Manual for assessing restored and natural coastal wetlands with examples from southern California

[From the U.S. Government Printing Office, www.gpo.gov]

_''XS@@ESSING
A M@ANUAL-... VQ
RES_T RED'. A NATURAL
C, -OAS :T-A-L- WETLANDS

WITH EXAMPLES FROM

SOUTHERN CALIFORNIA

by the

Pacific Estuarine Research Laboratory
(PERL)

Biology Department
San Diego State University
San Diego, California 92182-0057

1990

A Publication of the California Sea Grant College

GC
57.2 CZIC COLLECTION
C313
no.021

Dedicated to the memory of

Dr. Millicent Quarnmen,
estuarine researcher,
who insisted that restored wetlands
be assessed on the basis of
their functioning.

Cite as: Pacific Estuarine Research Laboratory. 1990. A manual for assessing restored
and natural coastal wetlands with examples from southern California. California
Sea Grant Report No. T-CSGCP-021. La Jolla, California.

U . S . DEPARTMENT OF COMMERCE NOAA
COW- COASTAL SERVICES CENTER
2234 SOUTH HOBSON AVENUE
CHARLESTON , SC 29405-2413

Property of CSC Library

Contributions

Principal Author, Compiler Technical Advice
Joy B. Zedler, Ph.D. Wetland and estuarine John Boland, Ph.D. Weiland ecologist,
ecologist, with interest in all aspects of salt reviewed the section on bird sampling
marsh structure and functioning. Responsible Brian Fink, M.S. Coastal plant propagation
for most of the writing--and all of the effors. expert, advice on salt marsh bird's beak
Theodore Griswold, M.S. Wetland
Section Author biologist; data analysis and interpretation.
Barbara Kus, Ph.D. Avian ecologist,
Ren6 Langis, Ph.D. Weiland ecologist, with provided methods for bird study
expertise in assessing nutrient dynamics * Christopher S. Nordby, M.S. Authority
Principal author of Section IU: Soils; co. on wetland fishes and invertebrates
author of Section IV.2: Water quality; Kathy Williams, Ph.D. Insect ecologist,
collaborator on the San Diego Bay case study; provided methods for insect study
and reviewer of most sections.

Data Collection
Illustrations: Mark Lung Patrice Ashfield. Birds, Tijuana Estuary,
Copy Editing: Bruce Nyden Sweetwater River Wetland Complex
Melissa Bukowski. Geographic Information
Assistance with manuscript format: System, Tijuana Estuary
Julie Reynolds, Dawn Makis Max Busnardo. Vegetation map, Tijuana Estuary
John Cantilli. Soil sulfide dynamics, soil redox
Carol Digiorgio. Benthos, Tijuana Estuary
Robert Espinoza. Herpetology, Tijuana Estuary
Agency Advice Peggy Fong. Algal dynamics at Tijuana Estuary
and in experimental mesocosms
US Fish and Wildlife Service, US Army Jeff Haltiner, Ph.D. Hydrologic expertise,
Corps of Engineers, National Marine Fisheries Sweetwater River Wetland Complex
Service, State Coastal Conservancy, California Kim Johnson. Insects, Tijuana Estuary
Coastal Commission, California Department of Sharon Lockhart. Weiland habitat area data for
Parks and Recreation, California Department of San Diego County
Transportation. Note: Input from staff Paul Little. Water quality, Tijuana Estuary, Los
members does not imply agency endorsemen . Peflasquitos Lagoon
Ricardo Martinez-Lara. Polychaete identification
David Nelson. Geographic Information System,
Tijuana Estuary
Jeff Newman. Birds, Tijuana Estuary
Research and Publication Support Regina Rudnicki. Macroalgae, Tijuana Estuary
and in enrichment experiments
This work is a result of research sponsored Sue Rutherford. Epibenthic invertebrates,
in part by the National Oceanic and Atmospheric Sweetwater River Weiland Complex
Administration (NOAA), National Sea Grant Theresa Sinicrope. Water quality, Sweetwater R.
College Program, Department of Commerce, Weiland Complex
under grant number NA85AA-D-SG140, project Kendra Swift. Restored marsh assessment, San
number R/CZ-82, through the California Sea Diego Bay wetlands
Grant College, in part by NOAA Office of Ernie Taylor. Mammals, Tijuana Estuary
Coastal Zone Management, Marine and Estuarine John Tiszler. Vegetation map, Tijuana Estuary
Management Division, under contract no. 85- Nils Warnock. Birds, Tijuana Estuary
0236-85-072-A, and in part by the California Philip Williams, Ph.D. Hydrologic expertise,
State Resources Agency, the California Tijuana Estuary
Department of Transportation, and the California Richard Wright, Ph.D. Geographic Information
State Coastal Conservancy. System, Tijuana Estuary
Malgorzata Zalejko. Nitrogen fixation, San
Support for publication was provided in part Diego Bay wetlands
by the NOAA Coastal Ocean Program. Emily Zedler, Sarah Zedler. Data entry, proofing

Table of Contents

1. Introduction ............................................................................. I

11. Strategies for wetland construction, restoration, enhancement ....... 3

1. Concerns .............................................................................. 3

2. Rationale for functional assessment ................................................ 5

3. Objectives of assessment ............................................................ 6

4. Criteria for "successful mitigation" as required in San Diego Bay ............. 12

5. Reference wetlands and reference data sets ....................................... 13

6. How should major restoration programs be undertaken? ....................... 17

III. Case study: Sweetwater River Wetlands Complex ....................... 19

IV. Sampling methods and comparative data from natural wetlands ..... 35

1. Hydrologic functions ................................................................ 37

2. Water quality .......................................................................... 40

3. Soils: Substrate qualities and nutrient dynamics .................................. 44

4. Vegetation composition and growth ................................................ 52

5. Marsh insects: Pollinators, predators, and prey .................................. 65

6. Aquatic invertebrates: Food chain support ........................................ 69

7. Fishes: Community dynamics, controlling factors .............................. 76

8. Birds ................................................................................... 81

9. Reptiles and amphibians ............................................................. 89

10. Mammals .............................................................................. 90

V. Recommendations for minimum monitoring ................................ 92

VI. Literature cited ......................................................................... 96

Introduction

experience needed to understand complex
1. Introduction ecosystems. In addition to drawing upon
their expertise, the text includes
This manual presents recommenda- information that has been summarized for
tions for assessing the structure and. various. conferences on wetland
functioning of coastal wetlands, with restoration (e.g., Zedler et al. 1988,
emphasis on salt marshes and tidal 1990), as well as information from
creeks. While the recommendations can manuscripts in progress (e.g., Nordby
be applied to many situations, the main and Zedler, in press; Langis et al., in
purpose of the manual is to standardize press) and recent studies carried out at
methods of assessing restored, enhanced, Tijuana Estuary
or constructed wetlands. The need. Thorough evaluation
The basic premise is that a man- procedures that can stand the test of
made wetland should provide the same scientific review are clearly needed.
values as the region's natural wetland Quammen (1986) recommended two
ecosystems. This is especially important types of monitoring to evaluate whether
in southern California, where native created wetlands compensate for the
wetland habitats have dwindled to small, losses in natural wetlands: first, an
disturbed, isolated remnants, most of assessment of compliance with resource
which can no longer sustain the agency recommendations and, second,
biodiversity that once characterized this long-term, scientific evaluation to test
region. Our remarks often emphasize predictions and hypotheses concerning
rare native species, but we do not suggest the development of natural ecosystem
that management be restricted to target functions. We suggest that the two
assessment goals be merged, such that
species (Zedler 1984). The threatened "compliance" becomes the successful
and endangered species that often drive "replacement of lost wetland functions."
restoration and management efforts are Projects would be , considered in
the obvious symptoms of more subtle compliance with mitigation policies once
changes in ecosystem structure and the constructed or modified wetland
function A whole-system approach at shows high potential for achieving natural
the regional scale is needed to assess, functional attributes. Furthermore, the
understand, and manage coastal southern information base upon which judgments
California wetlands. of compliance/success are made should
With restoration and enhancement be able to withstand scientific review.
plans growing in number and size, it is The problems resulting from erroneously
imperative that we understand how well concluding that a project is successful are
such projects are working if we are to too great for the assessment process to be
achieve. the overall objective of casual or short-term. A margin of safety
maintaining regional biodiversity. If is essential.
most projects are failing to provide the From the ecological perspective,
habitat required for native species to determining when a constructed wetland
persist in perpetuity, we must find better has attained the functions of a natural
restoration techniques, and we must wetland is neither simple nor quick. As
certainly halt mitigation practices that Odum (1987, p. 67) points out, "Too
allow the "replacement" of native wetland frequently, the success or failure ... is
habitat with artificially constructed. determined after a year or two's growth
wetlands. of the original, planted vegetation.
The list of scientists whose work and Unfortunately, dramatic unanticipated
advice have contributed to this manual changes may occur over the ensuing
indicates the broad range of skills and years .... it is not uncommon for the plant
community to become invaded and

Introduction

dominated by aggressive 'disturbance
species'.... The long-term result may be
a wetland environment which has limited
functional value for wildlife habitat
support or nutrient processing and which
lacks aesthetic attractiveness to the degree
originally planned." Likewise, Broome
et al. (1987, p. 197) concluded that
monitoring a constructed wetland for four
growing seasons was insufficient to
determine "if the created marsh reaches
equivalent levels of production for all
plant species and remains self-
sustaining."

Organization of this manual.
The rationale for requiring constructed
wetlands to meet strict assessment criteria
is developed in Section II. A case study
of the Sweetwater River Wetlands
Complex follows (Section III). For this
case, functional criteria were mandated by
resource agencies, and the Pacific
Estuarine Research Laboratory was
requested to undertake the detailed
assessment. The assessment is not yet
complete, but results to date show -how
the 4-5-year-old wetlands compare to
natural wedand remnants.

The assessment methods are grouped
under ten attributes of coastal wetlands
(Section IV). For each component of the
ecosystem, the reasons for study and
assessment are provided, followed by
methods recommended by PERL.

Recommendations for the minimum
sampling effort needed for monitoring
constructed wetlands are provided in
Section V. Reference data and sources of
additional information are then given.

The manual is not intended to be a
final statement, but a working document V.
that will continue to evolve as our
knowledge of coastal wetlands advances.
In all applications of the work, users
should check with PERL to insure that
the latest information is in hand.

Strategies for wetland construction, restoration and enhancement

11. Strategies for Too many projects have emphasized
single-species management. The planting
wetland construction, of mangroves, cordgrass, or eelgrass
restoration and seems to have become synonymous with
tidal wetland restoration, and transplant
enhancement survival rates have become measures of
"success." In southern California, the
1. Concerns endangered species law has had a curious
impact on restoration activities--it has
Coastal wetland restoration is an tended to suggest single-species man-
active endeavor in many regions of the agement and to imply that one wetland
country. New habitats are being con- type should be modified to meet the needs
structed using dredge spoils, grading of of a target species. Thus, we have seen
topography, or other techniques. proposals to plant cordgrass in pickle-
Wetland construction and restoration weed marshes, as mitigation for habitat
projects are often initiated to offset or destruction, because cordgrass provides
mitigate impacts of wetland destruction. habitat for the federally endangered light-
footed clapper rail. However, pickle-
When uplands are excavated to construct weed provides habitat for the State-listed
habitats as replacements for lost wet- Belding's Savannah sparrow, and con-
lands, the question is--are the lost wet- version of one vegetation type to another
land values replaced? When existing does not constitute ecosystem enhance-
wetlands are modified to restore or ment.
enhance wetland values, two questions
arise--are existing values maintained and We question the hypothesis that
are additional values provided? As per- constructed wetlands are carrying out the
ceptions of success or failure have broad range of natural wetland function5.
become more controversial, two things Unfortunately, existing data are inade-
have happened--criteria for project com- quate for testing this hypothesis.
pliance have become more, detailed (as Evaluations of both natural and con-
indicated by more complicated permit structed ecosystems have emphasized
requirements issued by the US Army structural, rather than functional,
Corps of Engineers), and researchers attributes. Furthermore, most of the
have become more interested in studying structural data are for plants. Fewer
man-made systems. evaluations of man-made wetlands have
Wetland scientists around the country considered the animal components.
(cf. regional reviews in Kusler and Three other factors add to the diffi-
Kentula 1989, Strickland 1986, Good culty of comparing man-made and natural
1987, Zelazney and Feierabend 198.7) are system functions. First, there may not be
concerned that restoration and habitat suitable models with which to compare
construction attempts are not replacing constructed systems--most coastal wet-
lost wetland values. Milton Weller of lands are already modified; their hydro-
Texas A&M states (in Kusler and Kentula logic regimes (inundation and salinity
1989), "In most situations we can patterns) have probably changed dramati-
provide the environmental needs to allow cally in the past century, and key species
dominant wetland plants and animals to may already have been eliminated. Is it
succeed, and the product will satisfy sufficient to strive for conditions that now
many if not most viewers. We cannot, exist in wetland remnants, or must we
however, expect to replace the complex understand what conditions allowed those
and diverse natural systems that are a wetland ecosystems to develop? The
product of many centuries of evolution latter seems to be necessary for plant
and randomness ...... communities where population establish-

