bergstrom, P.W., & Batiuk, R A . 1993. Assessing water quality with submersed aquatic vegetation. Bioscience, 43: 86-94. Downing, J.A. & Anderson, M.R. 1985. Estimating the standing biomass of aquatic macrophytes. Canadian Journal of Fisheries and Aquatic Sciences, 42: 1860-1869. Drew, E A . 1978. Carbohydrate and inositol metabolism in the seagrass, Cymodocea nodosa. New Phytologist, 81: 249-264. Drew, E. A . 1980. Soluble carbohydrate composition of seagrasses. In: Phillips, R.C. & McRoy, C P . (Eds.) Handbook of Seagrass Biology: An Ecosystem Perspective. Garland Press, New York, pp. 247-259. Drew, E.A. 1983. Sugars, cyclitols and seagrass phylogeny. Aquatic Botany, 15: 387-408. Duarte, C M . 1989. Temporal biomass variability and production/biomass relationships of seagrass communities. Marine Ecology Progress Series, 51: 269-276. Duarte, C M . 1991. Seagrass depth limits. Aquatic Botany, 40: 363-377. Duarte, C M . & Chiscano, C L . 1999. Seagrass biomass and production: a reassessment. Aquatic Botany, 65: 159-174. 127 Duarte, C M . , Martinez, R. & Barron, C. 2002. Biomass, production and rhizome growth near the northern limit of seagrass (Zostera marina) distribution. Aquatic Botany, 72: 183-189. Dumbauld, B.R., Armstrong, D.A., & Feldman, K . L . 1996. Life-history characteristics of two sympatric thalassinidean shrimps, Neotrypaea californiensis and Upogebia pugettensis, with implications for oyster culture. Journal of Crustacean Biology, 16: 689-708. Everett, R.A. , Ruiz, G . M . & Carlton, J.T. 1995. The effect of oyster mariculture on submerged aquatic vegetation: an experimental test in a Pacific Northwest estuary. Marine Ecology Progress Series, 125: 205-217. Feldman, K . L . , Armstrong, D.A., Dumbauld, B.R., DeWitt, T.H. & Doty, D.C. 2000. Oysters, crabs, and burrowing shrimp: review of an environmental conflict over aquatic resources and pesticide use in Washington State's (USA) coastal estuaries. Estuaries, 23: 141-176. Felger, R.S., Moser, E.W. & Moser, M . B . 1980. Seagrasses in Seri Indian culture. In: Phillips, R.C. & McRoy, C P . (Eds.), Handbook of Seagrass Biology: A n Ecosystem Perspective. Garland Press, New York, pp. 261-276. Fonseca, M.S. 1989. Sediment stabilization by Halophila decipiens in comparison to other sea grasses. Estuarine and Coastal Shelf Science, 29: 501-507. Fonseca, M.S. , Fisher, J.S., Zieman, J .C & Thayer, G.W. 1982. Influence of the seagrass, Zostera marina L. , on current flow. Estuarine, Coastal and Shelf Science, 15: 351-364. Fonseca, M.S., Meyer, D.L., & Hall, M.O. 1996. Development of planted seagrass beds in Tampa Bay, Florida, USA. II. Faunal components. Marine Ecology Progress Series, 132: 141-156. Frost, M.T. , Rowden, A . A . & Attrill, M.J . 1999. Effect of habitat fragmentation on the macroinvertebrate infaunal communities associated with the seagrass Zostera marina L . Aquatic Conservation: Marine and Freshwater Ecosystems, 9: 255-263. Gallegos, C L . & Kenworthy, W.J. 1996. Seagrass depth limits in the Indian River Lagoon (Florida, USA): Application of an optical water quality model. Estuarine, Coastal and Shelf Science, 42: 267-288. Gambi, M . C . 1988. Flowering in a Zostera marina bed off San Juan Island (Washington, U.S.A.) during winter. Aquatic Botany, 30: 267-272. 128 Ganter, B. 2000. Seagrass {Zostera spp.) as food for Brent geese (Branta bernicla): an overview. Helgoland Marine Research, 54: 63-70. Government of Canada. B i l l C-10. 2002. A n Act Respecting the Marine Conservation Areas of Canada. Griffis, R.B. & Chavez, F.L. 1988. Effects of sediment type on burrows of Callianassa californiensis Dana and C. gigas Dana. Journal of Experimental Marine Biology and Ecology, 117: 239-253. Griffis, R.B. & Suchanek, T.H. 1991. A model of burrow architecture and trophic modes in thalassinidean shrimp (Decapoda: Thalassinidea). Marine Ecology Progress Series, 79: 171-183. Halvorson, W.L. 1996. Changes in landscape values and expectations: what do we want and how do we measure it? In: Wright, R.G. (Ed.) National Parks and Protected Areas: Their Role in Environmental Protection. Blackwell Science, Massachusetts, pp. 15-30. Hansen, J.W., Pedersen, A .G.U. , Berntsen, J., Ronbog, I.S., Hansen, L.S. & Lomstein, B .A. 2000. Photosynthesis, respiration, and nitrogen uptake by different compartments of a Zostera marina community. Aquatic Botany, 66: 281-295. Harbo, R . M . 1999. Whelks to Whales: Coastal Marine Life of the Pacific Northwest. Harbour Publishing, British Columbia, 245 pp. Harrison, P.G. 1979. Reproductive strategies in intertidal populations of two co-occurring seagrasses (Zostera spp.) Canadian Journal of Botany, 57: 2635-2638. Harrison, P.G. 1982a. Spatial and temporal patterns in abundance of two intertidal seagrasses, Zostera americana den Hartog and Zostera marina L. Aquatic Botany, 12: 305-320. Harrison, P.G. 1982b. Comparitive growth of Zostera japonica Aschers. and Greabn. and Z. marina. L. under simulated intertidal and subtidal conditions. Aquatic Botany, 14: 373-379. Harrison, P.G. 1987. Natural expansion and experimental manipulation of seagrss (Zostera spp.) abundance and the response of infaunal invertebrates. Estuarine, Coastal and Shelf Science, 24: 799-812. Harrison, P.G. & Mann, K . H . 1975. Chemical changes during the seasonal cycle of growth and decay in eelgrass (Zostera marina) on the Atlantic coast of Canada. Journal of Fisheries Research Board of Canada, 32: 615-621. 129 Hillman, K. , Walker, D.I., Larkum, A.W.D. & McComb, A.J . 1989. Productivity and nutrient limitation. In: Larkum, A.W.D. , McComb, A.J . & Shepherd, S.A. (Eds.) Biology of Seagrasses. Elsevier, Amsterdam, pp.635-685. Hodge, J.E. & Hofreiter, B.T. 1962. Determination of reducing sugars and carbohydrates. In: Whistler, R.L. & Wolfram, M . L . (Eds). Methods in Carbohydrate Chemistry, Vol 1: Analysis and Preparations of Sugars. Academic Press, New York, pp. 380-394. Jacobs, R.P.W.M & Pierson, E.S. 1981. Phenology of reproductive shoots of eelgrass, Zostera marina, at Roscoff (France). Aquatic Botany, 10: 45-60. Johnson, B.L . 1999. The role of adaptive management as an operational approach for resource management agencies. Conservation Ecology, 3: 8. http://www.consecol.org/vol3/iss2/art8 Johnson, G.E. & Gonor, J.J. 1982. The tidal exchange of Callianassa californiensis (Crustacea, Decapoda) larvae between the ocean and the Salmon River Estuary, Oregon. Estuarine, Coastal and Shelf Science, 14: 501-516. Jones, S.E. & Jago, C.F. 1993. In situ assessment of modification of sediment properties by burrowing invertebrates. Marine Biology, 115: 133-142. Keddy, C.J. & Patriquin, D.G. 1978. An annual form of eelgrass in Nova Scotia. Aquatic Botany, 5: 163-170. Kentula, M.E . & Mclntire, C D . 1986. The autoecology and production dynamics of eelgrass (Zostera marina L.) in Netarts Bay, Oregon. Estuaries, 9: 188-199. Kikuchi, K. , Kawasaki, Y . & Sato, S. 2001. Effect of seasonal changes on the carbohydrate levels of eelgrass Zostera marina at Odawa Bay. Fisheries Science, 67: 755-757. Koch, E.W. & Beer, S. 1996. Tides, light and the distribution of Zostera marina in Long Island Sound, USA. Aquatic Botany, 53: 97-107. Kraemer, G.P. & Alberte, R.S. 1995. Impact of daily photosynthetic period on protein synthesis and carbohydrate stores in Zostera marina L. (eelgrass) roots: implications for survival in light-limited environments. Journal of Experimental Marine Biology and Ecology, 185: 191-202. Labadie, L . V . & Palmer, A.R. 1996. Pronounced heterochely in the ghost shrimp, Neotrypaea californiensis (Decapoda: Thalassinidea: Callianassidae): allometry, inferred function and development. Journal of Zoology, 240: 659-675. 130 Landres, P.B., Vemer, J. & Thomas, J.W. 1988. Ecological uses of vertebrate indicator species: a critique. Conservation Biology, 2: 316-328. Lee, K.S. & Dunton, K . H . 1997. Effects of in situ light reduction on the maintenance, growth and partitioning of carbon resources in Thalassia testudinum Banks ex Konig. Journal of Experimental Marine Biology and Ecology, 210: 53-73. Lee, K.S. & Dunton, K . H . 1999. Influence of sediment nitrogen-availability on carbon and nitrogen dynamics in the seagrass Thalassia testudinum. Marine Biology, 134: 217-226. Little, C. 2000. The Biology of Soft Shores and Estuaries. Oxford University Press, Oxford, 252 pp. MacFarlane, J .M., Quan, H.J., Uyeda, K . K . , & Wong,'K.D. 1996. Official Guide to Pacific Rim National Park Reserve. Blackbird Naturgraphics Inc., Calgary, 127 pp. MacGinitie, G.E. 1934. The natural history of Callianassa californiensis Dana. American Midland Naturalist, 15: 166-177. MacGinitie, G.E. & MacGinitie, N . 1949. Natural History of Marine Animals. McGraw-Hi l l , New York, 523 pp. Manning, R.B. & Felder, D.L. 1991. Revision of the American Callianassidae (Crustacea: Decapoda: Thalassinidea). Proceedings of the Biological Society of Washington, 104: 764-792. Marsh, J.A.Jr., Dennison, W.C. & Alberte, R.S. 1986. Effects of temeprature on photosynthesis and respiration in eelgrass (Zostera marina L.). Journal of Experimental Marine Biology and Ecology, 101: 257-267. Mather, R.G., Montgomery, W.I. & Portig, A . A . 1998. Exploitation of intertidal Zostera species by brent geese (Branta bernicla hrota): why dig for your dinner? Biology and Enviornment: Proceedings of the Royal Irish Academy, 98B: 147-152. McCrow, L.T. 1972. The ghost shrimp, Callianassa californiensis, in YaquinaBay, Oregon. M.Sc. Thesis, Oregon State University, 56 pp. McMillan, C. 1976. Experimental studies on flowering and reproduction in seagrasses. Aquatic Botany, 2: 87-92. McRoy, C P . & McMillan, C. 1977. Production ecology and physiology of seagrasses. In: McRoy, C P . & Helfferich, C A . (Eds.). Seagrass Ecosystems: A Scientific Perspective, Marcel Dekker, NewYork, p. 53-87. 