FWS/OBS-82/32 October 1983

lOlC C:c·,;.s,~ :,~c«,;

THE ECOLOGY OF TIDAL MARSHES OF THE PACIFIC NORTHWEST COAST: A COMMUNITY PROFILE

by

Denise M. Seliskar John L. Gallagher College of Marine Studies University of Delaware Lewes, DL 19958

Project Officers Wiley M. Kitchens, John Cooper, and Penny Herring National Coastal Ecosystems Team U.S. Fish and Wildlife Service 1010 Gause Boulevard Slidell, LA 70458

Perfonned for National Coastal Ecosystems Team Division of Biological Services Fish and Wildlife Service U.S. Department of the Interior Washington, DC 20240 Library of Congress Number: LC 83-600580

This report should be cited as follows:

Sel iskar, O. M., and J. L. Gallagher. 1983. The ecology of tidal marshes of the Pacific ft>rthwest coast: a community profile. U.S. Fish and Wildlife Service, Division of Biological Services, Washington, D.C. FWS/OBS-82/32. 65 pp. PREFACE

This tidal marsh community profile is chemical environment in which the organ­ part of a series of publications concern­ isms live. The third chapter treats the ing coastal habitats. The purpose of this biotic communities of tidal marshes. In profile is to describe the structure and the Pacific l'brthwest, there has histori­ ecological functions of tidal marshes of cally been more emphasis placed on struc­ the Pacific l'brthwest coast. This habitat tural aspects of the pl ant community than is ciassified by Cowardin et al. (1979) as on the functional roles for two reasons. occurring in the estuarine system, inter­ First, the diversity and zonation in the tidal subsystem, emergent v1etland class, Oregon and Washington Marshes are so and persistent subclass with a regularly striking that even ecologists oriented to­ or irregularly flooded water regime. The ward mineral cycling and energy fl ow can­ water chemistry is mixohal ine and the not help but think of indices of similar­ soils are mineral or organic. ity, line-point transects, and cluster analyses. Second, and perhaps coincident­ This profile is a synthesis of scien­ ally, most of the botanists workin'g in the tific infonnation and 1 iterature available coastal zone happen to be interested in on various aspects of tidal marsh ecology. cominuni ty structure. There has been much research on some topics and in these areas we will discuss The fourth chapter of the profile representative data. Little infonnation focuses on ecological interactions within specifically collected in the geographic tidal marshes and between marshes and reg ion is available on other topics. In adjacent systems. There is a paucity of the latter cases we have 1 ooked to other this type of infonnation from the wetlands regions where conditions are similar and of Oregon and Washington. Most of the have suggested probable conclusions when research on these topics in the region is future research on that topic is conducted from Canada and describes work relating to in the Pacific l'brthwest. salmon in British Columbia. Much of the local infonnation for this section was The coastal marshes of the region dot drawn from our experiences in the marshes the rocky coast. As seen on the cover of of Oregon and our discussions with others this publication, narrow inlets from the working there. Many of the ideas are sea open to protected bays where ma rs hes speculative in nature, and we have care­ fonn the ecotone between upland and the fully tried to point out when the data are estuarine habitat. Marsh plants probe the inadequate to make definitive statements. emerging mudflats for a suitable physical and chemical envi ronr.ient. On the upland Chapter 5 is a discuss ion of 11anage­ fringe, where soil aeration increases and r:ient practices. It examines both the his­ salinity drops, these pl ants lose their torical approach and future management competitive advantage to the upland plants. efforts built on the current understanding This profile is focused on these estuarine of the role of these discontinuous wet­ ecotones. lands in the coastal zone of the Pacific l'brthwest. The final chapter sur.inari zes The first two chapters of the text the major points about the Pacific l'brth­ discuss the development of Pacific l'brth­ west tidal marshes and compares them to west coastal wetlands and the physical and those at other sites.

i i i --

The authority used for plant names all other cases. in this profile is Hitchcock and Conquist ( 19 73}. The authority used for bi rd names Any questions, comments about, or is American OrAithologists 1 Union (AOU requests for this publication should be 1982). addressed to: The metric sys tern is used throughout this report except in those cases where Infonnation Transfer Specialist the convention of the field is to use Eng- National Coastal Ecosystems Team 1 ish units (i.e., soil horizons in inches, U.S. Fish and Wildlife Service fish weights in pounds). For ease in com­ W\SA-Sl idel l Computer Complex parison, English equivalents are given for 1010 Gause Boulevard tenperature, area, and tidal heights. A Slidell, Louisiana 70458 table of conversion values is provided for

iv CONTENTS

PREFACE • • • • iii FIGURES •••• vi TABLES • • • • vii ACKNOWLEDGMENTS • • • • viii CONVERSION FACTOR TABLE ix CHAPTER 1 PHYSICAL AND CHEMICAL ENVIRO!f.1ENT •••• 1 1.1 Regional Climate and Meteorology •• 1 1. 2 Tidal Regimes and Water Relations •••• 2 1. 2.1 The Estuaries ••••••••• 3 1. 2. 2 River Systems and Runoff ••• 4 1.2.3 Tidal Creeks ••••••••• 6 1. 3 Soils - Salinity and Fertility • • ••• 6

CHAPTER 2 MARSH DISTRIBUTION • • • • •• 10 2.1 Distribution of Tidal Marshes 10 2.2 Types of Ma rs hes • • • • • • • 11 CHAPTER 3 BIOTIC COMMUNITIES ••••••••••••••• 15 3.1 Vegetation - Zonation and Species Distribution • 15 3.2 Microbes • • • • • 22 3.3 Faunal Components 25 3.3.1 Invertebrates 25 3.3.2 Fi shes • 30 3.3.3 Birds 32 3.3.4 Mammals • 34 CHAPTER 4 ECOLOGICAL INTERACTIONS •••••• 37 4.1 Ecological Processes Within the Marsh 37 4.1.1 Primary Production • • • • • • • ••• 37 4.1.2 Decomposition • • • • • • • • • ••• 41 4.1. 3 Grazing • • ••• 42 4.1.4 Food Webs • • • • 42 4.1. 5 Uses of the Marsh • • • • • • 43 4.2 Mineral Cycling in the Marsh •••••.•••••.• 47 4.3 Interactions Between the Marsh and Adjacent Ecosystems • 48 CHAPTER 5 MANll.GEMENT • • • • • • • • • • • • • • . • • • • • 50 5.1 Sensitivity to Natural and Cultural Perturbations 50 5.2 His tori ca 1 Management Approach • • • • • • . • • • 51 5.3 Upland Marsh Boundaries ••.•••••• 54 5.4 Current Management Practices and Future Prospects 56 CHAPTER 6 SUMMARY AND COMPARISONS 58

REFERENCES 60

v FIGURES

Number 1 An annual precipitation pattern for coastal Oregon at Tillamook in 19 79 • . • • • . • . . . • . . • . . • • . • • . . 2 2 Typical daily tidal curve for Astoria, Oregon ••••.•••• 3 3 High tides at Netarts Bay, Oregon •••.••.•.•••••••• 4 4 Stream patterns in the marsh at the head of Netarts Bay, Oregon 6 5 Marsh soil moisture tension during the growing season at Netarts Bay, Oregon, at a depth of 10 cm • • • • . • • • • • • • • • • 9 6 Location of estuaries in Oregon and Washington •••••••••••• 11 7 Zone of accretion at the marsh-mudflat interface •••••.• 12 8 Pickleweed as a pioneer on the tidal flat in a low sandy marsh 20 9 Seaside plantain growing in a low sandy marsh ••••••••• 20 10 Circular clumps of arrow-grass in a low silty marsh •••••• 21 11 Lyngbye' s sedge marsh in a brackish estuary •••••••••• 21 12 Pacific sil verweed fran a mature high marsh •••••••••• 22 13 Typical zonation of marsh plants in a saline Pacific Northwest ti da 1 rna rs h • • • • • • • • • • • • • • . . • • • • • • • • • • 23 14 Possible zonation pattern in a brackish Pacific Northwest t1dal marsh .•.••...... •..•..•....• 23 15 Examples of the range of variations in vegetational zonation in Pacific tbrthwest tidal marshes •••••••••••••••••••• 24 16 Drift line in an Oregon low sandy marsh ••••••••••••• 26 17 Conceptual model of carbon flow between components ·of tidal marsh ecosystems in the Pacific Northwest •••••••••••••••• 38 18 Conceptual model of energy or material fl ow between tidal marshes and other coastal ecosystems in the Pacific Northwest •••••• 49 19 Lyngbye's sedge marsh invading a mudflat created by a log blocking a marsh creek • . . • • . • • . . . • • . . . • . . . . • . • . • . 51 20 Diked and undiked marshlands which are being used for pasture ••• 52 21 Transition from wetland to upland ••••••••••••••••• 55

vi TABLES Number 1 Classification of estuaries of the Pacific Northwest, drainage basin area, and mean annual discharge rates • • • • • • • • • • • . • • 5 2 Description of a typical Sulfihemist (Histosol) and a typical Sulfaquent ( Enti sol) from coastal marshes • • • • • • • • • • • • • • • 8 3 Areal extent of salt marshes associated with Oregon estuaries • . 11 4 Characteristics of the eight types of marshes described for Oregon by Akins and Jefferson (1973) . . . . • • . • . • . • . . • . • . . • . 13 5 Eight Oregon marsh community types and their typical plant species composition ...... 16 6 Common plant species found in Pacific Northwest tidal marshes • • . 17 7 Invertebrates characteristic of five Oregon marsh habitats. • • . • 27 8 Approximate densities and percentages of invertebrate taxa from five habitats of an Oregon marsh . • • • • • • • • • • • • • 29 9 Common and scientific names, marsh habitat, and life stage of some fishes associated with marshes of the Pacific Northwest • • • 31 10 Bird observations in salt marshes at Coos Bay . • • • • • • • • • • • • 33 11 Common and scientific names and relative abundance of birds associated with marshes of Coos Bay • • • • • • • • • • • • • • • • • • 35 12 Common and scientific names of some mammal species associated with the marshes of the Pacific Northwest •.••.••.••••••• ~ • 36 13 Representative processes of carbon flow in salt marsh communities • • • 39 14 Primary production estimates of selected Oregon marsh pl ants 41 15 Relative importance of prey items of the dominant fishes of Woodward Island marsh channels • • • • • • • • • • . • • • • 45 16 Landfills on lands lying between the line of ordinary high water and the line of ordinary low water in selected Oregon estuaries • • 53

vii ACKNOWLEDGMENTS

We are indebted to all those who have­ printed by Pam Palinski and Lewis Nelson. worked on the marshlands of the Pacific Pam Palinski and the NCET graphics group Northwest to develop the infonnati on we prepared the drawings. Rae Clark typed have drawn together. We thank all those the various drafts of the manuscript with who reviewed and improved the manuscript, speed and good cheer. especially Wiley Kitchens, John Cooper, and Penny Herrin9 of the National Coastal This report is dedicated to the mar­ Ecosystems Team (NCET) staff. riage of the authors which has survived the stresses of producing this document. The photographs were developed and

viii CONVERSION FACTORS

Metric to U.S. Customary Mu 1 t!.PJ.1. h To Obtain r.1il 1 imeters (mm) 0.03937 inches centir:ieters (cm) 0.3937 inches meters (m) 3.281 feet kilometers (km) 0.6214 mil es 2 square meters (m ) 2 10.76 square feet square kilometers (km ) 0.3861 square mil es hectares (ha) 2.471 acres liters (1) 3 0.2642 gal 1ons cubic meters (m 3) 35.31 cubic feet cubic meters (m ) 0.0008110 acre-feet milligrams (mg) 0.00003527 ounces grams (g) o. 03527 ounces kilograms (kg) 2.205 pounds metric tons (mt) 2205.0 pounds metric tons (mt) 1.102 short tons kilocalories (kcal) 3.968 BTU

0 Celsius degrees 1.8(C ) + 32 Fahrenheit degrees

U.S. Customary to Metric inches 25.40 mil 1 imeters inches 2.54 centimeters feet (ft) 0.3048 meters fathoms 1.829 meters miles (mi) 1.609 kilometers nautical miles (nmi) 1.852 kilometers 2 square feet (ft ) 0.0929 square meters acres 2 0.4047 hectares square mil es (mi ) 2.590 square kilometers gallons (gal) 3 3.785 1 i ters cubic feet (ft ) 0.02831 cubic meters acre-feet 1233.0 cubic meters ounces (oz) 28.35 grams pounds (1 b) 0.4536 kilograms short tons (ton) 0.9072 metric tons BTU 0.2520 kilocalories

0 Fahrenheit degrees 0.5556(F - 32) Celsius degrees

ix

CHAPTER 1 PHYSICAL AND CHEMICAL ENVIROfll.1ENT

The boundaries chosen for this pro­ During the winter the Aleutian Low, file on the tidal marshes of the Pacific one of the major pressure patterns con­ tbrthwest are the southern border of Ore­ trolling the Pacific Northwest climate, gon and the northern of Washington. Most expands southward over the Gulf of Al a ska of the coast (except Puget Sound) is rela­ and the Bering Sea from its usual position tively unfeatured, as is all the Pacific over southwest Alaska. Counterclockwise coast of the United States. Although the airflow around the low results in winds Pacific coast constitutes 22.5% of the from the west and southwest. Frontal general coastline of the country, not storms move eastward into the Pacific including Alaska, it contains only 14.4% Northwest, resulting in high precipitation of the detailed shoreline (U.S. Geological during the winter months. Average annual Survey 1970). The Pacific Northwest com­ precipitation along the coast is 180 to prises 7. 9% of the genera 1 Pacific coast. 200 cm (70 to 80 inches) per year with 80% The picturesque coast of the region has of this occurring between October and developed as the result of geological March. Prolonged periods of light to activity which has produced rocky head- moderate rain are typical. Snowfall is 1 ands, broad sand dune complexes, sandy rare along the coast. Dense cloud cover pocket beaches, and offshore rocks. Estu­ is common especially during the winter; aries project inland from the coast where 80% to 100% cloud cover is reported for major rivers enter the sea or where embay­ half the days of the year. Winds from the ments are cut off by headlands or baymouth south to southwest keep temperatures mild sandbars. The Coast Range was fanned from by bringing wanner air from the ocean. marine sediments and serves to modify the Maximum winter temperatures average 3° to coastal climate as does the California 7° C (38° to 45° F) along the Washington Current. The result is rather unifonn coast and in the 1 ow teens along the more coastal temperatures throughout the year. southern Oregon coast while minimum tem­ peratures range from -2° to 2° C (28° to 36° F) along the Washington coast and near 1.1 REGIONAL CLIMATE AND METEOROLOGY 5° C (41° F) for the Oregon coast (Oceano­ graphic Institute of Washington 1977). The Pacific tbrthwest coast has a The major pressure pattern control- mild, mid-latitude west coast marine cli­ 1 ing ~brthwest climate during the summer mate. The maritime influence of the is the North Pacific High which moves from Pacific Ocean provides moderate weather, its winter location off the southern Cali­ wanning the area in the winter and cooling fornia coast to off northern California it in the sumr.ler. Al so significant in during the summer. Clockwise airflow maintaining mild conditions is the protec­ around the high results in northwest tive barrier provided by the Cascade Moun­ oceanic winds which keep summer tempera­ tains which guards the coastal area tures mild. Coastal upwelling keeps near­ including the Olympic Mountains and the shore waters cool and fog often results Coast Range against the cold winter and where the cold water meets wanner offshore hot summer continental air masses to the waters. Along the Washington coast maxi­ east. mum temperatures range from 17° to 19° C (63° to 66° F); minimum temperatures aver­ species. This would be particularly true age 10° C (50° F). Precipitation during of plants, such as sal tgrass (Distichl is the summer months is light with o~y 20% spicata), which have the C4 type of uetab­ of the annual total occurring between ol ism and can effectively utilize very April and September (Oceanographic Insti­ high light intensity. Other plants such tute of Washington 1977). A typical annu­ as Pacific sil verweed (Potentil 1 a paci­ al precipitation pattern is that seen for fica) have the C3 type of 11etabolism, and Tillamook, Oregon, in 1979 (Figure 1). their photosynthetic processes are 1 ight saturated at much lower light intensities. The cl imate results in rather benign growth conditions for marsh plants and animals. Hard freezes are not common and 1.2 TIDAL REGIMES AND HATER RELATIONS ice scouring, which may be important in the development and detritus cycling of The tide along the coast of the Paci­ New England marshes, is extremely rare. fic ~brthwest consists of semidiurnal and Mild summer temperatures combine with oc­ diurnal components. The semidiurnal por­ casional rainfall to keep the interstitial tion occurs approximately every 12 hours salinity moderate when the lower high while the diurnal component occurs at tides occur. In Mid-Atlantic, Southeast, about 25-hour intervals. These phenomena Gulf coast, and southern California result in two high tides of unequal ampl i­ marshes, evaporation results in hypersa- tude per day: one higher high and one 1 ine conditions which lead to reduced lower high (Oceanographic Institute of pl ant growth. In the Pacific ~brthwest Washington 1977). Similarly, two unequal the fog and the high frequency of cloud lows occur each day. Figure 2 illustrates cover reduce 1 ight to the point ~~here a daily tidal height cycle for Astoria, photosynthesis may be limited for some Oregon. Figure 3 il 1 ustrates an annual 25 .,

' 20 "'i

(/) -Q) ..c u c :..:::;. 1.5,• z ,. 0 I' r-tr 1.0l..J ' 0... T u w 0:: , l ~ 0... 0.5,- 4 • ~ ~ I p ~ !> ' 11 j ) j ~ t~~ ~ ;' •'~ ~~ ~-r ~ ·1 T T l " 1 0 Jan FP.h Mar Aor Mov Jun Jul Aug Seo Oct Nov Dec Figure 1. An annual precipitation pattern for coastal Oregon at Tillamook in 1979 (from Seliskar 1981). Although rainfall occurs throughout the year, it is concen­ trated during periods when the Aleutian low moves southward. 2 10 allowing deeper, nutrient-rich (due to decomposition processes which occur at depth) water to rise to the surface. Upwelling is most prominent along the southern Oregon and northern California coast. Further northward it is masked on the surface by the Columbia River plume and fl ow from the Strait of Juan de Fu ca. t­ The Columbia River plume produces its own ::c ings are important in increasing productivity of the adjacent estuaries; thus nutrients from coastal 0 upwelling eventually bathe the tidal marshes as well (Proctor et al. 1980). ·2-t----.----.--r----.---...-...----.---.--r---.---...... 1.2.1 The Estuaries 0000 0400 0800 1200 1600 2000 2400 TIME Tidal marshes are subjected to vari­ Figure 2. Typical daily tidal curve for ous salinity and sediment regimes, depend­ Astoria, Oregon (U.S. Department of Com­ ing on the type of estuary they border. merce 1979). There are four types of Pacific l'brthwest estuaries: bar-built, bl ind, drowned­ river, and fjord. Bar-built estuaries are regime of high tides at Netarts Bay, Ore­ fonned by the accumulation of sand along gon. The diurnal range in tidal amplitude bars which, when connected to shore, varies from zero at the upriver portion of restrict water flow to coastal embayments. tidal influence to more than 3 meters (10 Netarts Bay, Sand Lake, Grays Harbor, and ft) at certain stations in some estuaries Will apa Bay are examples of bar-built es­ (U.S. Department of Commerce 1979). tuaries (Bottom et al. 1979). The higher tides occur during the The second type of estuary, the bl ind fall and winter and result in wetter soils estuary, develops from a bar-built type and more frequent input of salts to the when freshwater or tidal flow is low, usu­ intertidal substrates. The combination of ally during the summer, and beach sedi­ changes in atmospheric evaporative power, ments close the mouth of the river. Elk rainfal 1, and tidal inundation result in River, Pistol River, Sixes River, and the salinities in the lower elevation marsh Winchuck River in Oregon are examples being highest from July to December (Gal­ where such closures are common (Bottom et lagher and Kibby, unpublished data). In al • 1979). the higher elevation plant stands, salin­ ity was low and more unifonn throughout The third type of estuary is the the year. Saturation of the soils with drowned-river valley. These estuaries tide and rainwater is most common during were fonned upon the rising of sea 1 evel the winter. The resultant reduction in at the end of the 1 as t ice age and the soil oxygen comes at a time when the subsequent flooding of the former river pl ants are donnant so adverse effects are valleys. Coos Bay, Yaqu ina Bay, Nehal er.i minimal. Bay, and Siletz Bay exemplify this estuary type (Bottom et al. 1979). The Columbia Upwel 1 ing is an important feature of River Estuary, however, is associated with waters along the Pacific l'brthwest coast. a river that has a clean, stream-cut chan­ During the summer, preva i1 ing winds from nel rather than a drowned-river valley the north push surface water offshore (Duxbury 1971). Some estuaries such as

3 10

9

8 (]) 6 ...0 <[ 7 -+- (]) Lf 6

5------...-----~._.------i-o----.-----..-~...... -----. Jan Feb Mar Apr ...... May------Jun...... ---- Jul ...... Aug Sep Oct Nov Dec Figure 3. High tides at Netarts Bay, Oregon (from Sel iskar 1981). Highest high tides occur fall and winter.