Strategies for wetland construction, restoration and enhancement

ment limits biodiversity. If we lack the Sometimes, mitigation is required but
necessary blueprints for wetland devel- never implemented. A recent evaluation
opment, we can't expect to construct a of I I freshwater wetlands constructed in
system that mimics nature. Oregon included data on hydrology
(using the presence of water and saturated
Second, the functions of natural soil as indicators), to ography (slope),
. p
models have not been thoroughly studied. and plant species occurrences (Gwin and
We have far more information on the Kentula 1990). The vegetation at the
composition of benthic: macroinvertebrate constructed wetlands differed greatly
communities than on their psQ by shore- from species lists given in the permit file.
birds and other predators. There is con- Volunteer species (those not on the
siderable information on nutrient levels in planting list) made up 93-100% of the
salt marsh soils, but less understanding species present (ibid.). Similar results
of the sources and rates of inflow and were found in evaluations of Florida
outflow. Studies of one important pro- wetlands (Kentula, pers. comm.). The
cess, i.e. the net flux of detritus, suggest general conclusion is that the strucutral
that results differ from place to place. attributes of created wetlands are not as
Both the direction of net movement and planned.
the magnitude of organic matter flux vary
considerably from wetland to wetland, With recognition of these problems,
and even from site to site within a given there is increasing demand for improved
system. planning, better project implementation,
and monitoring of vegetation establish-
Finally, there hasn't been sufficient ment. Still, there is continuing concern
time for constructed wedand ecosystems that constructed wetlands do not replace
to mature--hence, it is difficult to assess lost functional values and that we don't
their future potential for performing know how to correct the situation.
natural wetland functions. Few man-
made systems have been in place "n Two representatives of the U.S. Fish
monitored for more than 5 years. While and Wildlife Service (Holmberg and
it may'be possible to construct tidal Misso 1986, p. 11) acknowledge that
wetlands that eventually mimic the func- "...large-scale artificial wetland creation
tions of natural wetlands, judgments are as a viable method of replacing functional
often made after only 1-3 years. Because natural wetlands has not been docu-
we don't know how long it takes, long- mented." Likewise, the Environmental
term assessments must be planned and Protection Agency has a "conservative
continued until we are reasonably confi- policy" on mitigation because of the
dent that expectations have been met. An scientific uncertainty associated with
additional reason for long-term assess- man-made wetlands (Ciupek 1986).
ment is that single measurements (one- Quarnmen (1986) reviewed several
time or one-year data sets) are.unlikely to reports on mitigation and concluded
describe salt marsh structure and func- "...that compliance is low and that the
tioning adequately, because of high tem- effectiveness of restoration to compensate
poral variability in climate and wetland for wetland losses cannot yet be
responses. Low plant biomass in one determined."
year may indicate poor potential for salt
marsh development, or it may simply The question is, what should be
reflect a year of below-average growing done? Golet (1986) recommends that
conditions. regulatory agencies take a conservative
stance on mitigation by rejecting propos-
Many salt marsh mitigation efforts als where loss of wetland is avoidable
have problems. In some cases, failures and reject proposals that result in net
are due to poor planning; in others, plans losses of wetland area. Instead, he
are not properly implemented. favors protecting wetlands in their natural

Strategies for wetland construction, restoration and enhancement

state and requiring any replacement wet- irreversibly, and threatened species may
lands to "recreatei as nearly as possible, move closer to extinction. However, the
the original wetlands in terms of size, risks are less if mitigation projects are
type, geographic location, and setting ...... mistakenly judged to be failures (i.e., if
Quarnmen (1986) recommends clearly wetland functions can be replaced but we
stated objectives and design criteria for don't realize it). It is better to be cautious
mitigation wetlands, with monitoring to than to assume success prematurely if we
be conducted and reported. Good (1987, are to prevent the net loss of in-kind
p. 107) calls for "More systematic moni- resources, as required by mitigation
toring and evaluation of existing and policy (US FWS 1988). Conclusive evi-
soon- to- be-constructed mitigation pro- dence that constructed wetlands can
jects ... to expand our knowledge base. replace natural wetlands would facilitate
Experiments need to be incorporated into restoration and enhancement efforts.
wetland restoration projects to evaluate Conclusive evidence that certain mitiga-
our hypotheses about what does and does tion procedures canno maintain resources
not work. Documentation and sharing of would reduce economic losses by helping
successes and especially failures is to resolve conflicts early in the planning
needed, not only among agencies, but in process.
the published literature."
Disillusionment with the functioning
To assist in the evaluation and docu- of restored marshes has led to stricter
mentation of mitigation efforts that are goals for mitigation and enhancement
already underway, and thus to improve projects in coastal wetlands and the need
the scientific basis for accepting or reject- for clearer methods of assessing
ing mitigation plans, we developed these "success." Functional, in addition to
recommendations for assessing how well structural, attributes of salt marshes are
constructed wetlands replace the func- being emphasized by both researchers
tions of natural wetlands in southern and resource agencies in southern
California. While other assessment California. A marsh constructed by
protocols have been developed (e.g., California Department of Transportation
Habitat Evaluation Program of the US as mitigation for highway expansion must
FWS; Adarnus and Stockwell 1983; provide self-sustaining populations of
Adamus et al. 1987), there is urgent need plants (including an endangered annual
to tailor procedures to the special environ- hemiparasite), vegetative cover that has
ments of this and region. Just as the list resilience and nitrogen-fixing capability,
of wetland functions needs to be refined and food chain support functions for
for the region's rare and threatened wetland-dependent endangered birds.
habitats, so do the procedures for The assessment protocol includes the
evaluating their presence. analysis of soil processes (decompo-
sition, organic matter accumulation,
nitrogen fixation, nitrogen trapping,
sulfide accumulation) and long-term
2. Rationale for functional monitoring of transplant expansion rates,
assessment reproductive potential of -the endangered
plant, and h abitat- specific uses by
invertebrates, fishes, and birds. In this
What constitutes "successful" mitiga- coastal region with many endangered
tion is highly controversial (Strickland species, functional values of lost wet-
1986, SF BCDC 1988, Zedler 1988a,b), lands must be replaced in accordance with
and judgments may differ depending on new and stricter methods of assessing
the evaluation criteria used and the refer- 11success."
ence data or comparison sites. If mitiga-
tion projects are erroneously judged to be
successful, natural resources may be lost

Strategies for wetland construction, restoration and enhancement

a. Provision of habitat for wetland-
3. Objectives of assessment' dependent species
b. Support of food chains
Scientists and managers recognize c. Transformation of nutrients
three classes of functional values for the d. Maintenance of plant populations
nation's wetlands (Adamus and e. Resilience (ability to recover from
Stockwell 1983)- hydrologic functions disturbances)
(e.g., flood peak reduction, shoreline f. Resistance to invasive species (plant
stabilization, groundwater recharge), or animal)
water guality improvement (sediment 9. Resistance to herbivore outbreaks
accretion, nutrient uptake), and food h Pollination
chain support (habitat and food, espe- i Maintenance of local gene pools
cially for commercially important fish and j Access to refuges during high water
shellfish and for esthetically appreciated k Accommodation of rising sea level.
birds). Many of these functions are less
important in southern California's salt a. Provision of habitat for
marshes (National Wetland Technical wetland -dependent species. For a
Council 1985, Onuf et al. 1978), because region with many rare and threatened
they are small in size (with little area to species, this is the most valued coastal
slow flood waters), and located on the wetland function. Several species of
coast and not upstream of potable water birds, mammals, insects, and a few plant
supplies. species are frequent management targets,
with maintenance of the entire ecosystem
For southern California, the decline in recognized as an essential management
quantity and quality of our coastal wet- goal.
lands has increased the importance of
providing habitat for communities of For the light-footed clapper rail
organisms that can live nowhere else; (Rallus longirostris levipes), the lower
several wetland species, including plants, marsh provides tall, dense cordgrass for
invertebrates, fishes, birds, and mam- cover and nesting sites, frequent tidal
mals, are threatened with extinction. inundation to deter mammalian predators,
Although providing habitat for rare and litter in the form of weaving materials for
endangered species is often the manage- nest construction, sufficient area for
ment goal, mitigators have not been territory establishment, and access to
required to guarantee their presence' food. In addition, the birds need higher
Rather, their "habitat" has been the goal elevation refuges during times of extreme
of several restoration and mitigation pro- high sea levels.
jects. In addition, there is considerable
habitat value for migratory birds that rest For Belding's Savannah sparrow
and feed in coastal wetlands (Onuf et al. (Passerculus sandwichensis beldingi) the
1978, Onuf and Quarnmen 1985). mid-to-upper marsh provides territorial
males with singing perches, females with
The need for strict assessment criteria suitable nesting sites and materials,
follows from the fact that there is so little proximity to food supplies, and distur-
information on how well constructed bance buffers. Sparrows also need
wetlands carry out the functions of their refuges during high sea level.
natural, southern California models.
Eleven functions of wetlands are consid- For salt marsh bird's beak
ered essential for restoration success (Cordylanthus maritimus ssp. maritimus)
(from Zedler et al. 1988, 1990): the upper salt marsh provides a regenera-
tion niche, (an annual supply of open
space for germination of this annual,
sensu Grubb 1977) and suitable host
species for the seedlings to parasitize.

Strategies for wetland constructiong restoration and enhancement

Adjacent upland habitats are needed to ing microbial activities, and disrupting the
support pollinator insects. soil surface. Benthic molluscs, worms,
and crustaceans consume foods produced
For wandering -skippers (Panoquina in, and tidally transported through, the
errans, a rare butterfly), the salt marsh marsh. Suspension feeders filter particles
provides dense patches of saltgrass from the water. Deposit feeders scrape
(Distichlis spicata), on which the larvae the soil surface and deposit fecal pellets
are host specific. and/or middens. Fishes and birds con-
sume algae, detritus, and invertebrates, As
b. Support of food chains. well as frogs and other fishes.
While Pacific Coast wetlands probably
are less important to coastal fisheries than c. Transformation of nutrients.
Atlantic Coast wetlands, shallow tidal These activities include microbial and
creeks do provide nursery grounds for chemical processes controlling the
Pacific halibut (C. Onuf, US FWS, pers. concentrations of nutrients and other
comm., S. Kramer, UCSD, pers. compounds and facilitating the biogeo-
comm.). Nordby (SDSU, pers. comm., chemical cycling of nutrients and the flow
Zedler and Nordby 1986) believes the of energy. Microbes play an important
fishery values are primarily in providing role in nutrient dynamics. Cyanobacteria
food for higher trophic levels, including (blue-green algae) and soil bacteria fix
terns, herons, and other fishes (e.g., nitrogen; these supplies may be essential
halibut, diamond turbot). The endan- for plant growth during times of low
gered California least tern (Sterna nitrogen influx. Nitrification produces
antillarum browni) depends on shallow nitrates, which are available for plant
waters for food. Elsewhere in the nation, uptake. Nitrates are also. substrates for
the nursery function that serves fish and denitrification, and a high nitrification rate
shellfish populations is often the foremost is central to rapid nitrogen cycling.
management goal. The support function Bacteria release nitrogen as gas (dehitrifi-
is perceived as follows: cation), thereby reducing con-centrations
of this nutrient during times of excess
Marsh plants produce organic carbon, influx or accumulation. Bacteria reduce
making it available to consumers and sulfates to sulfides (sulfate reduction)
decomposers. Epibenthic algae produce under anaerobic conditions; this may be
dissolved organic carbon (absorbed by followed by the precipitation of iron
invertebrate larvae) and highly digestible sulfides. Sulfate reduction thus plays a
biomass (consumed by invertebrates and significant role in organic matter
fishes). Macroalgae provide attachment decomposition; these microbial reducers
sites for topsmelt (Atherinops affinis) are available as food, and the reduced
eggs. Vascular plants produce organic sulfides store energy for other bacteria.
material for decomposers and herbivo-
rous insects; they also house many Nutrient transformations are not well
species of insects and spiders. Two known in Pacific coastal wetlands.
classes of decomposers modify the Because plant productivity of Atlantic
vascular plants; shredders break up dead Coast and Gulf of Mexico salt marshes is
plant, material, facilitating leaching and often nitrogen limited (Odum 1988), this
microbial growth, and microbial element has been the focus of most
organisms attack particles of litter assessments of nutrient dynamics.
(decomposition), making the detritus Recent studies suggest that nitrogen limits
more nutritious for higher consumers. plant growth and affects species interac-
tions in southern California (Covin
The burrowing benthic animals (clam 1984). Nitrogen dynamics should thus
species, ghost shrimp, polychaetes, be considered in evaluating the function-
arrow gobies) mix the sediments ing of constructed marshes. At present,
(bioturbation), aerating the soil, enhanc- little is known about nitrogen transforma-