131 Meling-Lopez, A .E . & Ibarra-Obando, S.E. 1999. Annual life cycles of two Zostera marina L. populations in the Gulf of California: contrasts in seasonality and reproductive effort. Aquatic Botany, 65: 59-69. Milne, D.H. 1995. Marine Life and the Sea. Wadsworth Publishing Company, USA, 459 pp. Moody, R. 1978. Habitat, population and leaf characteristics of Zostera marina L. on Roberts Bank, British Columbia. M.Sc. Thesis, University of British Columbia, Vancouver, 104 pp. Moore, K . A . , R.L. Wetzel and R J . Orth. 1997. Seasonal pulses of turbidity and their relations to eelgrass (Zostera marina L.) survival in an estuary. J.Exp.Mar.Biol & Ecology. 215: 115-134. Murphey, P.L. & Fonseca, M.S. 1995. Role of high and low energy seagrass beds as nursery areas for Penaeus duorarum in North Carolina. Marine Ecology Progress Series, 121: 91-98. Murphy, R.C. & Kremer, J.N. 1992. Benthic community metabolism and the role of deposit-feeding callianassid shrimp. Journal of Marine Research, 50: 321-340. Murray, L , Dennison, W.C. & Kemp, W . M . 1992. Nitrogen versus phosphorus limitation for growth of an estuarine population of eelgrass (Zostera marina L.). Aquatic Botany, 44: 83-100. Nelson, T.A. 1997. Interannual variance in a subtidal eelgrass community. Aquatic Botany, 56: 245-252. Nelson, T.A. & Waaland, J.R. 1997. Seasonality of eelgrass, epiphyte, and grazer biomass and productivity in subtidal eelgrass meadows subjected to moderate tidal amplitude. Aquatic Botany, 56: 51-74. Noss, R.F. 1990. Indicators for monitoring biodiversity: a hierarchical approach. Conservation Biology, 4: 355-364. Olesen, B. 1999. Reproduction in Danish eelgrass (Zostera marina L.) stands: size-dependence and biomass partitioning. Aquatic Botany, 65: 209-219. Olesen, B. & Sand-Jensen, K . 1994a. Demography of shallow eelgrass (Zostera marina) populations - shoot dynamics and biomass development. Journal of Ecology, 82: 379-390. Olesen, B . & Sand-Jensen, K . 1994b. Biomass-density patterns in the temperate seagrass Zostera marina. Marine Ecology Progress Series, 109: 283-291. 132 Orth, R.J, Luckenbach, M . , & Moore, K , A . 1994. Seed dispersal in a marine macrophyte: implications for colonization and restoration. Ecology, 75: 1927-1939. Palm, R. 2001. Unpublished data. Strawberry Isle Research Society, Tofino, British Columbia, www.island.net/~sisle/rshindex.htm. Peterson, C.H. 1977. Competitive organization of the soft-bottom macrobenthic communities of Southern California lagoons. Marine Biology, 43: 343-359. Phillips, R . C , Grant, W.S. & McRoy, C P . 1983a. Reproductive strategies of eelgrass (Zostera marina L.). Aquatic Botany, 16: 1-20. Phillips, R . C , McMillan, C. & Bridges, K .W. 1983b. Phenology of eelgrass, Zostera marina L., along latitudinal gradients in North America. Aquatic Botany, 15: 145-156. Poiner, I.R., Walker, D.I. & Coles, R.G. 1989. Regional studies - seagarsses of tropical Australia. In: Larkum, A.W.D. , McComb, A.J . & Shepherd, S A . (Eds.) Biology of Seagrasses. Elsevier, Amsterdam, pp 279-303. Posey, M . H . 1986a. Predation on a burrowing shrimp: distribution and community consequences. Journal of Experimental Marine Biology and Ecology, 103: 143-161. Posey, M . H . 1986b. Changes in a benthic community associated with dense beds of a burrowing deposit feeder, Callianassa californiensis. Marine Ecology Progress Series, 31: 15-22. Posey, M . H . 1987. Effects of lowered salinity on activity of the ghost shrimp Callianassa californiensis. Northwest Science, 61: 93-96. Posey, M . H . 1988. Community changes associated with the spread of an introduced seagrass, Zostera japonica. Ecology, 69: 974-983. Robertson, A.I. & Mann, K . H . 1984. Disturbance by ice and life-history adaptations of the seagrass Zostera marina. Marine Biology, 80: 131-142. Ruppert, E.E. & Barnes, R.D. 1994. Invertebrate Zoology, 6 t h Edition. Saunders College Publishing, Texas, 1056 pp. Sfriso, A . & Ghetti, P.F. 1998. Seasonal variation in biomass, morphometric parameters and production of seagrasses in the lagoon of Venice. Aquatic Botany , 61: 207-223. 133 Short, F.T. & Neckles, H.A. 1999. The effects of global climate change on seagrasses. Aquatic Botany, 63: 169-196. Short, F.T. & Short, C A . 1984. The seagrass filter: purification of estuarine and coastal water. In: Kennedy, V.S. (Ed.) The Estuary as a Filter. Academic Press, Orlando, pp. 395-413. Short, F.T. & Wyllie-Echeverria, S. 1996. Natural and human-induced disturbance of seagrasses. Environmental Conservation, 23: 17-27. Silverhorn, G .M. , Orth, R J . & More, K . A . 1983. Anthesis and seed production in Zostera marina L. (eelgrass) from the Chesapeake Bay. Aquatic Botany, 15: 133-144. Simenstad, C A . & Fresh, K . L . 1995. Influence of intertidal aquaculture on benthic communities in Pacific Northwest estuaries: scales of disturbance. Estuaries, 18: 43-70. Smith, R.D. & Alberte, R.S. 1989. Effects of anaerobiosis on in-vivo protein synthesis in the roots of a marine angiosperm Zostera marina L . Plant Physiology, 89 (4 Suppl.): 126. Sokal, R.R. & Rohlf, F.J. 1981. Biometry, 2 n d edition. W.H. Freeman and Company, San Fransisco, 859 pp. Soule, D.F. 1988. Marine organisms as indicators: reality or wishful thinking? In: Soule, D.F. & Kleppel, G.S. (Eds.) Marine Organisms as Indicators. Springer-Verlag, New York, pp. 1-11. Stevens, B. A . 1929. Ecological observation on Callianassidae of Puget Sound. Ecology, 10: 399-405. Suchanek, T.H. 1983. Control of seagrass communities and sediment distribution by Callianassa (Crustacea, Thalassinidea) bioturbation. Journal of Marine Research, 41: 281-298. Sumaila, U.R., Guenette, S., Alder, J., & Chuenpagdee, R. 2000. Addressing ecosystem effects of fishing using marine protected areas. ICES Journal of Marine Science, 57: 752-760. Swinbanks, D.D. & Luternauer, J.L. 1987. Burrow distribution of thalassinidean shrimp on a Fraser delta tidal flat, British Columbia. Journal of Paleontology, 61: 315-332. Swinbanks, D.D. & Murray, J.W. 1981. Biosedimentological zonation of Boundary bay tidal flats Fraser River Delta, British Columbia. Sedimentology, 28: 201-237. 134 Taiz, L. & Zeiger, E. 1998. Plant Physiology, 2" Edition. Sinauer Associates, Inc., Massachusetts, 792 pp. Tamaki, A . 1988. Effects of the bioturbating activity of the ghost shrimp Callianassa japonica Ortmann on migration of a mobile polychaete. Journal of Experimental Marine Biology and Ecology, 120: 81-95. Thayer, C. 1979. Biological bulldozers and the evolution of marine benthic communities. Science, 203: 458-461. Thayer, G.W., Wolfe, D.A. , & Williams, R.B. 1975. The impact of man on seagrass systems. American Scientist, 63: 288-296. Thompson, L.C. & Pritchard, A .W. 1969. Osmoregulatory capacities of Callianassa and Upogebia (Crustacea: Thalassinidea). Biological Bulletin, 136: 114-129. Thompson, R.K. & Pritchard, A .W. 1969. Respiratory adaptations of two burrowing curstaceans, Callianassa californiensis and Upogebia pugettensis (Decapoda, Thalassinidea). Biological Bulletin, 136: 274-287. Thorne-Miller, B . & Harlin, M . M . 1984. The production of Zostera marina L. and other submerged macrophytes in a coastal lagoon in Rhode Island, U.S.A. Botanica Marina, 27: 539-546. Touchette, B.W. & Burkholder, J .M. 2000. Overview of the physiological ecology of carbon metabolism in seagrass. Journal of Experimental Marine Biology and Ecology, 250: 169-205. van Katwijk, M . M . , Schmitz, G.H.W., Gasseling, A.P. & van Avesaath, P.H. 1999. Effects of salinity and nutrient load and their interaction on Zostera marina. Marine Ecology Progress Series, 190: 155-165. Van Lent, F. & Verschuure, J .M. 1994. Intraspecific variablity of Zostera marina L . (eelgrass) in the estuaries and lagoons of the southwestern Netherlands: I. Population dynamics. Aquatic Botany, 48: 31-58. Vermaat, J.E. & Verhagen, F.C.A. 1996. Seasonal variation in the intertidal seagrass Zostera noltii Hornem.: coupling demographic and physiological patterns. Aquatic Botany, 52: 259-281. Washington Department of Fisheries. 1970. Ghost shrimp control experiments with Sevin, 1960-1968. Technical Report 1. Washington Department of Fisheries, 62 pp. 135 Ward, D.H., Markon, C.J. & Douglas, D.C. 1997. Distribution and stability of eelgrass beds at Izembek Lagoon, Alaska. Aquatic Botany, 58: 229-240. Webster, P.J., Rowden, A . A . , & Attrill, M.J. 1998. Effect of shoot density on the infaunal macro-invertebrate community within a Zostera marina seagrass bed. Estuarine, Coastal and Shelf Science, 47: 351-357. Weitkamp, L .A. , Wissmar, R.C., Simenstad, C.A., Fresh, K . L , & Odell, J.G. 1992. Gray whale foraging on ghost shrimp (Callianassa californiensis) in littoral sand flats of Puget Sound, U.S.A. Canadian Journal of Zoology. 70: 2275-2280. Wenner, A . M . 1972. Sex ratio as a function of size in marine Crustacea. American Naturalist, 106: 321-351. Wenner, A . M . 1988. Crustaceans and other invertebrates as indicators of beach pollution. In: Soule, D.F. & Kleppel, G.S. (Eds.) Marine Organisms as Indicators. Springer-Verlag, New York, pp. 199-229. Yakimishyn, J. 2002. Unpublished data. Department of Geograrphy, University of Victoria, British Columbia. Zharova, N . , Sfriso, A . , Voinov, A . & Pavoni, B. 2001. A simulation model for the annual fluctuation of Zostera marina biomass in the Venice lagoon. Aquatic Botany, 70: 135-150. Zimmerman, R.C., Reguzzoni, J.L. & Alberte, R.S. 1995a. Eelgrass (Zostera marina L.) transplants in San Francisco Bay: role of light availability on metabolism, growth and survival. Aquatic Botany, 51: 67-86. Zimmerman, R.C., Kohrs, D.G., Steller, D.L. & Alberte, R.S. 1995b. Carbon partitioning in eelgrass: regulation by photosynthesis and the response to daily light-dark cycles. Plant Physiology, 108: 1665-1671. Zimmerman, R.C., Kohrs, D.G., Steller, D.L. & Alberte, R.S. 1997. Impacts of C 0 2 enrichment on productivity and light requirements of eelgrass. Plant Physiology, 115:599-607. 