Tillamook Bay show both bar-built and source of freshwater entering the Pacific drowned-river characteristics. Ocean from Oregon and Washington, runoff peaks in the late because of spring snow­ Puget Sound, a collection of estu­ mel t. During the summer the freshwater aries of several types, is the most plume moves southwest while in the winter complex of the Pacific Northwest systems. it moves northward (Proctor et al. 1980). Elements of the Sound are of the fjord Drainage basin areas and mean annual dis­ type carved out by glaciers (Duxbury charge rates for estuaries are included in 1971). Tab.le 1 lists the Pacific Table 1. Northwest estuaries and their classifi­ cation. The amount of freshwater entering the estuarine system is significant to the 1.2.2 River Systems and Runoff tidal marshes fringing the estuaries and will influence the species present and the The freshwater runoff patterns into net primary productivity. For example, the estuaries of the Pacific f'hrthwest, marsh flora and fauna in the Will apa Bay excluding the Columbia River Estuary, are and Sand Lake marshes are subjected to dependent upon precipitation patterns. high salinities due to their small drain­ Therefore, runoff is high during the age basins and low freshwater input. In rainy, winter season, while in the summer estuaries with larger drainage areas .such it is significantly 1ess. In the case of as the Columbia River, there is greater the Columbia River, which is the 1argest dilution of the salt water. An example of

4 Table 1. Classification of estuaries of the Pacific Northwest, drainage basin area, and mean annual discharge rates. (Modified from Roden 1967, cited by Proctor et al. 1980; Wilsey and Ham 1974, cited by Bottom et al. 1979; and Percy et al. 1974).

Mean annual Estuary Drainage basin discharge rates Estuaries classification area (mi 2 ) (106 acre ft/yr)

WASH I t£TON a Puget Sound Fjord Quillayute River Drowned-river 382.0 1.59 Hoh River Drowned-river 252.8 1.82 Queets River Drowned-river 443.8 2.99 Quinaul t River Drowned-river 264.0 2.02 Grays Harbor Bar-built 1) Humptul ips River 130.0 0.95 2) Chehalis River 1,813.8 5.63 Willapa Bay Bar-built 1) f'brth River 218.8 0.69 2) Willapa River 157.8 0.61 3) Na sell e River 89.1 0.38 OREGON Columbia River Stream-cut 255,406.2 184. 69 Necanicum River Drowned-river 87.0 Nehalem Bay Drowned-river 667.6 1.97 Tillamook Bay Drm-1ned-ri ver 540.0 Netarts Bay Bar-built 14.0 Sand Lake Bar-built 17.0 Nes tu cca Bay Drowned-river 91.1 o. 72 Salmon River Drowned-river 75.0 Siletz Bay Drowned-river 202.2 1.15 Yaquina Bay Drowned-river 253.0 Al sea Bay Drowned-river 356.6 1.30 Siuslaw River Drowned-river 340.4 0.64 Umpqua River Drowned-river 3,897. 6 5.83 Coos Bay Drowned-river 605.0 Coquille River Drowned-river 756.4 1. 79 Sixes River Bl ind 129.0 Elk River Bl ind 94.0 Rogue River Drowned-river 4,939.6 7. 72 Pistol River Bl ind 106.0 Che tco River Drowned-river 359.0

a·The symbol (-)indicates no reported values.

5 the seasonal increase in interstitial soil salinity can be seen in Netarts Bay, an NETARTS estuary with liftle freshwater input in BAY the summer. In eight pl ant stands the salinity rose from an average of 10 parts per thousand (ppt) in April to 22 ppt in July. The change was not uniform over the marsh; little change was noticed at the upper fringe while at lower sites salinity more than doubled (Gal 1 agher and Kibby in prep.).

1.2.3 Tidal Creeks Estuarine water moves in and out of the salt marshes by way of tidal creeks. According to Eilers (1975), most of the UPLAND creek systems are of the dendritic type - a major channel with minor branches and sub-branches. In the Nehalem Bay marsh, creeks do not often extend into the marsh area above 2.8 m above mean 1ower low water (MLLW). The density of marsh creeks is related to elevation, marsh type, and Figure 4. Stream patterns in the marsh at dra·inage (Eilers 1975). Generally, creek the head of Netarts Bay, Oregon. The development is most extensive in middle­ dashed 1 ine indicates the position of an aged marshes. Due to denser stabilizing old dike. vegetation in the upper portions of the marsh creek channel, meandering occurs only in lower marsh areas. An old dike is indicated by the dot­ ted 1 ine in Figure 4, and a number of Figure 4 depicts the stream pattern brackish water pools are seen in the upper in the marsh at the head of Neta rts Bay, reaches of the marsh. Such dikes were Oregon. These extensive channel systems pl aced on many marshes to make them dry provide access for aquatic organisms which enough for pasturing. Tide gates allowed may be using the marsh as a nursery ground creeks to drain precipitation and upland and/or feeding site. The stream systems runoff but blocked incoming tidal water. carry detritus from its site of production to the organisms in deeper water. They al so bring tidal subsidy to areas of the 1.3 SOILS - SALINITY AND FERTILITY marsh away from the main estuarine chan­ nel. Al though the tenn is incompletely The physical and chemical environ­ defined and not yet quantified, Cktum ments of Pacific Northwest marsh soils (1980) proposes that tidal energy subsi­ appear to be similar to those in coastal dizes the coastal marshes. That subsidy wetlands at similar latitudes on the may take the fonn of removing toxic organ­ Atlantic coast. Soils range from sand or ic materials, carrying oxygen and nutri­ silt to peat, depending on the age of the ents to the root zone, diluting salt near marsh and the type and amount of sediment the roots, or other undefined work. The load in the adjacent rivers. Their sal in­ sedge in Oregon r:iarshes is more vigorous ities likewise depend on the nature of the along the streambanks than is the back­ estuarine system in which they develop and marsh, a phenomenon which may be associ­ the position of their sites within the ated with tidal subsidy (Gallagher and estuary. The systems range from those Kibby 1981). dominated by freshwater from large rivers

6 sue~ as the .Col.umbia to those with pri­ lower elevation marshes. Seliskar (1981) marily oceanic influence. l't.ltrient input measured soil moisture along transects in to marsh soils from the ocean is 1 ikely the salt marsh fringing Netarts Bay, Ore­ greatest during seasonal coastal upwell­ gon, using pennanently implanted soil ing. Input from the land is greatest when moisture tensiometers. Soil moisture ten­ ~unoff .fr~ the nearby ~ills is high dur­ sion, a measure of dryness, is highest in ing ram in the fall, winter, and spring soils in the upper marsh (Figure 5). The and during snownelt in the Cascades in th~ large drops in tension closely coincide spring. with precipitation data for the area. High tensions develop between stonns and Extensive studies have not been made spring tides. of marsh soils in the Pacific rt>rthwest. Livennan (1982) excavated soil pits along Gal 1 agher (1980) described five fac­ transects from upland to tidal flat at tors which influence marsh soil salinity: Netarts Bay, Oregon. The soil was com­ (1) salinity of the flooding estuarine posed almost entirely of sand in the water; (2) tidal elevation; (3) the envi­ upland (dune) and tideflat. Between these rormental complex (including temperature, two areas 1 ies the tidal r.iarsh. In the pan evaporation, and rainfall); (4) soil middle marsh several soil horizons exist. texture (coarser soils are more easily The surface 0.5 m is a dark greyish-brown flushed); and (5) specific evapotranspira­ sandy loam. Below this is a layer of dark tion rate of the pl ant species. Soil grey clay loam approximately 0.25 m thick. salinity measured along transects at The next half meter consists of dark Netarts Bay, Oregon, decreased from lower greyish-brown sand beneath which is a dark to upper r.iarsh (Sel iskar 1981; Livennan grey-blue clay. These descriptions are 1982). The higher salinity at the lower ecological in nature but would be classi­ elevations is due to the frequent flooding fied as Entisols by pedologists. General­ by the es tua ri ne waters. As one moves ly marsh soils are classified in either landward, flooding of the marsh is less the order Entisol or Hi stosol. The pri­ frequent and rainwater dilutes the salts marily organic soils are the Histosol s, that a re present. Sa 1 in i ty may a 1 so and they are called Sulfihemists in coast­ decline with depth in the marsh. In some al areas where sulfides are produced from sandy r:iarshes low salinity ground water the reduction of seawater sulfates. Sul­ may surface at the interface of the marsh faquents are recently fonned mineral soils and mudflat. livennan (1982) found that (Entisol s) common to the salt marsh (Gal- in the marsh on the Netarts sand spit a 1 agher 1980). Table 2 illustrates a typi­ 1 ens-shaped freshwater body fl oats on top cal profile for a Histosol and an Entisol of a convex salt groundwater body. The from the marsh. Although the profiles freshwater lens recharges with winter pre­ were described from Atlantic coast cipitation, but during the summer its marshes, spot checks of characteristics in level approaches the surface of the salt Pacific Northwest marsh soils indicate groundwater. The point of freshwater dis­ that the profiles are similar. There is charge onto the marsh depends on the depth certainly no evidence that soils develop­ of the salt groundwater surface. ing in similar latitudes with similar types of flooding and salinity regir.1es The supply of nutrients provided by a would be different on the two coasts. soil depends, for the most part, upon the Details of detenninations of pH, col or, texture of the soil. The coarser the sub­ texture, etc., can be found in Breeding et strate, the lower the nutrients. Finer al. (1974), Darmody and Foss ( 1978), and textured salt marsh soils have the highest Gallagher (1980). nutrient content due to their greater cation exchange capacity (Gallagher 1980). Moisture content in salt marsh soils Sel iskar (1981) found that potassium (K}. is dependent upon precipitation and fre­ calcium (Ca), magnesium (Mg), and total quency of submergence, the 1 atter of nitrogen (total N) were much lower in a which, of course, is related to elevation. transect which consisted totally of sand Saturated conditions generally exist in than in any of the other more s i1ty tran-

7 fable 2. Description of a typical Sulfihemist (Histosol) and a typical Sul faquent (Entisol) from coastal h1arshes. These descriptions were written from Atlantic coast marshes, but checks of salinity, boundaries, pH texture, and structure on Pacific North­ west marsh profiles indicate similar soil descriptions would apply to them. Color designations were obtained using the Munsell soil color chart.

Depth Structure Salinity Horizon (inches) Color Texture consistency pH Boundary (ppt)

TYPIC SULFIHEMIST

Oi 1 0-8 10 YR 3/2 Organic, Massive, 6.6 Abrupt, 29. 3 5% mineral non-sticky smooth

Oe 1 0-15 10 YR 2/1 Drganic, Massive, 6.8 Abrupt, 27.1 10% slightly smooth mineral sticky

Oe 2 15-31 5 YR 3/2 Organic, Massive, 7.0 Abrupt, 27.2 20% slightly smooth mineral sticky

Oe 3 31-42 5 YR 2/2 Organic, Massive, 6.8. Gradual, 25.6 10% slightly smooth mineral sticky

Oe 4 42-63 5 YR 2/2 Organic, Massive, 7.0 no 25.4 15% slightly boundary mineral sticky TYPIC SULFAQUENT

A 11 0-8 10 YR 3/1 Clay 1oam Weak, sub­ 6.9 Gradual, 33 angul ar wavy blocky, sticky

A 12g 8-19 10 YR 3/1 Clay loam Massive, 7 .o Cl ear, 36 sticky wavy c lg 19-33 10 YR 4/1 Clay Massive, 6.9 Gradual, 38 sticky wavy

c 2g 33-50 5 GY 5/1 Clay Massive~ 6.8 Gradual, 39 sticky wavy

c 3g 50-60 5 GY 5/1 Silty clay Massive, 7.2 36 sticky

8 75 sition, compounds such as sulfides, organ­ "' ic acids, and al dehydes which are often I "' I toxic to some organisms may accumulate • II• I I I 65 I I I Sedimentation is important in a num­ I I I I ber of ways in the marsh. Where sea 1 evel I I is rising relative to the land, it pro­ 55 I I I I vides a mechanism for the plants to main­ I I I I I II tain their position relative to submerg­ I ELEVATION I I I I I I (m above MHW) ence. Further, it provides a mechanism 45 I I 11" I I I I for elevating mudflats to a height which I -1.01 I I I ..,.. __ ...... 0.76 I I I I will support marsh plant growth. Using I I I I ...... 0.18 (/) I I I maps and aerial photographs, Johannessen I I 0:: 35 I I I cC I I {1961) calculated the rate of lateral ad­ m I I I I I ~ I /. I vance of tidal marsh in the Coquille Estu­ I I I z I I LI.I I I I ary between 1887 and 1939 and found it to 0 25 I I I I be approximately 70 ft/yr. The rate has I I I I I I I I decreased since then to about 5 ft/yr. I I I I I I I I Between 1940 and 1960, the amount of marsh I I I I 15 I I I I I doub 1 ed in Ken tuck Slough of Coos Bay. I I I I I I Little net change has occurred in the \I I I I I Al sea Estuary al though there are areas of I I 5 ~ advancement and retreatment. Since 1875, marsh advancement has ranged from 0 to 27 ft/yr at Nehalem Bay (Johannessen 1961).

Figure 5. Marsh soil moisture tension Jefferson (1975) reported accretion during the growing season at r.ietarts Bay, rates of 0.5 to 1.7 cm/year in low, silty Oregon, at a depth of 10 cm (modified Oregon marshes. Deposition is uneven fror.1 Sel iskar 1981). between marshes because sediment loads vary from one estuary to another and cur­ rent velocities vary due to plant density, sects. In the sandy site, K, Ca, Mg, and water volume, and area of spreading. total N values were approximately 200 Within a marsh, deposition is usually parts per mil 1 ion {ppm), 1.0 mil 1 iequiva- greatest along the stream banks where the 1 ents (meq)/100 g, 2.3 meq/100 g, and current velocity first slows. Observa­ 0.05%, respectively, while in the siltier tions by the authors in the Fraser Estuary areas, the values were approximately 2,000 in British Columbia indicated that almost ppm, 8.0 meq/100 g, 25.0 meq/100 g, and a 11 the deposition occurs during the 0. 5% for K, Ca, Mg, and total N, respec­ spring freshet. Deposits of 5 cm/yr were tively. common. Stumpf (1981) has measured deposition rates in a Delaware salt marsh Marsh soils which are submerged fre­ and found that most of the annual quently are often anaerobic below a depth deposition was associated with stonn of 3 an. Oxygen is rapidly removed by the events, not day-to-day inundation. ifo decomposition process. As a result, chem­ suspect a few events each year are al so icals in these soils generally occur in a resonsible for most deposition in the reduced state. Duri nq anaerobic decompo- Pacific Northwest marshes.

9 CHAPTER 2 MARSH DISTRIBUTION

Steep relief characterizes the expanding, but found no such expansion at terrestrial side of the division between the Al sea. Eilers (1975) measured progra­ 1 and and sea and extends below the waves. dation of the wetlands at Nehalem Bay at a The heavy wave action dominating the maximum rate of 2.4 m/year. A possible exposed shores of the Pacific Northwest reason for this accretion is increased coast produces an inhospitable site for riverine sedimentation due to agriculture marshland development. There are, how­ and logging practices and perhaps erosion ever, two types of protected havens along brought about by forest fires (Akins and this high-energy coastline where marshes Jefferson 1973). Figure 7 illustrates a develop. marsh-mudflat interface where accretion is elevating a tidal fl at and the marsh is The most common tidal marshland sites expanding seaward. are those fringing the borders of rivers which run more or less directly into the Even though considerable progradation Pacific Ocean (Figure 6). The Salmon has taken place in past years, the loss of River Estuary, for example, has extensive marsh habitat is still greater than its fringing marshes. Historically, this type gain. For example, in the last 100 years of marsh extended many miles up the 90% of the Coos Bay salt marsh has been rivers, but diking has converted many of destroyed for various purposes including the upper river tidal wetlands into pas­ agriculture, industry, and residences. tures. A second setting for wetlands is The City of Coos Bay, Oregon, is built on in the lee of energent bay-mouth bars what was once marsh (Hoffnagle and 01 son which enclose broad river valleys where 1974). Sixty percent of the marshland low energy locations necessary for the bordering Puget Sound in the early 1800 1 s development of marshlands occur. Grays has been lost to dredge and fill projects, Harbor and Willapa Bay, Washington, and jetties, and marinas. This does not Tillamook Bay, Oregon, are examples of the include land loss before the area was latter type. first mapped (Meyer 1979). Diking has stopped the natural functioning of more salt marsh than any other type of intru­ 2.1 DISTRIBUTION OF TIDAL MARSHES sion. Once it was found that such areas provided productive farmland, most areas Jefferson (1975) and Ei1 ers ( 1975) of high marsh in Oregon were diked measured areas of existing salt marsh for (Jefferson 1975). About 50% of original 14 Oregon estuaries from aerial photo­ wetlands of Willapa Bay, Washington, have graphs and maps (Table 3). Jefferson been diked or filled (U.S. Department of (1975) stated that in Oregon there was a the Interior 1970). total of 29 km2 of undiked salt marsh, excluding the Columbia River. In 1 ight of the importance of the salt marshes to coastal ecosystems, marsh­ Salt marshes in the Northwest have lands are being reclaimed from agricul­ been shown to be prograding; Johannessen tural use in some pl aces by breaking the (1964) found that the Nehalem Bay, Tilla­ dikes. Mitchell (1981) conducted research mook Bay, Umpqua River, Coos Bay, and on the natural establishment and succes­ Coquille River marshes 'of Oregon were sion of salt marsh pl ants into a previ-

10 Table 3. Areal extent of salt marshes associated with Oregon estuaries (from Jefferson 1975 and Eilers 1975).

Acres of salt marsh Jefferson Eilers (1975) (1975)

a Columbia River Necanicuiil River 10 30 Nehalem Bay 440 601 Til 1 amook Bay 902 898 Netarts Bay 161 273 Sand Lake 415 670 Nes tucca Bay 203 225 Salmon River 366 184 Siletz Bay 264 360 Yaquina Bay 657 1102 Alsea Bay 561 652 Siuslaw River 761 785 Umpqua River 689 805 Coos Bay 1435 2239 Coquille River 257 393 Sixes River Elk River Rogue River 12 Pistol River Chetco River 2

TOTAL 7135 9217

aThe symbol (-) indicates no value re- ported.

rapidly on these newly flooded areas. SIXf.S kl\'t:: r.u;, klVER Federal action has also been taken to pro­ tect wetland environments and is discussed in Chapter 5.