Strategies for wetland constructiong restoration an& enhancement

tions in southern California salt marshes. plants to a specific date. Short-term
Although nitrogen fixers are present objectives, such as a requiring only the
among the abundant algal mats and near survival of transplants and not their
plant roots (rhizosphere), it isnot known establishment, persistence, and spread,
if the fixed nitrogen is available to allow -contractors to be freed of respon-
vascular plants. It is likely that processes sibility, even when the intent of
such as nitrogen fixation and denitrifica-, restoration is not fulfilled.
tion are limited by organic matter avail-
ability in the soil. If this hypothesis is A functional criterion, such as requir-
true, then the presence of organic matter ing self-maintaining populations, is
can determine the success of wetland needed. Self-maintenance requires that
restoration or creation. The techniques conditions favoring persistence, such as
for evaluating those processes are not adequate nutrients.(and processes such as
simple and must be standardized. Thus, nitrogen fixation) be present. Longer-
establishing methods for measuring term monitoring, and measurements of
nutrient transformation functions is an vegetative expansion and seed produc-
important part of our current research tion, should be employed.
program-
Persistence of plant populations
Nutrient dynamics are strongly linked develops through three mechanisms. The
to soil organic matter, which stores marsh substrate maintains seed banks;
nutrients and provides organic substrates these are especially important for short-
for bacteria involved in nitrogen fixation, lived, non-rhizomatous species, e.g.,
denitrification, and the sulfur cycle. Salt bird's beak, annual pickleweed
marsh vegetation tends to be nitrogen (Salicornia bigelovii ), sea-blite (Suaeda
limited throughout coastal environments, callfornica) allowing recovery from
including southern California (Covin and mortality events. Perennial species may
Zedler 1988). Floods provide an influx persist in part through longevity of
of nutrients, but this source is infrequent individuals. The rhizornatous species
and undependable in our Mediterranean- persist belowgound, even though indi-
type climate. It is unlikely that all of the vidual stems may die each year.
nitrogen required of marsh vegetation can Vegetatively reproducing species maintain
be supplied by tidal waters (Winfield potential for expansion of clones.
1980). Hence, nitrogen fixation is
important for salt marsh plants, especially e. Resilience. The region's high
for a region with hypersaline soils, where environmental variability leads to the need
plants may require extra nitrogen to for constructed wetlands. to be resilient,
regulate water uptake--some accumulate i.e., able to recover following extreme
amino.acids. as osmolites. The potential events, as well as human disturbances.
for nitrogen-fixation is higher where soil Coastal wetlands are subjected to natural
organic matter content is high (Zalejko hydrologic alterations, such as flooding
1989). The sulfur cycle is also affected and closure to tidal action. Human
by low organic matter (Cantilli 1989). impacts include street runoff, inflows of
Man-made wetlands are characteristically fertilizers, pesticides and toxic wastes,
lower in soil organic matter (Lindau and mechanical damage to vegetation,
Hossner 1981, Shisler and Charette increased sedimentation, -and encroach-
1984, Broome et al. 1986, Swift 1988, ment by pets.
R. Langis et al. in press), so the nutrient
dynamics of artificial wetlands may limit Alterations to a wetland's hydrologic
the development of vegetation. regime affects nearly every aspect of
ecosystem structure and function. In
d. Maintenance of plant 1984, Tijuana Estuary was closed to tidal
populations. "Successful restoration" flushing for 8 months--a result of long-
is often judged by the survival of trans- term human disturbances to the barrier

Strategies for wetland construction, restoration and enhancement

dune and storm-induced overwashing in dation. Organic soils that are wetted by
1983. The estuarine marsh became dry seawater retain some buffering capacity;
and extremely hypersaline (Zedler and they may prevent the development of acid
Nordby 1986). The clapper rail popula- sulfate soils upon exposure of sulfide-
tion neared extinction as its food and rich (H2S, FeS, FeS2) sediments. The
vegetative cover died out. After 5 years, organic matter provides an energy source
recovery is not yet complete--the annual for nitrogen fixers and decomposers.
pickleweed persisted only in a single, tiny
patch; cordgrass has not regained all of its f. Resistance to invasive
former distributional range; many of the species (plant or animal). The continual
macroinvertebrates failed to recolonize; threats of disturbance to topography and
the rail population has only half the num- salinity lead to the need for constructed
ber of nesting pairs as in pre-closure wetlands to resist invasive species (exotic
years. to the region or alien to the habitat). Once
established, many invasive species can
From this whole-system catastrophe, persist. A single season of altered
we learned that our remnant wetlands hydrology may be enough to allow estab-
have limited resilience. Because regional lishment and long-term presence of
biodiversity is already in jeopardy, it is unwanted species. The cattail (Typha
essential that any artificial wetlands be domingensis) is a local native of brackish
planned with that biogeographical consid- marshes that invaded the San Diego River
eration in mind, Persistence of species, salt marsh when the period of freshwater
and maintenance of the gene pool is a inflow was artificially prolonged by.
regional function. The extremes that reservoir discharges in 1980 (Zedler and
individual wetlands experience may Beare 1986). It was still present in 1990,
eliminate entire populations; for that even though current salinities would not
system to recover, propagules must arrive allow its establishment. Furthermore, a
from another. The presence of species in moderate increase in freshwater inflow at
nearby wetlands should speed recovery this time could stimulate vegetative
by increasing the availability of expansion now that the species is present
propagules. Such refuges, are especially in the marsh.
important for plant species that do not
develop persistent seed banks (e.g., The hydrologic regime (salinities and
Salicornia bigelovii). soil moisture regimes) usually precludes
establishment of species foreign to the
In order for the fish and benthic salt marsh, such as cattails (Typha
macroinvertebrate communities of a con- latifolia) and bulrushes (Scirpus spp.).
structed wetland to be resilient, there The mature marsh sod and vegetation
must be access to larvae; thus the site canopy function. to reduce seedling sur-
must have hydrologic connections with vival of exotic plants such as rabbit-foot
other systems in the area. For the insect grass (Polypogon monspeliensis) and
community to be resilient, constructed brass buttons (Cotula coronopifolia).
marshes should not be too isolated. In a Constructed salt marshes may lack suffi-
region where extreme and sometimes cient vegetative cover to prevent inva-
catastrophic events can't be avoided, it is sions. If wetlands are constructed in
essential that the wetlands retain connec- urban areas, freshwater inflows may be
tions with other systems to aid recovery augmented by street runoff, and soil
of sensitive populations. salinity regimes may favor germination
and seedling establishment of brackish
The accumulation of organic matter in marsh species. If the hydrologic regime
soils also plays an important role in con- cannot be corrected, weed control mea-
ferring resilience. Organic soils retain sures must be planned and implemented
moisture that may help plants survive as aliens appear. Most invaders will be
periods of infrequent or brief tidal inun-

Strategies for wetland construction, restoration and enhancement

easiest to manage if individuals are hand
pulled early in the invasion process. When the functions that confer
resistence to outbreaks are lacking and
Exotic animals are also of concern. herbivores get out of control, the vegeta-
Mosquitofish (Gambusia affinis) and tion is negatively affected. A cordgrass
sailfin molly (Poecilia latipinna) have (Spartina foliosa) planting on a dredge
been recorded from coastal wetlands in spoil island in San Diego Bay looked like
the San Diego area. The yellowfin goby a "successful" marsh for the first three
(Acanthogobiusflavimanus) is a current years; then, a population of scale insects
threat that appears to be moving south (Haliaspis spartina) irrupted and.nearly
along the California coast with extreme decimated the vegetation. Kathy
rapidity. A benthic, mussel from Japan Williams (SDSU, pers. comm.) attributes
(Musculus senhousia) has invaded the outbreak to a lack of predators. In
subtidal habitats of both Mission Bay and natural marshes a coccinelid beetle
San Diego Bay, where it seems to (Coleomegilla fuscilabris) appears to
displace native bivalves (D. Dexter, control outbreaks --under experimental
SDSU, pers. comm.). In studies along conditions, it feeds voraciously on scales.
San Diego Bay, comparing a constructed Neither parasites nor predators were
wetland with a natural marsh remnant, S. deliberately introduced to the marsh, and
Rutherford trapped many individuals of they did not become established on their
the exotic mussel in the artificial marsh, own. This was our first experience with
but found only one in the natural system. an insect outbreak, and no one anticipated
Once established, such exotics may it. We learned that resistance to defoliat-
preclude development of the native fauna. ing herbivores is an important function of
natural marshes. Because it may take
We do not know what defense pest species several years to cause notice-
mechanisms natural wetlands may have to able mortality or defoliation, the success
prevent or slow invasions of exotic of a marsh construction project cannot be
species. Where native species are already determined without long-term study.
abundant, there may be insufficient food
or space for invaders. The salinity h. Pollination. While many salt
regime may prevent survival or settling of marsh plants lack showy flowers and are
larvae that are alien to the salt marsh, wind pollinated, a few species rely on
such as mosquitofish. A dense insects. The annual salt marsh bird's
community of benthic organisms may beak (Cordylanthus maritimus ssp.
filter exotic larvae from the water column. maritimus) is pollinated by bees and flies,
on which this endangered species relies
g. Resistance to herbivore for successful seed set. For such insects
outbreaks. In native wetlands, insect to be present, their nesting habitat must
herbivores are diverse and abundant, but be also be available nearby--for the 5
population outbreaks are uncommon. species that potentially pollinate bird's
Native predators, including birds, carniv- beak (Lincoln 1985), salt flats and higher
orous insects, and parasitic wasps, keep ground with mammal burrows are the
herbivore populations in check. Where likely nesting sites.
such predators are lacking, herbivore
populations may go out of control. Brian Fink successfully transplanted
Herbivores also respond to the nutrient the bird's beak to a restored higher marsh
status of plants. Where nutrient-rich area at Sweetwater Marsh in 1990;
wastewaters flow into coastal wetlands, however, flowers did not produce seeds.
the nitrogen status of the salt marsh plants He attributed the population failure to lack
improves, and insect grazer populations of habitat for the pollinators (Fink,
may expand before predators are able to SDSU, pers. comm.).
crop the increased prey base..