136 Appendix 1: Seagrass population mean values (refer to Chapter 2) Appendices 1-3 contain the actual mean values used to construct the graphs in this thesis. A l l error values denote one standard error of the mean (S.E.). Table 1.1 Seagrass population parameters at four tidal strata, sampled eight times between July 2001 and August 2002. Strata M o n t h 2001 - 2002 Density # shoots/m 2 Reproduct ive % o f shoots L e n g t h c m W i d t h c m Biomass g dw / m 2 E d g e July 84 ± 13 8.3 ± 2 . 0 26.0 ± 1.3 0.40 ± 0 . 0 1 7.5 ± 1.6 Sept 66 ± 9 0.89 ± 0 . 4 3 32.6 ± 3 . 8 0.42 ± 0.03 4.6 ± 1.0 Dec 52 ± 7 0 32.3 ± 3 . 5 0.41 ± 0.03 1.9 ± 0 . 4 Feb 63 ± 11 0 20.5 ± 1.8 0.35 ± 0.02 2.5 ± 0 . 7 M a y 46 ± 6 4.9 ± 2 . 3 20.1 ± 2 . 4 0.39 ± 0 . 0 3 2.2 ± 0 . 9 June 41 ± 2 1.7 ± 0 . 8 23.3 ± 2 . 5 0.42 ± 0.02 1.6 ± 0 . 7 A u g 18 ± 5 8.6 ± 4 . 9 32.0 ± 3 . 0 0.49 ± 0.03 1.6 ± 0 . 5 Intertidal July 525 ± 3 1 5.4 ± 0 . 6 39.6 ± 2 . 8 0.47 ± 0.03 95.0 ± 18.6 Sept 413 ± 51 0 49.7 ± 6 . 4 0.44 ± 0.03 41.9 ± 10.7 Dec 322 ± 15 0 21.6 ± 1.3 0.33 ± 0 . 0 1 13.3 ± 1.6 Feb 346 ± 18 0 23.0 ± 0 . 9 0.34 ± 0 . 0 1 14.8 ± 1.7 M a y 350 ± 4 9 0.09 ± 0 . 0 6 29.1 ± 2 . 8 0.42 ± 0.02 21.8 ± 3 . 2 June 442 ± 2 1 1.7 ± 0 . 4 39.2 ± 2 . 6 0.48 ± 0 . 0 1 39.9 ± 2 . 7 A u g 414 ± 2 7 2.2 ± 0 . 4 66.7 ± 9 . 0 0.49 ± 0.03 78.1 ± 16.4 Slope July 161 ± 15 4.9 ± 1.1 76.9 ± 13.6 0.58 ± 0 . 0 5 80.1 ± 2 5 . 0 Sept 145 ± 17 0.87 ± 0.40 75.6 ± 8 . 3 0.54 ± 0 . 0 3 43.9 ± 8 . 1 Dec 75 ± 10 0 26.9 ± 3 . 7 0.42 ± 0.02 4.1 ± 1.0 Feb 44 ± 6 0 13.5 ± 1.7 0.32 ± 0 . 0 3 2.1 ± 0 . 5 M a y 38 ± 5 0 13.3 ± 0 . 5 0.22 ± 0 . 0 1 1.2 ± 0 . 1 June 187 ± 2 0 0.24 ± 0 . 1 1 28.2 ± 2 . 9 0.39 ± 0.04 11.7 ± 3.8 A u g 125 ± 9 1 . 8 ± 0 . 8 71.3 ± 5.1 0.61 ± 0 . 0 4 43.3 ± 8 . 5 C h a n n e l July 13 ± 3 0.69 ± 0 . 6 9 20.7 ± 2 . 5 0.26 ± 0 . 0 2 0.3 ± 0 . 1 Sept 23 ± 6 0 50.7 ± 8 . 4 0.50 ± 0 . 0 5 2.2 ± 1.0 Dec 15 ± 5 0 25.6 ± 3 . 4 0.44 ± 0.02 0.8 ± 0 . 4 Feb 16 ± 5 0 15.0 ± 0 . 9 0.34 ± 0 . 0 1 0.3 ± 0 . 1 M a y 37 ± 5 0 18.3 ± 2 . 4 0.29 ± 0.04 1.6 ± 0 . 7 June 47 ± 9 0 20.1 ± 1.6 0.28 ± 0.02 1.3 ± 0 . 6 A u g 34 ± 8 0.68 ± 0 . 4 7 38.7 ± 5 . 1 0.42 ± 0.04 7.7 ± 4 . 9 137 Table 1.2 Seagrass population parameters at five intertidal sites, sampled four times between April and August 2002. Site M o n t h 2002 Densi ty # shoots/m 2 Reproduc t i ve % of shoots L e n g t h c m W i d t h c m Biomass g dw / m 2 G B M a i n A p r i l 378 ± 3 0 0.05 ± 0.05 28.1 ± 3 . 0 0.35 ± 0.02 23.4 ± 4 . 4 M a y 362 ± 22 0.09 ± 0 . 0 6 29.1 ± 2 . 8 0.42 ± 0.02 21.8 ± 3.2 June 442 ± 2 1 1.7 ± 0 . 4 39.2 ± 2 . 6 0.48 ± 0.01 39.9 ± 2 . 7 August 4 1 4 ± 2 7 2.2 ± 0.4 66.7 ± 9 . 0 0.49 ± 0.03 78.1 ± 16.4 G B 1 A p r i l 1 8 0 ± 12 0.15 ± 0.15 48.9 ± 5 . 3 0.58 ± 0.04 28.4 ± 2 . 5 M a y 167 ± 1 1 0 49.0 ± 7 . 0 0.65 ± 0.06 40.6 ± 9 . 1 June 247 ± 16 2.0 ± 0 . 3 41.7 ± 3 . 3 0.54 ± 0.03 30.3 ± 4.0 August 223 ± 12 2.1 ± 0 . 5 98.7 ± 7 . 5 0.67 ± 0.03 87.9 ± 15.1 G B 2 A p r i l 181 ± 7 0 52.9 ± 5 . 2 0.52 ± 0.03 28.9 ± 4 . 5 M a y 181 ± 10 0 . 1 4 ± 0 . 1 4 48.2 ± 3 . 6 0.63 ± 0.03 33.6 ± 3 . 6 June 254 ± 15 2.6 ± 0 . 5 58.7 ± 4 . 9 0.62 ± 0.02 50.0 ± 4 . 3 August 244 ± 15 3.8 ± 0 . 7 80.5 ± 7 . 1 0.55 ± 0.03 64.9 ± 11.5 L I 1 A p r i l 335 ± 2 2 0.15 ± 0.11 38.9 ± 3 . 0 0.54 ± 0 . 0 3 48.7 ± 8 . 1 M a y 523 ± 29 4.1 ± 0 . 7 40.1 ± 4 . 3 0.48 ± 0 . 0 1 66.5 ± 10.2 June 493 ± 25 4.8 ± 0 . 7 55.5 ± 4 . 5 0.54 ± 0 . 0 2 83.2 ± 9 . 6 August 482 ± 19 3.8 ± 0 . 4 57.9 ± 3 . 5 0.48 ± 0.02 67.7 ± 9.4 L I 2 A p r i l 297 ± 16 0.24 ± 0 . 1 5 32.3 ± 2 . 8 0.51 ± 0 . 0 4 36.0 ± 5 . 3 M a y 293 ± 25 1.4 ± 0 . 4 44.7 ± 2 . 6 0.63 ± 0.02 66.3 ± 10.9 June 376 ± 24 1.7 ± 0 . 4 44.1 ± 4 . 5 0.61 ± 0 . 0 3 63.7 ± 11.0 August 2 8 6 ± 16 2.7 ± 0 . 6 109.0 ± 9 . 0 0.68 ± 0.02 179.6 ± 3 2 . 5 Appendix 2: Seagrass carbohydrate and chlorophyll mean values (refer to Chapter 3) Table 2.1 Overall estimates of mean carbohydrate concentrations in shoot and rhizomes of Zostera marina sampled in Clayoquot Sound in April and June 2002 [mean ± S.E. (n)] Date Shoots Rh izomes Sugar (mg/g) S ta rch (mg/g) Sugar (mg/g) S ta rch (mg/g) A p r i l 11.7 ± 0 . 5 (60) 16.0 ± 0 . 7 (37) 37.2 ± 3.9 (60) 21.6 ± 0 . 6 (37) June 32.9 ± 2 . 8 (45) 16.4 ± 1.0 (28) 140.2 ± 7.4 (45) 20.8 ± 1.0 (30) 138 Table 2.2 Mean estimates of carbohydrate concentrations in shoots and rhizomes of 5 populations of Zostera marina in Clayoquot Sound, in April and June 2002 [mean ± S.E. (n)] Date Site Shoots Rhizomes Sugar (mg/g) Starch (mg/g) Sugar (mg/g) Starch (mg/g) April GB Main 9.8 ± 0 . 8 (12) 14.7 ± 1.3 (7) 33.1 ±9 .5 (12) 21.9 ± 0 . 9 (7) GB 1 12.2 ± 1.0(12) 14.5 ± 1.0(8) 44.1 ±8 .6 (12) 24.0 ± 1.3 (8) GB 2 9.1 ±0 .7 (12 ) 14.2 ± 1.0 (7) 11.7 ±2 .8 (12) 19.7 ±0 .9 (8) LI 1 12.8 ± 0 . 9 (12) 17.3 ± 1.2 (7) 42.4 ± 6 . 0 (12) 21.1 ± 1.8 (7) LI 2 14.8 ±0 .8 (12) 19.2 ±2 .0 (8) 54.6 ±6 .5 (12) 21.5 ± 1.5 (7) June GB Main 11.6 ± 1.0 (9) 13.5 ±0 .9 (5) 103.4 ±20.9 (9) 18.4 ± 2 . 9 (6) GB 1 20± 1.6(9) 15.9 ±2 .0 (5) 136.4 ± 13.0 (9) 20.9 ± 2.4 (6) GB 2 52.9 ±4.1 (9) 15.5 ±2 .7 (6) 149.4 ± 18.7 (9) 20.8 ± 2.5 (6) LI 1 51.8 ±4 .3 (9) 15.0 ±2 .3 (6) 158.2 ± 14.1 (9) 23.5 ± 1.8(6) LI 2 28.2 ±3.1 (9) 21.5 ± 1.8 (6) 153.8 ± 10.3 (9) 20.1 ± 1.9 (6) Table 2.3 Overall estimates of mean total chlorophyll (a + b) concentrations (mg/g dry wt) and chlorophyll cr.b ratios in Zostera marina shoots sampled in Clayoquot Sound in Summer 2002 [mean ± S.E. (n)]. Parameter April May June July August Chi (a+b) mg/g 15.8 ±0 .4 (113) 10.9 ±0 .3 (150) 9.93 ± 0 . 2 (149) 12.8 ±0 .3 (150) 13.5 ±0 .4 (100) Chi a:b ratio 1.85 ±0 .04 1.66 ±0 .04 1.53 ±0.03 1.52 ±0 .02 1.29 ±0.02 139 Table 2.4 Mean estimates of shoot chlorophyll concentrations and ratios in 5 intertidal Zostera marina populations in Clayoquot Sound (3 sites in Grice Bay, 2 sites in Lemmens Inlet), in Summer 2002 [mean ± S.E. (n)]. Parameter Site A p r i l M a y June J u l y Augus t C h l (a+b) mg/g G B M a i n 1 1 . 0 ± 0 . 8 ( 1 5 ) 8.3 ± 0.4 (30) 6.9 ± 0.3 (30) 13.9 ± 0 . 7 (30) 12.9 ± 0 . 8 (20) G B 1 11.5 ± 0 . 7 (10) 12.0 ± 0 . 5 (30) 10.4 ± 0.5 (30) 10.0 ± 0 . 3 (30) 12.9 ± 0 . 8 (20) G B 2 16.1 ± 0 . 8 ( 2 9 ) 7.3 ± 0.5 (30) 11.7 ± 0 . 3 (30) 14.4 ± 0.4 (30) 14.9 ± 1.2 (20) L I 1 14.6 ± 0 . 5 (30) 11.1 ± 0 . 5 (30) 10.8 ± 0 . 6 (29) 11.7 ± 0 . 5 (30) 12.6 ± 0 . 6 (20) L I 2 20.6 ± 0.7 (29) 15.8 ± 0 . 7 (30) 9.8 ± 0.4 (30) 14.0 ± 0 . 4 (30) 14.3 ± 1.0 (20) C h l a:b ratio G B M a i n 1.69 ± 0 . 0 9 1.52 ± 0 . 0 4 1.54 ± 0 . 0 6 1.57 ± 0 . 0 4 1.26 ± 0 . 0 5 G B 1 1.68 ± 0 . 1 0 1.72 ± 0 . 0 3 1.36 ± 0 . 0 3 1.33 ± 0 . 0 3 1.09 ± 0 . 0 4 G B 2 2.01 ± 0 . 0 7 1.74 ± 0 . 1 6 1.44 0.03 1.51 ± 0 . 0 3 1.28 ± 0 . 0 5 L I 1 2.05 ± 0.09 1.75 ± 0 . 0 3 1.57 ± 0 . 0 6 1.64 ± 0.04 1.44 ± 0 . 0 4 L I 2 1.63 ± 0 . 0 2 1.57 ± 0 . 1 0 1.76 ± 0 . 0 8 1.56 ± 0 . 0 2 0.41 ± 0.05 Appendix 3: Ghost shrimp mean values (refer to Chapter 4) Table 3.1 Mean overall ghost shrimp density each sampling period between July 2001 and August 2002 (mean ± S.E., n = 168). Date 2001 - 2002 Juveniles per m 3 Males per m 3 Females per m 3 Adul t s per m 3 Individuals per m 3 M a y 376 ± 3 2 283 ± 23 336 ± 2 9 656 ± 47 1032 ± 6 7 June 762 ± 57 269 ± 24 379 ± 2 8 697 ± 45 1459 ± 7 0 Sept 687 ± 45 313 ± 2 2 500 ± 2 8 857 ± 4 5 1543 ± 7 1 N o v 814 ± 4 6 307 ± 26 433 ± 3 1 783 ± 5 2 1598 ± 7 6 Feb 744 ± 46 266 ± 2 1 321 ± 2 4 608 ± 39 1353 ± 7 2 A p r i l 555 ± 4 0 214 ± 2 0 292 ± 26 547 ± 43 1102 ± 7 1 June 812 ± - 5 8 274 ± 22 435 ± 29 733 ± 43 1546 ± 7 5 A u g 777 ± 5 6 339 ± 2 5 454 ± 30 831 ± 4 5 1609 ± 7 9 140 Table 3.2 Ghost shrimp population parameters at six intertidal zones at the two sites in Grice Bay, data from all sampling dates is combined [mean ± S.E. (n)]. Site Z o n e Density # ind /m 3 Biomass g dwt /m 3 L e n g t h m m Reproduct ion % carrying eggs Sex ratio % female Site 1 Zone 1 723 ± 58 34 ± 4 24.4 ± 0.4 (230) 30 ± 8 (30) 57 ± 4 Zone 2 1434 ± 5 1 60 ± 3 22.8 ± 0.3 (640) 24 ± 5 (52) 61 ± 3 Zone 3 1965 ± 59 116 ± 4 24.9 ± 0.3 (1009) 21 ± 4 ( 5 5 ) 62 ± 2 Zone 4 2210 ± 6 0 166 ± 6 25.8 ± 0 . 3 (1157) 32 ± 5 (56) 54 ± 2 Zone 5 1 9 2 4 ± 61 183 ± 12 26.8 ± 0 . 4 (1098) 44 ± 5 (47) 54 ± 2 Zone 6 1 0 2 2 ± 89 56 ± 9 22.5 ± 0.5 (497) 14 ± 9 (12) 54 ± 4 Site 2 Zone 1 1048 ± 67 38 ± 3 22.0 ± 0 . 4 (340) 15 ± 6 (35) 55 ± 4 Zone 2 2180 ± 9 7 78 ± 4 20.6 ± 0 . 3 (710) 22 ± 6 (45) 61 ± 3 Zone 3 2082 ± 60 124 ± 5 23.8 ± 0 . 3 (1143) 22 ± 5 (55) 62 ± 2 Zone 4 1632 ± 6 5 109 ± 9 24.2 ± 0.4 (902) 34 ± 5 (33) 57 ± 2 Zone 5 429 ± 55 2.5 ± 0 . 4 13.8 ± 0 . 2 (197) 0 (0 ) 0 Zone 6 215 ± 4 2 0.9 ± 0 . 2 12.8 ± 0 . 2 (90) 0 ( 0 ) 0 n = 112 n = 112 n = # o f whole shrimp n = # o f females Table 3.3 Ghost shrimp population parameters each sampling date at the two sites in Grice Bay, data from all zones is combined (mean ± S.E. (n), or n at bottom of column). Site Date 2001 -2002 Density # ind /m 3 Biomass g dwt /m 3 L e n g t h m m Reproduct ion % carrying eggs Sex ratio % female Site 1 M a y 1236 ± 9 8 96 ± 11 25.7 ± 0 . 6 (457) 65 ± 5 (56) 51 ± 3 July 1668 ± 7 8 94 ± 9 23.4 ± 0.4 (595) 17 ± 4 (68) 52 ± 3 Sept 1584 ± 81 110 ± 9 25.9 ± 0 . 4 (617) 0.8 ± 0.7 (76) 64 ± 3 N o v 1 7 4 7 ± 102 115 ± 9 25.1 ± 0 . 4 (614) 0(69) 59 ± 3 Feb 1564 ± 9 4 106 ± 14 24.0 ± 0 . 5 (633) 0(70) 54 ± 3 A p r 1249 ± 107 70 ± 8 23.8 ± 0 . 5 (502) 9 ± 2 (52) 59 ± 3 Jun 1620 ± 8 0 102 ± 8 25.1 ± 0 . 4 (633) 27 ± 4 (76) 62 ± 3 A u g 1701 ± 9 7 127 ± 11 26.9 ± 0.4 (580) 0.7 ± 0.7 (73) 58 ± 3 Site 2 M a y 828 ± 85 52 ± 8 24.8 ± 0.7 (275) 65 ± 6 (41) 54 ± 4 July 1 2 5 1 ± 111 50 ± 6 21.2 ± 0 . 