2. 2 TYPE OF MARSHES Akins a11d Jefferson (1973) describe eight types of tidal marshes in Oregon: (1) low sandy marshes, (2) low silty Figure 6. Location of estuaries in Oregon marshes, (3) sedge marshes, (4) immature and Washington (U.S. Geological Survey high marshes, (5) mature high marshes, 1970). (6) bulrush and sedge marshes, (7) inter­ tidal gravel marshes, and (8) diked salt ously diked area on the Salmon River marshes. These r1arsh types are the result Estuary. Marshes similar to the adjacent of the selection of marsh pl ant species natural systems seem to be developing which are best adapted to a particular

11 Figure 7. Zone of accretion at the marsh-mudflat interface. A patch of pickleweed is seen in the left foreground and tufts of arrowgrass are invading the mudflat in the cen­ ter of the photograph. combination of substrate, estuarine salin­ defined channels around clumps of plants, ity regime, and elevation. The charac­ which are discontinuous at the lower ele­ teristics of the eight marsh types are vations as in the low sandy marshes. summarized in Table 4. Sedge marshes al so form on silt and Low sandy marshes occur on sandy sub­ have a nearly level surface. They are strate with a gradual slope, typically on often found on isl and or delta edges with the 1ow-energy side of bay mouth sand elevations somewhat above the first two spits or as fringing marshes on isl ands marsh types. They are flooded by most with coarse textured sediments. They are high tides and drain via creeks in the flooded by nearly all high ti des and drain higher sedge marshes and diffusely in the diffusely (i.e., there are no tidal lower ones. Soil salinities are often low creeks) over the marsh surface. Near the during spring freshets. Vegetation is tidal fl at edge, the vegetation is scat­ continuous and low in diversity. tered, but becomes continuous up the slope. Immature high marshes are relatively level with some bare depressions and are Low ~ marshes develop on fine­ located on silty substrates. Organic riat­ textured sedinents, silt, or mud substrate ter accumulation is abundant here as it is in low energy parts of estuaries and are in sedge wetlands. Immature high riarshes relatively flat. These marshes develop in occur above sedge and low sandy marshes, areas of rapid sedimentation and are usually at least 40 cm above the tidal flooded by nearly all high tides. They flat; often the transition is an abrupt have diffuse drainage patterns with sorie rise. Many of the high tides, especially

12 Table 4. Characteristics of the eight types of marshes described for Oregon by Akins and Jefferson (1973).

Marsh Sub- Relative Vegetative type strate elevation cover Drainage Slope

Low sandy Sand Slightly above Continuous Diffuse Gradual toward tideflat except near upland tidefl at Low silty Silt or As above As above Diffuse As above clay plus small channels Sedge Silt Slightly above Continuous Deep Nearly 1 evel low silty marsh channels (20-30 on above tidefl at) Immature Peat 40 cm or more Continuous Channels Nearly 1 evel high over above ti defl at silt Mature Peat Slightly above Continuous Deep Nearly level high over immature high but with channels with depression silt pans Bulrush Silt or Similar to low Continuous Diffuse Nearly level and sedge sand silty in brack- ish to fresh tidal water Intertidal Gravel Variable Sparse Diffuse Sloped perhaps gravel and steeply sand Diked Peat As immature or Continuous Fonner Nearly level over mature high channels silt filled the higher highs, cover the soil surface. rainfal 1, tidal input, and evaporation. A well-defined system of channels drain These pools may become very saline or even and flood these marshes, which have few dry when evaporation rates are high and open areas in the vegetative canopy. neap tides do not carry water high enough to reach the depressions. Mature high Mature .b..i.9..!:!. marshes are level and marshes are found a meter or more above have developed extensive peaty soils. A the tidal fl at. dendritic network of steep-sided stream channels circulates water to the soil sur­ Bulrush and sedge marshes are low face on higher high tides. Shallow saline marshes in brackish parts of the estuary. pools (pans) produce openings in the The substrate is silt or sand, and inunda­ otherwise continuous sward of vegetation. tion occurs with most high tides. Drain­ Salinities fluctuate widely and depend on age is diffuse. Vegetation is continuous

13 and its composition is dependent on the characteristics due to seepage, high water salinity. tables, and perhaps residual salinity. Vegetation is continuous over the riarsh Intertidal gravel marshes are rare surface and the old dendritic stream chan­ fonns which develop on sand and gravel nel system has collapsed. bars near the mouths of relatively high­ energy estuaries with large volumes of In the recent classification system freshwater. Vegetation is discontinuous developed by the U.S. Fish and Wildlife and of a type which indicates low sal in­ Service (Cowardin et al. 1979), the tidal ities. The salt from the tidal water is marshes of the Pacific ~brthwest are probably leached by rainwater and fresh­ classified as follows: SYSTEM-estuarine; water runoff through the coarse sub­ SUBSYSTEM-intertidal; CLASS-emergent wet- strates. 1 and; SUBCLASS-persistent; WATER REGIME­ regul arly flooded or irregularly flooded; Diked salt marshes are rnanmade habi­ WATER CHEMISTRY-rnixohal ine (0.5 to 30.0 tats that develop when the tides are ex­ ppt); and SOIL-mineral or organic. Addi­ cluded from immature and mature high tional modifiers that pertain to some marshes. Although non-salt marsh pl ants northwest marshes are diked and artificial may invade, the area retains some wetland (dredge materials).

14 CHAPTER 3 BIOTIC C().tMUNITIES

This chapter describes the communi­ The 1 iterature on the second group, ties of organisms in the tidal marshes of the auto trophic and heterotrophic the Pacific Northwest. The marsh complex microbes, is sparse. In this section it includes the vegetated, relatively flat has been necessary to draw on the 1 itera­ portions which drain during low tide, the tu re from other regions of the country to shallow pans which may hold water for provide information on the probable micro­ months, and the creeks which circulate bial activities in the Pacific Northwest. water throughout some types of marsh. Although data from other areas are not as These creeks may be dry during low tide or good as firsthand studies, they do provide retain water throughout the tidal cycle. a basis for future research and some There are a number of macrophytic pl ant insight into the probable role these orga­ communities, and the species richness nisms play in the ma rs hes of Oregon and within the marshes of a single estuary is Washington. often high. Jefferson (1975} has identi­ fied approximately 70 species in the salt Al though the faunal communities are marshes of the region. The 1 iterature on the best described communities in the plant species composition and community Pacific l'brthwest marshes, the foodweb structure is relatively abundant. The relationships between populations is not marsh microbial communities on the decay­ very well known. There is, however, a ing plant material and in the soil have rather extensive Canadian 1 iterature from not been studied in any detail. Faunal which to draw information about the components of the systan have received fishes. more attention because of their economic importance. Extensive species lists of invertebrates, fishes, birds, and mammals 3.1 VEGETATION - ZONATION ANO SPECIES have been compiled. DISTRIBUTION Several aspects of seed-bearing vege­ Zonation in marshes of the Pacific tation are important. The first, zonation l'brthwest, as in marshes elsewhere, is and species distribution, is of obvious 1 argely a result of the pattern of tidal interest relative to marsh types and inundation. The role of inundation is roles. The second aspect is the upland complex and may involve a multitude of boundary, which, besides its scientific factors including soil salinity, moisture, interest, is assured particular importance aeration, and nutrient status. Burg et because of its role in defining the limits al. (1976), Disraeli and Fonda (1978) and of regulation relative to Section 404 of Ewing (1983), among others, have identi­ the Clean Water Act (Public Law 92-500). fied salinity, elevation, and inundation Because of its importance in wetland as important factors in controlling com­ management, we will defer the discuss ion munity composition. One of the most of boundaries until Chapter 5. Topics not widely studied effects of inundation is dealing with the structure of the plant development of the salinity regime to community, such as productivity, are sum­ which the plants are exposed. The salin­ marized in Chapter 4, which discusses eco­ ity gradient in the estuarine marshes has logical processes. vertical, 1 ateral, and 1 inear components.

15 The vertical gradient extends from the Table 5. Eight Oregon marsh community soil surface downward and may be important types and their typical pl ant species com­ in detennining which part of the root sys­ position (Akins and Jefferson 1973). tem is functional in water uptake. In the 1 ateral gradient from the mud fl at edge to the upland boundary, the higher salinities generally occur in the lower marsh with Marsh the upland edge being nearly fresh. Hori­ type Plant species zontally. salinity decreases from the river mouth to the limit of tidal influ­ ence. Low sandy Pickleweed, three-square, saltgrass, Jaumea, seaside Other effects of tidal waters which plantain, sandspurry, influence vegetational zonation include Lyngbye's sedge, milkwort aeration of the root zone and nutrient supply. Soil type is al so important, and Low silty Arrow-grass, spike-rush, in coarse sandy soils salt is removed from sand spurry the surface by the rapid percolation of rainwater. Clay soils have low percol a­ Sedge Lyngbye's sedge t ion rates, and penneability is decreased further when the cation exchange capacity Immature Tufted hairgrass, salt­ of the clays becomes saturated with sodium high grass ,arrow-grass, pickle­ ions. Under these conditions, soil peds weed, Lyngbye's sedge break down and the soil structure deterio­ rates in a massive form. making the move­ t1atu re Tufted hairgrass, Baltic ment of air and water through the sub­ high rush, creeping bentgrass, strate difficult (Gallagher 1977). gum plant, Pacific silver­ weed, orache The eight marsh types outl i ned by Akins and Jefferson (1973) and the typical Bulrush Bulrush, Lyngbye's sedge plant assemblages associated with them are and sedge 1 isted in Table 5. Table 6 1 is ts plant species common to Pacific f>brthwest tidal Intertidal Spike-rush marshes. In the low sandy marshes of Ore­ gravel gon, the areas closest to the tideflats are dominated by three-square, pickl eweed Diked Tufted hai rgrass, salt (Figure 8), sal tgrass. and al kal igrass rush, creeping bentgrass, (common on the northern part of the gum plant, Pacific Pacific ltlrthwest coast), along with some silverweed, orache blue-green algae and Cladophora (a green alga). Sal tgrass, seaside plantain (Figure 9), sandspurry, Lyngbye's sedge, and mil kwort are common at higher el eva­ tions in the low sandy marsh (Jefferson 1975). The parasitic plant dodder is The sedge marshes consist almost often found growing over sal tgrass and solely of monospecific stands of Lyngbye's pickleweed plants. sedge. Extensive sedge marshes are located on the Siletz River delta and the Circular patches of arrow-grass brackish portions of estuaries where the (Figure 10) and clumps of pickleweed are intertidal zone is silty. The tall creek­ common in the low silty marshes, as are bank pl ants are seen in Figure 11. In scattered spike-rush plants and sandspurry. many respects Lyngbye's sedge appears to Possible secondary invaders to these areas be a Pacific Northwest analogue of the are tufted hai rgrass and Lyngbye 's sedge east coast smooth cordgrass, Spartina (Jefferson 1975). alternifl ora. Lyngbye' s sedge grows in

16 Table 6. Common plant species found in Pacific Northwest tidal Marshes. L=low marsh; H=upper marsh.

Position in Common name Scientific nar.ie Family the marsh f'btes

Al kal igrass Puccinellia pumila Graninae L Most abundant in muddy areas, also on Atlantic coast Arrow-grass Triglochin maritimum Juncaginaceae L Often a pioneer plant at mudflat- marsh interface

Baltic rush Juncus balticus Juncaceae H Also on Atlantic coast

Bulrush Sci rpus val idus Cyperaceae L Found on muddy shores Creeping bentgrass Agrostis alba Graminae H Fonns meadow-like swards ,_. -..J Dodder Cuscuta sal ina Cu sc utaceae L Parasitic on many plants of the 1ower marsh

Eel grass Zostera marina Zosteraceae L A seagrass often seen as wrack which has been washed up into the marsh Gum pl ant Grindelia integrifolia Compos itae H A composite distributed from Alaska to northern California, fonnerly §_. stricta Jaumea Jaumea carnosa Compos i tae L Asucculent composite found in the low marsh

Limegrass El ymus moll is Graminae H Found in upper fringes of marsh and on sand dunes (continued) Table 6. Continued.

Posit ion in Common name Scientific name Family the marsh ~tes

Marsh clover Tri fol ium wonnskjolcl_i_i_ Leg um i nosae H A legu!:le of the high marsh and dunes Meadow barley Hordelll1 bracht:antherum Graninae H Common in mature high marshes Mil kwort Gl aux mari ti ma Primulaceae L Widespread in many marshes of Arctic and temperate North America

Orache Atripl ex patul a Chenopodiaceae H Also on Atlantic Coast Orthocarpus Ort ho carpus Scrophulariaceae L Paintbrush owl-clover, usually ..... castil l ejoides doesn't fonn monospecific stands ro Pacific silverweed Potentilla pacifica Rosaceae H Found in high marsh and moist sand dunes

Sal tgrass Distichlis spicata Graminae L In marshes and on beaches, also on Atlantic Coast Salt rush Juncus lesueurii Juncaceae H Found in high marsh and sand dunes Pickl eweed Salicornia virginica Chenopodiaceae L A succulent found in marshes and on beaches, also on Atlantic Coast

Sandspurry ~ularia canadensis Caryophyl l aceae L Cornr.ion from Alaska to northern California

Sandspurry Spergularia macrotheca Caryophyl 1 aceae L & 11 Common from Britis~ Columbia to Baja California (continued ) Table 6. Concluded.

Position in Common name Scientific name Family the marsh tbtes

Sandspurry Spergul aria marina Caryophyll aceae l Less conmon than the two previous species

Seas i de p1 a nta i n P1antago maritima Pl a ntag i naceae L Produces a 1arge fleshy root which is the perenniating organ

Lyngbye's sedge Ca rex 1yngbye i Cyperaceae L Occupies position similar to smooth cordgrass of Atlantic and Gulf Coasts S1 ough sedge Carex obnupta Cyperaceae Found in nearly freshwater fringes of saline marshes ...... '° Spike-rush Eleocharis parvu1a Cyperaceae L Widely distributed along the entire Atlantic coast as well

Spike-rush Eleocharis palustris Cyperaceae L Widespread in temperate and cold­ temperate regions of the northern hemisphere Three-square Scirpus americanus Cyperaceae l Widely distributed in U.S. and southern Canada Tufted hai rgrass Deschampsia cespitosa Grarti nae H World-wide distribution Wes tern dock Rumex occidental is Polygonaceae H Also occurs in Rocky Mountains Figure 8. Pickl eweed as a pioneer on the Figure 9. Seaside plantain growing in a tidal flat in a low sandy marsh. The low sandy marsh. These perennials produce lighter colored halo around the vegetation large fleshy tap roots which store re­ is sand trapped by the plant sterns. Sub­ serves during the winter season. strate levels within the vegetation are often 2-3 inches above the surrounding substrate. large monospecific stands, occupies a highest in this marsh type, to the point relatively low position in the intertidal where dune grasses begin, includes tufted zone, produces large amounts of detritus, hairgrass, Pacific silverweed (Figure 12), and is more productive along the stream­ marsh clover, and Baltic rush (Jefferson banks than in the back marsh. 1975}. Bulrush and Lyngbye's sedge are char­ Intertidal gravel marshes are less acteristic plants of the bulrush-sedge floristically diverse and are often domi­ marsh type. Bulrush becomes predominant nated by spike-rushes of several species. upstream where the water is fresher. At Tufted hairgrass, Baltic rush, and creep­ the mouth of the South Fork of the Siuslaw ing bentgrass characterize the mature high River there are 120 hectares (296 acres) marshes. Pacific silverweed (Figure 12), of this type (Jefferson 1975). Extensive gum plant, and orache are also common. areas of this r.iarsh type al so border the Many of Oregon's mature high marshes have Columbia River Estuary. been diked and thus fal 1 in the category diked salt marsh. Will apa Bay, Washing­ Tufted hai rgrass and saltgrass are ton, includes some marshes of this type most common to the immature high marsh. that have not been diked. Arrow-grass, Baltic rush, seaside plantain, pickleweed, and Lyngbye's sedge are less A typical zonation sequence for a prevalent. Three-square, milkwort, alkali­ high-salinity marsh, where the tidal water grass (in the northern part of the salinity is above 25 ppt, is shown in reg ion), sandspu rry, and Orthocarpus r.iay Figure 13. Sequences wil 1 vary depending al so be present. The community growing on substrate, salinity, and chance. Local

20 Figure 10. Circular ciurnps of arrow-grass Figure 11. Lyngbye's sedge marsh in a in a low silty marsh. These tall plants brackish estuary. Streamside plants may trap litter of eelgrass detached from beds be nearly twice as productive as those in on intertidal flats. the back marsh away fror1 the creeks.

intrusion of fresh groundwater may, of creekbanks and the other further back from course, drarnatical ly alter this pattern. the drainage channels. On the creekbank For exarnpl e, a freshwater table may the plants start growth earlier in the develop under the vegetation in the dune spring, are taller, and have a higher net complex of a bay-mouth bar estuary. This primary productivity than the back marsh water table may surface at some point in plants. This appears to be associated at the ma rs h which fo nned on the bay side of least in part with the storage and remobi- the bar. The result is a zone of low 1 ization of carbohydrates in the rhizomes interstitial soil salinity in an unusual (Gallagher and Kibby 1981). Whether the setting. Plants in this zone may be par­ fonns of sed9es are ecophenes resulting ticularly vigorous as a consequence of the from the action of environment on geneti­ reduced salinity stress, or the species cally sirnilar plants or ecotypes whose present may be shifted as a consequence of growth and behavior are the result of the changed environment. The authors genetic differences has not been investi­ measured a situation where the intersti­ gated. tial soil salinity was 28 ppt seaward of the zone, 8 ppt in the zone of freshwater Most of the evidence regarding a intrusion, and 18 ppt in the soil toward similar situation for smooth cordgrass the upland. (Spartina alterniflora} indicates that the differences are environmentally controlled A typical profile in a brackish area_ and that the plants are similar qeneti­ is depicted in Figure 14. A number _of cally (Smart 1982). There is, however, salinity regimes could produce the setting some evidence for certain genetic dif­ for such a sequence. Intrusion of upland ferences between smooth cordgrass pl ants freshwater could produce the proper inter­ gr01-m in a common garden. Preliminary stitial soil salinities, or similar condi­ rneasurenents by Gallagher, Grant, and tions could arise fror.i exposure to tidal Sor:iers indicate that the morphological and water of brackish salinity. The low ele­ productivity differences for the two vation is occupied by Lyngbye's sedge and growth fonns are rnai nta i ned for at least 5 the upper 1 evel s by a mature high marsh yea rs. association. There appear to be two fonns of Lyng bye's sedge: one found along the In some cases the vegetation zones in 21 green fonn, Enteromopha intestenalis, from a site in the Grays Harbor Estuary, Wash­ ington. His sampling program was aimed at the entire estuary and in most cases the exact habitat sampled (rock, mud, creek, epiphyte, log, marsh, soil) along a tran­ sect was not reported. Published informa­ tion on tidal marsh algal distribution and diversity is lacking. cur casual observa­ tions indicate that there will prove to be a rich flora (particularly the diatoms) and that many will show a general associa­ tion with particular marsh types. Popula­ tions of photosynthetic bacteria are obvious in some of the tidal pools, but we have not been able to find any records of their study in the region. We were unable to find studies of the populations of decomposer microbes or aerobic or anaero­ bic soil bacteria, In Atlantic coastal marshes, ATP has Figure 12. Pacific sil verweed from a often been used as a measure of microbial mature high marsh. These ~ants are biomass (Christian et al. 1981), and its structurally weak and decompose rapidly in concentration in marsh soil and tidal the fall after they die. creek water has been used to estimate the distribution of microorganisms with respect to depth and time. Christian (1976 as cited in Christian et al. 1981) the Pacific Northwest marshes are clear found ATP concentrations to decrease with and may comprise nearly monospecific depth in all marsh soils tested. ATP con­ stands (top of Figure 15). In other centrations were greatest in both the soil marshes the zones are diffuse and contain and water during the warmest months of the a mosaic of many species (bottom of Figure year. Christian et al. (1981) concluded 15). Although these zonation patterns that 79% of the standing stock of the within the marsh are primarily of ecologi­ microbial community in the marsh is asso­ cal interest, the boundary between the ciated with the soil. marsh and the upland is also of legal importance. This topic wi11 be further Microbes are found in the water as discussed in Chapter 5. well as in the soil or on plant material. In the water most microbial plankton are free floating. Some of the organisms, 3.2 MICROBES primarily bacteria, are attached to detri­ tus particles; it is these that are most Algal communities associated with active in consuming DOC (dissolved organic Pacific f'brthwest tidal marshes consist of carbon) produced by smooth cordgrass and filamentous algae as well as di atoms and phytoplankton. Heterotrophic activity are abundant at times during the year. In varies with the tide because the resuspen­ the more si1 ty substrates diatoms appear sion of detritus is greatest on an ebbing to be the most common forms on the creek­ tide and lowest at slack tide (Christian banks, and they al so cover the collapsed et al. 1981}. The microbial populations dead plant material. Macroscopic fonns in salt marsh soils were not found to be are most conspicuous in the shallow creeks nutrient limited but were inhibited by and in the tidal pools or pans. Thom increased salinity and decreased soil (1981) reported a macroscopic brown alga, moisture at the times when the marsh was Fucus distichl is spp. edentatus, and a not flooded (Christian et al. 1981).

22 HW-

TIDEPOOL

Figure 13. Typical zonation of marsh pl ants in a saline Pacific rtirthwest tidal marsh. The tidepools or pans may contain a diversity of algae. The lateral extent of the zones depends on the slope and may range from a few yards to hundreds of yards.

HW_.

TUFTED HAIRGRASS

LOWER VIGOR L YNGBYE'S SEDGE

HIGH VIGOR L YNGBYE'S SEDGE

Figure 14. Possible zonation pattern in a brackish Pacific rtirthwest tidal marsh.

23 Figure 15. Exar:ipl es of the range of variations in vegetational zonation in Pacific llbrthwest tidal marshes. The upper photograph shows distinct, nearly monospecific stands of three-square, tufted hairgrass, and slough sedge, going from lower left to upper right. The lower photograph shows diffuse zones of high plant diversity in a 1ow sandy marsh. The dark tall clumps of plants are arrow-grass and the shorter zones are mixed areas of pickleweed, gum weed, seaside plantain, saltgrass, and Orthocarpus.

