Benthic Autotrophy in Netarts Bay, Oregon

Home , Netarts Bay

NETARTS BA) OREGON

Final TechnicalReport to the Environ mental Protection Agency

BENTHIC AUTOTROPHY IN NETARTS BAY, OREGON

E ST U AR E In Reference to BIOLOGICAL Jiiiii( / EPA Research P ROCESSES \PRocE Grant No. R806780

(ACRO-

OregonState University

May I, 1983 7 ZOSTERA ALGAL / ss xposil PRIMARY PRIMARY \ 'ROD PROD1T

O(TRITAL) PRETIO

BENThIC AUTOTROPHY IN NETARTS BAY, OREGON

Final Report Submitted June 1, 1983, to the

Environmental Protection Agency 200 S.W. 35th Street Corvallis, OR97331

Prepared by

C. David Mclntire, Professor

and

Michael W. Davis, Mary E. Kentula, and Mark Whiting Research Assistants

Department of Botany and Plant Pathology Oregon State University Corvallis, OR 97331

in reference to

EPA Research Grant No. R806780 entitled:

"Relationships Between Nutrient Fluxes and Benthic Plant Processes in Netarts Bay, Oregon TABLE OF CONTENTS

Introduction

Background ...... Conceptual Framework for the Research Ecosystem Processes and Process Capacity ATentative Model of Estuarine Processes Netarts Bay ...... EPARhodanuLneStudiesofl978andl979 ...... Some Relevant Literature ...... Benthic Microflora ......

Zostera and Macroalgae ...... Some Relevant Laboratory Studies ...... Columbia River Project ......

TheAlgaiPrimaryProductionSubsystetu...... Production Dynamics of Sediment-Associated Algae in

Netarts Bay, Oregon ......

Sampling Strategy ......

Methods ...... , ...... , ...... Primary Production ......

Biomass . . . . . a * ...... a a . . Sediment Properties ...... Physical and Chemical Variables

Data Analysis ......

Results ......

Physical Properties ...... * ......

Organic Matter ...... a a . a ...... a . a a . a a

Chlorophyll a Concentration ...... a a a ......

Primary Production ...... a . a . . a aa a . a a Relationships Among Variables ...... Interpretation of AutotrophicPatterns ...... Experimental Studies of Estuarine Benthic Algae

Yaquina Bay . . . a a ...... a a a a . . . . . a a ...... Methods ......

Primary Production ...... a a a .a ...... a a a

Biomass . a a a a ...... a a a a . . . a a . . a a a . . a a a ...... a . Isolation of DiatomAssemblages

Physical Variables * ...... a . * a a a . a a a a. a a . . . a a a Results ...... a.... .aaaa.. .a.a..... I.. Experiments withlntact Sediment Cores Experiments with Isolated Epipelic Diatoms ...... Experiments with Isolated Macroalgae ...... Interpretation of Experiments ...... a . a ...... a a ...... Effects of Estuarine Infauna on Sediment-Associated Microalgae Study Site ...... Methods ...... -3-

-Experimental Design . Metabolic Activity ...... Chlorophyll a Analysis ...... Macrofaunal Abundance ...... Statistical Analysis ...... Results ...... Field Defaunation Experiment ...... Laboratory Defaunation Experiment ...... Interpretation of Defaunation Experiments ...... The Diatom Flora of Netarts Bay ...... Sampling ...... Methods ...... Data Analysis ...... Results ...... StructureofEnvironmentalflata ...... The Diatom Flora ...... AutecologyofSelectedTaxa ...... Community Patterns ......

The Zostera Primary Production Subsystem ...... Materials and Methods ...... DescriptionoflntensiveStudyArea ...... SelectionofQuadratandSampleSize...... Measurement of Biomass ...... MeasurementofPrimaryProduction ...... Short Marking Methbd ...... RadjoactiveCarbon(14C) Method...... Data Analysis ...... Morphometrics and Autecology of Zostera marina L. in Netarts Bay Sexual Reproduction ...... Vegetative Growth ...... ProductionDynamicsofZostera ...... Biomass ...... Primary Production ...... Relationship Between Zostera and Epiphytic Assemblages ...... Description of EpiphyticAssemblages ...... Biomass ...... Production ...... Components of Variance ...... Bioenergetics of the Zostera Primary Production Subsystem ...... Discussion ......

References ......

Appendixl...... INTRODUCTION

This final report was prepared primarily for the information of persons having the responsibility of judging the merit of research supported by a grant from the Environmental Protection Agency (EPA No. R806780). The

research program was initiated September 1, 1979 and the original project and

budget periods extended from this date to August 30, 1981.Extensions of the

project and budget periods were requested in May, 1981 and again in January,

1982. These extensions were granted, and the termination date for the project and budget periods was reset to August 30, 1982. Although scientists and

other individuals interested in the production dynamics of benthic plants in estuaries may find this report an informative summary of our work during the

prOject period, it is not intended to be a definitive publication and should not be evaluated as such. Instead, five manuscripts representing segments of the research presented in this report are in preparation for submission to refereed professional journals. The primary purpose of the final report is to integrate the various aspects of the research and to provide an outlet for the data that can be made available to other scientists.

The general objective of research supported by EPA Grant No. R806780 was to determine mechanisms that control the production dynamics of benthic plants and associated epiphytes in Netarts Bay relative to physical processes and changing patterns of nutrient distribution. In particular, we were especially concerned with biological processes that were related to the patterns of nutrient flux determined by the EPA dye studies of August, 1978 and January,

1979. These studies involved the introduction of Rhodamine dye and subsequent -5-

comparisons of the dye concentration with concentrations of substances of interest at selected stations in the estuary. Nutrients of interest included dissolved organic carbon (DOC), particulate organic carbon (POC), nitrate- nitrate (NO2-NO3), orthophosphorous (Ortho-P), ammonia nitrogen (NH3), and molybdate reactive silica (Si); in addition, temperature, salinity, turbidity, fluorescence at 460 nm and 660 nm, and Rhodamine concentrations also were monitored, The results of the EPA investigation suggested that nutrient dynamics were strongly related to biological processes associated with large areas of seagrass (Zostera marina) during August; but in January, corresponding patterns were more closely related to physical processes. Therefore, our general objective stated above was compatible with the EPA's research program at Netarts Bay and provided the basis for a detailed examination of benthic plant processes in the estuary.

For research purposes, it was convenient to partition our work on benthic plant processes in Netarts Bay into two subsystems: Zostera Primary Produc- tion and Algal Primary Production. These subsystems are part of a tentative conceptual model of the entire estuarine ecosystem, the details of which are described in the next section of this report. The research approach involved an intensive field investigation of the two subsystems during a two-year period and some concurrent experimental work on the Algal Primary Production subsystem. The sampling strategy in the field was carefully designed to take advantage of the existing knowledge of nutrient distribution and circulation in the estuary.

This report is organized into three major subsections: Background, The

Algal Primary Production Subsystem, and The Zostera Primary Production

Subsystem. In the Background section, we present a conceptual framework for the research, briefly describe the estuary, review the EPA dye studies of 1978

and 1979, and present a review of relevant literature. The Algal Primary

Production Subsystem section is concerned with the production dynamics of

benthic micro and macroalgae relative to sediment properties and other

physical and biological variables. In this section, we also present the

results of laboratory studies designed to establish functional relationships

between algal primary production and selected physical and biological

factors. The Zostera Primary Production Subsystem section presents the

ecological energetics of a seagrass population and associated epiphytic and

benthic microalgae. Also, information concerning plant morphometrics and

subterranean detritus is included in this section.

BACKGROUND

I. Conceptual Framework for the Research

Ecosystem Processes and Process Capacily

The ecological literature often refers to various physical and biological processes in contexts that are usually intuitively understandable without a formal, theoretically structure. However, for analytical purposes, it is desirable to formalize the process concept by definitions and the establish ment of a system for diagramming relationships. Here, the definitions apply only to biological processes, and physical processes are treated conceptually as driving or control functions.

Definition: A process is a systematic series of actions relevant to the

dynamics of the system as it is conceptualized (or modelled). -7-

The process concept is made more explicit by the diagram illustrated in

Figure 1. This structure is compatible with both autotrophic and heterotrophic processes (e.g., algal primary production and deposit feeding). In the figure, a process, the entire contents of the circle is elaborated into one state variable, the process capacity, and six variables representing inputs and outputs associated with the process. Theoretically, process capacity includes two components, a quantitative aspect which is the biomass at any instant of time involved in the process, and a qualitative aspect which relates to taxononiic composition and physiological state. Other variables associated with the process include resource input, cost of processing (C), waste discharge (W), losses to the process of decomposition (B), losses to the process of predation or grazing (P or C), and export (E). In the diagram, decomposition, predation, and export act directly on the process biomass and are represented by dashed arrows from the state variable box. In contrast, resource input, cost of processing, and waste discharge are associated with the process as a whole and are, therefore, represented by solid arrows connected to the perimeter of the circle.

At this point, the concept of process capacity needs further clarifi- cation. Relative to a given set of inputs, the potential performance of a unit of biomass associated with a process such as primary production (or deposit feeding) may change with temporal and spatial changes in the taxonomic composition and physiological state of that biomass. Or, stated from a population perspective, one combination of populations or parts of populations involved in a particular process may do things at different rates, i.e., may require a different parameterization than a different combination involved in the same process. Theoretically, process capacity expresses both quantitative RESOURCE -P INPUT

C w

COMPONENTS OF PROCESSCAPACITY: 1. QUANTITATIVE: biomass 2. QUALITATIVE: genetic information OTHER VARIABLES: C is COST OF PROCESSING W is WASTE DISCHARGE B is LOSS TO DECOMPOSITION P is LOSS TO PREDATION E is EXPORT

Figure 1. Schematic representation of the process concept illustrating a state variable,, the process capacity, andrelevant input and output variables. and qualitative aspects of the process state and, therefore, represents a unit that is time and spatially invariant relative to its relationships with other components of the system. The problem is to develop a set of rules that will map process biomass, a variable that changes qualitatively, into process capacity, a corresponding variable that maintains a constant functional potential. As yet, we have not invented an approach to the direct estimation of process capacity, and most current models represent only the quantitative aspects (biomass) of process capacity. In this report, we deal with qualitative differences by partitioning the biomass associated with the process of

primary production into several state variables (e.g., micro-algae, species of macroalgae, and seagrass).

The process viewpoint emphasizes the capacity of a system or subsystem to

process inputs and the mechanisms that control and regulate that capacity.

Ideally, parameter estimation is based on direct measurement of processes in

the field or laboratory rather than on the summation of activities of individual species populations. Moreover, the process approach is particularly

suitable for the investigation of benthic autotrophy in estuaries, as field

and laboratory methods are available to monitor primary production in complex

assemblages of mlcroalgae, without having to measure photosynthetic rates of

the many constituent taxa.

A Tentative Model of Estuarine Processes

A tentative conceptual model of an estuarine ecosystem is illustrated in

Figures 2-5. This preliminary model provided the structure that was necessary

to design a research strategy consistent with our project objectives. Since

ecosystem modelling is an iterative process, this structure undoubtedly will

change as more field and laboratory studies generate new information. -10-

Estuarine biological processes can be considered holistically in terms of

inputs and outputs relative to the entire ecosystem (Fig. 2). Also, the eco-

system can be investigated mechanistically, in this case as a system of three

coupled subsystems that can be uncoupled and investigated holistically or mechanistically in isolation after coupling variables have been carefully

identifed. This conceptualization of ecosystem processes is based on the FLEX modelling paradigm developed by W. S. Overton (1972, 1975, 1979) and related

to the general systems theory of Klir (1969). For our purposes, it is conven-

ient to partition estuarine biological processes into Water Column Processes,

Emergent (or Marsh) Plant Processes, and Mudflat (or Benthic) Processes. This report is primarily concerned with the dynamics of Mudflat Processes, and in

particular with the Primary Food Processes subsystem of Mudflat Processes.

A hierarchical arrangement of the subsystems of Mudflat Processes is illustrated in Figures 3 and 4. This conceptual model partitions Mudflat

Processes into two coupled subsystems: Primary Food Processes and Macro- consumption. Since our research was concerned with benthic autotrophy associated with estuarine tidal flats, the Primary Food Processes subsystem was the subsystem of interest and was partitioned conceptually into some of its subsystems. On the other hand, Macroconsumption includes such subsystems as Suspension Feeding, Deposit Feeding, and Predation (Fig. 2), processes not elaborated in Figures 3 and 4. Primary Food Processes has three subsystems:

Zostera Primary Production, Algal Primary Production, and Detrital Decomposi- tion. Zostera Primary Production consists of Macrophyte Primary Production- and Epiphyte Primary Production (Fig. 3), two subsystemsinvestigated at an intensive study site in Netarts Bay. Some variables associated with the Algal -11-

EMERG8 PLANT

ESTUARINE WATER- BIOLOGICAL COLUMN PROCESSES MUD FLAT PROC ESSES

DEPOSIT ZOST ERA ALGAL SUSPENSIcN\ PRIMAR'' PRIMAR'( FEEDING ) FEEDING

DETRITAL PR EDAT ION

Figure 2. Systems diagram of Estuarine BiologicalProcesses and associated subsystems. -12-

MUDFLAT PROCESS ES

PFP: PRIMARY FOOD PROCESSES MPP= MACROPHYTE MC = MACROCONSUMPTION PRIMARY PRODUCTION APP= ALGAL PRIMARY PRODUCTION ZPP ZOSTERA PRIMARY PRODUCTION EPP= EPIPHYTE DD = DETRITAL DECOMPOSITION PRODUCTION

Figure 3. Systems diagram of Mudflat Processes indicating details of the Zostera Primary Production subsystem. -13-

Primary Production subsystem are diagramed in Figures 4 and 5. In Figure 4, primary production is controlled directly by nutrient input, light energy input, respiratory losses, and a complex of other physical processes that have effects on photosynthesis and respiration. In addition, losses or gains of algal biomass (the process state variable) by the mechanisms of animal consumption, export, transfer to detrital processes, or imports also affect process dynamics.

As mentioned earlier, capacity for algal primary production is affected by both the biomass and the qualitative aspects of that biomass. In the research at Netarts Bay, the qualitative aspects were considered by expressing algal biomass as several state variables. For example, preliminary research indicated that three functional groups of algae are important primary pro- ducers on the tidal flats of the bay. Figure 5 represents the process of algal primary production in more detail. In this case, algal biomass is partitioned into three state variables: microalgae (the diatom flora and less prominant inicroalgae), the Enteromorpha Ulva flora, and the less abundant red and brown macroalgae (e.g., Gracilaria verrucosa). Process rates associated with the Enteromorpha Ulva are considerably higher than rates found for other functional groups, but this group is only conspicuous during several months in the summer. Therefore, the sampling strategy and experimental pro- cedures for investigating algal primary production were adjusted Co accommodate temporal, qualitative changes in the process state variable.

The structure illustrated in Figure 5 served as the conceputal basis for the research described in the Algal Primary Production subsystem section.

This diagram illustrates the variables that must be examined in order to -14-

MUDFLAT PFP MC PROCESSES

ZPP APP E

BIOMASS --o--- PF 0 L

R PF PHYSICAL FACTORS L LIGHT PFP = PRIMARY FOOD PROCESSES N NUTRIENTS

MC= MACROCONSUMPTION - Ra RESPIRATION DD= DETRITAL DECOMPOSITION C CONSUMPTION ZPP ZOSTERA PRIMARY PRODUCTION 1= IMPORT APP=ALGAL PRIMARY PRODUCTION E EXPORT

Figure 4. Systems diagram of Mudflat Processes indicating details of the Algal Primary Production subsystem. -15-

ALGAL PRIMARY PRODUCTION

LIGHT ENERGY

MICROALGAE

PHYSICAL FACTORS ESPIRATION ENTEROMORPHA GRACI LARIA ULVA

NUTRI ENTS

EXPORT'' GRAZING

DETRITAL DECOMPOSITION

Figure 5. The Algal Primary Production subsystem and associated state variables with relevant inputs and outputs. -16-

understand the dynamics of this subsystem within the coupling structure of the

entire estuarine ecosystem. A similar approach was taken for the examination

of the Zostera Primary Production subsystem. In summary, Algal Primary Pro-

duction and Zostera Primary Production were the target subsystems for the

research presented in this report; and the results of this research were

interpreted relative to the EPA dye studies of 1978 and 1979 and within a

coupling structure for the entire ecosystem.

II. Netarts Bay

Netarts Bay is the sixth largest estuary in Oregon with a surface area of

941 hectares of which 612 hectares are classified as tideland area (Fig. 6).

The permanently submerged land is small in area (ca. 329 hectares) and is

restricted to a relatively narrow channel extending from the mouth of estuary

to the mudflat at the south end. Netarts Bay drains an area of only 36.3 km2

and is partially exposed to waves at the throat. The depression which the

estuary now occupies was formed as a consequence of the differential erosion

of the soft sedimentary rock (the Astoria Formation) between the basalt head-

lands of Cape Meares and Cape Lookout during the Late Tertiary Period. The

sand spit which separates the bay from the Pacific Ocean is a remnant of three

sand dunes which have eroded as a consequence of sea level elevation after the

last period of glaciation. Because of the lack of any major tributaries, the

estuary exhibits a relatively high salinity with values usually above 30 0/00

and seldom below 14 0/00 at any location. Although Burt and McAllister (1959)

reported that the bay remained vertically well-mixed throughout the year, the

recent dye studies by the EPA indicate that horizontal mixing is more complex -17-

NETARTS BA)' OREGON

Figure 6. Nap of Netarts Bay, indicating thelocation of the channel at low tide. -18-

with only limited mixing occurring between ocean water and bay water during

each tidal cycle. Sediments introduced into Netarts Bay are estimated to

average only 2,250 tons annually.

III. EPA Rhodamine Studies of 1978 and 1979

Field studies of nutrient fluxes in Netarts Bay by the EPA and associated

scientists were performed in August, 1978 and January, 1979. Rhodamine dye

was introduced at a station in the southern part of the estuary at low tide,

and its distribution and concentration was monitored to examine patterns of

circulation and horizontal mixing in the system. In general, it was found

that at low tide the water retained in the channel the bay water -- was the

same water that was over the seagrass and mudflat areas during high tide. As

the tidal cycle moves ocean water into the estuary, the bay water is pushed

out of the channel and over the seagrass areas and tidal flats.Also, it was

found that relatively little mixing takes place between incoming ocean water

and the resident bay water, as detectable amounts of Rhodamine were found at

low tide ten days after its introduction.

Comparisons of nutrient concentrations at many sampling stations in the

channel and over the seagrass and tidal flats with Rhodamine concentrations

provided a basis for the examination of nutrient fluxes transported by

advection, within the water column, and fluxes between the water column and

the Zostera Primary Production and Algal Primary Production subsystems.

Furthermore, this kind of analysis provides a holistic view of the dynamics of

selected substances that can serve as a basis for the investigation of mech-

anisms at finer levels of resolution. -19-

The results of the August study strongly implicated the seagrass and associated organisms as biological components accounting for significant deviations of patterns in the concentrations of Si, NO2-NO3, and DOC from the pattern exhibited by the Rhodamine. During this time of year, there is evidence that there is a net flux of Si and NO2-NO3 from the water column to the

Zostera Primary Production subsystem and that this net uptake is operating on a time resolution greater than one tidal cycle but less than the time resolu- tions associated with advection or water column processes. Apparently there is a rapid net flux of ammonia from the Zostera Primary production and Algal

Primary Production subsystems to the water column on a time resolution of less than one tidal cycle. However, this substance is more patchy in distribution than Si or NO2-NO3, indicating detectable changes within the water column.

DOC exhibited a net flux from the bottom to the water column and was apparently exchanged on a time resolution similar to that of Si and NO2-NO3. The spatial and temporal pattern of temperature during the summer study followed the pattern of Rhodamine and clearly provided a contrast between bay water and ocean water. At this time, the bay maintained a temperature near 20°C while the ocean water was approximately 9°C.

During the period between the summer and winter studies, there was a massive export of plant biomass out of the grass beds; and by January the grass consisted of roots, rhizomes, and short remnants of leaf and stem tissue. Nutrient data from the winter study indicated that patterns were controlled primarily by physical processes, and NO2-NO3 and DOC fluxes between the bottom and the water column were operating on a relatively long time resolution (perhaps several weeks). A net flux of Si to the bottom was not -20

detected, and this substance was actually introduced into the water column

from some of the small tributaries.

Interpretation of the dye and nutrient studies suggested that pronounced

nutrient gradients of biological significance develop during the growing

season and extend from the channel along circulation transects over the sea

grass and mudflats as ocean water moves back into the channel on an incoming

tide. A detailed study of vertical and horizontal nutrient profiles along

such a transect was investigated by Nancy Engst of the EPA in collaboration

with Dr. Jorg Imberger and the P1. A float study was undertaken to determine

the pattern of circulation over the seagrass, and a suitable transect for

sampling was selected from the results of these preliminary observations.

This research is scheduled for completion by late spring or summer, 1983 and

will provide a more mechanistic view of the pattern and quantity of nutrient

exchange between the benthos and the water column.

IV. Some Relevant Literature

The following review of literature represents an arbitrary selection

of references that provide a good basis for comparing the results of this

study with other research being conducted by the Principal Investigator

and associates and with the results of studies by other scientists.More

specifically, the review covers selected field studies of tidal flat

primary production and some past laboratory work on benthic autotrophy.

Benthic Microflora

Relatively little information is available on rates of primary production

and respiration in assemblages of marine and estuarine benthic microalgae. -21-

Methodology employed in the estimation of these rates usually involves the isolation of a sample in a flask, bell jar, or chamber and the subsequent monitoring of carbon-14 uptake or changes in the concentration of metabolic gases in the medium (e.g., see Bott et al. 1978; Darley et al., 1976; Hall et al., 1979; Hunding and Hargrave, 1973; Marshall etal., 1973; Mclntire and

Wuiff, 1969; Pamatmat, 1965, 1968, 1977; Pomeroy, 1959; Van Raalte etal.,

1974). Such methods attempt to estimate the productivity of the entire algal assemblage and require certain assumptions that are usually violated to some degree depending on the particular situation. The difficulty in partitioning primary production among the constituents of the benthic microflora is one of the most complex problems in aquatic ecology. Steele and Baird (1968) described a method for measuring carbon assimilation by epipsammic organisms, and the method of Hickman (1969) is designed to separate epipelic and epipsammic algae for measurements of primary production. To our knowledge, there is still no satisfactory field method for partitioning community respiration into bacterial and producer respiration in assemblages of microorganisms.

The study by Steele and Baird (1968) indicated that rates of primary production on beach sand were relatively low, in the range of 4-9 g C m2 yr'.

There was an increase in chlorophyll a and organic carbon with depth below the low-water level to 13 m, but decreases in light intensity with depth apparently accounted for a corresponding decrease in the ratio of uptake to the concentration of chlorophyll a. In any case, organic carbon content under

1 m2 of beach sand to a depth of 20 cm was about 50 g. Furthermore, viable populations of diatoms were found to a depth of 20 cm at the low-water -22-

stations. This distribution of living organisms was attributed to mixing of

the sand by wave action and stimulated speculation concerning metabolic rates

of diatoms buried below the zone of effective light penetration for extended

periods. Dye (1978) compared rates of epibenthic algal production on sand

with the rates measured on mud in a South African estuary (the Swartkops

estuary). The mean rate of primary production in estuarine sand (53 g C m2

yr) was higher than that found for shifting beach sand by Steele and Baird

but less than half of the mean rate found for the muddy areas (116.5 g C nr2 yr).

Other estimates of total primary production for assemblages of benthic

microorganisms associated with sandy silt or mixtures of silt and clay are

higher than those reported above for epipsammic assemblages alone. Gr$ntved

(1960) investigated the productivity of the microbenthos in some Danish fjords

and estimated that the average carbon fixation was 116 g C m2 yr'. He also

found that average fixation was 142 mg C m2 (2 hrY' for 87 samples from

sand (mean water depth of 0.85 m) and 139 mg C n12 (2 hrY for 42 samples

from sandy silt (mean depth of 1.11 m). After correcting the data for depth

effects, it appeared that primary production was greater when the bottom material contained silt and ciay than when the substrate was 'pure' sand.

Moreover, maximum primary production was at water depths between 0.5 and

0.7 m, and rates at one locality (Naeraa Strand) for the period from November

to February were about one-third of the rates found for the rest of the years

In another study, Gr$ntved (1962) found that photosynthetic potential was

about four times higher on the exposed tidal flats in the Danish Wadden Sea

than in the Danish fjords. More recent work in the Western Wadden Sea (Cadee -23-

and Flegeman, 1974) indicated that mean annual primary production of the microflora associated with tidal flats was about 100 g Cm2yr. Primary production was correlated with temperature, solar radiation, and functional chlorophyll; and excretion was only 1% of the annual primary production. Further studies (Cadee and Hegeman, 1977) indicated that primary production was related to the tidal level of the stations, and annual rates varied from 29 g c m2 on the lowest tidal flat station to 188 g C if2 at the highest station.

The investigation of benthic primary production in the Eems-Dollard estuary and the Eastern Dutch Wadden Sea by Colijn (1974) indicated a seasonal periodicity in bioniass and production consisting of a spring maximum and a lower maximum in the autumn; winter values were low, and summer values were intermediate between spring and winter values. Colijn and van Buurt (1975) found that the photosynthetic rate of marine benthic diatoms in the field was saturated by a light intensity of approximately 10,000 lux, and at higher intensities no photoinhibition was found. Within a range of 4° to 20°C the photosynthetic rate increased about 10% per degree C.

Gross primary production of microalgae in the intertidal marshes on the coast of Georgia was measured by the oxygen method with bell jars (Pomeroy,

1959). The annual rate of gross production was estimated to be 200 g C if2; efficiency of photosynthesis ranged from 1to 3% at light intensities less than 100 kcal m2 hr and was 0.1% or less at intensities in excess of 300 kcal m2 hr'. Pamatamat (1968) using similar methods found that the primary production on an intertidal sandflat on False Bay, San Juan Island was comparable to the Danish Wadden Sea and the salt marshes of Georgia. In this case, photosynthetic efficiency over the year averaged 0.10, 0.11, and 0.12% -24-

of total incident radiation at the three stations under investigation. Also, rates of photosynthesis exhibited an endogenous rhythm apparently related to tidal cycle; rates were relatively high during flood and ebb tide and depressed during low and high water. Marshall etal. (1971) investigated primary production of the benthic microflora of shoal estuarine environments in southern New England and concluded that an annual rate of about 100 g C m2 yr1 is representative of both the Danish and southern New England shoals.

In contrast, Leach (1970) reported a value of only 31 g C m2 yr for an estuarine intertidal mudflat in northeast Scotland and suggested that the relatively low value for this region might be related to climatic factors.

The relative contribution of microalgal assemblages to the total primary production in marine and estuarine ecosystems has been examined in various field studies. Gallagher and Daiber (1974) investigated primary production of edaphic algal communities in a Delaware salt marsh and found that the annual rate varied from 38 to 99 g C m2, depending on the local composition of the associated macrophyte flora. Gross algal production was about one-third of the angiosperm net production, and the highest rates occurred when the angiosperms were dormant. Sources of autotrophic and allochthonous organic carbon available to the Nanaimo Estuary delta, British Columbia, were investigated by

Naiman and Sibert (1979). Annually, benthic microalgae produced 4-55 g C phytoplankton about 7.5 g C m2, macroalgae 0.9-7.5 g C m2, Zostera marina

26.9 g C m2, and Carex 564 g C in2. The angiosperms entered the food web as detritus, and allochthonous sources of carbon (dissolved and particulate organic matter) from the river contributed over 2000 g C m2 yr'. Joint

(1978) measured rates of primary production on the sediment surface and in the -25--

water column along the coast of England. The annual primary production was

143 g C m2 for the sediment and 81.7 g C m2 for the water column. Primary production on the sediment surface ceased when the mudflat was flooded by the tide. Burkholderetal. (1965) found that daily rates of primary production

(carbon-14 method) for different functional groups of benthic microalgae averaged 4.45 mg C rr12 (blue-green algae), 4.05 mg C m2 (diatoms) and 5.03 mg

C m (mixed flagellates and diatoms) during some studies in Long Island

Sound. In the Chukchi Sea near Barrow, Alaska, the primary productivity of the benthic microflora ranged from below 0.5 mg C m2 hr in winter when the sampling area was covered with ice to nearly 57.0 mg Cm2 hr' in August

(Matheke and Homer, 1974). The latter value was eight times the productivity of the ice algae and twice that of the phytoplankton.

Taylor and Gebelein (1966) investigated the vertical distribution of plant pigments in intertidal sediments at Barnstable Harbor, Massachusetts.

Highest concentrations of all pigments occurred in the upper 1 mm. Chloro- phyll a and c and fucoxanthin concentrations decreased with depth and were

20 and 50% of surface values at 5 cm; diatoxanthin, diadinoxanthin, and carotene concentrations did not decrease with depth. In a related study,

Taylor (1964) showed that 10% of the solar radiation penetrated to a depth of

1.5 mm in sand with a grain size between 63 and 177 -'m in diameter; 1% reached a depth of 3 I'm. Microalgae living on sediment in this area required only

12 g cal cm2 hr' to obtain their maximum photosynthetic rate and were able to photosynthesize at 90% of their maximum rate at a depth of 2 mm at noon on a clear day. Riznyk and Phinney (1972) also investigated the vertical distribution of chorophyll a and primary production in the intertidal sediments of Southbeach and Sally's Bend in Yaquina Estuary, Oregon. The sandy silt of

Southbeach had an estimated annual gross primary production of 275 to 325 g C m2yr, while the finer silt of Sally's Bend had estimated values of 0 to

125 g C m2 yr. These differences were attributed to the presence of large populations of bacteria and meiofauna in the fine, detritus-rich sediment of

Sally's Bend. The greatest biomass of microalgae on both tidal flats was found in the upper 1 cm of sediment, but viable diatoms were found throughout the length of the piston core sampler (9.1 cm in length). Chlorophyll a at

Southbeach ranged from a mean concentration of 9.3 hg cm3 at a depth of 7.8-

9.1 cm to 20.7 hg cm3 in the upper 1.3 cm; corresponding concentrations at

Sally's Bend were 2.9 and 4,7 jig cni3, respectively.

Summarizing the primary production studies cited above, we can say that annual rates of gross primary production by niicroalgae on tidal flats fre'- quently fall between 50 and 200 g C m2 yr', and values for the epipsammic assemblages associated with shifting beach sands may be as low as 10 g C n12 yr or less. Microalgae and plant pigments usually are concentrated in the upper few millimeters of sediment, but living cells are often found at depths of 20 cm or more. Vertical distribution apparently is related to the extent to which the sediment is mixed by water movements, although epipelic diatoms can exhibit a vertical migration in the upper few millimeters.There is some indirect evidence that primary production and the development of assemblages of autotrophic organisms may be inhibited to some extent when there is a large amount of organic detritus in the sediments. Bacterial activity in such sediments reduces the oxygen concentrations, alters the pH, and perhaps generates compounds that are toxic to some microalgae. -27-

Zostera and Macroalgae

Rates of net primary production of eelgrass populations reported in the literature vary considerably. Representative values expressed as g dry weight m2 yr are 10 to 58 (Massachusetts Conover, 1958), 765 (New York Burk- holder and Doheny, 1968), 273 to 648 (Alaska McRoy, 1966), 24 to 842 (Cali- fornia - Keller, 1963), 187 to 1078 (Puget Sound, Washington Phillips,

1972), and 304 (Netarts Bay, Oregon Stout, 1976). Examples of biomass estimates for eelgrass during the growing season expressed as g dry weightm2 are 186 to 324 (Alaska McRoy, 1966), 6 to 421 (California - Keller, 1963;

Waddel, 1964), 5 to 29 (Massachusetts Conover, 1958), 148 to 2470 (New

York - Burkholder and Doheny, 1968), 18 to 396 (Puget Sound, Washington

Phillips, 1972), and 288 to 467 (Netarts Bay, Oregon Stout, 1976). Until the study described in this report, Stout's investigation of the Zostera population of Netarts Bay was the only published report of an eelgrass study in an Oregon Estuary. In addition to the production and biomass values reported above, she found that (1) the populations could be partitioned into shallow-water and deep-water eelgrass; (2) shallow-water and deep-water eelgrass averaged 671 shootsm2 and 1056 shoots m2, respectively; (3) the percentages of biomass in roots and rhizomes were 46% (shallow-water plants) and 29% (deep-water plants); and (4) the percentage of reproductive shoots were 17% (shallow-water plants) and 21% (deep-water plants).

Zostera marina has also been investigated relative to its biogeography

(McRoy, 1968), energy flow to consumers (Thayer etal., 1972), nitrogen fixation (Goering, 1974), and its associated animal populations (Kikuchi and

Peres, 1973). Here, we are particularly interested in the studies of nutrient -28-

dynamics and the epiphytic flora. McRoy and Barsdate (1970) found that eelgrass absorbs phosphate through both roots and leaves and that the plant at times may pump phosphate from the sediments back into the watecolumn. Up- take of phosphate was greatest in the light, but it also occurred in the dark.

Additional work by McRoy, etal. (1972) indicated that rates of uptake and excretion of phosphorus by both roots and leaves of eelgrass are dependent on the orthophosphate concentration in the medium. In these studies, eelgrass abosrbed 166 mg P m2 day from the sediments, assimilated 104 mgn12 da1, and excreted 62 mg m2 day back into the water column. The nitrogen-fixing capacity of three species of seagrasses and associated organisms was investigated by McRoy, Goering, and Chaney (1973). Rates of fixation of Zostera,

Thalassia, and Syringodium were low or undetectable, and it was concluded that the process is unimportant in at least some seagrass systems. These results were in contrast to the relatively high rates of fixation reported for Zostera marina by Patriquin and Knowles (1972) and for some other wetland plants of

Oregon by Tjepkenia and Evans (1976). McRoy (1974) found that epiphyte productivity was closely related to that of its host plant (Zoster marina in this case). NO3,NH4+,and (NH2)2C0 were all absorbed by the root-rhizome system and transported to all parts of the plant, and there was a direct transfer of both carbon and nitrogen from Zostera to the epiphytes on the leaves.Penhale

(1977) found that average net primary production was 0.9 g C m2 day for eelgrass and 0.2 g C m2 day for its epiphytes. Furthermore, it has been reported that epiphytes can reduce the rate of photosynthesis of eelgrass by as much as 45% and that this effect is influenced by light intensity and HCO3 concentration (Sand-Jensen, 1977). Penhale and Smith (1977) also found that -29-

heavily epiphytized Zostera excreted only 0.9% of its photosynthate and that

excretion was much less in the dark than in the light; excretion increased

after desiccation.

Relative to ecosystem bioenergetics the importance of macroalgae in

Oregon estuaries is greatest in systems having rocky areas, particularly long

jetties (e.g., Yaquina Estuary). In estuaries without such areas (e.g.,

Netarts Bay), species diversity is low, and the macroalgal biomass is usually

dominated by species of Enteromoropha, Gracilaria, Fucus, and some of the

smaller forms (e.g., Ectocarpus and Polysiphonia). Complete species lists

along with notes on distribution are available for Yaquina Estuary (Kjeldsen,

1967) and Netarts Bay (Stout, 1976), and Phinney (1977) has published a

comprehensive list of the macrophytic marine algae of Oregon. Effects of

variations of salinity and temperature on the photosynthetic and respiratory

rates of Ulva expansa, Enteromorpha linza, Laminaria saccharina, Sargassum

muticum, Alaria marginata, and Odonthalia floccosa from Yaquina Estuary have

been investigated by Kjeldsen and Phinney (1971).

Some Relevant Laboratory Studies

Mclntire and Wulff (1969) and Wulff and Mclntire (1972) studied the

effects of illumination intensity, exposure period, salinity, and temperature

on the primary productivity of estuarine periphyton in a laboratorymodel

ecosystem. The laboratory system consisted of afiberglassed wooden trough,

3inlong, 76 cm wide, and 80 cm deep, with the bottom graduated in a "stair-

step" manner. Tidal cycles were simulated by periodically pumping seawater

into the system, and the illumination intensity was regulated by adjusting the -30-

height of a large lamp fixture over the trough. A respirometer chamber was

designed to monitor changes in dissolved oxygen concentration in water sur-

rounding a sample community. Such samples consisted of periphyton assemblages

that developed in the laboratory ecosystem on acrylic plastic plates or other

substrates of interest.

Results of experiments conducted with the laboratory ecosystem indicated

that periphyton biomass accumulated most rapidly on plastic substrates sub-

jected to relatively high illumination intensities without exposure to desic-

cation. In the absence of grazing, biomass ranged from 17.2 g m2 on a sub-

strate exposed to the air for 8 hr per day to 128 g m2 on a substrate with no

exposure; corresponding concentrations of chlorophyll a were 0.037 and 0.837 g m2, respectively. Primary productivity in periphyton assemblages exposed to

periods of desiccation was less under winter conditions than under correspond-

ing conditions in the summer; and productivity in assemblages not exposed to

desiccation was strongly affected by illumination intensity during both the

summer and winter experiments. Rates of primary production at 18,500 lux

ranged from about 0.08 to 1.00 g 02 m2 hr depending on the biomass and

chlorophyll concentration.

Since 1976, Admiraal and associates have reported the outcome of a nunber

of experiments related to the ecology of benthic diatoms in the Eems-Dollard

estuary, a part of the Dutch Wadden Sea (Admiraal 1977 a, b, c, d, e; Admiraal

and Peletier, 1979 a, b; 1980 a, b). Some of the results of these studies were:

1. Cell division rates in unialgal diatom cultures decreased when the

light intensity decreased below 5 E m2 daf1 or when the daily

photoperiods were shorter than 8 hours. -31-

2. The cell division rate was proportional to the incubation temperatures

between 4 and 20°C.

3. Light extinction in the sediment column was of critical importance

with respect to the photosynthetic rate and growth dynamics of natural

sediment inhabiting diatom populations.

4. The photosynthetic rate of cultures and natural diatom populations

were very high during incubation in media with salinities between

4 0/00 and 60 /oo. Differences among species were found only at

salinities below 8 0/00

5. The supply of phosphate and probably the supply of nitrogenous

compounds were not limiting in the regulation of the numbers of

benthic diatoms in the estuary.

6. Accumulation of oxygen and the depletion of carbon dioxide were

indicated as the cause of retarded growth in dense assemblages of

benthic diatoms.

7. Continuous discharges of organic wastewater stimulated the formation

of dense assemblages of benthic diatoms, while the associated high

concentrations of ammonia promoted the dominance of two ammonia

tolerant species.

8. The presence of sulfide near the surface of the intertidal sediment

eliminated certain sulfide sensitive species from the diatom

assemblage.

9. Six out of ten species of diatoms isolated from the estuary were

capable of heterotophic growth in the dark, and under limiting light

levels, the addition of organic substrates increased the division

rates of these species. -32-

Jonge (1980) investigated fluctuations in the organic carbon to chlorophyll a ratio for diatom assemblages isolated from the sediments of the Ems estuary. Values for this ratio over a 3-year period ranged from 10.2 to

153.9 with yearly averages and standard deviations of 40.3±13.8; 41.2±20.4, and 61.4 ± 22.0, respectively. The method of isolation involved the use of lens tissues and a filtration procedure which produced a suspension of epipelic diatoms with relatively little contamination by small animals or bacteria. This procedure was of particular relevance to our study, as it provided a living isolated flora of microalgae for estimates of respiration and for the establishment of a biomass to chlorophyll a ratio.

In a recent paper, Revsbechetal. (1981) compared the results of three different methods of measuring primary production of sediment-associated microalgae. The methods under Investigation were by an oxygen microprofile using a platinum microelectrode, H'4CO3fixation, and the standard oxygen exchange method. In a highly oxidized sediment, the three methods yielded almost identical results at low light intensities (20 1-'E m2 sec). The oxygen methods underestimated primary production at higher light intensities when there was conspicuous bubble formation. Also, the conventional oxygen method underestimated the primary production in sulfuretum-type sediments as compared to the other two methods. Measurements of the specific activity of

HCO within the photic zone showed a steep gradient of H14CO3at the sediment surface. Calculations of benthic primary production taking the actual specific activity into account yielded 2 to 5 times higher estimates than calculations using the specific activity in the overlying water. -33-

Columbia River Project

This project was initiated by the P.1. in September 1979 as part of the

Columbia River Estuary Data Development Program (CREDDP) and continued until

October 1, 1981. The general objective of the research was to investigate the production dynamics of benthic plants on the tidal flats of the Columbia River

Estuary. In particular, the work was concerned with effects of chemical and physical gradients on the structural and functional attributes of micro- and macro-vegetation and on the productivity and biomass of the benthic primary food supply. The research plan for the project included (1) a descriptive field study of the production dynamics of benthic plants on the tidal flats of the estuary; and (2) an investigation of mechanisms accounting for the observed dynamics in the field. Unfortunately, the unanticipated early termination of CREDDP allowed time and support only for a field investigation at selective intensive study sites. However, observational data obtained in the field provided a good basis for generating hypotheses concerning mechanisms that regulate autotrophic processes, hypotheses that were examined in part by the research described in this report.

Because of the enormous size of the area under investigation by the

Columbia River Estuary Data Development Program (ca., 150 square miles), the selection of a suitable sampling strategy for the field investigation of benthic autotrophy was a very difficult problem. Essentially, two alternative approaches were considered: (1) a broad survey involving the collection of non-replicated samples from as many locations as possible over the entire study area at perhaps two or at most, three different seasons of the year; or

(2) frequent replicated sampling at relatively few intensive study sites. The -34-

first approach provides an insight into variation over the largest possible spatial area but fails to yield information about local variation in space and time. The second approach allows the calculation of a variance structure for each site and gives much more information about temporal variation. Moreover, if the intensive study sites are representative of large areas in the estuary, the second approach can provide considerably more insight into process mechanisms than the first approach, particularly if concurrent measurements of physical variables are obtained along with the biological data. Data obtained under the direction of the P.r. were generated from a sampling program based on the second approach, replicated samples taken at monthly intervals from five intensive study sites. A similar approach was used for the field work described in this report.

Prior to the beginning of the field sampling program in April 1980, the estuary was surveyed to identify potential sites for intensive study. Five sites were selected on the basis of their relative positions along the salinity gradient, their sediment type, and whether or not they were representative of large, common habitat types in the estuary. The intensive study sites with corresponding CREDDP coordinates were Clatsop Spit (3-59-13), Youngs Bay (3-

53-10), Baker Bay (4-0-18), Grays Bay (3-40-17), and Quinns Island (3-29-14).

The sites at Clatsop Spit and Baker Bay were under marine influence, and surface salinities ranged from 32 0/00 at high tide during low freshwater discharge to 0 0/oo at low tide during high discharge. The Clatsop Spit site was located on the northern side of Clatsop Spit, approximately 1 km west of

Parking Lot D. This site was characterized by fine sand and relatively high current velocities. The Baker Bay site was located on the northern side of -35-

Baker Bay, near the Liwaco Airfield. The sediment was primarily coarse silt to very fine sand. The site at Youngs Bay was on the western side of the bay, approximately 1 km south of the mouth of the Skipanon River. Here, surface salinities varied from 0 to 10 0/00, grain size of the sediments ranged from medium silt to fine sand. Sites in Grays Bay and on Quinns Island were under strong freshwater influence, with surface salinities always near 00/00. The

Grays Bay site was located on the eastern side of the bay, approximately 1 km south of the mouth of Crooked Creek. The sediments were composed primarily of very fine sand. The site at Quinns Island was located on the eastern tip of the island. Because it was more exposed to river currents than the Grays Bay site, the sediments were coarser, ranging in grain size from very fine sand to medium sand.

At each intensive study site, 25-rn horizontal transects were identified and marked with wooden stakes. The transects were located in the high, mid, and low intertidal zones at each station and in the high marsh at all stations but Clatsop Spit where no marsh exists. The distance between the upper transect in the high marsh and the lowest intertidal transect varied from station to station depending on the slope of the tidal flat. The transects in the marsh and in the high, mid, and low intertidal regions were approximately

0.9 m, 0.7 in, 0.5 in, and 0.3 in above mean lower low water, respectively.

Field sampling was designed to generate estimates of primary production and chlorophyll a concentration in the sediments. The sampling strategy at each intensive study site involved the collection of sediment cores for the analysis of chlorophll a concentration and for measurements of primary production in a respirometer chamber. Six cores were obtained at random loca- -36-

tions along each transect, and each of these was subsampled in the laboratory

to obtain estimates of chlorophyll concentration in the upper cm, between

4.5 and 5.5 cm from the surface, and between 9 and 10 cm from the surface.

Concurrently, two sediment cores were obtained from each of the transects in

the marsh and in the upper and lower intertidal regions for measurements of

primary production and oxygen consumption. These cores were subsampled after

these measurements for estimates of chiorphyll a concentration and organic matter in the upper centimeter of sediment. Selected physical variables also

were monitored along with the measurements of primary production and chloro-

phyll concentration. The physical variables of interest included temperature,

light intensity, salinity, and five properties of the sediment: median grain

size, mean grain size, skewness, kurtosis, and the sorting coefficient.

Preliminary results from the analysis of data obtained from April through

October 1980 are summarized below:

1. Diatoms were the most abundant group of plants on the tidal flats in

the Columbia River Estuary. Although large numbers of diatom species

were found on each tidal flat under investigation, species composition

varied greatly among tidal flats. Blue-green algae were frequently

found growing beneath the emergent marsh plants in late summer at all

sites. Macrophytes were not conspicuous on the estuarine tidal flats.

Zostera marina L. and an unidentified species of Zostera had a patchy

distribution in Baker Bay, and Enteromorpha intestinalis var. maxima

J. Ag., a filamentous green alga, was abundant in marsh samples from

Youngs Bay and Baker Bay in April and May. A sparse growth of

Potamogeton foliosus Raf. and P. richardsonii (Benn.) Rydb. was -37-

observed on the tidal flats of Grays Bay during spring and summer,

while Ceratophyllum demersum L. and Elodea canadensis Michx. were

often abundant in marsh pools at Grays Bay.

2. The highest rates of gross primary production (GPP) were recorded at

the Youngs Bay site. During the period from May through October, the

mean of all measurements of GPP for Youngs Bay was 108.46 mg of carbon

fixed per square meter per hour at light saturation. The mean GPP (mg

2 c hr) was 53.89 at Baker Bay, 35.44 at Quinns Island, 32.36 at

Grays Bay, and 3.00 at Clatsop Spit.