5 (376) 6 ± 3 (44) 69 ± 4 Sept 1 5 0 2 ± 117 79 ± 9 23.1 ± 0.5 (530) 0(54) 61 ± 3 N o v 1 4 4 8 ± 112 59 ± 8 20.5 ± 0.4 (496) 0(54) 62 ± 4 Feb 1 1 4 1 ± 106 60 ± 7 22.7 ± 0.5 (423) 0(47) 55 ± 4 A p r 956 ± 9 2 48 ± 6 23.0 ± 0.6 (335) 11 ± 4 (37) 52 ± 5 Jun 1 4 7 1 ± 1 2 7 56 ± 6 20.8 ± 0.5 (448) 11 ± 3 (46) 60 ± 5 A u g 1517 ± 126 67 ± 7 22.4 ± 0.4 (499) 0(55) 57 ± 4 n = 84 n = 84 n = # of shrimp n = # of females 141 Appendix 4: Seagrass population statistics Appendices 4 - 6 contain the test statistics and p values used to determine significance as referred to in the text of this thesis. Bold values indicate significance. 4.1 Seagrass shoot density: (refer to section 2.3.1) Does shoot density change significantly over time at GB Main? Mann-Whitney pairwise analysis 7 dates; 21 pairs; a = 0.002; n = 24 (except Intertidal in September where n = 8) Date comparison E d g e Intertidal Slope C h a n n e l p value x1 p value x1 p value x1 p value x1 July - sept 0.522 0.410 0.151 2.066 0.322 0.980 0.466 0.532 ju ly - dec 0.096 2.765 <0.0005 24.095 <0.0005 16.188 0.197 1.667 July - feb 0.110 2.561 <0.0005 14.720 <0.0005 24.755 0.466 0.532 j u l y - may 0.025 5.018 0.018 5.626 <0.0005 26.602 <0.0005 12.634 j u l y - J u n e <0.0005 13.287 0.085 2.966 0.236 1.407 0.002 9.626 j u l y - aug <0.0005 24.415 0.010 6.594 0.031 4.648 0.026 4.942 sept - dec 0.353 0.863 0.048 3.924 0.002 9.573 0.149 2.085 sept - feb 0.392 0.734 0.286 1.138 <0.0005 18.430 0.286 1.139 sept - may 0.109 2.562 0.500 0.455 <0.0005 19.220 0.012 6.351 sept -june 0.004 8.177 0.777 0.080 0.103 2.656 0.030 4.708 sept - aug <0.0005 18.167 0.913 0.012 0.293 1.107 0.212 1.554 dec - feb 0.959 0.003 0.056 3.641 0.032 4.618 0.524 0.406 dec - may 0.536 0.384 0.765 0.089 0.010 6.658 0.001 11.819 dec - june 0.026 4.973 <0.0005 16.424 <0.0005 15.690 <0.0005 13.723 dec - aug <0.0005 13.280 0.025 5.008 0.001 10.293 0.002 9.569 feb - may 0.384 0.384 0.845 0.038 0.577 0.312 0.001 11.626 feb - june 0.015 5.941 0.010 6.647 <0.0005 22.636 0.001 10.237 feb - aug <0.0005 15.926 0.279 1.173 <0.0005 26.624 0.010 6.674 may - june 0.039 4.264 0.228 1.455 <0.0005 24.008 0.628 0.235 may - aug <0.0005 13.557 0.288 1.128 <0.0005 31.609 0.193 1.694 june - aug 0.110 2.553 0.240 1.383 0.013 6.229 0.291 1.113 142 Does shoot density change significantly over the summer at the five intertidal sites? Mann-Whitney pairwise analysis 4 dates; 6 pairs; a = 0.008; n = 24 Date comparison G B M a i n G B 1 G B 2 L I 1 L I 2 p value I1 p value x2 P value x1 p value I2 p value I1 apr - may 0.975 0.001 0.765 0.090 0.959 0.003 <0.0005 17.786 0.804 0.061 apr - june 0.085 2.966 0.001 10.227 <0.0005 12.890 <0.0005 14.793 0.008 7.027 apr - aug 0.386 0.750 0.010 6.707 0.002 9.706 <0.0005 15.685 0.643 0.215 may - June 0.027 4.918 <0.0005 12.889 0.001 11.307 0.279 1.173 0.016 5.772 may - aug 0.261 1.264 0.005 8.049 0.003 8.952 0.095 2.792 0.837 0.043 june - aug 0.240 1.383 0.322 0.981 0.635 0.225 0.710 0.138 0.003 9.130 Does seagrass density differ significantly between 4 tidal strata? Mann-Whitney pairwise analysis of variance 4 strata; 6 pairs; a = 0.008; n = 24 (except Intertidal in September where n = 8) strata comparison J u l y 2001 Sept 2001 Dec 2001 F e b 2002 p value x2 p value I1 p value x1 p value x2 edge - int <0.0005 35.306 <0.0005 17.115 <0.0005 35.281 O . 0 0 0 5 34.807 edge - slope <0.0005 12.307 <0.0005 11.940 0.160 1.970 0.342 0.903 edge - chan <0.0005 28.161 <0.0005 14.306 <0.0005 15.829 <0.0005 19.081 int - slope <0.0005 35.279 <0.0005 13.694 <0.0005 35.032 <0.0005 35.325 int - chan <0.0005 35.471 <0.0005 17.871 <0.0005 36.377 <0.0005 35.839 slope - chan <0.0005 30.489 <0.0005 25.616 <0.0005 20.242 <0.0005 12.876 strata comparison M a y 2002 J u n e 2002 A u g 2002 p value x1 p value x1 p value x1 edge - int <0.0005 24.204 <0.0005 35.063 <0.0005 35.400 edge - slope 0.317 1.003 O . 0 0 0 5 23.117 O . 0 0 0 5 34.183 edge - chan 0.307 1.045 0.331 0.945 0.101 2.685 int - slope <0.0005 25.550 <0.0005 31.233 <0.0005 35.036 int - chan <0.0005 25.647 <0.0005 35.290 <0.0005 35.306 slope - chan 0.901 0.015 <0.0005 21.362 O . 0 0 0 5 25.871 143 Does seagrass density differ significantly between 5 intertidal sites? Mann-Whitney pairwise analysis of variance 5 sites; 10 pairs; a = 0.005; n = 24 Site comparison A p r i l M a y J u n e A u g p value I2 p value 1Z p value I1 p value x2 G B m a i n - G B l <0.0005 22.405 <0.0005 28.536 <0.0005 26.395 O . 0 0 0 5 25.121 G B m a i n - G B 2 <0.0005 23.091 <0.0005 26.490 <0.0005 25.744 O . 0 0 0 5 19.570 G B m a i n - L I l 0.433 0.614 <0.0005 17.527 0.180 1.797 0.042 4.128 G B m a i n - LI2 0.091 2.862 0.042 4.127 0.093 2.826 0.001 11.164 G B l - G B 2 0.613 0.256 0.332 0.941 0.885 0.021 0.403 0.698 G B l - L I l <0.0005 19.667 <0.0005 30.894 <0.0005 30.109 <0.0005 32.402 G B l - L I 2 <0.0005 18.586 <0.0005 12.154 <0.0005 14.022 0.005 7.987 G B 2 - L I l <0.0005 20.876 <0.0005 30.552 <0.0005 30.320 <0.0005 31.819 G B 2 - LI2 O . 0 0 0 5 19.948 0.002 9.572 <0.0005 13.027 0.056 3.641 L I l - L I 2 0.337 0.920 <0.0005 21.626 0.003 9.134 <0.0005 28.865 4.2 Seagrass shoot size: (refer to section 2.3.2) Does shoot length change significantly over the year at GB Main? Mann-Whitney pairwise analysis of variance 7 dates; 21 pairs; a = 0.002 Date comparison E d g e Intertidal Slope C h a n n e l p value x2 n p value x2 n p value x 2 n p value x2 n j u l y - sept 0.493 0.470 9 0.257 1.284 5 0.929 0.008 9 <0.0005 12.173 9 j u l y - dec 0.245 1.352 9 O . 0 0 0 5 13.448 9 0.003 8.716 9 0.551 0.356 9 j u l y - feb 0.029 4.739 9 <0.0005 14.337 9 <0.0005 14.337 9 0.025 5.002 9 j u l y - may 0.221 1.495 9 0.018 5.550 9 <0.0005 16.228 9 0.149 2.087 9 j u l y - june 0.047 3.943 9 0.612 0.257 9 0.004 8.364 9 0.929 0.008 9 ju ly - aug 0.170 1.882 9 0.046 3.990 9 0.788 0.072 9 0.008 7.044 9 sept - dec 0.917 0.011 15 0.001 10.714 5 <0.0005 16.530 15 0.015 5.888 15 sept - feb 0.019 5.493 15 0.001 10.714 5 O . 0 0 0 5 21.389 15 <0.0005 21.774 15 sept - may 0.011 6.509 15 0.016 5.766 5 <0.0005 21.794 15 <0.0005 13.809 15 sept -june 0.097 2.752 14 0.176 1.830 5 <0.0005 16.355 15 <0.0005 15.530 15 sept - aug 0.787 0.073 15 0.458 0.550 5 0.740 0.110 15 0.330 0.956 15 dec - feb 0.024 5.111 15 0.330 0.950 15 <0.0005 12.876 15 0.038 4.302 15 dec - may 0.018 5.592 15 0.044 4.048 15 <0.0005 16.039 15 0.192 1.702 15 dec - june 0.077 3.127 14 O . 0 0 0 5 19.514 15 0.756 0.097 15 0.384 0.759 15 dec - aug 0.967 0.002 15 O . 0 0 0 5 21.007 15 O . 0 0 0 5 18.434 15 0.038 4.302 15 feb - may 0.756 0.097 15 0.130 2.293 15 0.934 0.007 15 0.782 0.077 15 feb - june 0.395 0.725 14 <0.0005 20.253 15 <0.0005 12.873 15 0.017 5.688 15 feb - aug 0.008 6.939 15 <0.0005 21.774 15 <0.0005 21.774 15 <0.0005 16.866 15 may - june 0.570 0.322 14 0.026 4.925 15 <0.0005 18.794 15 0.185 1.754 15 may - aug 0.010 6.725 15 <0.0005 13.942 15 <0.0005 21.794 15 0.001 11.027 15 june - aug 0.018 5.557 14 0.014 6.091 15 <0.0005 19.514 15 0.002 9.549 15 144 Does shoot width change significantly over the year at GB Main? Mann-Whitney pairwise analysis of variance 7 dates; 21 pairs; a = 0.002 Date comparison E d g e Intertidal Slope C h a n n e l p value x2 n p value x2 n p value x1 n p value x1 n July - sept 0.474 0.513 9 0.205 1.608 5 0.402 0.701 9 0.004 8.075 9 July - dec 1.000 0.000 9 <0.0005 13.483 9 0.018 5.557 9 <0.0005 12.613 9 July - feb 0.030 4.681 9 <0.0005 12.200 9 0.001 10.239 9 0.002 9.678 9 ju ly - may 0.743 0.108 9 0.232 1.427 9 <0.0005 15.809 9 0.607 0.265 9 J u l y - j u n e 0.849 0.036 9 0.788 0.072 9 0.011 6.436 9 0.855 0.034 9 ju ly - aug 0.009 6.781 9 0.811 0.057 9 0.676 0.175 9 0.006 7.610 9 sept - dec 0.633 0.229 15 0.003 8.563 5 0.009 6.749 15 0.587 0.294 15 sept - feb 0.105 2.629 15 0.008 7.114 5 <0.0005 13.713 15 0.130 2.294 15 sept - may 0.467 0.529 15 0.861 0.031 5 <0.0005 21.470 15 0.002 9.500 15 sept -june 0.600 0.275 14 0.457 0.552 5 0.008 7.074 15 0.002 9.851 15 sept - aug 0.338 0.919 15 0.826 0.048 5 0.280 1.162 15 0.349 0.876 15 dec - feb 0.226 1.466 15 0.493 0.470 15 0.018 5.636 15 0.001 10.533 15 dec - may 0.917 0.011 15 0.001 10.749 15 <0.0005 20.887 15 0.005 7.864 15 dec - june 0.630 0.231 14 <0.0005 21.789 15 0.212 1.555 15 <0.0005 14.278 15 dec - aug 0.058 3.599 15 O . 0 0 0 5 15.412 15 0.001 11.166 15 0.802 0.063 15 feb - may 0.190 1.720 15 0.001 10.230 15 0.006 7.689 15 0.018 5.621 15 feb - june 0.021 5.306 14 <0.0005 21.803 15 0.417 0.658 15 0.001 10.166 15 feb - aug 0.001 11.847 15 <0.0005 14.160 15 <0.0005 17.605 15 0.027 4.874 15 may - june 0.948 0.004 14 0.054 3.725 15 <0.0005 13.296 15 0.281 1.161 15 may - aug 0.096 2.778 15 0.328 0.955 15 <0.0005 21.837 15 0.024 5.121 15 june - aug 0.027 4.882 14 0.950 0.004 15 0.001 12.006 15 0.003 8.556 15 Does seagrass shoot length change significantly over the summer at the 5 intertidal sites? Mann-Whitney pairwise analysis of variance 4 dates; 6 pairs; a = 0.008; n = 15 (except LI 2 in June where n = 14) Date comparison G B M a i n G B 1 G B 2 L I 1 L I 2 p value x2 p value x2 p value x2 p value x2 p value x2 apr - may 0.901 0.015 0.868 0.028 0.633 0.228 0.885 0.021 0.006 7.612 apr - june 0.015 5.889 0.648 0.208 0.468 0.527 0.012 6.297 0.061 3.522 apr - aug <0.0005 15.367 <0.0005 14.404 0.004 8.312 0.002 9.807 O . 0 0 0 5 21.774 may - june 0.026 4.925 0.663 0.190 0.135 2.230 0.010 6.720 0.678 0.172 may - aug <0.0005 13.942 <0.0005 12.873 0.001 11.710 0.002 9.549 O . 0 0 0 5 21.389 june - aug 0.014 6.091 O . 0 0 0 5 17.724 0.011 6.507 0.576 0.314 O . 0 0 0 5 18.293 145 Does seagrass shoot width change significantly over the summer at the 5 intertidal sites? Mann-Whitney pairwise analysis of variance 4 dates; 6 pairs; a = 0.008; n = 15 (except LI 2 in June where n = 14) Date comparison G B M a i n G B 1 G B 2 L I 1 L I 2 p value x2 p value x1 p value x1 p value x1 p value x1 apr - may 0.016 5.795 0.262 1.256 0.085 2.975 0.221 1.497 0.004 8.220 apr- june O . 