Strategies -for wetland construction, restoration and enhancement

then the entire intertidal marsh will be
L Maintenance of local gene flooded with two feet of water or more.
pools. A species may be maintained Where a habitat refuge is lacking,
through transplantation or other means of survival through extreme events is
artificial propagation, but its genetic uncertain. While the refuge function of
integrity may be modified in the process. buffers needs further quantification, it is
Nearly every author for EPA's regional clear that buffers are necessary during
restoration reviews (Kusler and Kentula storm periods and for future intertidal
1989) expressed concern that local gene habitat as sea level rises.
pools be maintained in nature.
k. Accommodation of rising
The gene pool is threatened in at least sea level. Park et al. (1989) estimate
two ways--when local genetic diversity is that a one-meter rise in sea level would
not salvaged and when alien material is eliminate 65% of the coastal marshes and
brought in from other regions. In the swamps of the contiguous US, because
first case, the failure to retain a broad the upward migration of wetland ecosys-
representation of local plants or animals tems is restricted by coastal development.
may result in a wetland with low genetic Such a rise is probable by the year 2100.
diversity, with reduced material for In southern California, even a half meter
natural selection. In the second case, rise, would wipe out closer to 100% of
local gene pools may be reduced through the remaining wetlands, because the
competitive exclusion that would not upland transitional habitats needed to
normally occur. support the rising marshlands are all
developed. We're probably stuck with
Genetic integrity is not only important rising sea level, so restoration planning
for its own sake. Long-term maintenance should accommodate it. Instead of grad-
of the habitat may be at risk. A plan to ing an entire site to intertidal elevations,
.transplant cordgrass from San Francisco transitional habitats should be left, and
Bay to San Diego Bay, 500 miles to the connections to upland topography should
south, was criticized by PERL biologists be maintained. As Bill Niering of
who argued that the northern populations Connecticut College wrote (in Kusler and
might be less tolerant of high tempera- Kentula 1989), "Our basic goal is to
tures and hypersalinity. An extreme create a persistent functional wetland
event could wipe out the imported system. In some situations this may be
material, perhaps years after establish- more important than the creation of a
ment. A local source of cordgrass that specific wetland type since the present
was destined for destruction was physiognomy may merely be a
identified and used instead. momentary expression of the system's
vegetation potential in the future."
j. Access to refuges during
high water. California's Coastal Act
requires a 100-foot buffer around wet-
lands to reduce disturbances to the
protected ecosystem, although some have
considered bike trails and parking lots
compatible uses within the buffer zone.
The importance of having a functional
refuge is evident during extreme high
waters, when clapper rails and other
animals need to escape their flooded
habitat. Storms can raise water levels
well above predicted tidal heights, and if
storms coincide with the highest spring
tides of the year (as in January 1988),

Strategies for wetland construction, restoration and enhancement

cattails), and resident natives that undergo
population irruptions not seen in natural
4. Criteria for wetlands (e.g., scale insects on
"successful mitigation" cordgrass). Annual surveillance will be
needed for problem species, so that
as required in.San Diego Bay corrective measures can be initiated early
enough to be effective.

How similar should the functions of Finally, we recommend that an
constructed and natural wetlands be objective scientific panel review the
before the project is considered sampling or monitoring program used to
successful mitigation? In order to insure judge compliance/success. For the
no net loss of habitat values, it is Caltrans Connector Marsh, the US Fish
necessary to have high expectations for and Wildlife Service anticipated the need
wetlands constructed or modified for the for interim reviews of the mitigation
purpose of mitigating losses elsewhere. program. The federal agencies and
In the past, standards have varied from contractors must meet with the Service
project to project. Where standards are annually to review project status and
lax, there is greater risk that efforts to recommend any remedial actions (US
retain or improve upon wetland values FWS 1988, p. 23).
will also be inadequate.
In a recent decision, the US Fish and
High standards are necessary if Wildlife Service prescribed detailed
projects are to comply with agency objectives, including functional attributes,
policies, and detailed criteria for for evaluating the success of the
assessing success are needed to document mitigation project at the Connector
in detail what is happening on the Marsh, constructed by Caltrans in Chula
restoration site. In addition, scientific Vista. Three endangered species occur in
studies should be included to identify the affected area, the light-footed clapper
what causal factors are responsible. If, rail, the California least tern, and the salt
for example, nutrient concentrations are marsh bird's beak. When the Sierra Club
unusually low, an understanding of the charged the federal agencies with failure
reasons will help in identifying corrective to enforce the original mitigation
measures and in preventing similar requirements (primarily a transfer of land
problems in the future (Broome et al. from private to public ownership,
1987). subsequently ordered by the court;
Thompson 1988), it became possible to
Obviously, for a wetland designed to revise and update the "Section 7
provide habitat for use by endangered Consultation" under the Endangered
birds, judgments of success should Species Act, strengthening the require-
include the presence of those populations ments both for mitigation and for judging
and perhaps their self- maintenance. its success.
Annual censusing of rare and endangered
birds is needed plus 5- and 10-year The new requirements are set forth in
follow-up studies of the plant and animal the "Biological Opinion" (US FWS 1988)
communities and ecosystem functions for the Combined Sweetwater Flood
that develop on site. Control and Freeway Project. "The
Service shall deem that the wetland
Special attention should be paid to creation and modification projects are
problem species. At least three categories successful on showing that the channels
have been identified in constructed and emergent wetlands provide suitable,
wetlands: invasive exotics (e.g., rabbit- functional habitats for the California least
foot grass), invasive natives that are alien tem and light-footed clapper rail, and the
to the desired wetland community (e.g., emergent wetlands are also vegetated by

Strategies for wetland construction, restoration and enhancement

patches of salt marsh bird's beak and 75
percent of the native species currently 5. Reference wetlands and
occurring in the Sweetwater River reference data sets
Wetlands Complex." (ibid., p. 23) The
Service went beyond this general goal In order to compare constructed wet-
and provided specific criteria as follows: lands with natural ecosystems, there must
Channels need to provide "suitable be data on the structure and functioning
habitat for the California least tem and the of representative natural wetlands.
light-footed clapper rail." Forage items Because most southern California wet-
lands have been disturbed, there are no
need to be present for 2 years at levels of pristine wetland examples. Even in
75 percent of the density and diversity of nearby Baja California, where coastal
the prey base in comparable habitats with development is less extensive, there are
the Sweetwater River Wetlands Complex few examples of undisturbed habitat.
(SRWC). Wetland habitat for rails must Selection of wetlands to serve as models
include 7 home ranges for 2 years, each for restoration must therefore be based on
with non-overlapping areas of 2-4 acres, an understanding of how disturbance
including low, middle, and high marsh. affects wetland ecosystems. Two addi-
Lower marsh in each home range must tional features of Southern California
have at least I patch of cordgrass of 60- wetlands make it difficult to characterize
80 cm height and 90-100% cover that is the reference wetland or restoration tar-
go-100 m2 in size and able to maintain get. The first is spatial heterogeneity
itself (i.e., in place for 3 yr and with N- within and between wetlands; the second
fixation). The middle marsh must have is interannual variability, due to both
>70% cover with 75% of the typical natural and man-caused events.
native species at SRWC. The high marsh
must have <20% cover of weedy species The problem of spatial vari-
and maintain 5 separate patches (of I m2 ability. No two wetlands are identical in
and at least 10 m part, with >20 species lists, distributions, or rates of
plants/patch) of bird's beak that are self- various processes. Yet the spatial vari-
sustaining (stable or increasing in number ability among the region's wetlands is
and area) for 3 yr. An attempt should be more of an aid than a hindrance in under-
made to include high saltmarsh berms standing cause-effect relationships.
near areas of low saltmarsh. Current knowledge of wetland structure
These new requirements represent a and function comes from long-term
experience and study in several wetlands,
major advance in agency expectations for with each site contributing unique
successful mitigation. Their effectiveness information.
in insuring the replacement of in-kind
habitat will be evaluated along with the Four main study sites have con-
constructed habitat. tributed to our understanding, with the
longest period of record at Tijuana
Estuary (Table 11. 1). The results of these
studies indicate that no one wetland can
serve as a reference site for any man-
made habitat. Information from con-
structed wetlands needs to be compared
with the total available information,
preferably by a panel of scientists familiar
with the study sites, to insure thorough,
objective evaluation. Such a process has
been initiated by Caltrans, involving
scientists from the Pacific Estuarine

Strategies for wetland construction, restoration and enhancement

Research Laboratory and resource agency maps of habitat types in southern
staff to help assess wetland functions at California coastal wetlands, based on
the Connector Marsh. 1985 aerial photography. However, their
habitat classifications have not been
The problem of temporal vari- checked in the field. Sharon Lockhart
ability. It has taken several years to used the FWS draft maps to estimate the
understand just a few of the functions of areas of 23 coastal wetlands in San Diego
the Tijuana Estuary salt marsh, largely County. She outlined each of the poly-
because of the high interannual variability gons with a planimeter to obtain areas,
and the importance of extreme events in and summarized the acreage of each
triggering changes in the vegetation. Ten habitat type and each wedand. Acreages
years of data on plant distributions and included areas between the ocean inlet
abundances are still inadequate, because and the first major break in habitat toward
we have not witnessed all types of the east. Freeway 5 was the limit for
extreme events or human impacts. Our wetlands in San Diego and Mission Bays;
knowledge of vegetation control func- El Camino Real was the limit for Los
tions includes relationships with salinity- Peflasquitos Lagoon north to Buena Vista
inundation regimes, but is weak concern- Lagoon. The FWS habitat classifications
ing interactions with nutrients, soil redox were combined into 5 habitat types: bay,
conditions, heavy metals and other toxic channel, salt marsh, brackish/fresh
materials. Yet this is the best data set marsh, impounded waters, and other
available, and it serves to caution man- habitats (riverine and unidentified poly-
agers that high interannual variability may gons).
be the rule, rather than the exception, for
the region's wetlands. Excluding the 5,420 ha of subtidal or
bay habitat, this survey indicates 756 ha
It is thus apparent that short-term of salt marsh, 832 ha of brackish and
assessments must document trends, fresh marsh, 495 ha of impounded
rather than "average states." Monitoring wetlands, 163 ha of channels; and 32 ha
programs must take more than of other wedand habitat types in San
"snapshots" of the system under study; Diego County (Figure 11. 1). However, it
time series are needed to determine the is likely that these are underestimates
direction of developments. Rather than because many wedand areas may not be
focusing on one-time measures of trans- distinct on aerial photos. Comparative,
plant survival (often the only requirement field based data for wetlands at Tijuana
for monitoring mitigation sites), it is more Estuary were obtained in a PERL survey,
important to assess the ability of the sys- which included extensive field sampling
tem to respond to changing environments over large areas of disturbed wetland.
(e.g., expansion following winter stream- The map by PERL ecologists indicated
flow and low soil salinities) and the much larger areas of wedand at Tijuana
likelihood that populations can be main- Estuary than the FWS map (556.3 ha
tained through environmental extremes compared to 284.1 ha). For Tijuana
(e.g., measures of seed production; size Estuary, the FWS draft inventory
of seed bank), to identify any major shifts appeared to be a 50% underestimate of
in the community (e.g., loss of desired wedand area.
plant species), and to develop corrective
measures for problems that are seen For Tijuana Estuary, the areas of
(e.g., eradication of exotic invaders habitat from more detailed mapping,
before they become widespread). including ground truthing, are given in
Figure 11.2. The areas were obtained
The region's wetiand resources using the SDSU Geographic Information
have not been censused in detail. The US System.
Fish and Wildlife Service (National
Wetlands Inventory) has prepared draft

Strategies for wetland construction, restoration and enhancement

Table H.1. A list of sites and types of data currently gathered by scientists associated with
the Pacific Estuarine Research Laboratory. While there are additional data for these and other
sites, only sampling programs that use comparable methods are listed. TE = Tijuana Estuary;
SDR = San Diego River Marsh; LPL = Los Penasquitos Lagoon; CM = Connector Marsh and
PC = Paradise Creek, both on San Diego Bay. Similar vegetation and salinity data have been
obtained at Estero de Punta Banda, Baja California, by Silvia Ibarra-Obando at CICESE,
Ensenada, B.C.

Location: TE SDR LPL CCM P

Soil salinity data X X X X X

Streamflow data (USGS) X X X

Lower salt marsh veg. data Sept. July Sept. summer summer
Number of stations 102 250 86 64 7
Number of years 10 6 2 3 3
Nutrient dynamics data qtrly qtrly
Number of years 5 3 3

Upper salt marsh data Sept. Sept.
Number of stations 115 12
Number of years 4 5

Fish and invertebrate data qtrly qtrly
Number of stations 4-6 3
Number of years 4 5

Figure H.1. Comparison of habitat areas in 23 wetlands of San Diego County. Areas
exclude subtidal bay habitats.