In general, rates of GPP declined in June and July. This was

particularly evident at the upriver sites of Quinns Island and Grays

Bay. During this period, increases in scouring and sediment load in

the water column, which were attributed to early summer freshet and

the Mount St. Helens eruption, created an unstable and presumably an

unsuitable habitat for the benthic microflora. Therefore, rates of

GPP decreased at the lower intertidal levels, which are more exposed

to river flow. With a decline in freshwater discharge and in the

activity of Mount St. Helens, sediments stabilized and GPP increased

during August and September. Rates of GPP throughout the estuary

appear to be closely associated with sediment stability; the more

stable the substratum, the higher the rates of GPP.

Production of the benthic microflora in the marsh declined in

spring and early summer as emergent marsh plants began to develop and

shade the sediment surface. Subsequently, as the emergent vegetation

began to decline in August and September, production of the benthic

microflora began to increase with increasing available light. 3. Highest concentrations of chlorophyll a in the upper cm of the sedi-

ment were generally associated with sites with the greatest rates of

GPP. Mean concentrations of chlorophyll a (g chl a cm2) in the

upper cm for April through October were 31.77 in Youngs Bay, 23.03 in

Baker Bay, 11.90 in Grays Bay, 10.52 at Quinns Island, and 1.35 at

Clatsop Spit. Concentrations of chlorophyll a were usually highest in

the marsh and lowest in the lower intertidal zone. There were no

conspicuous seasonal changes in chlorophyll a concentration at the

intensive study sites. The highest and lowest monthly mean values

were recorded in April and July, respectively.

In general, it was found that chlorophyll a concentration in the

top cm of sediment was a relatively good predictor of benthic primary

production when the flora was composed of diatoms. The regression

equation derived from the field data is:

GPP = 2.8955 + 0.3001 CHLOR,

where GPP is the gross primary production expressed as mg C hr and

CHLOR is the chlorophyll a concentration in the top cm of sediment

expressed as mg. The corresponding R2 value for this relationship was

0.732, indicating that measurements of chlorophyll a concentration can

serve as a reasonably good tentative estimate of GPP for locations

where direct measurements of primary production are not possible.

This conclusion only applies to assemblages of benthic microalgae in

the intertidal area of the Columbia River estuary, as this relation-

ship is variable in a more marine system such as Netarts Bay (see a

later section of this report). -39--

4. The percentage of organic matter in each sediment sample expressed as

ashfree dry weight showed remarkably little variation at each site

during the study period. There was no apparent increase in organic

matter in the sediments as the emergent plant vegetation began to die

back in late summer and early fall. In general, highest percentages

of organic matter were associated with sediments in the marsh. Mean

percentages of ashfree dry weight in sediment samples for April

through October were 4.35 in Baker Bay, 3.26 in Youngs Bay, 2.12 in

Grays Bay, 1.42 at Quirins Island, and 0.42 at Clatsop Spit. The

higher percentages were associated with the finergrained sediments. -40-

THE ALGAL PRIMARY PRODUCTION SUBSYSTEM

Studies of algal primary production in Oregon estuaries have been limited, but some early studies included production measurements made in the laboratory using intertidal sediment (Riznyk and Phinney, 1972), artificial substrates (Mclntire and Wulff, 1969; Wuiff and Mclntire, 1972), and macroalgae (Kjeldsen and Phinney, 1971). Results reported here, together with the recent study in the Columbia River estuary (Amspoker and Mclntire, 1982), are the first field measurements of algal primary production for sediment-associated communities in Oregon estuaries.

The purpose of the research reported in this section was. to describe seasonal patterns of algal primary production and associated physical and biological variables in Netarts Bay. In addition, experiments were conducted at the O.S.U. Marine Science Center, Newport, Oregon, to examine mechanisms that accounted for the patterns observed in the field. Priorities for the experimental work were established by the conceptual framework described in an earlier Section. Because the research examined autotrophic processes that occur in many estuaries, the understanding of these processes in Netarts Bay is relevant to problems that extend beyond the geographical limits of this particular estuary.

Results of our research on the Algal Primary Production subsystem are reported in four subsections: 1. Production Dynamics of Sediment-Associated

Algae in Netarts Bay, Oregon; II. Experimental Studies of Estuarine Benthic

Algae; III. Some Effects of Estuarine Infauna on Sediment-Associated Micro- algae; and IV. The Diatom Flora of Netarts Bay. In subsection I, seasonal patterns of primary production for sediment-associated algae at three inten- -41-

sive study sites in Netarts Bay are described and interpreted relative to

selected physical variables. Laboratory studies of estuarine benthic algae are reported in subsection II. In particular, these experiments were con-

cerned with the relationship between algal primary production and light

intensity; also, the ratio of chlorophyll a to algal biomass was investigated

in isolated diatom assemblages. In subsection III, effects of infauna on

algal production dynamics are described from the results of defaunation

experiments in the field and laboratory. Subsection IV is concerned with the

taxonomic structure of the sediment-associated and epiphytic diatom flora of

Netarts Bay.

I. Production Dynamics of Sediment-Associated Algae in Netarts Bay, Oregon

This research was concerned with the investigation of seasonal changes in

algal biomass and patterns of algal primary production in Netarts Bay. Rele-

vant biological variables included microalgal biomass expressed as the chloro-

phyll a concentration in the sediment, the concentration of organic matter in

the sediment expressed as ash-free dry weight, and the rates of gross primary

production and community oxygen uptake as measured in light and dark chambers.

The sampling strategy was consistent with the known dynamics of physical and

chemical processes in Netarts Bay as determined by the EPA Rhodamine studies

(EPA, 1979).

Sampling Strategy

Netarts Bay was surveyed to identify potential sites for intensive study

which were representative of large areas of the bay. Three sites were -42--

identified on the basis of sediment type and location in the bay (Fig. 7).

The site near the mouth of the bay was characterized by medium sand (SAND site); the site along the western shore of the bay had fine sand (FINE SAND site); and the site along the eastern shore had coarse silt (SILT site).All sites were exposed to moderate wave action. Macroalgae were absent at the

SILT site, and blue-green algae occurred at the mean high water (MHW) level at the FINE SAND and SILT sites.

At each intensive study site, 50-rn horizontal transects were marked with wooden stakes. The transects were located at 0.5, 1.0, 1.5, and 2.0 m above mean lower low water (MLLW) at each site. The sampling strategy at each, site involved the monthly collection of sediment cores for the measurements of primary production and the concentration of chlorophyll a and organic matter.

On each sampling date, sediment cores were obtained randomly along each transect: three for measurement of primary production, six for analysis of chlorophyll a concentration, and three for analysis of organic matter concentration. Sampling was initiated in March, 1980 and continued for a period ending in March, 1981. Cores for the measurement of primary production were incubated in situ in field respirometers and subsampled after incubation for the measurement of chlorophyll a concentration in the top cm of sediment.

Cores for the measurement of chlorophyll a concentration and organic matter content were subsampled in the laboratory to obtain estimates of concentration in the top cm of sediment and at a depth of 4.0 to 5.0 cm from the surface.

Monthly measurements of physical variables also were made near the transects of all three intensive study sites. These variables included temperature, salinity, light intensity of photosynthetically active radiation, and change in sediment height. and N P4CIFIc OREGON 1000 -43- METERS I- 0 (A) the SAND site; (B) the FINE SAND site (C) the SILT site. study sites: Map of NetartS Bay indicating the location of intensive NE TARTS BAY OREGON BAY NE TARTS igure 7. -44-

Methods

1. Primary Production

Gross primary production and community oxygen uptake were estimated in the field from oxygen measurements in stirred, light and dark chambers designed to hold intact cores of sediment. Between May and September, measurements were made in plexiglas chambers which were 92 cm in height, had an internal diameter of 12.8 cm, and enclosed a sediment surface area of

128.61 cm2 (Fig. 8). Two light chambers and one dark chamber were used to estimate primary production at each site. Each chamber was inserted 30 cm into the sediment, filled with 5.7 1 of seawater from the bay and sealed.

Water in the chamber was circulated by a magnetic stirrer at a motor speed of

27 rpm. Rates of oxygen evolution and uptake were based on measurement periods of 4.0 to 7.5 hours between initial and final readings. Measurements of dissolved oxygen were made with an Orbisphere® salinity-corrected dissolved oxygen system by inserting the oxygen probe into a port on the side of the chamber.

Between October and March, estimates of primary production were made in plexiglas chambers which were 14.2 cm in height, hadan internal diameter of

6.8 cm, and enclosed a sediment surface area of 36.32 cm2 (Fig. 9).Three replicate chambers were inserted 5 cm into the sediment, withdrawn with intact sediment cores, and plugged with rubber stoppers. Each chamber was filled with 300 ml of seawater from the bay, sealed and placed in a water bath.

Water in the chambers was circulated by a magnetic stirrer at a motor speed of

300 rpm. Rates of oxygen evolution and uptake were based on measurement periods of 0.5 to 1.0 hour between initial and final readings. Readings were made by replacing the stirrer on top of the chamber withan oxygen probe. -45-

WATER LEVEL

0-RINC GASKE7

MOTOR- BATTERY UNIT

PORT MAGNE7 PORT FOR OXYGEN 0-RING METER PROBE GASKET INCUBATING MAGNET MATERIAL

LOWER COMPARrMENT1

I II I II II II II II II I I II I I II I I II II II I I II I I

Figure8. Diagram ofa respirometer designed for the measurement of primaryproduction in a seagrasS communitY. -46-

TURBINE V WATER MAGNET SOURCE 0-RING SEAL MAGNET PORT FORSTIRRER AND PROBE

-PORT FOR FILLING

SEAWATER

ALGAE AND SEDIMENT

RUBBER BUNG

PLEXIGLAS CHAMBER FOR 02 METABOLISM

Figure 9. Diagram of a raspirometer designed for the measurement of benthjc primary productjo in the laboratory. -47-

Rates of net community production or oxygen uptake were measured when the chambers were exposed to full sunlight or darkened by covering the water bath with black plastic, respectively.

Simultaneous incubations of intact sediment in both types of chambers indicated that chamber type had no significant effect on the rate of primary production. All production rates were corrected for seawater-associated oxygen production in the light, and oxygen uptake in the dark using the light and dark BOD bottle method for measuring plankton metabolism (Strickland and

Parsons, 1972). Plankton metabolism was negligible except during August.

Gross primary production was estimated by adding the rate of community oxygen uptake in the dark to the rate of net community production in the light for an equivalent period of time. Estimated rates of gross primary production expressed as mg 02 rn2 h were converted to mg Cm2 h' while assuming a photosynthetic quotient of 1.2, i.e., mg C = 0.312 x mg °2 Estimated rates of community oxygen uptake expressed as mg °2m2 h were converted to mg C m2 h while assuming a respiratory quoteint of 1.0, i.e., mg C = 0.375 x mg

02 (Westlake, 1965).

2. Biomass

Microalgal biomass was expressed as the concentration of chlorophyll a in the sediment. Samples for chlorophyll a analysis were collected using a 12-cm long plastic corer with an internal diameter of 2.3 cm. The corers were gently pressed 10 cm into the sediment and then were withdrawn carefully with an intact core. The cores were capped, frozen, and transported back to the laboratory. En the laboratory, the frozen sediment was extruded from the cores, and sections from the top cm and 4 to 5 cm from the surface were -48-

excised with a knife. A core section was placed in a mortar with 0.5 ml of

saturated magnesium carbonate solution and 10 ml of 90% acetone (v/v). The

slurry was ground with a pestle for one minute, poured into a screw-capped

test tube, and kept in the dark at 4°C for 24 hours. The extract was centri-

fuged, and chlorophyll a concentration was measured using the method of

Strickland and Parsons (1972). The Lorenzen equation (corrected for phaeopig- ments) was used to calculate mg chlorophyllaiti2. These values are uncor-

rected for chlorophyllide (Whitney and Darley, 1979).

Macroalgal biomass was sampled by harvesting six replicate 0.25 in2 quad-

rats randomly along each 50-rn transect. Samples were rinsed in fresh water, weighed, dried at 70°C for 48 hours, and reweighed.Dried samples were ashed at 450°C for 24 hours and reweighed to determine ash-free dry weight, an estimate of organic matter. Chlorophyll a concentration in the macroalgae was determined as described above using approximately 1.0 g wet weight of algae per sample.

3. Sediment Properties

Sediment was sampled for organic matter content using the same coring devise described above for the chlorophylla samples. Samples were collected, frozen, and transported back to the laboratory. In the laboratory, the frozen sediment was extruded from the corers and sections from the topcm and 4 to

5 cm from the surface were excised with a knife. Core sections were dried at

70°C for 48 hours and weighed. These sections then were ashed at 450°C for

24 hours and reweighed. A correction for loss of sediment water of hydration after ashing was determined by adding distilled water to the ashed sediment, and drying at 90°C for 24 hours and reweighing. -49-

Bimonthly samples from the top cm of sediment were taken for grain size analysis (Buchanan and Kain, 1971). Sediment was wet-sieved with a 63 1-'m mesh sieve. The coarse fraction was dried at 70°C for 24 hours and then sieved through a set of graded sieves to determine fraction weights above 63 1Am.

Fractions below 63 urn were examined by a pipette analysis. Sediment statistics included mean grain size, the sorting coefficient, arid a measure of skewness (Inman, 1952). Mean grain size was expressed in phi-units where =

-log2 of the grain size in mm. The sorting coefficient expresses the uni- formity of grain size, i.e., the better sorted sediments exhibit the less uniform distribution of size classes. Skewness measures the degree of symmetry in the grain size distribution, with positive values indicating skewness to the smaller grain size and negative values indicating skewness to the larger grain size.

4. Physical and Chemical Variables

Light intensity was measured in iE m2 sec photosynthetically active radiation with a Licor® quantum meter. Bimetal or thermistor probes were used to measure temperature, while salinity was measured with a temperature-compensated AO Goldberg® refractometer. Samples for analysis of nitrite-nitrate

(NO2-NO3), ammonia (NH4), orthophosphate (PO4) and molybdate-reactive silica

(S102) were taken July, August, and September from water in the respirometer chambers before and after incubation. These samples were analyzed by methods for autoanalysis of nutrients (Strickland and Parsons, 1972).

5. Data Analysis

The chlorophyll and production data were analyzed by a three-way analysis of variance (ANOVA) with sediment type, tidal height, and time treated as main -50-

effects; the three-way interaction mean square was usedas the error term in

calculating F-values. All statistical analyses were performed with a Control

Data Corporation Cyber 170/720 computer at the Oregon State University

Computing Center using the SPSS system (Nie et al., 1975) and the REGRESS

subsystem of SIPS (Rowe and Brenne, 1981).

Results

1. Physical Properties

Sediment characteristics were determined for the intensive study sites

(the SAND, FINE SAND, and SILT sites in Fig. 7) from March, 1980to March,

1981 in Netarts Bay. Mean grain sizes in phi-units () were 2.28 at the SAND site, 2.66 at the FINE SAND site, and 3.79 at the SILT site (Table 1).Mean values for the sorting coefficient were 0.34 (well-sorted) at the SAND site,

0.64 at the FINE SAND site, and 1.13 (poorly-sorted)at the SILT site; while corresponding mean skewness values were 0.25, 0.10, and -0.04, respectively.

In summary, sediment at the SAND site was medium to fine sand, well- sorted, with the most positive skew; sediment at the FINE SAND sitewas fine to very fine sand, with medium sorting anda positive skew; and sediment at the SILT site was very fine sand to coarse silt, poorly sorted andwith very little skew.

Monthly change in sediment height was measured at the three intensive study sites from April to October (Fig. 10). There was little net change in sediment height, except at the SAND site at 1.0 m above MLLW and at the FINE

SAND site at 2.0 in, and 0.5inabove MLLW. Monthly changes in sediment level usually ranged from 0 cm to 3 cm, changes that could account for burialor -51-

Table 1. Results of sediment grain size analysis for samples obtained every two months from March 1980, to March 1981, at intensive study sites. Tidal level is aboveE.LW; 0.5 m (1), 1.0 m (2), 1.5 bi(3), and 2.0 m (4). Mean grain size (MEAN) is in phi-units, and sorting (SORT) and skewness (SKEW) are dimensionless. Statistics were calculated according to Inman (1953). Values are annual means () of n replicates and the standard error of the mean (SE).

EAN SORT SKEW

Site Level SE SE SE

SAND All 23 2.28 0.03 0.34 0.25 0.43

1 5 2.28 0.10 0.31 1.25 1.25

2 6 2.27 0.03 0.36 0.47 1.20

3 6 2.22 0.05 0.33 -0.54 0.50

4 6 2.34 0.03 0.35 0.00 0.01

FINE SAND All 21 2.66 0.10 0.64 0.10 0.65

1 3 3.04 0.35 0.70 -0.05 0.22

2 6 2.64 0.14 0.61 0.15 0.11

3 6 2.52 0.25 0.64 0.06 0.07

4 6 2.64 0.13 0.65 0.17 0.08

SILT All 23 3.79 0.11 1.13 -0.04 0.17

1 5 4.03 0.09 1.24 0.10 0.03

2 6 3.95 0.13 1.00 -0.54 0.61

3 6 3.78 0.34 1.08 0.01 0.08

4 6 3.44 0.19 1.20 0.29 0.07

I I I

include SANI) (solid circle), FINE/SAND (triangle), and SILT (open circle). (open SILT and (triangle), FINE/SAND circle), (solid SANI) include

tlonthlv changes in sediment height at intensive study sites in.Netartn flay. in.Netartn sites study intensive at height sediment in changes tlonthlv FIgure 10. FIgure Sites

MONTH

M A S A J J

0 I I I I -5.0 1 z I

0.5 N + MLLW + N 0.5

I I I- I I z

(9 . hi

ON + MLLW + ON 10

(I) hi

U' I I I I I I

a. I

- a J +5Ø J I.5M + MLLW + I.5M

___ hi I I -5.0 d

I I I I I 0

ML - I I I I Ml I +YiO +5.0 -53-

export of surface chlorophyll a and organic matter at certain times and sites.

The SILT site was the most stable site with respect to change in sediment

level during the measurement period.

Temperatures of the air, exposed sediment, and water were usually greater

than 10°C (Fig. 11). Air, exposed sediment, and water temperature ranges were

from 0°C to 24°C, from 2°C to 19°C, and from 8°C to 18°C, respectively.

Photosynthetically active radiation ranged from 0.5 to 60 Em2 daf1 and was chiefly influenced by day length and cloud cover (Fig. 12). Extinction

coefficients were calculated from light intensity measurements taken at high

tide at all three intensive study sites (Table 2). These coefficients ranged

from 0.400 to 0.870 and were generally lower at the SAND and-FINE SAND sites

than at the SILT site.

An example of the effect of water depth on light available at the sedi-

ment surface is shown in Figure 13. Measurements of light intensity at the

water surface, water depth, and the extinction coefficient were made in

Netarts Bay for a day in July and a day in September. The light intensity

at the sediment surface was calculated for each hour between 8:00 a.m. and

4:00 p.m. PST, at three tidal elevations, by substituting the appropriate

light intensity, water depth and extinction coefficient into the expression:

= ie; where 1 is the light intensity at depth z, 10 is the light

intensity at the surface and Tiis the extinction coefficient. A clear,

sunny day occurred in July; while in September, the sun only appeared between

11:00 a.m. and 12:30 p.m. PST. Light intensity was not limiting for photo-

synthesis between 9:00 a.m. and 4:00 p.m. during the day in July; but in

September, the light intensity often fell below light-saturation of photo-

synthesis which was approximately 1E m2h1 (see results of laboratory study

described in a later section). -

H I z I LiJ - I iJX i II I I I range of temperature at all intensive study during the period from April, 1980, to March, MONTH 111111 jili A SO ND J FM III I IJi J Bars Indicate daily 11111 II 'II Bay. I I I II AM J

I I l0 10 Temperature of air, sediment and water 1981 in Netarts sites on the sampling dates. Ui a- Ui I C) 0 uJ Figure 11.

May, 1981 at Netarts Bay. Netarts at 1981 May,

period from May, 1980, to 1980, May, from period PhotosynthetIcally active radiation (E nr2 day-i) for the for day-i) nr2 (E radiation active PhotosynthetIcally

Figure 12. Figure J JASON J MONTH J FM AM FM J

S ._____

. S. ------0 . 0 0/ S N 0 10/ 0 -56-

Table 2. Extinction coefficients ()a measured at high tideat the SAND, FINE SAND and SILT intensive study sites. Light intensity at theurfac and 1.0 in depth is photosynthetically active radiation (Em sec ). High tide was between 11:00 and 13:00 PST.

Light Intensity Site Date Surface 1.0inDepth

SAND 13May 1375 737 0.624 iiJune 1250 670 0.624 iiJuly 1350 804 0.518 10August 1000 509 0.675 7September 1500 804 0.624

FINE SAND 12May 570 281 0.707 10June 2100 1206 0.555 10July 1800 1206 0.400

9August 820 496 0.503 6September 1500 804 0.624

SILT 11May 615 295 0.735 9June 800 335 0.870 9July 460 241 0.646 8August 870 375 0.842 5September 680 375 0.595

a) n= in I- in I; where I= surface light intensity and I= light intensity at 1.0indepth. -57-

I I I

oO°\\ 4I N

3 I HIGH / I I..'\ OMID 1 I I/SEPTI £LOW 2- I /

, a

L _L 9:00 11:00 1:00 3:00 PS.T Figure 13. Effect of water depth on light available to sediment- associated algae in Netarts Bay for a day in July and a day in September at three intertidal heights: HIGH (1.5 m above MLLW), MID (1.0 in above MLLW) and LOW (0.5 m above MLLW). High tide was at noon (2.0inabove flLW) Values were calculated using measured water levels, extinction coefficients, and light intensities at the water surface. The line at 1.0 E m2 hr- approximates light saturation of photosynthesis. -58-

Light transmission through the sediment at all three of the intensive study sites was determined by measuring light transmission through 1 mm thicknesses of sediment in petri dishes (Haardt and Nielsen, 1980). Reduction of photosynthetically active radiation to 1.0% of surface light intensity was reached at 2.55 mm at the SAND site, 2.00 mm at the FINE SAND site, and 1.30 mm at the SILT Site. Salinity in the water column of Netarts Bay ranged from 28 to 34 0/00 during the study. Values for interstitial water usually varied between 25 and 35 /oo. However, interstitial water at 2.0 m above fLLW in all sediment types reached values as low as 5 /oo during rain storms or periods of terrestrial runoff from ground water seepage. During July, August, and September nutrient concentrations (NO2-No3, NH4, PO4, and SiO3) in the water from the respirometer chambers were measured before and after in situ incubation (Table 3). The only significant changes in concentrations during the incubation period occurred when Enteromorpha prolif era (Setch.) S.&G. was present in the chambers. For example, the concentration of NO2-NO3 decreased from 3.23 to 1.05 I.M during the six-hour incubation period in August at the SAND site.

2. Organic Matter There was a significant positive correlation (n = 34, r = 0.82; p < .01) between sediment mean grain size in phi-units () and the organic matter concentration (AFDW) in the top cm of sediment. The corresponding regression equation expressing mean grain size as a predictor of organic matter was AFDW =

-89.54 + 109.21 1?. An analysis of variance indicated that there were significant differences (P < .01) in organic matter concentration in the top cm of sediment associated with differences in sediment type (Appendix I). The SILT site had the highest mean organic matter concentration followed by the FINE SAND site and the SAND site, respectively (Table 4). There also were significant differences (P < .01) in the concentration of organic matter associated with the effects of tidal height, time, and a two-way interaction between sediment type and tidal height (Table 4, Figs. 14 and 15). Organic matter concentration was highest during the summer. At the SILT and FINE SAND sites, the transects at 2.0 m above MLLW had the highest concentrations of organic matter; while at the SAND site, the highest concentrations were at 0.5 m above MLLW. The ratio of organic matter in the top cm of sediment to that at the 4 to 5 cm depth was calculated and used as an index of mixing of organic matter in the sediment. No significant differences related to the effects of sediment type, tidal height, or time were found in the bay sediments relative to this ratio. The mean value of the ratio for all sediment types was 1.50 indicating that organic matter usually was present at both depths in similar concentrations. -59-

Table 3. Initial nutrient concentrations (iiM) in respirometer chambers at the SAND, FINE SAND and SILT intensive study sites during sampling in July, August and September.

July August September

NO2-NO3

SAND 3.23 3.23 0.81 FINE SAND 0.16 0.32 0.16 SILT 0.48 0.24 0.16

NB4

SAND 0.56 1.11 2.78 FINE SAND 0.56 2.22 2.22

SILT 1.11 1.67 2.22

0.48 0.48 0.34 0.39 0.39 0.39 0.39 0.39 0.39

14.46 14.46 10.51

11.83 11.17 17.09 13.14 12.49 15.77 Table 4. Concentration of organic matter (g m2) in the top cm of sediment, expressed as ash-free dry weight, in Netarts Bay at the SAND, FINE

SAND and SILT sites. Intertidal levels are above MLLW: 0.5in(1),

1.0 m (2), 1.5in(3), and 2.0 (4). Values are site and tidal level means () and standard errors (SE) of n observations.

Site SE

SAND 43 134.05 8.56 Level 1 10 196.95 15.71 2 11 138.88 14.72 3 11 119.73 9.93 4 11 86.36 7.41

FINE SAND 38 203.88 33.07 Level 1 6 182.11 23.20 2 10 201.53 20.30 3 11 205.38 16.82 4 11 216.38 17.02

SILT 43 377.76 13.93 Level 1 10 385.99 19.09 2 11 350.07 20.18 3 11 325.61 19.42 4 11 450.11 33.59 -61-

lOM t MLLW datum ScC. SANO T±SE £FINE SAND 50 0. o SILT

400

E 4 a' 301 0 U-

201

r MA M J J A SO N 0 J FM MONTH

_1 O.5M i MLLW datum SAND ± SE iFINE SAND

o SI LI

E a' 300 0 U- 200

too

r ri U M J A S 0 N 0 J F M MONTH Figure 14. Concentration of organic matter in thetop cm of sediment expressed as ash-free dry weight (AFDW)at intensive study sites from March, 1980, to March, 1981. Sites are SAND, FINE SAND, and SILT at 1.0m and 0.5 in above MLLW. Values are means of three replicates. -62-

2.OM + MLLW datum

600 S SAND 7 SE £ FINE SAND 0 SiLT 500 a N TI/f 400 E /t1

0 Li. /

FM AM .J J A SO ND J FM MONTH

I.5M + MLLW datum

SAND ±SE £ FINE SAND

500 o SI LI a N

300 0 U.

F MA M J J A SON D J FM MONTH Figure 15. Concentration of organic matter in the top cm of sediment expressed as ash-free dry weight (AFDW) at intensive study sites from March, 1980, to March, 1981. Sites are SAND, FINE SAND, and SILT at 20 m and 1.5 m above MLLW. Values are means of three replicates. -63-

Sediment-associated total organic matter was separated into its component

fractions (Table 5). The total represents monthly measurements of organic matter concentration in the top cm of sediment. Microalgal biomass was calcu-

lated using a conversion factor between chlorophyll a and organic matter determined in the laboratory throughout the year with epipelic diatoms iso-

lated from the sediment (see page 109 of this report). For these calcula-

tions, microalgal organic matter concentration = 166.98 x chlorophyll a con-

centration. Animal organic matter was determined by harvesting in April. The

concentration of detritus was calculated by the subtraction of microalgal and

animal organic matter from the total. The results of these calculations indi-

cated that detritus was the major component of organic matter in these sedi-

ments.

3. Chlorophyll a Concentration

An analysis of variance indicated that there were significant differences

(P < .01) in the chlorophyll a in the top cm of sediment associated with the

effects of sediment type, tidal height, time, and a two-way interaction

between sediment type and tidal height (Appendix I, Figs 16 and 17). The SILT

site had the highest mean chlorophyll a concentration, followed by the FINE

SAND site and the SAND site, respectively (Table 6). Chlorophyll a concentra-

tion was highest during the spring at the SAND and FINE SAND sites; while

during the fall and winter, the concentration was highest at the SILT site.

At the SILT and FINE SAND sites, the 2.0 m transects had the highest concen-

trations of chlorophyll a; while at the SAND site, the highest concentrations

were found for the transect at 1.0 m above MLLW. -64-

Table 5. Ranges of component fractions of organic matter (g m2) in the top cm of sediment, expressed as ash-free dry weight at the intensive study sites. Total organic matter and animal organic matter were measured by direct sampling. Microalgal organic matter was calculated from chlorophyll a concentrationsa. Detrital organic matter was calculated by the subtraction of microalgal and animal organic matter from the total organic matter.

Intensive Study Site Fraction SAND FINE SAND SILT

Total 69.8-268.4 128.0-303.2 256.5-575.0

Microalgae 1.6-23.2 4.3-31.8 4.3-48.0

Animals 1.3-11.8 3.9-8.2 1.1-5.5

Detritus (66.9-233.4) (119.8-263.2) (251.1-521.5)

a) Microalgal organic matter concentration = chlorophyll a concentration x 166.98 (see pages of this report). -65-

MLLW datum SANO !±SE £FINE SAND

o St LI N

at -J

F MA M J A [1 M MONTH

0.5 M f MLLW datum

SAND ±SE LEINE SAND

0 Si LI 'S

at -J

MA M J A M MONTH Figure 16. Concentration of chlorophyll a in the top cm of sediment (CHL a) at intensive study sites from March, 1980, to March, 1981. Sites are SAND, FINE SAND, and SILT at 1.0 m and 0.5 above MLLW. Values are means of six replicates. -66-

2.OM + MLLW datum

SAND I tSE £FINE SAND

OS I LT 1 a N zoo

z-J U

L J I I I I F' M AM .J J A SON DJ FM MONTH

j I I I F I I I I F

I.5M I- MLLW datum .SAND TtSE £ FINE SAND 051 LI 200 a N

iIOO-J U

F MA M J 1J A SON D J FM MONTH

Figure 17. Concentration of chlorophyll a in the top cm of sediment (CHL a) at intensive study sites from March, 1980, to March, 1981. Sites are SAND, FINE SAND, and SILT at 2.0 m and 1.5 m above MLLW. Values are means of six replicates. -67-

Table 6. Concentration of chlorophyll a (mg mg2) in the top cm of sediment in Netarts Bay at the SAND, FINE SAND and SILT sites. Intertidal levels are above MLLW: 0.5 m (1), 1.0 m (2), 1.5 m (3), and 2.0 m (4). Values are site and tidal level means (i) and standard errors (SE) of n observations.

Site SE

SAND 45 46.18 6.02 Level 1 10 79.71 11.59 2 12 66.09 13.24 3 12 25.12 4.83 4 11 16.94 2.81

FINE SAND 42 74.72 7.37 Level 1 6 83.12 20.16 2 12 54.73 13.26 3 12 73.90 12.41 4 12 91.34 3.89

SILT 46 93.70 11.07 Level 1 10 49.50 8.87 2 12 69.16 20.89 3 12 82.86 9.61 4 12 165.91 24.45 The ratio of chlorophyll a concentration in the topciii of sediment to

that at the 4 to 5 cm depth was calculated as an index of mixing of chloro-

phyll a in the sediment. High chlorophyll a ratios resulted from a paucity of

chlorophyll a in the 4 to 5 cm depth sediment, while relatively low values

indicated that chlorophyll a was mixed to the 4 to 5cm depth. There were

significant differences (P < .01) in the chlorophylla ratios associated with

the effects of sediment type, tidal height, anda two-way interaction between

sediment type and time (Appendix I). The SAND site had the highest mean chlorophyll a ratio followed by the SILT and the FINE SAND sites (Table 7).

At all sites, the mean ratio was higher at the transects 2.0 m above MLLW than at the other intertidal tansects.

4. Primary Production

An analysis of variance indicated that there were significant differences

(P < .05) in gross primary production of microalgae associatedwith the effect of time (Appendix I). Maximum gross primary production occurred in the summer when Enteromorpha sporelings were abundant in the sedimentcommunity (Figs. 18 and 19).

When production by Enteromorpha sporelings was included in the calculations, the FINE SAND site had the highestmean rate of gross primary production followed by the SAND site and the SILT site (Table 8). If samples with

Enteromorpha sporelings were excluded, all sediment types had similarmean rates of gross primary production.

Because of the difference in growth form, the contribution of Entero- morpha prolifera to total primary production at the intensive study siteswas estimated by experiments with isolated plants in the laboratory (see details Table 7. Ratio of the chlorophyll a concentration in the top cm of sediment to the concentration at the 4 to 5 cm depth in Netarts Bay at the SAND, FINE SAND and SILT sites. Intertidal levels are above MLLW: 0.5 m (1), 1.0 m (2), 1.5 m (3), and 2.0 (4). Values are site and tidal level means () and standard errors (SE) of n observations.

Site SE

SAND 43 41.73 13.49 Level 1 10 2.25 0.38 2 11 2.40 0.49 3 11 56.13 39.80 4 11 102.53 26.94

FINE SAND 38 9.31 13.86 Level 1 6 1.21 0.21 2 10 3.25 1.81 3 11 1.27 0.22 4 11 27.27 50.23

SILT 43 11.50 17.35

Level 1 10 1.53 0.22 2 11 3.29 4.19 3 11 1.51 0.24 4 11 38.77 69.74 -70-

I I I I I I I I I I I I I

I.OM+ MLLW datum SAND 200- £ FINE SAND oSILT E

C-,

QQI 0 a- w

FM AM J J A MONTH

I I 1 I I I I I I I I I

O.5M +MLLW datum

. SAND

200- tSE £ FINE SAND - oSILT E 0

100- - a- a-

1 I L t i i L I I F M A M J J A SO ND J MONTH Figure 18. Gross primary production (GPP) at intensive study sites from Nay, 1980, to March, 1981. Sites are SAND, FINE SAND, and SILT at 1.0 m and 0.5 m aboveMLLW. Values are means of two replicates from May to September and three replicates from October to March. = z 0 C V d iRON Os \\ iio ddD

T/ 0 0 5w) Sites are SAND, FINE SAND, and Values are means of two replicates fl's D 40 2 w U) U, +1 o o c'J ]NIJ (1 SILI at 1.5 m above MLLW. to from May to September and three replicates from October ONVS from Gross primary production (GPP) at intensive study sites May, 1980, to March, 1981. March. 4 -J It) Figure 19. -72-

Table 8. Intertidal gross primary production (mg C m2 hr') in Netarts Bay at the SAND, FINE SAND and SILT sites. Intertidal levels are above MLLW: 0.5 m (1), 1.0 m (2), 1.5 m (3), and 2.0 (4). Values are site and tidal level means () and standard errors (SE) of n observations assuming a photosynthetic quotient of 1.2.

Site SE

Including Enteromorpha sporelirigs:

SAND 25 37.43 7.94 Level 1 7 35.80 7.26 2 8 62.16 21.28 3 8 23.04 3.38 4 2 8.98 2.34

FINE SAND 22 47.15 13.84 Level 1 4 84.88 65.75 2 8 40.78 15.53 3 8 47.13 19.11 4 2 27.10 14.56

SILT 26 25.06 3.97 Level 1 7 20.07 5.75 2 8 19.34 4.46 3 8 21.42 4.25 4 3 53.70 22.12

Excluding Enteromorpha sporelings:

SAND 23 27.82 3.25

FINE SAND 20 28.05 5.12

SILT 26 25.06 3.97 -73-

in a later section). This alga was conspicuous from June to August at the

SAND and FINE SAND sites, and the maximum biomasses at these sites were 149

and 162 g C m2, respectively (Table 9). Enteromorpha was never present at

the SILT site. The estimated contribution of Enteromorpha to the annual net

primary production at the SAND and FINE SAND sites was 2,712 and 2,886 g

Cm2yr', respectively,an enormous elaboration of organic matter. While

growth of Enteromorpha appeared to be related to seasonal increases in

temperature and day length, it is not possible to rule Out nutrient dynamics

associated with the upwelling phenomenon as a possible cause of the high

production of this alga during the summer months. Its absence at the SILT

site was probably related to the relatively high turbidity and instability of

the sediments at this location.

There were significant differences (P K .01) in community oxygen uptake

associated with the effect of sediment type (Appendix I). The FINE SAND site had the highest mean rate of community oxygen uptake followed by the SILT

site and the SAND site (Table 10). There were also significant differences

(P < .05) in community oxygen uptake associated with the effects of time and a

two-way interaction between sediment type and tidal height (Fig. 20). Maximum

community oxygen uptake in all sediment types was detected in the summer when

temperature was relatively high.

The ratio of daily gross primary production to daily community oxygen uptake was calculated using data from the primary production estimates of

this study. Daylength at the sediment surface ranged from 6 h in December to

12.2 h in June. This ratio is an index of benthic autotrophy relative to

heterotrophic activity in the sediment community, the higher values indicating -74-

Table 9. Maximum biomass (BIOMASS) of Enteromorpha prolifera, and the annual

contribution of Enteromorpha prolifera to the rates of gross

primary production (GPP), respiration (RESP) and net primary

production NPP), at the SAND and FINE SAND sites in Netarts Bay.

SAND FINE SAND

BIOMASS (g Cm2) 149 162

GPP (g c m2yr1) 3691 3880

RESP (g C m2yr) 907 994

NPP (g C m2yr') 2712 2886 -75-

Table 10. Carbon equivalent of the intertidal community oxygen uptake (mg c m2 hr') in Netarts Bay at the SAND, FINE SAND and SILT sites. Intertidal levels are above MLLW: 0.5 m (1), 1.0 m (2), 1.5 m (3), and 2.0 (4). Values are site and tidal level means () and standard errors (SE) of n observations assuming a respiratory

quotient of 1.0.

Site SE

SAND 25 11.49 1.86

Level 1 7 13.81 2.31 2 8 17.72 3.57 3 8 6.57 2.33 4 2 1.20 1.20

FINE SAND 22 27.16 3.87

Level 1 4 21.18 6.50 2 8 28.98 6.49 3 8 29.05 8.12 4 2 24.63 5.76

SILT 26 18.34 2.37

Level 1 7 22.74 5.56 2 8 17.23 3.48 3 8 20.67 4.22 4 3 7.93 2.33 -76-

I.5M+ MLLW £ £ SAND 50 £ FINE SAND /o \/ 0 SILT

I.OM+MLLW 0 A_A )5O \ 0 !° 0 0.5M +MLLW

0 50

0 ------_ - a

M J A S U N 0 J F M MONTH Figure 20. Community oxygen uptake (OUPTK) at intensive study sites from May, 1980 to March, 1981. Sites are SAND, FINE SAND, and SILT at 1.5 in, 1.0inand 0.5inabove MLLW. Values are single observations from May, 1980, to September, 1980, and means of three replicates from October, 1980, to March, 1981. -77-

relatively greater gross primary production. The SAND site had the highest mean ratio followed by the FINE SAND site and the SILT site(Table 11). There were no significant differences in the mean values of this ratioassociated with effects of either tidal height or time.

The ratio of gross primary production to chlorophyll a concentration in the top cm of sediment was calculated to give a measure of production per unit of microalgal chlorophyll a (Table 12). The mean value for this ratio ranged from 0.51 (SILT site) to 1.21 (FINE SAND site). There were no significant differences among the mean values for this ratio associated with the effects of sediment type, tidal height, or time (Appendix I).

5. Relationships Among Variables

Table 13 gives Pearson product-moment coefficients of correlation (r) for selected pairs of physical and biological variables monitored in Netarts Bay during the study. While such correlations are often unrelated to causation, they do provide a useful initial insight into the structure of the data. At the 5% significance level, 13 r values were judged to be significantly different from zero. However, relationships suggested by values below 0.50 are undoubtedly weak or spurious, so such tests are essentially useless for our purpose. Here, we adopt the view that the coefficients are simply an indication of the degree to which variables covary and indicate some of the higher values that are relevant for data interpretation.

In general, correlations among the variables in Table 13 were low; and only 4 out of 28 values were greater than 0.50. Among physical variables, daylength and temperature had the highest r value (0.74). Mean sediment size in phi units was the only physical variable that had a correlation greater

than 0.50 with a biological variable; its correlation with organic matter concentration in the top cm of sediment was 0.82. Among biological variables, -78-

Table 11. Ratio of daily gross primary production to daily community oxygen uptake in Netarts Bay at the SAND, FINE SAND and SILT sites. Intertidal levels are above MLLW: 0.5 m (1), 1.0 m (2), 1.5 m (3), and 2.0 (4). Values are site and tidal level means (T) and standard errors (SE) of n observations.

Site SE

SAND 25 3.19 1.69 Level 1 7 1.13 0.17 2 8 1.56 0.40 3 8 7.24 5,19 4 2 0.71 0.33

FINE SAND 22 0.77 0.20 Level 1 4 1.42 0.86 2 8 0.65 0.26 3 8 0.69 0.21 4 2 0.27 0.21

SILT 26 0.75 0.21

Level 1 7 0.44 0.14 2 8 0.48 0.07 3 8 0.51 0.09 4 3 2.86 1.47 -79-

Table 12. Ratio of gross primary production to chlorophyll a concentration (mg C hr1 mg) in Netarts Bay at the SAND, FINE SAND and SILT sites. Intertidal levels are above MLLW: 0.5 m (1), 1.0 m (2), 1.5 m (3), and 2.0 (4). Values are site and tidal level means () and standard errors (SE) of n observations.

Site SE

SAND 25 0.95 0.15

Level 1 7 0.55 0.10 2 8 1.16 0.41 3 8 1.17 0.17 4 2 0.67 0.36

FINE SAND 22 1.21 0.45

Level 1 4 2.13 1.62 2 8 1.13 0.62 3 8 1.30 0.83 4 2 0.41 0.23

SILT 26 0.51 0.08 Level1 7 0.57 0.12 2 8 0.54 0.15 3 8 0.54 0.16 4 3 0.31 0.08 Table 13. pairsA matrix of physicalof Pearson and product-moment biological variables coefficients monitored of correlation in Netarts (r)Bay forfrom selected March sedimentcores(AFDW),cm1980 of to (CHLR),sedinent March chlorophyllmean 1981.grainof (CHLS), gross sizea concentrationprimaryorganic (SEDMEAN), productionmatter daylengthin concentrationthe (GPP), top (DAYL), cm community of in sedimentand the water oxygentop in cmtemperature theuptakeof respirometersediment (OUPTK), (TEMP). The variables are chlorophyll a concentration in the top CHLS CHLS - AFDW 0.40* 0.81*CHLR 0.46* GPP 0.08OUPTK SEDMEAN 0.29* -0.07 DAYL -0.07 TEMP CHLRAFDW 0.18 - -0.05 0.63* 0.080.13 -0.10 0.82* _0.31* 0.12 _0.26* 0.15 GPPOUPTK - 0.42* _0.24* 0.15 0.38*0.01 -0.01 0.45* TEMPDAYLSE DME AN - 0.11 - 0.74*0.07 The * symbol indicates that the value is significantly different from zero at the 57. level. there was a relatively high correlation between chlorophyll a concentration in the top cm of sediment and that in the top cm of cores used in the respirometer (0.81) and between gross primary production and chlorophyll a concentration in the respirometer cores (0.63).

Because of the time and problems associated with the direct measurement of gross primary production (GPP) in the field, it was of interest to examine other variables as potential predictors of GPP. From the correlation analysis, the only likely predictor was the chlorophyll a concentration in the top cm of sediment (CHLR). The linear and curvilinear equations corresponding to each intensive study site and to the pooled data for all sites are presented in Table 14. These equations represent the prediction of GPP for microalgal assemblages in the absence of Enteromorpha sporelings. For this relationship, the highest R2 value was found for the SILT site (0.55), but all values were relatively low. However, the relationship derived from the pooled data from Netarts Bay resembles the linear equation obtained for data from five intensive study sites on the Columbia River estuary, although regression analyses indicated that the curve fit was more satisfactory for the Columbia

River data (Table 14). Since the y-intercept was relatively high for the

Netarts Bay data (9.57 mg C hr1), it was desirable to examine several other models. From a biological perspective, it makes sense to force the line through the origin and investigate a linear and curvilinear relationship. In

Table 14, such models are presented for both the Netarts Bay and Columbia

River data; and residual mean squares are provided as a basis for within estuary comparisons. Based on data from this study, the most satisfactory -82-

Table 14. Linear and curvilinear equations expressing gross primary production (GPP) as a

function of the concentration of chlorophyll a (CHLR)in the top cm of sediment.

Units for GPP and CHLR are mg hr and mg, respectively. The models correspond to

the SAND, FINE SAND and SILT sites at Netarts Bay, and to the pooled data for

Netar-t-s Bay and the Columbia River Estuary.

Residual Mean Location R2 Square

Netarts Bay

SAND GPP =10.47+ 0.41CHLR 23 0.50

FINE SAND GPP =-3.56+ 0.65CHLR 20 0.37

SILT GPP = 7.60+ 0.30CHLR 26 0.55

Pooled Data:

1. GPP = 9.57+ 0.35CHLR 69 0.40 220.7

2. GPP = 8.22+ 0.40CHLR -0.00025 (CHLR)2 69 0.40 223.6

3. GPP= 0.48CHLR 69 - 246.6

4. GPP = 0.61CHLR -0.0012(CHLR)2 69 - 228.9

Columbia River

Pooled Data:

1. GPP= 2.90+ 0.30CHIR 56 0.73 474.0

2. GPP =-2.80+ 0.38CHLR -0.00017 (CHLR)2 56 0.74 464.2

3. GPP= 0.31 HLR 56 - 468.8

4. GPP = 0.36CHLR -0.00014 (CHLR)2 56 - 457.5 model for predicting GPP in Netarts Bay is model 4, i.e., the quadratic function forced through the origin:

GPP=0.61 CHLR-0.0012 (CHLR)2

In the case of the Columbia River estuary, either the linear are quadratic functions are probably satisfactory. In summary, chlorophyll a concentration in the top cm of sediment is a relatively good predictor of GPP in the Columbia

River estuary, a system under the influence of high freshwater discharge; while in Netarts Bay, a more marine system, such predictions are less reliable. The disruptive influence of burrowing marine animals is one obvious explanation for the difficulty in predicting GPP at certain locations in Netarts Bay.

Mean sediment grain size was a relatively good predictor of the organic matter concentration in the top cm of sediment. This relationship was improved slightly by adding the chlorophyll a concentration in the top cm of sediment to the model. The linear equations are

AFDW=-89.54 109.21 SEDMEAN and

AFDW=-123.50 104.41 SEDMEAN+0.92 CHLR, where AFOW is the organic matter concentration (g m2), SEDNEAN is the grain size in phi units, and CHLR is the chlorophyll a concentration (mgm2); corresponding R2 values are 0.67 and 0.73.