0 0 0 5 15.367 0.835 0.043 0.032 4.576 0.983 0.000 0.049 3.874 apr - aug 0.004 8.211 0.016 5.823 0.618 0.248 0.191 1.709 <0.0005 13.496 may - june 0.054 3.725 0.110 2.552 0.819 0.053 0.011 6.405 0.394 0.728 may - aug 0.328 0.955 0.589 0.292 0.053 3.746 0.884 0.021 0.054 3.726 june - aug 0.950 0.040 0.012 6.318 0.070 3.275 0.015 5.898 0.052 3.779 Does seagrass shoot length vary between the four tidal strata? Mann-Whitney pairwise analysis of variance 4 strata; 6 pairs; a = 0.008 strata comparison J u l y 2001 Sept 2001 Dec 2001 F e b 2002 p value x1 n p value x1 n p value x1 n p value x2 n edge - int 0.002 9.281 9 0.040 4.208 5 0.029 4.742 15 0.191 1.707 15 edge - slope 0.022 5.276 9 <0.0005 13.172 15 0.191 1.707 15 0.014 6.092 15 edge - chan 0.015 5.093 9 0.078 3.018 15 0.221 1.497 15 0.025 5.019 15 int - slope 0.047 3.947 9 0.127 2.333 5 0.395 0.724 15 <0.0005 14.404 15 int - chan 0.001 10.399 9 0.513 0.429 5 0.548 0.362 15 <0.0005 17.724- 15 slope - chan 0.002 9.837 9 0.034 4.476 15 1.000 0.000 15 0.350 0.872 15 strata comparison M a y 2002 June 2002 A u g 2002 p value x2 n p value x2 n p value x2 n edge - int 0.044 4.049 15 <0.0005 15.436 14 0.001 11.710 15 edge - slope 0.101 2.687 15 0.275 1.191 14 <0.0005 18.430 15 edge - chan 0.678 0.172 15 0.221 1.495 14 0.443 0.589 15 int - slope <0.0005 14.737 15 0.028 4.834 15 0.310 1.033 15 int - chan 0.007 7.228 15 <0.0005 18.788 15 0.011 6.507 15 slope - chan 0.333 0.939 15 0.051 3.801 15 <0.0005 12.577 15 146 Does seagrass shoot width vary between the four tidal strata? Mann-Whitney pairwise analysis of variance 4 dates; 6 pairs; a = 0.008 strata comparison J u l y 2001 Sept 2001 Dec 2001 F e b 2002 p value t n p value x2 n p value x2 n p value x2 n edge - int 0.030 4.695 9 0.930 0.008 5 0.071 3.270 15 0.560 0.340 15 edge - slope 0.008 7.061 9 0.040 4.236 15 0.480 0.500 15 0.851 0.035 15 edge - chan <0.0005 12.318 9 0.339 0.914 15 0.428 0.629 15 0.832 0.045 15 int - slope 0.171 1.875 9 0.220 1.506 15 <0.0005 12.756 15 0.692 0.157 15 int - chan <0.0005 12.305 9 0.695 0.155 15 <0.0005 12.644 15 0.689 0.160 15 slope - chan <0.0005 12.293 9 0.371 0.800 15 0.573 0.318 15 0.882 0.022 15 strata comparison M a y 2002 June 2002 A u g 2002 p value x2 n p value x 2 n p value x2 n edge - int 0.724 0.125 15 0.003 8.565 14 0.967 0.002 15 edge - slope 0.001 10.697 15 0.294 1.103 14 0.016 5.816 15 edge - chan 0.074 3.182 15 <0.0005 15.709 14 0.144 2.134 15 int - slope <0.0005 19.217 15 0.020 5.405 15 0.028 4.848 15 int - chan 0.010 6.626 15 <0.0005 19.623 15 0.218 1.519 15 slope - chan 0.349 0.877 15 0.024 5.103 15 0.003 8.821 15 Does seagrass shoot length vary between the five intertidal sites? Mann-Whitney pairwise analysis of variance 5 sites; 10 pairs; a = 0.005; n = 15 (except LI 2 in June where n = 14) Site comparison A p r i l M a y June A u g p value x2 p value x2 p value t p value x2 G B m a i n - G B l 0.001 10.876 0.029 4.773 0.407 0.688 0.010 6.720 G B m a i n - G B 2 0.001 11.440 0.001 11.998 0.003 8.795 0.141 2.168 G B m a i n - L I l 0.025 5.019 0.054 3.721 0.003 8.795 0.852 0.035 G B m a i n - LI2 0.191 1.707 0.001 11.016 0.383 0.762 0.002 9.294 G B l - G B 2 0.575 0.314 0.663 0.190 0.017 5.688 0.062 3.485 G B l - L I l 0.221 1.497 0.290 1.119 0.071 3.255 <0.0005 13.781 G B l - L I 2 0.006 7.608 0.950 0.004 0.861 0.030 0.548 0.362 G B 2 - L I l 0.036 4.389 0.081 3.036 0.724 0.124 0.010 6.720 G B 2 - LI2 0.012 6.299 0.468 0.527 0.089 2.897 0.049 3.882 L I l - LI2 0.281 1.163 0.125 2.356 0.081 3.048 <0.0005 15.691 147 Does seagrass shoot width vary between the five intertidal sites? Mann-Whitney pairwise analysis of variance 5 sites; 10 pairs; a = 0.005; n = 15 (except LI 2 in June where n = 14) Site comparison A p r i l M a y June A u g p value x2 p value x2 p value x 2 p value x2 G B m a i n - G B l <0.0005 18.264 <0.0005 12.585 0.065 3.414 0.001 11.486 G B m a i n - G B 2 0.001 11.440 <0.0005 21.813 O . 0 0 0 5 16.224 0.044 4.065 G B m a i n - L I l <0.0005 15.698 0.059 3.566 0.031 4.655 0.884 0.021 G B m a i n - L I 2 <0.0005 13.954 O . 0 0 0 5 21.794 <0.0005 12.365 <0.0005 12.475 G B l - G B 2 0.480 0.499 0.835 0.043 0.071 3.267 0.005 7.775 G B l - L I l 0.062 0.062 0.001 10.611 0.619 0.248 <0.0005 17.768 G B l - L I 2 0.298 1.082 0.917 0.011 0.285 1.145 0.561 0.338 G B 2 - L I l 0.520 0.414 <0.0005 20.294 0.003 8.572 0.042 4.138 G B 2 - L I 2 0.708 0.140 0.868 0.028 0.512 0.431 0.002 9.315 L I l - L I 2 0.724 0.125 <0.0005 20.276 0.046 3.968 <0.0005 19.899 4.3 Seagrass shoot biomass (refer to section 2.3.3) Does seagrass biomass change significantly over the year at GB Main? Mann-Whitney pairwise analysis of variance 7 dates; 21 pairs; a = 0.002 Date comparison E d g e Intertidal Slope C h a n n e l p value x2 n p value x2 n p value x2 n p value x2 n ju ly - sept 0.114 2.497 9 0.053 3.738 5 0.245 1.352 9 0.736 0.113 9 j u l y - dec 0.002 9.800 9 <0.0005 15.254 9 0.003 9.068 9 0.104 2.640 9 ju ly - feb 0.006 7.688 9 <0.0005 14.792 9 O . 0 0 0 5 12.588 9 0.293 1.107 9 j u l y - may 0.003 9.068 9 <0.0005 12.588 9 <0.0005 12.588 9 0.157 2.006 9 j u l y - june 0.001 10.730 9 0.008 7.041 9 0.011 6.422 9 0.420 0.651 9 j u l y - aug 0.001 10.607 9 0.325 0.968 9 0.297 1.089 9 0.039 4.239 9 sept - dec 0.101 2.684 15 0.026 4.954 5 <0.0005 12.577 15 0.207 1.590 15 sept - feb 0.178 1.817 15 0.032 4.573 5 <0.0005 14.720 15 0.258 1.278 15 sept - may 0.049 3.882 15 0.061 3.522 5 <0.0005 12.284 15 0.461 0.544 15 sept - june 0.023 5.150 14 0.827 0.048 5 0.003 8.551 15 0.690 0.160 15 sept - aug 0.032 4.573 15 0.407 0.688 5 0.885 0.021 15 0.155 2.027 15 dec - feb 0.694 0.155 15 0.633 0.228 15 0.089 2.893 15 0.656 0.198 15 dec - may 0.507 0.441 15 0.049 3.882 15 0.017 5.688 15 0.006 7.560 15 dec - june 0.132 2.268 14 <0.0005 21.007 15 0.206 1.600 15 0.017 5.659 15 dec - aug 0.418 0.657 15 <0.0005 17.033 15 <0.0005 21.389 15 0.002 9.508 15 feb - may 0.431 0.621 15 0.093 2.822 15 0.694 0.155 15 0.008 6.975 15 feb - june 0.089 2.897 14 <0.0005 20.253 15 0.011 6.507 15 0.036 4.376 15 feb - aug 0.418 0.656 15 O . 0 0 0 5 17.033 15 <0.0005 21.774 15 0.002 9.291 15 may - june 0.471 0.519 14 0.001 11.428 15 0.006 7.608 15 0.607 0.264 15 may - aug 0.819 0.052 15 0.001 10.602 15 <0.0005 21.774 15 0.277 1.182 15 june - aug 0.570 0.323 14 0.178 1.817 15 0.001 11.149 15 0.164 1.932 15 148 Does seagrass biomass change significantly over the summer at the intertidal sites? Mann-Whitney pairwise analysis of variance 4 dates; 6 pairs; a = 0.008; n = 15 (except LI 2 in June where n = 14) Date comparison G B M a i n G B 1 G B 2 L I 1 L I 2 p value I1 p value x2 p value x1 p value x1 p value x1 apr - may 0.820 0.052 0.663 0.190 0.373 0.795 0.206 1.600 0.078 3.108 apr - june 0.015 5.888 0.787 0.073 0.004 8.073 0.012 6.297 0.036 4.389 apr - aug 0.003 8.551 <0.0005 13.475 0.011 6.507 0.110 2.550 <0.0005 20.628 may - june 0.001 11.428 0.852 0.035 0.007 7.381 0.191 1.707 1.000 0.000 may - aug 0.001 10.602 0.007 7.157 0.019 5.492 0.852 0.035 O . 0 0 0 5 12.284 june - aug 0.178 1.817 <0.0005 12.284 0.604 0.269 0.178 1.817 <0.0005 13.762 Does seagrass biomass differ significantly between the 4 tidal strata? Mann-Whitney pairwise analysis of variance 4 strata; 6 pairs; a = 0.008 strata Ju ly 2001 Sept 2001 Dec 2001 F e b 2002 comparison p value x2 n p value x2 n p value x2 n p value x2 n edge - int <0.0005 12.789 9 0.003 8.550 5 O . 0 0 0 5 19.149 15 <0.001 20.253 15 edge - slope 0.019 5.476 9 <0.0005 13.475 15 0.097 2.753 15 0.644 0.228 15 edge - chan <0.0005 12.803 9 0.021 5.366 15 0.004 8.488 15 <0.001 14.044 15 int - slope 0.453 0.563 9 0.965 0.002 5 <0.0005 13.475 15 <0.001 21.389 15 int - chan <0.0005 12.803 9 0.002 10.024 5 <0.0005 21.025 15 <0.001 22.189 15 slope - chan <0.0005 12.803 9 <0.0005 16.903 15 0.001 12.016 15 0.002 9.811 15 strata comparison M a y 2002 June 2002 A u g 2002 p value x2 n p value x2 n p value x2 n edge - int <0.0005 18.430 15 <0.0005 21.000 14 <0.0005 21.823 15 edge - slope 0.494 0.468 15 0.001 10.430 14 <0.0005 21.823 15 edge - chan 0.406 0.689 15 0.541 0.374 14 0.394 0.726 15 int - slope <0.0005 21.774 15 <0.0005 13.781 15 0.078 3.108 15 int - chan <0.0005 20.664 15 <0.0005 21.779 15 <0.0005 18.075 15 slope - chan 0.053 3.752 15 0.001 10.604 15 <0.0005 17.033 15 149 Does seagrass biomass differ significantly between the 5 intertidal sites? Mann-Whitney pairwise analysis of variance 5 sites; 10 pairs; a = 0.005; n = 15 (except LI 2 in June where n = 14) Site comparison A p r i l M a y June A u g p value x2 p value x2 p value x2 p value x2 G B m a i n - G B l 0.330 0.950 0.221 1.497 0.029 4.742 0.419 0.654 G B m a i n - G B 2 0.373 0.795 0.021 5.299 0.049 3.882 0.917 0.011 G B m a i n - L I l 0.014 6.091 <0.0005 14.090 <0.0005 13.475 0.694 0.155 G B m a i n - LI2 0.093 2.822 0.002 9.549 0.074 3.202 0.004 8.073 G B l - G B 2 0.663 0.190 0.950 0.004 0.006 7.608 0.221 1.497 G B l - L I l 0.071 3.255 0.029 4.742 <0.0005 16.692 0.300 0.950 G B l - L I 2 0.237 1.397 0.078 3.108 0.003 9.069 0.005 7.839 G B 2 - L I l 0.065 3.407 0.011 6.507 0.010 6.720 0.756 0.097 G B 2 - L I 2 0.272 1.208 0.078 3.108 0.694 0.154 <0.0005 13.475 L I l - L I 2 0.330 0.950 0.756 0.097 0.097 2.750 <0.0005 13.781 4.4 Seagrass reproduction (refer to section 2.3.4) Does seagrass flowering intensity change significantly over the year at GB Main? Mann-Whitney pairwise analysis of variance 7 dates; 21 pairs; a = 0.002; n = 24 (except Intertidal in September where n = 8) Date comparison E d g e Intertidal Slope C h a n n e l p value x2 p value t p value x2 p value x2 j u ly - sept <0.0005 13.283 <0.0005 16.394 0.001 10.723 0.317 1.000 j u l y - dec <0.0005 20.418 O . 0 0 0 5 37.712 <0.0005 22.268 0.317 1.000 j u l y - feb <0.0005 20.418 <0.0005 37.712 <0.0005 22.268 0.317 1.000 ju ly - may 0.028 4.808 O . 0 0 0 5 36.124 <0.0005 22.268 0.317 1.000 j u l y - j u n e 0.002 10.041 <0.0005 19.716 <0.0005 16.428 0.317 1.000 j u l y - aug 0.007 7.218 O . 