24 Up to this point the discussion of nitrogenous oxides to ammonia may be an microorganisms has dealt with an aerobic important process in soils where deni tri­ enviro11nent; the soils of the salt marsh, fication takes place (Wiebe et al. 1981). however, are predo~inantly anaerobic. Fennentation, sulfate reduction, dissimi- Methane is another product produced 1atory nitrogenous oxide reduction and in anaerobic marsh soils. In the process methanogenes1s. are the major processes' of rnethanogenesis, carbon dioxide or a occurring in anaerobic tidal marsh soils methyl group serves as the electron (Wiebe et al. 1981). Fennentation is acceptor. Competition between populations carried out by facultative or obligate of methanogenic and sulfate reducing bac­ anaerobes. Facultative anaerobes can teria for substrates regu1 ates methano­ function either in the presence or absence genesis. Nitrogenous oxides al so play a of oxygen, but obligate anaerobes can grow regulatory role, thereby inhibiting metha­ only in an anaerobic environment. In the nogenesis. By limiting the substrates breakdown of organic matter during fennen­ used by sulfate reducers, denitrification tation, an organic compound, rather than has been shown to increase methanogenesis oxygen, acts as the terminal electron (Wiebe et al. 1981). Another important acceptor. Alcohols and acids are included factor discussed by Wiebe et al. (1981) is ac1ong the possible end products of this water flow -- the tennination of water process. These end products may then flow reduces sulfate concentration and serve as substrates for other anaerobes increases methane production. Thus the nearby (Wiebe et al. 1981). four anaerobic processes and the associ­ ated bacterial populations occurring in Sulfate acts as the terminal electron salt marsh soils are coupled to one acceptor during sulfate reduction in another. anaerobic soils. The bacteria that carry out this reaction release sulfides into the soil. One can often smell the sul­ 3.3 FAUNAL Cet-1PONENTS fides in marsh soils containing a large quantity of organic matter. Sul fate Among the numerous faunal surveys of reducers belong to only three genera: various types which have been carried out Desulfovibrio, Desulfuromonas, and Desul­ for the Pacific t-brthwest tidal marshes fomacuium {Wiebe et al. 1981). Ewing are those by Higley and Holton 1981, Hoff­ (1983), working in the Skagit Estuary nagle et al. 1976, Stout 1976, and Roye marsh in Puget Sound, found reducing con­ 1979. We have divided the fauna into the ditions in the soils at most stations. following groups for discussion: inverte­ Areas near pan and creek bottoms had the brates, fishes, birds, and mammals. Much lowest redox potentials. In New England, research is still needed with these marshes resemble those in the Pacific groups, especially in quantifying popul a­ rt>rthwes t in temperature regime and often tion sizes and cycles. in peat development. Sulfate reduction in the northern Atlantic marshes is greater 3.3.1 Invertebrates than in those further south. 1-bwa rth and Giblin (1983) found rates in Georgia to be Information on the invertebrate fauna significantly lower than those in Massa­ of the Pacific t-brthwest marshes is frag­ chusetts. Sul fate reduction appears to be mentary. The most comprehensive studies the major fonn of respiration in the soils are those of Hoffnagle et al. (1976), who in both states. We know of no similar studied the Coos Bay marshes, and Higley evaluations made in the Pacific rt>rthwest. and 1-blton (1981), who investigated the marshes of Siletz and Netarts Bays, Ore­ Nitrate is the terminal electron gon. The 1atter authors determined the acceptor in denitrification and N? or N?O trophic structure of invertebrate marsh is the end product; the result is~ a loss communities by sampling soil infauna of of fixed nitrogen from the soi 1 • Anaero­ low and high marshes; invertebrate fauna bic conditions are also required for this of 1 ow (up to 15 cm above ground 1 evel) process. Dissimilatory reduction of and high vegetation (samp1 ed by terres-

25 trial sweep nets) of the low and high Mites and ticks (Order Acarina) were marshes; fauna of the marsh debris line abundant in the vegetation closest to the (Figure 16); fauna of the submerged marsh; ground (O to 15 cm) in all marsh types. and fauna of marsh pans, tidal creeks, and Collembolan, Homopteran, and Coleopteran tidal flats. They found that ol igochaetes insects also inhabitated these areas. and dipteran larvae dominate the fauna of Higher in the canopy (above 15 cm above the marsh soils. This contrasts with ground level) of all marsh types Acarina, marsh soils of the Atlantic coast where Araneae, Homoptera, Diptera, and Hymenop­ polychaetes were much more abundant than tera were common. Acarina, Araneae, Col- ol igochaetes ( Cammen 1976, as cited in 1 embol a, and Amphipoda inhabited the Higley and Holton 1981). The dominant debris line. Overall, taxonomic diversity amphipod in the Pacific f'brthwest was was greatest in the high marsh, decreasing Corophium salmonis, common where low-silt to the low marsh and then to the debris marsh merged with tidal flat habitat. line. Table 7 lists the invertebrate inhabitants and the stage of the life cycle found in Oregon tidal marshes. Table 8 gives approximate densities for these animals for several habitats. Acarina and ol igochaetes were moder­ ately abundant in marsh vegetation sub­ merged at high tide. Coleoptera, Homop­ tera, Hemiptera, and Collembola were also among terrestrial taxa collected from the submerged vegetation. Aquatic crusta­ ceans, occupying marsh pans and tidal creeks, included amphipods (Corophium spp., Anisogammarus confervicolus, Orches­ tia traskiana), the isopod, Gnorimo­ sphaeroma lutea, and two Cumaceans (Hemi- 1 eucon spp. and Cumell a spp.). Some amphipods move inland before advancing tides and find shelter in dead eelgrass. Infaunal composition of tidal creeks and tidal pans of the marsh were similar (Higley and Holton 1981). Polychaeta, /\mphipoda, Tanaidacea, and Isopoda were found, many of whose species are common to the Atlantic Coast as well. A major dif­ ference in fauna between the Atlantic and the Pacific l'brthwest coastal marshes is the obvious scarcity of decapods in the l'brthwest. Atlantic marshes are teeming with fiddler crabs (Uca) on the order of 80 to 200/m2 (Montague et al. 1981), along with others such as the blue crab (Calli­ nectes sapidus). Higley and Holton (1981) found only one decapod, Hemigrapsus oregonensis, which was inhabiting tidal Figure 16. Ori ft line in an Oregon low creeks in the sedge and mature high sandy marsh. Wrack consists primarily of marshes. eelgrass. These deposits are greatest after storms late in the growing season Mollusks common to northwest tidal when eelgrass in the adjacent beds is marsh creeks include Alderia modesta, senescing and burdened with epiphytes. Macoma balthica (Higley and HoltoilT981),

26 Table 7. Invertebrates chardcteristic of five Oregon riarsh habitats (modified from Higley and Holton 1981). A= adult, L =larvae, N =nymphs, (-)=not divided into families but members of the order are present.

_filg_h rna rs h Low marsh Debris 1 ine Pan Tidal creek NJ. of NJ. of No. of No. of NJ. of Order Stage families Stage families Stage families Stage families Stage families

Cnideri a A 1 A

Turbellaria A

Nemertea A

~matoda A - A - A Po 11chaeta A 2 A 2 A 6 01 i go cha eta A - A - f\ - A N '-.I Gastropoda {\ 1

3ivalvia A 1 Araneae A - A - A - {\ Acarina A - A - A - A Os trdcoda A Copepoda A 2 A 3 Cirripedia A 1 A 1

Cumacea A 1 A 1 A

Tanaidacea A Isopoda A - A 1 A 1 A (continued) Table 7. Concluded.

High marsh Low marsh Debris line Pan Tidal creek No. of No. of No. of No. of No. of Order Stage far.iil i es Stage families Stage far,1ilies Stage families Stage fanil ies

Amphipoda A 1 A - A 1 A - l\ Decapoda A 1 Coll embol a A 5 A 3 Oiplura (Entognatha) A Orthoptera A

Thysanoptera A - A - A Nco Odonata N

Hemiptera A 2 A,N 4,1 A,N 1,1 A 1 A,N 2,1

Homoptera A 4 A 4 A 1

Col eoptera A 9 A 4 A 3 A 2 A 1 Trichoptera L 1 L 1 lepidoptera A - L 1 A Diptera A,L 15,5 A,L 11,6 A 4 L 7 A,L 7,6 Hymenoptera A - A - A 1 Chil opoda A Table 8. Approximate densities and percentages of invertebrate taxa from five habitats of an Oregon marsh (modified from Higley and Holton 1981.) (*) = taxa making up <10% of collection;(-)= no data available.

2 rtimber of animalsLm {%} Immature Mature Invertebrate Low sand Low silt Sedge high high MARSH SOIL (Feb.)

01 igochaeta * 5,680(22) 2,990(30) 4,570(61) 990(20) Amphi poda * 11,610(45) * * * Di ptera (larvae) 23,880(76) 7,740(30) 4,480(45) 1,870(25) 2,570(52) Areneae * * * * 490(10) Aca ri na 3,770(12) * * * * LOWER CANOPY VEGETATION (Sept.) (ground 1 evel to 15 cm) .A.ca ri na 450(84) 1, 080( 69) 3,770(95) 1,710(62) 2,670(54) Isopoda * 240(15) * * * Coll embol a * * * 830(30) 1,730(35) Homoptera * 250(15) * * * UPPER CANOPY VEGETATION (Sept.) (above 15 cm above ground 1 evel) Araneae * * * 17(15) * Acarina 30(33) 680( 47) 510(52) 14(12) * Homoptera 10(10) 570(40) 320(33) 25(21) 110(53) Di ptera 40(45) * * 40(35) * Hymenoptera * * * 18(16} 40(18) DEBRIS LINE (Aug.)

Acarina 10,210(63) Coll embol a 3,240(20) SUBMERGED VEGETATION (Feb.) (at high tide) 01 igochaeta * * 690(22) 380(49) Acarina 8,030(96) * 1,410(45) * Isopoda * 6,480(76) * * Diptera * * 940(30) 160(21) (continued)

29 Table 8. Concluded.

2 l'llmber of animal s/m (%) Low Immature Mature Invertebrate sand Low silt Sedge high high

PAN \~ATER {April) 01 igochaeta 120(11) 1,370(46) Copepoda 580(54) * Amphipoda 350(32) * Diptera (1 arvae) * 1,190(40) TIDAL CREEK SOIL (l'bv.) Polychaeta 20,220(10) 7,660(20) 01 i gochaeta 145,590(72) 18,380(48) Amphi;Joda * 5. 740(15) TI DAL CREEK WATER ( l'bv.) Cnidaria * 50(10) Nemertea * 130(28) Polychaeta 1,070(22) * 01 i go cha eta 1,220(25) 100(20) Cumucea 190 (1 O) * Amphipoda 970(20) 120(26)

Macom~ inconspicua, and the soft-shelled 3.3.2 Fishes clam Mya arenaria (Macdonald 1977). Ma coma ~uta and Cryptomya ca 1 ifo rni ca The role of tidal narshes as nursery were al so found in tidal creeks of Grays and feeding grounds for salmon and other Harbor, Washington (Macdonald 1977). The fish has yet to be proven in Pacific major conclusion in the species inventory l'brthwest marshes. However, •:!vidence to carried out by Hoffnagle et al. (1976) was support their role in the fishes 1 1 ife that in a particular marsh one or two spe­ cycle is beginning to accumulate. Inven­ cies of invertebrates dominated while the tories of fish have been made in various rest were fe\~ in number and random in dis­ estuaries of the Pacific Northwest, but tribution. much 1 ess inforr.iation is available on the fish that inhabit the tidal ~arshes (Table It is not known if or how the tidal 9). Hoffnagle et al. (1976) surveyed fish marshes are important to the economically species of tidal creeks of the marshes of important invertebrates of the estuary Coos Bay, Oregon, a11d al so carried out gut such as the Dungeness crab and Japanese analysis to obtain infor.nation on feeding oyster. Large nu;nbers of small Dungeness habits. The two do•ninant fishes of these crabs have been found in the upper reaches marshes were the shiner perch and the of the Coos Bay Estuary (Roye 1979). More staghorn scul pin. Major food sources study, however, is necessary to ascertain included ar.iphipods, especially Corophium, whether the tidal narsh plays a valuable in the case of the staghorn scul pin and role in the survival of these animals. ha rpacticoid copepods for the shiner

30 Table 9. Common and scientific names, marsh habitat, and life stage of some fishes associated with marshes of the Pacific Northwest (Hoffnagle et al. 1976; Levy et al. 1979; Northcote et al. 1979; Higley and Holton 1981).

Common name Scientific name Marsh habitat Life stage

Anadromous species Chinook salmon Oncorhynchus tshawytscha Sedge marsh - edge Juvenile Chl.111 salmon Oncorhynchus keta Low marsh - level portion Juvenile Coho sal man Oncorhynchus kisutch Marsh creek Juvenile Long fin smelt Spirinchus thaleichthys Marsh creek Pink salmon Oncorhynchus gorbuscha Marsh creek Juvenile Sockeye salmon Oncorhynchus nerka Marsh creek Juvenile

,_.w Marine species Northern anchovy Eng raul is mo rdax Sedge marsh - edge Shiner perch Cymatogaster aggregata Sedge marsh - edge and creek Juvenile Staghorn scul pin Leptocottus annatus High marsh - edge, creek and pan Juvenile to Low marsh - edge and creek young adult Starry flounder Platichthys stellatus Low marsh - 1 evel portion, edge and creek Surf smelt Hypomesus pretiosus Low marsh - 1 evel portion Juvenile

Freshwater species Pearnouth chub Mylocheilus caurinus Marsh creek Prickly scul pin Cottus asper Marsh creek Threespine stickleback Gasterosteus acul eatus High and 1 ow marsh - edge, Juvenile to creek and pan adult perch. Detritus particles from salt marsh 3,3.3 Birds plants are the major food of Corophium; thus there is a link between fish and Salt marshes of Pacific Northwest marsh plants. In Hoffnagle's study, juve­ estuaries are located along the Pacific nile fish dominated the sites, which also Flyway for migratory waterfowl. Along supports the nursery role of the tidal with the tidal flats and open water, the marshes. marshes provide prime habitats for feeding and wintering. Magwire (1976a) conducted Higley and Holton (1981) found that the most comprehensive survey on marsh the species diversity of fish in the Pa­ utilization by birds of the Pacific tbrth­ ci fie ltlrthwest marsh habitats was not as west in a summer study at Coos Bay, Ore­ great as in those along the Atlantic gon. He studied six marshes at Coos Bay coast. They suggested that perhaps this by dividing them into the following zones: is due to lower salinity (and therefore shrubs and trees at the upper marsh edge, fewer marine species are temporary resi­ high marsh, middle marsh, low marsh, mud­ dents) or to the 1 esser extent of f'brth­ flat, and water. Bird observations were west marshes. In both high and low made as he traversed each marsh from up- marshes at Netarts and Siletz Bays, stag­ 1 and to mud fl at. He found that 28 bi rd horn scul pin and threespine stickleback species utilized the marsh proper, in­ dominated. Other species captured by cluding the air space over the marsh. seine and trawls included juvenile surf These are listed in Table 10 according to smelt and juvenile chum salmon. In addi­ the location in the marsh where they were tion to the shiner perch and the three­ observed. spine stickleback in a slough adjoining a sedge marsh, nine other species were cap­ The five swallow species (barn, tured but in lower numbers. Among these cliff, rough-winged, violet-green, and less common species were northern anchovy, tree) are migratory birds and are commonly starry flounder, juvenile chinook salmon, seen over the marsh from April to August. and the shiner perch (Higley and Hal ton They have large mouths and prey on flying 1981) • insects. tbne of the swallows build their nests in the marsh, but the barn swallow Most infonnation on fish residents of uses mud taken from the marsh edge to Oregon and Washington estuaries comes from build its nest, usually under building ang 1 er catch data. Comprehensive surveys eaves (Magwire 1976a). have not been done in most cases; studies investigating the utilization of the The widely distributed song sparrow, marshes by these fish are especially lack­ the most common bird Magwire observed ing. For this information we have to rely over the marsh. often nests in Marsh on the studies done in British Columbia plants, in snags found in the marsh, or in and assume that many of their conclusions the brush adjacent to the upland. These are transferrable to other Pacific ttirth­ birds can be seen probing for worms and west estuarine systems, at least until insects in the marsh and its creeks such studies are completed in Oregon and (Magwire 1976a). Washington. Chinook and coho salmon (chum salmon are found in a few of the estua­ The finches were observed to eat ries), steel head trout, cutthroat trout, seeds of arrow-grass but were seen only shad, and sturgeon frequent many of Ore­ infrequently in the marsh (Magwire 1976a). gon's estuaries (Kreag 1979a, b; Roye 1979). Other common fish include various The northern harrier builds its nest species of perch, starry flounder, Pacific on the marsh surface and soars over the staghorn scul pin, Pacific herring, surf marsh searching for small mammals. The smelt, and northern anchovy. Many other red-tailed hawk generally lives and hunts species have been reported in these estu­ outside the marsh but was occasionally aries; at least 66 species are known to seen in the upper part of the marsh (Mag­ Coos Bay. wire 1976a).

32 Table 10. Bird observations in salt marshes at Coos Bay (modified from Magwire 1976a).

Species Upland Marsh Mud fl at Water High Middle Low

Mallard (Anas platyrhynchos) x x

Turkey vulture (Cathartes ~) x x x x x x ft>rthern harrier (Circus cyaneus) x Red-tailed hawk (Buteo jamaicensis) x x x Great egret (Casmerodius albus) x

Great blue heron (Ardea herodias) x x x x x x Green-backed heron (Butorides striatus) x x x x x

Virginia rail (Rallus limicola) x x x x x

Kil 1 deer ( Charadrius vociferus) x

Sandpipers a x x

Band-tailed pigeon {Columba fasciata) x x

Common nighthawk (Chordeiles minor) x x x x x

Be 1 ted kingfisher ( Ceryl e a le yon) x x

Barn swallow (Hirundo rustica) x x x x x x

Cliff swallow (Hirundo pyrrhonota) x x x x x x

lt>rthern rough-winged swallow (Stelgidopteryx serripennis) x x x x x

Violet-green swallow (Tachycineta thalassina) x x x x x

Tree swallow (Tachycineta bicolor) x x x x x x American crow (Corvus brachyrhynchos) x x x

Marsh wren (Cistothorus palustris) x x x American robin (Turdus migratorius) x x x x European starling (Sturnus vulgaris) x x x x x Common yellowthroat (Geothlypis trichas) x Red-winged blackbird (Agelaius phoeniceus) x Purple finch (Carpodacus purpureus) x American goldfinch (Carduelis tristis) x x x Song sparrow (Melospiza melodia) x x x x x

aSandpipers include the western sandpiper (Calidris mauri). least sandpiper (Calidris minutilla), white­ rumped sandpiper (Calidris fuscicollis), and Baird's sandpiper (Calidris bairdii).