Interpretation of Autotrophic Patterns

Sediment characteristics of Netarts Bay remained relatively constant at the intensive study sites during the study period. This would be expected for a bay which does not receive large seasonal inputs of terrestrial runoff and -84-

associated silt. Sediment type ranged from medium sand to coarse silt and

corresponded to sources of material for the bay. Changes in sediment height in

Netarts Bay during the spring and summer corresponded to areas with conspicuous

wave activity and strong water currents. Most of these areas were on the western side of the bay. Frostick and McCave (1979) found an association

between algal growth and sediment accretion, with subsequent erosion of sediment when the algae died. Wave action also strongly influenced the amount of

erosion. No simple association between algal growth and sediment accretion was

found in Netarts Bay, although it was observed that Enteromorpha did trap large

amounts of sediment. Mucus secreted by diatoms, blue-green algae, and

flagellates also helps to bind sediment (Coles, 1979; Frostick and McCave,

1979).

Organic matter concentration in the top cm of sediment was related to mean grain size, with the highest mean concentration found at the SILT site and the

lowest at the SMD site. Seasonal changes in organic matter corresponded to

changes in plant, animal, and detrital biomass. Similar seasonal changes were observed in other estuaries (e.g., Edwards, 1978; Cadée and Hegeman, 1977). In

Netarts Bay, there were no large differences between organic matter concentration in the top cm of sediment and that at the 4 to 5 cm depth. However, differences in the chemical composition of organic matter may vary with depth as sediment is disturbed by benthic animals (Johnson, 1977).

In Netarts Bay, large amounts of Enteromorpha and Zostera were transported to the strand line at 1.0 m above MLLW during the summer and the fall. This accumulation occurred primarily on the shores furthest away from the mouth of the bay. The material was fragmented and incorporated as organic matter into the intertidal sediments rapidly during these seasons. Josselyn and Mathison

(1980) also observed large biomasses of macrophyte litter at the strand line in

Great Bay, New Hampshire, reporting that seaweeds composed 35 to 85% of the material thoughout the year.

In general, chlorophyll a concentration in the top cm of sediment was

higher at the SILT site than at the SAND site. A similar pattern was found for

other estuaries (Cad&e and Hegeman, 1977; Coles, 1979; Colijn and Dijkema,

1981). In Netarts Bay, there was a relatively high concentration of chloro-

phyll a in the top cm of sediment at 2.0 m above MLLW, a pattern apparently

related to abundant soil moisture and the lack of benthic animals. Blue-green

algae also were common at this location. At lower tidal heights at theSILT

site (< 1.5 m aboveNLLW),the chlorophyll a concentration in the top cm of

sediment was the same or lower than that at the FINE SAND and SAND sites. At

these tidal heights, the flora was dominated by epipelic and epipsammic dia-

toms. Chlorophyll a concentration in the top cm of sediment at the FINE SAND

site at 2.0 in aboveMLLWwas maximum during the summer, a maximum that was

probably related to high moisture, abundant blue-green algae, and the lack of

benthic animals. The SAND site had low chlorophyll a concentrations, a pattern

associated with a wave-exposed beach which lacked moisture during exposure to

the air.

Animal holes in the sediment, made primarily by Callianassa, were abundant

at all study sites at the tidal height of 1.5 m aboveMLLW. These perforations

resulted in a decrease in chlorophyll a concentration in the top cm of sediment

expressed on an areal basis. Animal holes were more abundant at theSILTsite

than at the FINE SAND and SAND sites at tidal heights less than 1.5 m above

MLLW. However, at these tidal heights, benthic animals were abundant at all

sites. Other factors limiting microalgal biomass at the study sites were scouring and turbidity associated with water movements and interactions with

Enteromorpha and Zostera. At times, both Enteromorpha and Zostera shaded sediment-associated microalgal populations and also competed with microalgac for space. Large "ropes" of Enteromorpha became entangled in Zostera beds and trapped large amounts of sediment, an event which inhibited the growth of microalgae. Similar factors apparently limited the production of microalgae during long-term studies in salt marshes (Gallagher and Daiber, 1974; Van

Raalte, et al., 1976).

The ratio of the chlorophyll a concentration in the top cm of sediment to the concentration at the 4 to 5 cm depth was highest at 2.0 m above MLLW at all three intensive study sites. The 2.0 m transect at the SAND site was a relatively barren beach strand line, while the 2.0 m transects at the FINE SAND and SILT sites had compacted subsurface sediments. At these sites, high ratios apparently were associated with the relative stability of these sediments and the poor survival of subsurface populations of microalgae once they have been buried in compacted subsurface sediment. At transects lower than 2.0 m above

MLLW, animal activity and scouring disturbed the sediment and lowered the chlorophyll a ratio.

The complicated temporal pattern of chlorophyll a distribution apparently was controlled by sediment moisture, animal activity, macrophyte cover, water currents, and wave action. Maximum chlorophyll a concentrations were observed in the spring during a period of relatively high microalgal growth and inactiv- ity of benthic animals and again in the fall during fair weather and a decrease in animal activity. Experiments in the field and in the laboratory have indicated that benthic animals inhib±ted xnicroalgal biomass accumulation signifi- -87-

candy (see later section of this report; Admiraal, 1977d; Coles, 1979). In other estuaries, chlorophyll a concentration near the sediment surface also was related negatively to wave action, water currents, and storm activity

(Pamatmat, 1968; Gallagher and Daiber, 1974; Colijn and Dikjema, 1981).

Maximum gross primary production occurred during the summer at all three intensive study sites. Although large plants of Enteromorpha were excluded from production measurements, sporelings were present in the sediment at the

SAND and FINE SAND sites and presumably contributed significantly to sediment- associated gross primary production in the summer. When Enteromorpha was excluded, the mean hourly rates of gross primary production were similar for all intensive study sites. This is in contrast to the mean chlorophyll a concentration in the top cm of sediment which was highest at the SILT site.

The growth of blue-green algae at 2.0 m above MLLW contributed large biomasses at this site but apparently did not increase the hourly rate of gross primary production.

Other studies have detected a diurnal rhythm of sediment-associated algal

primary production (Pamatmat, 1968; Gallagher and Daiber, 1974). However, no

such rhythms were found during the daylight hours in Netarts Bay. Other factors which were not measured in Netarts Bay, but could possibly explain variation in primary production, include microalgal migration in the sediment, sediment stability, grazing pressure, and import-export of microalgal biomass

(Cadêe and Hegeman, 1974; Pamatmat, 1968).

Light intensity at the intertidal sites was usually 10 to 20% of full

sunlight, an intensity that was above the level necessary for light saturation of photosynthesis (Admiraal, 1977a; Colijn and van Buurt, 1975). Exceptions to this included days of heavy cloud cover and fog and periods of wind-induced wave activity which stirred up the sediments in certain areas.Apparently photosynthesis was controlled by daylength, which was determined by cloud cover, turbidity, water depth, and sunrise-sunset times. The control of photosynthesis in sediment-associated microalgae by daylength also has been observed in other estuaries (Cad&e and Hegeman, 1974; Admiraal and Peletier,

1980; Pamatmat, 1968).

Maximum community oxygen uptake occurred during summer, corresponding to increases in temperature, organic matter concentration and metabolic activity.

A similar pattern was found by Pamatmat (1968) and Duff and Teal (1965). The

FINE SAND site apparently had the most favorable conditions for the growth of animals and plants. Apparently community oxygen uptake was not limited by substrate concentration, as the mean organic matter concentration was highest at the SILT site while the mean oxygen uptake was maximum at the FINE SAND site. Some earlier work indicated that community oxygen uptake may be correlated with organic matter concentration (Pamatmat, 1975; Edwards, 1978).Much of the organic matter at the SILT site was refractory (e.g., twigs and leaves) and derived from terrestrial sources. This material was probably relatively unavailable to the detrital food web. Further research is necessary to partition the various organic components in estuaries into available and unavailable fractions (Johnson, 1977).

The highest ratios of daily gross primary production to daily community oxygen uptake occurred at the SAND site when sporelings and young plants of

Enteromorpha were dominant. The communities at the SILT and the FINE SAND sites had ratios below unity and were, therefore, supported in part by detrital inputs. The ratio of gross primary production to chlorophyll a concentration in the top cm of sediment was relatively constant throughout the year at all three intensive study sites. Chiorophyllides which are included in the measurement of chlorophyll a represent a possible source of variation in the estimated ratio of gross primary production to chlorophyll a concentration, as chlorophyllides are not photosynthetically active pigments (Whitney and Darley,

1979). Moreover, chlorophyll a sampled from the top cm of sediment included microalgal biomass, which was below the level of 1% of the surface light intenstiy (1.3 to 2.6 mm depth) an intensity at which photosynthesis is equal to or below respiratory losses. Also, diurnal rhythms of microalgal migration and photosynthetic activity control the proportion of biomass able to photosynthesize in the surface sediments (Pamatmat, 1968; Taylor, 1964); and seasonal variations in the ratio of carbon to chlorophyll a may account for fluctuations in the ratio of gross primary production to chlorophyll a (Jonge,

1980). Jonge (1980) found a negative correlation between microalgal growth rate and the ratio of carbon to chlorophyll a in estuarine sediment for a period of one year.

A summary of the seasonal and annual production dynamics of microalgal assemblages for the three intensive study sites is presented in Tables 15-17.

Mean microalgal biomass, gross primary production, and community oxygen uptake were calculated from data reported in Figures 16, 17, 18, 19, and 20, assuming a carbon to chlorophyll a ratio of 83.49. Estimates of microalgal respiration, net primary production, and non-algal oxygen uptake are based on the assumption that the hourly rate of algal respiration is 29% of the hourly rate of gross primary production. These assumptions are derived from experimental work 1

Table 15. Estimates of seasonal sediment-associated mean microalgal biomass and microalgal production in Netarts Bay at the SAND site. The tidal levels are 0.5 m, 1.0 m and 1.5 in above MLLW. The variables, expressed as g Cm2, include: mean seasonal and annual biomass of microalgae (BIOMASS), and the seasonal (90 days) and annual (360 days) gross primary production (GPP), microalgal respiration (RESP), net primary production (NPP), community oxygen uptake (OUPTK) and non-algal community oxygen uptake (NON-ALGALOtJPTK). Seasonal rates were calculated by multiplying hourly rates by daylenth and adding daily rates for the appropriate period. The hourly rate of iuicroalgal respiration was assumed to be 29% of the mean hourly rate of gross primary production, a value determined by laboratory experiments (see page in this report). NON-ALGALOUPTKequals OUPTK - RESP.

NON-ALGAL Season BIOMASS GPP RESP NPP OIJPTK OUPTK

Winter 0.5in 4.61 9.54 8.97 0.57 13.44 4.47 1.0in 5.00 18.00 17.17 0.83 9.78 ** 1.5in 1.44 18.60 17.83 0.77 4.54 **

Spring 0.5 m 7.32 30.45 21.30 9.15 32.61 11.31 1.0 m 4.16 32.76 23.30 9.46 45.47 22.17 1.5in 2.30 16.97 11.91 5.06 14.79 2.88

Summer 0.5in 6.50 56.67 34.01 22.66 39.24 5.23 1.0 m 10.20 125.18 75.57 49.61 60.04 ** 1.5in 2.17 21.59 13.03 8.56 23.46 10.43

Fall 0.5 in 6.44 10.88 8.84 2.04 24.35 15.51 1.0in 5.99 28.53 23.81 4.72 24.74 0.93 1.5in 1.85 17.70 16.38 1.32 7.23 **

Annual 0.5 in 6.22 107.54 73.12 34.42 109.64 36.52 1.0in 6.34 204.47 139.85 64.62 140.03 23.10 1.5in 1.94 74.86 59.15 15.71 50.02 13.31

**Indicationof a negative valuefor NON-ALGAL OUPTK;negative valuesare assumed tobe zerofor the estimation ofannual NON-ALGAL OUPTK. -91-

Table 16. Estimates of seasonal sediment-associated mean microalgal biomass and microalgal production in Netarts Bay at the FINE SAND site. The tidal levels are 0.5 m, 1.0 m and 1.5 m above MLLW. Variables and method of calculation are the same as described for Table 15.

NON-ALGAL BIOMASS GPP RESP NPP OUPTK OUPTK

- 9.50 9.00 0.50 13.00 4.00 2.59 15.83 14.77 1.06 26.09 11.32 5.62 11.88 11.01 0.87 26.16 15.15

7.07 21.92 14.06 7.86 45.62 31.56 5.70 28.85 17.73 11.12 54.33 36.60 5.69 29.21 18.77 10.44 58.22 39.45

3.35 101.92 64.50 37.42 36.57 ** 4.88 75.11 44.66 30.45 83.37 38.71 4.95 114.72 67.52 47.20 124.78 57.26

- 18.00 12.00 6.00 21.00 9.00 4.05 25.91 20.09 5.82 84.35 64.26 4.08 6.83 5.51 1.32 15.89 10.38

5.21 151.34 99.56 51.78 116.19 44.56 4.31 145.70 97.25 48.45 248.14 150.89 5.09 162.64 102.81 59.83 225.05 122.24

**Indication of a negative value for NON-ALGAL OUPTK; negative values are assumed to be zero for the estimation of annual NON-ALGAL OUPTK. -92-

Table 17. Estimates of seasonal sediment-associated mean microalgal biomass and aicroalgal production in Netarts Bay at the SILT site. The tidal levels are 0.5 m, 1.0 m and 1.5 m above MLLW. Variable and method of calculation are the same as described for Table 15.

NON-ALGAL BIOMASS GPP RESP NPP OIJPTK OUPTK

5.81 21.46 19.91 1.55 28.85 8.94 14.50 14.20 13.28 0.92 39.66 26.38 5.99 13.65 12.99 0.66 35.16 22.17

4.33 20.60 13.59 7.01 46.98 33.39 4.14 10.88 7.83 3.05 18.15 10.32 5.68 22.86 15.15 7.71 34.44 19.29

2.21 11.32 6.96 4.36 25.84 18.88 2.37 27.63 16.82 10.81 49.25 32.43 3.50 14.32 8.89 5.43 56.96 48.07

3.42 10.64 9.04 1.60 28.79 19.75 3.33 11.02 7.74 3.28 35.96 28.22 4.84 37.08 19.23 17.85 48.06 28.83

3.94 64.02 49.50 14.52 130.46 80.96 6.09 63.73 45.67 18.06 162.83 117.16 5.00 87.91 56.26 31.65 174.62 118.36 -93-

presented in a later section of this report (page 109). In several cases for the SAND and FINE SAND sites, the latter assumption generated a negative value for non-algal oxygen uptake, indicating that microalgal respiration was probably less than the assumed 29%.

In general, the annual rates of gross primary production for the three intensive study sites were well within the range of values reported in the literature for similar habitats (Table 18). Annual gross primary production varied from 74.86 to 204.47 g C m2, both extremes occurring at the SAND site. Maximum microalgal production at the SAND and FINE SAND sites occurred in the summer at tidal levels where sporelings of Enteromorpha were prominent.

Production at the SILT site, where Enteromorpha did not occur, was less variable with the maximum in the fall at 1.5 m above MLLW. This site also exhibited an unusually high mean biomass at 1.0 m above MLLW during the winter when gross primary production was a little below the average for all seasons.

Possible explanations for this inconsistency include: (1) an underestimation of net primary production either by an over estimation of microalgal respiration or by an under estimation of gross primary production; (2) over estimation of microalgal biomass; and (3) microalgal heterotrophy. If rates of gross primary production were markedly higher on exposed sediment than on inundated sediment, our method of measurement could have provided an under- estimate for this variable. Furthermore, heterotrophic growth of sediment- associated microalgae has been reported by Admiraal and Peletier (1979),

Jorgensen et al. (1980), McLean et al. (1981), and Hellebust and Lewiri

(1977). In particular, the work of Admiraal and Peletier (1979) suggest that at least some diatoms associated with fine sediment particles can grow almost as rapidly in the dark as in the-light if a suitable organic substrate is available. The potential of the diatom flora at the SILT site for hetero- -94-

Table 18. Annual rates of benthic microalgal primary production reported in

selected studies.

Rate Study Location (g c m2 yr')

1. Steele & Baird (1968) beach sand 4-9

2. Pomeroy (1959) Georgia salt marsh 200

3. Grontved (1960) Danish fjords 116

4. Cadee & Hegeman (1974) Western Wadden Sea 100

5. Cadee & Hegeman (1977) Western Wadden Sea 29-188

6. Pamatmat (1968) False Bay, San Juan Island 143-226

7. Gallagher & Daiber (1974) Delaware salt marsh 38-99

8. Marshall et al. (1971) Southern New England 100

9. Leach (1970) Northern Scotland mudflat 31

10. Riznyk & Phinney (1972) Yaquina estuary, Oregon: Southbeach (sandy silt) 275-325 Sally's Bend (fine silt) 0-125

Netarts Bay: SAND 129* FINE SAND 153* SILT 72*

*Values represent the mean of values for the three tidal levels (0.5 in,1.0 m, and 1.5inabove MLLW). -95-

trophic growth is unknown. However, the survival of diatoms at 18 cm below the surface of the sediment was reported by Riznyk (1969) for assemblages in

Yaquina Bay, suggesting the possibility of heterotrophic nutrition or a temporary decrease in metabolic rate during burial (McIntyre et al., 1970).

Sources of carbon for non-algal community oxygen uptake probably included both algal and non-algal organic matter. At the SAND, FINE SAND, and SILT sites, 100%, 50%, and 20%, respectively, of the carbon for metabolism could have been derived from microalgae which were available during the entire year.

The remaining carbon was probably supplied by detrital sources from other areas. Other studies have found that detritus is a major source of carbon to estuarine secondary production (Pomeroy et al., 1977; Tenore, 1977). Sources

of detrital carbon to Netarts Bay included Enteromorpha, Zostera , bacteria, fungi, animals, and microalgae. There did not appear to be major contribu- tions of carbon from the marsh because of the limited export of particulate matter from the area. However, studies have shown that some of this high marsh production may be exported as energy to the intertidal habitats in the form of dissolved reduced inorganic sulfur compounds and used chemotrophically by anaerobic organisms while their carbon source is carbon dioxide (Howarth and Teal, 1979).

The contribution to net primary production by Enteromorpha (2800 g c m2 yr1) was a large seasonal input of organic matter into Netarts Bay, especially during the summer and fall. Similar patterns of Enteromorpha growth were observed in other estuaries (Conover, 1958; Nienhuis, 1970). By comparison, microalgae supplied a much smaller proportion of organic matter to the total net primary production of algae in the bay (15 to 65 g C m2 yr). However, this microalgal contribution is also important and represents a

significant source of food which is available throughout the year for benthic

organisms (Baillie and Welsh, 1980; Ribelin and Collier, 1980).

In conclusion, it was evident that microalgal production and biomass in

Netarts Bay was controlled by daylength, temperature, water currents, sediment

moisture and stability, animals, and macrophyte cover.Nutrients did not

appear to limit microalgal growth, apparently because of the abundance of

nutrients in the intertidal sediments (Admiraal, 1977b, d; Cardon, 1981). The

possibility of heterotrophic growth in microalgae was suggested by the

presence of relatively high biomasses of microalgae during periods of low net

primary production in the winter on sediment with high concentrations of

organic matter. The sediment communities at the FINE SAND and SILT sites

obtained at least half or more of their carbon from detrital sources other

than in situ microalgal net primary production. However, Enteromorpha and

Zostera production in the bay provided large supplies of carbon for sediment

community metabolism during the summer and fall.

II. Experimental Studies of Estuarine Benthic Algae

Laboratory studies were designed to investigate relationships between

sediment-associated estuarine algae and physical factors affecting their growth. Specific objectives of this research included: (a) the determination

of the relationship between light intensity and gross photosynthesis in assemblages of sediment-associated algae; (b) the determination of the relationship between temperature and community metabolism in sediment-associated algal assemblages; and (c) the determination of the relationships between biomass and metabolic rates and between biomass and chlorophyll a in populations of epipelic diatoms and macroalgae isolated from sediment-associated -97-

algal assemblages. Experiments corresponding to these objectives were conducted at the Oregon State University Marine Science Center (Newport, Oregon) and nearby tidal flats in Yaqunia Bay. Results from these experiments were used to help understand mechanisms that accounted for the production dynamics of algal assemblages in Netarts Bay and for the estimation of parameters compatible with our structural model of the Algal Primary Production subsystem.

Yaguina

Yaquina Estuary (44°35' N;124°04' W) drains 656 km2 of the Coast

Range (Fig. 21) and is Oregon's fifth largest estuary, with a surface area of

1583 hectares, of which 548 hectares is tideland and 1035 hectares is subtidal. The estuary is subjected to mixed semi-diurnal tides and is classified as a partly-mixed system in the winter and spring and a well-mixed system during the summer and fall. Maximum tidal range is 3.0 m. Mean low water

(MLW) and the mean high water (MHW) are 0.5 m and 2.0 m above the mean lower low water datum (MLLW), respectively. Daylight hours in Newport vary from

8.5 perday in June to 15.5 per day in December. The salinity pattern in the estuary is complex. In the summer, upwelled bottom water from off the Oregon coast combines with insignificant fluviatile input to account for a relatively high salinity (33 to 35 0/00) in Yaquina Bay. In the winter, high freshwater discharge is responsible for salinities as low as 8 0/00 at the Marine Science

Center during low tide. During a complete tidal cycle, 707 of the water in the bay is exchanged with ocean water (Goodwin et al., 1970). During an Figure 21. OCEAN

METERS I Map ofYaquinaBay with arrowsindicatinglocationofintensive study sites. . - -- o YAQLJ/NA BA '- . .. I :. .. -\ % wg S / -98-- Y:: . 5f . . I . - gri' ... :::.. ,, --I . ------I OREGON : : .. OREGON N S.

entire year, water temperature over the sediment flats usually ranges from 6°C to 20°C, and the air temperature over exposed sediment flats ranges from 0°C to 30°C.

Methods

1. Primary Production

Gross primary production and community oxygen uptake were estimated in the laboratory from oxygen measurements in stirred, light and dark chambers designed to hold intact cores of sediment. Between March and September, measurements were made in plexiglas chambers which were 32.5 cm long and had an internal diameter of 14.5 cm (Fig. 22). Each chamber held three intact sediment cores, and these cores were 5 cm deep with a total sediment surface area of 136.08 cm2. The chambers were filled with 5.7 1 of filtered seawater, sealed and placed into a water bath. Water in each chamber was circulated through tygon tubing by a pump. Rates of oxygen evolution were based on measurement periods of 0.5 to 1.0 hour between initial and final readings.

Measurements of dissolved oxygen were made polarographically with an Orbi- sphere salinity-corirected dissolved oxygen system by inserting the oxygen probe into a pot connected to the water circulation line. Between October and

May, estimates of primary production were made in vertical plexiglas chambers described in an earlier section of this report (Fig. 9 and page 44); the methods were essentially the same as those used for the field work.

Rates of net community production or oxygen uptake were measured when either type of chamber was exposed to light or darkened by covering with a sheet of black plastic, respectively. All chambers were preiricubated in the dark for 1.0 hour before incubations under sunlight or incandescent and cool- ALGAE AND SEDIMENT GASKET ATER FLOW 02 PROBE PORT CORERHOLDER CORER PVC RUBBERBUNG 0/ / I PUMP CHAMBER COOLING COILS FJgire 22. DIagram of a plexiglas chamber from measuring oxygen productl9n and uptake of intact PLEXIGLASFOR 02 METABOLISM CHAMBER sediment cores and macroalgae. L...... I IJ! - - - - -r -- - -..--'- _i - -10 1-

white flourescent lamps. Fiberglass screening was used as a neutral density filter to reduce light Intensity when required. Simultaneous incubations of intact sediment cores in both types of chamber indicated that there were no significant differences in primary production attributable to chamber type.

Computations of gross primary production from the oxygen data were made following the same procedure described in an earlier section (see page 47 of this report).

2. Biomass

Microalgal biomass expressed as the concentration of chlorophyll a was determined following the procedure described earlier on page 47 of this report. Macroalgal biomass samples were rinsed in fresh water, weighed, dried at 70°C for 48 hours, and reweighed. Dried samples were ignited in a muffle furnace at 450°C for 24 hours and reweighed to estimate weight of organic matter. The chlorophyll a concentration in macroalgae was determined by the method described for uiicroalgae using approximately 1.0 g wet weight of algae per sample. Carbon content was estimated by multiplying the weight of organic matter by 0.5 (Vollenweider, 1974).

3. Isolation of Diatom Assemblages

Motile, epipelic diatoms were isolated from sediment samples obtained each season during an entire year at intensive study sites in Yaquina Bay using the method of Jonge (1980). Approximately 1000 cm3 of the top cm of sediment were taken from the field and transported to the laboratory. This sediment was mixed with 100 ml of sand-filtered, UV-treated seawater (30 to

33 0/00 salinity) and spread out in a shallow tray, 40 by 50 cm. Clean white quartz sand (125 to 250 m particle diameter) was spread over the sediment in -102-

a 1 nun layer, and three layers of lens tissue were placed over the sand and the sediment. The entire tray was covered with clear plastic and incubated for 36 hours at 15°C and a light intensity of 150 11E m2 sec1

After the incubation period, the lens tissue was lifted off the sediment and mixed with 500 ml of sand-filtered, UV-treated seawater. The mixture was filtered through four layers of 2.5 mm thick foam plastic and then through a

55 m Nitex(u mesh net. The foam plastic filtered Out lens tissue fibers, and the net filtered out small animals. Microscopic inspection confirmed that the light brown-colored suspension contained primarily diatoms and few bacteria or flagellates.

The diatom suspension obtained from the isolation procedure was used to establish relationships between primary production and chlorophyll a, organic matter and chlorophyll a and for measurements of algal respiration. Four replicate 50-nil screw-cap test tubes with a diatom sample were incubated for

0.5 hr at a light intensity of 210 1E m2 sec, a temperature of 14°C, and a salinity of 30 to 33 /oo. Samples also were incubated in the dark for 0.5 hr under similar conditions. Measurements of oxygen concentration and calculation of production were performed as described above. After these measurements, 50 ml samples of diatom suspensions were filtered on pre-ashed glass fiber filters, and chlorophyll a was extracted from four filters in acetone using methods described above. Also, four filters were dried at 70°C for

24 hrs and reweighed to determine dry weight. Then these filters were ignited in a muffle furnace at 450°C for 24 hr and reweighed to determine weight of organic matter. Carbon content was estimated by multiplying the weight of organic matter by 0.5 (Vollenweider, 1974). -103--

4. Physical Variables

Light intensity was measured in -'E m2sec1 photosynthetically active radiation using a Licor® quantum meter. Temperature was measured using bimetal or thermistor probes. A temperature-compensated AOGoldberg® refractometer was used to measure salinity.

Results

1. Experiments With Intact Sediment Cores

The relationship between light intensity and gross primary production of intact sediment cores was investigated in March, August, and September, 1980

(Figure 23). Experiments A, B, and C were conducted outdoors under natural sunlight and fiberglass screens, while experiment F was performed in the laboratory under a fluorescent-incandescent lamp fixture and fiberglass screens. Mean microalgal biomasses, expressed as mg chlorophyll am2 in the top cm of sediment, for two replications (i.e., two respirometers with intact sediment cores) during experiments A, B, C, and F were 134, 199, 213, and 115, respectively.

Data from the experiments suggested that the relationship between light intensity and gross primary production was approximately linear at intensities between zero and250 PE m2 sec, except in experiment C where the linear segment was between zero and about 150 UE m2 sec. The equation for a linear segment from zero to 248'E m2 sec' estimated from data pooled for all four experiments is

GPP = 0.59 I, where GPP is the rate of gross primary production expressed as mg Cm2 hr and I is the light intensity. To estijnate the asymptotic maximum rate (max

4Il p1 text (or curve fitting procedure. fitting curve (or text

B (triangles), C (solid squares), and F (hexagons). F and squares), (solid C (triangles), B

See Intensity. Data correspond to experiments A (open squares), (open A experiments to correspond Data

Figure 23. Figure Relatlonntiip between grons primary production primary grons between Relatlonntiip

and light and

(tE m2 sec') m2 LIGHT INTENSITY

600 200 400 800 0

'I.I.IsJ

[SJ 0

(I) (I)

a- 50

>- a-

0100

a::

0 0

I- 0

z'50

200

tJ -c L. 250 -105-

and the shape of the curves above 248 UE m2 sec, several functions were examined for experiments A, B, and C. The rectangular hyperbola, a function commonly used as a model of photosynthesis-light relationships (Lederman and

Tett, 1981), generated'max estimates of 284 (A), 269 (B), and 172 (C) g

C m hr1, values thatwere inconsistent with the distribution of the data points. This model forced the curve through the origin and, in this case,

A more satisfactory estimate was provided relatively high estimates of 'max obtained by the exponential model: -B2 I GPP=B0-B2e where is equal to 1'max Curves generated by this function for experiments

A, B, and C are plotted in Figure 23. The curves for experiments A and B intersect the linear segment at 267 and 260 3iE m2 sec, while the curve for experiment C does not intersect the linear segment; the latter curve is ter- minated at an intensity of 157 liE m2 sec. Estimates of maxfor experiments A, B, and C from the exponential model are 233, 181, and 136 mg C m2 hr, respectively.

Data in Figure 23 indicate that the onset of light saturation of photosynthesis (1k) occurs between 200 and 400 jE m2 seC'. Here is defined as the light intensity at which extrapolations of the linear segment and the light-saturated region of the GPP-intensity curve intersect (Talling, 1957).

For our purposes, 1k = Pm/O.S9, where maxis estimated by the exponential model and 0.59 is the slope of the linear segment. Therefore, 1k values for experiments A, B, and C are 395, 307, and 231 E m2 sec, respectively.

The effect of temperature on rates of gross primary production and oxygen uptake in assemblages of sediment-associated microalgae was investigated in -106-

May, 1981 (Tables 19 and 20). Each experiment consisted of three replications, i.e., three respirometers with intact sediment cores; and sediment cores for each replication were collected from two intertidal levels in

Yaquina Bay: 1.9 m and 1.0 m above MLLW. The mean chlorophyll a concentration in the top cm of the sediment cores was 151 mg m2 and the corresponding standard deviation was 21.7 mg m2.Water temperature in Yaquina Bay during April and May varied between 11° and 14°C. The temperature range under investigation was from 7°C to 17°C, a range between the maximum and minimum annual values recorded for the bay. The light intensity during the primary production measurements was 600 .iE if2 sec, an intensity well above the values determined during the experiments illustrated in Figure 23. Data are reported as hourly rates and Q10 values, where 10/(t -t ) 10-r2r1 The temperature coefficient (Q10) is a multiplier that predicts the rate for a

10°C change in temperature, t2 and t1 are the upper and lower temperatures of the range under consideration, and r2 and r1 are metabolic rates corresponding to t2 and t1, respectively.

Rates of gross primary production varied betwen 24.68 and 252.30 mg Cif2 hr depending on temperature, experiment number, and replication (Table

19). Therefore, estimates of Q10 at light saturation were based on a set of samples representing a wide range of photosynthetic capacities. The mean Q10 value for three replications of each of four experiments (n = 12) was 2.05 with a standard error of 0.15. There was no significant correlation between the mean rate of gross primary production for a particular replication and its -107- Table 19. The rate of gross primary production (mg C m2 hr') at different water temperatures and corresponding Q10 intensityexpressedvalues for as wasintact mg 600 chlorophyll sedimentliE m2 sec. coresa n12. Data from corresponds Yaquina Bay. to samples from 1.9 m and 1.0 m above MLLW. Temperature range was from 7°C to 17°C; salinity was 30 0/00, and light Mean biomass for three replications at each temperature is Experiment TidalLevel BiomassMean Temperature GPP Q10 GPPReplication 2 Q10 GPP Q10 GPP Mean Q10 +1.1 152.52137.51 17.0 8.0 107.72 76.54 1.46 147.41107.72 1.42 164.42113.39 1.51 139.85 99.22 1.46 +1.1 181.65154.02 14.0 7.0 246.63119.06 2.83 249.47141.74 2.24 252.30175.76 1.68 249.47145.52 2.25 +1.9 151.92127.00 16.0 7.0 85.0436.85 2.53 102.05 48.19 2.30 113.39 51.03 2.43 100.1645.36 2.45 +1.9 162.13144.12 15.0 8.0 49.3524.68 2.69 52.4333.93 1.86 55.5240.10 1.59 52.4332.90 2.04 -108- Table 20. The30eachand 0/00. correspondingrate temperature of oxygen IsQio uptake expressed values (OUPTK) for as Intact mgexpressed chlorophyll sediment as carbon acores m2. equivalentfrom Yaqulna (mg Bay. C m Data corresponds to samples from 1.9 m and 1.0 m above MLLW. Temperature range was from 7°C to 17°C; salinity was Mean biomasshr') for at three different replications water temperatures at Replica t io n Experiment LevelTidal Bloma ssMean Temperature OUPTK 1 Q1 0 0 UPTK 2 Q1 OUPTK 3 Q 10 OUPTK Mean Q10 I +1.1 137.51152.52 17.0 8.0 40.8817.04 2 .64 44.2920.44 2.36 6137.48 .33 1.73 48.8424.99 2.24 +1.1 181154.02 .65 14.0 7.0 54.5230.67 2.27 54.5237.48 1 .71 61.3340.89 1.78 56.7936.35 1.92 3 +1.9 151127 .92.00 16.0 7.0 23.85 6.81 4.03 27.26 6.81 4.67 30.67 6.81 5.32 27.26 6.81 4.67 4 +1.9 162.13144.12 15.0 8.0 25.9511.12 3.36 29.6622.24 1.15 37.0729.66 1.38 21 30.89.01 1.96 -109-

corresponding Q10 value (r = -0.08 with 10 d.f.). However, the mean Q10 value was significantly higher for samples obtained at + 1.9 m aboveMLLWthan for samples obtained at + 1.0 m aboveMLLW(t = 22.27 with 10 d.f.); the corresponding means were 2.23 and 1.87, respectively.

Temperature coefficients for rates of oxygen uptake in the dark were more variable than values associated changes in the rate of gross primary production (Table 20). The mean Qç value for measurements of oxygen uptake (n =

12) was 2.70 with a standard error of 0.39. There was a significant negative correlation between the mean uptake rate for a replication and the corresponding Qi value (r = -0.72 with 10 d.f.). Also, the mean Q10 value for samples from 1.9 m aboveMLLWwas significantly higher than the mean for samples from 1.0 aboveMLLWCt = 10.04 with 10 d.f.); these means were 3.32 and 2.08, respectively.

2. Experiments With Isolated Epipelic Diatoms

The procedure for isolating living epipelic diatoms from sediment samples provided the opportunity to estimate some useful ratios for sediment-associated microalgal assemblages. The ratios of interest were (1) biomass as ash- free dry weight to chlorophyll a(AFDW/CHLOR);(2) gross primary production to chlorophyll a(GPP/CHLOR);and (3) net primary production to gross primary production in the light (NPP/GPP). The ratioAFDW/CHLORprovided a basis for estimating autotrophic biomass in sediment-associated microalgal assemblages when such assemblages consist primarily of diatoms, while NPP/GPP provided

corresponding estimates of respiratory losses.

Results of experiments with isolated epipelic diatoms are presented in

Tables 21-23. AFDW/CHLOR values varied from 107.55 to 254.98 with a mean for -110-

all experiments of 166.98 and a standard error of 8.20 (Table 21). The lowest mean value for a particular experiment was obtained during January, and the most variation among replications occurred in an experiment conducted in

February. Estimates of GPP for the calculation of GPP/CHLOR and NPP/GPP were based on the assumption that respiration in the dark was equal to respiration in the light. GPP/CHLOR values were considerably higher than values obtained for intact sediment cores (Tables 14 and 22) and were more similar to values reported for algal cultures and natural populations of phytoplankton (See

Table 17 in Parsons et al., 1977). The mean value for seven experiments

(28 replications) was 5.12, and the associated standard error was 0.40. The mean ratio of net primary production to gross primary production for seven experiments was 0.71 (Table 23) indicating that respiration was approximately

29% of GPP for an equivalent period of time.

3. Experiments With Isolated Macroalgae.

Rates of primary production and biomass for sediment-associated macroalgae are summarized in Table 24. Three species of algae were studied:

Enteromorpha prolifera (Mull.) J.Ag.; Ulva expansa (Setch.) S.&G.; and

Gracilaria verrucosa (Huds.) Papenf. Rates of respiration per gram dry weight were similar for all three species, while rates of gross and net primary production were higher for Enteromorpha and Ulva than for Gracilaria.Maximum biomass at the intensive study sites and chlorophyll a concentration per gram dry weight were much higher for the green algae.However, gross primary production per mg chlorophyll a was higher for Gracilaria than for Ulva and

Enteromorpha. Mean net primary production for Enteromorpha, Ulva, and

Gracilaria, calculated from the specific rate and biomass estimates was 4.7, -111-

Table 21. Ratio of blomass expressed as ash-free dry weight to chlorophyll a concen-

tration in assemblages of epipelic diatoms isolated by the lens paper

method from intertidal sediment samples from Yaquina Bay. Data include

the ratio for each replication and the mean ratio and standard error for

each experiment.

Date of Replication Experiment s.E.

10/1/80 209.96 192.85 177.42 177.25 189.37 7.78

10/28/80 176.22 189.86 186.16 189.11 185.34 3.14

1/21/80 107.55 132.44 135.25 122.76 124.50 6.25

2/26/80 113.51 254.98 161.70 126.22 164.10 31.96

4/2/80 108.20 190.17 143.23 168.84 171.62 10.51

Pooled 166.98 8.20 -112-

Table 22. Ratio of gross primary production (mg C hr) to chlorophyll a (mg) in assemblages of epipelic diatoms isolated by the lens paper method from intertidal sediment samples from Yaquina Bay. The light intensity, temperature, and salinity during the experiments were 21O1 E m2 sec, 14°C, and from 30 to 33 0/00, respectively. Data include the ratio for each replication and the mean ratio and standard error for each experi- men t.

Date of Replication

Experiment S.E.

2/26/81 5.58 5.47 6.91 6.48 6.11 0.35

3/14/81 2.29 2.43 2.43 3.54 2.67 0.29

4/2/81 3.07 3.49 2.86 3.18 3.15 0.13

5/9/81 3.13 2.23 2.78 4.17 3.08 0.41

10/30/81 10.49 7.37 6.44 7.01 7.83 0.91

11/9/81 6.91 6.40 6.43 6.92 6.67 0.14

12/11/81 6.27 5.57 6.10 7.46 6.35 0.40

Pooled 5.12 0.40 -113-

Table 23. Ratio of net primary production to gross primary production in assemblages of epipelic diatomsisolatedby thelens paper methodfrom intertidal sediment samples fromYaquinaBay. The light intensity,temperature, and salinity during theexperimentswere210U E m2 sec1,14°C, and from 30- 33 0/00, respectively. Dataincludethe ratio for eachreplicationand the mean ratio andstandard errorforeach experiment.

Date of Replication Experiment S.E.

2/26/81 0.86 0.86 0.87 0.86 0.86 0.00

3/14/81 0.41 0.50 0.50 0.69 0.53 0.06

4/2/81 0.88 0.87 0.88 0.86 0.87 0.00

5/9/81 0.30 0.40 0.55 0.30 0.39 0.06

10/30/81 0.80 0.73 0.81 0.82 0.79 0.02

11/9/81 0.81 0.76 0.71 0.74 0.76 0.02

12/11/81 0.82 0.76 0.79 0.74 0.78 0.02

Pooled 0.71 0.03 -114-

Table 24. Biomass and rates of primary production for sediment-associated macroalgae in Yaquina Bay. Species include: Enteromorpha prolif era, Ulva expansa, and Gracilaria verrucosa. Measurements were made at 13° to 15°C, a salinity of 30 0/00, and a light intensity of 600E m2 sec. Variables include: respiration (RESP), net primary production (NPP), gross primary production (GPP), dry weight(DW),ash-free dry weight(AFDW),carbon (C) calculated asAFDWx 0.5, and chlorophyll a concentration(CHLR). Values are expressed as means of six replications with corresponding standard errors (SE). Biomass values correspond to the period of maximum standing crop which was from June to August for Enteroinorpha, July to September for Ulva, and March to May for Gracilaria.

Enteroinorpha Ulva Gracilaria

Variable S.E. S.E. S.E.

1. RESP (mg C g DW'hr) 1.4 0.3 1.4 0.4 1.0 0.2

2. NPP (mg C g DW'hr') 9.5 1.2 10.3 0.4 7.6 1.2

3. GPP (mg C gDW1hr) 10.9 0.9 11.7 0.8 8.5 1.1

4. DW(g m2) 497.8 69.6 411.9 65.9 19.5 2.2

5. AFDW (gf2) 323.6 48.5 267.7 25.2 13.1 2.2

6. C (gif2) 161.8 24.3 133.9 8.03 6.5 1.1

7. CHLR/DW(mg g) 2.5 0.0 1.9 0.0 0.9 0.1

8. CPP/CHLR(mg hr'mg) 4.4 0.4 6.2 0.5 9.4 0.9

9. Daylight NPP/GPP 0.87 0.04 0.88 0.02 0.88 0.03

*Npp(g C m2 hr) 4.7 4.2 0.2

*Estimated from 2 and 6 -115-

4.2, and 0.2 g C m2 h, respectively. Light saturation of macroalgal photo-

synthesis was similar to that for rnicroalgae (200 to 400-IE m2 sec). No inhibition of photosynthesis was noted at full sunlight (1500 to 2000 1E m2

sec1), althoughsome bleaching of thalli was observed in the field when macroalgae were exposed to full sunlight at low tide.

Interpretation of Experiments

Results of the experiments relating gross primary production to light intensity indicated that the sediment associated microalgal assmeblages reached their light-saturated rate at an intensity between 10% and 20% of the intensity at full sunlight (1500 to 2000 1E m2 sec). Similar results were reported by Admiraal (1977) and Colijn and van Buurt (1975). No inhibition of photosynthesis was observed at 950liE m2 sec, suggesting that the assemblages, which consisted primarily of motile epipelic diatoms, were able to make vertical adjustments in the sediment that optimized growth and survival.

Since the light intensity is about 1% of the surface intensity at a depth of

2 mm in the sediment (see page 58 of this report), such vertical adjustments in position require relatively little time. For example, a diatom moving at mean speed of 10 In sec', a reasonable estimate for motile forms (Harper,

1977), can descend from the sediment surface to a depth of two 2 mm in about

3 mm.

Light-saturated rates of primary production for experiments A, B, and C apparently were not determined by the chlorophyll a concentration in the top cm of sediment, as the lowest rate (experiment C) was found for the assemblage with the highest chlorophyll concentration. The chlorophyll a concentrations in the experimental sediment cores were relatively high during all three -116-

experiments, considerably higher than mean concentrations found between 0.5 m

and 1.5 in above NLLW at the three intensive study sites at Netarts Bay (Table

6). Therefore, it is possible that microalgal assemblages in tidal flats

under the influence of marine water can realize their maximum productive

capacity at chlorophyll a concentrations below 100 mg ui2. Also, there is a

strong possibility that the measurable chlorophyll in the top cm of sediment

is not all involved in the photosynthetic process, either because of burial

below the 2 mm level or because the analytical method fails to discriminate

between chlorophyll a and chlorophyllides. A recent investigation of benthic

autotrophy in the Columbia River estuary revealed a linear relationship

between gross primary production and chlorophyll a concentration in the top

cm up to concentrations as high as 400 mg ni2 (Table 14). However, the max-

imum rate of gross primary production measured during the Columbia River study

was 199 mg C m2 hr', a value less than the estimated value for exper-

iment A. These data indicate that sediment chlorophyll in the lower Columbia

River is less active in the photosynthetic process than that in Netarts Bayor

Yaquina Bay. There is no clear explanation for this difference, although the

study areas on the Columbia River are subjected to severe physical stress

during periods of high freshwater discharge.

tn general, an increase in temperature stimulated community oxygen uptake more than gross primary production. Similar observations were reported by

Duff and Teal (1965), Pamatmat (1968), and Davies (1975). Phinney and

Mclntire (1965) found that temperature can affect the rate of photosynthesis

at light saturation in lotic periphyton assemblages. The Q10 value of 2 reported for the lotic assemblages was remarkably similar to the mean value of

2.05 obtained for the sediment-associated assemblages from Yaquina Bay.How- -117-

ever, it is doubtful that temperature fluctuations accounted for a change in

maxfrom 136 mg C m2 hr (experiment C) to 233 mg C m2 hr (experiment

A). Therefore, differences in 1'max were apparently related to differences in the physiological state of the experimental assemblages, although mechanisms responsible for this physiological variation are unclear.

From the experimental work and observations at Netarts Bay and the

for sediment-associated diatom assemblages under Columbia River Estuary, max the most favorable conditions for growth, i.e., optimum resources, temperature, and biomass, must be in the neighborhood of 240 mg C m2 hr1. If the assemblages are exposed to an average input of light energy equivalent to

10 hr day above the saturation intensity for photosynthesis, the upper limit of gross primary production, i.e., the autotrophic potential for such communities, is approximately 900 g C m2 yr'. Mclntire (1973) estimated that max under optimal conditions for lotic periphyton was about 1 g 02 m2 hr.

Assuming a P.Q. of 1.2, the autotrophic potential for epilithic periphyton in streams is estimated to be 1,150 g C m2 yr. Values reported in the literature for sediment associated microalgal assemblages are usually about

25% or less of the estimated upper limit of 900 g C m2 yr (Table 18).

Therefore, factors such as temperature, turbidity, sediment instability, animal activity, toxic metabolic byproducts, and nutrient limitation apparently prevent these assemblages from realizing their autotrophic potential.

However, in the case of Yaquina Bay, some recent experiments indicate that sediment associated microalgal assemblages are not nutrient limited (Cardon,

1981).

The mean ratio of gross primary production to the concentration of chlorophyll a in the top cm of sediment for the experimental work with intact cores was 0.66 mg C hr mg. This value was similar to the mean value found -118-

for the pooled field data from Netarts Bay (0.48 mg C hr' mg) and to values found in other studies of sediment-associated microalgae (Colijn and van

Buurt, 1975; Admiraal, 1977). The mean value for this ratio for pooled samples from the Columbia River estuary was 0.31 m C hr mg' (Table 14), indicating that the relationship between primary production and chlorophyll concentration is variable and may be an unreliable basis for predicting production dynamics in a particular estuary.