500-

400-

300-
ca

200-

0- 1 F I I I I I I I

COCO :30COOO WW2>'-O
0 A2, 0 -0 - 0 0
0
0 a Mn 2 cc
LLI CD (z 6 .0 :1M LU .5 .2 > 0 0
CQ r Cn 0- 0) M (D C$) 0 -0 a)
0 CV .!2 'V -0 Z-) M a) :3
c (V CD
0
Cn W X: 0 E (L
0 0 E cn tc Im :3
M 0
CO (D0 as CO (V , 2
cn LL (L CI) :3 as 0 LL
CO E 05 Cid
D CO
0 W
0 C/)

Strategies for wetland construction, restoration and enhancement

Table 11.2. San Diego County wetland habitats, deterrnined from US FWS draft maps of
the National Wetland Inventory by S. Lockhart. Data are hectares (1 ha = 2.471 ac). Br/Fr
Brackish/Freshwater.
Saline Br/Fr Impounded
Location Bay Channel Marsh Marsh Waters Other

Tijuana Estuary 0.0 223.6 171.1 88.1 0.0 1.4
San Diego Bay 4483.2 .2 11.0 23.3 430.9 0.0
Sweetwater 0.0 1.9 116.1 18.6 0.0 1.0
Famosa Slough 0.0 2.6 1.5 0.0 0.0 0.0
San Diego River 62.6 0.0 47.1 19.8 0.0 1.1
Mission Bay 620.1 0.0 35.3 39.9 0.0 0.0
Los Pefiasquitos 0.0 3.1 130.7 50.0 0.0 0.0
San Dieguito 30.0 0.0 28.7 36.7 0.0 0.8
San Elijo 0.0 31.6 87.6 74.2 0.0 0.0
Batiquitos 42.7 99.9 8.6 52.7 0.0 0.0
Aqua Hedionda 105.9 0.1 31.4 26.9 0.0 0.0
Buena Vista 0.0 0.0 0.0 15.6 64.6 0.0
Lorna Alta 0.0 0.0 0.0 0.7 0.0 0.0
San Luis Rey 0.0 0.7 0.0 82.6 0.0 2.0
Oceanside Harbor 85.3 0.0 0.2 0.0 0.0 0.0
Santa Margarita 30.4 0.0 87.3 79.0 0.0 0.0
Cockleburr Cyn 0.0 0.0 0.0 2.0 0.0 0.0
French Canyon 0.0 0.0 0.0 6.3 0.0 0.0
Aliso Creek 0.0 0.0 0.0 10.2 0.0 0.0
Unnamed Canyon 0.0 0.0 0.0 3.8 0.0 0.0
Las Pulgas Cyn 0.0 0.0 0.0 46.8 0.0 0.6
San Onofre Creek 0.0 0.0 0.0 8.8 0.0 0.0
San Mateo Creek 0.0 0.0 0.0 56.3 0.0 32.2

Figure 11.2. Area of several wetland habitat types at Tijuana Estuary, as determined by
PERL from detailed ground truthing of the 1986 aerial photo.

Other wetlands
Brackish/Fresh A
Riparian scrub -
Transition
Mudflat
Salt panne-.
Salt marsh -
Channels
Beach
0 100 2@O

Hectares

Strategies for wetland construction, restoration and enhancement

Two major questions will be answered,
6. How should major in sequence, using replicate experimental
restoration programs be marshes: First, how will different
degrees of tidal influence affect pickle-
undertaken? weed marsh? Second, how important is
topographic complexity to wedand func-
It is recommended that major restora- tioning--should small tidal creeks be
tion projects, especially those involving excavated within the marsh plain to pro-
techniques or habitat types for which vide habitat for a diverse macrobenthos
there are no previous examples, begin and food chain?
with an experiment to test methods of
restoration on the site. A small effort in To answer the first question, 24 small
planting vegetation or modifying the tidal marshes (lxlO-m mesocosms;
topography in various configurations, Figure 11.3) will be constructed adjacent
followed by evaluation over a growing to an existing tidal channel. Each will be
season, will help rule out techniques that assigned to one of 6 treatments (4x repli-
don't work in the short-term and will help cation). Soil and vegetation attributes
identify promising approaches that can be (including both vascular plant and algal
developed further producers) will be followed through one
growing season. Study of the ecosystem
All projects should be continually responses will determine the degree of
evaluated with a long-term functional tidal flushing and freshwater inflow
assessment program. An adaptive needed to manage the existing pickleweed
management approach will improve the ecosystem and to construct new pickle-
restoration program through time; that is, weed marshes.
collection of information about system
development, identification of problems, To answer the second question, a 20-
and experimentation with new restoration acre area adjacent to an estuarine channel
techniques, will provide the information will be graded down to approximately 4.5
necessary to incorporate corrective ft MLLW for development as an
measures. experimental intertidal marsh. Replicate
subareas (n=3) will be constructed with
The Tijuana Estuary Tidal Restoration and without tidal creeks. Results from
Program provides a good example. A this experiment (vegetation and soil
plan to restore full tidal flushing to about development, invertebrate colonization,
200 ha (500 ac) of the estuary was devel- and bird use) will guide implementation
oped by two hydrologists, Phil Williams of later phases of the project. Since
and Mitch Swanson (1987). Review of funds have not been identified for the
the initial, "minimum dredging plan" by larger restoration program, it is likely that
PERL biologists indicated that some areas several years of data will be available by
where dredging was proposed were criti- the time results are needed. Meanwhile,
cal habitat for sensitive species. information will add to the technical
Alternative sites for new channels and information base and guide restoration
intertidal marsh habitats were then projects elsewhere.
sought, and a compromise was developed
to reduce biological impacts but require a
more costly dredging program. The
revised plans will soon be available as an
Environmental Impact Report/Statement
(EIR/EIS for Tidal Restoration of Tijuana
Estuary, prepared for the California State
Coastal Conservancy). The EIR/EIS
proposes a low-cost experimental phase
to precede major dredging and grading.

,17

Strategies for wetland construction, restoration and enhancement

Figure 11.3.
-2
Map of 24 tidal mesocosms; at P": '.'0'
A_A_A_ 3 ;'
%
Tijuana Estuary. Each of the
experimental blocks (1-4) includes
6 hydrologic treatments: 4
i;. isle
wt i
3 tidal flushing regimes, each
with and without freshwater inflow.

Reduced tidal flushing Full tidal flushing Prolonged flooding

6
P-1. ' WIT M.... C ...... S, rIki i
Tidal influence restricted Daily tidal inundation Daily tidal inflow,
by one-way flap gate that and drainage drainage impeded by
reduces inflow, but allows height of outlet
drainage

View of tidal mesocosm construction plan

Tide
gate 7.3 ft MLLW

5.3 ft MLLW
I Orn (33ft) lp_@, rn (3.3ft)

Interval view after planting Saficornia virginica
Tide
qate

U
f'r

Assessment Case Study

College Program, Department of Com-
111. Case Study: merce, under grant number NA85AA-D-
SG140, project number R/CZ-82,
Sweetwater River through the California Sea Grant College
Wetlands Complex program, with matching funds from the
California State Resources Agency.

Free sulfide in natural and
constructed salt marshes (from
Our research site is within the largest Cantilli et al. 1989 and Cantilli 1989).
and most controversial (Thompson 1988) The presence of only trace amounts of
mitigation project on San Diego Bay. free sulfide in a man-made marsh on San
The project combines highway widening, Diego Bay suggests that its biogeo-
construction of a new freeway inter- chemical functioning is not equivalent to
change, and excavation of a flood control that of a natural salt marsh.
channel. Mitigation for lost wetland habi-
tat is being carried out by the California Sediments in the artificial system and
Department of Transportation (Caltrans) an adjacent natural marsh were sampled
and includes construction and enhance- near the seaward edge of cordgrass
ment of intertidal salt marsh habitat (cf. (Spartina foliosa) growth. Concentra-
US ACE 1982, Swift 1988) within the tions of free sulfide (112S, HS) in the
City of Chula Vista (32L'10'N, 117L' natural marsh were significantly greater
10'W). The mitigation marshes include than in the constructed marsh (Figure
the "Connector Marsh," which was Ell. 1), with levels up to 3 mM/L at 25-cm
constructed as a hydrologic link between depth. Further, data on cordgrass aerial
Paradise Creek and the Sweetwater biomass from marshes at San Diego Bay
Marsh, and Marisma de Naci6n, a 17- and Tijuana Estuary suggest that free
acre marsh excavated from the "D Street sulfide is not directly phytotoxic. Sulfide
fill" in early 1990. The marsh, channel, is a product of sulfate reduction, a
and creek habitats are within the range of process that typically dominates anaerobic
three birds that are on the federal decomposition and carbon cycling in salt
endangered species list: the light-footed marshes. This process may be limited by
clapper rail (Rallus longirostris levipes), low sedimentary organic matter content in
the California brown pelican (Pelecanus the man-made marsh (Table 111. 1). The
occidentalis), and the California least tern near-absence of sulfate reduction in the
(Sterna antillarum browni) and one artificial marsh may affect energy flow
endangered plant, the salt marsh bird's and export, as well as the retention of
beak (Cordylanthus maritimus ssp. heavy metals and anthropogenic sulfur.
maritimus). Mitigation marshes at this
site must provide functional habitat for
the rail, the tern, and the bird's beak (US
FWS 1988). Table 111.1. Means of percent organic
carbon in marsh soils at depths of 0-5 cm.
While the assessment study is not yet and 5-10 cm. Values were obtained by
complete, results are available for soils, weight loss (dry wt) on ignition at 7000 C
nutrients, vegetation, and epibenthic for one hour.
invertebrates (Langis et al. in press;
Zedler and Langis in press). Compari- Dgpth Natural Artificial
sons of fishes, channel benthos, and
birds are in progress. 0-5 cm 8.14 4.33
The work has been sponsored by 5-10 cm 9.24 3.98
Caltrans and NOAA, National Sea Grant

Assessment Case Study

Figure HI.I. Pore water concentrations of fi-ee sulfides in the artificial (A) and natural
(N) marshes at depths of 5, 15, and 25 cm. Bar I s.e.

0.7-

0.6-

0.5-

E 0.4-
a)
_0
0.3-

U)
0.2-

0.1-

0.0-
A-5 A-15 A-25 N-5 N-15 N-25

Site and Depth

higher-- generally twice as high--in the
Potential nitrogen inputs (N- natural marsh than in the man-made
fixation) in natural and man-made salt marsh for five of six sampling periods
marshes (modified from Zalejko et al. (Table 111.2, Figure 111.2). In September
1989, and Langis et al. in press). Data 1988, there was. no difference between
indicate that nitrogen fixation provided the marshes. The rates of nitrogen in the
more nitrogen in the natural salt marsh natural marsh, although greater than in
(Paradise Creek Marsh) than in the man- the constructed marsh, were still very low
made (Connector Marsh) salt marsh by comparison with Atlantic Coast wet-
located at San Diego Bay. lands. While nitrogen-fixation at this one
natural marsh may not be representative
Using the acetylene reduction tech- of the region, the indication is that nitro-
nique, potential N-fixation rates were gen limits the growth of cordgrass. N-
measured in both the root zone of cord- fixation is one mechanism by which
grass (Spartina foliosa) and on the sedi- nitrogen can be supplied continuously
ment surface in association with blue- without the herbivore-stimulating effects
green algae. Nitrogen fixation was mea- that sometimes accompanies fertilizer
sured at the same intertidal elevation in treatments. Lower N-fixation rates in the
both marshes to eliminate differences due constructed marsh were related to lower
to soil moisture. Rates of N-fixation on soil organic matter levels (see next page).
the sediment surface were significantly

Assessment Case Study

nitrogen fixation. At this study site, the
Rates of N-fixation in the root zone of natural marsh was 2.3 times higher in
cordgrass were often similar for the con- above-ground plant biomass (Langis and
structed and natural marshes (Table Zedler in press). Other studies (e.g.,
111.3). Rates were significantly higher in Covin 1984) have shown that nitrogen is
the natural marsh for only two out of five limiting to salt marsh plant growth. N-
sampling times. Langis et al. (in press) fixation is thus a very important process
attribute the high rates in the constructed because it introduces new nitrogen into
marsh to the low nitrogen concentrations salt marsh ecosystems. A positive
found there. Nitrogen fixation is known feedback control system is suggested--as
to shut down when nitrogen is abundant cordgrass begins growth, organic matter
in the environment. becomes available to N-fixers in the
rhizosphere, and N-fixation is stimulated.
Significant positive correlations were Increased nitrogen levels then stimulate
found between N-fixation rates and per- more rapid growth of cordgrass. On the
cent organic matter and belowground soil surface, similar interactions with
biomass (Figure 111.3) showing the microalgae may be present.
importance of primary producers for

Table 1H.2. Rates of N-fixation in surface cores (1 cm deep) for several constructed and
natural marsh sites, expressed as nmol C2H2/hr/m2. Data are means (and s.e.) for 5-12
samples from the Connector and Paradise Creek marshes adjacent to San Diego Bay. Data
are from Langis et al. (in press).

Constructed Marshes Natural
Date Islands North Bank Marsh

Feb. 1988 15.0 (4.62) 7.3 (4@62) 33.0 (4.30)
Apr. 1988 42.6 (22.74) 15.6 (6.35) 118.9 (32.28)
July 1988 45.4 (2.16) 47.6 (4.94) 70.2 (9.87)
Sept. 1988 34.7 (0.99) 46.5 (7.61) 34.0 (1.15)
Feb. 1989 24.5 (2,1) 26.8 (2.68) 61.5 (8.70)
Mar. 1989 no data 19.5 (0.88) 24.7 (1.49)

Table 111.3. Rates of N-fixation within the cordgrass rhizosphere (10 cm deep) for con-
structed and natural marshes. Data are nmol C2H2ft/m2 as in Table 111.2.