33 Common to the higher marshes is the larger mammals. This omnivore feeds on long-billed marsh wren, nesting a foot fruits, fish, invertebrates, and small or two above the ground in the tall vege­ vertebrates and mammals, as well as eggs. tation of this zone. This bird feeds Another marsh visitor is the black-tailed on marsh insects, snails, and spiders in deer, which browses in brush areas gener­ the grass and on the ground (Magwire ally outside the marsh. Matted marsh 1976a). grass and the many tracks indicate that the deer utilize the marsh as a refuge. Virginia rails bui1d their well­ Deer have been observed by the authors concealed nests close to the ground in feeding on arrow-grass in the marshes. the marsh. Their long curved bills are Beaver live at the edge of one of the especially suited to probing in the mud marshes of Coos Bay, where the freshwater for wonns, sna i1 s, larvae, etc. They al so meets the estuary. They feed on bark of feed on pl ant seeds. Soras rails were deciduous trees at the marsh's edge. al so occasionally sighted in the marsh Other large mammals reported to frequent areas of Coos Bay (Magwire 1976a). marsh areas of Coos Bay are muskrat, mink, river otter, weasels, grey fox, coyote, The great blue heron is seen in those and bobcat (Magwire 1976b). The abundance marshes adjacent to large mudflats. Her­ of these animals in the marsh is primarily ons feed on fish and invertebrates in related to the proximity of urban areas. shallow water at the edge of the marsh and The closer the urbanization, the fewer the in marsh creeks. They seek refuge in the animals. A list of the common and scien­ marsh to res to re and preen their feathers tific names of those species frequently but do not nest there. Their nests are associated with the marshes is given in built in large rookeries in Sitka spruce Table 12. or hemlock (Magwire 1976a). Magwire (1976b) captured six species Based on a census made each December, of small mammals in a study of low sandy, Roye (1979) associated the following birds low sedge, immature high, and high marshes with Coos Bay salt marshes: American at Coos Bay. The vagrant shrew made up wigeon, black-bellied plover, willet, 71% of the captures; the next most abun­ common gol deneye, northern harrier, bald dant species was the deer mouse, compris­ eagle, red-tailed hawk, great blue heron, ing 23% of the small mammal catch. Four green heron, American coot, tundra (or other species were captured more rarely - whistling) swan, Canada goose, gadwall, the Oregon meadow mouse, western red­ snowy egret, Virginia rail, long-billed backed mouse, the black rat, and the Trow­ curlew, pectoral sandpiper, knot, Allerican bridge shrew. bittern, Great egret, sora common snipe, and mallard (see Table 10). The relative Magwire (1976b) correlated the rela­ abundance of these birds in the ma rs hes tionship between captures and marsh type. has been determined by Roye (1979) and is The deer mouse was most abundant in the given in Table 11. Many of these species high marsh. Its numbers declined as dis­ utilize other estuarine habitats as well. tance into the marsh from the terrestrial side increased. The deer mouse apparently 3.3.4 Mammals resides in the dense trees and shrubs just beyond the edge of the marsh and feeds on The most comprehensive study on mam­ fruits, seeds, and insects. The vagrant r.1a l populations of tidal marshes in Oregon shrew is an insectivore and prefers the was done by Magwire (1976b) in the salt dense herbaceous cover of the immature marshes of Coos Bay. Mammals are primary high marshes and sedge marshes. Here they consumers, predators, and scavenge rs in build their nests directly on the ground the salt marsh food web. The larger mam­ in the highest part of a particular area. mals generally range widely and are there­ There was no correlation between abundance fore not restricted to marsh habitats. and distance into the marsh; the vagrant Magwire ( 197 6b) found the raccoon to be shrew was observed throughout the entire the most frequent marsh visitor of the marsh. However, more were captured near

34 Table 11. Common and scientific names and relative abundance of birds associated with marshes of Coos Bay (Roye 1979).

Relative a Common name Scientific name abundance

American bittern Botaurus lentiginosus Rare American coot Ful ica americana Abundant American wigeon Anas americana Abundant Bald eagle Haliaeetus leucocephalus Rare Black-bellied pl over Pluvialus sguatarola Common Canada goose Branta canadensis Rare Great egret Casmerodius albus Co11111on Common goldeneye Bucephala clangula Uncommon Common snipe Gallinago gallinago Uncommon Gadwall Anas strepera Uncommon Great blue heron Ardea herodias Common Green-backed heron Butorides striatus Uncommon Red knot Cal idri s canutus Uncommon Long-billed curlew tilmenius americanus Rare Mallard Anas platyrhynchos Common rib rthern harrier Circus cyaneus Uncommon Pectoral sandpiper Calidris melanotos Rare

Red- ta i 1ed hawk Buteo jamaicensis Uncommon Snowy egret Egretta thul a Rare Sora Porzana carol ina Rare Virginia ra i1 Rall us l imicola Uncormion Tundra swan Cygnus columbianus Rare Wil 1et Catoptrophorus semipalmatus Uncommon aabundant = ..'.':_50/day/observer, common = 10-49/day/observer, uncommon = 0-9/day/observer, rare = .::_5/day/observer.

35 Table 12. Common and scientific names of some mammal species associated with the marshes of the Pacific Northwest.

Common Name Scientific Name Common Name Scientific Name

Residents Visitors Black rat Ra ttus rattus Beaver Castor canadensis Black tailed deer Odocoilous hemionus Deer mouse Peromyscus col umbi anus manicul a tus Bobcat Lynx rufus

Oregon meadow Microtus oregonii Coyote Canis 1a trans mouse Grey fox Urocyon cinereoargenteus Shrew Sorex vagrans Mink Mus tel a vi son Muskrat Ondatra zibethicus Trowbridge shrew Sor~ trowbridgi i Raccoon Procyon 1otor River otter Lutra canadens is Western red- Cl ethrionomys backed mouse occidental is Weasel Mustel a spp. logs which had been washed into the narsh. adjacent to Netarts Bay, only one mammal, the vagrant shrew, was encountered (Stout During tidal inundation the deer 1976). The paucity of mammal life in that mouse seeks higher terrestri~ ground marsh was attributed to the lack of fresh­ while the '/agrant shrew remains in its own water. Dew, precipitation, and food flu­ 1ocal e in the tops of t'ie emergent marsh ids provide the sr.iall rnarunal inhabitants vegetation or on top of 1ogs or other de­ of the salt marsh with freshwater (Magwire bris. As might be expected, these marsh 1976b). An interesting adaptation of the inhabitants are more vulnerable to preda­ deer mouse is that those captured from a tion by marsh hawks and owls during tidal salt riarsh can survive indefinitely on a inundation (~agwire 1976b). diet of dry food and seawater, while those taken from a r.iountain region cannot (Fis­ In a survey done in the salt marshes ter, as cited in Magwire 1976b).

36 CHAPTER 4 ECOLCXJICAL INTERACTIONS

Previous chapters have examined the 1 ack of macro invertebrates, particularly distribution, types, areal expanse, and snails and crabs (Uca} in the X3 compart­ development of tidal mars hes, the marsh ment. The anaerobic heterotrophs, x10, 2 flora and fauna, and the physical and are bacteria. Similarly, X , the 1 iving chemical environment in which the biota marsh pl ants, encompasses al 1 types found l i ve. From this we know where the ma rs hes in the P~cific Northwest marshes. The are and what the 1 iving conditions are for algae, X°, includes floating (phyto­ the biota which were discussed in Chapter plankton} and attached (micro- and macro­ 3. We have examined James Bonner's first benthic} algae. The most diverse compart­ 4 question of ecology, "Who lives where and ment is X , POC. This ranges from dead why?" We al so touched on his second marsh pl ants and algae to imported litter question "who eats whom?" This chapter (as eel grass, drift logs, or mammalian establishes a framework into which various feces} to fine particulate matter from any short food chains fit to develop a holis­ of these sources. tic view of the tidal marsh ecosystem. Potential interactions between the marsh Table 13 1 ists some typical processes and adjacent ecosystems will be examined, by which carbon is transferred between the and in a broad sense we wil 1 examine ecological components shown in Figure 17. Bonner's final ecological question, "What For example, the export of aerobic hetero­ is the biology of togetherness?" trophs in the air (F 3,5} could represent the loss of insects directly or through insectivorous birds (Pfeiffer and Wiegart 4.1 ECOLCXJICAL PROCESSES WITHIN THE MARSH 1981}. Quantitative data on most flux notes listed in Table 13 and the standing A generalized conceptual model of crops of the functional compartments in tidal marsh components and their interac­ Figure 17 are not available for salt tions is presented in Figure 17. The marshes of the Pacific rt>rthwest. CA!r model was assembled from data and observa­ experience has shown that the marshes of tion from Pacific Northwest marshes and the Pacific Northwest are structurally and experience in Atlantic Coast marshes. Al­ functionally similar to eastern Spartina though the model is based on carbon cycl­ al terniflora marshes. Studies of produc­ ing, any material (or energy) could be so tion, decomposition, and grazing support discussed using the diagram. this concept. The compartments of the model are 4.1.1 Primary Production lumped categories of functionally similar organisms. For example, X3 , the aerobic There are two major primary producer heterotrophs, is made up of birds, in­ groups in the system: emergent marsh sects, deer, and other herbivores in the plants (or angiosperms) and algae {benthic aerial component. The same compartment in algae, phytoplankton, and macro-algae). the water column consists of fish, amphi­ The emergent macrophytes appear dominant. pods, mussels, oysters, and bacteria. Algal standing crop, biomass, and produc­ Within the sediments, X3 is comprised pri­ tivity have not been quantified. Oata marily of meiofauna and bacteria. An does exist for the nearby mudflats and important difference between Pacific seagrass beds, but we feel the marsh algal Northwest and East Coast marshes is the production is much more like that of

37 AIR

f!! z :3 WATER 0. COLUMN :r: ~ --~~--1~~-­ < :E 0z :I>

., .. " # : • ~ • • .. • : .. . .: .. _... - ...... ' . , . . . .. ( ·•· ·,\ .. ~ :t ... : ~ ~ . . . .. 4 , ·~ • ~ J . - . .. .

i:igure 17. Conceptual model of carbon flow between components of tidal marsh ecosystems in the Pacific ft>rthwest.

the Middle Atlantic marshes than that of converted to a carbon base (approximately the geographically closer Pacific rt>rth­ 45% of dry weight), these 1 ast two data west tidal flats, seagrass beds, or south­ sets are similar to those measured by Gal­ ern California marshes. We would expect lagher and Kibby (in prep.). Kibby et al. it to average approximately lOOg C/m 2/yr. ( 1980) have depicted many of the pl ants in the Pacific rt>rthwest marshes and given Macrophytic seed-bearing p1 ants in production estimates (Table 14). They Oregon have a wide range of productivity al so have devised a series of field tech­ from lOOg C/m 2/yr to as much as lOOOg niques to be used in making rapid primary C/m2/yr (Gallagher and Kibby in prep.). production estimates of wetland sites. Others working along the Pacific flbrthwest coast have found similar ranges in primary As is the case in many middle Atlan­ productivity. Burg et al. (1980), for ex­ tic salt marshes, the location of plant ample, working in southern Puget Sound stands relative to marsh creeks affects found production ranging from 90 gram dry their productivity. Al though the aerial weight (gdw)/m 2/yr for a salt marsh sand­ growth rates of streamside and back marsh spurry (Spergularia marina) association to Lyngbye's sedge are similar, the seasonal 1390 gdw/m2/yr for the Lyngbye's sedge production is greater for the streamside association. Eilers (1975) measured net plants (Gallagher and Kibby 1981). The primary productivity ranging from 520 to streamside plants develop a canopy from 1940 gdw/m2/yr. When these values are underground reserves in the late winter

38 Table 13. Representative processes of carbon flow in salt marsh communities. ~umerical designations refer to direction of flow between the compartments of Figure 17; for instance, F (1,2) shows carbon flow from the atmosphere (Xl} to plants (X2} by the pho­ tosynthetic pathway.

Component Fluxes Ecological processes

Aerial F (1,2) Marsh plant photosynthesis

F (2,1) Marsh plant respir~tion F (2,3) Grazing (primarily by insects), pathogenic infection F (2,4) Plant death, fragmentation F (5,4)/(4,5) POC import, export (by wind, gravity, animals) F (5,3)/(3,5) Immigration, emigration (primarily mammals, birds, insects)

F (3,4) Fecal production and death F (4,3) Detrital consumption F (3,1) Microbial and animal respiration

Water column F (2,7) Leaching from leaves and stems (on site in marsh F (2,3) Grazing (possibly by insects in air pockets), and in tidal creeks) pathogenic infection F (7,5)/(5,7) Tidal DOC export and import F (7,4) DOC flocculation, sorption on POC F ( 4, 7) Lea chi ng from POC

F (5,3)/(3,5) Immigration or emigration of animals (fish, crustaceans), waterborne microbes F (7,3) DOC uptake (primarily by microbes) F (3,7) DOC release by excretion F (3,6) Microbial and animal respiration

F (3,4) Fecal production and death

(continued)

39 Table 13. Concluded.

Component Fluxes Ecological processes

Water column (cont.) F (4,3) Detrital consumption and microbial activity F (5,4)/(4,5) POC import and export by tidal water (often as litter) F {8,7) Excretion and extracellular release

F {8,4) Algal death, fragmentation F {2,4) Plant death, fragmentation F ( 6, l) co2 release

F (1,6) co2 solution F {9,6) Diffusion from sediments F {8,5)/(5,8) Transport by tidal water

F {8,6) Al gal res pi ration F (6,8} Al gal photosynthesis ------Sediment F {2,11) Leaching from roots and rhizomes F (11, 7) Leaching from substrate F (2,4) Root and rhizome death and fragmentation

F (4,11) Leaching {primarily from roots and rhizomes) F (11,3) DOC uptake {primarily by microbes)

F (11, 10) DOC uptake by anaerobic bacteria

F (10,11) Cellular release

F {4,3) Detrital consumption

F (3,9) Bacterial and benthic animal respiration

F (10,9) Anaerobic bacterial respiration and methanogenesis F {9,6) Inorganic carbon exchange

40 and early spring, a time when 1 ight and Table 14. Primary production estimates temperature conditions ordinarily are not of selected Oregon marsh plants (from optimum. The rapid spring growth from Kibby et al. 1980). underground reserves in roots and rhizomes correlates with the ultimate high annual production of these streamside sedges. Net When temperature and 1 ight become more primary favorable as the growing season pro­ Common Scientific production gresses, the streamside plants have a name name (g/mZ/yr) 1 arger canopy to intercept 1 ight and pho­ tosynthesize than do the back marsh plants. Although the growth rate per unit Lyngbye's Carex lyngbyei 1,850 of tissue is similar, total growth is sedge faster in the streamside plants where more 1 eaf area is present when growth condi­ Pickl eweed Sal icornia 1,660 tions become favorable. vi rginica These differences in the management Saltgrass Distichlis 1,300 of underground reserves may be coupled spicata with environmental factors such as better waterflow through the streamside soils Paci fie Potentil la 900 which results in the removal of toxic sub­ silverweed pacifica stances and the aeration of the root zone. The greater sediment deposit ion and Arrow-grass Triglochin 900 reworking rates which seem to occur on maritima some of the streambanks in the younger marshes may al so be important. These Three-square Scirpus 550 actions produce a fertilization of sorts americanus for the plants near the streams. Bal tic rush Juncus bal ticus 450 We found that growth increased in back marsh Lyngbye's sedge in response to nitrogen fertilization (unpublished data). Although they have not been tested, it is 1 i kely that at 1east some of the other of the season's primary production appears Pacific flbrthwest species would respond to be buried in the Fraser River delta in similarly. Other nutrients, light, tem­ southwestern Canada. perature, and salinity may be factors altering marsh production, but data to 4.1.2 Decomposition resolve this question are not available for the Pacific Northwest marshes. Gallagher et al. (in press, a) have measured the decomposition rates of marsh The major source of belowground car­ grasses in Oregon. Rates were comparable bon in most sys terns is from root and rh i­ to those measured by Gallagher and Pfei f­ zome growth. In southeastern Atlantic f er {1977) in the southeast Atlantic coast coast marshes, production ranges widely in spite of the higher annual temperature but may equal or exceed aerial net primary at the lower latitude site. In addition production (Gallagher and Plumley 1979). to cool winter temperatures which regulat­ In local situations where silt deposition ed decomposition, another major factor was from turbid freshets buries large quanti­ the moisture content of the dead pl ant ties of aboveground plant material in the tissue (POC). Those plants which had the sediment, aerial production can also con­ greatest percentage of supportive tissue tribute significant quantities to below­ decomposed slower than those that collaps­ ground carbon pools. We have not observed ed to the ground soon after they died. such sites in Washington or Oregon but Tissue nitrogen correlated with decomposi­ have seen situations where a large portion tion rate in several cases.

41 We have no direct evidence for the 10% of the plants are grazed; most biomass release of DOC frcxn living marsh plants in eventually enters the detritus foodweb Oregon and Washington. Gal 1 agher et al. (Pfeiffer and Wiegert 1981). Aerobic (1976) have demonstrated that the process heterotrophs which include insects and am­ occurs in smooth cordgrass marshes of the phipods feed on POC in the litter layer in Atlantic coast; it may be presumed to also the air, water column, and sediments. occur in pl ants in Pacific r-brthwest marshes. Higley and Holton (1981) have shown that there is a diverse community of in­ Leaching of DOC from the POC compart­ sects in Oregon marshes, but the flux of ment takes pl ace as the detritus ages or material through them has not yet been is broken down. In Oregon marshes rates measured. Insect studies indicated that of DOC leaching from eelgrass litter de­ Diptera and Homoptera (with large numbers creased more rapidly with aging in the Pa­ in the lower marsh) comprised most of the cific sil verweed marsh than in the pickle­ insects caught in sweeps in the marshes of weed zone. The consumption of POC of Coos Bay. Energy fl ow studies indicated either plant or animal origin by aerobic l eafhoppers {Cicadell idae) consumed 7 .3% heterotrophs may take the form of bacteri­ of saltgrass production with 2% being al decomposition. Gallagher and Pfeiffer assimilated; thus, assimilation efficiency (1977) found that the release of DOC to was 27.5% (Hoffnagle et al. 1976). Si!'1i­ the water column from standing dead smooth lar values are reported by Smalley (1960, cordgrass was inversely related to micro­ as cited in Hoffnagle et al. 1976) for bial respiration rates. This was inter­ insects of Atlantic coast smooth cordgrass preted as an uptake of labile DOC by the marshes. Therefore both insect assem­ bacteria. When microbial populations were blages of northwest marshes and their high, little DOC was rel eased into the energetics resemble those found in east surrounding water. Respiration was al so coast marshes. found to vary with plant _species. Micro­ bial respiration was greater on dead Hoffnagle et al. (1976) have studied smooth cordgrass than on dead black the role of marsh grass detritus in the needlerush. Christian et al. (1981) sug­ diets of the bay riussel (Mytilus edul is) gested that an equilibrium exists between and the Japanese oyster (Crassostrea microbial populations and product ion of ~). In order to simulate the natural substrate. DOC in estuarine water in­ diet, detrital particles were prepared by creases when the microbial populations are grinding dried plant material in filtered control led by predation by protozoa and seawater. The relative importance of dif­ detritivores; however, it never reaches ferent food sources for bay mussels (as levels high enough to be transported out measured by survival) is as follows: of the particular tidal stream in which it Lyngbye's sedge, saltgrass, pickleweed, a was produced. mixture of 4 plants (the above 3 plus bul­ rush), phytoplankton, bulrush, and eel­ Chris ti an et al. {1981) used oxygen grass. For the Japanese oyster, pickle­ uptake to measure microbial metabolism. weed appeared to be the most important Respiration rates of subtidal sediments followed by saltgrass, phytoplankton, bul­ and of soils in the area of tall smooth rush, lyngbye's sedge, mixed marsh plants, cordgrass were comparatively high {Q 1 o = and eel grass. 1.20) relative to soils in the short smooth cordgrass areas (Q 10 = 1.27); all 4.1.4 Food Webs varied seasonally. The low Q10 values indicate the predominant role that physi­ For many specific problems it is use­ cal rather than biological processes play ful to decompose small parts of the gener­ in the transfer process. al model into submodels. For example, food webs involving salmon ids are of com­ 4.1.3 Grazing mercial importance in the Pacific North­ west and have been described by a number In smooth cordgrass marshes less than of vmrkers including Northcote et al.