Gross primary production of diatom assemblages isolated from the sediment was very similar to the gross primary production reported for phytoplankton assemblages (Steeman Nielsen and Hansen, 1959), indicating physiological similarities in photosynthesis between benthic and pelagic diatoms. Similar values have been obtained for berithic diatoms grown in liquid pure culture

(Admiraal, 1977a). Apparently, conditions in the sediment prevent epipelic diatoms from attaining photosynthetic rates that they can realize isolated in a liquid suspension. Hourly net primary production averaged 71% of hourly gross primary production for isolated diatoms, assuming the respiratory rates in the light and dark were equal. Colijn and van Buurt (1975) measured net primary production of isolated diatoms, but their measurements did not include an estimate of respiration. Such information is useful in field studies when it is desirable to partition algal respiration from community oxygen uptake, as community oxygen uptake includes microbial, plant, and animal respiration as well as chemical oxygen uptake.

Unfortunately, there is no satisfactory method for determining the ratio of net primary production to gross primary production in assemblages of sediment-associated microalgae. Our tentative estimate of 71% from suspensions of isolated epipelic diatoms may have been affected by bacterial contamination and the change from sediment contact to conditions in a culture vessel. -119-

Although bacteria were not observed by direct microscopic examination of the isolated suspensions, they were probably responsible for some oxygen uptake during the experiments. Moreover, the availability of oxygen is undoubtedly variable among the different sediment types and between a given sediment type and conditions in the culture vessel, differences that limit the use of the laboratory results for indirect estimates of net primary production and algal respiration in the field. Pomeroy (1959) also estimated microalgal respiration by an indirect method and found net primary production to be not less than 90% of gross primary production. In our experiments with macroalgae, rates of respiration were about 15% of the rate of gross photosynthesis for an equivalent period of time. Estimates for macroalgae were probably more reliable than those for the microalgal assemblages, as the ratio of macroalgal biomass to the biomass of bacterial contaminants was relatively large and experimental conditions for the macroalgae were more compatible with the natural conditions in the estuary than the environment of a culture vessel.

The measurement of the ratio of biomass (ash-free dryweight) to chioro- phyll a concentration in isolated assemblages of epipelic diatoms provided a basis for estimating autotrophic biomass in sediment-associated microalgal assemblages when diatoms dominated the flora. Assuming one-half of the biomass was carbon, the mean value for this ratio was approximately 84 mg

C mg chlorophyll a. Jonge (1980) reported values for the Ems-Dollard estuary that ranged from 10.2 to 153.9 mg C mg1, andfound that the ratio varied seasonally and from year to year. Mean valuesfor three successive years were 40.3, 41.2, and 61.4 mg C mg, all of whichare less than the mean estimated for the assemblages from Yaquina Bay. Howeer, Jonge's values were based on direct measurements of organic carbon, whilein the research reported -120-

here, biomass was measured as the weight lost after ignition, and carbon was estimated indirectly. In any case, estimates of autotrophic biomass in the sediment by the multiplication of chlorophyll concentrations by a biomass: chlorophyll ratio will be affected strongly by the physiological state of the biomass that was involved in the experimental determination of the ratio.

Therefore, it is desirable to determine the ratio during the season when the biomass estimates are required, and to measure the biomass of the experimental assemblages in the same units needed for the field estimates.

Ratios of gross primary production to chlorophyll a concentration for sediment associated macroalgae were similar to values obtained for isolated diatom assemblages, but from 10 to 20 times higher than corresponding values for micralgae in intact sediment. Production of Enteromorpha and Ulva in

Yaquina Bay and Netarts Bay was higher than corresponding values reported for o'ther estuaries (Littler et a].., 1979; King and Schramm, 1976; Buesa, 1977;

Conover, 1958), although specific rates of photosynthesis (i.e., the rate per unit of biomass) in the Oregon plants were similar to rates found for material from other geographical locations. Production estimates obtained during this study for Gracilaria were similar to values reported in the literature

(Lapointe et al., 1976; Vawes et a].., 1978). Also, light saturation of macroalgal photosynthesis was similar to that for microalgal photosynthesis and that measured in other studies (King and Schramm, 1976; Dawes et a].., 1978;

Ramus and Rosenberg, 1980). No inhibition of photosynthesis at full sunlight was observed.

The net daily primary production of sediment-associated algae was estimated to compare the productive potential of green macroalgae, red macroalgae, and sediment-associated microalgae (Table 25). The daylengths corresponded to -121-

Table 25. Daily net primary production of sediment-associated algae in Yaquina Bay. Algae are: Enteromorpha prolifera, Ulva expansa, Gracilaria verrucosa and sediment-associated microalgae (pre- dominately diatoms). Values are estimated from Tables 23 and 24, and from field observations. Dayiength is expressed as mean hours during the growing season, and biomass as mean values during the growing season. The growing season for Enteromorpha and Ulva was between June and August; for Gracilaria between March and May; and for microalgae between April and September.

Enteromorpha and Ulva Gracilaria Microalgae

DAYLENGTH 12 12

(hours)

B I OMAS S 148.5 6.5 8.3 (g c ni2)

NET PRODUGT ION 4.50 0.20 0.045 (g C m2 h1)

RESPIRATION 0.63 0.02 0.023 (g C m2 h')

DAILY NET PRODUCTION 46.44 1.50 0.26 (g C m2 d1) z GROWTH/DAya 27.21 20.76 3.09

% Growth/Day= 100 (in (Nt/No))/t; where N0 is the biomass at time zero, Nt is the biomass at time t and t is in days. -122-

a period of maximum growth and biomass for each functional group at the intensive study sites. Net primary production and rates of respiration per hour

for macroalgae were taken from Table 24. Microalgal biomass was estimated

from chlorophyll a concentrations at the intensive study sites during the

period of maximum growth. Net primary production and rates of respiration per hour were calculated from measurements of gross primary production and the mean ratio given in Table 23. Daily net primary production was estimated by subtracting the respiration during the dark hours of the day from the net

primary production during the daylight hours.

Macroalgae were very productive (20 to 27% growth per day), especially

Enteromorpha and Ulva, which accounted for large seasonal inputs of organic matter into Oregon estuaries. Mean sediment-associated microalgal production in the summer was much less than that of macroalgae (3.09% growth per day).

Admiraal and Peletier (1980) found similar growth rates in intact sediment- associated diatom populations. Macroalge are probably able to grow at a much greater rate than sediment-associated microalgae because of their ability to live on the sediment and in the water column and their greater rate of photosynthesis per unit chlorophyll a. The macroalgae can absorb nutrients from the sediment at low tide and maximize light capture by becoming suspended in the water column during periods of inundation (Welsh, 1980). Sediment-associated microalgae can also become suspended during inundation, but their growth is probably limited by turbidity created by sediment particles associated with the microalgae arid their mucus (Baillie and Welsh, 1980).

III. Effects of Estuarine Infauna on Sediment-Associated Microalgae

This section presents the results of research conducted in collaboration with Henry Lee, Marine Division of the Environmental Research Laboratory -12 3-

(E.P.A.), Marine Science Center, Newport, Oregon. Dr. Lee was supported by a

National Research Council Fellowship with E.P.A. Also, we gratefully acknowledge helpful suggestions by D. J. Specht, R. C. Swartz, and G. E.

Walsh, all of the Environmental Research Laboratory.

Importance of biotic factors in regulating sediment-associated microalgae is poorly understood. Many epifaunal and infaunal deposit-feeders ingest and assimilate microalgae (Sanders et al., 1962; Levinton, 1980). Experimental manipulations of the estuarine gastropods Hydrobia spp. (Fenchel and Kofoed,

1976), Nassarius obsoletus (Say) (Pace et al., 1979), and Bembicium auratum

(Quoy and Gainard) (Branch and Branch, 1980), demonstrated that epifaunal deposit-feeders can regulate microalgal biomass. A freshwater epibeñthic amphipod, Hyalella azteca (Sauggure), either stimulated or depressed microalgal production depending on amphipod density (Hargrave, 1970). Experimental studies of infaunal regulation of microalgae include the investigation by

White et al. (1980) who found that the sand dollar, Mellita quinquiesperforata

(Leske), had no significant effect on chlorophyll a concentrations in sediment and the study by Coles (1979), who gave qualitative evidence for infaunal regulation of microalgae.

In some preliminary experiments, we observed that defaunated sediment developed a golden-brown "diatom" layer within afewdays. These observations indicated that microalgal colonization was rapidandthat infauna regulated niicroalgal production. To test these hypotheses weexamined microalgal colonization of defaunated sediment in the fieldandlaboratory. Here, we present results of these experiments and discusstherole of infauna in controlling sediment-associated microalgae. -124-

Study Site

Experiments were performed on an intertidal sand flat in Yaqunia Bay adjacent to the Oregon State University Marine Science Center and in a laboratory at the Marine Science Center. The field study site was 1.0 to 1.1 m above MLLW and was exposed twice daily for an average exposure of 6 hours per day. Maximum tidal amplitude in Yaqunia Bay is three meters. Sediment mean grain size was 2.64 phi, a fine sand. The microalgal community was composed primarily of a diverse assemblage of pennate diatoms (Amspoker and Mclntire,

1978). During the summer, the macroalgae Enteromorpha prolifera (Mull.) J.

Ag. and Ulva expansa (Setch.) S.&G. covered a large portion of the study site. Rates of gross primary production by sediment-associated microalgae at this location ranged from 20 to 150 mg C m2 hr'.

Methods

1. Experimental Design

The effect of infauna on microalgae was evaluated by comparing microalgal biomass and production of an undisturbed sediment community with that of defaunated sediment. Defaunated sediment was sediment collected at the field site, sieved through a 1-mm mesh screen, and frozen for approximately one month. In the field experiment, defaunated sediment was placed in cylinders of fiberglass screening (1.5-mm mesh). These cylinders had a surface area of

411.9 cm2, were 10 cm deep, and were completely embedded within the sediment.

Eight defaunated sediment cylinders and eight control plots (unmanipulated sediment) were randomly assigned to 57 cm x 57 cm plots within a 285 cm x

228 cm block. The experiment was initiated in June, 1980. Macroalgae were -125-

periodically removed from the study site. Defaunated and control treatments were sampled 1, 10, and 40 days after transplanting the defaunated sediment into the field. At each sampling date, two defaunated cylinders and two control plots were sampled. Three cores (45.6 cm2, 4 cm deep) were taken within each defaunated cylinder or control plot. These cores were brought into the laboratory and used to assess metabolic activity, chlorophyll a concentration, and macrofaunal abundance.

In the laboratory experiment, cores (36.3 cm2, 4 cm deep) were either taken in the field or filled with defaunated sediment and then maintained in flowing seawater. Microalgae growing on defaunated sediment came from an initial inoculum of suspended microalgae originating from immersion of the control sediment into the water bath and by surviving microalgae. The seawater was sand-filtered, UV-treated, and maintained at 11.0 ± 1.0°C and a salinity of 30 0/00. Light was supplied by cool-white fluorescent and incandescent sources at 150p E m2 sec photosynthetically active radiation with a 12-hour light and 12-hour dark photoperiod. Light was measured with a

Licor® quantum meter and salinity with an A0 Goldberg® temperature-compensated refractometer. Metabolic activity, chlorophyll a concentration, and macrofaunal abundance were determined 1, 10, and 40 days after initiation of the experiment. Three cores of both treatments were sampled on each of the three dates. The laboratory experiment was initiated in January, 1981. To determine the effects of maintaining cores in the laboratory, metabolic activity, chlorophyll a concentration, and animal abundance cere measured in three cores taken from the field one day before termination of the experiment (i.e., day

39). -126-

2. Metabolic Activity

Gross primary production (GPP) and community oxygen uptake (OUPTK) were

measured in the laboratory using changes of oxygen in stirred, light and dark

chambers designed to hold intact cores of sediment (see page 44 of this

report). In the field experiment, the three cores from a defaunated or

control plot were incubated in 6.0-1 plexiglas chambers (Fig. 8). In the

laboratory experiment, a 300-mi plexiglas chamberwas used for each intact

core (Figure 9). The large chamber required 1 hour dark and1 hour light

incubation for measurements of OUPTK and net community02 production, respec-

tively. Incubations were performed at 13 to 14°C, a light intensity of 210 i.iE m2 sec', anda salinity of 30 0/00. The small chamber required one-half hour for each condition. Oxygen was measured polarographically using an

Orbisphere® salinity-compensated02 system. GPP was calculated by adding

OJJPTK to net community 02 production foran equivalent period of time. Calcu- lated rates of GPP in tug 02m2hr were converted to mg Cm2hr' assundng a photosynthetic quotient of 1.2 (i.e., mg C 0.312 x tug 02), and OUPTK rates in mg 02 m2 hr were converted to mg C m2 hr assuming respiratory quotient of 1.0 (i.e., mg C 0.375 tug 02) (Westlake, 1965).

3. Chlorophyll a Analysis

The concentration of chlorophyll a in sediment sampleswas analyzed using the method of Strickland and Parsons (1972). After metabolic measurements, a core (4.15 cm2, 3 cm deep) for chlorophyll a analysis was taken from each larger core. These smaller cores were frozen intact and the top cmwas sectioned off and then emersed in 90% acetone. The concentration of chlorophyll a in the acetone extract was determined witha spectrophotometer, and the Lorenzen equation was used to calculate tug chlorophyllam2, corrected -127-

for pheaopigments. These values are not corrected for chlorophyllide, but are suitable for comparing the effect of grazing on adjacent plots (Pace et al.,

1979; Whitney and Darley, 1979).

4. Macrofaunal Abundance

After removing the subcores for the chlorophyll analysis, the remaining sediment in each field core (40.05 cm2) or laboratory core (32.15 cm2) was sieved through a 1.0 mm mesh screen. The macrofaunal residue was preserved in buffered 10% seawater formalin which was dyed with rose bengal. The macro- fauna was sorted later and identified under 12x magnification.

5. Statistical Analysis

The significance of changes over time or differences between treatments was tested using SPSS programs (Nie et al., 1975). One-way analysis of variance (ANOVA) was used to test for temporal changes with the defaunation and control treatments analyzed separately. Two-way ANOVA was used to test for time, defaunatlon, and time by defaunation interaction effects. The small-sample t-test was used to test differences between treatments at specific dates. Statistical significance was defined at P < .05. Infaunal abundance was transformed using the expression y = log10 (x + 1), where x was the numerical abundance. Values for CHLR, GPP, and OUPTK were not transformed. tn the field experiment, values obtained for three cores from each defaunated cylinder or control plot were averaged, resulting in two replications for each of the two treatments at each sampling date. In the laboratory experiment, each core was analyzed independently, resulting in three replications for each treatment at each sampling date. -128-

Results

1. Field Defaunation Experiment

Chlorophyll a concentration in control sediment decreased significantly during the 40-day experiment, whereas the concentration in defaunated sediment increased significantly during this period (Fig. 24). This pattern resulted in a significant time by defaunation interaction. Colonization of defaunated sediment by microalgae was evident as golden-brown patches where defaunated sediment had been transplanted. Neither gross primary production (Fig. 25) nor community oxygen uptake (Fig. 26) varied significantly over time or between control and defaunated sediments. Total infaunal abundance in control sediment did not vary significantly over time, while animals in defaunated sediment increased (P > .05) to control levels by day 40 (Fig. 27). A tanaid,

Leptochelia dubia (Kr6yer) constituted > 63% of the individuals in defaunated and control sediment. Other abundant infaunal taxa were spinoids, capitellids, amphipods, and the venerid bivalve Transennella tantilla (Gould). Epi- faunal species were not abundant at this site.

2. Laboratory Defaunation Experiment

The chlorophyll a concentration in control sediment did not vary significantly over time (Fig. 28). Two-way ANOVA indicated significant time and defaunation effects. By day 40, the chlorophyll a concentration was approximately four times greater In defaunated sediment than In control sediment (P

< .05). Gross primary production in control sediment did not vary significantly over 40 days, but increased dramatically (P < .01) with time in defaunated sediment (Fig. 29). Two-way ANOVA indicated significant time and time by defaunation interactioneffects. Both effects resulted from the rate -129-

o 0 0 o 0 o U) 0 1) (\J (\J 2 LI) 200

CONTROL i'tSE I 41 ;- 150 £ DEFAUNATED

E lao =-J C..) 50

14-1

l0 DAY JUNE FIELD EXPERIMENT

7±SE CONTROL £ DEFAUNATED

0 0 N 0 2 co (\1 60 0N E 40 c. 3-

l0 DAY JUNE FIELD EXPERIMENT

Figures 24 and 25. Concentratiofl of chlorophyll a and rate of gross primary production during the field defaunation experiment in June, 1980. Values are means of two replicates. -130-

80N N 0N 0 20 E E

cL 0 0 40 10 7±SE CONTROL 20 £ DEFAUNATED

I 10 40 DAY JUNE FIELD EXPERIMENT

tSE CONTROL TANAIDS 80,000 ACONTROL TOTAL 0DEFAUNATED TANAIDS ADEFAUNATED TOTAL 60,000

(1) J 40,000 z

20,000

I. I0 D AY JUNE FIELD EXPERIMENT

Figures 26 and 27. Rate of community oxygen uptake and abundance of animals during the field defaunation experiment in June, 1980.Values are means of two replicates. -131-

250

tSE 'CONTROL 200 6DEFAUNATED

N 50 E -j 100 0

10 DAY JAN. LAB EXPERIMENT iI

±SE CONTROL - £ DEFAUNATED

N 200

0 C,

!4o 3- a- 3- 0 (' C., 00 / 50

DAY JAN. LAB. EXPERIMENT

Figures 28 and 29. Concentration of- chlorophyll a and rate of gross primary production during the laboratory defaunation experiment in January, 1981. Values are means of three replicates. -132-

of gross primary production in defaunated sediments starting from zero and reaching relative high values by day 10. By day 40, the rate was approximately two times greater in defaunated sediment than in control sediment.

Oxygen uptake in the dark by control sediment did not vary significantly over time, while values for this variable increased in defaunated sediment by day

10 and then decreased to values similarto the controls by day 40 (Fig. 30).

Two-way ANOVA indicated significant defaunation and time by defaunation interaction effects on oxygen uptake.

Total infaunal abundance in control sediment did not vary significantly over time, whereas infaunal abundance in defaunated sediment was zero at days

1 and 10 and then increased to 311 tanaids per m2 by day 40 (Fig. 31).

Leptochelia dubia constituted more than 75% of the individuals in control sediment. Other abundant taxa in control sediment included spionids, capitellids, and amphipods.

There were no significant differences in chlorophyll a concentration, primary production, community oxygen uptake, or total infaunal abundance between control cores maintained in the laboratory for 40 days and cores taken in the field on day 39.

Interpretation of Defaunation Experiments

Microalgal colonization of defaunated sediment in the field was rapid as indicated by the return of gross primary production (GPP) to control levels within 10 days. Chlorophyll a (CHLR) also returned to control level within

10 days though It is not possible to separate the amount of chlorophyllides initially in the defaunated sediment from the import of viable microalgae.

Total infaunal density returned to control levels by day 40, as expected from -133-

oJ c,.J

2O GO E

Afri }- 010 0 ±SE CONTROL £ DEFAUN.ATED 20

I0 DAY JAN,LAB.EXPERIMENT

SCONTROL TANAIDS £CONTROL TOTAL DEFALJNATED TANAIDS DEFAUNATED TOTAL 1E

0, -J

I0 DAY JAN. LAB EXPERIMENT

animals Rate of community oxygenuptake and abundance of Figures 30 and 31. January, 1981. during the laboratorydefaunation experiment in Values are means of threereplicates. -134--

previous experiments (H. Lee and J. Lee, unpublished). Placement of de-

faunated sediment in the field simulated localized natural disturbancessuch as those caused by epibenthic fishes and invertebrates orbioturbation (Lee

and Swartz, 1980). The rapid recovery of both microalgal production and bio-

mass indicates that this microalgal assemblage was resilient tosmall-scale disturbances. This resilience was probably facilitated both by transport of microalgae into disturbed areas (Tenore, 1977; Baillie and Welsh, 1980) and by

rapid growth of colonizing microalgae (Admiraal and Peletier, 1980). Rapid

colonization on a localized scale does not imply that microalgae are resistant

to a generalized stress (e.g., pollution) or overlappinglocalized disturb-

ances such as occurs in a Callianassa bed.

We hypothesized that removal of infauna would stimulate microalgal

production and biomass. This result was not readily observed in the field

experiment. The possible import of microalgae into defaunated sediment during

initial stages of colonization and export during later stages of colonization

and initial presence of chlorophyllides in defaunated sediment would tend to

minimize differences in chlorophyll a concentration and primary production

between control and defaunated sites. Recolonization by infauna could also

minimize differences between treatments. To better evaluate influences of the

infauna on microalgal biomass and metabolism, a laboratory experiment was

performed which eliminated the confounding effects of import-export of micro-

algae and infaunal colonization. Similarity of levels of chlorophyll, gross

primary production, community oxygen uptake, and infaunal abundance between

control cores held in the laboratory for 40 days and the natural field commun-

ity suggested that maintenance of cores did not affect community structure or -135-

function. Therefore, the laboratory experiment was considered an appropriate method to determine effects of infauna on sediment-associated microalgae.

In the laboratory, removal of infauna caused a significant increase in microalgal biomass and production. The relatively high concentration of chlorophyll a in defaunated sediment by day 1 was not functional as indicated by the low rate of primary production. Growth of microalgae in the defaunated sediment was indicated by an increase in primary production by day 10, although the chlorophyll concentration remained constant. By day 40, chlorophyll concentration and the rate of gross primary production in defaunated sediment were significantly higher than in control sediment.These results suggested that natural densities of infauna can control both microalgal biomass and production, as has been found in several other studies of estuarine epifaunal gastropods (Fenchel and Kofoed, 1976; Pace et al.,

1979; Branch and Branch, 1980). Natural infaunal density at the study site exceeded the level at which microalgal and microbial populations are stimulated by consumption of senescent cells or nutrient additions (Margrave,

1970, 1976; Cooper, 1973; Porter, 1976; Pace et al., 1979).

Grazing is the most likely mechanism by which infauna controlled microalgal biomass and production. The most abundant species at the study site,

Leptochelia dubia, contained both broken and whole diatom frustules in its gut, indicating herbivory. The same food resource was found for another tanaid, Leptochelia rapax (Harger) (Kneib et al., 1980). Burial of cells by sediment turnover is another mechanism by which infauria can limit microalgae. However, Leptochelia does not appear to turn over large quantities of sediment (Myers, 1977), and large sediment reworkers such as Callianassa and

Upogebia were not abundant at our study site. The process of defaunating -136-

sediment may have increased nutrient concentrations which in turn could have stimulated microalgal growth. However, field and laboratory nurient-addition experiments at the study site indicated that sediment-associated microalgae were not nutrient-limited (Cardon, 1981).

Patterns of oxygen uptake were difficult to interpret. In the field, oxygen uptake in defaunated sediment recovered to control values between days

10 and 40. In the laboratory, oxygen uptake of defaunated sediment was initially identical to control sediment, increased to greater than control levels between days 1 and 10, and then decreased to control values between days 10 and 40. Causes for the different patterns in the field and laboratory experiments were not apparent. The low macrofaunal density (0 to 311 m2) and diversity (0 to 1 species) in the defaunated sediment in the laboratory normally would constitute strong evidence for a highly degraded benthic community. Yet oxygen uptake at day 40 was the same for control and defaunated sediment. This similarity of community metabolism between these radically different benthic communities indicates that there is no simple relationship between oxygen uptake and community structure. Also, the effect of heterotrophic microorganisms on these experiments is unknown.

We have presented experimental evidence for a controlling effect of infaunal animals on estuarine sediment-associated microalgae.As suggested by Pace et al. (1979), herbivory is one of the factors regulating microalgal biomass and production although its relative importance, compared to other factors, is unknown. Differences in infaunal abundance and activity may be one of the factors influencing both spatial and temporal patterns of microalgae (Cadêe and Hegeman, 1974; Coles, 1979).The influence of animals on plant communities is a widespread phenomenon and should not be ignored when assessing controlling factors of plant growth (Brook, 1955, 1975). -137-

IV. The Diatom Flora of Netarts Bay

Figure 5 indicated that microalgae and several macroalgae are important groups of autotrophs in Oregon estuaries. Among the microalgae, the diatoms are by far the most abundant and diverse group, with representative taxa found in the water column, in the sediment, and growing as epiphytes on Zostera marina and macroalgae. In this section the taxonomic structure of the diatom assemblages of Netarts Bay is examined and related to selected environmental variables. Such information relates to patterns of primary production and to patterns of water circulation in the estuary. Also, the species composition of estuarine diatom assemblages can be used as an indicator of certain aspects of the chemical and physical environment.

Sampling

The distribution and relative abundances of diatom taxa were examined in planktonic, epiphytic and sediment collections during a one year period from

February 1980 to March 1981. Samples were collected once each month on dates that corresponded to favorable tide and weather conditions. Sampling stations were established at sites A, B, and C in Netarts Bay (Fig. 7), which represented different sediment types (i.e., sand, fine sand, and silt).

Samples of the sediment-associated diatom flora were obtained from four tidal heights along an intertidal transect at each site. These sampling stations were the same stations that were established to study microalgal primary production (see page 41). Epiphytic diatoms were collected from a study area near site B. In this case, samples were obtained from three locations, 1.1 m,

1.2 m, and 1.4 m above MLLW, along an intertidal gradient across a -138-

large Zostera bed. A detailed description of this Zostera bed is reported in

a later section and in Figures 32-34. On each sampling date, collections of

planktonic diatoms were obtained from: (1) bay water at low tide near site B;

(2) ocean water at high tide near the mouth of the bay; and (3) from water

near Site B at high tide.

Methods

Samples of planktonic diatoms were collected by pouring 10 1 of water

through a 101.im mesh plankton net. Planktonic diatoms were cleared of

pigments in ethanol, and sea salt and other dissolved solids were removed from

the samples by successive washings in distilled water. After washing, a sub-

sample of the diatom suspension was transferred to a microscope coverslip and

allowed to air dry. Coversllps with the dry diatom deposits were mounted on microscope slides using Cumarone resin (Holmes etal., 1981).

One shoot of Zostera marina and its associated epiphytic diatoms was

collected from each of the three intertidal sampling stations near site B on

each sampling date. These samples were placed in screw-top bottles and were

transported to the laboratory for processing.

Benthic sediment cores were obtained at the intertidal stations at site

A, B, and C on each sampling date. These samples were taken by pushing a

2.3 cm diameter plastic pipe into the sediment and extracting the upper 2-3 cm

of sediment and its associated microflora. The cores were capped, transported

to the laboratory in a vertical position, and frozen until the cleaning

procedure was initiated. Diatoms in the top cm of each core were isolated by

the cleaning procedure. -139-

Organic materials in the epiphytic and sediment samples were removed by nitric acid digestion in a Kjeldahl apparatus. Large sediment particles were separated from the diatoms by rapid decanting, and the clean diatom frustules were allowed to settle in beakers for at least 4 hr. After several washings with distilled water, the isolated frustules were mounted on microscope slides following the procedure described above for the planktonic taxa.

Epiphytic and sediment-associated diatoms were identified and counted with a Zeiss research microscope at 1200x magnification using brightfield illumination. Approximately 500 valves were identified for each sample, a sample size that was compatible with the statistical methods used to report the results (Mclntire and Overton, 1971). Epiphytic samples from all

12 months were counted for a total of 36 samples. Of the 116 benthic samples collected, samples from every other month were counted for a total of

64 samples. Results of the examination of plankton samples are reported as species lists.

Since the diatom collections were obtained concurrently with the sampling programs designed to study microalgal and Zostera primary production and biomass, measurements of physical and biological variables obtained during these investigations also were used to interpret distributional patterns in the diatom flora. Variables of particular interest included intertidal height, chlorophyll a and organic matter concentrations in the top cm of sediment, the ratio of chlorophyll a concentration in the top cm of sediment to that at a depth of 4-5 cm, day length, water temperature, mean sediment particle size, the sediment sorting coefficient, and the sediment skewness coefficient.

Methods associated with the collection of these data are described on pages

44-49 of this report. -140-

Data Analysis

The analysis of distributional patterns in the diatom flora of Netarts

Bay involved:

1. A principal component analysis of physical and biological

environmental data;

2. Calculation of niche breadth values (Mclntire and Overton, 1971) for

selected epiphytic and sediment-associated taxa;

3. The regression of the relative abundance of selected epiphytic and

sediment-associated taxa against environmental variables and principal

components of environmental variables;

4. Comparisons of assemblages pooled as epiphytic, sand taxa, fine sand

taxa, and silt taxa by SIMI, a measure of similarity (Mclntire and

Moore, 1977);

5. Ordination of samples and taxa by reciprocal averaging(Hill, 1973);

and

6. Correlation of ordination scores for sites with selected environmental

variables.

All data analyses were performed on a Control Data Corporation CYBER

170/720 computer system at the Oregon State University Computer Center using the AIDNX, ORDIFLEX, CORRELX, and PARTL computer programs, and SIPS, a statistical interactive programming system. -14 1-

Results

1. Structure of Environmental Data.

Environmental variables were organized in two data sets, one with environmental variables corresponding to the epiphyte samples and another with variables corresponding to the sediment samples. Variables related to the epiphyte samples included intertidal height, daylength, and water temperature.

The Interrelationships among these variables were investigated by correlation analysis. Tidal height was virtually uncorrelated with daylength (r= 0.00) and with water temperature (r = 0.08), while the covariance between daylength and water temperature, variables related toseason, was high (r 0.82). The environmental data were matched with 18 samples of epiphytic diatoms for alternate months.

Environmental data corresponding to sediment samples consisted of nine variables, namely intertidal height (TIDE), surface chlorophylla (CHLA), surface organic matter (OM), the chlorophyll ratio(RATIO),daylength (DAYL), water temperature (TEMP), mean particle size (PHI), and the sorting and skewness coefficients(SORTand SKEW). Covariances among the sediment properties PHI, SORT and OM were relatively high, and correlation coefficients for these variables were 0.67 (PHI-OM), 0.78(SORT-OM),and 0.83(PHI-SORT).

Moreover, daylength and water temperature were closely correlated with each other (r = 0.76). The variables tidal height, surface chlorophyll, the chlorophyll ratio, and the skewness coefficient were not highly correlated with each other and with other variables (correlation coefficients below

0.40). The environmental data were matched with 64 samples of sediment- associated diatoms from alternate months. -142-

The relatively large environmental data set associated with the benthic diatom samples was summarized by the concentration of variance into three prinicipal components. The interpretation of the principal components was achieved by the examination of factor loadings, i.e., the correlations between the components of interest and the original variables (Table 26).

Relatively high correlations (r > 0.8) between the first principal component and OM, PHI and SORT indicated that this axis was an expression of sediment properties. The second axis was highly correlated (r > 0.9) with daylength (DAYL) and with water temperature (TEMP), and was interpreted as a seasonal component. The third axis was highly correlated (r = 0.81) with intertidal height (TIDE).

The low communalities for surface chlorophyll (CHLA), the chlorophyll ratio (RATIO) and the skesmess coefficient (SKEW) indicated that these measures were not expressed appreciably by the first three components.

Eigenvalues indicated that the first three principal components accounted for

66% of the total variation in the environmental data, and 33% was expressed by the first axis alone.

2. The Diatom Flora

Planktonic, epiphytic, and sediment-associated diatom samples contained a total of 340 taxa (species and varieties of species) from 73 genera. Of these taxa, 50 were planktonic and 294 were either epiphytic, associated with sediment, or both. Only five taxa, namely Thalassiosira 1, Thalassiosira decipiens, Thalassiosira pacifica, Thalassionema nitzschioides and Skeletonema costaturn, occurred among planktonic, sediment and epiphytic assemblages. -143-

Table 26. Factor loadings for the first three principal components generated from the environmental variables associated with sediment samples. Variables are intertidal height (TIDE), chlorophyll a concentration (CHLA), sediment organic matter (OM), chlorophyll ratio (RATIO), daylength (DAYL), water temperature (TEMP), mean sediment particle size (PHI), sorting coefficient (SORT), and the skewness coefficient (SKEW). Loadings with an absolute value greater than 0.8 are underlined for emphasis.

PCA1 PCA2 PCA3 r2 r3 Communality

TIDE -0.16 -0.18 0.81 0.72

CHLA -0.50 -0.12 0.44 0.45

OM -0.88 -0.10 0.07 0.78

RATIO 0.57 -0.07 0.40 0.48

DAYL -0.03 0.93 0.13 0.88

TEMP -0.00 0.93 0.13 0.88

PHI -0.85 -0.03 -0.09 0.73

SORT -0.91 -0.11 0.11 0.85

SKEW -0.21 -0.03 0.38 0.15

Elgenvalues 2.94 1.80 1.12

% of variance extracted 33% 20% 13% -144-

Approximately 53,850 diatom valves were identified and counted in

64 benthic samples and 36 epiphytic samples. Epiphyte samples collected monthly from three intertidal stations resulted in a count of 19,463 valves and the identification of 123 taxa. Sediment samples from every other month along four intertidal stations resulted in the identification of 34,851 valves that represented 282 taxa. The overlap between benthic and epiphytic assemblages was 111 taxa, or 38% of the total for the two assemblages.

As is usually the case, most of the species within the assemblages were rare. Taxa represented by five or fewer valves accounted for 38% of the total number of taxa identified. Such rare taxa contribute relatively little information to the analysis of distributional patterns, as their occurrence in a sample may represent allochthonous inputs, not a reflection of local environment. Consequently, specific taxa of interest were chosen for the statistical analysis on the basis of their abundance, and the rarer taxa were eliminated. The 72 taxa chosen for ordination accounted for 95% of all epiphyte valves counted and 91% of all valves counted for the sediment samples. The 36 sediment-associated taxa chosen for the regression analysis accounted for 79% of all valves counted for the sediment samples. The 22 epiphytic taxa chosen for a regression analysis with environmental variables accounted for 94% of all epiphytic valves counted.

In general, the taxa found in plankton samples were benthic or epiphytic taxa that were dislodged and suspended in the water column by tidal currents and turbulence. Common benthic species in these samples included Paralia sulcata, Anaulus balticus, Navicula digitoradiata, Navicula cancellata,

Nitzschia socialis and Nelosira moniliformis. Epiphytic taxa found as -145-

tychoplankton included Cocconeis scutellum, Cocconeis scutellum v. parva,

Synedra fasciculata and Navicula directa. Euplanktonic species were abundant only in samples from February, March and August of 1980. In these samples, the species were typical of those found in the neritic plankton along the

Oregon Coast, and the presence of marine plankton in Netarts Bays apparently was due to seawater transport into the bay by tidal fluxes.

Samples of plankton from the three months that had marine neritic floras are compared in Table 27. Although samples from February and March shared fewer taxa (26%) than the February and August samples (38%), the February and

March samples were dominated by the same species, namely Nitzschia seriata, N. pungens, Chaetoceros compressus, Skeletonema costatum, Thalassiosira decipiens, T. 1, and Rhizosolenia setigera. August plankton samples were dominated by several species of Chaetoceros, contained different species of

Nitzschia and Thalassiosira than February and March samples, and lacked species of Rhizosolenia. Dominant species in the August samples included

Chaetoceros coinpressus, C. constrictus, C. socialis, Nitzschia pacifica,

Eucampia zodiacus, Thalassiosira nordenskioeldii, and T. pacifica.

The relative abundances and niche breadth values for 72 common benthic and epiphytic taxa are presented in Table 28. The most abundant epiphytes were Navicula salinicola, Navicula tripunctata v. schizonemoides, Navicula frustulum v. subsalina, and Synedra fasciculata; while the most abundant sediment-associated taxa were Achnanthes hauckiana, and Opephora pacifica. In this case, niche breadth values may range from 1.0 if a species was present in only one sample, to 82 if equally abundant in all 82 samples. Although

Paralia sulcata was not very abundant in the samples, it had a relatively high -146-

Table 27. A list of planktonic species found in Netarts Bay during months in

1980 when the flora was dominated by marine neritic taxa. Plus

signs (+) indicate that a taxon was present.

February March August

Actinocyclus octonarius + Actinocyclus splendens + Asterionella japonica + Bacteriastrum delicatulum Bacteriastrum hyalinum Biddulphia longicuris + Chaetoceros armatum + Chaetoceros compressus + + Chaetoceros constrictus + + + Chaetoceros curvisetus + Chaetoceros decipiens + Chaetoceros didymus + Chaetoceros lacinosus + Chaetoceros lorenzianus + Chaetoceros radicans + Chaetoceros socialis + + Chaeto ceros vanheurcki + Corethron hystrix + Coscinodiscus curvatulus Coscinodiscus eccentricus Coscinodiscus radi atus + ± + + August + + + + + + + + ± + + + March +

-f + + + + + + + + + + + + + + + + February -147- semispinosa Coscinodiscus sublineatus Coscinodiscus Lauderia borialis Leptocylindrus danicus Lithodesmium undulatus Navicula planamembranacea Rhizosolenia alata Rhizosolenia hebatata v. Schroederella delicatula Stephanopyxis nipponica Ditylum brightwellii Eucampia zodiacus Hemlaulus hauckii Navicula complanatula Nitzschia delicatissima Nitzschia pacifica Nitzschia pungens Nitzschia seriata Pleurosigma normanii Rhizoselenia setigera Skeletonenia costatum Stephanopyxis palmeriana Thalassionema nitzschioides Thalassiosira aestivalis Thalassiosira decipiens Thalassiosira nordenskioeldii Thalassiosira rotula Thalassiosira 1 Thalassiosira pacifica Thalasslosira longissima Table 27 (continued) Table 27 -148-

Table 28. A list of 72 selected taxa, their total relative abundance (NT), and their relative abundance in epiphyte (NE) and benthic (NB) samples. Also listed is the niche breadth values (B) for each taxon in relation to 82 epiphyte and benthic samples.

NT NE NB

Achnanthes 1 440 0 440 31.44

Achnanthes 11 B 918 0 918 20.26

Achnanthes hauckiana 3244 9 3235 30.86

Achnanthes latestriata 172 0 172 19.63

Achnanthes lemmermanni 621 2 619 36.51

Amphora 35 72 0 72 13.84

Amphora coffeiformis 441 4 437 43.20

Amphora exigua 349 2 347 19.87

Amphora laevis v. perminuta 155 0 155 14.99

Amphora libyca 148 0 148 22.65

Amphora micrometra 499 0 499 19.28

Amphoraproteus 79 0 79 9.03

Amphorasabyii 1695 7 1688 41.31 Amphoratenerrima 678 104 574 38.48

Anorthoneis eurystoma 118 1 117 8.32

Bacillaria paradoxa 192 171 21 7.44 Berkeleya rutilans 473 373 100 9.32

Cocconeis 11 A 1649 2 1647 40.87

Cocconeis 11 C 336 2 334 32.89

Cocconeis J 991 11 980 28.39 Cocconeis costata 460 390 70 14.40 Cocconeis placentula v. euglypta 1365 38 1327 42.48

Cocconeis scutellum 798 585 213 19.92 Cocconeis scutellum v. parva 529 481 48 9.91 Cymbellonitzschia hossamedinii 1117 0 1117 21.64 -149-

Table 28 (continued) NT NE NB

Fragilaria pinnata 96 0 96 10.04 Fragilaria striatula v. californica 212 2 210 2.54 Gomphonema oceanicum 458 413 45 9.55

Gyrosigina prolongum 42 1 41 3.97

Hantzschia 1 38 0 38 6.44

Hantzschia marina 62 0 62 6.58

Melosira inoniliformis 278 37 241 6.18

Melosira nummuloides 82 17 65 4.54

Navicula 3 71 2 69 3.36

Navicula 16 206 0 206 3.14

Navicula 109 460 16 444 24.27

Navicula 150 201 129 72 14.93

Navicula 199 1100 0 1100 17.30

Navicula ammophila v. minuta 144 0 144 16.84

Navicula directa 542 535 7 6.90

Navicula diserta 340 0 340 41.91

Navicula diversistriata 332 2 330 15.86

Navicula forcipata 120 0 120 15.01 Navicula gottlandica 162 0 162 9.06

Navicula gregaria 857 4 853 32.34

Navicula grostchopfi 76 0 76 5.47

Navicula patrickae 214 1 213 20.18

Navicula sauna 48 0 48 9.66 Navicula salinicola 3665 1760 1905 41.12

Navicula tripunctata 69 0 69 2.49 Navicula tripunctata v. schizonemoides 1229 930 229 24.86

Nitzschia 2 82 0 82 16.51

Nitzschia 5 100 62 38 19.64

Nitzschia 47 117 0 117 3.18 -150-

Table 28 (continued) NT NE NB

Nitzschia 171 223 202 21 3.72

Nitzschla brevirostris 78 78 0 3.08 Nitzschia dissipata v. media 314 207 107 29.97

Nitzschia frustulum 41 0 41 12.92 Nitzschia frustulum v. subsalina 2456 1060 1396 37.13 Nitzschia fundi 1524 472 1052 42.75

Nitzschia pseudohybrida 274 191 83 13.86

Nitzschia punctate 21 0 21 16.23

Nitzschia rostellata 232 191 41 8.54

Opephora pacifica 3503 6 3497 54.23

Opephora perminuta 262 1 261 15.80

Opephora schultzi 237 3 234 34.15

Paralia sulcata 565 18 546 45.06

Rhoicosphenia curvata 56 49 7 11.60

Rhopalodia musculus 65 3 62 8.51 Synedra fasciculata 851 803 48 15.14

Thalassiosira 1 1037 119 918 42.65

Trachysphenia australis 246 2 244 11.62 -151-

niche breadth value because of its presence in numerous sediment and epiphyte samples. In contrast, Opephora pacifica derived its large niche breadth value from its high relative abundance in numerous sediment samples.

Several species had low niche breadth values, indicating a restricted distribution, and yet made a substantial contribution to the samples in which they were found. Fragilaria striatula v. californica constituted approximately one-fifth of the total valves in two samples from the SAND site (1.0 above MLLW) in August and October. In all other samples, this taxon was absent or rare. Navicula 3 was abundant only in February in samples from the

SAND Site at 1.5 m and 2.0 m above MLLW. Although Navicula 16 was found only in sand samples and was usually rare, this diatom represented about two-fifths of all the valves identified from 2.0 m above MLLW in April. Nitzschia 171 and Nitzschiabrevirostris were epiphytic taxa with restricted distributions. Nitzschia brevirostris was abundant only in the three epiphyte samples from October, and Nitzschia 171 was abundant in the September epiphyte samples where it was found in the colonial muselage of Bacillaria paradoxa.

3. Autecology of Selected Taxa.

The relationship between the relative abundance of 22 selected epiphytic taxa and three environmental variables are presented in Table 29. The coefficient of determination (R2) for the multiple regression of each taxon against all three variables ranged from 0.02 for Cocconeis scutellum to 0.81 for Nitzsehia fundi; nine of the 22 taxa had R2 values 0.50 or greater.Most of the covariance between species abundance and the environmental variables was associated with seasonal fluctations in the flora, as the relative abundance of nine taxa were correlated (r > 0.50) with daylength and water -152-

Table 29. The relationship between 22 selected epiphyte taxa and three environmental variables: intertidalheight(TIDE), daylength (DAYL), and water temperature (TEMP). Correlation coefficients are given for each taxon,as well as thecoefficient of determination (R2) for the multipleregression of theabundance of eachtaxon against the three variables.

TIDE DAYL TEMP R2

Amphora tenerrima 0.19 0.48 0.37 0.27 Bacillaria paradoxa 0.02 0.11 -0.14 0.17 Berkeleya rutilans 0.11 0.53 0.42 0.30 Cocconeis costata 0.08 0.34 0.12 0.20 Cocconeis scutellum -0.08 -0.01 0.05 0.02 Cocconeis scutellum v. parva 0.30 0.70 0.58 0.58 Gomphonema oceanicum -0.16 -0.07 -0.43 0.44

Navicula 150 0.07 0.50 0.28 0.30 Navicula directa -0.01 -0.46 -0.39 0.22

Navicula salinicola 0.24 -0.38 0.03 0.52 Navicula tripunctata v. schizonemoides -0.23 -0.74 -0.57 0.61 Nitzschia 5 0.03 -0.22 -0.49 0.32 Nitzschia 171 0.05 0.12 -0.03 0.07 Nitzschia brevirostris 0.05 -0.24 -0.58 0.53 Nitzschia dissipata v. media -0.09 -0.14 -0.57 0.66 Nitzschia frustulum v. subsalina 0.18 0.81 0.68 0.69

Nltzschia fundi 0.09 0.89 0.81 0.81 Nitzschia pseudohybrida 0.02 -0.50 -0.70 0.50 Nitzschia rostellata -0.08 -0.10 -0.37 0.27 Rhoicosphenia curvata -0.21 0.23 0.08 0.12 Synedra fasciculata -0.05 0.06 -0.37 0.55 Thalassiosira 1 0.13 0.10 0.05 0.03 -15 3-

temperature (Table 29). Navicula tripunctata v. schizonemiodes, Nitzschia frustulum v. subsalina, and Nitzschia fundi were abundant throughout the year, but had a distinct maximum occurrence during winter, spring, or summer, respectively. Cocconeis scutellum v. parva and Berkeleya rutilans were absent or rare in the summer or fall, but were common during winter and spring.

Several species of Nitzschia were common in the October samples, namely N. brevirostris, N. dissipata v. media, N. pseudohybrida, N. rostellata and

Nitzschia 5. Nitzschia 171 was present only in September samples, and therefore had a distinct seasonality in spite of its low correlations with daylength and water temperature. Correlations of relative abundances of the

22 epiphytic taxa with tidal, height were relatively weak, only Cocconeis scutellum v. parva exhibited a weak relationship with intertidal position (r

0.30).