Constructed Marshes Natural
Date Islands North Bank Marsh

Dec. 1987 no data 3.0 (0.51) 7.5 (0.82)
Feb. 1988 105.5 (30.63) 44.3 (16.00) 104.3 (51.31)
Sept. 1988 34.7 (0.99) 46.5 (7.61) 34.0 (1.15)
Apr. 1988 96.2 (20.56) 77.0 (17.60) 161.9 (30.78)
Sept. 1988 251.7 (20.78) 251.8 (40.56) 73.7 (9.54)

Assessment Case Study

Figure 111.2. Mean N-fixation
(acetylene reduction) on the 120-
sediment surface at individual sites
in the artificial and natural salt 100-
marsh during July 1988. Units
are C2H4 nmoles/g soil per hr
(*10-4). Data are from Zalejko 80-
(1989) and Zalejko et al. (1989).
60-

40-

20-
Artificial Marsh Natural
Marsh

0.012- (a) R = 0.82
Figure 111.3. Relationships P = 0.0001
between rates of N-fixation
(acetylene reduction) and (a) soil 0.010-
organic matter content and (b)
belowground biomass. Each point 0.008-
represents an individual
measurement. Units are C2H4 0.006-
nmoles/g soil per hr (* 10-4). Data
are from Zalejko et al. (1989). 0.004

0.002
0.1 0.2 0.3
Organic Matter (% dry wt)

40- (b) R = 0.55
P = 0.005

30 -

20 -

10 -

0
0 2 4
Belowground Biomass (g dry wt,'

Assessment Case Study

higher (Figure 111.6), and percent organic
Nitrogen dynamics in natural vs. matter in sediments was 2.4 times higher in
man-made salt marshes (modified from the natural marsh. Total nitrogen in soils
Langis and Zedler 1989, and Langis et al. was highly correlated with percent organic
in @press). Nitrogen levels in sediments, matter (Figure 111.7).
pore-water and above-ground vegetation in
the constructed salt marsh were compared Clearly, nutrient dynamics in the
with those in the natural Paradise Creek constructed marsh are not comparable to
Marsh. Sampling sites were the same as those of the natural marsh. Since in both
for nitrogen fixation (above). marshes sediment nitrogen levels (TKN)
were correlated with percent soil organic
Nitrogen levels (ammonia, nitrate plus matter, we believe that belowground
nitrite, and total Kjeldahl nitrogen [TKNI) organic matter must accumulate in soils for
were significantly higher (P<0.01) in the nitrogen levels to reach those of the natural
natural than in the man-made marsh in all marsh.
the compartments examined. Phosphorus
levels were likewise higher in the natural Through time, we expect the marsh soil
marsh (Figure IIIA). The magnitude of to develop and for organic matter and
differences in nitrogen concentrations nutrients to accumulate. However, in two
between the natural reference marsh and the years of sampling nitrogen concentrations,
constructed marsh were as follows: 2.4 there was no indication of an increase.
times higher for sediment total N and Much longer-term studies are needed to
extractable NH4; 9.4 times higher for poTe- determine how rapidly wetland soils can
water NH4 (Figure 111.5) and 0.2 times develop and how long it may take for
higher for foliar nitrogen in cordgrass. constructed marshes to achieve the nutrient
Aboveground biomass of plants was much status of natural marshes.

Figure IHA Total nitrogen (TKN) and phosphorus determined in sediment core samples
in May 1988 from the natural and man-made marshes. Differences in total nitrogen means
were highly significant (P<0.001) while total phosphorus values- were not. The low N/P
ratios are indicative of nitrogen limitation. Bar = 1 s.e.

NATURAL

V MAN-MADE
-P
E

0
TOT-N TOT-P TOT-N TOT-P

Assessment Case Study

Figure IH.5. Ammonium levels of pore-water samples collected September 16 and
November 15, 1988. Differences between sites were highly significant (P<0.001). Bar1
s.e.

4-
NATURAL

E 2-

MAN-MADE

0
SEPT16 NOV15 SEPT 16 NOV 15

Figure M.6. Above-ground biomass of cordgrass collected July 15, 1988 from 0.25-m2
quadrats. Differences between sites were highly significant (P<0.001). Bar I s.e.

140-

120-

C4 100-
E
La
C4 80-

60-

40-

20-

0
NATURAL MAN-MADE

Assessment Case Study

Figure 1H.7. Regression -of sediment total nitrogen (TKN) over percent organic matter.

5
y 0.1094 + 1.1058x R = 0.97

4-
W

3
A 0

Cr
0
0-

0
0 1 2 3 4.
TKN (mg/g so! I dry wt)

Epibenthic invertebrate distri- After four years, this marsh had not
butions in natural vs. man-made salt developed its natural food chain support
marshes (modified from Rutherford and function. As cordgrass cover increases,
Zedler 1989; details in Rutherford 1989). it should facilitate recruitment of the
There were significantly more individuals invertebrate community; however, it is
(2-4x) and more invertebrate species in a not yet clear whether the constructed
natural marsh at San Diego Bay than in marsh provides the required quantities
the 4-yr-old man-made marsh, comparing and qualities of food for native
the same low-marsh elevations (Figures invertebrate populations to develop their
111.8-111.9). All animals were trapped in natural abundances and to persist in
litterbags filled with dried cordgrass. The perpetuity.
most abundant species was a larval
Dipteran, Pericoma sp., which was
significantly more abundant (up to 9x) in
the natural marsh (ANOVA, p< 0.05).
An anemone, Diadumene franciscana,
was found only in the natural marsh. In
the man-made marsh, there were
significantly more Hemigrapsus crabs in
all sites sampled. In the natural marsh (at
ca. 0.3 m above MSL), areas with 80-
100% cover of cordgrass supported twice
as many invertebrates as areas with 0-
20% cover. At an elevation of 0.5 m
above MSL, the numbers of species and
individuals were similar for areas with Polychaete worm
high cover at the two marshes.
(Nephthys caecoides)

Assessment Case Study

Figure 111.8. Mean number of individuals per litterbag in areas of low elevation with
high (80-100%) vs. low (0-20%) cover of cordgrass vs. high elevation with high cover.

300- MAN-MADE
NATURAL

CO 200-
M
W

> 100-
a
z

LL
0

0-
lo elev./hi cov lo elev./Io cov hi elev,

MARSH SITE

Figure 111.9. Mean number of species per litterbag in areas of low elevation with high
(80-100%) vs. low (0-20%) cover of cordgrass vs. high elevation with high cover.

MAN-MADE
12 NATURAL

10-

8-
Uj

IL
CL
Cn 4-
U.
0
2

0
lo elev./hi cov lo elev./lo cov hi elev.
MARSH SITE

Assessment Case Study

Channel fishes and benthic Connector Marsh: yellowfin gobies
invertebrates. The requirements for (Acanthogobius flavimanus) and sailfin
functional equivalency of the channel fish mollies (Poecilia latipinna). With the
and invertebrates in the Biological exception of one sailfin molly found in
Opinion (US FWS 1988) were that the F/G Street marsh, these species were
constructed marshes provide 75% of the found exclusively in the constructed
species and 75% of the individuals found marshes. Generally, opportunistic
in the reference marshes. An extensive nonnative species such as these are much
sampling program was designed to more successful in colonizing disturbed
understand how the constructed channel habitats. Little is known about the
system has developed and to assess its specific impact that these species have on
functioning in relation to natural the native species. Yellowfin gobies are
channels. identified as voracious predators, and it
may be assumed that they have some type
Channel organisms were sampled at of impact on the native community
three stations in the Connector Marshes through predation. Likewise, sailfin
and four in the natural (reference) mollies are very similar to killifish in their
marshes within SRWC (Table 111.4). life history, and may have a competitive
Fish and invertebrates were sampled in effect on this species. The effect of these
June 1989, October 1989, and January species on the channel community should
1990. At each station, blocking nets be investigated further.
were used to create a sampling area
approximately 10 rn long. A beach seine The sediments and channel
was repeatedly pulled through the sample configuration of each station differed,
area until the fish catch per unit effort providing a spectrum of habitat types
declined. Individuals were identified, within the wetlands complex (Table
measured and released. Benthic IIIA). Because of this, a generalized
invertebrates were sampled adjacent to comparison of the Connector Marsh
each station with 9 cores using a clam stations with all of the reference marshes
gun (0.018 m2 surface area) to a depth of would be insufficient to evaluate the
20 cm. Samples were preserved and specific stations. Instead, a similarity
retained for laboratory identification. index was used to compare the individual
sites.
The fish community. In com-
paring the total fish caught over the three For each sampling station, the
sampling dates, the natural marshes numbers of individuals sampled was
contained 15 species of fish. The Con- totaled for each species over the three
nector Marsh stations, when combined, sampling dates. The relative abundance
contained all of these species and more. of each species was calculated for each
The average number of individuals caught station (Table 111.5). From these relative
at the connector marsh stations over the abundances, an index of % similarity was
three sampling periods was 630, calculated (Sim. = 1[min. a,b], where a
compared to an average catch of 610 fish and b are relative abundances in the two
at the reference marsh stations. This samples being compared).
indicates that for year 5, with 3 sampling
periods, the Connector Marsh has 100%
of the species and similar numbers of
individuals as are found in the reference
marshes at SRWC.

The evaluation of functional
equivalency is tempered by the
occurrence of two exotic species in the

Assessment Case Study

Table HIA Description of channel stations at the Connector Marsh and other sites of the
Sweetwater River Wedand Complex (SRWC).

DescriRtion
Constructed channels
(Connector Marsh, CM)

Stn.2 South CM Coarse mud bottom. Steep bank on one side, gently sloping
east channel on the other. Max depth at low tide is approx. 0.75 m.
Occasional deeper areas with 50 gal. barrels on the bottom.
Stn.3 South CM Soft mud, sloping banks on both sides. Completely intertidal
.west channel (max depth at low tide is approx. 0.0 m).
Stn.7 North CM Soft mud, with medium sloping banks on both sides.
Max depth at low tide is approx. 1.0 m.

Natural channels

Stn. 1 . Isla Flaca, Sandy mud with gentle, slope on one side and steep, eroding
Eastern tip bank on the other. Max depth at low tide is approx.
0.75 rn. Sweetwater River flows diectly out this course.
Stn.4 Sweetwater Marsh, Soft mud/silt channel; some areas with dense clarn shells (dead).
main channel Steep banks up to the marsh plain on both sides.
Max depth at low tide is approx. 1.25 m.
Stn.5 E St. Marsh, Soft mud channel with varying depths. Generally steep banks
main channel on both sides to a marsh plain. Max depth at low tide is
approx. 0.75 m.
Stn.6 F/G St. Marsh, Very soft mud/silt bottom. Low bank that is steep on one side
main channel and sloping on the other. Sluggish but regular tidal
flushing. Maxdepthatlow tide is approx. 0.75 m.

Comparisons of the fish communities
within the Connector Marsh (Table 111.5)
indicate 72% similarity (of relative
abundances) between the two stations in Table 111.5. Similarity among
the west channel (South west and North Connector Marsh stations.
west). Both western channel stations Comparisons are based on relative
were dominated by topsmelt. These abundances of individuals for 3 seasonal
similarity comparisons also indicate that samples in 1989-90. (Sim. = Y,[min.
the fish community in the eastern channel a,b], where a and b are relative
(South east) is only about 20% similar to abundances in the two samples being
each of the other CM stations. The compared)
eastern channel is, shallow, has a soft-
bottom, and is dominated by arrow CM Stations % Similarijy
gobies. The eastern channel has changed
considerably since construction of the South west x North west 72.2%
Connector Marsh. Prior to 1984, it was Southwest x Southeast 20.4%
the main tidal channel that directed flows. Northwest x Southeast 20.8%

Assessment Case Study

to Paradise Creek. After dredging of the Table 111.6. Similarity of fish
western channel, flows shifted to that sampled in natural and constructed
deeper course, and sediments accreted in wetland channel stations. Data are
,the east channel. It is understandable that for three seasonal samples in 1989-90.
the eastern channel has a different fish Similarities are based on relative
community from the deeper channel that abundances of species, as in the previous
was recently constructed. table.