42 ( 19 79 ) from whom the fo 11 owing ex amp 1 es nile chinook salmon is fish, and in the were taken. At least six fish species and case of peamouth chub adults, 19% of their size classes (leopard dace, largescale diet consists of benthic suspension feed­ sucker adults, prickly scul pin juveniles, ers and 25% of fish. Of course the ben­ mountain whitefish, juvenile chum salmon, thic suspension feeders and the fish (via and starry flounder juveniles) of the low­ zoopl ankton) are dependent on suspended er Fraser River estuary may be included in detritus coming from the benthic detritus a food ~1eb consisting of three compart­ source. ments: The stomach contents of fish collect­ ed at Netarts and Siletz Bays were analyz­ benthophagous ed by Higley and Holton (1981). Terres­ detritus - benthos fishes trial prey, including adult insects and spiders in small quantities and dipterous larvae and pupae, were consumed only by In contrast, the 1 argescal e sucker ( <300 the chum salmon. Chum salmon al so fed on mm), chinook salmon (<50 mm), threespine flatfish 1 arvae and cumaceans (Hemil eucon stickleback, peamouth chub (<75 mm), red­ spp.). Staghorn sculpin, threespine s ide shiners, and juvenile sockeye salmon stickleback, and juvenile chum salmon con­ are involved in a more complex food web. sumed primarily aquatic animals, amphip­ Terrestrial insects and northern squawfish ods, harpact icoid copepods, cumaceans, juveniles contribute up to 43% of the diet ol igochaetes, and polychaetes. Starry of the latter two species. The resulting flounder ate mostly decapod 1arvae, adult foodcha in is: Callianassa, and amphipods. The major food of the rest of the fish in the 1arge marshes and adjoining habitats consisted detritus--• benthos benthophagous of amphipods, isopods, tanaids, poly­ 1 fishes chaetes, cumaceans, copepods, dipterous larvae and pupae, and fish. zooplankton and/or terrestrial The tidal creeks in the Skagit River insects salt marsh of Washington's Puget Sound provide feeding grounds for 1 arge numbers Where sma 11 fishes are prey the food of chum and chinook fry (200 to 800 fry web becomes : per 100 m in April) (Congleton and Smith 1976). Their diets consist mostly of 1 argely or Corophium, harpacticoid copepods, and in­ detritus- benthos--.. partially sect adults, larvae, and pupae. The high benthophagous density of fry and the fact that most were fishes f collected with full stomachs supports the idea that this marsh serves as an impor­ zoopl ankton / I tant feeding area. In addition, Cong1eton and/or small (cited by Meyer 1979) has found that prey terrestrial insects - fishes numbers increase in the marsh by 1 ate March before the epibenthic population in Prickly scul pin (>50 mm), white sturgeon Skagit Bay has had a chance to build up. ( 125 to 720 mm), starry flounder (>100 Thus, the marsh provides food for salmon mm), stag horn scul pin (100 to 300 mm), juveniles during the period of time that northern squawfish (>150 mm), Dolly Varden food reserves in the bay are in short sup­ ( 181 to 430 mm), and peamouth chub ( 75 to ply. 150 lilll) are involved in this food web. 4.1.5 Uses of the Marsh Chinook salmon juveniles {50 to 110 nm in length) and peamouth chub adults are The most extensive work on the utili­ not strictly benthic detritivores. A zation of tidal marshes of the Pacific 1 arge portion (32%) of the diet of juve- lt>rthwest by various fish has been con-

43 ducted by researchers at the Westwater detritus based. Estuarine fish of the Research Center, University of British major anns of the Fraser River rely pri­ Columbia in Vancouver. Most of the work marily on the benthos, i.e., secondary has been done on the marsh creeks of the production. Most benthic invertebrates in Fraser River Estuary. Only 32% of the turn feed on the detritus entering the Fraser's origina1 wetland area has not estuary from terrestrial sources upstream been diked, and there is a great deal of in the watershed and from adjacent interest in learning the value of the wet- marshes. Marsh vegetation occurs along 1ands in order to protect them from fur­ the islands, sloughs, and side channels, ther urban-industrial development. They but the diked river banks lack large are located on the Pacific Flyway and are amounts of macrophytic vegetation. r-brth­ therefore thought to be critically impor­ cote et al. (1979) felt that in temperate tdnt to migratory waterfowl as wel 1. In areas such as the Pacific Northwest feed­ order to assess whether or not marshes act ing on the detritus itself would prove un­ as rearing grounds for juvenile salmon, reliable for the fish since the amount of Westwater Research Center scientists detritus in the river and estuary varies studied fish diets, the utilization of the seasonally. In tropical estuaries, where estuary, and habitat characteristics asso­ organic matter input is more constant, ciated with high numbers of salmon. Most fish do feed mainly on detritus. of the remainder of the discussion in this section is based on results reported in ~rthcote et al. (1979) analyzed the Westwater publications. stomach contents of 21 species of fish collected over a one-year period on the Chinook salmon spawn the 1ength of mainstream lower Fraser River. The bulk the Fraser River as far as 1000 km up­ of the diets of Most of these fish came stream from the estuary. There are two from the benthos. Chironomid larvae were types of chinook adults: smaller (~15 lb) a dominant food item for most of the resi­ red-fleshed fish that return from the dent species as well as for the juvenile ocean during the summer, and 1 a rger (~ 20 salmon migrants. Terrestrial insect pro­ 1b) white- fl es hed fish that return from duction was utilized only by the sockeye the ocean during the fal 1. White chinook ju ven i1 es. As they migrated downstream, spawn in the lower Fraser River while red 90% (by weight or numbers) of their diet chinook spawn in the upper river trioutar­ consisted of insects, mainly dipterans. ies. Those spawning in the upper river In general. diets did not vary greatly produce fry which remain in the freshwater seasonally. The fishes of this area uti- tributaries for either 3 months or one 1 ize many different prey types, as do many year. These fry spend little time in res­ north temperate fishes. tt>rthcote et al. idence in salt marshes on their way to the (1979) identified 21 prey classes. A giv­ ocean. However, the recently hatched en species may use up to 12 of the possi­ white chinook fry of the lower river do ble 21 classes, but an individual of that use the marshes as rearing grounds (Levy species restricts its prey to between 1.2 and tt>rthcote 1981). Levy and lt>rthcote and 3.5 prey classes, probably a result of suggest that perhaps the difference in particular prey being most abundant in a marsh utilization stems from the proximity localized area. of the spawning site to the estuary. Low­ er river chinook might have a higher rate Table 15 ranks prey items of the dom­ of survival if they spend a period of time inant fishes of the Fraser's Woodward Is­ in the productive marsh waters. On the land marsh tidal channels. Few salmon fry other hand, the chinooks from further up were found in the stomachs of fish preda­ the river might benefit more from a longer tors in the tidal channels. Although fish residence time in the freshwater which are important in the diet of the 1 arge would prepare them for their longer and staghorn scul pin, salmon fry made up only more rigorous journey down the turbulent a small proportion of their diet. Perhaps river. the marsh habitat is providing the young salmon with refuge from predators (Levy et The food web of the Fraser Del ta is al. 1979).

44 Levy et al. ( 1979) al so related prey fry (43 mm mean length) fed on both har­ preference to predator size. The diet of pacticoid and insect pupae; and the 1 arg­ the smal 1 pink sal non fry ( 34 mm mean es t fry, the Chi nook ( 49 mm mean 1 ength), length) collected in Fraser Marsh creeks fed mostly on insect pupae. consisted mostly of harpacticoids; chum

Table 15. Relative importance of prey items of the dominant fishes of Woodward Island marsh channels. Ranking index = (frequency + volume percentage) percent occurrence. (modified from Levy et al. 1979).

Order of relative importance* Species 1 2 3 4 5

Anadromous species

Chinook smolt a g h c b

Chun fry h f i c g

Pink fry f h i c g

Sockeye fry g h b c Chinook fry h ; g d c

Long fin smelt c b d e Marine species

Staghorn scul pin c e b a d Starry flounder e c d h

Pacific herring b c Freshwater Species

Threespine stickleback i c f b h

Prickly scul pin b e i c d

*a = Fish. b = Mysid shrimp (Neomysis mercedis). c = Amphipod (Anisogammarus confervicolus). d = Amphipod (Corophium spinicorne). e = Isopod (Gnorimosphaeroma oregonensis). f = Harpacticoid copepod. g =Insect adult. h = Insect pupa. i = Insect larvae.

45 Occurrence, distribution, and resi­ weight gains and the fact that the pink dency patterns were obtained for the domi­ salmon were the first to leave the marsh nant fish species of the marsh tidal chan­ tidal channels while the chinook remained nels of Woodward Island in the Fraser the longest, indicate that juvenile pink River by Levy et al. {1979). Fish were salmon did not have extended marsh resi­ trapped in these marsh channels when the dency. On the other hand, chum and es pe­ tide began to ebb. Captured salmonids ci ally chinook salmon did. were marked with fluorescent grit and used in recapture studies for residency deter­ The Fraser estuary contains 3,563 minations. During 124 sampling days, 31 hectares (8,801 acres) of marsh. As dis­ species of fish were collected. Juvenile cussed previously, juvenile chum and chi­ salmon (pink, chum, and chinook) were nook salmon use these areas as rearing taken between February and August but were grounds. Hundreds of mil 1ions of these most abundant between March and June. two species migrate down the river annual­ Five species occurred most consistently: ly, and huge numbers of them are present longfin smelt, starry flounder, staghorn in the marshes between March and June each sculpin, threespine stickleback, and pea­ year (levy and Northcote 1981). In an mouth chub. Differences in maximum densi­ effort to protect at least the most valua­ ties and timing of the seasonal peak ble parts of the marsh from further devel­ occurred between channels. opment and alteration, it is necessary to know what habitat characteristics provide Levy and Northcote (1981) were able for the richest and most used rearing to describe two fish communities occupying grounds. Levy and Northcote (1981) found tidal channels of the Fraser marshes. that consistently low catches of chinook Salmonids, sculpins, and starry flounders fry were caught in certain marsh tidal were associated with large tidal channels channels. To detennine which factors are while sticklebacks and peamouth chubs important in attracting these fish to cer­ occupied those tidal channels far into the tain other marsh channels, Levy and North­ marsh. levy et al. (1979) found that dis­ cote (1981) measured 22 habitat character­ tribution within the tidal channels al so ; sties. They found that 12 of these were varied. Sticklebacks were concentrated significantly correlated with the number far into the tidal channel whereas stag­ of chi nook fry captured. The significant horn sculpins and starry flounders were characteristics were: mouth width, sta­ most abundant near the mouth of the chan­ tion width (width of channel at sampling nel. The juvenile salmon were abundant at site), channel length, channel order (a both ends of the channel. On ebbing tides ranking based on a series of 22 possible the first of the salmon to emigrate from features), sub-channel length, mean angle the tidal channels were the pink fol lowed of the channel bank, tidal slope, high by the chum; the chinook remained in the tide height, tidal channel area, bank ele­ channel the longest. Juvenile salmon, vation, area of subtidal refugia, and the especially chinook, migrated al so between number of hours of tidal channel submer­ channels. gence prior to samp 1i ng. Many of these variables are highly correlated with each Size of the juvenile salmon changed other. over the sampling period. Juvenile chi­ nook showed the greatest increase in aver­ I t may be poss i b1 e to pi npoi nt the age size. Juvenile chum showed a small best rearing ground areas using the above increase and juvenile pink showed no infonnation. For example, bank elevation change. was inversely related to the catches of chinook fry, indicating that the fry are Large numbers of juvenile chinook more likely to utilize low elevation marsh were recaptured after 1 month, while few areas {levy and Northcote 1981). pink salmon were recaptured. Recapture rates of juvenile chum salmon were inter­ Chinook juveniles varied in size as mediate (Levy et al. 1979). These recap­ well as in recapture rate in various marsh ture patterns, coupled with differential tidal channels in the estuary. During

46 their survey, Levy and Northcote (1981) dents? If survival rates of estuary resi­ found that the tidal channels of Hoodward dents prove to be better than the others Island were the site of the chinook having and if the marshes are found to be under­ the largest size, the highest recapture utilized, then perhaps artificially pro­ rate, and the greatest abundance. Wood­ duced fry could be given a better chance ward Island al so has the largest area of of survival by being released directly rearing habitat - the greatest marsh tidal into the marsh rather than at sites fur­ channel and backwater area. It is there­ ther up the river (Levy and Northcote fore considered to be the most important 1981). rearing ground in that area of the estuary. 4.2 MINERAL CYCLING IN THE MARSH The Sixes and Rogue Rivers lack the Mineral cycling in marshes of the marshes characteristic of the Fraser River Pacific Northwest has not been studied ex­ estuary. In the Sixes River in Oregon, tensively. Gallagher and Kibby (1980) Reimers (1973) found that chinook reared have measured fluxes of trace metals from in the estuary had a higher survival rate soil to plants, but the turnover was not to the adult stage than did those reared followed through the rest of the foodweb. outside the estuary. Shoaling occurs at Tjepkema and Evans (1976) have measured the mouth of that estuary and inhibits rates of nitrogen fixation in various flow to the ocean. This causes low shore­ plant species. Their studies focused on lands to become inundated and, with the Bal tic rush, but tests with tufted hair­ entrapment of nutrient-rich ocean water, grass, seaside plantain, and pickleweed increases the productivity of the estuary also produced positive results. As a (Ratti 1979). result, Tjepkema and Evans (1976) specu­ lated that the association between nitro­ Opposite results were found in the gen fixation and wetland plants is wide­ Rogue River in Oregon. Twenty years ago, spread. Although the bacteria were pri­ extensive shoaling occurred at the mouth marily associated with the rhizomes and of the Rogue River which encouraged juve­ large roots, neither their nature nor nile chinook to spend considerable time their exact location is known. We have no rearing in the Rogue Estuary. Now, due to reason to believe that processes of nitro­ the jetties and channelization at the gen fixation by anaerobic bacteria and de­ mouth, the chinook spend less time rearing nitrification are proceeding any differ­ there (Ratti 1979). ently than in Atlantic coast marshes. The Pacific Northwest soils range from organic The effects of shoaling and associat­ to mineral and are often saturated with ed increases in productivity in the Sixes saline water. Physical conditions are may produce some of the same effects as similar; rates, but not types, of pro­ occur in the Fraser. Perhaps the Sixes cesses probably differ. and the Fraser estuaries are best suited for those types of salmon which benefit Ammonification and nitrification from the use of such productive marsh likewise have not been evaluated, nor have areas and therefore spend time there on various aspects of the phosphorus cycle. their journey to the ocean. Data collected on gradients of phosphorus in Netarts Bay led us to believe the tidal Questions that still need answering marsh may be exporting phosphorus during include: do those fish that reside in the the summer months. Similar measurements estuary and its tidal marshes for some with nitrogen fonns indicated an import period of time have better survival rates from nearshore upwelling. The nitrogen, than those that don't?; what are the popu­ sulfur, and phosphorus cycles in marshes lations of those areas of the tidal have been recently summarized by Whitney marshes not yet sampled?; what proportion et al. (1981) and Wiebe et al. (1981) for of the migrating fry reside in the estua­ Atlantic coast marshes. There is no rea­ ries' marshes?; which of the river's son to suspect that they are greatly dif­ stocks contribute to the estuary resi- ferent in marshes of the Paci fie North­ west.

47 4. 3 I ITTERACTI ONS BETWEEN THE MARSH AND ~ith mammals. We have observed deer graz- ADJACEITT ECOSYSTEMS 1ng some marsh plants such as arrow-grass t~fted hairgrass, and sal tgrass. Upland One of the important features of birds may feed on marsh insects and marsh tidal marshes is their ability to interact plant seeds, although documentation for readily with adjacent ecosys terns. This is these pathways is scanty for Pacific made possible by the flooding and ebbing f'brthwest marshes. There is additional of the tidal waters which sol ubil ize many nutrient interaction associated with run­ substances and carry them to and from the off of rainwater which carries dissolved marsh. Because water is dense, it can nutrients and particulate matter from the carry particulate materials ranging in higher elevation system to the next lower size from microscopic detritus to Doug- one. Direct exchanges al so occur between 1 as-fir 1 ogs. The drift 1 ogs frequently uplands and tidal flats or stream channels strand and accumulate in many parts of the when either directly abuts high ground. upper intertidal zone; they are a major Where the topography is steep, direct feature of Pacific rt>rthwest marshes. interactions are more common. Because of their slow turnover rate and large mass, logs can be a long-term dis­ Exchanges between tidal marsh and ruptive force as they move during major tidal· flat are primarily in the form of storm events. DOC and POC from marsh plants and dissolv­ ed nutrients leaching from them and possi­ bly from the soil. Tidal flats may export Figure 18 represents a conceptual dissolved organic and inorganic compounds model of some of the major flows of energy to the marsh on rising tides. Eelgrass and nutrients between coastal uplands, and macroalgae perform the role of marsh tidal marshes, tidal flats, and subtidal pl ants at the tidal fl at elevation. In­ stream channels (the part of the estuary vertebrates at the sediment-air and sedi­ always containing water) in the Pacific ment-water interface and within the sedi­ rt>rthwest. Like all conceptual models, it ment play a much more important role on is a simplification of reality. We have the flats than in the marsh. As in all not specifically shown a microbial com­ ecosystems, fungi and bacteria play a partment but have assumed they are ubiqui­ pivotal role in the degradation process of tous. Documentation of the magnitude of whole plants to inorganic nutrients fluxes is poor in all cases. Soil, water, through POC and DOC. and air zones are shown where appropriate; we depict the major biotic ecosystem com­ Tidal flats probably interact more ponents as being part of one or several of with the subtidal stream channels than these zones. Interactions between each with the marshes because tides connect zone connected by water are shown with them daily whereas tides carry materials fused arrows to represent an easy exchange to the marshes less frequently. Although because of the water. Connections with the extent of this interaction has not coastal uplands are represented with been precisely quantified, there is a closed arrows because exchanges are not as rather 1arge export of eel grass and, to a readily accomplished. 1 esser extent, macroal gae from the tidal flats to the marsh (Gallagher et al. in press, b). The amount of eelgrass being The principal direct exchanges be­ exported to the marsh is greater than tween the coastal upland and the tidal would be expected on the basis of normal flats and stream channels are through the tidal interactions; much of the eelgrass movements of birds and mammals from the is broken loose from the beds by wave upland which feed in the lower zones on action during storms when tides are high fish and invertebrates. Their excrement in the fa 11 • These events return a po r­ may represent a transport in either direc­ t ion of the nutrients lost from the marsh t ion depending on where forag i ng and ex­ by tidal exchange and, to an extent, close creting occur. Between the upland and the the cycling within the tidal fl at/marsh tidal marsh similar exchange may occur complex and reduce the loss of energy- and

48 Figure 18. Conceptual model of energy or material flow between tidal marshes and other coastal ecosystems in the Pacific l't>rthwest. I nutrient-rich compounds to the subtidal either by bacteria which are later consum­ stream channels. ed by invertebrates or directly by inver­ tebrates which a re food for various spe­ Exchanges from tidal flats to sub­ cies of juvenile salmon. The residence tidal stream channels occur twice daily. time of the fish in the marsh-bordered The interactions involve the full spectrum estuary depends on the species. of materials from nutrients to fish on their way to feed in tiny marsh creeks. These ecological interactions within Based on the previously discussed infonna­ the tidal marshes and between the marshes tion from the British Columbia estuaries, and adjacent ecosystems must be understood it seems probable that the major role before wise decisions about wetlands man­ played by the wetlands in this coastal agement can be made. It is not enough fisheries food web is to provide a nursery just to know what organisms will be di­ ground. That role is providing a shelter­ rectly affected. The second and third ed habitat and producing POC and DOC which order interactions may be the ones on enter the stream channels to be utilized which the decision should be based.