Relationships between the distribution of 36 sediment-associated taxa and the environmental data are summarized in Table 30. Twenty-two of these taxa were at least weakly associated (r > 0.38) with the first principal component of the environmental data matrix, and 13 taxa have correlation coefficients of

0.50 or greater. This component expressed sediment properties, especially OM,

PHI, and SORT (Table 26). Negative correlation coefficients for Achnanthes 1,

Achnanthes 11B, Amphora micrometra, Navicula gottlandica, and Opephora schultzi indicated that these taxa were associated with sediments that were composed of finer particles, were poorly sorted, and had a high concentration of organic matter. The high positive correlations for Achnanthes latestriata

Amphora proteus, Anorthoneis eurystoma, Cocconeis J, Cymbellonitzschia hossamedinii, Navicula ammophila v. minuta, Navicula 16, and Navicula Table 30. The relationship between 36 sediment-associated taxa -154- and the first three principal regressionr3)components are given of forthe eachspeciesenvironmental component. variables data againstmatrix. R is the the first coefficientCorrelation of three principal factors coefficients determination (r1, for the r2 and derivedagainstcoefficient from all theenvironmentalof determination components analysis, where variables. for the multiple R regression= rof species + r + r and R2 is the abundance PCI r1 PC2 r2 PC3 r3 Rf 2 R 2 AchAchnanthesAchnanthes nanthes haucklana 111 B -0.47-0.67 -0.24-0.03 0.06 -0.07 0.400.02 0.440.550.46 0.610.660.48 AchnanthesAmphoraAchnanthes latestriatlemmermanni exigua a 0.490.440.55 -0.26-0.21 0.12 -0.07-0.18 0.40 0.320.360.39 0.480.460.43 Amphora microuietralibycalaevis V. perminuta -0.74 0.39 -0.07-0.01 0.28 -0.00-0.18-0.05 0.190.230.55 0.640.290.33 AmphoraAmphora proteus sabyiltenerrima -0.20 0.290.59 -0.11 0.180.00 -0.17-0.42 0.28 0.220.140.44 0.220.340.79 CocconeisAnorthoneis J11 eurystomaA -0.20 0.680.58 -0.07-0.02 0.10 -0.19-0.36 0.25 0.520.170.40 0.240.580.67 Table 30 (continued) PCi PC2 PC3 Cocconeis placentula v. euglypta r2 r3 R MelosNavlculaCyinbellonitzschf ira 3 a moniliformis hossamedinli -0.05 0.28 0.170.13 -0.10-0.10-0.46 0.040.29 0.120.47 Navicula 10916199 -0.22 0.260.53 -0.20-0.01 0.59 -0.26 0.200.12 0.460.320.12 0.550.220.41 NaviculaNavicuia amrnophiladiversistriata v. minuta ott1andica 0.570.580.18 -0.13-0.32 0.02 -0.32 0.040.00 0.450.340.13 0.560.410.32 Navicula saliuicolagroschopfigregaria -0.44-0.52 0.01 -0.00-0.02 0.20 -0.12 0.440.20 0.390.080.28 0.510.430.26 Nitzschia'Titzschia 47 frustulum v. subsaitna -0.39-0.24-0.44 -0.32-0.29 0.05 -0.27 0.320.04 0.280.230.27 0.310.320.34 OpephoraNitzschia perininuta pacificafundi -0.09-0.34-0.41 -0.16 0.340.49 -0.05 0.160.17 0.150.44 0.300.250.54 ThalassiosiraTrachyspheniaOpephora schultzi I australis -0.62 0.320.13 -0.00-0.32 0.24 -0.45 0.140.03 0.120.280.48 0.250.480.61 -156-

diversistriata indicated that these taxa were associated with sandy, well- sorted sediments with low organic content. In some cases, the linear model did not expose obvious patterns in the data. For example, Amphora exigua,

Amphora laevis v. perminuta and Amphora libyca had from 77% to 82% of their total relative abundance in sand samples and yet exhibited weak correlations with the first principal factor (r = 0.39 to 0.49). Melosira moniliformis was an extreme case, with 63% of its valves in sand samples and 96% of its valves in sand and fine sand samples together, and a correlation with the sediment factor of only -0.05.

The second principal component is correlated with daylength and temperature, variables that express seasonal changes. En general, species correlations with this axis were weak except for Navicula 109 and Nitzschia fundi. These taxa were most common in the summer.

The third principal component is an expression of intertidal height.

Species correlations with this component also were weak, with no correlation coefficient greater than 0.50 and most values less than 0.30. Achnanthes hauckiana, Achnanthes lemmermanni, and Navicula groschopfi had a positive correlation with this component Cr values between 0.40 and 0.44); while

Amphora sabyii, Cocconeis placentula v. euglypta, Cymbellonitzschia hossamedinii and Thalassiosira 1 had negative correlations with this component

(r values between -0.39 and -0.46). The former group of species was associated with high intertidal stations, whereas the latter group was associated with the lower stations.

Taxa mentioned above that were associated with factors 1, 2, or 3 have R2 values ranging from 0.22 to 0.55 for the multiple regression of relative abundance against the first three principal components of the environmental -157-

data. The total for the multiple regression of each of these taxa against all nine environmental variables ranged from 0.34 to 0.79.

4. CommunIty Patterns

Both Jaccard and SIMI indices of similarity were used to compare the

species composition of epiphytic and benthic assemblages (Table 31).

The Jaccard index indicated that the epiphyte samples had from 32% to 35%

of their taxa in common with samples from the SAND, FINE SAND, or SILT

sites. Samples from the three sediment types shared a greater percentage of

taxa among themselves, in this case from 52% to 59%. In contrast to the

Jaccard index, SIMI reflects the relative abundances of species as well as

their joint occurrences in samples; a high value usually indicates that the

assemblages have the same dominant taxa. SIMI values ranged from 0.18 to 0.84

and emphasized the differences between the epiphyte samples and the sediment

samples (SIMI = 0.18 to 0.40) as well as the similarities among sediment

samples (SIMI = 0.56 to 0.84). Epiphytic samples were more similar to samples

from the SILT site than to samples from the SAND or FINE SAND sites, while

assemblages from sand were more similar to those from fine sand than from

silt.

Species and samples were ordered along ordination axes by the method of

Reciprocal Averaging (RA). The data for this analysis consisted of species

abundances counts for a total of 82 samples, of which 18 were epiphytic

collections and 64 were from the sediment. The first ordination axis

separated the epiphytic samples from the sediment samples; while the second

axis represented a continuum relative to sediment type, with samples from the

SAND site at the top right of the graph, those from the FINE SAND site at the

center right, and assemblages from the silt site at the bottom right (Fig.

32). -158-

Table 31. A comparison of the similarity of pooled epiphytic and benthic

samples using SIMI and Jaccard indices. SIMI compares the presence

and relative abundance of species among samples and is found in the

lower left half of this table. The Jaccard index uses only

presence-absence data and forms the upper right half of the table.

The pooled samples are epiphytes (EP) and the assemblages from the

Sand (SA), Fine Sand (FS) and Silt (SI) sites.

EP 0.35 0.32 0.32

SA 0.18 0.54 0.52

FS 0.29 0.62 0.59

St 0.40 0.56 0.84 -159-

The correspondence of site and species ordinations can be examined by a comparison of Figure 32 with Figure 33. Epiphytic species are found on the left side of Figure 33 (group 1). These species are Nitzschia brevirostris,

Rhoicosphenia curvata, Navicula directa, Gomphonema oceanicum, Synedra fasciculata, Bacillaria paradoxa, Cocconeis costata, Cocconeis scutellum V. parva, Nitzschia 5, Navicula tripunctata v. schizonemoides, Nitzschia dissipata v. media, Nitzschia psuedohybrida, Nitzschia rostellata, Nitzschia

171, Navicula 150 and Cocconeis scutellum. Sediment-associated species are found on the right side of the diagram, with species from sand on the top, species from fine sand in the center, and species from silt on the bottom.

Taxa that were abundant in all sediment types (group 8) are in the center near the right margin of the figure. These taxa are found mixed with species that had greater fidelity to specific substrates (mainly groups 5 and 6). Species that were found in benthic as well as epiphytic samples are in the center of the figure (group 9). These taxa are Navicula salinicola, Navicula 109,

Navicula 3, Nitzschia frustulum v. subsalina, Nitzschia fundi, Melosira nummuloides, Melosira moniliformis, Thalassiosira 1, Berkeleya rutilans, and

Amphora tenerrima. A group of five species that were virtually restricted to the upper intertidal stations at the SAND site are found in the upper right- hand corner of Figure 33 (group 2). These taxa are Hantzschia marina,

Hantzschia 1, Anorthoneis eurystoma, Amphora proteus and Navicula 16. Taxa that were most abundant at the mid-intertidal stations at the SAND site are labeled as group 3. These taxa were Gyrosigma prolorigum, Achnanthes latestriata, Navicula forcipata, Amphora exigua, Cocconeis J, and Navicula diversistriata. Interspersed with members of group 3 were taxa that were E EPIPHYTES S S M=FS= = SILT(MUD)SANDFINE SAND S ss S SS S S - 60 E EEE E E E E F SS S5S F S 4 40 EE E E E F F F F FF F C0' F M N F NF F F N 20 MM MM N MM MF F F 0 0 20 I 40 RA AXIS I 60 I M Fl 80 MM 100 Figure 32. andOrdinationSamples the SILT were of site epiphyteobtained (M). andfrom sediment-associated Zostera leaves (E), diatom the SAND samples site by(S), reciprocal the FINE SAND (F), averaging. I- I,J 2 3 44 5 33 8 8

In 7 3434 66 58855 68 80 77 76 6 777 N- 8 9 9 N- epiphyte and sediment samples 9 species groups discussed in the text. 9 9 60 1 9 9 RA AXIS Numbers refer to 40 III species associated with the 20 illustrated in Figure 32. Ordination of diatom o 0' 80 60 20 .i.Is c'J (1) 4 4 '40 Figure 33. -16 2-

found in the sand samples, regardless of tidal height. These species (group

4) included Navicula ammophila v. minuta, Amphora laevis v. perininuta, Amphora libyca, Trachysphenia australis and be1lonitzschia hossamedinii. Iear the right center of the figure, there are species that were common in the low intertidal region at the SAND site. These taxa (group 5) also were common at the mid and low intertidal stations at the FINE SAND site.These taxa were

Amphora tenerrima, Thalassiosira 1, Navicula 199, Achnanthes lemmermanni,

Navicula diserta, and Cocconeis placentula v. euglypta. Amphora tenerrima and

Thalassiosira 1 were closely related to this group, but were classified as members of group 9 because they were also common epiphytes. Several taxa that were common at the FINE SAND site also were common in samples from the SILT site. These taxa (group 6) were Amphora sabyii, Cocconeis hA, Achnanthes hauckiana, Achnanthes 1, Navicula patrickae and Nitzschia punctata. Four other taxa were associated with this group but were classified as members of group 9 because they were also found as epiphytes.These taxa included

Navicula salinicola, Navicula 109, Nitzschia fundi, and Nitzschia frustulum V. subsalina. Taxa that were most abundant in the samples from the SILT site are found at the lower right of Figure 33. These taxa, classified as group 7, were Navicula tripurictata, Opephora schultzi, Amphora 35, Nitzschia frustulum,

Nitzschia 2, Navicula gottlandica, Achnanthes 1IB, Amphora micrometra,

Navicula salina, Navicula groschopfi, Rhopalodia musculus, Fragilaria pinnata, and Nitzschia 47. Species that were common in all three sediments were placed in group 8 and included Opephora pacifica, Opephora perminuta, Amphora coffeiformis, Paralia sulcata, Cocconeis 11C, Navicula gregaria, and

Fragilaria striatula v. californica. -163-

To obtain better resolution in the ordinations and to relate the sample ordinations to patterns in the environmental data set, the data matrix was divided into the epiphytic samples and the sediment samples, and an RA ordination was performed separately on each data subset.

A plot of epiphyte samples relative to the first and second RA axes is presented in Figure 34. The samples from winter and spring are found on the lower left of the diagram. There was a change in community structure in June that caused most of the June samples to be oriented in the upper left of the figure. The epiphyte samples from August and October are located on the right end of RA axis 1, an orientation that also illustrated temporal changes and discontinuities in species composition and relative abundance.

An RA ordination of epiphytic taxa corresponded closely with the pattern of sample ordinations in Figure 34 and helped to illucidate the seasonal changes in community composition (Figure 35). The taxa on the lower left, identified by four-letter acronyms, are Navicula directa, Gomphonema oceanicum, Paralia sulcata, Rhoicosphenia curvata, Cocconeis scutellum, and

Cocconeis costata. These taxa were spring and winter species that had a maximum relative abundance in February or March. Taxa that were abundant in spring and summer are found in the upper left of the figure. These taxa,

Berkeleya rutilans, Nitzschia frustulum v. subsalina, Cocconeis scutellum v.

arva, Nitzschia fundi, Navicula 109, and Cocconeis placentula v. euglypta, were abundant from April through August and had maximum relative abundances from May through July. Taxa that were common from August through October are found in the center of the diagram and to the far right. Rhopalodia musculus,

Navicula 150, Melosira moniliformis, Melosira nummuloides, Bacillaria J J F:0: DECEMBEFEBRUAR YR 0=U:J:4= OCTOBERAUGUSTJUNEAPRIL U U)c'J 60 J L!J U 4>< A A 40 A 0 20 - F F 0 0 F 20D 40 RA AXIS I 80 $00 Figure 34. duringordinationOrdination which methodof the samples samples was reciprocalof wereepiphytic obtained. averaging diatoms andassociated the letters with Zosterarefer to leaves. the month The

the species discussed in the text. the in discussed species the samples illustrated in Figure 34. Figure in illustrated samples The acronyms represent acronyms The Ordination of epiphytic diatom species associated with the with associated species diatom epiphytic of Ordination Figure 35. Figure

RA AXIS I AXIS RA

001 08 40 0 C)

NDIR

NPSU NBRE

NSCH

GOMO

NIT5

PSUL o NDIS NROS CSUJ NSAL RCUR SFAS

BPAR

ccos

N171 BRUT N150 (I) NSUB RMUS

CFR ('.J NFUN MNUM

N109 VldD [SM -166- paradoxa, Nitzschia 171, Opephora perminuta, and Nitzschia rostellata were most abundant in October. Synedra fasciculata, Navicula salinicola, and

Navicula tripunctata v. schizonemoides are found just below and to the left of the center of the diagram and were abundant during all seasons.

An RA ordination of sediment samples and species resulted in a distributional continuum that was related to sediment type. The relative positions of sediment samples and species were similar to the ordinations of the entire data set (Fig. 32 and 33).

In order to interpret the distributions of samples and taxa in relationship to the physical environment, PA axes from the sample ordinations were correlated with environmental variables.

Two RA axes generated from an ordination of epiphyte samples were correlated with tidal height, daylength, and water temperature (Table 32).

The highest correlations were between the second RA axis and daylength (r =

0.90) and between this same axis and temperature (r = 0.85). The first RA axis is weakly correlated with the same variables (r < 0.25). Therefore, the second axis is an expression of seasonal variation in the epiphytic flora, while in the first axis is uninterpretable relative to these environmental variables.

The examination of pattern in the sediment samples relative to the physical environment also was investigated by correlation analysis (Table

33). The first RA axis for the benthic samples was highly correlated with sediment properties, as r = -0.75, -0.68, -0.77 for correlations between this axis and organic matter, mean particle size, and the sorting coefficient, respectively; the first PA axis was correlated, to a lesser degree, with the chlorophyll ratio (r = 0.53). -16 7-

Table 32. Correlations between environmental variables and two PA axes generated from an analysis of epiphyte data. The environmental variables are intertidal height (TIDE), daylength (DAYL), and water temperature (TEMP). The coefficient of determination (R2) is given for the multiple regression of each PA axis against the three environmental variables.

Correlation Coefficients

Variable RA1 RA2

TIDE -0.04 0.25 DAYL -0.05 0.90 TEMP -0.24 0.85

R2 0.13 0.90 -168-

Therefore, this axis contrasts assemblages associated with fine sediments that were poorly sorted, high in organic content, and with low chlorophyll ratios, with assemblages associated with coarse, well sorted sediments that were low in organic content and had high chlorophyll ratios. The stations with high chlorophyll ratios were the high intertidal stations, especially at the SAND site.

The second RA axis was weakly associated with tidal height (r = -0.45) and the chlorophyll ratio (r = -0.37). This axis contrasts low intertidal stations which had relatively low chlorophyll ratios with the high intertidal stations.

A multiple regression analysis indicated that 76% of the variation in the sediment sample scores for the first BA axis was associated with the nine environmental variables (Table 33). Approximately 46% of the variation in sample scores for the second BA axis was associated with the same environmental variables. However, weak correlations of this axis with etwironmental variables made the interpretation of this axis uncertain. -169-

Table 33. Correlations between environmental variables and two RA axes generated from an analysis of sediment diatom data. The environmental variables are intertidal height (TIDE), chlorophyll a in the top cm of sediment (CHLA), organic matter (OM), chlorophyll ratio (RATIO), daylength (DAYL), water temperature (TEMP), mean sediment particle size (PHI), the sorting coefficient (SORT), and the skewness coefficient (SKEW). The coefficient of determination (R2) is given for the multiple regression of eachRA axis against the nine environmental variables.

Correlation Coefficients

Variable RA1 RA2

TIDE -0.01 -0.45 CHLA -0.41 -0.15 OM -0.75 -0.11 RATIO 0.52 -0.37 DAYL 0.03 -0.01 TEMP -0.02 -0.03 PHI -0.68 -0.23 SORT -0.77 -0.25 SKEW -0.07 -0.01

R2 0.76 0.46 -170-

THE ZOSTERA PRIMARY PRODUCTION SUBSYSTEM

The research reported in this section examined the production dynamics of a Zostera marina L. bed in Netarts Bay. Specific objectives Included: (1) a description of the autecology of Zostera in Netarts Bay; (2) an investigation of macrophyte-epiphyte relationships, and (3) the monitoring of the primary production in Zostera and its epiphytes in the intertidal region over a growing season. Again, the process model presented in an earlier section served as a conceptual framework for the research.

The results of the research on the Zostera Primary Production subsystem are presented in six major subsections: I - Materials and Methods; II -

Morphometrics and Autecology of Zostera marina L.; III Production Dynamics of Zostera; IV - Relationship Between Zostera and Epiphytic Assemblages; V -

Bioenergetics of the Zostera Primary Production subsystem; and VI -

Discussion. Since many of the methods and materials for the study were relevant to both the morphometric and production aspects of the research, such information is presented in one introductory subsection, Subsection I. In

Subsection II, patterns of sexual reproduction and vegetative growth of

Zostera in Netarts Bay are presented along with a general description of the plant's natural history and autecology. Subsections III, IV, and V are concerned with the bioenergetics of Zostera marina and associated epiphytes.

I. Materials and Methods

Description of Intensive Study Area

The EPA studies in Netarts Bay, reviewed above on pages , were used to generate hypotheses for further investigations at a finer level of resolu- -171-

tion. A detailed study of the vertical and horizontal nutrient profiles of the water column over a transect from the channel through the seagrass beds to the open mudflat was initiated by EPA. Related biological aspects were examined by the research presented in the next four subsections of this report. Interpretation of the results of the nutrient profile study in relation to the biological processes study required that both aspects be conducted during the same time period at the same location. A site appropriate for both projects was selected on the basis of the results of a field study of the circulation pattern of water over the seagrass beds.

Criteria used in the selection included: (1) the presence of a large expanse of Zostera marina L. that was representative of the Zostera beds in the estuary; (2) evidence of a straight line flow of water over the Zostera beds during an incoming tide; and (3) accessibility for sampling.

The intensive study area in Netarts Bay included approximately 37,500 m2 located within the shellfish reserve and research area managed by the Oregon

State Department of Fisheries and Wildlife (Township 2S, Range lOW, in the vicinity of Whiskey Creek). This region lies near the north end of the large intertidal Zostera bed that occupies the southern and western regions of the bay (Fig. 36).

Three transects over a range of tidal heights were chosen within the intensive study area. These were labeled transects 1 through 3 (Fig. 37 and

38). Transect 1 was 1.1inabove MLLW and was located within the Zostera bed away from the region of obvious influence from the channel. The other transects were located during an incoming tide by placing a stake in the sediment at the water's leading edge when the water at the next lower transect was -17 2-

NETARTS BAY, OREGON

'44

C-) 0 0

0 Q

INTENSIVE STUDY 4REI

Figure 36. Location of Zostera beds and the intensive study area in Netarts Bay. -173-

75m / /

ZOSTERA POND

cc'

SOUTH DRAJNAGE C', 'CHANNEL I OLD -

OYSTER TRANSECT1 BEDS j

NORTH DRAINAGE MAINCHANNEL

Figure 37. Map of the area of intensive study, indicating the location of the sampling transects relative to the main channel. ZOSTERA BED OPEN MUDFLATii - / / i TRANSECT 2 (1.2m) TRANSECT (+1.lm) 0 DISTANCE IN METERS 40 80 120 160TRANSECT 3 (+1.4m) 200 240 Figure 38. Cross section of the area of intensive study indicating the location and height of the sampling transects. -17 5-

15 cm deep. This procedure located transect 2 at 1.2 m above MLLW, and

transect 3 at 1.4 m above MLLW. Transects 1and 3 represented the upper and

lower limits of the Zostera bed within the intertidal region; transect 2

represented a region of transition. Moreover, transect 2 was at the edge of a

large pool of water that was created at low tide by the damming effect of the

larger Zostera shoots in the area. Therefore, this transect was located

between an area that was regularly exposed at low tide, and one that did not

drain completely at low tide.

The transects were established by positioning stakes with an Abney level

and measuring tape at 5 m intervals at the appropriate elevation. Transects

1, 2, and 3 were 75 in, 100 m, and 100 m in length, respectively. The ends of

the transects were located away from any obvious influence of the drainage

channels that bordered the study area. The elevation at each transect was

checked relative to predicted values for the height of the high tide for

Tillamook County beaches. Stakes marked at 1 cm intervals along their length

were placed in the channel at the intensive study area and at each of the

transects. The difference in the depth of the water in the channel at slack

low tide and at the following slack high tide was determined, and the depth of

the water at each transect was measured. The ratio between the predicted

height of the high tide and the measured height was used to calculate the ele-

vation of each transect relative to MLLW.

A series of 0.5 x 0.5 m sample sites were designated along each transect.

Transect 1 had 150 sample sites and transects 2 and 3 each had 200 sample

sites. The sites to be sampled along each transect on a particular date were

chosen from a random number table without replacement. Individual random

number tables of the proper size for each transect were generated by a computer using a random iterative process. -176-

Selection of Quadrat and Sample Size

Choice of quadrat size and sample size for biomass determinations was

based on data collected during the 1979 growing season. Quadrats larger than

400 cm2 were considered unsuitable because the large biomass from sucha

quadrat limited the number of replicates that could be processed before the

material deteriorated. Quadrat sizes from 100 to 400 cm2 were tested. Counts

of shoot density along each transectwere used as an indicator of the variance

among biomass samples. These counts were taken at the beginning and end of

the growing season, i.e., in June and September. In general, variance at all

transects decreased with an increase in quadrat size from 100 to 400 cm2

(Table 34); 400 cm2 (20 x 20 cm) quadratswere used for all subsequent

samples. The harvest of a 400 cm2 quadrat did no obvious, lasting damage to

the system. During the subsequent growing season, harvested areas were

recolonized by vegetative shoots from neighboring regions.

The shoot density data from June and September 1979 alsowas used to

select a suitable sample size. Factors considered in choosing the number of

quadrats per sample for each transect included the level of precision of the

estimate of the mean of the biomass, and the cost in time and materials of

processing the samples. For the intensive study area, it was determined that

a sample size of seven 400 cm2 quadrats had a standard error that was less

than 207. of the mean (Table 34). Also, seven quadrats per transect could be harvested by one person during the low tide in twoor three consecutive days.

In 14 days or less, the plant material fromseven 400 cm2 quadrats per transect could be processed to the point where it could be stored indefinitely without deterioration until all necessary measurements could be made. This -17 7- TRANSECTTable 34. Sample statistics from shoot density data used to determine quadrat size (n = 7). 1 3 JUNQUADRAT SIZE (cm ) E, 1979 2 100 200 300 400 100 200 300 400 100 200 300 400 STANDARDMEAN ERRO R 957104 1186 138 1238 130 1236 117 170529 443109 410 90 518 99 1171 333 224879 208919 1007 193 SEP =STANDARD x% OF MEAN ERRO 11 12 10 10 32 25 22 19 28 25 23 19 MEANTEMBER, 1979 657 686 710 44 721 847 112900 140762 108954 1286 310 1143 265 1052 206 1025 188 =STANDARDSTANTARD X% OF MEAN ERRO ERRO R 1492 1281 6 24 3 182 21 12 18 11 24 23 20 18 -178-

was an important consideratthn in choosing a samplesize, because fresh samples of Zostera could not be refrigerated longerthan 14 days without significant deterioration. Therefore, the standardsample size selected for this study was seven quadrats per transect. However, depending on weather conditions, time and help available for field work,and the biomass of the

Zostera, as few as five quadrats at a transect weresampled at certain times.

Measurement of Biomass

The intensive study site was visited monthly from May 1979 through June

1981. Biomass samples were taken monthly from April 1 to September 8, 1980, on February 12, 1981 and between April 6 and May 31, 1981.

Individual sample sites were located by extending a 5 m rope marked at

0.5 m intervals between two of the stakes along a transect (Fig. 39).A 20 x

20 cm quadrat frame made of polyvinyl chloride (PVC) welding rod was positioned on the right-hand side of the site, approximately 10 cm back from the rope (Fig. 39B). Shoots inside the quadrat were moved away from the edges, and the sediment and rhizome mat along the outer edge of the quadrat were cut through a trowel. Zostera and associated sediment were then placed in a 1 mm mesh sieve, and the sediment washed from the roots with water.

Therefore, coarse particulate organic matter (CPOM) was retained where CPOM was defined as particles larger than 1 mm in diameter. Plants with associated epiphytes were put in labeled plastic bags with a minimum of water, while the litter was retained in the sieve. The remaining sediment in the quadrat was dug out to a depth of about 20 cm below the rhizosphere and was added to the sieve. The sand was washed from the sieve and the remaining litter material was placed in a labeled plastic bag. All material was kept on ice until it could be refrigerated after returning to the laboratory. -'79-

TRANSECT SEGMENT NUMBERS 10 0 200 190 80 I7 160 80 70 60 50 40 30 20 I I I 1

I STAKES AT 5m INTERVALS ALONG A lOOm TRANSECT /

70 60 I I t

á9 I I 70 67 66 63 62 61 I. A 5m INTERVAL WITH 10 O.5rn TRANSECT SEGMENTS

A O.5m TRANSECT SEGMENT (Th H ROPE WITh MARKERS ---- I

A = AREA USED FOR SHOOT PRIMARYPRODUCTION MEASUREMENT

B= PLACEMENT OF 400 cm2 QUADRAT FOR BI0?AASS SAMPLE

Figure39. A 100-rn sampling transect at three levels of spatial resolution: I. 200 0.5-rn transect segments; II. 5-rn segment of transect with 10 0.5-rn transect segments; and III. The details of one sampling site at a transect segment. -180-

During the time each sample was taken, selected physical properties of

the bay and ocean water were measured. Surface water temperature was measured

with a hand-held thermometer. A temperature compensated A0 Goldberg

refractometer was used to determine salinity. Light intensity at the water

surface and at a depth of 0.25 in were measured with a LICOR Underwater

Photometric Sensor and Quantum Radiometer Photometer.These values were used

to calculate extinction coefficients.

Zostera bioniass from each sample quadrat was divided into a series of

subsamples to reduce the biomass of the material to be handled atone time and

to minimize exposure to room temperature. Each subsample was placed in a 1 mm mesh sieve, and individual shoots were cut from the rhizome at the nodes.

This material was transferred to a second 1 mm mesh sieve along with loose leaves, shoots and seedlings.

The sieve containing the leaves, shoots and seedlings, i.e., the aboveground material, was carefully dipped several times intoa basin of distilled water to rinse away salt and any remaining sand. The shoots were sorted and the following information was recorded: the number of vegetative shoots, reproductive shoots, and seedlings; the number of leaves per vegetative shoot; and the number of leaves and flowers per reproductive shoot.Leaves were cut from their sheaths, and the sheaths were placed into labeled 55x 15 mm aluminum weighing dishes. These containers had been bent to fit inside a 60 x

20 mm plastic petri plate. To minimize handling errors, a weighing dish was kept inside a petri dish unless a weight was being determined.The leaves were laid out on a labeled 20 x 20 cm glass plate until an area of approximately 16 x 20 cm was covered. The length and width of the area covered were measured to the nearest 0.5 cm and recorded. The glass plates -18 1-

were covered with leaves and then were coated with distilled water, placed in wooden racks and transferred to a freezer. When the water was frozen, the

plates were wrapped in aluminum foil and stored in a freezer. Reproductive

shoots were processed in the same manner. Stems and flowers were combined with the sheaths from the vegetative shoots.

The sieve with the belowground material, i.e., roots, rhizomes and

litter, was thoroughly washed in a basin of tap water to remove any remaining

sand, and then was dipped several times in distilled water to complete rinsing away the salt. Material from this sieve was sorted by inspection into living

and dead roots and rhizomes, and each fraction was placed into labeled

containers. Other detritus in the sieve was combined with dead rhizome material, along with unsprouted Zostera seeds. Notes were taken on the

quality of the litter.

All sorted material was kept frozen and stored until it was lyophilized.

The process of lyophilization was of particular importance because it

facilitated the removal of epiphytes from the leaves (Penhale, 1977). After

lyophilizing, epiphytes were removed from the leaves and glass plate with a

scraper made from a plastic coverslip glued between two glass microscope

slides with 3-4 mm of Its edge protruding. Epiphytes and leaves were each

placed in containers and the dry weight of all sorted lyophilized material was

determined.

To determine the ash-free dry weight of the leaves, sheaths, roots and

rhizomes, and epiphytes, all the material of one kind from the sample quadrats

at a transect was combined and ground to a powder in a Waring blender. This

homogenized material was subsampled to fill a maximum of three porcelain

crucibles (COORS, size 0). The naterial in the crucibles was weighed, ashed -182-

for 48 hours at 500°C in a muffle furnace, and reweighed to determine the weight of the ash. To determine the time necessary to completely ash the material, crucibles with organic matter were repeatedly ashed until they came to constant weight. The difference between the dry weight of the material before ashing and the weight of the ash was recorded as ash-free dry weight.

Measurement of Primary Production

1. Shoot Marking Method

Net primary production of Zostera shoots were measured by a modified shoot marking technique (Zieman, 1974). Biweekly samples of 14 to 15 shoots per transect were taken, with two or three shoots marked at each sample site

(Fig. 39A). Individual shoots typical of the area were selected. A 0.5 m stake of white PVC welding rod bent into a ring at one end was inserted into the sediment, and the base of a shoot was encircled with the ring.Another

PVC stake extending 15 cm above the sediment was placed in the area of the marked shoots to make the region easy to locate, since the PVC rings at the sediment surface were not readily visible (Fig. 40A). Each shoot was marked by inserting a 22 gauge hypodermic needle above the level of the youngest sheath through all the leaves at the same time. The distance between the PVC ring at the base of the shoot and the needle was measured to the nearest

0.5 cm. After the needle was removed, the number of leaves and lateral shoots were counted. A marker was placed midway along the length of the next to youngest leaf (Fig. 40B). This marker, made from a 20 cm length of one pound test nylon monofilament, had a loop formed by a slipknot at one end; a 1 cm -183--

NEEDLE

LEVEL OF .PLASTIC NEEDLE MiRK BEAD

DISTIANCE SU PKNOT RECORDED

Figure 40. Shoot marking technique. A. Zostera shoot with next to youngest leaf marked with monofilament line. B. The monofilament leaf marker with plastic bead. -184-

length of 15 pound test monofilament was glued to the other end to serve as a needle. To make the marker more visible, a plastic colored bead was threaded onto the line and glued in place below the slipknot. The needle end of the marker was passed through the leaf and through the loop at the other end. The loop then was pulled closed, securing the marker in place around the leaf.

After 12 to 16 days, the shoots were harvested at the level of the PVC ring, and placed in individual, labeled plastic bags until they could be refrigerated at the laboratory. At the same time, new shoots were marked for the next two-week period.

In the laboratory, individual shoots were rinsed in distilled water to remove sand and salt. The leaf with the plastic marker was located; and the number of leaves present, the number of new leaves, the number of leaves lost, and the number of lateral shoots were determined (Fig. 41). The length and width of each leaf above its sheath was measured relative to its position on the shoot. Then the distance from the base of the shoot to the original location of the needle mark was measured, and the leaves were cut free at this point. If the PVC ring obviously had moved during the incubation time, the level of the needle mark in the oldest leaf was used as the reference point, as this leaf usually did not grow during the measurement period.Leaves were laid out in the order of their age, and their needle marks were located. The distance between the original level of the mark (i.e., the cut end of the leaf) and its new position was the leaf's contribution to net primary production. This new growth was removed from the leaf and placed on a 20 x 20 cm labeled glass plate. The length and width of this tissue were measured relative to the position of the leaf on the shoot. Glass plates with this NEW GROWTH = NEEDLE MARK ORIGINAL-LEVEL)F NEED LE MARK OLDEST p-YOUNGEST LEA DIS ARDED IF FOUR LEAVES WERE MARKED 1 ORIGINALLY THEN: LEVELSHOOT OF CUTOFF PVC RING AT 1 LEAFLEAF WAS WAS SLOUGHED PRODUCED Figure 41. Shoot marking technique indicating the above-ground net primary production (new growth) and leaf export. -185-

material were treated in the same manner as those processed with Zostera

leaves for the biomass determination. Epiphytes removed from the leaves were

discarded, and the dry weight of the new growth per shoot was determined.

2. Radioactive Carbon (14C) Method

The relative net primary production of Zostera and the epiphytes was

based on in Situ 14C incorporation measurements (Wetzel, 1974).The Bittaker

and Iverson (1976) design for an incubation chamber was adapted for use with

this method. This chamber also was used for the measurement of algal primary

production; its structure is described in an earlier section of this report

(Fig. 8). Light and dark chambers were used for the field measurements.

Light chambers were constructed of transparent, tubular Plexiglass, while dark

chambers were constructed of white, tubular Plexiglass.The exterior of each

dark chamber was painted silver and wrapped with silver duct tape to insure no

leakage of light.

The '4C-sodium bicarbonate (NaH14CO) used was obtained from New England

Nuclear Corporation as a powder with a specific activity of 7.5 mCi/mmol. The

powder was dissolved in sterile, distilled water in a stoppered volumetric

flask. This solution was standarized in April, 1982 and was stored at 5° C until use.

At each transect, measurements were obtained from two light chambers and

one dark chamber. Monthly measurements were made from June 29 to August 24,

1980 and on May 31, 1981. At each sample site a clump of Zostera was

selected. A 30 cm length of PVC pipe (i.d. = 10.5 cm, o.d. = 11.5 cm) was

placed around the clump and was inserted into the sediment until about 2 cm

remained above the surface. To contain the leaves of the Zostera shoots, a bag made of nylon was fastened around the top of the pipe with a rubber band. When the water from the incoming tide was approximately 50 cm deep at

the sample site, the incubation unit of the chamber was positioned around the net and the PVC pipe. Next, the net was removed, and the incubation unit was

inserted 30 cm into the sediment. When the incubation unit filled with water,

the stirring unit was attached, thereby sealing the chamber. Then 7.3 11Ci of a standardized solution ofNaH14CO3 solutionwas injected into the incubation unit and the time was recorded. An external water sample was taken, fixed with mercuric chloride, and later the total available inorganic carbon in the

sample was determined by the EPA Water Testing Laboratory using a split beam

infrared spectrometer.

During the incubation period, light intensity (PEm2 s1)was measured

continuously using a LICOR Spherical Quantum Sensor and Printing Integrator.

In addition, salinity, surface water and air temperature, and light intensity

at the water surface and at a depth of 25 cm were measured hourly over the

Zostera bed.

After a period of 4 to 7.5 hrs, the stirring unit was removed, and the

time was recorded. The incubation unit then was pulled from the sediment, and

the PVC pipe containing the incubated core of Zostera was removed. The

Zostera core was transferred to a1 mm sieve, and the associated sediment was

washed away from the roots and rhizomes. The Zostera was rinsed in distilled

water to remove the salt, wrapped in foil, placed in a labeled plastic bag,

and frozen quickly on dry ice.

Samples were stored in a freezer until they were lyophilized. Dried

Zostera was sorted into roots and rhizomes, leaves, and epiphytes. The dry -187--

weight of each fraction was determined. Samples were then exposed to

concentrated HC1 fumes to remove residual labeled inorganic carbon (Allen,

1971). Material from each sample was ground to a powder in a Waring blender

and subsamples of 0.1 to 0.2 g were placed in filter paper cones and

compressed into pellets. The pellets were combusted in a Packard TRI-CARB

Oxidizer (Model 306) where evolved CO2 was chemically trapped in 4 ml of

Packard CARBOSORB. This was transferred into a scintillation vial with 12 ml

of Packard PERMAFLUOR V. The vials were counted in a Packard TRI-CARE Liquid

Scintillation Spectrometer System (Model 2425). Counts were corrected for

recovery from the oxidizer (92% for Zostera, 91% for epiphytes), counting

efficiency (59% for aboveground Zostera, 56% for belowground Zostera, 54% for

epiphytes), and background. Assimiliation of '2C was calculated according to

an equation of Pehale (1977), but disintegrations per minute (dpm) were

substituted for counts per minute (cpm).

Data Analysis

Morphometric and biomass data were analyzed by multiple regression

analysis (Snedecor and Cochran, 1967) and principal components analysis

(Cooley and Lohnes, 1971). Data analyses were performed on a Control Data

Corporation CYBER 170/720 computer at the Oregon State University Computer

Center. The computer programs used were part of the REGRESS and MULTIVARIATE

subsystems of the Systems Interactive Programming System (SIPS).

II. Morphometrics and Autecology of Zostera marina L. in Netarts Bay

The vegetative shoot of Zostera marina L. comprises an extensive rhizome

system that bears erect, leafy shoots. The linear leaves with basal sheaths EB

are 3-12 mm broad and up to 12 dm long (Hitchcock and Cronquist, 1973).

The rhizome has a meristem associated with each leaf that is positioned immediately below the node (Tomlinson, 1974). At each node two bundles of roots are formed. Since Zostera persists in natural habitats primarily through vegetative reproduction, a continually active meristem is necessary to maintain the populations. This requirement is termed meristem dependence

(Tomlinson, 1974). The shoot apical meristem produces leaves dichotomously, giving the shoot a laterally, flattened appearance. An internode remains small until the leaf associated with one of the nodes becomes the next to the oldest leaf on the plant. At this time the internode elongates, and the roots are produced at the node. Consequently, the shoot is pushed ahead through the sediment by the growth of the youngest internodes along the rhizome. The life of an internode is about 90 days during the growing season in North Carolina

(Kenworthy, personal communication). New shoots are produced by the development of axillary buds in the nodes of the oldest leaves.With the loss of the leaf whose node produced it, and with the branching of the rhizome, these shoots become independent from the parent shoot.

Sexual Reproduction

Events in the life of Zostera marina L. occurred with seasonal regularity in Netarts Bay. Seeds were found in the sediments throughout the year, but gerimination was restricted to spring (Fig. 42, Table 35), starting between mid-February and early April, and concluding by the end of June. Seedlings were not found in the sample taken on February 12, 1981, but were present at all transects in the samples taken in early April 1980 and 1981. They dis- 100 CsJ -90 SEEDLING DENSITY az70-J(9(I) 0(I)Luw6O 50 N TRANSECT 21 (+(+1.lm) 1.2 m) 2:Lu 20 [I TRANSECT 3 (-ft4m) ii APR1 MAY2 JUNT JUN29 JUL26 AUG24 SEP24 Figure 42. DensityEach point of seedings represents along the meanthe ofthree 5 to transects 7 obse DATE rvat ions.during the 1980 sampling period. 21 25 (10) (25) MAY 30 5 (5) 85 (19) MAY 3 1981 25 70 (14) (24) APR 6

C I 0 FEB 12

0 0 0 SEP 24

C 0 0 AUG 24 -190- 0 0 0 JUL 26

1980 0 C 0 JUN 29 (8) 25 4 29 (4) (11) JUN 1 79 60 36 (26) (15) (13) MAY 2 (9) 25 86 21 (22) (14.4) APR 1 Values in parentheses represent the standard error of mean 5 to 7 observations." Dashes indicate no data available. TRANSECT Table 35. Mean number of seedlings per square meter during the period from April 1980 through Ma! 1981. -19 1-

appeared from the samples at all transects by the end of June. Moreover, seedling data indicated that sexual reproduction did not play a major part in the growth and maintenance of the Zostera bed in Netarts Bay.This is compatible with the conclusion of McRoy (1966) and Phillips (1972). However, although the majority of the new shoots are produced by vegetative reproduction, the seedlings are probably important to the maintenance of the genetic diversity within the population. At no time did seedlings represent more than

7% of the shoot population at all transects (Table 36).The mean percentage of the total shoot density that was represented by seedlings when they were present was 2.7% (S.E. = 0.4%).

In samples obtained during 1980 and 1981, the presence of reproductive shoots was variable (Fig. 43, Table 37). In 1980, flowering shoots were present from April to September. In the low intertidal region, reproductive shoots were first seen in April, extending up to the mean tidal height of the winter higher low tide. Flowers were seen throughout the bed in May, and they first appeared in the samples on June 1at transect 1 and on June 29 at transects 2 and 3. Flowers were absent in the samples taken from all transects on

September 24. In 1981, reproductive shoots observed during February in pools at the higher intertidal region (+1.4inabove MLLW) did not become numerous enough to appear in the samples until May 30. Data related to reproductive shoot densities also suggested that sexual reproduction did not play a major role in the growth and maintenance of the Zostera bed in Netarts Bay. When present, reproductive shoots representd no more than 6% of the total shoot density at all transects (Table 38). The mean percentage of the total shoot density that was represented by flowering shoots when they were present was

2.6% (S.E. = 0.5%). -19 2- Table 36. Percentage of the total shoot density represented by seedlings during the period from April 1980 through May 1981. The percentages were calculated from the ratio of the TRANSECT means of 5APR to 17 observations. MAY 2 JUN 1 JUN 29 1980 JUL 26 AUG 24 SEP 24 FEB 12 APR 6 1981 MAY 3 MAY 30 21 3.02.7 5.13.7 3.10.2 0 0 0 0 1.8 2.5 0.7 Dashes indicate no data available.3 3.2 6.7 3.1 0 0 0 0 2.2 0.5 1.8 csJ REPROD UCTIVE SHOOT DENSITY U) 7o TRANSECT 1(-i-1.lm) >Ig6oIiJ(I) S TRANSECT 3(+14m)2(12m) w00:: 4o wco0c3OIL zIo APR 1 JUN 1 Figure 43. Densitysampling of period.reproductive shoots along the three transects during the 1980 MAY2 Each point represents the mean of 5 to 7 observations. DATEJUN29 JUL26 AUG24 SEP24 40 (19) (28) 100 MAY 30

0 I C MAY 3

C I C 1981 APR 6

C I C FEB 12

C C C SEP 24 5 25 50 (5) (17) (21) AUG 24 -194- (7) (7) 40 14 11 (13) JUL 26 65 29 32 1980 (20) (50) (41) JUN 29 (7) 39 0 0 JUN 1

0 0 0 Values in parentheses represent the standard error of mean 5 to 7 observations. MAY 2

C C C APR 1 May 1981.

N Dashes indicate no data available Table 37. Mean number of reproductive shoots per square meter during the period from April 1980 through TRANSECT

Dashes indicate no data available. data no indicate Dashes

0.8 5.3 3.0 5.7 0

1.0 3.0 2.8

0 0 0 0 o 0 0

0.5 1.9 3.4 2.8

1.1 0- 0- 0- 0 0 0 1Z TRANSECT MAY 2 MAY JUL 26 JUL JUN 1 JUN SEP 24 SEP AUG 24 AUG MAY 30 MAY APR 1 APR MAY 3 MAY JUN 29 JUN

FEB 12 FEB

9 IdV 1980

1861 observations. 7 to 5 of means * from April 1980 through May 1981. May through 1980 April from The percentages were calculated from the ratio of the of ratio the from calculated were percentages The Percentage of the total shoot density represented by reproductive shoots during the period the during shoots reproductive by represented density shoot total the of Percentage Table 38. Table -19 5- -19 Vegetative Growth

Because the majority of the shoots in the bed at any time were vegetative, the results of this study related primarily to the biology of the vegetative shoots of Zostera marina in Netarts Bay.

Short, narrow leaves were produced by each shoot in the autumn, and in this form the plant survived the winter. In Netarts Bay the change to the winter morphology began in September and was complete at the upper intertidal transects (2 and 3) by the end of November and at transect 1by the end of

December. Growth in the leafy portions of the plant resumed sometime in late

February or March when the transition from the narrow-bladed form to the broader-bladed form occurred. The winter leaves were systematically sloughed off as the summer leaves were produced. The leaves produced during the period of transition were usually 1-2 mm wider at the base than at the tip, while the summer leaves had a uniform width along their entire length from May until the change to the winter morphology began in September.

There was a regular pattern in the production and sloughing of leaves from vegetative shoots in Netarts Bay. The time interval between the initiation of two successive leaves on one shoot was termed the plastochrone interval (PT). The Pt for the period between samples was calculated from the expression proposed by Jacobs (1979):

number of shoots marked X observation period in days number of new leaves on marked shoots -19 7-

The time interval between the sloughing of two successive leaves on one shoot was termed the export interval (El). The average El for a sample period was calculated from the expression:

number of shoots marked X observation period in days number of leaves sloughed from marked shoots

Trends for the entire growing season were examined by using data from transects 1 and 3 from June through October 1980, and from April through June

1981 (Table 39). The value of the Pt ranged from 7.0 to 25.6 days, while the value of the El ranged from 7.1 to 23.3 days. In general, the P1 was shorter than or equal to the El during April and May. This resulted in an increase in number of leaves per shoot during that period, because leaves were being produced faster than they were being shed. From June through October, the trend reversed and the El was shorter than the Pt.The net result of this pattern was a reduction in the number of leaves per shoot, since during that period leaves were lost faster than they were produced.A plot of the average number of leaves per shoot against time illustrated the effect of these patterns for the entire growing season in 1980, and for the time measured in

1981 (Fig. 44). The relatively low number of leaves per shoot determined for

June 29, 1980 was probably due to experimental error, as inexperienced new workers were involved in the laboratory work at that time. There was little correlation between number of leaves per shoot and other variables associated with Zostera and its epiphytes (Table 40). Therefore, the number of leaves per shoot was not a linear function of these other variables, but instead, was more closely related to changes in the P1 and El. -19 8-

Table39.Meanplastochroneinterval(P1)and meanexport interval (El)expressed asdays fortheperiod from June 1980 throughJune 1981. Valueslistedare themeans of 14 or15observations.