0 0 0 5 14.283 0.007 7.278 0.589 0.292 sept - dec 0.039 4.263 1.000 0.000 0.020 5.444 1.000 0.000 sept - feb 0.039 4.263 1.000 0.000 0.020 5.444 1.000 0.000 sept - may 0.218 1.518 0.407 0.688 0.020 5.444 1.000 0.000 sept -june 0.607 0.265 0.001 10.714 0.586 0.297 1.000 0.000 sept - aug 0.824 0.049 0.001 10.711 0.614 0.254 0.153 2.043 dec - feb 1.000 0.000 1.000 0.000 1.000 0.000 1.000 0.000 dec - may 0.005 7.960 0.153 2.043 1.000 0.000 1.000 0.000 dec -june 0.020 5.444 <0.0005 26.240 0.039 4.263 1.000 0.000 dec - aug 0.039 4.264 <0.0005 26.238 0.010 6.676 0.153 2.043 feb - may 0.005 7.960 0.153 2.043 1.000 0.000 1.000 0.000 feb - june 0.020 5.444 <0.0005 26.240 0.039 4.263 1.000 0.000 feb - aug 0.039 4.264 <0.0005 26.238 0.010 6.676 0.153 2.043 may - june 0.448 0.576 O . 0 0 0 5 21.857 0.039 4.263 1.000 0.000 may - aug 0.417 0.660 <0.0005 22.071 0.010 6.676 0.153 2.043 june - aug 0.856 0.033 0.448 0.575 0.296 1.093 0.153 2.043 150 Does seagrass flowering intensity change significantly over the summer at the intertidal sites? Mann-Whitney pairwise analysis of variance 4 dates; 6 pairs; a = 0.008; n = 15 (except LI 2 in June where n = 14) Date comparison G B M a i n G B 1 G B 2 L I 1 L I 2 p value I1 p value I1 p value t p value x2 p value x2 apr - may 0.820 0.052 0.663 0.190 0.373 0.795 0.206 1.600 0.078 3.108 apr - june 0.015 5.888 0.787 0.073 0.004 8.073 0.012 6.297 0.036 4.389 apr - aug 0.003 8.551 <0.0005 13.475 0.011 6.507 0.110 2.550 <0.0005 20.628 may - june 0.001 11.428 0.852 0.035 0.007 7.381 0.191 1.707 1.000 0.000 may - aug 0.001 10.602 0.007 7.157 0.019 5.492 0.852 0.035 <0.0005 12.284 june - aug 0.178 1.817 <0.0005 12.284 0.604 0.269 0.178 1.817 <0.0005 13.762 Does seagrass flowering intensity differ significantly between the four tidal strata? Mann-Whitney pairwise analysis of variance 4 strata; 6 pairs; a = 0.008; n = 24 (except Intertidal in September where n = 8) strata comparison J u l y 2001 Sept 2001 Dec 2001 F e b 2002 p value x2 p value t p value p value edge - int 0.507 0.440 0.226 1.468 1.000 1.000 edge - slope 0.364 0.823 0.785 0.074 1.000 1.000 edge - chan <0.0005 15.963 0.039 4.263 1.000 1.000 int - slope 0.234 1.415 0.169 1.896 1.000 1.000 int - chan O . 0 0 0 5 30.976 1.000 0.000 1.000 1.000 slope - chan <0.0005 17.313 0.020 5.444 1.000 1.000 strata comparison M a y 2002 June 2002 A u g 2002 p value x2 p value x2 p value t edge - int 0.042 4.116 0.008 7.129 0.003 8.931 edge - slope 0.005 7.960 0.505 0.444 0.674 0.177 edge - chan 0.005 7.960 0.020 5.444 0.351 0.871 int - slope 0.153 2.043 <0.0005 15.657 0.023 5.154 int - chan 0.153 2.043 <0.0005 26.240 <0.0005 16.057 slope - chan 1.000 0.000 0.039 4.263 0.144 2.135 151 Does seagrass flowering intensity differ significantly between the five intertidal sites? Mann-Whitney pairwise analysis of variance 5 sites; 10 pairs; a = 0.005; n = 15 (except LI 2 in June where n = 14) Site comparison A p r i l M a y June A u g p value I1 p value ll p value x1 p value x1 G B m a i n - G B l 0.330 0.950 0.221 1.497 0.029 4.742 0.419 0.654 G B m a i n - G B 2 0.373 0.795 0.021 5.299 0.049 3.882 0.917 0.011 G B m a i n - L I l 0.014 6.091 O . 0 0 0 5 14.090 <0.0005 13.475 0.694 0.155 G B m a i n - LI2 0.093 2.822 0.002 9.549 0.074 3.202 0.004 8.073 G B l - G B 2 0.663 0.190 0.950 0.004 0.006 7.608 0.221 1.497 G B l - L I l 0.071 3.255 0.029 4.742 <0.0005 16.692 0.300 0.950 G B l - L I 2 0.237 1.397 0.078 3.108 0.003 9.069 0.005 7.839 G B 2 - L I l 0.065 3.407 0.011 6.507 0.010 6.720 0.756 0.097 G B 2 - LI2 0.272 1.208 0.078 3.108 0.694 0.154 <0.0005 13.475 L I l - L I 2 0.330 0.950 0.756 0.097 0.097 2.750 <0.0005 13.781 Appendix 5: Seagrass carbohydrate and chlorophyll statistics 5.1 Seagrass carbohydrates (refer to section 3.3.1) Do carbohydrate concentrations different between shoots and rhizomes of the same plant? Non-parametric paired t-test: Wilcoxon Site A p r i l J u n e p value Z n p value z n Glucose G B M a i n 0.136 1.490 12 0.021 2.310 9 G B 1 0.005 2.824 12 0.008 2.666 9 G B 2 0.937 -0.078 12 0.011 2.547 9 L I 1 0.003 2.981 12 0.008 2.666 9 L I 2 0.003 2.981 12 0.008 2.666 9 al l 5 <0.0005 5.330 60 <0.0005 5.774 45 Starch G B M a i n 0.018 2.366 7 0.225 1.214 5 G B 1 0.018 2.366 7 0.500 0.674 5 G B 2 0.018 2.366 7 0.249 1.153 6 L I 1 0.128 1.521 7 0.046 1.992 6 L I 2 0.310 1.014 7 0.753 -0.314 6 al l 5 <0.005 4.275 35 0.010 2.573 28 152 Does glucose and starch concentrations differ significantly between April and June in the shoots and rhizomes at the five intertidal sites investigated? Mann-Whitney Analysis of Variance a = 0.05 site Glucose S ta rch p value Chi2 n p value Chi2 n Shoots G B M a i n 0.118 2.444 9 0.380 0.771 5 G B 1 0.002 9.778 9 0.372 0.798 5 G B 2 <0.0005 14.727 9 0.391 0.735 6 L I 1 O . 0 0 0 5 14.727 9 0.317 1.000 6 L I 2 0.001 10.227 9 0.366 0.817 6 all 5 <0.0005 43.537 45 0.828 0.051 28 Rh izomes G B M a i n 0.007 7.293 9 0.519 0.417 6 G B 1 O . 0 0 0 5 13.136 9 0.199 1.653 6 G B 2 <0.0005 13.136 9 1.000 0.000 6 L I 1 <0.0005 14.727 9 0.668 0.184 6 L I 2 <0.0005 14.727 9 0.475 0.510 6 all 5 <0.0005 58.877 45 0.420 0.651 30 153 Does carbohydrate concentration in the shoots and rhizomes differ between the five intertidal sites? Kruskal-Wallis non-parametric analysis of variance used to determine i f there are any significant differences between sites. Mann-Whitney pairwise analysis used to determine between which pairs these significant differences lie (not necessary to perform i f Kruskal-Wallis test is non-significant. 5 sites; 10 pairs; a = 0.005 Shoot glucose A p r i l June C h i 2 p value n C h i 2 p value n Kruska l -Wal l i s 17.540 0.002 12 36.484 <0.0005 9 G B m a i n - G B l 1.920 0.166 12 8.750 0.003 9 G B m a i n - G B 2 0.270 0.603 12 12.789 <0.0005 9 G B m a i n - L I 1 4.320 0.038 12 12.789 O . 0 0 0 5 9 G B m a i n - LI2 9.720 0.002 12 10.388 0.001 9 G B l - G B 2 4.083 0.043 12 12.789 <0.0005 9 G B l - L I l 0.030 0.862 12 12.789 <0.0005 9 G B l - L I 2 2.803 0.094 12 6.786 0.009 9 G B 2 - L I l 6.750 0.009 12 0.018 0.895 9 G B 2 - LI2 11.603 0.001 12 11.558 0.001 9 L I l - LI2 1.470 0.225 12 9.827 0.002 9 Shoot starch A p r i l June C h i 2 p value n C h i 2 p value n Kruska l -Wal l i s 6.445 0.168 7 9.039 0.600 5 R h i z o m e glucose A p r i l June C h i 2 p value n C h i 2 p value n K r u s k a l - W a l l i s 16.704 0.002 12 7.047 0.133 9 G B m a i n - G B l 1.203 0.273 12 G B m a i n - G B 2 0.403 0.525 12 G B m a i n - L I l 1.613 0.204 12 G B m a i n - LI2 3.413 0.065 12 G B l - G B 2 7.680 0.006 12 G B l - L I l 0.053 0.817 12 G B l - L I 2 0.963 0.326 12 G B 2 - L I l 11.603 0.001 12 G B 2 - LI2 13.653 <0.0005 12 L I l - L I 2 1.470 0.225 12 R h i z o m e starch A p r i l June C h i 2 p value n C h i 2 p value n Kruska l -Wal l i s 5.610 0.230 7 2.013 0.733 6 154 5.2 Seagrass shoot chlorophyll (refer to section 3.3.2) Does chlorophyll concentration in seagrass shoots change significantly over time? Sites are lumped; Mann-Whitney Pairwise A N O V A ; 5 dates; 10 pairs; a = 0.005 C h l o r o p h y l l a dates p value Chi2 n apr - may <0.0005 74.139 113 apr - june <0.0005 120.870 113 apr - ju ly <0.0005 46.430 113 apr - aug <0.0005 40.250 100 may - june 0.031 4.628 149 may - ju ly <0.0005 17.548 150 may - aug 0.004 8.190 100 june - ju ly O . 0 0 0 5 58.270 149 june - aug <0.0005 31.189 100 j u l y - aug 0.334 0.933 100 C h l o r o p h y l l b dates p value Chi2 n apr - may <0.0005 35.864 113 apr - june <0.0005 48.822 113 a p r - j u l y 0.027 4.866 113 apr - aug 0.340 0.909 100 may - june 0.440 0.596 149 may - j u l y O . 0 0 0 5 29.611 150 may - aug <0.0005 48.354 100 june - j u l y <0.0005 46.550 149 june - aug <0.0005 66.035 100 j u l y - aug 0.001 10.384 100 T o t a l C h l o r o p h y l l dates p value Chi2 n apr - may <0.0005 61.503 113 apr - june <0.0005 95.762 113 a p r - j u l y <0.0005 28.169 113 apr - aug <0.0005 13.217 100 may - june 0.107 2.603 149 may - j u l y <0.0005 22.989 150 may - aug <0.0005 23.287 100 june - j u l y O . 0 0 0 5 56.809 149 june - aug <0.0005 48.210 100 j u l y - aug 0.803 0.803 100 155 C h l o r o p h y l l a:b ratio dates p value Chi2 n apr - may <0.0005 26.924 113 apr - june <0.0005 62.563 113 apr - ju ly <0.0005 71.505 113 apr - aug <0.0005 114.270 100 may - june <0.0005 12.528 150 may - ju ly 0.001 11.735 150 may - a u g <0.0005 71.251 100 june - ju ly 0.443 0.588 150 june - aug <0.0005 40.600 100 ju ly - aug O . 0 0 0 5 53.623 100 Does chlorophyll concentration vary significantly between the intertidal beds investigated? Mann-Whitney pairwise analysis of variance 5 sites; 10 pairs; a = 0.005 C h l o r o p h y l l a sites p value C h i 2 n G B m a i n - G B l 0.020 5.378 120 G B m a i n - G B 2 <0.0005 22.017 125 G B m a i n - L I l <0.0005 29.836 125 G B m a i n - LI2 O . 0 0 0 5 62.231 125 G B l - G B 2 <0.0005 12.554 120 G B l - L I l <0.0005 17.059 120 G B l - L I 2 <0.0005 48.938 120 G B 2 - L I l 0.965 0.002 139 G B 2 - LI2 <0.0005 12.604 139 L I l - LI2 <0.0005 15.310 139 C h l o r o p h y l l b sites p value C h i 2 n G B m a i n - G B l 0.001 10.808 120 G B m a i n - G B 2 <0.0005 16.113 125 G B m a i n - L I l 0.009 6.912 125 G B m a i n - LI2 <0.0005 45.005 125 G B l - G B 2 0.036 4.421 120 G B l - L I l 0.457 0.553 120 G B l - L I 2 O . 0 0 0 5 27.972 120 G B 2 - L I 1 0.012 6.369 139 G B 2 - L I 2 0.001 11.812 139 L I l - L I 2 <0.0005 34.094 139 T o t a l C h l o r o p h y l l sites p value C h i 2 n G B m a i n - G B l 0.005 7.924 120 G B m a i n - G B 2 O . 0 0 0 5 20.077 125 G B m a i n - L I l <0.0005 19.516 125 G B m a i n - L I 2 O . 0 0 0 5 56.147 125 G B l - G B 2 0.002 9.717 120 G B l - L I l 0.013 6.184 120 G B l - L I 2 <0.0005 41.302 120 G B 2 - L I 1 0.267 1.231 139 G B 2 - LI2 0.001 12.059 139 L I l - L I 2 <0.0005 22.788 139 C h l o r o p h y l l a:b ratio sites p value C h i 2 n G B m a i n - G B l 0.010 6.604 120 G B m a i n - G B 2 0.391 0.735 125 G B m a i n - L I l <0.0005 26.282 125 G B m a i n - LI2 0.013 6.234 125 G B l - G B 2 0.001 10.382 120 G B l - L I l <0.0005 47.886 120 G B l - L I 2 O . 