The comparison of CM sampling CM Sweet- E F/G Isla
stations indicates that there are at least Channel water St. St. Flaca
two habitat types represented at the
Connector Marsh: 1) a mudflat/channel North west 56% 41% 37% 34%
habitat (east channel) that is dominated by South west 64% 47% 27% 39%
arrow gobies; and 2) deeper, subtidal South east 56% 49% 15% 71%
channel habitats (west channels of North
and South CM) that are dominated by
topsmelt. Comparing species lists for the
Connector Marsh and reference channels
did not reveal this difference. the new channel of the constructed marsh
does not closely resemble that of natural
The three Connector Marsh stations channels. The highest similarity was
were then compared to those for the 71%, comparing the accreting channel
natural marshes (Table 111.6). Data ftorn along the eastern side of South CM with
the deeper, western channel of both the Isla Flaca channel, which receives
North CM and South CM are most like freshwater inflows from Sweetwater
those from the channel at Sweetwater River. The most obvious difference
Marsh (56-64% similar). The between reference channel fish
Sweetwater Marsh channel has a high communities and CM channel
proportion of topsmelt. In contrast, the communities is that several species tend
shallow, soft-bottom habitat of the to share dominance in the reference
eastern channel of CM is most similar to channels, while CM channels are more
Isla Flaca. Both of these stations are heavily dominated by a single species. A
distinguished from the others by their more thorough evaluation of this pattern
large, adjacent intertidal mudflats, and is needed.
both are dominated by arrow gobies.
Benthic invertebrate commun-
Conclusions regarding fish ity. Benthic invertebrates were sampled
communities. The preliminary con- adjacent to each fish sampling station
clusion, based on 3 seasonal samples, is with 9 cores using a clam gun (0.0 18 m2
that the US FWS criteria for fish species surface area) to a depth of 20 cm. An
and abundance have been met, since all of additional station was established in
the native fish species have been shown Vener Pond where shorebirds are known
to occur in the constructed marsh to feed. Invertebrate samples were
channels and since densities of fishes are preserved and retained for laboratory
very similar to reference channels. identification.
Two functional differences need to be Sampling of the benthic invertebrate
explored further. First, non-native community of the constructed marsh is
species have colonized the constructed not complete. Laboratory examinations
marshes, and their influence on native and counts have been done only for the
populations is unmeasured. Second, the June and October 1989 sampling periods.
fish community (relative abundances) of

Assessment Case Study

The following summary of findings is individuals as in the natural benthic
thus preliminary. sampling sites. These calculations are
clearly preliminary and are not
In all, 27 taxa were captured and conclusions of the study.
identified throughout the wetland
complex; of these, 14 were found in
numbers of fewer than 5 in the combined
June and October samples (Table 111.7). The bird community. The object-
Despite the use of 9 cores per sampling ives of marsh restoration include
station (a total area of 0. 16 m2), many provision of foraging habitat for
stations had very low densities. The endangered birds. Understanding how
results obtained to date show high spatial the entire bird community uses the
and temporal heterogeneity in the sand restored marsh is important to the overall
and mud cores, with several station functional assessment. Not only can we
samples having few animals, and one learn what functions the restored marsh
station (Vener Pond in June) containing provides, we can also explore
173 individuals of a small snail. relationships among species and species
Combining both sample dates, totals groups to understand why certain uses
encountered ranged from 20 to 276 occur or fail to develop.
animals.
Extensive bird surveys were
The total number of individuals undertaken (Ashfield and Kus, unpub.
sampled was 683; standardizing the data). Detailed analyses and results will
numbers to area sampled indicates that be provided in an M.S. thesis (Ashfield,
samples from the constructed channel had in prep.). The following is an initial
54% as many individuals as the natural evaluation of the data from the Connector
benthic sampling stations. Marsh, the Paradise Creek, and the Bay
Shoreline. Sampling took place twice per
Of the 27 taxa, 11 were found in both month for 13 months from March 1999
constructed and natural channel stations; through March 1990 (except once per
there were 20 taxa in the natural channels month in November 1989 and January
and 18 in the CM sampling stations. 1990), for a total of 24 low-tide surveys.
Assuming no effect of greater sampling We report on the information for low-tide
area (an assumption that may not be met surveys only (high-tide uses were also
since there were 4 natural stations and 3 recorded). We include only the
constructed stations), the similarity of waterbirds (excluding gulls).
species lists is 58% (using the 2w/[a+b]
similarity index). Comparing the 11 taxa Two sampling sites were selected for
of the constructed channel to the 20 found comparison with the Connector Marsh
in natural channels suggests that only (CM): Paradise Creek (PC) and the Bay
55% of the naturally occurring taxa have Shoreline (BS), which is just west of
established in the Connector Marsh Gunpowder Point. The rationale for
channels by the October sampling date. comparing CM to PC is that PC habitat
These are preliminary calculations, and was damaged and replacement of
the similarities will most likely change functional losses there was part of the
once the full-year data set is analyzed. objective of constructing CM. The
reason for comparing CM to BS is that
Preliminary summary of ben- CM has a larger proportion of intertidal
thic invertebrates. Of the two flat than PC.
seasonal samples for which animals have
been identified and counted, the
constructed channels have 55% of the
species and 54% of the densities of

Assessment Case Study

Table 111.7: Channel invertebrates found at the Sweetwater River Wetlands Complex,
June plus October 1989. Data are numbers per sampling station (number/O. 162 m2).

F/G E SW Vener Isla S.isl. North S.isl.
Taxa D= Pond Flaca west IsIs. east TOW
Bivalves
Macoma nasuta
Tagelus californianus 1 0 2 0 0 1 2 4 10
Protothaca stanunea 4 16 37 0 2 18 10 25 112
Musculista senhousei 1 1 3 0 5 3 2 2 17
Gastropods 2 0 0 0 0 0 0 0 4
Assiminea californica
Cerithidea californica 0 0 0 0 0 6 0 0 6
Acteocina inculata 9 1 13 16 1 30 3 13 86
Bulla gouldiana 16 0 0 173 11 0 0 0 200
Nassarius sp. 0 0 0 0 0 0 0 1 1
Crustacea 0 0 0 0 1 0 0 0 1
Hemigrapsus oregonsis
Amphipoda 1 1 4 0 0 0 0 1 7
Orchestia traskiana
Anthozoa 0 0 0 0 0 9 0 0 9
Ceriantharia
Polychaetes 1 0 0 0 0 0 0 0 1
Capitellidae
Unknown capitelli&
Cirratulidae 0 0 0 0 0 2 2 1 5
Cirratidus cirratus
Tharyx sp. 0 0 1 0 0 0 0 0 1
Eunicidae 2 0 0 0 0 1 0 0 3
Eunice valens
Glyceridae 0 1 0 0 0 0 0 0 1
Nereidae 0 0 1 0 0 0 0 0 1
Nereis sp.
Orbiniidae 0 0 0 0 4 6 1 5 16
Scoloplos armiger
Phyllodocidae 0 0 15 0 10 0 0 0 25
Eteone sp. 0 0 0 0 2 1 0 0 3
Spionidae 0 0 0 0 0 0 1 0 1
Polydora nuchalis
Polydora cornuta 11 0 0 26 0 3 0 0 40
Polydora sp. 2 0 0 2 0 12 0 6 22
Streblospio benedicti 26 0 0 59 0 12 4 10 ill
Fly larvae 0 0 0 0 1 0 0 1 2
Chordata 0 0 0 0 0 0 0 2 2
Ascidacea
Diplostoma sp.
I Q Q Q Q 0 Q D
Totals 77 20 76 276 37 104 25 71 688

Assessment Case Study

western sandpipers, and in the area of
The three census sites are not identical intertidal flats, Because western
in size or physiography- The Paradise sandpipers are so abundant at BS, their
Creek site was larger but, because it is numbers mask the importance of other
dominated by salt marsh habitat, it had species. The largest total for dowitcher
the smallest area of intertidal flats. Field sightings was at BS, as were totals for
estimates of the areas sampled at CM and willets and marbled godwits.
BS were very similar at 6-7 ha. Of these,
the BS site was entirely intertidal flat CM shares 18 species with PC and 18
habitat, while CM was a combination of with BS (Table 111.8). The data on
flats, channel water, and marsh. relative numbers of sightings show
moderate similarity for the top ten
Comparisons of CM with both species, which comprised 80-90% of all
PC and BS. Overall, the number of sightings. The similarity between CM
water- associated bird sightings was and PC (77% for species lists and 62%
lowest in PC (less than 1/3 that at CM), for sightings, Table 111.8) was greater
despite the fact that this census area was than the similarity between CM and BS
larger than CM and BS (Table 1111.8). In (57% for species lists and 53% for
the low-tide censuses, excluding gulls, sightings). Thus, despite the fact that
PC had 23 species of waterbirds, CM had CM and BS are more alike in having large
24, and BS had 39. Using these data, areas of intertidal flats, CM and PC
PC and CM had about the same species appear more similar in their bird
richness, but attracted fewer species than communities. As an intertidal flat, CM
BS. Note that all sites would have longer functions less well than BS, both from
species lists if gulls, high-tide surveys, the standpoint of numbers of species and
-and land birds were included. numbers of sightings. As a channel-
marsh system, CM and PC are
Neither the lower numbers of moderately similar in bird communities,
sightings nor fewer species seen at PC with substantially more numbers of
can be attributed to sampling area, since water-associated birds seen at CM than at
PC was the largest site; hence, the finding PC.
of fewer sightings at PC than at both CM
and BS is real. The extensive cover of
salt marsh vegetation at Paradise Creek
and its single large channel (rather than a
highly dissected tidal creek system) is not
optimal habitat for large numbers of
water- associated birds. PC does,
however, support dowitchers, willets,
and western sandpipers in fair
abundance.

The largest number of sightings was
at BS (nearly 3x as many as at CM and
over 9x as many as at PC). With its high
species richness and abundance of
waterbirds, BS is clearly the most-used
waterbird habitat of the three comparison
sites. CM is intermediate in waterbird
support functions, between PC and BS.

The three sites differed in the number Fiddler crab
of species encountered, in sightings of (Uca crenulata)

Assessment Case Study

Table 111.8. Sightings of water-associated birds at three wetlands of the
SRWC. Data are for 13 months in 1989-90. Data are for low-tide and water-associated
species excepting gulls. The similarity of species lists was calculated as 2w/a+b, where
w=no. of species in common, a=no. of species at PC and b=no. of species at CM.
Relative data are provided only for species with at least 5% of the sightings in at least one
of the three sites. Similarity of relative sightings was calculated as Y,(mininum a,b), where
a=rel. sightings at PC and b = rel. sightings at CM, summed for all species, not just the
ones listed here).

Summ4a PC CM BS

Total sightings; relative to x 0.31x x 2.88x
Number of species sighted 23 24 39

Relative number of sightings (% of site tot 1)
au
Western sandpiper 15.5 38.9 40.6
Dowitcher spp. 25.0 29.1 11.7
Willet 20.0 3.5 7.2
Marbled godwit 4.9 2.1 8.5
Least sandpiper 2.1 9.7 0.7
Dunlin 5.7 4.8 1.7
Killdeer 4.3 6.5 0.1
Red knot 0 0 8.7
American coot 0 7.0 0
Bufflehead 6.6 0.1 -0

Sum for above 10 species (%) 84.1 91.7 79.2

Site cg=ari@ons (all species) CM with PC CM with BS

Number of species in common 18 18
Similarity of species lists 76.6% 57.1%
Similarity of rel. numbers of sightings. 62.5% 52.7%

Conclusions from the bird some of the qualities of the best mudflat
census data. The avifauna of SRWC is site of SRWC. Overall, there were more
abundant and species rich. The sightings at CM than PC, but far fewer
Connector Marsh provides habitat that is than at BS.
somewhat like that of the marsh-
dominated Paradise Creek site and Overall evaluation of functional
somewhat like that of the Bay Shoreline. equivalency. It is not easy to create
The relative numbers of bird sightings at cordgrass ecosystems that are function-
CM are more similar to those of PC than ally equivalent to natural ones. In a
the Bay Shoreline, indicating that the CM recent attempt to simplify the complex
site can support many of the marsh birds data comparing the two types of systems
(e.g., dowitchers, willets) but that it lacks at SRWC, we summarized 11 data sets