TRANSECT1 TRANSECT2 TRANSECT3 P1 El P1 El P1 El

1980 JUN14-29 15.0 10.0 8.3 13.6 9.0 7.9

JUN29-JUL13 21.0 8.8 12.9 10.0 13.3 10.7

JUL13-26 18.0 12.0 14.4 14.4 13.0 11.4

JUL26-AUG12 25.0 16.0 20.0 12.8 14.2 9.1

AUG12-24 16.5 19.0 14.7 11.0 21.7 10.8

AUG24-SEP8 17.0 19.0 16.0 13.1 11.7 16.0

SEP8-24 25.6 16.0 19.6 11.0

SEP24-OCT7 22.0 15.0 19.6 12.3 10.5 7.6

OCT7-23 19.0 15.0 12.4 11.4 18.0 18.0

1981 APR6-18 9.0 15.0 10.0 14.0

APR18-MAY3 9.5 9.4 7.0 18.7 -u MAY3-16 12.2 11.8 8.0 7.8

MAY16-30 10.0 11.7 7.0 23.3

MAY30-JUN17 13.0 8.5 9.4 7.1

JUN17-JUL1 18.0 11.6 10.5 8.4

* Dashes indicate no data available. ru LEAVES PER SHOOT TRANSECT2TRANSECT 1 (+1.(1.2m) im) U)0F- TRANSECT3(+1.4m) LI-IiiI -J>Lu z2m0LuLi. Figure 44. AprilAverage through number September, of leaves 1980.per vegetative0 shoot at the three transects during the period from APR I MAY2 JUN 1 Each point represents the mean ofDATE 5 to 7 observations. JUN29 JUL26 AUG24 SEP24 -2 00-

Table4O. Correlation coefficients (r) relating the number of leaves per shoot to selected variables. Values greater than 0.2 are significantly different than zero where P < 0.01.

Variables

Total shoot density(shootsm2) 0.166

2 Average area of a vegetative leaf(cm ) -0.253

Aboveground biomass(g dry wt. m2) 0.075

Belowground bioinass(g dry wt.m2) 0.005 -2 Total Zostera biomass (g dry wt. m ) 0.045

Epiphyte biomass (gash-free dry wt.m2) 0.010

Epiphyte leaf (g dry wt. epiphyte/g dry wt. leaf) 0.072 -20 1

The P1 measured the average length of time that a leaf spent in each position on a shoot. Therefore, the PT multipled by the mean number of leaves per shoot equalled the average lifetime of a leaf on a shoot.To examine the changes in the length of the lifetime of a leaf, the sampling period was divided into growth phases according to changes in the ratio between the P1 and the El (Table 41). Plants along transect 1 exhibited three phases. In

April, the P1 (9 days) was shorter than the El (15 days), in May the P1 and El were equal (each 12 days), and from June through October the P1 (20 days) was longer than the El (15 days). Transect 3 had two phases. In April and May the P1 (9 days) was shorter than the El (13 days), and from June through

October the Pt (14 days) was longer than the El (11 days).Transect 2 was only measured from June through October when the P1 (15 days) was longer than the El (12 days). The average lifetime of a leaf was calculated for each phase for each transect (Table 42). Comparison of the lifetime of a leaf for each transect revealed that from June through October, leaves remained on the shoots in the lower intertidal region (transect 1) longer than they did on shoots in the upper intertidal region (transects 2 and 3). In general, the time a leaf remained on a shoot increased from April to September.

Growth of a leaf in relation to Its age or position on the shoot also was anlayzed. Data were standardized by expressing the amount of growth of each leaf as a proportion of the total growth of the shoot. Since the absolute age of each leaf measured was not known, relative ages were used. The youngest leaf on a shoot was designated as number one, and each successive leaf was numbered in order from the youngest to the oldest. This avoided problems when comparing shoots with different numbers of leaves, because any group of leaves -2 02-

Table 41. Mean plastochrone interval (P1), mean export interval (El), andmean number of leavesper shoot (LVS)for periods fromAprilto June, 1981and from Juneto October,1980. P1and El are expressedas days. The values inparenthesesrepresentthe standard error of the meanfor 9 to26 observations.

1981 1980 APRILTO JUNE JUNETO OCTOBER TRANSECT P1 El LVS P1 El LVS

12.0 11.4 3.7 19.9 14.5 2.8 (1.4) (0.9) (0.1) (1.3) (1.2) (0.1)

- - - 15.3 12.3 2.7 - - - (1.3) (0.5) (0.4)

3 8.7 13.2 4.0 13.9 11.4 2.9 (0.6) (2.7) (0.1) (1.5) (1.3) (0.1)

TOTAL 10.3 12.3 3.9 16.5 12.8 2.9 BED (0.9) (1.4) (0.1) (0.9) (0.8) (0.1)

Dashes indicate no data available. -203-

Table 42. Mean lifetime of a leaf (days) for each transect during the growing season. P1 = mean plastochrone interval; El = mean export interval; LVS = mean number of leaves pershoot.*

PIEI

TRANSECT 1 Period April May June-October P1 9.2 11.1 19.9 LVS 3.7 3.6 2.8 Lifetime 34.0 40.0 55.7

TRANSECT 2 Period - June-October P1 - - 15.3 LVS - - 2.7 Lifetime - 41.3

TRANSECT 3 Period April-May June-October P1 8.7 13.9 LVS 4.0 2.9 Lifetime 34.8 40.7

Dashes indicate no data available. -204-

in the same position relative to the youngest leaf on their respective shoot

were approximately the same age. At all transects throughout the study the

greatest proportion of growth occurred while a leaf was in position 2 (Table

43). During the first phase of growth (April and May) the youngest and next

to youngest leaf accounted for 48 to 65% of the total growth of the shoot.

During the last phase of growth (June to October), these leaves accounted for

75 to 95% of the growth of the shoot. As the growing season progressed, the

number of leaves per shoot decreased, the Pt became longer, the average

lifetime of a leaf became longer, and an average leaf spent more time in each

position. Therefore, as fall approached, more of the growth occurred at

positions 1 and 2 and less occurred at position 3, and more than half of the

total growth of a shoot was due to the growth of leaves two Pt or less old.

III. Production Dynamics of Zostera

Biomass

The general pattern for the accumulation of the total biomass of Zostera,

i.e., the summation of aboveground and belowground biomasses, was similar at

transects 1 and 3 (Fig. 45). In 1980, total biomass increased along both

transects through the spring to a maximum in July, and then declined (Table

44). The maximum total biomass at transect 1 was 463.1 g dry weight if2,

while that at transect 3 was 142.8 g dry weight if2. In the spring of 1981,

the rate of increase of total biomass at transects 1and 3 was higher than the

rate of increase in the spring of 1980 (Table 44). Total biomass at transect

1 on May 30, 1981 was equivalent to that on June 29, 1980. Total biomass at

transect 3 on May 30, 1981 was equivalent to that on July 26, 1980. In -205-

Table 43.Percentage of total shoot growth distributed among leaves of different relative ages,wherel youngest leaf on shoot. Values in parentheses represent the standard error of the mean of 14 or 15 observations.

APRIL-MAY JUNE JULY-OCTOBER

TRANSECT 1

%1 21.0 33.7 41.5 (1.4) (3.8) (1.9) %2 41.3 46.3 50.0 (2.6) (2.6) (1.7) %3 28.5 19.0 8.5 (1.3) (1.5) (1.3) %1 + 2 64.8 80.0 91.5 (0.5) (2.1) (1.3) %1 + 2 + 3 93.3 99.0 99.9 (1.3) (0.6) (0.1)

TRANSECT 2 c 35.9 (2.2) %2 48.4 (2.0) %3 13.6 (2.3) %1+2 84.3 (3.6) %l+2+3 97.9 (1.3)

TRANSECT 3

%1 14.3 25.7 27.1 (2.8) (4.3) (3.8) %2 34.3 43.7 48.1 (2.0) (3.2) (1.2) %3 32.5 24.3 22.1 (1.9) (4.3) (3.0) %1 + 2 48.5 72.7 75.3 (4.0) (2.0) (3.7) %1 + 2 + 3 81.0 96.7 97.4 (3.5) (2.0) (0.9)

Dashes indicate no data available. -206-

Table 44. Mean total fro Zos tera biomass, expressed as g 30, 1981. dry weight m2 at the three transects observations.during the Theperiod standa rdm Apr11error 1 of 1980 to Maythe mean is 1980 in parentheses.* The values are the means of 5 to 7 1981 TRANSECT APR259.5 1 MAY269.9 2 JUN321.8 1 362.5JUN 29 463.1JUL 26 AUG321.2 24 SEP252.0 24 119.6FEB 12 APR292.6 6 MAY 305.43 MAY353.9 30 (25.8) 45.6(9.9) (12.4) 84.0 (21.4)(31.4) 82.0 113.9(34.3)(54.2) 147.3(23.4)(23.1) 128.0(24.3)(11.5) 166.9(14.4)(22.5) (15.9) (38.1) (12.4) (25.1) *Dashes indicate no data available. (13.9) 64.5 (12.0) 45.4 65.4(5.0) 103.6(38.2) 142.8(26.8) (19.3) 88.7 (22.0) 57.5 (15.7) 47.1 (17.5) 62.7 (29.2) 63.8 147.5(39.4) -207-

contrast, total biomass of transect 2 increased throughout the summer of

1980. The pattern for the increase of total biomass at transect 2 during 1980 closely followed that for transect 3 until July 26. Subsequently, total biomass at transect 3 decreased, while that for transect 2 continued to increase (Fig. 45). Field notes taken during April and May described a sparse, patchy distribution of shoots at transects 2 and 3; evidence of erosion and burial of shoots were also noted. An impoundment of water in the low intertidal region was formed by the large, dense Zostera shoots grawing in a basin created by sand deposition along the edges of the bed. This area was termed the Zostera pond (Fig. 37). Its boundaries were made obvious during the lowest tides by the presence of about 15 cm of water, and during high tides by calm water in the region, caused by the dampening effects of the

leaves. As the summer progressed and the shoots increased in size throughout

the bed, the boundaries of the Zostera pond increased to include transect 2, and that transect began to look more like transect 1. Furthermore, at

transect 3, plants growing in areas where pools formed were in better

condition than plants growing in areas completely exposed at low tide.

When total biomass was compartmentalized into aboveground and belowground biomass, some new patterns emerged. Aboveground and belowground biomass at

transect 2 and aboveground biomass at transects 1 and 3 reflected the pattern described for total biomass, while the pattern for belowground biomass at

transects 1 and 3 was different (Tables 44, 45, and 46). In 1980, at transect

1, belowground biomass changed relatively little from April 1 to June 29

(Table 46). A maximum belowground biomass of 206.9 g dry weight m2 was recorded a month later on July 26. Belowground biomass decreased during late 500 TOTAL ZOSTERA BIOMASS N 3(+1.4m)2(+1.2m)I (+1.lm) aw0 E 200300 APR1 MAY2 JUN1 JUN29 JUL26 AUG24 SEP24 FEB12 APRG 1981 MAY3 MAY30 Figure 45. toTotal Ma y 30, 198Zos tera biomass1. at three transects in Netarts Bay Eachduring point the representsperiod from the 1980April mean 1,of 19805 to 7 observations. DATE -209-

summer and fall, and increased again between February 12 and April 6, 1981.

In contrast to the rapid accumulation of total and aboveground biomass observed at transect 1 in the spring of 1981, belowground biomass did not reach a level comparable to that of spring 1980 until May 3, 1981. At transect 3 there was a small peak in belowground biomass in April during both

1980 and 1981, followed by a larger peak that occurred with the maximum aboveground biomass (Tables 45 and 46). In 1980, belowground biomass declined during late summer, fall and winter.Therefore, an increase in belowground biomass during April 1981 preceded the increase in aboveground biomass at transect 3.

There was a striking difference between aboveground biomass at transects

1 and 3 during 1980 and 1981 (Table 45). At transect 1, the aboveground biomass for May 30, 1981 was 25% higher than that recorded on June 1, 1980.

At transect 3, the aboveground biomass on May 30, 1981 was 48% higher than the maximum aboveground biomass reached in 1980.

In general, at all three transects belowground biomass was a more con- servative factor than aboveground biomass. The percentage increase of aboveground biomass was compared to that for the belowground biomass for the period from April 1to September 24, 1980 (Table 47). The percentage increase of aboveground biomass ranged from 161% at transect 1 to 714% at transect 3. The percentage increase of belowground biomass ranged from 55% at transect 1to

185% at transect 2. Therefore, the percentage increase of aboveground biomass was much higher than the percentage increase of belowground biomass at all transects. -2 10- Table 45. Mean abovegroun d Zostera b 980ioinass to expressed as g dry 1981. The weightvalues arem2 atthe the means of 5 three transects to 7 observations. during Thethe standardperiod from er ror of theApril 1, 1 mean is 1980 May 30, in parentheses.* 1981 TRANSECT APR 1 98.7 MAY114.8 2 162.1JUN 1 JUN197.8 29 JUL256.2 26 173.9AUG 24 119.4SEP 24 FEB 12 57.6 APR166.5 6 MAY181.0 3 MAY202.4 30 (10.9) 11.3(2.7) 26.6(3.1)(5.4) (21.4) 37.8(9.3) (19.9)(37.0) 60.8 (10.0)(21.1) 71.0 (10.8) 53.5(7.6) (13.0) (7.8)70.4 (7.9) - (21.6) - (5.1) (16.6) Dashes indicate no data available. (1.0) 7.2 (2.6) 8.1 30.3(4.1) (21.6) 49.5 57.1(8.7) 34.9(7.3) 22.0(8.3) 22.3(7.7) 21.6(8.4) (14.1) 32.4 (23.5) 84.4 -211- Table 46. Mean belowgroun Thethe standardperiod from er rord Zostera of theApril 1, biomass1980 to expressed Maymean 30, is 1981.inas parentheses.*g dry weight m2 at the three transects during The values are the means of 5 to 7 observations. TRANSECT 160.8APR 1 154.1HAY 2 151.7JUN 1 164.4JUN 29 1980 206.9JUL 26 147.2AUG 24 132.5SEP 24 FEB12 62.0 APR126,1 6 1981 MAY3124.4 MAY30151.6 (15.2) 34.3(7.4) (14.5) 58.5(1.0) (12.6)(14.1) 44.2 (16.0)(18.9) 53.1 (15.5)(11.7) 76.3 (14.3) 74.5(5.0) (14.6) 96.6(8.1) (8.2) - (18.5) - (7.8) (12.1) Dashes indicate no data available. (10.4) 71.7 37.3(9.8) (5.2)35.1 (17.6) 54.1 (22.3) 85.6 (12.5)53.7 (14.3) 35.5 24.8(8.3) (12.1) 41.1 (15.3) 31.4 (16.6) 63.2 Table 47. Percentage1, 2, and increase3 during ofthe aboveground period from and April belowground 1 to September biomass 24, along 1980. transects Biomass wasthe expressed ratio of asthe g meansdry weight of 5 tom 7 observations. Aboveground Blomass 2 The percentages were calculated from Belowground Biomass Transect 98-256Range Difference 158 % Increase 161 133-206 Range Difference 73 % Increase 55 11-71 7-57 6050 546714 35-8534-97 6350 143185 -2 13-

Comparing aboveground and belowground biomass as a proportion of total

biomass gave additional insights into seasonal trends.At all transects,

belowground bioinass comprised the greater proportion of the total biomass at

the beginning and end of the sampling period in 1980 (Table 48). At transect

1, both aboveground and belowground biomass accounted for between 40 and 60%

of the total biomass over the entire sampling period, each fraction averaging

approximately 50%. In contrast, belowground biomass at transects 2 and 3

always comprised more than 50% of the total biomass in 1980, and April 1980 it

represented 90% of the total biomass at transect 3 (Table 48).

Changes in total biomass were the result of changes in shoot density and

leaf or plant size along the transects. Leaf size was monitored as the change

in the average area of a leaf from a vegetative shoot.Mean plant size was

calculated by dividing total Zostera biomass by mean shoot density. Multiple

regression analysis of the relationship between total biomass and plant size

and density, and between total biomass, and leaf size and density indicated

that there was a linear relationship between the dependent and independent

variables (Table 49). Plant size and density were better predictors of total

biomass than were leaf size and density. This was understandable because

plant size included both aboveground and belowground portions of the plant,

while leaf size represented only the aboveground parts. These relationships

can be used to obtain an estimate of total biomass through a combination of

field counts of density and an estimate of leaf or plant size rather than by harvesting and sorting material from quadrats.

Average leaf size expressed as cm2 increased from April 1to August 24,

1980 at all transects (Fig. 46, Table 50). The relatively small leaf size measured on September 24, 1980 reflected the completion of the sloughing of -214- Table 48. Percentagetransects offor total the periodblomass from corresponding April 1, 1980 to abovegroundto May 30, 1981.and belowground material at the three The percentages were calculated from the ratio of the means ofAPR 5 1to 7 observations.* MAY 2 JUN 1 JUN 29 1980 JUL 26 AUG 24 SEP 24 FEB 12 APR 6 1981 MAY 3 MAY 30 TRANSECT 1 BelowgroundAboveground 62.137.9 56.543.5 50.449.6 46.353.3 44.955.1 45.954.1 52.747.3 51.848.2 43.156.9 40.759.3 42.857.2 TRANSECT 2 BelowgroundAboveground 73.926.1 68.331.7 50.449.6 50.449.7 49.950.1 57.740.9 42.058.0 TRANSECT 3 Aboveground 9.9 21.6 46.8 41.9 44.1 41.7 37.5 47.2 34.4 50.8 57.2 Dashes indicateBelowground no data available. 90.1 78.4 53.2 58.1 55.9 58.3 62.6 52.5 65.6 49.2 42.8 -215-

Table 49.Multiple regression of mean total Zostera biomass (TZOS) against mean leaf size (AREA), mean plant size (SIZE), and mean shoot density (TSHD) for each transect. R2 is the coefficient of determination.

TRANSECT R2 MODEL

0.66 TZOS 18.54 + 6.73 AREA + 0.07 TSHD

0.80 TZOS -101.41 - 1.29 SIZE + 1.22 TSHD

0.86 TZ?DS 69.74 + 7.36 AREA - 0.03 TSHD

0.98 TZOS 78.04 + 1.08 SIZE + 0.07 TSHD

0.84 TZOS -27.37 + 0.53 AREA + 5.00 TSHD

0.98 TZOS -97.63 + 0.01 SIZE + 1.05 TSHD LEAF SIZE TRANSECT (i-1.lm) 3O40 TRANSECT (1.4m)(+1.2m) a:Lii 20 0 APRI MAY 2 JUN1 JUN 29 AUG 24 JUL26 SEP24 FEB12 3MAY30 Figure 46. bothMean7 observations. sides,leaf size along as theexpressed three transectsby the average in Netarts area ofBay. a leaf from a vegetative shoot, considering 1980 DATE Each point represents the mean of 5 to as - 7.2 (1.8) (0.7) 14.5 MAY 30 - 8.0 5.7 (0.6) (0.2) The values MAY 3 1981 - 5.5 3.3 (0.5) (0.6) APR 6 - 3.0 2.3 (0.11) (0.3) FEB 12 1, 1980 to May 30, 1981. (2.4) (3.8) (2.1) 30.3 16.4 10.6 The standard error of the mean is in parentheses. (4.0) (2.0) (2.9) 42.1 18.6 12.1 AUG 24 -2 17- (1.6) (1.7) (1.3) 37.2 15.1 12.0 JUL 26 1980 7.6 (4.6) 36.7 (1.8) (1.3) 12.0 JUN 29 6.0 5.8 (1.7) (0.6) (0.4) 12.3 JUN 1 4.6 2.5 (1.8) 15.0 (0.2) (0.3) MAY 2 4.8 2.5 1.8 (0.4) (0.2) (0.3) represent the means of 5 to 7 observations. cm2 for the three transects period from April APR 1

I Dahses indicate no data available. Table 50. Mean leaf size measured as the average area of a from vegetative shoot and expressed TRANSECT -2 18-

the large leaves produced during the sampling period in 1980 and the beginning of the change to the small, narrow leaves typical in the winter. In the spring of 1981, at transect 1, leaf size increased at a lower rate than it did in the spring of 1980. Mean leaf size on Nay 30, 1981 was similar to that for

May 2, 1980 (Table 50). In contrast at transect 3, mean leaf size increased at a higher rate in the spring of 1981 than it did in the spring of 1980.

Mean leaf size on May 30, 1981 was similar to that for June 29, 1980 (Table

50).

The changes in plant size were similar in pattern to changes in total biomass at all transects during the sampling period in 1980 (Fig. 45 and

47). En the spring of 1981 at transects 1 and 3, plant size increased at a lower rate than in the spring of 1980. At transect 1, the mean plant size on

May 3, 1981 was similar to that measured on April 1, 1980, and at transect 3 the mean plant size for April through June 1, 1980 was not reached until May

30, 1981 (Table 51).

Differences between the leaf and plant sizes measured in 1980 and those measured in 1981 can be explained in conjunction with the changes in shoot density. At transects 2 and 3, during the sampling period in 1980, shoot density, plant size, leaf size and total biomass follow the same general patterns of change (Fig. 45, 46, 47 and 48). At transect 1, during the same period, shoot density decreased as plant size and leaf size increased. Total bioniass was greater in transect 1 than at transects 2 and 3 in April 1980 because the shoot density at transect 1 was three times greater than that at transects 2 and 3 (Fig. 45 and 48) when plant size was similar at all three transects (Fig. 47). Although shoot density decreased from April through rr.i PLANT SIZE RIANSECT Cl) 300 RANSECTRIANSECT 2(l.2m)3(+1.4m)1(-H.lm) J-J 200 .1 10101 0 APR1 MAY2 JUN29 AUG24 JUN1 JUL26 SEP24 FEB12 APR6 1981 MAY3 MAY30 Figure 47. observations.periodMean size from (mg) April 1, 1980 to May 30, 1981. of a vegetative plant along the three transects in1980 Netarts Bay during the DATE Each point represents the mean of 5 to 7 -2 20- Table 51. Meanratioperiod size of from ofthe a April meansvegetative 1,of 19805 toplant to7 observations.May expressed 30, 1981. as milligrams at the three transects for the The percentages were calculated from the TRANSECT APR 1 MAY 2 JUN 1 JUN 29 1980 JUL 26 AUG 24 SEP 24 FEB 12 APR 6 1981 MAY 3 MAY 30 12 58.788.7 127.0 70.7 160.9 88.2 119.4194.9 105.2330.8 313.4 68.8 152.4253.3 66.8 76.1 88.7 104.2 Dashes indicate no data available. 81.7 84.7 81.4 101.1 104.4 90.0 115.0 42.0 54.5 70.1 88.8 i.IiI. SHOOT DENSITY 3000 TRANSECT 3(i-1.4m2(+1.2m1(+lim) 0F-(I) 2000 (I)0 [I$IIS] 0 APR1 MAY2 JUN29 AUG24 I JUNI I 1980 I JUL26 SEP24 FEB 12 I I I V I I APR6 I MAY I 3 MAY30 I I - - - Figure 48. Meanfrom shoot April density 1, 1980 (shoots to May m2) 30,- along 1981.- the three- transects in Netarts Each point represents the mean of 5 to 7 observations.DATE 1981 Bay during the period -222-

September at transect 1, total biomass continued to increase in response to the increase in leaf and plant size until July 26 (Fig. 45, 46, 47 and 48).

Therefore, the data indicated that shoot density was independent of plant size until some threshold was reached, above which plant size has a negative effect on density.

Mean leaf area per unit area of substrate expressed as m2 m2 measured the combined effects of shoot density and plant size. Both sides of the leaf were considered in this measurement. When plant size and density combined to produce a mean leaf area per unit of substrate of between 7.5 and 11 m m2,

shoot density began to decrease at transect 1 (Tables 52 and 53). Transects 2 and 3 never reached this mean leaf area per unit area of substrate. This response was probably the result of self-shading within the Zostera bed in the region of transect 1. In addition, plant sizes were relatively large in the samples taken on June 29 and July 26, 1980 at transect 1. This suggested that there had been a selective loss of small shoots prior to that time, possibly due to shading from the large shoots.

Data from all transects for 1980 and 1981 were pooled to examine the properties of the entire Zostera bed. Sample statistics in the variables of interest were listed in Table 54. Relationships that existed for each transect were still evident when the data were pooled (Table 55). Aboveground and belowground biomass were highly correlated with each other and with total biomass. Total shoot density and the average area of a vegetative leaf were less highly correlated with total biomass. However, since they represented components of biomass, i.e., number of shoots and size of aboveground parts, when they were combined in a multiple regression against total Zostera biomass they -2 23- Table 52. Mean shoot density expressed as Valuesshoots m2 at the the meansthree transects during the of 5 to 7 observations. period Standardfrom Aprilerror of1, the 1980 mean is to May in30, parentheses.1981. 1980 represent 1981 TRANSECT APR 29251 MAY 22125 JUN 12000 JUN 291860 Jul 261400 AUG 241025 SEP 24 995 1790J1D IL LtL AITh 3845 U L'Utl 3445 ) IiLtlA1 3395 J'J (148)(293) 789 (134)(176) 1188 (212) (97) 929 (236)(253) 954 (295)(156) 1400 (162)(160) 886 (120)1095(87) (136) - (387) - (143) - (359) - Dashes Indicate no data available. (148) 782 (140) 536 (111) 804 (308) 1025 (140) 1368 (228) 896 (185) 500 (359) 1135 (436) 1150 (428) 910 (387) 1761 6.0 (1.2) (1.5) 15.9 MAY 30 during observa- 2.4 (0.4) (0.9) to 7 11.0 MAY 3 1981 7.5 1.4 (1.3) (0.5) APR 6 three transects means of 5 1.8 0.9 (0.1) (0.3) at the FEB 12 m2 7.5 4.6 1.5 (0.8) (0.6) (0.6) as SEP 24 Values represent the 4.1 2.3 (0.9) (0.8) (0.1) 11.1 expressed AUG 24 1981. parentheses.* -224- is In 5.4 4.6 (1.9) (0.9) (0.6) 15.7 May 30, JUL 26 substrate 1980 the mean 4.1 3.0 (2.4) (1.1) (1.3) 1980 to 14.4 area of JUN 29 9.1 2.4 1.7 (1.1) (0.6) (0.2) error of April 1, JUN 1 1.8 0.5 (1.4) (0.3) (0.2) 11.9 Standard MAY 2 4.9 0.6 0.3 (0.6) (0.1) (0.1) the period from tions. APR 1

I-, Dashes indicate no data available. Table 53. Mean leaf area per unit TRANSECT -225-

Table 54. Sample statistics for Zostera biomass and selected morphometric data corresponding to the entire Zostera bed over the whole sampling period from April 1 to September 24, 1980 and from February 12 to May 30, 1981 (n = 175).

Standard Mean Error Range

Total Biomass2

(g dry wt. m ) 165.3 9.5 0-554

Aboveground Bomass

(g dry wt. m ) 79.2 5.5 0-326

Belowground Bi9mass

(g drywt. m ) 86.0 4.4 0-240

Shoot Density (shoots m2) 1454.1 75.9 0-4950

Average Area of Vegetative Leaf* Lcm2) 11.5 0.8 0-57

Both sides considered. -2 26-

Table 55.A matrix of correlation coefficients relating Zostera biomass and selected morphometric variables for the pooled data set, i.e., all samples from the three transects. TZOS = total Zostera biomass; ABGB = aboveground biomass; BLGB = belowground biomass; TSHD = total shoot density; AREA = average area of a vegetative leaf, both sides considered. Values greater than 0.2 are significantly different than zero where P < 0.01.

TZOS ABGB BLGB TSHD AREA

TZOS 1.00 0.97 0.95 0.67 0.64

ABGB 1.00 0.84 0.65 0.65

BLGB 1.00 0.63 0.56

TSHD 1.00 0.03

AREA 1.00 -227-

accounted for a large proportion of the variance in total biomass (Table

56). Again, this supported the idea presented earlier that measurement of shoot density and the average area of a vegetative leaf can be used to obtain an estimate of total Zostera biomass.

Primary Production

Shoot net primary production (g dry weight m2 sample period), as measured by the leaf marking method, had a similar pattern at all transects throughout the study (Fig. 49, Table 57). During the period from June 14 to

October 23, 1980, there were two peaks in net primary production of the shoot. The first occurred between June 29 and July 26 and was reflected by peaks in total biomass at transects 1and 3 (Fig. 45).

A second and smaller peak in net primary production of the shoot occurred between August 24 and September 24. This maximum was not reflected in the measurements of total bioinass at transect 1 and 3 for the same period because of a decrease in shoot density and plant size that occurred at the same time

(Fig. 47 and 48). Total biomass at transect 2 increased throughout the sampling period in 1980. The decrease in net primary production of the shoot at all transects between July 13 and August 24 was concurrent with a bloom of

Enteromorpha prolifera in the area (see Davis, 1982). As the Enteromorpha drifted through the Zostera bed with the incoming or outgoing tide, the long leaves of the seagrass became tangled in the ropes of the alga, and the

Zostera was uprooted as it was dragged along. In areas where Enteromorpha had become caught in the sediment or on some object, the Zostera in the vicinity became buried from an increase in the sedimentation caused by the slowing of -2 28-

-I Table 56.Multiple regression of total Zostera biomass (TZOS) against total shoot density (TSHD) and the average area of a vegetative leaf (AREA) for the entire Zostera bed. R2 is the coefficient of determination.

MODEL

0.40 TZOS = 43.2 + 0.08 TSHD

0.40 TZOS = 81.8 + 7.3 AREA

0.83 TZbS= -36.0 + 0.06 TSHD + 4.2 AREA JUL1 1981 MAY3 APR6 tAY3O DATE SHOOT Each point represents the mean of 14 or 15 observations. OCT23 SEP24 AUG24 1980 JUL26 TRANSECT2(+I.2m) TRANSECT3(1.4m) 'TRANSECT 1 (il.lm) JUN29 JUNI 0 0 0 0 0 (\J E I w >- clOO N Shoot net primary production expressed as g dry weight m2 along three transects for the period from June 1980 to July 1981, Figure 49. -2 30- Table 57. valuesperiod1;JulyShoot 1, representnet1981. the NPP-DAYprimary = productionshoot net along the three transects duringNPP-PER = shoot means of 14primary or 15 productionobservatIons.net primary expressed production expressed as g dry as gthe dry period weight nr2 dayl. weightfrom June 14, 1980 m2 sample The to Sampling 1980 Period JUN 2914- JULJUN 29-13 JUL 13-26 AUGJUL 1226- AUG 12-24 AUGSEP 24- 8 SEP 248- OCTSEP 724- OCT 237- TransectTransect 12 NPP-PERNPP-DAY 129.1 47.6 8.6 153.4 66.010.2 57.181.1 6.2 69.332.0 4.3 31.741.7 3.5 35.353.8 3.4 43.468.7 4.3 31.061.4 4.4 20.244.4 2.8 Transect 3 NPP-DAYNPP-PER 36.0 2.63.2 39.5 4.42.5 48.7 3.74.8 36.3 2.32.0 13.5 2.6 2.2 1.9k2.7 * 27.1 2.2 14.8 1.3 Sampling 1981 Period APR 6-18 MAYAPR 18-3 NAYMAY 3-16 MAY 3016- MAYJUN 30- 17 1.0 JULJUN 117 0.1 1.9 0.9 Transect 1 NPP-DAYNPP-PER 108.9 9.1 113.8 7.6 122.2 9.4 150.0 10.7 230.5 13.6 114.2 8.2 Transect Based on data from 3 NPP-DAYNPP-PER two shoots. 22.5 1.6 30.0 2.1 39.6 2.8 65.1 4.7 42.1 2.5 37.2 2.7 -231

the water by the tangled mass of the Enteromorpha. Eventually even the

Enteromorpha was buried, and an oozing, sulfurous mound of sediment marked the

place where there had been a patch of Zostera. The disappearance of

Enteromorpha from the Zostera bed in early August was followed by an increase

in net primary production of the Zostera shoots (Fig. 49).

The rate of shoot net primary production was greater in 1981 than in 1980

(Fig. 49). The maximum rate of shoot net primary production was reached in

early June in 1981 as compared to early July in 1980. At transect 1, the

maximum rate of shoot net primary production in 1981 was 50% higher than that

of 1980, while that for transect 3 was 21% higher (Table 57).

At transect 1, the accumulation of total biomass was divided into five

phases. Between February 12 and April 6, 1981, there was a large increase in

total Zostera biomass that was the result of a large increase in shoot density

(Fig. 45 and 48). During the same period of time, mean leaf size increased

slightly from 3.0 cni2 to 5.5 cm2. Mean plant size also increased slightly

from 66.8 mg to 76.1 mg. A similar pattern was noted in 1980. Presample data

from February 23, 1980 gave a mean density of 1387.5 shoots m2 witha

standard error of 143.9 shoots m2. Subsequently, density increased to 2925

shoots m2 by April 1, 1980 (Table 52). Therefore, primary production in the early spring at transect 1 was first channeled into the production of new shoots. Then the change in leaf morphology from small, narrow leaves typical in the winter to long, wide leaves typical in the summer occurred during April and May 1981. It was only after this change was completed that the sharp increase in shoot net primary production was initiated during the month of May

1981, thereby making a shift in growth strategy to the production of above- -2 32-

ground parts. The sharp decrease in net primary production of shoots observed at transects 1and 3 during June 1981, coincided with a bloom of Enteromorpha, just as the decrease in net primary production of shoots during July and

August 1980 coincided with a similar bloom. Therefore, the same pattern of growth was followed during both 1980 and 1981. The stages involved in this general pattern are summarized in Figure 50.

Conditions for the growth of Zostera were better at transect 1 than at transects 2 and 3. Maximum mean total biomass, mean shoot density, mean leaf size, mean plant size and mean shoot net primary production at transect 1 were always greater than at transects 2 and 3 (Table 58). During the summer low tides, Zostera growing in the region of transect 1 was never exposed. Even during the lowest tides, 10 to 15 cm of water remained in the Zostera pond.

This layer of water protected the plants from exposure to the air. This conclusion was further supported by the relatively high shoot density and relatively large shoots observed in areas with pooled water during low tide at transect 3. In the winter, there was a noticeable difference in the density and condition of plants located above and below the level of the higher low tide. Transect 1 was located below this elevation, while transects 2 and 3 were located above it. Shoots in the region below the average height of the winter higher low tide appeared to be protected. These shoots were exposed to the air during low tide and no more than once a day during the winter, while the rest of the Zostera bed was exposed twice each day. The edges of the bed showed evidence of erosion, and plants in this area often had their rhizomes exposed as the sand around them was washed away. During the entire year there was evidence of sand deposition in the high intertidal region, especially -233-

DECEMBER TO WINTER GROWTH FORM FEBRUARY (SHORT, NARROW LEAVES)

MARCH INCREASE IN SHOOT DENSITY

CONVERSION TO SUMMER APRILaMAY GROWTH FORM LONG, 'NIDE LEAVES)

LATE MAY HIGH RATE OF GROWTH OF THROUGH AUGUST ABOVEGROUND PARTS

SEPTEMBER CONVERSION TO WINTER THROUGH NOVEMBER GROWTH FORM (SHORT, NARROW LEAVES)

Figure 50. The annual phases of growth exhibited by Zostera marina L. growing along transect 1 in Netarts Bay, Oregon. -234- Table i8. Maximumdensity mean along values the transectsof total biomass,in 1980 andleaf 1981. size, shoot size, and shoot Total Biomass Leaf Size Shoot Size Shoot Density 1980 (g dry wt. nr2) (cm2) (mg) (shoots m2) Transect 21 186463 42.118.6 152.4313.4 14002925 1981 Transect 3 143 12.1 115.0 1368 Transect 13 148354 14.5 7.2 104.2 88.8 16613845 -235-

along transect 3. Neither erosion nor sedimentation was as evident in the region of the bed where transect 1 was located. Therefore, a combination of

factors apparently contributed to the differences between plants that grew in

the region of transect 1 and those located in the region of transects 2 and 3.

IV. Relationship Between Zostera and Epiphytic Assemblages

Description of Epiphytic Assemblages

Epiphytic assemblages on Zostera in Netarts Bay were primarily composed of diatoms. Epiphyte biomass expressed as ash-free dry weight m2 from pooled data representing the entire Zostera bed had a high correlation with biomass expressed as dry weight. The mean ratio of ash-free dry weight to dry weight was 0.24, which is a typical value when the flora consists of diatoms

(Mclntire and Phinney, 1965). The ratio of ash-free dry weight to dry weight for each transect also indicated a diatom flora, except for the samples in

1980 taken on June 1 and 29 (Fig. 51). By inspection under a microscope it was noted that during this time of the year Smithora naiadum (Anderson)

Hollenb. and Ectocarpus sp., or other non-diatom algae were prominent components of the epiphyte community. The addition of these taxa increased the organic matter content of the epiphytic assemblage, as the percentage ash- free dry weight for most algae other than diatoms is usually greater than 60%

(e.g., see Davis, 1982). The percentage ash-free dry weight for samples obtained in February and April, 1981 also was high.During this period, the combination of a small biomass of epiphytes and unusually brittle Zostera leaves created a problem of contamination of the epiphytic samples with particles of Zostera leaves. EPIPHYTE % AFDW 1L700 80 uTT RANSECT 3(+1.4m)2(+l.2m)1 (-'-I.lm) zLUI-w60(9 0LU 402030 APR1 MAY2 JUN1 JUN29 JUL26 AUG24 SEP24 Figure 51. Percthetran meansectsentage of during ash-free3 subsamples. the dryperiod weight from associated April through with DATE Septemberthe biomass 1980. of epiphytes at the three Each point represents -2 37-

The taxonomic structure of the diatom assemblages epiphytic on Zostera marina L. in Netarts Bay is described byWhiting (1983) and on pages 137 to 170 of this report. From November through July, the flora was dominated by species of Cocconeis, Synedra, Navicula, Nitzschia, Gomphonema and Rhoico- sphenia. From August through October, Cocconeis, Gomphonema and Rhoicosphenia virtually disappeared from the samples, and different species of Navicula and

Nitzschia became the dominant taxa. The Shannon-Weaver diversity index revealed that the epiphytic flora was generally low in species richness and high in dominance.

B ioma s s

The interpretation of changes in epiphyte loads was related to the units that were used to express biomass. Epiphyte biomass expressed as g dry weight m2 also included inorganic material,e.g., debris and dead diatom frustules.

Therefore, dry weight expressed the actual load of material that was distributed over the leaves. Biomass expressed as ash-free dry weight m2 was considered an indicator of the organic matter present.

Epiphytes were prominent on the Zostera leaves from April through

October. During the winters of 1979 and 1980, field notes indicated that the

Zostera leaves were essentially free from epiphytes.This conclusion was supported by biomass measurements taken on February 12, 1980, when epiphyte biomass for transects 1 and 3 were 0.4 and 0.7 g dry weight m2, respectively.

During the sampling period from April through September 1980, epiphyte biomass expressed as dry weight was the greatest at transect 1. There was little net increase or decrease in epiphyte hioniass at the transects from May through

September 1980 (Fig. 52, Table 59). 100 EPIPHYTE BIOMASS TRANSECT 1 (-t-1.lm 60 TRANSECT 23 (+(+1.4m 1.2m >- 4O a0:: 20 0 APRI JUN1 JUL26 SEP24 FEB12 MAY3 Figure 52. Epiphytefrom April biomass 1980 expressedthrough May in 1981.,g dry Eachweight point m represents the mean of 5 to 7 observations. MAY2 1980JUN29 AUG24 DATE -2 for all threeAPR6 transects over the period 1981 MAY30 8.6 34.3 (6.5) (2.1) MAY 30 Standard 1.8 (9.2) (0.9) 65.0 MAY 3 1981 0.7 (3.0) (0.3) 16.0 APR 6 0.4 0.7 (0.1) (0.2) FEB 12 6.7 (3.7) (2.9) 87.0 24.1 (12.6) SEP 24 8.8 5.1 (2.7) (1.4) 44.1 (11.4) AUG 24 Values are the means of 5 to 7 observations. -239- (6.1) (3.9) 80.7 24.7 11.0 (22.3) JUL 26 1980 5.5 5.1 (2.4) (3.3) 81.4 (25.2) JUN 29 (5.7) (2.3) (4.6) 23.4 10.1 10.0 JUN 1 8.0 72.1 29.3 (9.0) (3.6) (11.0) MAY 2 8.4 1.5 (7.1) (3.3) (0.5) 66.1 error of the mean is in parentheses. APR 1

N-I C'1 Dashes indicate no data available. TRANSECT Table 59. Mean epiphyte blomass (g dry wt. m2). -2 40-

The principal difference between the pattern in epiphyte biomass

expressed as g dry weight m2 and that expressed a g ash-free dry weight m2 was an obvious increase in organic matter relative to dry weight biomass at

transect 1 in June and July 1980 (Fig. 53, Table 60). This was the result of

the increase in percentage ash-free dry weight that occurred when the flora

changed from exclusively diatoms to an assemblage that included Smithora and

Ectocarpus. Therefore, biomass as ash-free dry weight reflected shifts in the

composition of the flora that were not evident from biomass data expressed as dry weight.

Epiphyte biomass was significantly correlated (r > 0.6, P < 0.01) with

Zostera aboveground, belowground and total biomass, and with the average area of a vegetative leaf and shoot density (Table 61). However, epiphyte biomass was not correlated with the number of leaves per shoot (P > 0.01). Changes in number of leaves per shoot occurred independent of changes in Zostera biomass, and of changes in leaf area and shoot density which were closely related to changes in the growth rate of the host plant. Therefore, the data Indicated that the pattern in the epiphyte biomass was closely related to the accumulation of Zostera biomass.

Epiphyte load was examined by expressing epiphyte biomass as g dry weight per g dry weight of Zostera leaf. Epiphyte loads were highest in April and

May 1980; a maximum value of 2.3 g epiphytes/g leaf was measured at transect

2 in May (Fig. 54). From June 1 through September 1980, the loads averaged

0.4 g epiphytes per g leaf with a standard error of 0.06 g epiphytes per g leaf. Therefore, during the period of time when Zostera biomass was highest, the epiphyte loads were the lowest, because of an increase in Zostera biomass during a time when there was no net increase in epiphyte biomass. 30 EPIPH Y TE - AFDW TRANS ECT2ECu (+1.2m) (tl.lm) F- E 2025 TRANS ECT3(+1.4m) a:w>-. (I)LLrw 0 APRI MAY 19802 JUN1 JUN I29 I JUL26 SEP24 I AUG 24 I FEB12 i-f 1981APR6 MAY3 MAY 30 I Figure 53. Epiphytetoduring 7 observ thebi ations.periodomass expfr omressed April as 1980 g ash-free to May 30,dry 1981. DATE Eachm poi-2 at t nthree represe trans ntsects the in mean Netarts of 5 Bay -242- Table 60. observations.Mean epiphyte biomass expressedStandard error of 1980 in g ash-freethe dry mean is in parentheses.* wt. m2. Values are the means 1981 of 5 to 7 TRANSECT APR 1 13.8 MAY 2 13.6 JUN 1 9.6 JUN 29 27.5 JUL 26 21.5 AUG 24 12.4 SEP 24 17.3 FEB 12 0.2 APR 6 5.0 MAY 3 15.5 MAY 30 8.0 (0.6)(1.5) 1.5 (1.2)(2.0) 4.0 (0.9)(2.4) 4.0 (1.2)(8.5) 2.7 (1.4)(5.9) 5.7 (3.2)(0.7) 2.3 (1.0)(2.5) 6.6 (0.0) - (0.9) - (2.2) - (1.5) Dashes indicate no data available. (0.1) 0.3 (0.5) 1.2 (1.1) 2.4 (1.5) 2.3 (0.9) 2.6 (0.4) 1.3 (0.7) 1.6 (0.1) 0.4 (0.1) 0.3 (0.3) 0.6 (0.6) 2.3 -243-

Table 61. Correlation between epiphyte biomass and total shoot density (TSHD), average number of leaves per shoot (LVES), average area of a vegetative leaf (AREA), aboveground biomass (ABGB), belowground biomass (BLGB), and total Zostera biomass (TZOS). Epiphyte biomass is expressed as dry weight (EPDW) and ash-free dry weight (EPAF). Values greater than 0.2 are significantly different than zero where P < 0.01.

VARIABLE EPDW EPAF

TSHD 0.39 0.38

LVES 0.01 -0.01

AREA 0.57 0.63

ABGB 0.67 0.74

BLGB 0.70 0.72

TZOS 0.71 0.76 2.5 A EPIPHYTE LOAD Jhi 2.0 TR ANSECTANSECT 1 (+1.lm)2(+1.2m) C')ILhi 1.5 3(+1.4m) IH>-Lii hiIL [$1 APRI MAY2 JUN29 AUG24 I JUN1 I JUL26 SEP24 FEB12 MAY3 I I I APR6 MAY30 1981 I I Figure 54. Epiphyterepresentsalong three load the transectsexpressed mean of foras5 to gthe dry7 observations.period weight from of epiphytesApril 1980 per through g dry Mayweight 1981. of 1980 DATE Each poinZos tera tleaf -245-

Epiphyte biomass was influenced by the regular loss of that portion of the standing crop which has colonized the oldest Zostera leaf on a shoot. At certain times during the study, the loss of a particular group of leaves had more impact on the epiphyte biomass than the loss of leaves at other times.