49 CHAPTER 5

MAMGEMENT

5.1 SENSITIVITY TO W\TURAL AND CULTURAL red-fleshed chinook (Levy and Northcote PERTURBATIONS 1981). Levy and Northcote (1981) also suggest that it may be possible to use the There is concern that the true value resident juvenile population in the estu­ of the tidal marsh to estuarine productiv­ ary as an index of the abundance of ity will be realized too late - after many returning adult chinook stock. of the remaining marshlands are destroyed by landfills and development. Only re­ Many marshes, diked or undiked, are cently have investigators begun to study grazed by domestic animals; these prac­ marsh utilization by salmon and other tices can al so alter the system apprecia­ fishes and assess quantitatively the sig­ bly. N:>t only is pl ant production of nificance of this habitat to western fish detritus reduced, but nutrient cycles are stocks. In the Fraser River estuary most altered and the soils physically compacted of the fish populations are dependent on a during grazing. The consequence of these relatively small array of benthic inverte­ alterations has not been thoroughly brates, many of which are tidal marsh assessed in the Paci fie Northwest. The creek inhabitants (N:>rthcote et al. 1979). role of the marsh in absorbing stonn tides Altering the species composition of the is little changed, and the grazed, undiked benthos with various land or water manage­ marsh continues to perform all its func­ ment projects which would change the tions in the coastal zone al though in a supply or timing of detritus fl ow from somewhat altered form. upstream could be very detrimental to the food supp 1y of the fishes downstream Logging operations near the estuaries (Northcote et al. 1979). may also lead directly to damage when marshes are used as log storage areas. Di king the marshes has been a long Escaped logs may end their journey on practiced method of increasing agricultral marshes; their effects may be long- land but one which may decrease fisheries. 1asting. (With decomposition half-1 ives Levy and l'brthcote (1981) concluded that measured in decades, logs can smother juvenile salmon use the entire length of marsh plants and block creek channels tidal channels and, therefore, that diking thereby changing the hydrology of the even a portion of a marsh area can signif­ systems. Figure 19 shows a creek filling icantly reduce the estuary 1 s capacity as a in as a result of the channel being rearing ground. Recall from Chapter 4 blocked by a log. As the siltation pro­ that the white-fleshed chinooks of the cess continues here, the Lyngbye's sedge Fraser River reside in the estuary for a marsh wil 1 probably completely close the period of weeks or months before migrating gap between the two sides of the fonner to sea while the red-fleshed chinooks, channel. originating further upstream, spend a 1onger time in the freshwater and are in Marshes tend to be very sensitive; the estuary only briefly. Estuarine popu­ even paths made by scientists traveling to lations of juveniles make an important study sites may be visible several years contribution to the commercial and recrea­ after the study ends. tbt only are the tional catch of white-fleshed chi nook and pl ant stems themselves crushed, but after probably to some extent to that of the several tr,ips, the rhizor:ie mat in the soft

50 Figure 19. Lyngbye's sedge marsh invading a mudflat created by a log blocking a marsh creek.

sediments is damaged. Reinvasion of these because of the soil changes, different damaged areas appears to be surprisingly drainage patterns, and grazing (Jefferson slow. In the lower edges of the sedge 1975; Mitchell 1981). marshes, the primary colonizer may be arrow-grass which is al so a common invader Salt marshes have been altered by on tidal flats adjacent to low silty fil 1 ing and spoil disposal as wel 1. In marshes. comparison to the relatively limited extent of the coastal estuaries. a trenen­ dous acreage of intertidal area has been 5. 2 HISTORICAL MANAGEMENT APPROACH filled in the Pacific ttirthwest. Accord­ ing to the U.S. Amy Corps of Engineers Historically, man's greatest ir.tpact (USAGE 1976, cited by Proctor et al. on Pacific Northwest tidal marshes has 1980), 1,554 ha (3,840 acres) in Grays been caused by diking in an effort to Harbor have been used for dred9e material expand land usable for agriculture. Salt disposal. A total of 283 ha ( 700 acres) marshes were diked and drained by the in the Columbia River Estuary have been earliest white settlers and used as crop fi11 ed (Sear.ian 1977, cited by Proctor et and pasture lands. Almost all old, high al. 1980). A large percentage of the marsh areas of Oregon have been diked original marshland of the Skagit River {Jefferson 1975). Today many of these delta in Puget Sound has been destroyed or areas are stil 1 used for grazing (Figure altered. Today less than 203 of these 20). and marsh hay is stil 1 cut. Once the marshes remain (Congleton and Smith 1976). marsh is diked, the soi1s compact and About 40% (2,509 ha, 6,200 acres) of become aerobic; the soil salinity gradu­ emergent marsh of Will apa Bay has been ally decreases. The plant community altered by diking and filling (USACE 1976, slowly changes from a halophytic plant cited by Proctor et al. 1980). Hoffnagle community to a damp meadow community and Olson (1974) estimated that 903 of the

51 Figure 20. Diked (upper) and undiked (lower) marshlands which are being used for pasture.

52 original salt marsh of Coos Bay has been Table 16. Landfills on 1 ands lying be­ altered by diking and filling. Landfills tween the line of ordinary high water and in other Oregon estuaries are sur.imarized the line of ordinary low water in selected in Table lG. Oregon estuaries (modified from Percy et al. 1974). Logging practices have ~so taken their toll on the ~brthwest marshes. Hoffnagle and 01 son (1974) have described the effects of logging on the Coos Bay Estuary Acres Use salt marshes. Eros ion due to cl ear cut­ ting and slash burning increased sedimen­ tation in the bay; this in turn led to Al sea 24.8 Marine oriented/ increased dredging and the dur:1pi ng of recreation dredged material on marshes. Logging also results in the accumulation of logs and Nehalem 7.3 Residential I woody debris. recreation

Only a very small percentage of the Nes tucca 0.7 Erosion control for tbrthwest's original marshes remain resident i a 1 property unchanged, especially in the highly devel­ oped areas. The slow, natural prograda­ Salmon 0.1 Parking area/boat tion of some of the marshes cannot make up 1 au nch for the historical destruction of this valuable resource. Sandlake 3.0 Diked, agricultural

While wetlands have been perturbed, Siusl aw 40.6 '·1a ri ne-or i ented altered, and destroyed since the area was industry first settled, the history of their pro­ tection is much shorter. Bean ( 1978) sur.1- Tillamook 102.1 Industry marized the role of the Federal Goverrment in wetland protection. In cooperation Umpqua 97.5 Marina and harbor/ with the States, the Federal Goverll!lent marine-oriented deep­ began acquiring wetlands in 1929 with the water navigation and Migratory Bird Conservation Act. About industry this same time, however, the Federal Gov­ errtnent started a huge hydroelectric Yaquina 202.1 Marine-oriented deep­ development program with the enactment of (below water navigation and the Federal Power Act of 1920. This, Toledo) industry along with the Flood Control Act of 1936, essentially cancelled the wetlands preser­ TOTAL 478.2 vation benefits of the Migratory Bird Con­ servation Act. In an effort to reduce the adverse effects of the 1920 and 1936 Acts on wildlife, the Fish and Wildlife Coordi­ nation Act, originally passed by Congress the Federal Water Pollution Control Act in 1934, was amended in 1958 to include ,drnendments of 1972 (Public law 92-500), the mitigation concept. In essence, this the Federal Govermient became the princi­ meant that if public hunting and fishing pal regulator of the devel Oi)llent of wet- were disturbed due to a Federal water 1 ands. Under Section 404 of these llr.lend­ development project, private lands of ments, the Environmental Protection Agency similar recreational quality must be pur­ and the U.S. Anny Corps of Engineers have chased to replace those lost. the responsibility of setting wetland boundaries and protecting navigable waters Until the late 1960's the Federal by protecting some wetlands. Pennits must Goverrment ignored private activities be issued by the U.S. Anny Corps of Engi­ affecting wetlands. With the passage of neers for the discharge of dredge or fil 1

53 material onto wetlands contiguous with and t'ie transition zone from undisputed navigable waters or into the waters them­ wetland to upland is gradual (Figure 21). selves. The Coastal Zone Management Act In _this case the answer is not easily has made Federal funds available to defined, and the resolution becomes a encourage codstal States to develop policy decision rather than a scientific management programs for their coastal determination. zones and resources. The State may then be eligible for Federal grants-in-aid to The U.S. Anny Corps of Engineers and put their management programs into effect. the U.S. Environmental Protection Agency One of the remaining critical problems in have the responsibility to develop guide- tidal marsh management is locating the 1 ines by which wetland boundaries may be upland marsh boundaries as indicated in delineated for determining where permits Sect ion 404 of Public Law 92- 500. are required. Several approaches have been proposed and taken by various inves­ tigators and agencies to delineate wet- 5.3 UPLAI{) MARSH BOUNDARIES 1ands. Some states define intertidal wet­ lands with respect to a tidal datum, e.g., It has been realized in recent years mean low water (Wass and Wright 1969; that the protection of wetland ecosysteris Virginia Code, Sec. 62. 1-13,2). After an is of paramount importance because they investigation of eight coastal marsh are very productive, are used by fish as sites, the National Oceanic and Atmos­ rearing and feeding grounds, serve as pheric Administration (NOAA 1975) sug­ stonn buffers, and improve water quality gested the upper 1 imit of the marsh be of adjacent strea17t channels. In order to defined at 0.76 m (2.5 ft) above mean high protect a wetland from intrusions such as water (MHW). With further study this cri­ filling, diking, draining, and pollution, terion was found to be satisfactory only the limits of that wetland must first be for the Atlantic and gulf coasts but un­ defined. Herein lies the difficult tech­ satisfactory for the West coast, where the nical problem of quantifying the 1egal ecotone spanned elevations of up to one definition. meter. NOAA therefore set the upper marsh 1 irni t as being the average of the upper In the United States, regulations set and lower limits of the transition zone, forth by the U.S. Anny Corps of Engineers which in turn was based on floristic com­ define wetlands as: position • • • • areas that are inundated or Floristic criteria involve basing the saturated by surface or ground water upper wetland limits on the presence of of a frequency and du ration suffi­ certain species. Frenkel et al. (1978) cient to support, and that under identified the transition zone in Oregon normal circumstances do support, a by a strong dominance of Pacific silver­ prevalence of vegetation typically weed and the presence of Achillea mille­ adapted for 1 ife in saturated soil fol ium, Angelica lucida, Aster subspica­ conditions. Wetlands generally in­ tus, Oenanthe samentosa, marsh clover, clude swamps, marshes, bogs, and and Vici a gigan~Some statutes specify similar areas. (USACE 1981) a 1 ist of indicator pl ants which may be used to identify a wetland (New York Envi­ In the field, this definition alone be­ ronmental Conservation Law, Sec. 25-0103, comes inadequate. In those situations cited by Lagna 1975). Frenkel et al. where the trans it ion from the wetland to (1981) used a combination of floristic and the upland is abrupt, it is relatively vegetational (pl ant coverage) criteria to easy to determine the upper 1 imit of the establish lists of upland and wetland marsh s i nee the change from wetland so i1 s plants and to develop a quantitative meas­ and vegetation to upland types occurs in a ure integrating floristic and vegetational short distance (Figure 21). However, the data. They could thus detenni ne up1 and problem of cl ear detennination of the limits by shifts in plant community struc­ upper limit becomes very difficult in ture. areas where the grade of the slope is low

54 Figure 21. Transition from wetland to upland. The transition zone in the upper photo­ graph is sharp and the transition from low silty rnarsh to 'Jpland takes place in a few meters. The left hand side of the lov-1er photograph depicts a gradual transition from a low sandy marsh to upland. The transition is spread over 20-30 m.

55 Remote sensing techniques have al so The USACE Waterways Experiment Sta­ been proposed as a means of defining wet- tion's Wetlands Group is currently recom­ 1ands for jurisdictional purposes (Reimold mending a three-faceted approach to deter­ et al. 1972). Different species of plants mining wetlanSands, in the Columbia a id in setting wetland l 1mits. She found River Estuary (39 km or 24 mi from the morphological and anatomical differences mouth) has been pl anted with :narsh pl ants within species depending on the moisture (sedge, bulrush, Juncus effusus, Carex reg ir.1e under which the pl ants were growing obnupta, and tufted ha i rgrass) (Kirby in the field. Pl ant height changes, 1978; McVay et al. 1980). degree of branching and flowerin'J, and amount of aerenchymous (air conducting) Oregon has one of the inost rigorous tissue varied with soil moisture content. fill and removal laws in the nation. The Root distribution within the soil also Oregon Land-Use Program is based on a varied with soil moisture. A larJe per­ series of legally enforceable "goals." A ce11tage ( _:.:,65%) of root and rhizome bio­ requirenent of their Estuarine Resources mass in the upper 10 an of a soil core Goal states that: served as ,rn indicator of high soil moi.s­ ture for four salt marsh pl ant species When dredge or fill activities are t~st:~d (Seliskar, in press). These vari- pennitted in intertidal or tidal 011s chan9es are most useful after a gross marsh areas, their effects shall be 1 imit has been set and a more precise mitigated by creation or restoration Jel ineation is needed. of another area of similar biological

56 potential to ensure that the integ­ merged time equival ence 11 method lo rel ate rity of the estuarine ecosystem is surface area and time submerged was devel­ maintained. (Oregon LCDC 1976) oped at Oregon State University. As a result, 26 to 28 ha ( 65 to 70 acres) of diked marsh was determined as the proper Guidelines to the above requirement area to be restored to mitigate the 1oss (Oregon LCDC 1976) state that, if possi­ of 13 ha (32 acres) of sandy bottom (LaRoe ble, the mitigation site should be near 1978). Taylor and Frenkel (1979) pre­ the dredge or fill site in order to be the dicted that approximately 12% of the miti­ most similar ecologically. If this is not gation site, that lying below 1.4 m possible, then other areas with similar (4.6 ft), elevation, would become tidal (in order of importance) salinity, tidal mudflat. They expected that in 40% of the exposure and elevation, substrate, current mitigation site, the area between the ele­ velocity, orientation to solar radiation, vations of 1.4 and 2.4 m (4.6 and 7.8 ft), and slope characteristics should be se­ the diked marsh vegetation would be re­ lected for mitigation. If neither of placed by those species nonnally found on these aforementioned mitigation site cri­ silty substrates. The area above 2.4m teria can be met, then the restoration of (7.8 ft) (48% of the site) was thought to areas or resources of the greatest scar­ be suitable for the development of high city compared to past abundance must be salt marsh. To date, neither the filling undertaken. Dredge material isl ands, nor the creation has taken place. diked marshes and those wetlands separated from ci rcul at ion by causeways, or other Perhaps the most powerful tool to fills are suggested as potential mitiga­ have recently received widespread accept­ tion sites. LaRoe (1978) stated that the ance in solving problems relating to tidal Estuarine Resources Goal was carefully marsh use is the mitigation procedure. conceived and worded and is intended to Al though the price of rehabilitating prevent any further net loss of intertidal wetlands degraded in a previous era or habitats in Oregon. converting upland to marsh is high, the economic value of being able to destroy As an example, the mitigation re­ certain wetland sites during estuarine quirement was first tested when the city resource development is immense, and high of North Bend requested a pennit to fil 1 a mitigation cost can usually be justified 13 ha (32 acres) sandy bottom area of Coos economically. Judicious application of Bay. The mitigation site chosen was a this procedure makes it possible to con­ third priority type - diked marsh in a tinue economic growth in the coastal zone different part of the estuary. The ques­ without destroying the natural resources tion of how much marsh to mitigate for which make that economic development pos­ 13 hectares (32 acres) of sandy bottom was sible and which drew people to the coasts a difficult problem (LaRoe 1978). A "sub- in the first place.

57 CHAPTER 6

SUMMARY ANO CQ'1PARISONS

Tidal marshes of the Pacific North­ fie Northwest, as along the Atlantic and west do not fonn a continuous band paral- gulf coasts, the addition of new sediments 1 el to the coast as do the wetlands along to the wetlands has been decreased by up­ much of the Atlantic and gulf coasts. river dams, channelization, and diking of These discontinuous marshes may develop on river banks. Sed irients are often trapped the intertidal flood plains of estuaries in upstream reservoirs or sent offshore and may extend miles inland. Alternative­ rather than being trapped in the coastal ly they may arise along the quiet borders zone. In those areas where spring fresh­ of sandspits which have enclosed anbay­ ets bring new sediments to the marshes, ments along the rocky coast. Tidal the nutrient cycles which take place in marshes arising in riverine systems tend the decaying peat are subsidized by the to be fresher than those behind sand bars seasonal input of nutrients. The coastal since they are fed by rivers or streams upwelling phenomenon, which is not found which extend well into the Coastal Range on the Atlantic and gulf coasts, provides or even into the Cascade Mountains. a mechani~n for bringing nutrients in deep coastal water back to the Paci fie estu­ Snov.anelt from the Cascades produces a aries via flood tides. large freshwater pulse into the Columbia River and Puget Sound which may last sev­ Like the Atlantic coast, the tides of eral months in the spring. In the Colum­ the Pacific rt>rthwest have two highs and bia and many other rivers, the storage lows daily, but unlike those in the east, capacity of hydroelectric dams has se­ they are vastly unequal in height. This verely curtailed freshwater runoff during complex pattern of flooding produces a snownel t or periods of high rainfal 1. high diversity of environmental gradients Where pulses of freshwater do occur, marsh in moisture and salinity that affects both soil salinities may be seasonally de­ plant and animal distribution. .~long the pressed thus allowing seeds of plants gulf coast where tides are diurnal and of otherwise inhibited by saline conditions relatively low amplitude, wind and stonn to genninate. Adult pl ants may al so be tide-related water surges play a much influenced when salinity stress is reduced greater role in the hydrology of the during the early season growth. This may extensive, nearly fl at marshes than they result in greater primary productivity do in the Pacific rt>rthwest. than would occur without the freshwater. Those marshes behind the sandspits devel­ Like the New England riarshes, the oped on bay mouth bars, on the other hand, saline marshes of the Pacific Northwest are often exposed to salinities only have a high angiosperm diversity compared s1 ightly lower than the coastal water to the extensive monospecif i c stands of since often only small freshwater sources gulf coast and Atlantic coast marshes are included in the drainage basin. south of Cape Cod. The upland border transitions in the Pacific North\'lest tend The substrates on which Pacific to be sharper than those of many other r.brthwest marsh fauna and flora depend coastal areas because of the sharp topo­ vary from very sandy to primarily s i1 t and graphic relief in the region. Few studies clay. More mature r:iarshes have built have defined the moisture tolerance of their own substrate of peat. In the Paci- tidal species in the region, and there are

58 still some problec1s 1-1ith setting the upper ~n ~oth directions. F.elgrass beds can ~oundary of the marsh. Macroal gae are rnfluence the mush by acting as filters prominent in the occasional tidal pools on for nutrients C011ing into the wetl '1nds the 1.1a rsh surface and pemanent pools ; n fror1. coastal upwel 1 ifllj or by trap;ling the streams. nutrients leaching fro1,1 decomposing 11arsh grasses. further, the seagrasses "lay he Mic rufl o ra 1 assemblages of the region carried by the tides to the narsh where '.lave not been as well studied as those in they deco•npl)se, thereby cycling the nutri­ the 101arshes of southern California or the ents between tidal narsh and adjacent sea­ Atlantic and gulf coasts. There are, how­ grass beds. ever, subst.:rntial algal popu1 ations on !10th the na rsh surf ace and the bare banks liistorical ly, tidal marsh r:1anag(!t"1ent bordering the creeks. Fungal and bacte­ in the Pacific rt>rthwest has involved the rial populations in the soil or in the exploitation of the wetlands for agrietll­ dead pl ants have not been studied, but tural purposes. The high intertidal wet- deco11position rate experiments indicate 1 ands have been grazed and 11any of the processes involving these organisms are as lower wetland habitats have been "rc­ rapid as in the Southeast Atlantic coastal cl aimed" by the construction of dikes. t ida 1 marshes. These "reclaimed" lands were primarily used for pasture and hay crops in the Faunal components on the marsh sur­ dairy regions of the coast. Recently the face are doninated by insects on the marsh value of these coastal wetlands has been pl ants and by amphipods in the eel grass recognizei:l, and 1aws protecting then from «rack coninunity. NJtably absent are the Deing fil I ed or annexed to the adjacent fiddler crdbs and sna i1 s common in the upland have been enacted. Mitigation pro­ :~arshes of the Mid-Atlantic and Southeast. c1~dures have become popular in recent Bird and mammal utilization of the Pacific years; restoration of damaged marsh or the tt>rthwest wetlands is extensive in some creation of new ones on dredged material areas, but perhaps the most important has frequently been the proposed solution food web trophic linkages with vertebrates when coastal developrient requires the occur with the fishes in the tidal creeks. destruction or alteration of existing wet- Data frcx:1 southwestern Canada document the 1 ands. use of tidal marsh habitats by juveniles of com:nercial ly foiportant salmon ids. The challenges for scientists, manag­ Juvenile salmonids an1 other fish feed on e rs. and deve 1 ope rs in the coming yea rs organisms such as harpacticoid copepods, are complex and difficult. How can we which in turn depend on marsh detritus for have maxirnu•11 economic growth in the their sustenance. The residence time of coastal zone witnout destroying the eco- the numerous species of salnonids is vari­ 1og ical and aesthetic base which inspired a)l e and depends on species and estuarine that development? What is the necessary conditions. In addition to the salmonids, areal extent of the various components there are a number of other important necessary to sustain the natural and forage fish which are found as temporary anthropogenic systems? What is the best or pennanent residents of the tidal marsh spatial distribution of marsh, Mudflat, streams. seagrass bed, and stream channel for a particular site? How can we manage the Energy and materials which flow 1i1arsh-estuarine complex to 1:iaxirnize pro­ between the r.1a rsh and upland ecosystems ductivity? The solution to these problems are dependent on animal migration and involves not only the develo~nent of sound ~1ater flow from the uplands. Adjacent manager~ent pl ans, but al so the establish~ ecosystens at lower elevations, such as ment of a val 1d scientific base on which the mud and sand f1 ats and subtidal estu­ to make these plans. Perhaps the greatest arine channels, can additionally interact danger is that we will bypass the latter via tidal waters which can carry materials in our rush to devise the for.:ier.