0 0 0 5 22.690 120 G B 2 - L I 1 <0.0005 14.618 139 G B 2 - LI2 0.195 1.677 139 L I l - L I 2 <0.0005 13.259 139 Appendix 6: Ghost shrimp statistics 6.1 Ghost shrimp distribution (refer to section 4.3.1) Does ghost shrimp density change significantly over time? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002; n = 84 Date comparison S i t e l Site 2 p value t p value I1 may - Jul 0.003 8.665 0.008 7.145 may - sept 0.008 6.935 <0.0005 25.325 may - nov 0.001 11.912 <0.0005 14.674 may - feb 0.023 5.148 0.032 4.587 may - apr 0.958 0.003 0.253 1.308 may - j un 0.004 8.493 <0.0005 13.926 may - aug 0.001 10.100 <0.0005 14.961 ju l - sept 0.900 0.016 0.005 7.785 j u l - nov 0.374 0.792 0.184 1.764 j u l - feb 0.475 0.510 0.676 0.174 j u l - apr 0.005 7.778 0.101 2.688 ju l - jun 0.965 0.002 0.198 1.658 j u l - aug 0.545 0.367 0.135 2.239 sept - n o v 0.160 1.976 0.102 2.667 sept - feb 0.621 0.245 0.003 8.550 sept - apr 0.024 5.093 <0.0005 16.718 sept - jun 0.700 0.149 0.207 1.593 sept - aug 0.311 1.028 0.187 1.742 nov - feb 0.176 1.831 0.099 2.727 nov - apr 0.002 10.071 0.002 9.411 n o v - j u n 0.297 1.086 0.986 O . 0 0 0 5 nov - aug 0.577 0.312 0.851 0.035 feb - apr 0.034 4.470 0.319 0.992 feb - jun 0.520 0.415 0.087 2.928 feb - aug 0.247 1.341 0.076 3.146 a p r - j u n 0.009 6.857 0.007 7.212 apr - aug 0.005 8.046 0.004 8.357 jun - aug 0.611 0.259 0.876 0.024 Does ghost shrimp biomass change significantly over time? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002; n = 84 Date comparison S i t e l Site 2 p value x1 p value I1 may - j u l 0.441 0.595 0.278 1.179 may - sept 0.102 2.680 0.008 6.949 may - nov 0.051 3.809 0.096 2.777 may - feb 0.646 0.211 0.263 1.255 may - apr 0.123 2.376 0.647 0.210 may - jun 0.191 1.712 0.120 2.422 may -aug 0.021 5.342 0.010 6.649 j u l - sept 0.204 1.610 0.057 3.609 j u l - nov 0.104 2.638 0.451 0.567 j u l - feb 0.466 0.532 0.800 0.064 j u l - apr 0.008 7.060 0.398 0.714 j u l - j u n 0.329 0.955 0.615 0.253 j u l - aug 0.037 4.331 0.094 2.800 sept - nov 0.637 0.227 0.259 1.272 sept - feb 0.128 2.311 0.128 2.321 sept - apr 0.001 10.237 0.013 6.191 sept - j un 0.666 0.186 0.173 1.843 sept - aug 0.444 0.585 0.691 0.158 nov - feb 0.089 2.897 0.791 0.070 nov - apr <0.0005 14.304 0.099 2.717 nov - j un 0.376 0.783 0.899 0.016 nov - aug 0.597 0.279 0.315 1.008 feb - apr 0.043 4.089 0.507 0.439 feb - jun 0.185 1.759 0.784 0.075 feb - aug 0.026 4.968 0.207 1.593 a p r - j u n 0.001 10.353 0.214 1.545 apr - aug <0.0005 15.393 0.016 5.835 jun - aug 0.196 1.676 0.271 1.212 Does ghost shrimp density differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003; n = 112 Zone comparison Site 1 Site 2 p value I1 p value I1 1 to 2 <0.0005 67.218 O . 0 0 0 5 72.111 1 to 3 <0.0005 120.381 <0.0005 93.980 1 to 4 <0.0005 140.246 <0.0005 40.031 1 to 5 <0.0005 110.380 <0.0005 50.640 1 to 6 0.075 3.174 <0.0005 82.334 2 to 3 <0.0005 40.416 0.970 0.001 2 to 4 O . 0 0 0 5 72.246 <0.0005 14.965 2 to 5 O . 0 0 0 5 39.723 <0.0005 130.674 2 to 6 <0.0005 13.194 <0.0005 127.733 3 to 4 0.007 7.337 <0.0005 23.774 3 to 5 0.760 0.093 O . 0 0 0 5 148.514 3 to 6 <0.0005 51.080 <0.0005 131.954 4 to 5 0.012 6.252 <0.0005 110.626 4 to 6 O . 0 0 0 5 73.432 <0.0005 119.026 5 to 6 <0.0005 48.364 0.012 6.253 Does ghost shrimp biomass differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003; n = 112 Zone comparison Site 1 Site 2 p value x 2 p value x2 1 to 2 <0.0005 43.424 <0.0005 52.494 1 to 3 <0.0005 125.062 <0.0005 126.951 1 to 4 <0.0005 150.629 <0.0005 27.954 1 to 5 <0.0005 107.960 <0.0005 116.292 1 to 6 0.073 3.205 <0.0005 132.177 2 to 3 <0.0005 78.907 <0.0005 49.121 2 to 4 <0.0005 133.858 0.031 4.643 2 to 5 <0.0005 79.384 O . 0 0 0 5 157.441 2 to 6 <0.0005 40.134 <0.0005 167.695 3 to 4 <0.0005 38.598 0.056 3.661 3 to 5 <0.0005 25.730 O . 0 0 0 5 170.366 3 to 6 <0.0005 59.616 <0.0005 174.625 4 to 5 0.437 0.604 <0.0005 136.592 4 to 6 <0.0005 77.066 <0.0005 161.57 5 to 6 <0.0005 67.483 0.001 10.423 160 6.2 Ghost shrimp population structure (refer to section 4.3.2) Does ghost shrimp length change significantly over time? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002 Date comparison Site 1 Site 2 p value t n p value x1 n may - j u l 0.008 6.980 457 O . 0 0 0 5 14.991 275 may - sept 0.363 0.827 457 0.267 1.231 275 may - nov 0.853 0.034 457 <0.0005 18.182 275 may - feb 0.045 4.007 457 0.169 1.890 275 may - apr 0.109 2.571 457 0.163 1.950 275 may - jun 0.884 0.021 457 <0.0005 16.893 275 may - aug 0.031 4.640 457 0.058 3.604 275 j u l - sept <0.0005 20.613 595 O . 0 0 0 5 15.131 376 j u l - nov 0.001 10.897 595 0.881 0.022 376 j u l - feb 0.280 1.167 595 O . 0 0 0 5 12.318 376 j u l - apr 0.185 1.758 502 0.003 8.642 335 j u l - j u n <0.0005 14.092 595 0.601 0.274 376 j u l - aug <0.0005 36.807 580 0.001 10.391 376 sept - nov 0.326 0.963 614 <0.0005 16.663 496 sept - feb O . 0 0 0 5 13.330 617 0.925 0.009 423 sept - apr 0.002 9.544 502 0.983 O . 0 0 0 5 335 sept - j u n 0.310 1.029 617 0.001 10.102 448 sept - aug 0.095 2.780 580 0.559 0.341 499 nov - feb 0.040 4.226 614 O . 0 0 0 5 14.252 423 nov - apr 0.063 3.459 502 0.001 11.457 335 nov - j un 0.878 0.024 614 0.358 0.845 448 nov - aug 0.007 7.395 580 <0.0005 13.272 496 feb - apr 0.886 0.021 502 0.998 O . 0 0 0 5 335 feb - jun 0.002 9.278 633 0.006 7.596 423 feb - aug <0.0005 27.958 580 0.776 0.081 423 a p r - j u n 0.015 5.891 502 0.016 5.852 335 apr - aug <0.0005 22.426 502 0.738 0.112 335 jun - aug 0.007 7.251 580 0.007 7.238 448 161 Does ghost shrimp length differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003 Zone comparison S i t e l Site 2 p value t n p value x1 n 1 to 2 0.075 3.162 230 0.003 9.093 340 1 to 3 0.089 2.894 230 0.015 5.943 340 1 to 4 0.173 1.858 230 0.070 3.273 340 1 to 5 0.407 0.687 230 O . 0 0 0 5 145.495 197 1 to 6 <0.0005 25.560 230 <0.0005 112.190 90 2 to 3 <0.0005 24.890 640 <0.0005 39.772 710 2 to 4 <0.0005 34.477 640 <0.0005 42.550 710 2 to 5 <0.0005 32.049 640 <0.0005 81.504 197 2 to 6 0.015 5.946 497 <0.0005 75.765 90 3 to 4 0.045 4.021 1009 0.187 1.744 902 3 to 5 0.007 7.187 1009 <0.0005 143.235 197 3 to 6 <0.0005 28.443 497 <0.0005 103.866 90 4 to 5 0.182 1.785 1098 <0.0005 156.628 197 4 to 6 <0.0005 38.580 497 <0.0005 110.606 90 5 to 6 O . 0 0 0 5 47.132 497 0.002 9.359 90 162 Does JUVENILE ghost shrimp density change significantly over time? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002; n = 84 Date comparison Site 1 Site 2 p value I1 p value x2 may - j u l <0.0005 15.633 <0.0005 13.627 may - sept 0.021 5.360 <0.0005 28.162 may - nov O . 0 0 0 5 16.603 <0.0005 38.649 may - feb O . 0 0 0 5 24.237 0.001 10.294 may - apr 0.041 4.169 0.013 6.107 may - jun 0.002 9.630 <0.0005 26.407 may - aug 0.008 7.005 <0.0005 25.223 j u l - sept 0.059 3.569 0.162 1.952 j u l - nov 0.938 0.006 0.011 6.456 j u l - feb 0.277 1.183 0.763 0.091 j u l - apr 0.087 2.923 0.160 1.979 j u l - j un 0.352 0.867 0.077 3.118 j u l - aug 0.312 1.021 0.212 1.559 sept - nov 0.031 4.653 0.143 2.150 sept - feb 0.001 11.242 0.208 1.585 sept - apr 0.852 0.035 0.002 9.510 sept-jun 0.245 1.351 0.479 0.500 sept - aug 0.436 0.607 0.935 0.007 nov - feb 0.134 2.245 0.014 6.013 nov - apr 0.146 2.118 <0.0005 17.118 nov-jun 0.311 1.027 0.727 0.122 nov-aug 0.341 0.907 0.242 1.370 feb - apr 0.008 7.020 0.288 1.127 feb - j un 0.026 4.932 0.041 4.195 feb - aug 0.035 4.447 0.117 2.460 a p r - j u n 0.537 0.382 0.001 10.652 apr - aug 0.534 0.388 0.004 8.300 jun - aug 0.962 0.002 0.628 0.234 Does A D U L T ghost shrimp density change significantly over time? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002; n = 84 Date comparison Site 1 Site 2 p value I1 p value I1 may - j u l 0.403 0.698 0.864 0.029 may - sept 0.018 5.615 0.023 5.176 may - nov 0.011 6.401 0.942 0.005 may - feb 0.417 0.659 0.752 0.100 may - apr 0.089 2.900 0.652 0.204 may - jun 0.058 3.593 0.773 0.083 may - aug 0.018 5.599 0.075 3.179 ju l - sept 0.097 2.747 0.037 4.345 j u l - nov 0.030 4.689 0.913 0.012 j u l - feb 0.027 4.892 0.909 0.013 j u l - apr 0.006 7.554 0.488 0.481 ju l - jun 0.190 1.715 0.885 0.021 j u l - aug 0.072 3.235 0.159 1.987 sept - nov 0.531 0.393 0.042 4.153 sept - feb <0.0005 13.532 0.015 5.863 sept - apr <0.0005 16.746 0.005 7.727 sept - j un 0.776 0.081 0.039 4.244 sept - aug 0.864 0.029 0.464 0.536 nov - feb <0.0005 12.565 0.959 0.003 nov - apr <0.0005 16.602 0.557 0.345 nov - j un 0.319 0.994 0.911 0.013 nov - aug 0.629 0.233 0.131 2.285 feb - apr 0.315 1.008 0.523 0.408 feb - j un 0.003 9.020 0.792 0.069 feb - aug <0.0005 13.671 0.071 3.258 a p r - j u n <0.0005 13.489 0.399 0.713 apr - aug <0.0005 16.901 0.029 4.759 jun - aug 0.556 0.346 0.174 1.852 Does JUVENILE ghost shrimp density differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003; n = 112 Zone comparison Site 1 Site 2 p value I1 p value 1 to 2 <0.0005 82.124 O . 0 0 0 5 72.766 1 to 3 <0.0005 107.580 O . 0 0 0 5 63.401 1 to 4 <0.0005 128.941 <0.0005 38.141 1 to 5 <0.0005 108.165 0.104 2.641 1 to 6 <0.0005 39.193 <0.0005 26.714 2 to 3 0.001 10.112 0.025 5.038 2 to 4 O . 0 0 0 5 24.812 <0.0005 16.688 2 to 5 <0.0005 13.950 <0.0005 70.044 2 to 6 0.206 1.598 <0.0005 116.473 3 to 4 0.064 3.433 0.020 5.440 3 to 5 0.479 0.502 O . 0 0 0 5 59.860 3 to 6 0.001 11.719 <0.0005 115.328 4 to 5 0.306 1.048 O . 0 0 0 5 43.117 4 to 6 <0.0005 21.875 <0.0005 98.281 5 to 6 <0.0005 13.546 0.003 8.740 Does A D U L T ghost shrimp density differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003; n = 112 Zone comparison Site 1 Site 2 p value x2 p value x1 1 to 2 <0.0005 25.379 <0.0005 20.157 1 to 3 <0.0005 80.500 O . 0 0 0 5 74.243 1 to 4 <0.0005 96.502 0.011 6.515 1 to 5 <0.0005 56.433 O . 