The low epiphyte biomasses at transects 1 and 2 on June 1 and August 24, 1980, and at transect 1 on May 30, 1981, were probably related to the loss of

Zostera leaves that carried a larger proportion of the epiphyte biomass than was usually associated with the leaves that were sloughed. The highest loads of epiphytes in the study were recorded for the sample taken on May 2, 1980

(Fig. 54). One can predict the time when the leaves present in the Zostera bed at the time of the samples were sloughed by considering the mean lifetime of a leaf (34 days), the mean PT (9 days), and the mean number of leaves per shoot (4) during the time period under consideration (Fig. 55A). A leaf initiated by a shoot with four leaves on April 28 would be sloughed 34 days later on May 31. The oldest leaf on such a shoot would have initiated growth three Pr earlier on April 2, and would have been sloughed 34 days later on May

6. Therefore, most of the leaves present in the Zostera bed on May 2 were probably sloughed together with their heavy epiphyte load before June 1,

thereby accounting for the decrease in epiphyte biomass measured at that time. The loss of epiphyte biomass that occurred between July 26 and August

24, 1980 was explained using the same reasoning. The leaves that would have been sloughed during that period were in positions 1 and 2 on the shoot during the time of maximum shoot production (Fig. 55B). Leaves in positions 1 and 2 on a shoot together accounted for 75 to 92% of the shoot's growth from July to

October (Table 44). Therefore, the leaves that were sloughed between July 26 A. MEAN P1= 9OAYS MEAN a LVESXLIFE =34 4 VA b YS d SAMPLEDBIOMASSAPR1 I II I LOADMAXIMUM MEASURED EPIPHYTE j MAYI 4l DECREASEBIOMASS MEASURED EPIPHYTE JUN 10 8. MEAN P1 MEANMEAN LVES LIFE 3 /6 DAYS 46 DAYS N) JUN 201 30 1b JULIO 20 30 ('AUG9 19 (-I b 29 LEAFHIGHESTRATE GROWTH OF MAXIMUMBIOM ASS MEASUREDZOSTERA KEY BIOMASSDECREASE MEASURED IN EPIPHYTELEAF IDENTIFIER Figure 55. Time line of leaf initiation and sloughing explaining the sharp decreases in epiphyte biomass 2 = LEAF /DEN/FIERLEAF SLOUGHED IN/nA rED plastochroneofloadisobserved somedue of totheof on the the study.June interval priorlargest 1 and sloughing (P1),leavesAugust lifetime produced24,of the1980. ofmajority during a leaf the of(LIFE), thestudy. leaves and number that carriedof leaves the per highest shoot epiphyte(LVES). B. The decrease in epiphyte biomass on August 24, 1980 is due to the loss A. The decrease in epiphyte biomass on June 1, 1980 Factors considered included the -247-

and August 24, 1980 were among the largest leaves produced that year. Since they were also the oldest leaves on the shoot, they probably carried a large portion of the epiphyte biomass. In addition, notes made while sorting the sample taken in August indicated that many amphipods were present, and that the Zostera leaves showed signs of herbivory. Perhaps the process of grazing also accounted for at least some of the decrease in epiphyte biomass in August

1980.

Production

Measurements of epiphyte primary production using the method were made monthly from June to September, when the summer lower low tides were early in the day. The results from the 14C measurements were used to establish a ratio between the rates of 12C assimilation for the Zostera and the epiphytic assemblage rather than as a direct estimate of epiphyte production. It was assumed that Zostera and epiphyte respiratory losses per unit biomass were the same. The rates of accumulation for Zostera and epiphytes and their ratios are presented in Table 62.The mean ratio of the rates of 12C assimilated expressed as mg 12C assimilated (g dry weightY' h for the epiphytic assemblage and Zostera was 0.60 with a standard error of

0.08.

The ratio was used to estimate the net primary production of the epiphytes from the net primary production of Zostera obtained with the shoot marking period. The net primary production for the epiphytic assemblage was1 calculated for each time interval during which the net primary production of

Zostera was measured. First, the specific growth rate for Zostera was -21i8- Table 62. Netl2three primary transects production on .July of 2 (AFDW).assimilated (g dry we Values 11 ste d 6are lght)and the AugZostera h andust the 24, epiphyte 1980,mennso[ and assemblage 2May or 31,3 replicate 198I.measured(I)RY saniples.*usingWT) and the as mg '2C assimilated (g ash-tree dry wclght Production is expressed as mg 11C method along Epiphyte DRY WF Production AFDW Zostera DRY WT Production AFI)W ZosteraEpi phyte DRY wr Product IonProductIon: AFDW July 26, 1980 Transect 0.6 2.2 1.3 1.8 0.5 1.2 23 18.5 0.4 80.4 1.7 40.8 2.0 57.1 2.8 0.40.2 0.61./s August 24, 19 Transect 80 71.1 253.9 82.3 111.0 0.9 32 22.919.0 91.673.1 24.834.0 47.636.0 0.70.8 Play 31, 1981 Transect 161.9 702.6 272.0 361.2 0.6 1.9 23 85.6 329.2 111.0 146.9 0.8 2.2 -2 49-

calculated by dividing the net primary productionof the aboveground portions of Zostera by the mean aboveground biomassfor the period of time under consideration. Then the specific growth rate of the epiphytwas calculated by multiplying the specific growth rate of Zosteraby the mean ratio of the rates Of'2C assimilation for Zostera and the epiphyce assemblage (0.60).

Finally, net primary production of theepiphytes was calculated by multiplying the mean biomass of the epiphytes by thespecific growth rate of the epiphytes. These values are listed in Table 63. The net primary production of the epiphyte assemblage at transect 1ranged from 5.1 to 13.7 g ash-free dry weight m2 month (18.7 to 40.5 g dry weightm2 m1) during the period from June 1980 through June 1981, while thatfor transect 3 ranged from 0.4 to

2 2.7 g ash-free dry weight 2month (1.1 to 8.1 g dry weight month).

Components of Variance

To gain added insight into therelationships among the morphotnecric and biornass variables, principal components analysis(PCA) was performed on a pooled data set for the three transects. Variables in the data matrix included shoot density (shootstn2), number of leaves per vegetative shoot, mean area of a vegetative1ef (cm2), aboveground andbelowground Zostera biomass (g dry weightm2), epiphyte biomass (g ash-free dry weighttn2), and epiphyte load (g dry weight perm2 leaf surface). Only the first three components, with eigenvalues greaterthan or equal to one, were interpretable.

The first three principal componentsaccounted for 82.9% of the variation in the seven variables (Table64). Factor loadings indicated thatprincipal component 1 (PCi) primarily expressesshoot density, average area of a vege- -250-

Table 63. Net primary production of the epiphyte assemblage measured along the three transects from June 1, 1980 to May 30, 1981. Production is expressed as g dry weight m2 (DRY WT) and g ash-free dry weight in2 (AFDW) for each period of time.*

TRANSECT

DRY WT AFDW DRY WT AFDW DRY WT AFDW

R1Ai]

JUN1-JUN29 36.7 13.7 7.8 3.5 6.8 2.3

JUN29-JUL26 40.5 12.2 16.6 6.0 8.1 2.7

JUL26-AUG24 18.7 5.1 10.0 2.5 5.6 1.4

AUG24-SEP24 29.5 7.1 13.6 3.6 7.0 1.7

1981

APR 6-MAY 3 28.4 7.8 1.1 0.4

MAY 3-MAY 30 39.7 9.4 2.6 0.8

*Dashes indicate no data available -25 1- Table 64. ThecorrespondingFactor (PCI,analyses loadings PC2, was PC3)eigenvlue correspondingbased of theon and 175morphometric accumulated and percentage biomass observations.to the first three variables,principal of the variance.thecomponents Shoot density (shoots m2) Variables 0.644 PCi 0.557 PC2 -0.162 PC3 AverageNumber ofarea leaves of a pervegetative vegetative leaf shoot (cm2) -0.002 0.704 0.804 0.120 Aboveground biomass (g dry wt. in2) 0.9220.953 -0.570 0.081 -0.132-0.017 0.008 EpiphyteBelowground biomassload biomass (g (gdry ash-free (gwt. dry m2 wt. leafdry m2) wt.area) m2) 0.0090.856 0.0270.1160.070 0.9760.285 AccumulatedEigen value % of variance 48.6 3.403 67.3 1.305 82.9 1.093 -2 52-

tative leaf, aboveground and belowground Zostera biomass and epiphyte biomass.

These variables all were expressions of the biomass associated with the

Zostera Primary Production subsystem (Fig. 3). Total biomass for the Zostera

Primary Production subsystem expressed in g ash-free dry weight m2 (ZPP) was regressed against PCi, generating an R2 value of 0.97; the regression model was ZPP = 125.87 + 53.03 PCi. Factor loadings also indicated that the second principal component (PC2) was an expression of the inverse relationship between the average area of a vegetative leaf on the one hand and the shoot density and number of leaves per vegetative shoot on the other hand, i.e., the morphometrics of a vegetative shoot. The third principal component (PC3) was highly correlated with epiphyte load (r = 0.98). When correlation coefficients were calculated for this group of variables, epiphyte load was not correlated with any of the other variables. In summary, this set of data generated three independent (orthogonal) components of variance: autotrophic biomass, leaf-shoot density (morphometrics), and epiphyte load.

V. Bioenergetics of the Zostera Primary Production Subsystem

In the Background section of this report, a conceptual model of the Mud- flat Processes subsystem was presented as a hierarchy of coupled subsystems

(Fig. 3). This model was used to organize the examination of the bioenergetics of the seagrass ecosystem in Netarts Bay. The following analysis considers several levels of resolution. At the finest level of resolution, the Macrophyce Primary Production subsystem is decomposed into aboveground and belowground Zostera. The gains and losses of biomass for each of these components are partitioned and are presented as an energy budget. The Macro- -253-

phyte Primary Production subsystem then is integrated with the Epiphyte

Primary Production subsystem to form the Zostera Primary Production

subsystem. The gain of biomass from the primary production of this subsystem

is expressed as annual net production.

Information accumulated on the growth of Zostera marina L. was

synthesized with respect to the bioenergetics of the Macrophyte Primary

Production subsystem. The changes in its state variable, the macrophyte

biomass, were examined in terms of the inputs and outputs that affected its

value. These inputs and outputs were related to the dynamics of the

Macrophyte Primary Production subsystem, as primary production of Zostera was

related to the physiological activities of the plant.The inputs and outputs

that affect Zostera biomass also represented the couplings between the

Macrophyte Primary Production subsystem and the rest of the subsystems in the

conceptual model. Consequently, the perspective gained by such an approach

contributed to the understanding of the role that Zostera marina played in the estuary.

Changes in aboveground biomass were caused by inputs from shoot net

primary production and outputs as lost shoots, sloughed leaves and grazing.

Biomass accumulation due to shoot net primary production and losses due to

sloughed leaves were measured directly and were partitioned separately in the energy budget. Shoot net primary production for each sampling period was calculated according to the following expression:

Shoot net Rate of net Mean shoot primary = primary produc- x density production tion per shoot (shoots ni 2) (g dry wt (g dry wt per shoot) -254-

Monthly shoot net primary production, estimated as g dry weight m2, was calculated by summing the values for the constituent sampling periods. Losses due to sloughing of leaves expressed as g dry weight m2 for the time period under consideration were calculated according to the following expression:

Mean leaf Number of Mean number tvIeafl dry loss per leaves lost of shoots weight per in2 per shoot per leaf lost

The mean dry weight per leaf of the leaves lost was estimated as the mean weight of the largest leaf on a shoot at the beginning of the sampling period.

The energy budgets for abovegrourid Zostera are presented in Tables 65, 66 and 67. The column labeled "difference" represents biomass losses not accounted for directly in this study, and includes biomass lost as entire shoots and to grazing.

At all transcts during each time interval under consideration, biomass losses were at least 50% of the shoot net primary production. From July through October 1980, the biomass losses were equal to or higher than the shoot net primary production. This indicated the great capacity of the Zostera to turnover its aboveground biomass. The lifetime of a leaf ranged from 34 to

55 days during the course of this study. Therefore, every 34 to 55 days each shoot had an entire set of new leaves. With this in mind, it was not sur- prising that 35 to 100% of the biomass loss was due to sloughed leaves. The biomass loss from leaf export occurred in response to the changes in leaf size. En April and May the small leaves of the winter growth form were shed, and leaves exported during this period represented a smaller proportion of biotnass loss than when the larger summer leaves were lost from June to October. Table 65. Energy1 between April 1, 1980 and July 1, 1981. budget accounting for the gains and losses of aboveground biomass along transect -255- Values are expressed as g dry weight m2 for the period of time under consideration.* NET PRIMARY BIOMASS LOSS AS APR DATE 1, 1980 BIOMASS 98.7 EBIOMASS PRODUCTION LOSS LEAVES DIFFERENCE MAYJUN 1,2, 19801980 162.1114.8 +47.3+16.1 ** JULJUN 26,29, 1980 197.9256.2 +58.3+35.7 234.5220.8 176.2185.1 170.3 98.0 87.1 5.3 AUGSEP 24, 1980 119.4173.9 -54.5-82.3 122.5111.0 177.0193.3 106.3 98.0 70.795.3 FEBOCT 12,23, 19811980 57.670.2 ** -12.7-49.2 105.8 155.0 80.6 74.4 MAYAPR 3,6, 1981 181.0166.5 +108.9 +14.5 222.7 208.2 89.6 118.6 MAYJUL 1,30, 1981 1981 226.0202.4 ** +23.6+21.3 344.7272.2 321.1250.9 127.2 88.9 193.9162.0 1%* * Dashes indicateExtrapolated. no data available. Table 66. Energythe2 between period budget Aprilof accounting time 1 andunder October for consideration. the 23, gains 1980. and losses of aboveground biomass along * Values are expressed as g dry weight m2 for transect DATE BIOMASS ABIOMASS PRODUCTIONNET PRIMARY BIOMASS LOSS LOSS AS LEAVES DIFFERENCE MAYAPR 2,1, 1980 26.611.3 +15.3 JUN 29,1, 1980 1980 60.837.8 +23.0+11.2 85.3 62.3 17.9 44.4 AUGJUL 24,26, 1980 53.571.0 -17.5+10.2 123.1 63.7 112.9 81.2 43.330.8 37.982.1 OCTSEP 23,24, 1980 60.070.4 -10.4+16.9 51.278.8 61.661.9 53.149.8 12.1 8.5 ** Dashes indicateExtrapolated. no data available. -25 7- Table 67. Energy3 between April 1, budget accounting for the 1980 and July 1,gains 1981. and losses Values areof abovegroundexpressed as g drybiomass weightalong transect m2 DATE for the BIOMASS period of time under consideration.*LBI0MASS BIOMASS LOSS AS DIFFERENCE APR 1, 1980 7.3 - - - - MAYJUN 1,2, 1980 30.3 8.1 +19.2+22.2+0.8 66.5 - ** 47.3 - 25.2- ** 22.1 - JULJUN 26,29, 1980 57.149.5 + -22.17.6 49.888.2 80.671.9 41.527.4 30.453.2 AUGSEP 24, 1980 35.022.0 -13.0 0 41.855.9 41.868.9 14.917.4 27.251.5 FEBOCT 12,23, 19811980 22.322.0 ** +0.3 -- - - - MAYAPR 3,6, 1981 32.421.6 +10.8-0.7 52.5 41.7 10.6 31.1 MAYJUL 1,30, 1981 1981 60.084.4 ** +52.0-24.4 104.7 79.3 103.7 52.7 59.419.7 44.333.0 Dashes Extrapolated. indicate no data available. -2 58-

An energy budget also was constructed for belowground Zostera biomass.

The net primary production of belowground Zostera was measured between April and May 16, 1981, according to the method of Jacobs (1979) and Kenworthy

(personal communication). The net production of the belowground Zostera during this period of time was 219.8 g dry weight m2 at transect 1 and 45.3 dry weight m2 at transect 3. Since a direct measurement of belowground net primary production was made for only a short time period during this study, an alternative method of estimating belowground net primary production was adopted. For the development of an energy budget, shoot net primary production was used to estimate belowground net primary production. Since both variables were measured concurrently from April 6 to May 16, 1981, these data were used to establish the relationship between the net production of aboveground and belowground Zostera. The specific growth rate for the net production of aboveground and belowground Zostera was calculated by dividing the net production (g dry weight m2) by the mean biomass (g dry weight m2) for the period of time under consideration. The ratio between specific growth rate for aboveground Zostera and the specific growth rate for belowground Zostera was calculated for that period of time. For transect 1 this ratio was 0.90, while for transect 3 it was 0.46. It was assumed that the ratio between the specific growth rates of ahoveground and belowground Zostera remained constant throughout the study. The specific growth rate for shoots was calculated for each sample period, and the corresponding specific growth rate for belowground

Zostera was estimated using the appropriate ratio. The estimate of the specific growth rate for belowground Zostera then was multiplied by the mean belowground biomass for the corresponding period of time to obtain the net -259-

production of belowground Zostera. The energy budgets based on these esti-

mates are presented in Tables 68 and 69.

At transect 1, throughout the study period, the biomass loss of below-

ground Zostera was greater than or nearly equal to the net primary production.

At transect 3, the biomass loss was generally between 25 and over 100% of the

net primary production for belowground Zostera. Therefore, as was the case

for aboveground Zostera, belowground Zostera biomass was regularly being

turned over.

The large bioniass loss from the Macrophyte Primary Production subsystem

represented the coupling of this subsystem with the Detritial Decomposition

subsystem, as well as the Consumer Processes subsystem (Fig. 3). Losses of belowground biomass were primarily retained within the Zostera bed, while the

losses of aboveground biomass not retained within the Zostera bed were trans-

ported to other regions within the estuary, to the salt marsh and to the ocean. It was through this export of primarily aboveground biomass that

Zostera had an impact as an input of organic matter on regions distant from its source.

At a higher level of resolution, processes associated with the Zostera

Primary Production subsystem were described. The growing season for Zostera in Netarts Bay was defined in April through October, which is typical of temperate seagrass beds (Phillips, 1972; Sand-Jensen, 1975; Stout, 1976).

Since production data for months representing an entire growing season were available only for transects 1 and 3, data from those regions were used as a basis for the following estimates. -260-

Table 68.Energy budget accounting for the gains and losses of belowground biomass alongtransect 1 betweenApril 1, 1980and July1, 1981. Values are expressedas g dry weight nr2 forthe periodof time under consideration.

BIOMASS DATE BIOMASS tBI0MASS ODU

APR1,1980 160.8 - 6.7 MAY2,1980 154.1 -2.4 - JUN1,1980 151.7 +12.7 174.5 161.8 JUN29,1980 164.4 +42.5 167.1 124.8 JUL26,1980 206.9 -59.7 79.7 139.4 AUG24,1980 147.2 -14.7 105.1 119.8 SEP24,1980 132.5 -15.0 125.0 140.0 OCT23,1980 117.5** -55.5 - - FEB12,1981 62.0 +64.1 - APR6,1981 126.1 1.7 150.3 152.0 MAY3,1981 124.4 +27.2 179.4 152.2 MAY30,1981 151.6 -21.6 197.1 218.7 JUL1,1981 130.0

* Dashes indicates no data available. ** Extrapolated. -26 1-

Table 69.Energy budget accounting for the gains and losses of belowground biomass along transect 3 between April 1, 1980 and July 1, 1981. Values are expressed as g dry weight nr2 for the period of time under consideration.

NET PRIMARY BIOMASS DATE BIOMASS BIOMASS PRODUCTION LOSS

APR1,1980 71.7 -34.4 MAY2,1980 37.3 - 2.2 - JUN1,1980 35.1 +19.0 31.2 12.2 JUN29,1980 54.1 +31.5 55.9 24.4 JUL26,1980 85.6 -31.9 34.8 66.7 AUG24,1980 53.7 -18.2 40.1 58.3 SEP24,1980 35.5 ** - 5.5 29.5 35.0 OCT23,1980 30.0 -5.2 - FEB12,1981 24.8 +16.2 - APR6,1981 41.0 - 9.6 25.3 34.9 MAY3,1981 31.4 +31.8 18.9 -12.9 MAY30,1981 63.2 * -13.2 39.6 52.8 JUL1,1981 50.0

*Dashes indicate no data available. Extrapolated. -262-

The total net primary production of aboveground and belowground Zostera and epiphytes for the entire growing season was obtained from the production values measured for the period from April through October during the course of the study. Production expressed as g dry weight m2 day was calculated by dividing by 214 days, or the length of the growing season. The dimensions of the Zostera bed were measured and the bed was stratified into regions representative of transect 1and transect 3. The total area of the study site was

17,562 m2. Of this total, an area of 10,062 m2 was designated as equivalent to transect 1 and an area of 7500 m2 was designated as equivalent to transect

3. The net primary production for each region was then estimated by multiplying its area by the production m2.

The turnover time was calculated from the following expression:

Mean biomass Net production Turnover time during the period during the (days) from t1 - t0 period from

(gm2) ti tO (g m2 day)

where t0 is the time at the beginning of the measurement period and t1 is the time at the end of the measurement period.

Measurements of Zostera biomass expressed in g dry weight were converted to g C by multiplying by 0.38, the proportion of the dry weight that is carbon

(Westlake, 1963). Measurements of epiphyte biomass expressed as g dry weight were converted to g C using the expression proposed by Davis (1982):

g dry proportion ash- g carbon x 0.50, weight free dry weight

where 0.50 is the proportion of the ash-free dry weight that is carbon. -263-

The region represented by transect 1 accounted for 80% of the primary production of the Zostera Primary Production subsystem (Table 70).Although the mean biomasses for aboveground and belowground Zostera were nearly equivalent in this region, aboveground biomass accounted for 58% of the net primary production of the macrophyte, and was turned over about three more times during the season than the belowground material. In contrast, the region similar to transect 3 maintained a mean belowground biomass that was 30% higher than the mean aboveground biomass (Table 71). The aboveground biomass turned over almost three times as fast as the belowground biomass, and its rate of production was almost double that of the belowground biotnass.

Therefore, the plants in two regions of the bed had different growth strategies.

The net primary production by the Epiphyte Primary Production subsystem only accounted for 8% of the net primary production associated with the

Zostera Primary Production subsystem during the growing season (Table 72).

The mean turnover time for epiphyte assemblages was more than twice that of the aboveground Zostera.

Stout (1976) reported that 161 ha in Netarts Bay were occupied by shallow eelgrass beds. Her description of a shallow eelgrass bed was similar to that for transect 3. She also reported that 176 ha were occupied by deep eelgrass beds, i.e., equivalent to transect 1. To calculate the production of the entire bay associated with each region, the net primary production values expressed in g dry-weight m2 for each region was multiplied by 10,000 times the area of the region. Therefore, an estimate of total annual net primary production for the eelgrass beds in Netarts Bay was 6.0 x io6 kg. -264- Table 70. Annualprimarytransect net production 1.primary production expressedexpressed for asas gtheg drydry region weightweight of m mlthe da A growing season from April to October ; N PP/AREAy1; NPPnr2 =was aZostera assume be nnuald.d represented net =primary annual by net NPPd1 = net meanexpressedgrowingturnedproduction biomass over;season. as forexpressed gramsd-OVER the carbon. entire= asthe g numberareadry weight(10,062.5 of days form2) theexpres bi The values in parentheses are the nr2; X-OVER correspondiomass=sed the as numbe tokg tud ngmoverrry measurementsof weight; times during biomassXBIO the = Abovegroun d NPPd1 6.6 NPPm21413.4 NPP/AREA 14,000 153.9 XBIO X-OVER 9.2 d-OVER 23.3 Belowgroun ZosZostera tera d (1.8)(2.5) 4.7 1003.1(381.2)(537.1) 10,000(5300) (3800) 148.8(56.5)(58.5) 6.7 31.7 Macrophyte SubsysProductiPrimary te mon (4.3)11.3 2416.5(918.3) 24,000(9100) (115.0)302.7 8.0 26.8 Epiphy te SubsysteProductiPrimary mon (0.1) 0.9 (29.7)247.4 2,500(300) 57.7(6.9) 4.3 49.9 Zos tera ProductiPrimarySubsyste mon 12.12(4.4) (948.0)2713.8 26,500(9400) (121.9)360.4 7.5 28.4 -265- Table 71. primaryAnnualtransect netproduction 3.primary expressedproductionexpressed as asfor gg drythedry weightregionweight nr2;ofm2 theda NPP/AREA = annual net primary A growing season from April to October v-i; NPPmwasZostera assum 1) 2ed.ed represented by = annual net NPPd1 = net asseason.over;biomassproduction grams d-OVER expressedcarbon. for = the asnumberentire g dry of areaweight days (7500 for m2)the expressedbiomass to as turnover kg dry weight;during theXBIO growing = mean The values in parentheses are the corresponding measurements expressed in2; X-OVER = the number of times biomass turned d NPPd1 2.2 NPPm2 461.3 NPP/AREA 35,000 XBIO33.6 X-OVER 13.7 d-OVER 15.6 AbovegrounBelow g ro un Zostera d (0.4)(0.8) 1.1 (175.3) 244.1(92.8) 18,000(1300) (700) (18.6)(12.8) 49.0 5.0 42.9 Mac rophy te PrimaryProductiZos tera on 3.3 705.4 5,300 82.6 8.5 25.1 Epiphyte PrimarySubsyste m (1.3) (268.1) (2000) (31.4) Zos tera SubProducti sy ste mon (0.01) 0.1 36.9(4.4) (30) 280 (0.7) 5.8 6.3 33.9 PrimarySubsProducti y s t e mon (1.3) 3.4 (272.5)742.3 (2030) 5,600 (32.1) 88.4 8.5 25.2 -266- Table 72. weightNPP/AREAAprilAnnual m2to net Octob=day prinet ermary was produ assprimary1; NPPuf2 p roductionumed.ction = net for th primaryNPPd produ for the ent = entirenet ireprimaryction area expre pro(1Zostera be ssedductiond.7,562.5 as gexpressed dry weight as nc2;g dry A growing season from -2) expressed as pondingnumberkgto dry turnover weight; ofmeasur time du ementsrings biomass the exp gXBIO=me ressedrowingturnedan biomass expressed asover;season. grams d-OVER carb The on.values= the numbein pasgdryw arenthesesreight of days m; X-OVER forare thethe = biomass corres-the Aboveground NPPd1 8.8 NPPm21874.7 NPP/AREA17,500 187.5 XBIO X-OVER 10.0 d-OVER 21.4 Belowground ZosZostera tera (2.2)(3.3) 5.8 1247.2(474.0)(712.4) (4500)11,800(6600) (75.1)197.8(71.3) 6.3 33.9 Macrophy te ProductioPrimarySubsystem n 14.6(5.6) (1186.4)3121.9 (11,100)29,300 (146.4)385.3 8.1 26.4 Epiphyte ProductioPrimarySubsys tern n (0.1) 1.0 284.3(34.1) 2,800(300) 63.5(7.6) 4.5 47.8 Zostera ProductioPrimarySub sys tern n 15.6(5.7) (1220.5)3406.2 (11,400)32,100 (154.0)448.8 7.6 28.2 -267-

VI. Discussion

The pattern of growth of Zostera marina L. in Netarts Bay, Oregon is typical of Zostera growing in temperate regions. However, the results of this study also support the conclusion of McMillan and Phillips (1979) that seagrass populations reflect the selective influence of local habitat conditions. In Netarts Bay, the initiation of growth in the spring is marked by shoot proliferation, change to the summer morphology and rapid growth of the leaves as Tutin (1942), Phillips (1972), Sand-Jensen (1975), Jacobs

(1979), and Harrison (1982) also found. Flowering begins during the spring and early summer. Maximum vegetative growth occurs after the maximum density of reproductive shoots is reached (Phillips, 1972). In the late summer and into fall, there is a gradual decline in shoot density and biomass. Despite agreement on this general pattern of events, reports differ as to the timing and magnitude.

Sexual reproduction does not play a major role in the maintenance of

Zostera beds. Reproductive shoots comprise on the average from 2.6% (this study) to 15% of the total shoots in an area (Phillips, 1972; Sand-Jensen,

1975; Aioi, 1980; Harrison, 1982). The timing of events associated with the sexual reproduction of Zostera in Netarts Bay is very similar to that in Puget

Sound (Phillips, 1972). Flowers appear in the bed from March to May and disappear between August and October. Although Phillips (1972) found seeds germinating throughout the year, the time of peak seed germination in Puget

Sound (April to July) coincides with the time that seedlings were found in

Netarts Bay. -26 8-

The leaf marking technique permits the analysis of the growth of individual leaves on a vegetative shoot. This study agrees with previous investigations that most of the growth of a leaf occurs when it is the youngest and next to the youngest leaf on the shoot (Sand-Jensen, 1975; Mukai, etal.,

1979; Jacobs, 1979). Differences in the time it takes for a leaf to reach its maximum growth reflect local differences in the P1 and number of leaves per shoot, i.e., lifetime of a leaf (Table 73). The life of a leaf on a vegetative shoot in Netarts Bay is most similar to that reported by Jacobs

(1979) for Roscoff, France. He reported a change in the number of leaves per shoot and the PT during the course of his study. As in this study, he noted an increase in the number of leaves per shoot in the spring, and a decrease during the summer and into the fall. Moreover, he found a negative correlation between the PT and isolation. This was not the case for Netarts Bay, where the PT was longer from June through August 1980, when photosynthetically active radiation was maximum (Fig. 56). Instead, changes in the lifespan of a leaf in Netarts Bay were related to the regularly occurring annual changes in the plant between the winter and summer shoot morphology.

Biotnass and density of Zostera in Netarts Bay, Oregon is comparable to that for Zostera in other temperate regions (Table 74). The highest densities and biomasses were measured along transect 1 and are similar to the highest densities and biomasses reported in the literature.

Reported values for production were converted to carbon following the recommendations of Westlake (1)63). Values from the literature for aboveground Zostera productionrange from 0.5 to 8 g Crn2 day'. The results of this study are within theseranges, with 0.6 to 4.2 g Cin2 day for aboveground Zostera productionand 1.0 to 6.6 g C m2 day for total Zostera production (Table 75). 6050 16 -JIOrF-2O 4 JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY MONTH Figure 56. Maximumasthree days. transectsphotosynthetically from June active1980 to radiation May 1981. (PAR) and mean plastochrone Interval (P1) of the PAR values were adapted from Davis (1982). PAR is expressed as E.m2.d; P1 is expressed -269-

Table 73. Plastochrone interval(P1), meannumber of leaves per vegetative shoot (LVES)and meanlifetime of a leaf (LIFE) from data for variouslocalities. P1 and LIFE are expressed in days.

LOCATION P1 LVES LIFE REFERENCE

Japan 8 5.5 44 Mukai, et al. (1979)

Denmark 14 3.9-4.5 56 Sand-Jensen (1975)

France 13-19 2.1-4.4 40-57 Jacobs (1979)

Oregon 9-20 2.7-4.0 34-56 This Study

Calculated from data Table 74. Reported biomass and density values measured for -270- Zostera marina L. LOCALITY numbersBiomass isof expressedshoots m2. in g dry weight m2; densit BIOMASS DENSITY y is expressed REFERENCEa S TheDenmark Netherlands 157-443272-960 1-116 * 1200-1800 0-3600 - NienhuisPetersen,Sand-Jensen, and 1913 DeBree, 1975 1980 CanadaFrance 200-470 5-15 580-700 50-160 Harrison,Jacobs, 19791982 NorthNew York Carolina 175-545247-2062 50_185* - BurkholderPenhale,Dillon, 1971 1977and Doheny, 1968 California 30-73012_421* 6-192* HardingWaddell,Keller, and1963 1964 Butler, 1979 AlaskaWashington 186-324116-231 62-1840 599-4576710-2101151-893 McRoy,Phillips, 1970a,1966 1972 l970b Oregon 120-463288-467 995-3845671-1056 ThisStout, Study 1976 Transect 1 Only aboveground Zostera considered. 47-14846-167 500-1761789-1414 TransectTransect 3 2 -27 1- Table 75. Net production values reported for the shoots of Zostera marina L. LOCALITY productionand for the iswhole expressed plant (abovegroundin g C nr2 day-. plus belowg PRODUCTION METHOD REFERENCEround material) Net TheDenmark Netherlands O.O_3.5*1.4_3.7* 2-7.3 MarkingBiomass NienhuisSand-Jensen,Petersen, and 1913 DeBree, 1975 1980 NorthFrance Carolina 0.6-1.20.5-1.70.6_3.3* Marking14C'4C Method Method Penhale,Dillon,Jacobs, 197919711977 AlaskaWashington 0.7-4.03.3-3.8 8.0 02Biomass Method McRoy,Phillips, l97Oa,1966 1972 l97Ob Oregon 0.5-1.40.7-1.71.4-4.1 MarkingMarking This Study Transect 231 Summation of aboveground and belowground production. 0.9_l.9*2.5_6.6* Transect 31 -27 2-

As in this study, relationships among changes in shoot density, total

Zostera biomass and aboveground production are reported in the literature.

One such relationship is that described for transects 2 and 3 where changes in shoot density, total Zostera biotnass and aboveground production followed a similar pattern (Phillips, 1972; Neinhuis and De Bree, 1980; Aioi, 1980;

Harrison, 1982). Another is the pattern described for transect 1 in this study where shoot density is highest in the spring and decreases during the summer as total Zostera biomass and aboveground production increases (Sand-

Jensen, 1975; Jacobs, 1979).

Sand-Jensen (1975) attributed the decrease in shoot density that he observed to the loss of reproductive shoots during the summer. However, he reported that shoot density decreased from 1800 shoots m2 to about 1200 shoots m2, and that reproductive shoots accounted for no more than 4% of the total density or about 72 shoots nc2. Therefore, loss of reproductive shoots cannot totally account for the change in density that he observed.Jacobs

(1979) did not attempt to explain the pattern.

These patterns are explained by the hypothesis presented in this study that there is a threshold leaf area per unit area of substrate. Results of this study indicated that this threshold value is between 7.5 and 11.0 m2 leaf m2 area of substrate. When the leaf area per unit area of substrate is below the threshold value, shoot density, total Zostera biomass and aboveground production are positively correlated.When the leaf area per unit area substrate is above the threshold value, shoot density is negatively correlated with total Zostera biomass and aboveground production. For example, the leaf areas reported by Phillips (1972) range from 2 to 8 m2 leaf -27 3-

area of substrate. Therefore, the leaf area at Puget Sound during the time of his study was below the threshold value suggested by this study, and as the hypothesis predicts, the shoot density and biomass values that he measured increased and decreased together. Considering the conclusion of

Dennison (1979) that Zostera adjusts to a decrease in light levels by decreasing of leaf area, this threshold leaf area must represent a critical reduction of light In the Zostera bed which results from self-shading.

Light is considered an important factor in the determination of the distribution and production of Zostera marina (Burkholder and Doheny, 1968;

Sand-Jensen, 1975; Jacobs, 1979; and Mukai etal., 1980). Sand-Jensen (1975) clearly illustrated that the pattern of leaf production was closely related to that of insolation, but not that of temperature. The correlation between insolation and leaf production is also supported by this study (Fig. 57).

This relationship explains the earlier and higher shoot production reported for 1981 as compared to 1980. Rates of leaf production for May 1981 were comparable to those for July 1980, and photosynthetically active radiation for

May 1981 was comparable to that for July 1980.

The curves for photosynthetically active radiation and mean shoot production had a similar shape except during August 1980, when there was a relatively sharp decrease in shoot production. This decrease was attributed to the bloom of Entermorpha prolifera in Netarts Bay. No other study has reported such an interaction between the seagrass community and the drift algae.

There are few determinations of the biomass and productivity of the epiphytes of seagrasses. The most complete study of the production of Zostera 200180 160 0zj U-.50a: I4O8 0 -40 8OI00ztE a:'>-cJa:° -J(9t-20 I0 6040b Lu 0Zc) a: 0 JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY MONTH 20o I Figure 57. Maximum photosynthetically active radiation (PAR) and mean shoot net primary month.productionas Em2d-. for Shoot the threenet primary transects production from June is1980 expressed to May 1981. as g dry weight m2 PAR values were adapted from Davis (1982). PAR is expressed -274-

and its epiphytes is that of Penhale (1977). Using the method she estab-

lished that the ratio between the mean rates of the production for epiphytes

and Zostera was 0.74, i.e., 0.65 mg C (g dry weight' h' for the epiphytes

divided by 0.88 mg C (g dry weighty' h1 for Zostera. The corresponding

value for this ratio determined in this study was 0.60.

Production values for Zostera and epiphytic assemblages determined

through the coupling of the '4C method and the leaf marking method are

comparable to those reported for by other investigators using only the

method (Table 76). The only production values for the epiphytic assemblages

are those reported by Penhale (1977) and this study. Mean primary production

of epiphytes in North Carolina is at the lower end of the range reported in

this study. The mean biomass of aboveground Zostera and epiphytes reported by

Penhale (1977) was 105 g dry weight m2, while the mean biomass of aboveground

Zostera and epiphytes found in this study was 251 g dry weight m2. In both

studies 25% of the biomass was epiphytes. Therefore, the production values

reported in this study are higher than those reported by Penhale (1977)

because of the greater biomass in Netarts Bay.

It has been postulated that the constant replacement of the leaf biomass on a shoot of seagrass is a control on the development of stands of epiphytes

(Sand-Jensen, 1977; ott, 1980). Until this work, no study of seagrasses has monitored the biomass of the epiphytic assemblage and related it to the

pattern of the loss of leaves from the macrophyte. Sharp decreases in epiphyte biomass measured on June 1 and August 24, 1980 were related to the sloughing of particular groups of leaves. When a transition in the growth pattern of Zostera occurred, there was a greater loss of leaf biomass than -27 5-

Table 76.Comparisons of estimates of the primary production of Zostera and associated epiphytic assemblages obtained by the 14C method. Production is expressed as g C nr2 day

COMPONENT LOCALITY PRODUCTION REFERENCE

Zostera North Carolina 1.7 Dillon, 1971 0.9 Penhale, 1977

Alaska 11.0 McRoy, 1974

Oregon 1.0-6.6 This Study

Epiphytes North Carolina 0.2 Penhale, 1977

Oregon 0.3-4.9 This Study -27 6-

usual, and a decrease in epiphyte biomass was noted. Since new leaves are continually being produced on a shoot, the epiphyte biomass that is lost is also continuously replaced as the new tissue is colonized.

Phillips (1972) demonstrated that individual shoots of Zostera grown in culture without epiphytes had a regular pattern of initiation and loss of leaves. Therefore, he hypothesized that the timing of the loss of leaves is inherent to the Zostera plant and is not tied to any critical load of epiphytes. Results of the Principal Components Analysis done in conjunction with this study supported this hypothesis. Epiphyte load was highly correlated with one of the principal components, while the other variables considered were correlated with the two other principal components. These principal components represented orthogonal axes. Therefore, epiphyte load may be independent of these variables or may be associated by complex non-linear relationships.

There are few studies in the literature that attempted to partition the changes in the biomass of Zostera as was done in this study. This is primarily because the techniques for measuring belowground production were not developed until recently. Sand-Jensen (1975) and Jacobs (1979) reported annual production of the Zostera beds in their localities partitioned into aboveground, belowground and total Zostera (Table 77). The values for Netarts

Bay are higher, although the Zostera biomasses for all three areas are similar.

This study was designed to augment the ongoing work of the EPA relative to physical processes in the estuary. The coupling of these studies was designed to test the hypothesis that nutrient dynamics in Netarts Bay during the growing season (April through October) are closely related to the -277- TaiDle 77. Meanand belowgroundvalues of annual neZostera. t primary production values partioned into aboveground (g dry wt. PRODUCTION (g dni2) DURATION OFEXPERIMENT REFERENCE BelowgroundAboveground Total 16081116 492 572183389 12 months Jacobs, 1979 AbovegroundBelowground Total 1097 241856 415328 87 Apr.-Oct. Sand-Jensen, 1975 BelowgroundAboveground Total 3121.91247.21874.7 1186.4 474.0712.4 Apr.-Oct. This Study -27 8-

biological processes associated with the Zostera Primary Production subsystem.

The future application of the results of this study will involve the development of a holistic conceptualization of the estuary. Data related to the benthic algal flora reported in an earlier section can be integrated with data

from the Zostera decomposition work done by the EPA (unpublished data), and used to describe changes in biomass associated with the Algal Primary Produc- tion and the Detrital Decomposition subsystems. These two subsystems can be

then coupled with the dynamics of the Zostera Primary Production subsystem to describe the dynamics of the Primary Food Processes, a subsystem at a coarser

level of resolution.

In addition to describing the changes in biomass associated with each of

the subsystems and their couplings, an understanding of the dynamics of the

physical processes under investigation by the EPA will provide a knowledge of

the driving variables that affect these subsystems. This synthesis of

biological and physical processes will contribute to the understanding of the

fundamental mechanisms regulating processes common to many estuaries. -279-

REFE RENCES

Admiraal, W. 1977a. Experiments with mixed populations of benthic estuarine

diatoms in laboratory microecosystems. Botanica Marina 20:479-485.

Admiraal, W. 1977b. Influence of light and temperature on the growth rate of

estuarine benthic diatoms in culture. Marine Biology 39:1-9.

Admiraal, W. 1977c. Influence of various concentrations of orthophosphate on

the division rate of an estuarine benthic diatom, Navicula arenaria, in

culture. Marine Biology 42:1-8.

Admiraal, W. 1977d. Salinity tolerance of benthic estuarine diatoms as

tested with a rapid polarographic measurement of photosynthesis.Marine

Biology 39:11-18.

Admiraal, W. 1977e. Tolerance of estuarine benthic diatoms to high concen-

trations of ammonia, nitrite ion, nitrate ion and orthophosphate.Marine

Biology 43:307-315.

Admiraal, W., and Peletier, H. 1979a. Influence of organic compounds and

light limitation on the growth rate of estuarine benthic diatoms. Br.

phycol. J. 14:197-206.

Admiraal, W., and Peletier, H. 1979b. Sulphide tolerance of benthic diatoms

in relation to their distribution in an estuary. Br. phycol. J. 14:185-

196.

Admiraal, W., and Peletier, H. 1980a. Influence of seasonal variations of

temperature and light on the growth rate of cultures and natural popula-

tions of intertidal diatoms. Mar. Ecol. Prog. Ser. 2:35-43.

Admiraal, W., and Peletier, H. 1980b. Distribution of diatom species on an

estuarine mudflat and experimental analysis of the selective effect of

stress. J. exp. mar. Biol. Ecol. (in press). -28 0-

Aioi, K. 1980. Seasonal changes in the standing crop of eelgrass (Zostera

marina L.) in Odawa Bay, Central iapan. Aquat. Bot. 8:343-354.

Aioi, K., Mukai, M., ICoike, I., Ohtsu, M., and Hattori, A. 1981. Growth and

organic production of eelgrass (Zostera marina L.) in temperate waters of

the Pacific Coastof Japan. [I. Growth and analysis in winter. Aquat.

Bot. 10:175-182.

Allen, H. L. 1971. Primary productivity, chemo-organotrophy, and nutritional

interactions of epiphytic algae and bacteria on macrophytes in the

littoral of a lake. Ecol. Monogr. 41:97-127.

Amspoker, M. C., and Mclntire, C. D. 1978. Distribution of intertidal

diatoms associated with sediments in Yaquina Estuary, Oregon. J. Phycol.

14:387-395.

Arasaki, M. 1950a. The ecology of Amamo (Zostera marina) and Koamamp

(Zostera nana). Bull. Jap. Soc. Sci. Fish. 15:567-572.

Arasaki, M. 1950b. Studies on the ecology of Zostera marina and Zostera

nana. II. Bull. Jap. Soc. Sd. Fish. 16:70-76.

Baillie, P. W., and Welsh, B. L. 1980. The effectof tidal resuspension on

the distribution of intertidal epipelic algae in an estuary. Est. Coast.

Mar. Sd. 10:165-180.

Backman, T. W., and Barilotti, D. C. 1976. Irradiance reduction: Effects on

standing crops of the eelgrass Zostera marina in a coastal lagoon. Mar.

Biol. 34:33-40.

Bayer, R. D. 1979. Intertidal zoriation of Zostera marina in the Yaquina

Estuary, Oregon. Syesis 12:147-154.

Bayer, R. D. 1980. Birds feeding on herring eggs at the Yaquina Estuary,

Oregon. Condor 82:193-198. -281-

Bella, D. A. 1971. Mathematical modeling of estuarine benthal systems.

Proc. 1971 Technical Conference on Estuaries of the Pacific Northwest.

OSU Sea Grant Circular #2. Corvallis, OR.

Berner, R. A. 1977. Stoichiometric models for nutrient regeneration in

anoxic sediments. Limnol. Oceanogr. 22:781-786.

Biebl, R., and McRoy, C. P. 1971. Plasmatic resistance and rate of

respiration and photosynthesis of Zostera marina at different salinities

and temperatures. Mar. Biol. 8:48-56.

Bittaker, H. F., and Iverson, R. L. 1976. Thalassia testudinum productivity:

A field comparison of measurement methods. Mar. Biol. 37:39-46.

Blegvad, H. 1914. Food and conditions of nourishment among the communities

of invertebrate animals found on or in the sea bottom in Danish waters.

Rept. Danish Biol. Sta. 22:41-78.

Blegvad, H. 1916. On the food of fish in the Danish waters within the Skaw.

Rept. Danish Biol. Sta. 24:17-72.

Bott, T. L., Brock, J. T., Cushing, C. E., Gregory, S. V., King, D., and

Petersen, R. C. 1978. A comparison of methods for measuring primary

productivity and community respiration in streams. Hydrobiologia 60:3-12.

Branch, G. M., and Branch, M. L. 1980. Competition in Bembicium auratuin

(Gastropoda) and its effect on microalgal standing stock in mangrove mud-

flats. Oecologia (Ben.) 46:106-114.

Brook, A. J. 1955. The aquatic fauna as an ecological factor in studies of

the occurrence of freshwater algae. Rev. Algol. n.s. 3:141-145.

Brook, A. J. 1975. Aquatic animals aren't hungry in winter, or why Cymbella

blooms beneath the ice. J. Phycol. 11:235. -282-

Brown, C. L. 1962. On the ecology of aufuichs of Zostera marina in Charles-

town Pond, Rhode Island. r1.S. Thesis, Univ. of Rhode Island, Kingston.

52 p.

Buchanan, J. B., and Kain, J. M. 1971. Measurements of the physical and

chemical environment. In: N. A. Holmes and A. D. McIntyre (Eds.) Method!

for the study of marine benthos, IBP Handbook No. 16, Blackwell Sci. Pub.,

Oxford, pp. 30-56.

Buesa, R. J. 1977. Photosynthesis and respiration of some tropical marine

plants. Aquatic Bot. 3:203-216.

Burkholder, P. R., Repak, A., and Sibert, J. 1965. Studies on some Long

Island Sound littoral communities of microorganisms and their primary

productivity. Bulletin of the Torrey Botanical Club 92:378-402.

Burkholder, P. R., and Doheny, T. E. 1968. The biology of eelgrass.

Contrib. 3, Dept. Conserv. Waterways, Hampsted, Long Island. Contrib.

1227, Lamont Geological Observatory, Palisades, N.Y. 120 p.

Cadee, G. C., and Hegeman, J. 1974. Primary production of the benthic micro-

flora living on tidal flats in the Dutch Wadden Sea. Neth. J. Sea Res.

8:260-291.

Cadee, G. C., and Hegeman, J. 1977. Distribution of primary production of

the benthic ndcroflora and accumulation of organic matter on a tidal flat

area, Balyzand, Dutch Wadden Sea. Neth. J. Sea. Res. 11:24-41.

Cardon, N. C. 1981. Effect of nutrient enrichment on marine benthic diatoms

in Yaquina Bay, Oregon. M.S. Thesis, Oregon State University, 77 pp.