59 REFERENCES

Akins, G.J., and C.A. Jefferson. 1973. Burg, M.E., D.R. Tripp, and E.S. Rosen­ Coastal wetlands of Oregon. A natural berg. 1980. Plant associations and resources inventory report to the Ore­ primary productivity of the Nisqual ly gon coastal conservation and develop­ salt marsh on southern Puget Sound, ment commission, Salem. Washington. f'brthwest Sci. 54:222-235. American Ornithologists' Union (AOU). Christi an, R. R. , R. B. Hanson, J • R. Ha 11 , 1982. Thirty-fourth supplement to the and W.J. Wiebe. 1981. Aerobic mi­ American Ornithologists' Union check- crobes and meiofauna. Pages 113-135 in 1 ist of N::>rth Anerican birds. Auk L.R. Pomeroy and R.G. Wiegert, edS. (Suppl.) 99. pp. 1CC-16CC. The ecology of a salt marsh. Springer­ Verl ag, New York. Bean, M.J. 1978. The federal role in wetland protection: an overview. Congleton, J.L., and J.E. ~ith. 1976. Pages 103-107 in J.H. Montanari and Interactions between juvenile salmon J.A. Kusler, co-chairmen. Proceedings and benthic invertebrates in the of the national wetland protection sym­ Skagkit salt marsh. In C.A. Simenstad posium, June 6-8, 1977, Res ton, Vir­ and S.J. Lipovsky, eds. Fish food ginia. Tech. Rep. U.S. Fish Wildl. habits studies, proceedings of the Serv. Biol. Serv. Programs, FWS/OBS- first Pacific Northwest technical work­ 78/97. shop, Astoria, Oregon, October 13-15, 1976. Washington Sea Grant, Seattle. Bottom, D., B. Kreag, F. Ratti, C. Roye, and R. Starr. 1979. Habitat cl assifi­ Cowardin, L.M., V. Carter, F.C. Golet, cation and inventory methods for the and E.T. LaRoe. 1979. Classification management of Oregon estuaries. Oregon of wetlands and deepwater habitats of Department of Fish and Wildlife, Port- the United States. Tech. Rep. U.S. 1 and. 109 pp. Fish Wildl. Serv. Biol. Serv. Program, FWS/OBS-79/31. 103 pp. Breeding, C.H.J., F.D. Richardson, and S.A.L. Pilgrim. 1974. Soil survey of Darmody, R.G., and J.E. Foss. 1978. New Hampshire tidal marshes. New Hamp­ Tidal marsh soils of Maryl and. Mary­ shire Agricultural Experiment Station, l and Agricultural Experiment Station, University of New Hampshire, Durham. University of Maryl and, College Park.

Burg, M.E., E. Rosenberg, and D.R. Tripp. Disraeli, D.V., and R.W. Fonda. 1978. 1976. Vegetation associations and pri­ Gradient analysis of vegetation in a mary productivity of the Nisqually salt brackish marsh in Bellingham Bay, Wash­ marsh on southern Puget Sound, Washing­ ington. Can. J. Bot. 56:1308-1326. ton. Pages 109-144 in S.G. Hennan and A.M. Wiedemann. Contributions to the Duxbury, A.C. 1971. The earth and its history of the southern Puget Sound oceans. Addison-Wesley Publishing region, Washington. Evergreen State Co., Reading, Mass., 381 pp. College, Olympia, Wash.

60 Eilers, H.P. 1975. Plants, plant commu­ Gallagher, J.L., and H.V. Kibby. In pre­ nities, net production and tide levels: paration. Aerial primary production the ecological bi ogeography of the and mineral dynamics in estuarine marsh Nehalem salt marshes, Tillamook County, plant stands in Oregon. Oregon. Ph.D. Thesis. Oregon State University, Corvallis. 368 pp. Gallagher, J.L., and W.J. Pfeiffer. 1977. Aquatic metabolism of the stand­ Ewing, K. 1983. Environmental controls ing dead pl ant communities in salt and in Pacific Northwest intertidal marsh brackish water marshes. Limnol. plant communities. Can. J. Bot. 61: Oceanogr. 22: 562-564. 1105-1116. Gallagher, J.L., and F.G. Plumley. 1979. Federal Water Pollution Control Act. Underground bi amass profiles and pro­ 1972. Public Law 92-500. ductivity in Atlantic coastal marshes. Am. J. Bot. 66: 156-161. Frenkef, R.E., T. Boss, and S.R. Schuller. 1978. Transition zone vege­ Gallagher, J.L., W.J. Pfeiffer, and L.R. tation between intertidal marsh and Pomeroy. 1976. Leaching and microbial upland in Oregon and Washington. Final utilization of dissolved organic carbon report prepared for the U.S. Environ­ from leaves of Spa rtina a lterni flora. mental Protection Agency. Department Estuarine Coastal Mar. Sci. 4:467-471. of Geography, Oregon State University, Corvallis. Gallagher, J.L., H.V. Kibby, and K.W. Skirvin. In press, a. Respiration of Frenkel, R.E., H.P. Eilers, and C.A. microbial communities associated with Jefferson. 1981. Oregon coastal salt the dead plants in Oregon estuarine marsh upper limits and tidal datums. marshes. Estuarine Coastal Mar. Sci. Estuaries 4(3):198-205. Gallagher, J.L., H.V. Kibby, and K.W. Gallagher, J.L. 1977. Zonation of wet­ Skirvin. In press, b. Zostera inter­ lands and tidelands. Pages 752-758 in action with an Oregon salt marsh. J.R. Cl ark. Coastal ecosystem manage­ Aquat. Bot. ment: a technical manual for the con­ servation of coastal resources. John Higley, D.L., and R.L. Holton. 1981. A Wiley and Sons, New York. 928 pp. study of the invertebrates and fishes of salt ma rs hes in two Oregon es tua­ Gallagher, J.L. 1980. Salt marsh soil ri es. Final report prepared for the development. Pages 28-34 2.!l J.C. Lewis Department of the Army. School of and E.W. Bunce, eds. Rehabilitation Oceanography, Oregon State University, and creation of selected coastal Corvallis. habitats: proceedings of a workshop. U.S. Fish vJildl. Serv. Biol. Serv. Pro­ Hitchcock, C.L., and A. Cronquist. 1973. gram, Washington, D.C., FvJS/OBS-80/27. Flora of the Pacific Northwest. Uni­ versity of Washington Press, Seattle. Gallagher, J.L., and H.V. Kibby. 1980. 730 pp. Marsh plants as vectors in trace metal transport in Oregon tidal marshes. Am. Hoffnagle, J., and R. Olson. 1974. The J. Bot. 67: 1069-1074. salt marshes of the Coos Bay Estuary. Spec. Rep. Oregon Institute of Marine Gallagher, J.L., and H.V. Kibby. 1981. Biology, University of Oregon, Charles­ The streamside effect in a Carex ~- ton. 87 pp. byei estuarine marsh: the possible role of recoverable underground Hoffnagle, J., R. Ashley, B. Cherrick, t1. reserves. Estuarine Coastal Shelf Sci. Gant, R. Hall, C. Magwire, M. Martin, 12:451-460. J. Schrag, L. Stunz, K. Vanderzanden, and B. Van Ness. 1976. A comparative

61 study of salt marshes in the Coos Bay Kreag, R.A. 1979b. Natural resources of Estuary. A National Science Foundation Netarts Estuary. Vol. 2, No. 1. Ore­ Student Originated Study. University gon Department of Fish and VJildlife, of Oregon, Eugene. 334 pp. Portland. 45 pp.

Howarth, R.W., and A. Giblin. 1983. Lagna, L. 1975. The relationship of Sul fate reduction in the salt marshes Spartina al terni flora to mean high at Sapelo Island, Georgia. Limnol. water. Marine Science Research Center, Oceanog. 28:70-82. State University, New York at Stony Brook, New Sea Grant Institute, NYSSGP­ Huffman, R.T. 1981. Multiple parameter RS-75-002. approach to the field identification and delineation of aquatic and wetland LaRoe, E.T. 1978. Mitigation: a con- ecosystems. Tech. Rep. EL-81, U.S. cept for wetland restoration. Pages Army Engineer Waterways Experiment Sta­ 221-224 in J.H. Montanari and J.A. tion, Vicksburg, Miss. Kusl er, co-chairmen. Proceedings of the national wetland protection sympo­ Jefferson, C.A. 1975. Pl ant communities sium, June 6-8, 1977, Reston, Virginia. and succession in Oregon coastal salt Tech. Rep. U.S. Fish Wildl. Serv. marshes. Ph.D. Dissertation. Oregon Biol. Serv. Program. FWS/OBS-78/97. State University, Corvallis. 192 pp. Levy, D.A., and T.G. flbrthcote. 1981. Johannessen, C.L. 1961. Some recent The distribution and abundance of juve­ changes in the Oregon coast: shoreline nile salmon in marsh habitats of the and vegetation changes in estuaries. Fraser River Estuary. Tech. Rep. 25. Pages 100-151 in S.N. Dicken, ed. Some University of British Columbia, West­ recent physical changes in the Oregon water Research Centre, Vancouver, B.C. coast. Department of Geography, Uni­ 117 p. versity of Oregon, Eugene. Levy, D.A., T.G. flbrthcote, and G.J. Johannessen, C.L. 1964. Marshes pro- Birch. 1979. Juvenile salmon utiliza­ grading in Oregon: aerial photographs. tion of tidal channels in the Fraser Science 147:1575-1578. River Estuary, British Columbia. Tech. Rep. 23. University of British Colum­ Kibby, H.V., J.L. Gallagher, and W.D. bia, Westwater Research Centre, Van­ Sanville. 1980. Field guide to evalu­ couver, B.C. 70 p. ate net primary production of wetlands. U.S. Environmental Research Laboratory, Lewis, O.H., and M. Liverman. 1979. Corvallis, Oreg. EPA-600/8-80-037. Evaluation of soil moisture conditions in a coastal wetland. A progress re­ Kirby, C.J. 1978. Wetland development, port prepared by the U.S. Environmental an alternative. Pages 138-142 .i!!. J.H. Protection Agency, Corvallis, Oreg. Montanari and J.A. Kusler, co-chainnen. Proceedings of the national wetland Liverman, M.C. 1982. Multivariate protection symposium, June 6-8, 1977, analysis of a tidal marsh ecosystem at Reston, Virginia. Tech. Rep. U.S. Netarts Spit, Tillamook County, Oregon. Fish Wildl. Serv. Biol. Serv. Program. M. S. Thesis. Department of Geography, FWS/OBS-78/97. Oregon State University, Corvallis. Kreag, R.A. 1979a. Natural resources of Macdonald, K.B. 1977. Plant and animal Coquille Estuary. Vol. 2, fib. 7. Ore­ communities of Pacific North Anerican gon Department of Fish and Wildlife, salt marshes. In V.J. Chapman, ed. Portland. 48 pp. Wet coastal ecosystems. Elsevier Sci­ entific Publishing Co., Amsterdam.

62 Magwire, C. 1976a. Survey of bird spe- Oceanographic Institute of Washington. cies in and around the salt marshes of 1977. A summary of knowledge of the the Coos Bay Estuary. Pages 177-185 in Oregon and Washington coastal zone and Hoffnagle et al. A comparative study offshore areas; Vol. 1, Chapters 1-3. of salt marshes in the Coos Bay Estu­ Seattle, Wash. ary. A National Science Foundation Student Originated Study. University Odum, E.P. 1980. The status of three of Oregon, Eugene. ecosystem-1 evel hypotheses regarding salt marsh estuaries: tidal subsidy, Magwire, C. 1976b. Mammal populations outwell ing, and detritus-based food of the Coos Bay salt marshes. Pages chains. Pages 485-495 iD_ V.S. Kennedy, 191-200 in Hoffnagle et al. A compara­ ed. Estuarine perspectives. Academic tive study of salt marshes in the Coos Press, New York. Bay Estuary. A National Science Foun­ dation Student Originated Study. Uni­ Oregon Land Conservation and Development versity of Oregon, Eugene. Commission (LCDC). 19 76. Statewide ~anning goals and guidelines. Depart­ McVay, M.E., P.E. Heilman, D.M. Greer, ment of Land Conservation and Develop­ s. E. Brauen, and A.S. Baker. 1980. ment, 1175 Court Street N.E., Sal em, Tidal freshwater marsh establishment on Oreg. 97310. dredge spoils in the Columbia River Es­ tuary. tl. Environ. Qual. 9: 488-493. Percy, K.L., O.A. Bella, C. Sutterlin, and P.C. Kl ingeraan. 1974. Descrip­ Meyer, J. H. 1979. A review of the tions and infonnation sources for Ore­ l iterat\.lre on the value of estuarine gon estuaries. Sea Grant College Pro­ and shoreline areas to juvenile sal m­ gram, Oregon State University, Cor­ onids in Puget Sound, Washington. U.S. vallis. Fish Wildl. Serv. Fish. Assist. Office, Olympia, \tJash. 24 pp. Pfeiffer, l~.J., and R.G. Wiegert. 1981. Grazers on Spartina and their pred­ Mitchel 1, D.L. 1981. Salt marsh rees- ators. Pages 87-112 ..i.r! L.R. Pomeroy tablishment following dike breaching in and R.G. Wiegert, eds. The ecology of the Salr.1on River Estuary, Oregon. a salt marsh. Springer-Verlag, t-ew Ph.D. Dissertation. Department of York. Botany and Plant Pathology, Oregon State University, Corvallis. Pomeroy, L.R., 14.M. Darley, E.L. Ounn, J.L. Gallagher, E.B. Haines, and D.M. Montague, C. L., S.M. Bunker, E.B. Haines, ltJhitney. 1981. Primary production. M.L. Pace, and R.L. Wetzel. 1981. Pages 39-69 in L.R. Pomeroy and R.G. Aquatic 'llac roconsumers. Pages 69-85 iD_ Wiegert, eds. The ecology of a salt L.R. Pomeroy and R.G. Wiegert, eds. marsh. Springer-Verlag, New York. The ecology of a salt marsh. Springer­ Verl ag, Ne~1 York. Proctor, C.M., J.C. Garcia, D.V. Galvin, G.C. Lewis, L.C. Loenr, and A.M. Massa. National Oceanic and Atmospheric Adminis­ 1980. An ecological characterization tration (NOAA). 1975. The relation­ of the Pacific Northwest coastal re­ ship between the upper limit of coastal gion; Vol. 2, Characterization atlas­ n1arshes and tidal datum. Rockville, regional synopsis. U.S. Fish Wildl. Maryl and, National Ocean Survey (a Serv. Biol • Serv. Program. FWS/OBS- pilot study conducted for the EPA). 79 /12. f'brthcote, T.G., N. T. Johnston, and K. Ratti, F. 1979. Natural resources of Tsumura. 1979. Feeding relationships Rogue Estuary. Vol. 2, No. 8. Oregon and food web structure of 1ower Fraser Department of Fish and Wildlife, Port- River fishes. Tech. Rep. 16. Univer­ 1 and. 33 pp. sity of British Columbia, Westwater Research Centre, Vancouver, B.C. 73 pp.

63 Reimers, P.E. 1973. The length of resi­ Oregon Department of Land Conservation dence of juvenile fall chinook salmon and Development. in Sixes River, Oregon. Res. Oep. Oreg. Fish Comm. 4(2):1-43. Teal, J.M., and J. Kanwisher. 1966. Gas transport in the marsh grass, Reimold, R.J., .J.l. Gallagher, and D.E. Spartina al terniflora. J. Exp. Bot. Thompson. 1972. Coastal mapping with 17: 355-361. remote sensors. Pages 99-112 in pro­ ceedings of a symposium on coastal map­ Thom, R.M. 1981. Primary productivity ping. tlrnerican Society of Photograrn­ and carbon input to Grays Harbor metry, Washington, D.C. 319 pp. Estuary, Washington. EnvironMental Resources Section, U.S. Amy Corps of Roye, C. 1979. Natural resources of Engineers, Seattle. Coos Bay Estuary. Vol. 2, No. 6. Oregon Department of Fish and Wildlife, Tjepkema, J.D., and H.J. Evans. 1976. Portland, 87 pp. Nitrogen fixation associated with Juncus balticus and other plants of Rudy, P., ,Jr., and L.H. Rudy. 1981. Oregon wetlands. Soil Biol. Biochem. Oregon estuarine invertebrate animals. 8: 505-509. Draft MS. Contract 79-111. U.S. Fish Wildl. Serv., Portland, Oreg. 225 pp. U.S. Anny Corps of Engineers (USACE). 1981. Digest of water resources pol i­ Sel iskar, D.M. 1981. Morphological and cies and authorities. DAEN-CWR-R, anatomical responses of selected Pamphlet rt>. 1165-2-1, Office of the coclstal salt marsh plants to soil Chief of Engineers, Department of the moisture. M.S. Thesis. Department of Anny, Washington, D.C. 270 pp. Botany and Plant Pathology, Oregon State University, Corvallis. U.S. Department of Commerce. 1979. Tide tables 1979 - high and 1 ow water pre­ Sel iskar, D.M. In press. Root distri- dictions, west coast of North and South bution as an indicator of upper salt fl4nerica. U.S. Department of Commerce, inarsh wetland limits. Hydrobiol og ia. National Oceanic and Atmospheric Ad­ ministration, and National Ocean Sur­ Smart, R.M. 1982. Distribution and vey. environmental control of productivity and growth fonn of Sgartina al terni­ U.S. Department of the Interior, Fish and fl ora (Loisel). In 0.N. Sen, ed. Wildlife Service. 1970. Willapa Bay, Contributions to ~the ecology of Washington. Pages 213-248 in National halophytes. (In press.) estuary study, Vol. 3. - Stout, H., ed. 1976. The natural U.S. Geological Survey. 1970. The resources and human utilization of national atlas of the United States. Netarts Bay. A National Science Foun­ U.S. Department of the Interior. dation Student Originated Study. Oregon State University, Corvallis. Wass, M.L., and T.D. Wright. 1969. 247 pp. Coastal wetlands of Virginia. Spec. Rep. in Applied Mar. Sci. and Ocean Stumpf, R. B. 1981. Salt marsh sed imen­ Eng. rt>. 10. Virginia Institute of tation during stonn free conditions. Marine Science, Glouster Point. EOS 62(17) :303. 154 pp. Taylor, A.H., and R.E. Frenkel. 1979. Whitney, O.M., A.G. Chalmers, E.B. Ecological inventory of Joe Ney Slough Haines, R.B. Hanson, L.R. Pomeroy, and Marsh restoration site. Department of B. Sherr. 1981. The cycles of nitro­ Geography, Oregon State University, gen and phosphorus. Pages 163-183 in Corva 11 is. A report submitted to the L.R. Pomeroy and R.G. Wiegert, edS:-

64 The ecology of a sdl t marsh. Springer­ 1981. Anaerobic res pi rat ion and fer­ Verl ag, New York. :nenta t ion. Pages 13 7-159 .in. L. R. Pomeroy and R.G. Wiegert, eds. The Wiebe, IJ.J., R.R. Christidn, J.A. Hansen, ecology of a salt marsh. Springer­ G. King, B. Sherr, and G. Skyring. Verl ag, New York.

65 ,---:-.50272- -101 REP()~': DOCUMENTATION 11. REPORT NO • I z. 13. Rec;oien\'s ACC<»;on No. PAGE . - FWS/OBS-82/32 I 4.. Title ~nd Subtitle l 5. Report Date The ecology of tidal marshes of the Pacific Northwest coast: I October 1983 a community profile

7. Author(s.) D.M. Seliskar. and J.L. Ga 11 agher :: ,, "''"'"' '"'"'""'" "'" "' ~j 9. Perlorming: Organization Name and Address l 10, Pro1ect/iasio:./Work Unit No. University of Delaware I 1------I College of Marine Studies j l ~- Ccntr.acHC} or Grant(G) No. Lewes, DL 19958 I (C) I I (G) I 12. Spot'i:s.oring Orc;anizatiori Name and Address 113. Type of Report t.. Period Covere~ National Coastal Ecosystems Team I Division of Biological Services, Fish and Wildlife Service l U.S. Department of the Interior JK I Washington, DC 20240 I l S. Supplementary Notes

· 16. Abstrect (Limit: 200 words.} Tidal marshes of the Pacific Northwest develop (1) on the fringes of estuaries of rivers which empty more or less directly into the ocean or Puget Sound and (2) behind baymouth bars which form between adjacent headlands on the rocky coast. Floristically these wetlands are more complex than marshes of equivalent salinity along the gulf, Southeast and Mid-Atlantic coasts. Pacific Northwest marshes resemble more closely those of northern California and parts of New England. Primary productivity ranges widely as do rates of plant decomposition. Detritus appears to fuel a productive food web, the best documented of which involves salmonids feeding in the marsh creeks. Except for insects, macrofauna are not common on the marsh surface. A conceptual model of functional groups within the marsh, and a model depicting interactions between the marshes and adjacent ecosystems are presented. To date fewer quantitative ecological studies have been done on these wetlanas than on those in parts of California or the gulf and Atlantic coasts. This situation is rapidly changing and the emerging picture is that Pacific Northwest marshes are similar in function to those more intensively studied. Organisms and relative magnitudes of the fluxes are different, but processes are the same.

17. Oocurnl!!nt An.&ly~i~ a. Descriptors

Ecology, Biological productivity, Plants (botany)

b. IC::i!:ntifiers/Open·Ended Terms Tidal marshes, Puget Sound, Ecological model, Plant decomposition, Wetlands

c. COSATl field/Group

21. No. of P2g~s 18. Availability St.ate-ment j 19. Security Ciass (Thi~ Re-:.or:.; I Unclassified 65 -- 22. Price Unlimited i 20. Security Cla~s. (Thi5 Pa€,e) ! Unclassified ,. OPTIONAL FORM .12 (< 77) (See ANSl-l..39.18) (formerly NTIS-35)