0 0 0 5 160.395 1 to 6 <0.0005 16.838 <0.0005 163.015 2 to 3 <0.0005 43.287 <0.0005 20.686 2 to 4 <0.0005 61.644 0.546 0.365 2 to 5 <0.0005 23.977 <0.0005 176.137 2 to 6 <0.0005 49.773 <0.0005 178.028 3 to 4 0.068 3.342 <0.0005 15.266 3 to 5 0.381 0.766 <0.0005 189.728 3 to 6 <0.0005 75.855 <0.0005 191.154 4 to 5 0.009 6.838 <0.0005 117.067 4 to 6 <0.0005 88.737 <0.0005 120.301 5 to 6 <0.0005 63.635 0.156 2.009 Does the proportion of females significantly exceed the proportion of males? X 2 test Expected values are 50 % males and 50 % females a = 0.05 Critical x2value = 3.84 Site Date p value x2 Site 1 M a y 0.975 0.001 July 0.975 0.002 September 0.900 0.080 November 0.900 0.029 February 0.950 0.005 A p r i l 0.900 0.032 June 0.900 0.062 August 0.900 0.025 Site 2 M a y 0.950 0.006 July 0.750 0.142 September 0.900 0.047 November 0.900 0.058 February 0.950 0.010 A p r i l 0.975 0.001 June 0.950 0.041 August 0.900 0.024 Site 1 Zone 1 0.900 0.017 Zone 2 0.900 0.052 Zone 3 0.900 0.061 Zone 4 0.950 0.008 Zone 5 0.950 0.007 Zone 6 0.950 0.006 Site 2 Zone 1 0.950 0.010 Zone 2 0.900 0.046 Zone 3 0.900 0.056 Zone 4 0.900 0.018 Zone 5 insufficient data Zone 6 insufficient data Does the proportion of female ghost shrimp differ significantly between sampling dates? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002 Date comparison Site 1 Site 2 p value I1 n p value x1 n may - j u l 0.776 0.081 52 0.009 6.803 47 may - sept 0.037 4.353 52 0.168 1.898 48 may - nov 0.335 0.930 52 0.200 1.644 46 may - feb 0.894 0.018 52 0.610 0.260 48 may - apr 0.219 1.508 52 0.704 0.145 43 may - jun 0.071 3.252 52 0.191 1.707 48 may - aug 0.504 0.446 52 0.465 0.534 48 j u l - sept 0.060 3.547 57 0.063 3.450 47 j u l - nov 0.557 0.345 57 0.165 1.928 46 j u l - feb 0.684 0.166 57 0.020 5.369 47 j u l - apr 0.299 1.079 55 0.006 7.574 43 j u l - j u n 0.114 2.492 57 0.148 2.096 47 j u l - aug 0.946 0.005 57 0.030 4.700 47 sept - nov 0.219 1.513 56 0.702 0.146 46 sept - feb 0.044 4.069 56 0.496 0.463 53 sept - apr 0.506 0.442 55 0.094 2.807 43 sept - j u n 0.986 O . 0 0 0 5 56 0.715 0.134 49 sept - aug 0.155 2.024 56 0.704 0.144 55 nov - feb 0.145 2.123 52 0.274 1.197 46 nov - apr 0.854 0.034 52 0.107 2.605 43 nov - j un 0.764 0.090 52 0.958 0.003 46 nov - aug 0.329 0.953 52 0.417 0.659 46 feb - apr 0.210 1.570 51 0.449 0.573 43 feb - j un 0.118 2.441 51 0.316 1.007 49 feb - aug 0.805 0.061 51 0.711 0.137 53 a p r - j u n 0.721 0.128 49 0.120 2.416 43 apr - aug 0.295 1.097 49 0.290 1.118 43 jun - aug 0.299 1.078 54 0.471 0.520 49 167 Does the proportion of female ghost shrimp differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003 Zone comparison Site 1 Site 2 p value 3? n p value xz n 1 to 2 0.452 0.564 85 0.433 0.614 98 1 to 3 0.604 0.269 85 0.468 0.527 98 1 to 4 0.232 1.430 85 0.788 0.072 82 1 to 5 0.227 1.463 85 insufficient data 1 to 6 0.548 0.361 41 insufficient data 2 to 3 0.724 0.124 111 0.858 0.032 105 2 to 4 0.007 7.250 112 0.225 1.471 82 2 to 5 0.007 7.319 105 insufficient data 2 to 6 0.109 2.570 41 insufficient data 3 to 4 0.002 9.290 111 0.064 3.439 82 3 to 5 0.002 9.417 105 insufficient data 3 to 6 0.076 3.145 41 insufficient data 4 to 5 0.848 0.037 105 insufficient data 4 to 6 0.959 0.003 41 insufficient data 5 to 6 0.824 0.050 41 insufficient data 168 6.3 Ghost shrimp reproduction (refer to section 4.3.3) Does the proportion of reproductive females vary significantly over time? Mann-Whitney pairwise analysis of variance 8 dates; 28 pairs; a = 0.002 Date comparison Site 1 Site 2 p value x2 n p value x2 n may - j u l <0.0005 44.697 56 <0.0005 40.808 41 may - sept <0.0005 97.008 56 <0.0005 62.575 41 may - nov <0.0005 95.313 56 <0.0005 62.575 41 may - feb O . 0 0 0 5 95.330 56 <0.0005 56.209 41 may - apr O . 0 0 0 5 53.955 52 <0.0005 30.237 37 may - j un <0.0005 29.226 56 <0.0005 32.799 41 may - aug <0.0005 95.417 56 <0.0005 63.473 41 j u l - sept <0.0005 21.535 68 0.005 7.757 44 j u l - nov <0.0005 24.760 68 0.005 7.757 44 j u l - feb <0.0005 25.090 68 0.009 6.779 44 j u l - apr 0.470 0.521 52 0.152 2.053 37 j u l - jun 0.009 6.898 68 0.039 4.250 44 j u l - aug <0.0005 22.843 68 0.005 7.896 44 sept - nov 0.176 1.828 69 1.000 O.0005 54 sept - feb 0.173 1.855 70 1.000 O.0005 47 sept - apr <0.0005 18.227 52 <0.0005 16.146 37 sept - j u n <0.0005 51.227 76 <0.0005 21.923 46 sept - aug 0.590 0.290 73 1.000 O.0005 54 nov - feb 1.000 <0.0005 69 1.000 <0.0005 47 nov - apr <0.0005 22.403 52 O . 0 0 0 5 16.146 37 nov - jun <0.0005 52.937 69 O . 0 0 0 5 21.923 46 nov - aug 0.331 0.945 69 1.000 O.0005 54 feb - apr O . 0 0 0 5 22.705 52 <0.0005 14.175 37 feb - j un <0.0005 53.587 70 <0.0005 19.312 46 feb - aug 0.327 0.959 70 1.000 O.0005 47 a p r - j u n 0.001 12.010 52 0.562 0.336 37 apr - aug <0.0005 19.826 52 <0.0005 16.427 37 j un - aug O . 0 0 0 5 51.681 73 <0.0005 22.294 46 169 Does the proportion of reproductive females differ significantly between zones? Mann-Whitney pairwise analysis of variance 6 zones; 15 pairs; a = 0.003 Zone comparison Site 1 Site 2 p value x2 n p value xl n 1 to 2 0.534 0.386 30 0.283 1.153 35 1 to 3 0.737 0.113 30 0.154 2.033 35 1 to 4 0.335 0.929 30 0.001 11.877 33 1 to 5 0.044 4.058 30 insufficient data 1 to 6 0.281 1.160 12 insufficient data 2 to 3 0.709 0.139 52 0.755 0.098 45 2 to 4 0.048 3.921 52 0.008 7.098 33 2 to 5 0.001 10.462 47 insufficient data 2 to 6 0.445 0.583 12 insufficient data 3 to 4 0.042 4.153 55 0.010 6.693 33 3 to 5 0.001 11.588 47 insufficient data 3 to 6 0.250 1.323 12 insufficient data ' 4 to 5 0.094 2.807 47 insufficient data 4 to 6 0.037 4.331 12 insufficient data 5 to 6 0.005 7.978 12 insufficient data 6.4 Ghost shrimp and other infauna (refer to section 4.3.4) Does Cryptomya californica density regress significantly on Neotrypaea californiensis density? Regression analysis Linear equation r 2 p value A d u l t Neotrypaea only: June y = 0.154x + 40.3 0.57 < 0.0005 August y = 0.213x + 217.8 0.49 <0.0005 A H Neotrypaea: June y = 0.126x+ 114.8 0.40 <0.0005 August y = 0.190x +259.8 0.49 <0.0005 170 6.5 Burrow counts (refer to section 4.3.5) 2 3 Does the number of burrow holes per m predict the number of ghost shrimp per m or the dry weight of ghost shrimp per m 3 ? At each sampling station, a mean burrow density and a mean ghost shrimp density and biomass values were estimated. A regression analysis of these means was performed. B u r r o w density (holes /m 2 ) as correlated with: Month F ratio p value r2 Linear equation y = ghost shrimp variable x = burrow density n Ghost shr imp density ( ind/m 3 ) July 01 16.296 0.016 0.803 y= 1.80x- 127 6 Sept 01 9.863 0.005 0.342 y = 0.75x+ 1070 21 Nov 01 29.038 <0.0005 0.617 y= 1.93x + 605 20 Apr 02 3.730 0.095 0.348 y = 0.94x + 518 9 June 02 4.608 0.085 0.480 y = 0.90x4 816 7 All dates 30.184 0.002 0.139 y = 0.68x4- 1176 65 Summer 19.014 0.001 0.594 y= 1.27x4-464 15 Ghost s h r i m p biomass (g dw/ m 3 ) July 01 6.832 0.059 0.631 y = 0.18x-59.2 6 Sept 01 7.829 0.011 0.292 y = 0.09x 4- 27.8 21 Nov 01 37.348 <0.0005 0.675 y = 0.19x- 13.4 20 Apr 02 5.726 0.048 0.450 y = 0.09x4-2.3 9 June 02 57.983 0.001 0.921 y = 0.16x-35.2 7 All dates 35.863 <0.0005 0.363 y = 0.11x4- 18.1 65 Summer 11.896 0.004 0.478 y = 0.17x-24.5 15 Does the number of burrow holes per m 2 predict the number of ghost shrimp per m 3 or the dry weight of ghost shrimp per m ? Burrow density was estimated for each core taken and a regression analysis was performed on these values. B u r r o w density (holes /m 2 ) as correlated with: Month F ratio p value r2 Linear equation n Ghost shr imp density ( ind/m 3 ) June 02 10.568 0.006 0.430 y = 0.76x + 430 16 Aug 02 6.404 0.016 0.144 y = 0.62x 4- 967 40 Ghost s h r i m p biomass (g d w / m 3 ) June 02 4.838 0.045 0.257 y = 0.13x-30 16 Aug 02 0.226 0.637 0.006 y = -0.03x + 226 40 171 6.6 Mudflat parameters (refer to section 4.3.6) Does the sediment grain size predict the density of biomass of ghost shrimp in Grice Bay? Regression Analysis: mean sediment grain size per core regressed upon ghost shrimp density or biomass per core. Sediment grain size (mm) as correlated with: F ratio p value r 2 n Linear equation y = ghost shrimp variable x = sediment grain size Ghost shrimp density (ind/m 3) 5.681 0.018 0.029 192 y = 8216x + 221 Ghost shrimp biomass (g dw/m 3 ) 17.242 <0.0005 0.083 192 y = 1194x-91 Does the depth of the grain size predict the density of biomass of ghost shrimp in Grice Bay? Regression Analysis: The depth of each core was regressed upon the density or biomass of ghost shrimp in that core. B u r r o w depth (m) as correlated with: F ratio p value r 2 n Linear equation y = ghost shrimp variable x = sediment grain size Ghost shrimp density (ind/m 3) 278.64 O . 0 0 0 5 0.173 1334 y = 3363x + 364 Ghost shrimp biomass (g dw/m 3 ) 521.19 <0.0005 0.281 1334 y = 370x - 34 Does the mudflat elevation predict the density of biomass of ghost shrimp in Grice Bay? Mean mudflat elevation based on zone, was regressed upon mean ghost shrimp density and biomass. The ghost shrimp density distribution as a function of mudflat elevation increased with depth until 1.2 m above M L L W and then decreased. Therefore, there are two linear functions to describe ghost shrimp density. Ghost shrimp biomass followed a linear function from the upper limit of the ghost shrimp distribution to 0.8 m above M L L W . M u d f l a t elevation (m) F ratio p value r 2 n Linear equation y = ghost shrimp variable x = water depth as correlated with: Ghost shr imp density ( ind/m 3 ) shallow (> 1.2m) 93.02 <0.0005 0.520 88 y = -1629x + 4442 deep (< 1.2m) 139.84 O . 0 0 0 5 0.578 104 y = 1627x4-350 Ghost s h r i m p biomass (g dw/m 3 ) shallow ( > 0.8m) 487.93 < 0.0005 0.801 123 y = - 1 0 5 x 4 257 172 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0091050/manifest

Comment

Related Items