Castenholz, R. W. 1961. The effect of grazing on marine littoral diatom

populations. Ecology 42:783-794. -28 3-

Charman, K. 1979. Feeding ecology and energetics of the dark-bellied brant

goose (Branta bernicla bernicla) in Essex and Kent. In: R. L. Jefferies

and A. J. Davy (eds.) Ecological processes in coastal environments

Blackwell Sci. Pub., Oxford, pp. 451-465.

Colby, J. A., and Mclritire, C. D. 1978. Mathematical documentation for a

lotic ecosystem model. Internal Report 165, Coniferous Forest Biome,

Ecosystem Analysis Studies, U.S./International IBP. Seattle, WA. 52 p.

Coles, S. M. 1979. Benthic microalgal populations on intertidal sediments

and their role as precursors to salt marsh development. In: R. L.

Jefferies and A. J. Davy (Eds.) Ecological Processes in coastal

environments, Blackwell Sci. Pub., Oxford, pp. 25-42.

Colijin, F. 1974. Primary production, biomass and species composition of

epipelic diatoms in the eastern part of the Dutch shallows. Br. Phycol.

S. 9:217.

Colijin, F., and Buurt, G. van. 1975. Influence of light and temperature on

the photosynthetic rate of marine benthic diatoms. Mar. Biol. 31 :209-214.

Colijin, F., and Dijkema, K. S. 1981. Species composition of benthic diatoms

and distribution of chlorophyll a on an intertidal flat in the Dutch

Wadden Sea. Mar. Ecol. Prog. Ser. 4:9-21.

Conover, S. T. 1958. Seasonal growth of benthic marine plants as related to

environmental factors in an estuary. Pub. Inst. Mar. Sd. 5:97-147.

Conover, S. T. 1968. The importance of natural diffusion gradients and

transport of substances related to benthic marine plant metabolism. Bot.

Mar. 11:1-9.

Cooley, W. W., and Lohnes, P. R. 1971. Multivariate data analysis. Wiley,

New York. 364 p. -284-

Cooper, D. C. 1973. Enhancement of net primary productivity by herbivore

grazing in aquatic laboratory microcosms. Limnol. Oceanogr. 18:31-37.

Cooper, S. W. 1978. The drift algae community of seagrass beds in Redfish

Bay, Texas. Cont. Mar. Sci. 21:125-132.

Cummins, K. W. 1974. Structure and function of stream ecosystems.

Bioscience 24:631-641.

Davis, M. W. 1982. Production dynamics of sediment-associated algae in two

Oregon estuaries. Ph.D. Thesis, Oregon State Univ., Corvallis. 135 p.

Dawes, C. S., Moon, R. E., and Davis, N. A. 1978. The photosynthetic and

respiratory rates and tolerances of benthic algae from a mangrove and salt

marsh estuary: a comparative study. Est. Coast. Mar. Sci. 6:175-185.

Dayton, P. K. 1971. Competition, dispersal and persistence of annual

intertidal alga, Postelsia palmaeformis Ruprecht. Ecology 54:433-438.

Dayton, P. K. 1975. Environmental evaluation of ecological dominance in a

rocky intertidal community. Ecol. Monogr. 45:137-159.

DeBoer, J. A., Guigli, H. S., Israel, T. L., and D'Elia, C. F. 1978.

Nutritional studies of two red algae. I. Growth rate as a function of

nitrogen source and concentration. J. Phycol. 14:261-266. den Hartog, C. 1971. The dynamic aspect in the ecology of seagrass

communities. Thalassia yugoslavica 7:101-112. den Hartog, C. 1977. Structure, function and classification in seagrass

communities. In: C. P. NcRoy and C. Helfferich (Eds.) Seagrass

Ecosystems: A Scientific Perspective, Marcel Dekker, N.Y. pp. 89-121.

Dennison, W. 1979. Light adaptations of plants: A model based on Zostera

marina L. M.S. Thesis, Univ. Alaska, Fairbanks. 56 p. -285-

Dillon, R. R. 1971. A comparative study of the primary productivity of

estuarine phytoplankton and macrobenthic plants. Ph.D. Thesis, Univ.

North Carolina, Chapel Hill. 112 p.

DiToro, D. M., Thomann, R. V., O'Conner, D. J., and Mancini, J. L. 1977.

Estuarine phytoplankton biomass models-verification analyses and pre-

liminary applications. In: E. D. Goldberg, I. N. McCave, J. J. O'Brien,

J. H. Steele (Eds.) The Sea, Vol. 6. Wiley-Interscience, New York.

pp. 696-1020.

Drum, R. W., and Webber, E. 1966. Diatoms from a Massachusetts salt marsh.

Bot. Mar. 9:70-77.

Duff, S., and Teal, J. M. 1965. Temperature change and gas exchange in Nova

Scotia and Georgia salt marsh muds. Limnol. Oceanogr. 10:67-73.

Dye, A. H. 1978. Epibenthic algal production in the Swartkops Estuary.

Zool. Afr. 13:157-158.

Edwards, P. 1969. Field and cultural studies on the seasonal periodicity of

growth and reproduction of selected Texas benthic marine algae. Cont.

Mar. Sci. 14:59-114.

Edwards, P., and Kapraun, D. F. 1973. Benthic marine algal ecology in Port

Aransas, Texas area. Cont. Mar. Sc 17:15-52.

Edwards, R. R. C. 1978. Ecology of a coastal lagoon complex in Mexico. Est.

Coast. Mar. Sci. 6:75-92.

EPA, 1979. Unpublished data, Wetlands Research Program. Corvallis, Oregon.

Estrada, M., Valiela, I., and Teal, J. M. 1974. Concentration and distri-

bution of chlorophyll in fertilized plots in a Massachusetts salt marsh.

J. expt. mar. Biol. Ecol. 14:47-56. -28 6-

Fenchel, T. 1969. The ecology of marine microbenthos. IV. Structure and

function of the benthic ecosystem, its chemical and physical factors and

the microfaunal communities with special reference to the ciliated

protozoa. Ophelia 6:1-187.

Fenchel, T., and Kofoed, L. H. 1976. Evidence for exploitative interspecific

competition in inudsnails (Hydrobidae). Oikos. 27:367-376.

Fitzgerald, W. J., Jr. 1978. Environmental parameters influencing the growth

of Enteromorpha clathrata (Roth) J. Ag. in the intertidal zone on Guam.

Bot. Mar. 21:207-220.

Frostick,L. E., and McCave, I. N. 1979. Seasonal shifts of sediment within

an estuary mediated by algal growth. Est. Coast. Mar. Sci. 9:569-576.

Gallagher, J. L., and Daiber, F. C. 1974. Primary production of edaphic

algal communities in a Delaware salt marsh. Limnol. Oceanogr. 19:390-395.

Coering, J. J., and P. L. Parker. 1972. Nitrogen fixation by epiphytes on

sea grasses. Limnol. Oceanogr. 17:320-323.

Goldberg, E. D., McCave, I. N., O'Brien, J. J., and Steele, J. H. 1977. The

Sea. Vol. 6. Marine Modeling. J. Wiley and Sons. N.Y. 1048 pp.

Goodwin, C. R., Emmett, E. W., and Glenne, B. 1970. Tidal study of three

Oregon estuaries. Bull. 45, Engineering Experiment Station, Oregon State

University, Corvallis, Oregon. 42 pp.

Grntved, J. 1960. On the productivity of microbenthos and phytoplankton in

some Danish fjords. Meddr. Danm. Fiskeri-og Havunders. N.S. 3:55-92.

Grntved, J. 1962. Preliminary report on the productivity of microbenthos

and phytoplankton in the Danish Wadden Sea. Meddr. Damn. Fiskeri-og

Havunders 3:347-378. -287-

Grundinanis, V., and Murray, J. W. 1977. Nitrification and denitrification in

marine sediments from Puget Sound. Limnol. Oceanogr. 22:804-813.

Haardt, J., and Nielsen, G. 1980. Attenuation measurements of monochromatic

light in marine sediments. Oceanologica Acta 3:333-338.

Hall, C. A. S., Tempel, N., and Peterson, B. J. 1979. A benthic chamber for

intensely metabolic lotic systems. Estuaries 2:178-183.

Hall, M. 0., and Elseman, N. J. 1981. The seagrass epiphytes of the Indian

River, Florida. I. Species list with descriptions and seasonal

occurrences. Bat. Mar. 24:139-146.

Harding, L. W., Jr., and Butler, J. H. 1979. The standing stock and produc-

tion of eelgrass, Zostera marina, in Humboldt Bay, California. Calif.

Fish and Game 65:151-158.

Hargrave, B. T. 1969. Epibenthic algal production and community respiration

in the sediments of Marion Lake. J. Fish. Res. Bd. Can. 26:2003-2026.

Hargrave, B. T. 1970. The effect of a deposit-feeding amphipod on the

metabolism of benthic microflora. Limnol. Oceanogr. 15:21-30.

Hargrave, B. T. 1976. The central role of invertebrate feces in sediment

decomposition. En: The Role of Terrestrial and Aquatic Organisms in

Decomposition Processes. Anderson, J. M. and Macfayden, A. (Eds.).

Blackwell Sci. Publ., Oxford. pp. 301-321.

Marlin, N. M. 1971. Epiphytic marine algae: Interactions with their hosts.

Ph.D. Thesis, Univ. Washington, Seattle. 194 p.

Harlin, M. M. 1973. Transfer of products between epiphytic marine algae and

host plants. J. Phycol. 9:243-248.

Harlin, M. N. 1975. Epiphyte-host relations in seagrass communities. Aquat.

Bat. 1:125-131. -288-

Harlin, M. M. 1980. Seagrass epiphytes. In: R. C. Phillips and C. P. McRoy

(Eds.) Handbook of Seagrass Biology: An Ecosystem Perspective, Garland

STPM Press, New York.

Harper, M. A. 1977. Movements. In: D. Werner (Ed.) The biology of diatoms,

Blackwell Sd. Pub., Oxford, pp. 224-249.

Harrison, P. G. 1982. Spatial and temporal patterns in abundance of two

intertidal seagrasses, Zostera americana den Hartog and Zostera marina L.

Aquat. Bot. 12:305-320.

Hartman, R. T., and Brown, D. L. 1967. Changes in internal atmosphere of

submersed vascular hydrophytes in relation to photosynthesis. Ecology

48: 252-258.

Hellebust, J. A., and Lewin, J. 1977. Heterotrophic nutrition. In: D.

Werner (Ed.) The biology of diatoms, Blackwell Sci. Pub., Oxford, pp. 169-

197.

Hickman, M. 1971. The standing crop and primary productivity of the

epiphyton attached to Equisetum fluviatile L. in Priddy Pool, North

Somerset. Br. Phycol. J. 6:51-59.

Hitchcock, C. L., and Cronquist, A. 1973. Flora of the Pacific Northwest:

An Illustrated Manual. University of Washington Press, Seattle. 730 p.

Hossell, J. C., and Baker, J. H. 1979. Estimation of the growth rates of

epiphytic bacteria and Lemna minor in a river. Freshwater Biol. 9:319-

327.

Howarth, R. W., and Teal, J. M. 1979. Sulfate reduction in a 'Tew England

salt marsh. Liinnol. Oceanogr. 24:999-1013.

Hunding, C., and Hargrave, B. T. 1973. A comparison of benthic microalgal

production measured by C-14 and oxygen methods. J. Fish. Res. Board Can.

30: 309-3 12. -289-

Inman, D. L. 1952. Measures for describing the size distribution of

sediments. J. Sed. Petrol. 22:125-145.

Jacobs, R. P. W. M. 1979. Distribution and aspects of the production and

biomass of eelgrass, Zostera marina L., at Roscoff, France. Aquat. Bot.

7:151-172.

Johnson, R. 0. 1977. Vertical variation in particulate matter in the upper

twenty centimeters of marine sediments. J. Mar. Res. 35:273-282.

Joint, I. R. 1978. Microbial production of an estuarine mudflat. Est.

Coast. Mar. Sci. 7:185-195.

Jones, J. A. 1968. Primary productivity by the tropical marine turtle grass,

Thalassia testudinum, and its epiphytes. Ph.D. Thesis, Univ. Miami, Coral

Gables. 206 p.

Jonge, V. . de. 1980. Fluctuation in the organic carbon to chlorophyll a

ratios for estuarine benthic diatom populations. Mar. Ecol. Prog. Ser.

2:345-353.

Jorgensen, B. B. 1977. The sulfur cycle of a coastal marine sediment

(Limfjorden, Denmark). Litunol. Oceanogr. 22:814-832.

Jorgensen, N., Mopper, K., and Lindroth, P. 1980. Occurrence, origin and

assimilation of free amino acids in an estuarine environment.Ophelia,

Suppi. 1:179-192.

Josselyn, M. N., and Mathieson, A. C. 1980. Seasonal flux and decomposition

of autochthonous macrophyte litter in a north temperate estuary.

Hydrobiol. 71:197-208.

Keller, M. 1963. Growth and distribution of eelgrass (Zostera marina L.) in

Humboldt Bay, California. M.S. Thesis, Humboldt State College. 53 p.

Keller, M., and Harris, S. W. 1966. The growth of eelgrass in relation to

tidal depth. J. Wild. Manage. 30:280-285. -290-

Kelly, R. A. 1976. Conceptual ecological model of the Delaware Estuary. In:

B. C. Patten (Ed.) Systems Analysis and Simulation in Ecology Vol. [V.

Academic Press, New York. pp. 3-45.

Kikuchi, T., and Peres, J. M. 1973. Animal communities in the seagrass beds:

A review. Submitted to: International Seagrass Workshop, Leiden, Nether-

lands, 22-26 October 1973.

King, R. J., and Schramm, W. 1976. Photosynthetic rates of benthic marine

algae in relation to light intensity and seasonal variations. Mar. Biol.

37:215-222.

Kjeldsen, C. 1967. Effects of variation in salinity and temperature in some

estuarine macroalgae. Ph.D. Thesis, Oregon State University, Corvallis,

OR, 157 pp.

Kjeldsen, C. K., and Phinney, H. K. 1971. Effects of variations in salinity

and temperature on some estuarine macroalgae. In: Proceedings of the

Seventh International Seaweed Symposium. J. Wiley and Sons. N.Y.,

pp. 301-308.

Klir, C. J. 1969. An approach to general systems theory. Van Nostrand

Reinhold Co., N.Y. 323 pp.

Kneib, R. J., Stiven, A. E., and Haines, E. B. 1980. Stable carbon isotope

ratios in Fundulus heteroclitus (L.) muscle tissue and gut contents from

North Carolina Spartiria marsh. J. expt. mar. Biol. Ecol. 46:89-98.

Kreag, R. A. 1979. Natural resources of Netarts Estuary. Estuary Inventory

Report, Vol. 2, No. 1. Oregon Dept. of Fish and Wildlife. Newport,

Oregon. 45 p.

Krumbein, W. C., and Pettijohn, F. J. 1983. Manual of Sedimentary

Petrography. Appleton-Century-Crafts, New York. 549 pp. -29 1-

Kulm, L. D., and Byrne, J. V. 1966. Sedimentary response to hydrography in

an Oregon estuary. Mar. Geol. 4:85-118.

Lapointe, B. E., Williams, L. D., Goldman, J. C., and Ryther, J. H. 1976.

The mass outdoor culture of macroscopic algae. Aquaculture 8:9-20.

Leach, H. J. 1970. Epibenthic algal production in an intertidal mudflat.

Limnol. Oceanogr. 15:514-521.

Lee, H. II. and Swartz, R. C. 1980. Biological processes affecting the

distribution of pollutants in marine sediments. Part II. Biodeposition

and bioturbation. In: R. A. Baker (Ed.) Contaminants and sediments, Vol.

2, Ann Arbor Sd. Pub., pp. 555-606.

Levinton, J. S. 1980. Particle feeding by deposit feeders: models, data and

a new perspectus. En: K. R. Tenore and B. C. Coull (Eds.) Marine benthic

dynamics, Univ. of S. Carolina Press, Columbia,pp. 423-439.

Littler, M. M., Murray, S. N., and Arnold, K. E. 1979. Seasonal variations

in net photosynthetic performance and cover of intertidal macrophytes.

Aquatic Bot. 7:35-46.

Lubchencho, J. 1978. Plant species diversity in a marine intertidal

community: importance of herbivore food preference and algal competitive

abilities. Amer. Nat. 112:23-39.

Main, S. P., and Mclntire, C. D. 1974. The distribution of epiphytic diatoms

in Yaquina Estuary, Oregon (USA). Botanica Marine 17:88-99.

Marshall, N., Oviatt, C. A., and Skauen, D. M. 1971. Productivity of the

benthic microflora of shoal estuarine environments in southern New

England. tnt. Rev. ges. Hydrobiol. 56:947-956. -29 2-

Marshall, H., Skauen, D. M., Lampe, H. C., and Oviatt, C. A. 1973. Primary

production of benthic microflora. In: A Guide to the Measurement of

Marine Primary Production under some Special Conditions. UNESCO, Paris.

pp. 37-44.

Matheke, G. B. M., and Homer, R. 1974. Primary production of the benthic

microalgae in the Chukchi Sea near Barrow, Alaska. J. Fish. Res. Board

Can. 31:1779-1786.

McConnell, R. J., Durbin, 3. T., Misitano, D. A., and Sanborn, H. R. 1973.

Checklist of aquatic organisms in the lower Columbia and Willamette

Rivers. Northwest Fisheries Center, U.S. Nati. Mar. Fish. Ser. 5 pp.

Mclntire, C. D. 1968. Structural characteristics of benthic algal communi-

ties in laboratory streams. Ecology 49:520-537.

Mclntire, C. D. 1973. Periphyton dynamics in laboratory streams: a

simulation model and its implications. Ecol. Monogr. 43:399-420.

Mclntire, C. D. 1978. The distribution of estuarine diatoms along environ-

mental gradients: A canonical correlation. Estuarine and Coastal Marine

Science 6:447-457.

Mclntlre, C. D., and Wulff, B. L. 1969. A laboratory method for the study of

marine benthic diatoms. Limnol. Oceanogr. 14:667-678.

Mclntire, C. D., and Colby, 3. A. 1978. A hierarchical model of lotic

ecosystems. Ecol. Monogr. 48:167-190.

Mclntire, C. D., and Phinney, H. K. 1965. Laboratory studies of periphyton

production and community metabolism in lotic environments. Ecol. Monogr.

35:237-258.

Mclntire, C. D., Garrison, R. L., Phinney, H. K., and Warren, C. E. 1964.

Primary production in laboratory streams. Limnol. Oceanogr. 9:92-102. -293-

Mclntire, C. D., and Overton, W. S. 1971. Distributional patterns in

assemblages of attached diatoms from Yaquina Estuary, Oregon.Ecology

52:758-777.

Nclntire, C. D., and Moore, W. W. 1977. Marine littoral diatoms: Ecological

considerations. In: D. Werner's (Ed.) The Biology of Diatoms. Blackwell

Scientific Publications. p. 333-371.

McIntyre, A. D., Munro, A. L. S., and Steele, J. H. 1970. Energy flow in a

sand ecosystem. En: J. H. Steele (Ed.) Marine food chains, Univ. Calif.

Press, Berkeley, pp. 19-31.

McLean, R. 0., Corrigan, J., and Webster, J. 1981. Heterotrophic nutrition

in Melosira nummuloides, a possible role in affecting distribution in the

Clyde Estuary. Br. phycol. J. 16:95-106.

McMillan, C., and Phillips, R. C. 1979. Differentiation in habitat response

among populations of New World seagrasses. Aquat. Bot. 7:185-196.

McMillan, C, Zapata, 0., and Escobar, L. 1980. Suiphated phenollc compounds

in seagrasses. Aquat. Bot. 8:267-278.

McRoy, C. P. 1966. The standing stock and ecology of eelgrass, Zostera

marina, Izembeck Lagoon, Alaska. M.S. Thesis, Univ. Washington, Seattle.

138 p.

McRoy, C. P. 1969. Eelgrass under winter ice. Nature 224:818-819.

McRoy, C. P. 1970a. On the biology of eelgrass in Alaska. Ph.D. Thesis,

Univ. Alaska, Fairbanks. 156 p.

McRoy, C. P. 1970b. Standing stocks and other features of eelgrass (Zostera

marina) populations on the coast of Alaska. J. Fish. Res. Board Canada

27:1811-1821. -2 94-

McRoy, C. P. (Ed.). 1973. Seagrass ecosystems: Research recommendations of

the International Seagrass Workshop. National Science Foundation,

Washington. 62 p.

McRoy, C. P. 1974. Seagrass productivity: Carbon uptake experiments in

eelgrass, Zostera marina. Aquaculture 4:131-137.

McRoy, C. P., and Barsdate, R. J. 1970. Phosphate absorption in eelgrass.

Limnol. Oceanogr. 15:14-20.

McRoy, C. P., Barsdate, R. J., and Nebert, M. 1972. Phosphorus cycling in an

eelgrass (Zostera marina L.) ecosystem. Limnol. Oceanogr. 17:58-67.

McRoy, C. P., and Goering, J. J. 1974. Nutrient transfer between the sea-

grass Zostera marina and its epiphytes. Nature 248:173-174.

McRoy, C. P., and Hefferich, C. (Eds.). 1977. Seagrass Ecosystems: A

Scientific Perspective. Marcel Dekker, Inc., New York. 314 p.

McRoy, C. P., and Lloyd, D. S. 1981. Comparative function and stability of

macrophyte-based ecosystems. In: A. R. Longhurst (Ed.). Analysis of

Marine Ecosystems. Academic Press, London. pp. 473-489.

McRoy, C. P., and McMillan, C. 1977. Production ecology and physiology of

seagrasses. In: C. P. McRoy and C. Hefferich (Eds.). Seagrasses

Ecosystems: A Scientific Perspective. Marcel Dekker, Inc., New York.

pp. 53-87.

Mills, E. L. 1967. The biology of an ampeliscid crustacean sibling species

par. J. Fish. Res. Bd. Can. 24:305-355.

Mime, L. J., and Mime, M. J. 1951. The eelgrass catastrophe. Sci. Am.

184:52-55.

Misitano, D. A. 1973. A checklist of zooplankton in the lower Columbia and

Willamette Rivers, July to October 1973. Northwest Fisheries Center.

13. S. Mar. Fish. Ser. 11 pp. -29 5-

Montgomery, J. R., Zimmerman, C. F., and Price, M. T. 1979. The collection,

analysis and variation of nutrients in estuarine pore water. Est. Coast.

Mar. Sd. 9:203-214.

Moore, W. W., and Mclntire, C. D. 1977. Spatial and seasonal distribution of

littoral diatoms in Yaquina Estuary, Oregon (USA). Botanica Marina 20:99-

109.

Morgan, J. B., and Flolton, R. L. 1977. A bibliography of estuarine research

in Oregon. Oregon estuarine Research Council, No. 1. Oregon State Sea

Grant, Corvallis, Oregon. 141 p.

Moul, E. T., and Mason, D. 1957. Study on diatom populations on sand and

mudflats in the Woods Hole area. Biol. Bull. 113:351.

Mukai, H., Aioi,1. K., lizumi, H., Ohtsu, M., and Hattori, A. 1979. Growth

and organic production in eelgrass (Zostera marina L.) in temperate waters

of the Pacific Coast of Japan. I. Growth analysis of spring-summer.

Aquat. Bot. 7:47-56.

Myers, A. C. 1977. Sediment processing in a marine subtidal sandy bottom

community. 1. Physical aspects. J. Mar. Res. 35:609-632.

Naiman, R. J., and Sibert, J. R. 1979. Detritus and juvenile salmon

production in the Nanaimo Estuary: III. Importance of detrital carbon to

the estuarine ecosystem. J. Fish. Res. Board Can. 36:504-520.

Newell, S. Y. 1981. Fungi and bacteria in or on leaves of eelgrass (Zostera

marina L.) from Chesapeake Bay. Applied and Environ. Microbiology

41:1219-1224.

Nie, N. H., Hill, C. H., Jenkins, J. S., Steinbrenner, K., and Brent, D. H.

1975. SPSS: Statistical package for the social sciences. 2nd. Ed.

McGraw-Hill, Inc., NY. 675 p. -29 6-

Nienhuis, P. H. 1970. The benthic algal communities of flats and salt

marshes in the Grevelingen, a sea-arm in the southwestern Netherlands.

Neth. J. Sea. Res. 5:20-49.

Nienhuis, P. H. 1980. The eelgrass (Zostera marina L.) subsystem in brackish

Lake Grevelingen: Production and decomposition of organic matter.

Ophelia, Suppi. 1:113-116.

Nienhuis, P. H., and De Bree, B. H. H. 1977. Production and ecology of eel-

grass (Zostera marina L.) in the Grevelingen Estuary, The Netherlands,

before and after closure. Hydrobiologia 52:55-66.

Nienhuis, P. H., and De Bree, B. H. H. 1980. Production and growth dynamics

of eelgrass (Zostera marina L.) in brackish Lake Grevelingen (The Nether-

lands). Neth. J. Sea Res. 14:102-118.

Nixon, S. W., and Oviatt, 0. A. 1972. Preliminary measurements of midsummer

metabolism in beds of eelgrass, Zostera marina. Ecology 53:150-153.

Odum, H. T. 1956. Primary Production in Flowing Waters. Limnol. Oceanogr.

1:102-117.

Ogata, E., and Natsui, T. 1965. Photosynthesis in several marine plants of

Japan as affected by salinity, drying and pH with attention to their

growth habitats. Bot. Mar. 8:199-217.

Ostenfeld, C. H. 1905. Preliminary remarks on the distribution and biology

of the Zostera on the Danish seas. Bot. Tidskr. 27:123-125.

Ostenfeld, C. H. 1908. On the ecology and distribution of grasswrack

(Zostera marina) in Danish waters. Rept. Danish Biol. Sta. 16:1-62.

Ott, J. A. 1980. Growth and production in Posidonia oceanica (L.) Delile.

Mar. Ecol. 1:47-64. -29 7-

Overton, W. S. 1972. Toward a general model structure for a forest

ecosystem. In: J. F. Franklin, L. J. Dempster and R. H. Waring (Eds.).

Proceedings of the Symposium on Research on Coniferous Forest Ecosystems.

Pacific Northwest Forest and Range Experiment Station, Portland, OR.

pp. 37-47.

Overton, W. S. 1975. The ecosystem modeling approach in the coniferous

forest biome. In: B. C. Patten (Ed.). Systems Analysis and Simulation

in Ecology. Vol. 3, Academic Press, New York. pp. 117-138.

Overton, W. S. 1977. A strategy of model construction. In: C. A. S. Hall

and J. W. Day Jr. (Eds.), Ecosystem Modeling in Theory and Practice: An

Introduction with Case Histories. John Wiley and Sons, NY. p. 50-73.

Pace, M. L., Shimmel, S., and Darley, W. M. 1979. The effect of grazing by a

gastropod, Nassarius obsoletus, on the benthic microbial community of a

salt marsh mudflat. Est. Coast. Mar. Sci. 9:121-134.

Paine, R. T. 1971. Energy flow in a natural population of the herbivorous

gastropod Tegula funebralis. Limnol. Oceanogr. 16:86-98.

Pamatmat, M. M. 1965. A continuous flow apparatus for measuring metabolism

by benthic communities. Lininol. Oceanogr. 10:486-489.

Pamatmat, N. M. 1968. Ecology and Metabolism of a benthic community on an

intertidal sandflat. mt. Revue Ges. Hydrobiol. 53:211-298.

Parsons, T. R., Takahashi, M., and Hargrave, B. T. 1977. Biological

oceanographic process. 2nd ed. Pergamon Press, N.Y. 332 p.

Patriquin, D. 1973. Estimation of growth rate, production and age of the

marine angiosperm Thalassia testudinum Konig. Carib. J. Sci. 13:111-123. Penhale, P. A. 1976. Primary productivity, dissolved organic carbon

excretion, and nutrient transport in an epiphyte-eelgrass (Zostera marina)

system. Ph.D. Thesis, North Carolina State Univ., Raleigh. 82 p.

Penhale, P. A. 1977. Macrophyte-epiphyte biomass and productivity in an

eelgrass (Zostera marina L.) community. J. Exp. Mar. Biol. Ecol. 26:211-

224.

Penhale, P. A., and Thayer, G. W. 1980. Uptake and transfer of carbon and

phosphorus by eelgrass (Zostera marina L.) and its epiphytes. J. Exp.

Mar. Biol. and Ecol. 42:113-124.

Percy, K. L., Sutterlin, C., Bella, D. A., and Klingeman, P. C. 1974.

Descriptions and information sources for Oregon estuaries. Oregon State

University Sea Grant. Corvallis, Oregon. 294 p.

Petersen, C. G. J. 1891. Fiskenes biologiske Forhold I Holbaek Fjord. Rept.

Danish Biol. Sta. 1:1-63.

Petersen, C. G. J. 1913. Om Baendeltangens (Zostera marina) Aarsproduktion i

de danske Farvande. Mindeskr. Steenstr. Fds. Kbn. 9:1-20.

Petersen, C. G. J. 1915. A preliminary result of investigations in the

valuation of the sea. Rept. Danish Biol. Sta. 23:29-33.

Petersen, C. G. J. 1918. The sea bottom and its production of fish food. A

survey on the work done in connection with valuation of the Danish waters

from 1883-1917. Rept. Danish Biol. Sta. 25:1-82.

Petersen, C. G. J., and Boysen-Jensen, P. 1911. Valuation of the sea, 1.

Animal life of the sea bottom, its food and quantity. Rept. Danish Biol.

Sta. 20:1-81.

Pettitt, J., Ducker, S., and Knox, B. 1981. Submarine pollination. Sd Am.

224:134-144. -299-

Phillips, R. C. 1972. Ecological life history of Zostera marina L.

(eelgrass) in Puget Sound, Washington. Ph.D. Thesis, Univ. Washington,

Seattle, 154 p.

Phillips, R. C. 1978. Seagrasses and the coastal marine environment.

Oceanus 21:30-41.

Phillips, R. C. 1979. Ecological notes on Phyllospadix (Potamogetonaceae) in

the Northeast Pacific. Aquat. Bot. 6:159-170.

Phillips, R. C., and McRoy, C. P. (Eds.). 1980. A Handbook of Seagrass

Biology: An Ecosystem Perspective, Garland STPM Press, New York. 352 p.

Phinney, H. K. 1977. The macrophytic algae of Oregon. In: R. W. Krauss

(Ed.). The marine plant biomass of the Pacific Northwest coast. Oregon

State Univ. Press, Corvallis. 397 pp.

Platt, T., Denman, K. L., and Jassby, A. D. 1977. Modeling the productivity

of phytoplankton. In: E. D. Goldberg, I. N. McCave, J. J. O'Brien, and

J. J. Steele (Eds.). The Sea. Vol. 6: Marine Modeling. 3. Wiley and

Sons. N.Y. 1048 p.

Pomeroy, L. R. 1959. Algal productivity in salt marshes of Georgia. Limnol.

Oceanogr. 4:386-397.

Pomeroy, L. R., Bancroft, K., Breed, 3., Christian, R. R., Frankenberg, D.,

Hall, J. R., Maurer, L. G., Wiebe, W. J., Wiegert, R. G., and Wetzel, R.

L. 1977. Flux of organic matter through a salt marsh. In: M. Wiley

(Ed.) Estuarine process Vol. 2, Academic Press, N.Y.,pp. 270-279.

Pomeroy, W. M., and Stockner, 3. G. 1976. Effects of environmental

disturbance on the distribution and primary production of benthic algae on

a British Columbia estuary. J. Fish. Res. Bd. Can. 33:1175-1334. -300-

Porter, K. G. 1976. Enhancement of algal growth and productivity by grazing

zooplankton. Science 192:1332-1334.

Pritchard, D. W. 1967. What is an estuary: physical viewpoint. In: G. H.

Lauff (Ed.) Estuaries, pub. no. 83 Amer. Assoc. Adv. Sd. Washington,

D.C., pp 3-5.

Ramus, J., and Rosenberg, G. 1980. Diurnal photosynthetic performance of

seaweeds measured under natural conditions. Mar. Biol. 56:21-28.

Revsbech, N. P., and Jorgensen, B. B. 1981. Primary production of microalgae

in sediments measured by oxygen microprofile, H14CO3Fixation, and oxygen

exchange methods. Limnol. Oceanogr. 36:717-730.

Ribelin, B. W. and Collier, A. W. 1980. Ecological considerations of

detrital aggregates in the salt marsh. In: R. J. Livingston (Ed.)

Ecological processes in coastal and marine systems, Plenum Press, N.Y.,

pp. 47-68.

Riley, G. A., Stommel, H., and Bumpus, D. F. 1949. Quantitative ecology of

the plankton of the Eastern North Atlantic. Bull. Bingham Oceanogr. Coil.

12: 1-169.

Riznyk, R. Z., and Phinney, H. K. 1972. Manometric assessment of inter-

stitial microalgal production in two estuarine sediments. Oecologia

(Ben.) 10:193-203.

Round, F. E. 1971. Benthic marine diatoms. Oceanogr. Mar. Biol. Ann. Rev.

9:83-139.

Rowe, K., and Brenne, R. 1981. Statistical interactive programming system

(SIPS). Statistical Computing Report No. 7. Oregon State Univ.,

Corvallis. 151 p.

Ryther, J. H. 1956. Photosynthesis in the ocean as a function of light

intensity. Limnol. Oceanogr. 1:61-70. -301-

Sanders, H. L., Goudsmit, E. M., Mills, E. L., and Hampson, G. E. 1962. A

study of the intertidal fauna of Barnstable Harbor, Massachusetts.

Limnol. Oceanogr. 7:63-79.

Sand-Jensen, K. 1975. Biomass, net production and growth dynamics in an

eelgrass (Zostera marina L.) population in Vellerup Vig, Denmark. Ophelia

14:185-201.

Sand-Jensen, K. 1977. Effects of epiphytes on eelgrass photosynthesis.

Aquat. Bot. 3:55-63.

Sawyer, C. N. 1966. The sea lettuce problem in Boston Harbor. J. Wat. Poll.

Contr. Fed. 37:1122-1133.

Sedell, J. R., Triska, F. J., and Triska, N. S. 1975. The processing of

conifer and hardwood leaves in two coniferous forest streams. I. Weight

loss and associated invertebrates. Verh. Internat. Verein. Limnol.

19:1599-1609.

Setchell, W. A. 1929. Morphological and phenological notes on Zostera marina

L. Univ. of California, Publ. Bot. 14:389-452.

Sheldon, R. W., and Parsons, T. R. 1967. A practical manual on the use of

the Coulter Counter in Marine Science. Coulter Electronics, Toronto.

66 pp.

Sieburth, J. McN., and Thomas, C. D. 1973. Fouling in eelgrass (Zostera

marina L.). J. Phycol. 9:46-50.

Snedecor, G. W. and Cochran, W. G. 1967. Statistical methods (6th Ed.).

Iowa State University Press, Ames, Iowa. pp. 593.

Sondergaard, M. 1979. Light and dark respiration and the effect of the

lacunal system on ref ixation of CO2 in submerged aquatic plants. Aquat.

Bot. 6:269-283. -3 02-

Steele, J. H. 1974. The structure of marine ecosystems. Harvard Univ. Press

Cambridge. 127 p.

Steele, J. H., and Millin, M. M. 1977. Zooplankton dynamics. In: E. D.

Goldberg, I. N. McCave, J. J. O'Brien, J. H. Steele (Eds.). The Sea, Vol.

6. Wiley-Interscience, N.Y. pp. 857-890.

Steele, J. H., and Baird, I. E. 1968. Production ecology of a sandy beach.

Limnol. Oceanogr. 13:14-25.

Steeman Nielsen, E., and Hansen, V. K. 1959. Light adaption in marine

phytoplankton populations and its interrelation with temperature.

Physiol. Plant. 12:353-370.

Stout, H. (Ed.). 1976. The natural resources and human utilization of

Netarts Bay, Oregon. Oregon State University, Corvallis, Oregon. 247 p.

Strickland, J. D. H., and Parsons, T. R. 1972. A practical handbook of sea-

water analysis. Fish. Res. Bd. Can. Bull. 167 (2nd edi.): 1-310

Sullivan, M. J. 1976. Long-term effects of manipulating light intensity and

nutrient enrichment on the structure of a salt marsh diatom community.

J. Phycol. 12:205-210.

Sullivan, M. J. 1979. Epiphytic diatoms of three seagrass species in

Mississippi Sound. Bull. Mar. Sci. 29:459-464.

Sullivan, M. J., and Daiber, F. C. 1975. Light, nitrogen and phosphorus

limitation of edaphic algae in a Delaware salt marsh. J. expt. mar. Biol.

Ecol. 18:79-88.

Taylor, W. R. 1964. Light and photosynthesis in intertidal benthic diatoms.

Helgolander wiss. Meereseunters 10:29-37.

Taylor, W. R., and Gebelein, C. D. 1966. Plant pigments and light penetra-

tion in intertidal sediments. Helgolander wiss. Meereseunters 13:229-237. -303-

Tenore, K. R. 1977. Food chain pathways in detrital feeding benthic

communities: a review, with new observations on sediment resuspension and

detrital recylcing. In: B. C. Coull (Ed.) Ecology of marine benthos,

Univ. S. Carolina Press, Columbia, pp. 37-53.

Thayer, G. W., Adams, S. M., and LaCroix, M. W. 1975. Structural and

functional aspects of a recently established Zostera marina community.

tn: L. E. Cronin (Ed.). Estuarine research, Vol. 1, Chemistry, biology.

Thayer, G. W., Wolfe, D. A., and Williams, R. B. 1975. The impact of man on

seagrass sytems. Amer. Scientist 63:288-296.

Tjepkema, J. D., and Evans, H. J. 1976. Nitrogen fixation associated with

Juncus balticus and other plants of the Oregon wetlands. Soil Biol.

Biochem. 8:505-509.

Tomlinson, P. B. 1974. Vegetative morphology and meristem dependence--the

foundation of productivity in seagrasses. Aquaculture 4:107-130.

Tutin, T. G. 1942. Zostera. J. Ecol. 30:217-226. van den Ende, C., and Haage, P. 1963. Der Epiphytenbewuchs von Zostera

marina und der bretonischen Kuste. Bot. Mar. 5:105-110.

Vanderborght, J. P., Wollast, R., and Billen, G. 1977a. Kinetic models of

diagenesis in disturbed sediments. Part I. Mass transfer properties and

silica diagenesis. Limnol. Oceanogr. 22:787-793.

Vanderborght, J. P., Wollast, R., and Billen, G. 1977b. Kinetic models of

diagenesis in disturbed sediments. Part II. Nitrogen diagenesis.

Limnol. Oceanogr. 22:794-803.

Van Raalte, C. D., Stewart, W. C., and Valieta, I. 1974. A C-14 technique

for measuring algal productivity in salt marsh muds. Botanica Marina

17:186-188. -3 04-

Van Raalte, C. D., Valiela, I., and Teal, J. M. 1976. Production of

epibenthic salt marsh algae: Light and nutrient limitation. Limnol.

Oceanogr. 21:862-872.

Vollenweider, R. A. 1974. A manual on methods for measuring primary produc-

don in aquatic environments. IBP Handbook No. 12. Blackwell Sci. Pub.,

Oxford. 225 p.

Waddell, J. E. 1964. The effect of oyster culture on eelgrass (Zostera

marina L.) growth. M.S. Thesis, Humboldt State College. 48 p.

Welsh, B. L. 1980. Comparative nutrient dynamics of a rnarsh-mudflat

ecosystem. Est. Coast. Mar. Sci. 10:143-164.

Westlake, D. F. 1965. Some basic data for investigations on the productivity

of aquatic macrophytes. In: C. R. Goldman (Ed.) Primary Production in

Aquatic Environments, Mem. 1st. Ital. Idrobiol. Suppi. 18, Univ. Calif.

Press, Berkeley, pp. 231-248.

Wetzel, R. G. 1974. The enclosure of macrophyte communities. In: R. A.

Vollenweider (Ed.). IBP Handbook No. 12, A Manual on Methods for

Measuring Primary Production in Aquatic Environments. (2nd ed.). Oxford:

Blackwell Scientific PubI. pp. 100-107.

Wetzel, R. G., and Allen, H. L. 1972. Functions and interactions of

dissolved organic matter and the littoral zone in lake metabolism and

eutrophication. In: Z. Kajak and A. Hillbricht-Ilkowska (Eds.).

Productivity Problems of Freshwaters, Polish Scientific Pubi., Warsaw.

pp. 333-347.

White, D. D., Findley, R. H., Fazio, S. D., Bobbie, R. J., Nickels, J. S.,

Davis, W. M., Smith, G. A., and Martz, R. F. 1980. Effects of bioturba-

tion and predation by Mellita guinguiesperforata on sedimentary microbial

community structure. In: V.S. Kennedy (Ed.) Estuarine perspectives,

Academic Press, N.Y., pp. 163-171. -305-

Whiting, M. C., in progress. Distributional patterns and taxonomic structure

of diatom assemblages in Netarts Bay, Oregon. Ph.D. Thesis, Oregon State

Univ., Corvallis.

Whitney, D. E. and Darley, W. M. 1979. A method for the determination of

chlorophyll a in samples containing degradation products. Limnol.

Oceanogr. 24:183-186.

Wiegert, R. C., R. R. Christian, J. L. Gallagher, J. R. Hall, R. D. H. Jones,

and Wetzel, R. L. 1975. A preliminary ecosystem model of coastal Georgia

Spartina marsh. In: L. E. Cronin (Ed.) Estuarine Research, Vol. I.

Academic Press, New York. pp. 583-601.

Williams, J. E. 1959. A quantitative study of the Zostera marina populations

of the York River, Virginia. Nat. Sci. Found. Program, Virginia Fish.

Lab., 15 June 1959 to Sept. 1959.

Williams, R. B. 1964. Division rates of salt marsh diatoms in relation to

salinity and cell size. Ecology 45:877-880.

Wolff, W. J., Haperen, A. N. N. van, Sandee, A. J. J., Baptist, H. J. M., and

Saeijs, H. L. F. 1976. The trophic role of birds in the Crenvelingen

Estuary, The Netherlands, as compared to their role in the saline Lake

Grevelingen. In: G. Persoone and E. Jaspers (Eds.) 10th. European

symposium on marine biology Vol. 2, Universa Press, Wetteren, pp. 673-689.

Wood, E. F., Odum, W. E., and Zieman, J. C. 1969. Influence of seagrasses on

the productivity of coastal lagoons. In: Coastal Lagoons, A Symposium,

UNAM-UNESCO. Univ. National Autononia Mexico, Mexico, D.F. pp. 495-502.

Woodin, S. A. 1977. Algal "gardening" behavior by nereid polychaetes:

effects on soft-bottom community structure. Mar. Biol. 44:39-42. -306-

Wuiff, B. L., and Mclntire, C. D. 1972. Laboratory studies of assemblages of

attached estuarine diatoms. Lirnnol. Oceanogr. 17:200-214.

Zapata, 0., McMillan, C. 1979. Phenolic acids in seagrasses. Aquat. Bot.

7:307-317.

Zedler, J. B. 1980. Algal mat productivity: comparisons in a salt marsh.

Estuaries 3:122-131.

Zieman, J. C. 1974. Methods for the study of the growth and production of

turtle grass, Thalassia testudinum Konig. Aquaculture 4:139-143.

Zieman, J. C., and Wetzel, R. G. 1980. Productivity in seagrasses: Methods

and rates. In: R. C. Phillips and C. P. McRoy (Eds.). Handbook of

Seagrass Biology. Garland STPM Press, New York. pp. 87-115. -307-

Appendix I. Summary tables for the analysis of variance associated with the concentration of organic matter in the top cm of sediment (AFDW), the concentration of chlorophyll a in the top cm of sediment (CHLS), the ratio of chlorophyll a concentration in the top cm of sediment to that of the 4-5 cm depth (RATIO), gross primary production (GPP), community oxygen uptake (0UPTK), the ratio of gross primary production to community oxygen uptake (GPP/CR), and the ratio of gross primary production to the chlorophyll a concentration in the top cm of sediment (ASSIM). T level refers to tidal level and Sediment refers to Intensive study site (SAND, FINE SAND and SILT).

Mean Variable Source of Variation DF Square

AFDW Main Effects: 15 90787.815 17.67 .001 Sediment 2 2010000.000 391.23 .001 T Level 3 33000.744 6.42 .005 Time 10 20897.522 4.07 .001

2-Way Interactions: Sediment by T Level 6 66479222 12.94 .001 Sediment by Time 20 4464.136 0.87 n.s. T Level by Time 29 8192 .516 1.59 .10

Error 53 5137.643

CHLS Main Effects: 16 50412.211 5.99 .001 Sediment 2 153618.336 18.26 .001 T Level 3 44862.896 5.33 .005 Time 11 32615.977 3.88 .001

2-Way Interactions Sediment by T Level 6 99244.404 11.80 .001 Sediment by Time 22 9656.430 1.15 n.s. T Level by Time 31 7815.200 0.93 n.s.

Error 57 8413.847 -308-

Appendix I (continued)

Mean Variable Source of Variation DF Square

RATIO Main Effects: 15 46467.122 5.13 Sediment 2 92638.673 10.22 T Level 3 130603.101 14.41 Time 10 15117.787 1.67

2-Way Interactions: Sediment by T Level 6 26368.678 1.77 Sediment by Time 20 7625.724 2.91 T Level by Time 29 19719.394 0.84

Error 53 9065.748

GPP Main Effects: 11 9734.859 2.22 Sediment 2 11198.406 2.55 T Level 2 1503.358 0.34 Time 7 11141.728 2.54

2-Way Interactions: Sediment by T Level 4 3234.046 0.74 Sediment by Time 14 3347.967 0.76 T Level by Time 13 3114.239 0.71

Error 23 4388.343

OUPTK Main Effects: 11 1223.388 3.81 Sediment 2 2433.902 7.59 T Level 2 204.966 0.64 Time 7 1093.928 3.41

2-Way Interactions: Sediment by T Level 4 959.737 2.99 Sediment by Time 14 406.282 1.27 T Level by Time 13 403.892 1.26

Error 23 320.851 -309--

Appendix I (continued)

Mean Variable Source of Variation DF Square

ASSIM Main Effects: 11 6.055 1.61 n.s. Sediment 2 6.794 1.81 n.s. T Level 2 0.583 0.16 n.s. Time 7 7.076 1.88 n.s.

2-Way Interactions: Sediment by T Level 4 0.677 0.18 n.s. Sediment by Time 14 3.803 1.01 n.s. T Level by Time 13 2.801 0.75 n.s.

Error 23 3.759