Assessment Case Study

that were gathered when the constructed dated, and the habitat islands disappear
marsh was 4-5 years old (Zedler and from sight.
Langis in press). The resulting "index"
(Table M.9) indicates that the constructed The shortcomings of the constructed
marsh was less than 60% functionally marsh were not anticipated; on the
equivalent to the natural reference wedand contrary, the site benefitted from exten-
(Paradise Creek Marsh). sive planning and biological advice from
several agencies. Detailed assessments,
such as the work reported here, had never
Table 1H.9. Functional equivalency of been done in our cordgrass marshes, and
the constructed and natural cordgrass no one knew what site characteristics
marshes comparing soils, nutrients, needed to be measured to predict what
plants, and epibenthos. Soil and plant would limit ecosystem development.
data are from Langis et al. (in press)
except for plant heights (unpub. PERL The quantitative comparisons of con-
data). Epibenthic invertebrate data are structed and natural cordgrass marshes
from Rutherford (1989). @Langis et al. in press) revealed signif-
icant differences in substrate character-
Data set istics that help to explain why trans-
planted cordgrass is growing poorly.
Organic matter content 51 The sandier soils (probably from alluvial
Sediment nitrogen (inorganic N) 45 outwash) have little organic matter and
Sediment nitrogen (TKN) 52 little nitrogen. The low organic matter
Pore-water nitrogen (inorganic N) 17 content limits nitrogen fixation and
Nitrogen fixation (surface cm) 51 nutrient recycling by microbes. Thus,
N fixation (rhizosphere) 110 cordgrass growth is limited. Because
Biomass of vascular plants 42 levels of these two causal factors did not
Fohar nitrogen concentration 84 increase during the study, we reserve
Height of vascular plants 65 judgment on how long it will take for
Epibenthic invertebrate numbers 36 functional equivalency to reach acceptable
Epibenthic invertebrate species lists 78 levels. (Funds have been requested to
continue comparisons through age 9.)
Average of comparisons -57%
95% confidence limits 40-74% Our understanding of the importance
of soil organic matter and nitrogen
suggests corrective measures for future
restoration sites, and experiments are
Inclusion of data on fishes, channel now underway to test the ability of a
benthos, and birds failed to increase the variety of soil amendments to accelerate
overall similarity. Although fish species the development of constructed cordgrass
lists were 100% similar for the two sites' marshes. As the science of habitat
the bird for which the marsh was restoration advances, it should be
designed (light-footed clapper rail) was possible to achieve greater than 60%
not yet using the constructed marsh. functional equivalency. At the same time,
as scientists continue to understand the
A major difference between the details of how natural wetland
constructed and natural sites is easily seen ecosystems function, expectations for
during high tides. Tall cordgrass in the restoration sites will also rise and the
natural marsh extends above the water, attributes that are assessed will need to
providing a refuge for terrestrial arthro- expand accordingly.
pods and cover for birds hiding among
the plants. In contrast, cordgrass in the
constructed marsh is completely inun-

Sampling methods and comparative data from natural wetlands

with personnel from the Pacific Estuarine
IV. Sampling methods Research Laboratory who monitor San
Diego County wetlands in order to access
and comparative data the most recent reference data sets.
from natural wetlands Second, use a hierarchical approach,
insuring broad coverage with the most
Very few restoration or mitigation general descriptors of ecosystem
projects have included adequate monitor- condition (e.g., areas covered by tides)
ing programs, either in the extent of and selected, individual sampling sites for
information gathered or the length of time additional characteristics (e.g., plankton).
sampling has continued. In a review of
Section 404 permits (Clean Water Act) in Third, obtain aerial photos annually
Washington State, only 31% of the and identify changes from year to year.
permits were found to require monitoring; The US Army Corps of Engineers
those that did were typically of short photographs the coastline each year and
duration (usually I year), and only has air photos available for purchase.
required data on vegetation coverage (cf. Determine the tidal condition on the date
Josselyn et al. 1989). Likewise, the and time that the photo was taken from
California Coastal Commission's tide tables. Photos at maximum high tide
Wetlands Task Force (Ray et al. 1986) will indicate tidal coverage but obscure
found that the effectiveness of mitigation vegetational changes, while photos at low
efforts has not been documented--56% of tide will show the development of tidal
permits reviewed required some kind of creeks in areas formerly dominated by
monitoring, but efforts ranged from plants. Both will be useful for
simple visual observations to water interpreting changes in the overall
quality changes, with results that were ecosystem. Use photos to identify
not comparable. In many cases, results locations where tidal creeks are
were not available and it was suspected developing, to track the expansion of
that monitoring was not performed. vegetation onto newly graded tidal flats,
to document changes in the linkages
There are three basic recommenda- between the estuary and the surrounding
tions for an assessment program: landscape, and to quantify the portion of
the adjacent landscape that is developed.
First, use standard methods and Aerial photo analyses should extend to
compare data with existing monitoring areas beyond the wetland to include
programs. The methods proposed for potential enhancement sites.
monitoring are taken from published
literature and/or have been used to Recommendations for measuring
monitor other southern California coastal ecosystem conditions and functions may
wetlands (Tijuana Estuary, San Diego exclude some measures that are common
River Marsh, Los Penasquitos Lagoon, in other regions because their measure-
Mugu Lagoon). The most important ment is too difficult or the results are not
considerations are the t),pe of gear used to meaningful in this region. The prime
collect organisms (e.g., mesh size of fish example is net annual primary pro-
seines), time of sampling (e.g., soil ductivity of vascular plants (NAPP) and
salinity varies seasonally), size of sample algae. Both are highly variable, the
unit (e.g., data on frequency of former on an annual basis, the latter on a
occurrence of plant species vary with weekly basis. In addition, NAPP is
quadrat size), and location of sampling grossly underestimated but not to the
station (small changes in topography same degree for each species in the region
affect species composition). Persons (Onuf et al. 1978). Thus, it is nearly
who supervise and carry out sampling impossible to obtain data that can be
should interact (e.g., exchange reports) compared from site to site.

Sampling methods and comparative data from natural wetlands

Table IV.1. Ecosystem attributes to be considered in assessing how well constructed
wetlands replace the functions of natural wetlands:

Attribute and Measures Reason for Analyses

Hydrology*
Current velocity and distance from tidal inlet Tidal circulation
Water levels at various tidal cycles Tidal lags, inundation regime
Salinity of water and soil* Relation with strearnflow

Topography
Elevation% slope Emsion, accretion

Soils
Texture Drainage, resilience of soil
Organic matte? Nutrients, resilience of soil
Toxic substances Biological accumulation
Redox Indicates drainage, organic
matter, anaerobiosis
Sulfides and pH Potential for acid sulfate
soil formation

Nutrient dynamics
Nitrogen fixation rates Availability to producers
Inorganic nitrogen in sediments and pore water Potentially limiting nutrient
Denitrification rates Nitrogen recycling
Organic matter decomposition Nutrient mineralization
Nitrogen mineralization rates

Algae
Cover by dominant type* Food for invertebrates
Potential for nuisance blooms

Vascular plants
Total stem length (m/m2) of cordgrass Estimates standing crop
Cover of vascular plants* Shifts in dominance
Density of rare annual plants Population persistence and growth

Consumers
Decomposers and shredders Food chain support
Aquatic insects Indicators of water quality,
food chain support
Terrestrial insects, especially pollinators Food chain support; control
and predatory insects of herbivorous insects, pollination
Fishes* and invertebrates* Food chain support
Birds* Food chain support
Reptiles and amphibians Food chain support
Mammals Food chain support

*Highest priority components to assess

Sampling methods and comparative data from natural wetlands

much tidal circulation was improved by
1. Hydrologic excavation of intertidal habitat.
Placement of tidal staffs (simple,
functions graduated measuring sticks) facilitates
location of data loggers at comparable
elevations. Measurements are made in
both man-made and reference wetlands
The most important forcing function simultaneously to determine if the
of a coastal wetland is its hydrology, and constructed system behaves similar to a
readers are referred to hydrologists to natural wetland. Measurements at
understand the hydrodynamics of coastal increasing distances from the tidal inlet
wetlands. In almost every management help to reveal any obstructions or
planning effort, the bulk of the effort bottlenecks in the main channels. The
goes into the characterization of the identification of a sill or hardpan that
hydrology of the system and use of restricts tidal flows will help determine
models to predict changes in the depth whether there is potential for increased
and circulation of the system under tidal prism and whether such an
different management (e.g., dredging) obstruction needs to be removed.
regimes. This manual gives only the Finally, measurements before and after
briefest introduction to hydrologic the construction of new intertidal habitats
functions. should show the effects of the newly
created marsh on total-system hydrology.
Objectives. The objectives of a To characterize tidal flushing, the tidal
hydrologic survey are often to determine heights are plotted with time (24-hr cycle)
the system's tidal prism and the size of for each location. On the coast, the
tidal prism needed for self-maintenance of minimum tidal amplitude occurs in spring
the ocean inlet, to characterize tidal and fall, and the maximum in summer
flushing, and to understand the influence and winter; lowest tide levels occur in
of tidal circulation on the ecosystem. daytime in winter and at night in summer.
Three hydrologic features are of special Tidal conditions are modified within the
ecological interest- -inundation patterns, estuary. Reduced amplitudes and more
salinity regimes, and water column delayed peaks occur at increasing
stratification. Understanding the flooding distances from the ocean inlet. Increased
and drainage patterns of tidal channels is amplitudes and shorter lag times are
also extremely helpful in selecting anticipated following dredging or grading
sampling times for fishes and other operations that increase tidal prisms, and
channel organisms. Knowing how the such measurements would indicate
system deviates from predicted tide levels improved tidal flushing.
on standard tables helps one arrive at the
sampling station when the water levels Elevation. The hydrologic condi-
are appropriate for the type of data being tions of intertidal sites are determined by
collected. Maximum and minimum water measuring elevations relative to the
levels within a tidal system can differ by National Geodetic Vertical Datum, i.e.,
2 hours or more. the 1929 mean sea level, which is an
Inundation regime. Remote data average of data for the preceding 19
loggers are used to measure depth and years. Distributions of salt marsh plants
duration of tidal inundation (to estimate are often referenced to this standard
tidal lags and tidal amplitude damping) datum; however, it must be remembered
for selected tidal cycles prior to wetland that the inundation regimes of a specific
alteration and following construction of elevation may differ for various locations
wetland habitats. Measuring tidal within an estuary. Because tidal maxima
amplitudes and lag times will show how and minima are damped at the inland
extent of tidal creeks, and because peaks

Sampling methods and comparative data from natural wetlands

lag behind conditions at the ocean inlet, possibly warmer) water may float over
the inundation regime for an elevation of the more saline/cooler seawater. In late
0 ft or m NGVD at the mouth may differ summer, the pattern of stratification may
significantly from that at the most inland reverse, with warmer hypersaline water
edge of the estuary. In addition, overlying seawater.
microtopographic features of the intertidal
zone may impede inundation or drainage, To characterize water column
such that one site at I ft NGVD may be stratification, water salinities and
well drained, while another impounds temperatures are taken monthly at selected
high tide waters. Thus, elevation is a sampling stations. Temperature is first
general indicator of inundation regimes, taken at the surface and at the bottom; if
not a precise measure of habitat there are differences, additional sampling
conditions. Elevation should be at 10-cm vertical intervals should be done
measured as a general descriptor, but one to determine where the thermocline
should not assume that the same absolute exists. Since estuarine water tem-
elevation will have the same peratures vary with the tidal condition,
environmental conditions. the time of day, the storm condition, and
the season, these measurements are more
11igh precision is needed in elevation useful for determining whether the water
surveys. Salt marsh vegetation is column is stratified, than for char-
extremely sensitive to slight differences in acterizing an "average" water temper-
tidal inundation, and plants that thrive at ature. To obtain average water temper-
one elevation may yield to another species atures, we recommend using continuous
if the topography is 10 cm (4 inches) too sampling with a data logger to record
high or too low. For mapping of wetland conditions during both spring and neap.
habitats and for marking sites for tides, over 24-hour periods. Such
vegetation transplantation, 30-cm. (1-foot) detailed data would be needed for
contours are the coarsest intervals that are scientific studies and modeling of water
useful. circulation.

Elevations are measured relative to Additional indicators of poor tidal
benchmarks near the estuary. Pro- flushing include phytoplankton blooms
fessional surveyors may be needed to (pea-soup green water) and dense mats of
establish benchmarks near study sites, if macroalgae. Macroalgal cover can be
these are not present. An automatic level estimated at the water salinity stations,
(e.g., Wild Instruments) and calibrated distinguishing epibenthic from floating
stadia rod are used to measure and/or algae (see water quality methods).
mark additional elevations at the site.
Elevations of all systematic sampling Reference data. Williams and
stations are determined in this manner. Swanson (1987) compared tidal flushing
While reference data are nearly alwa i
data from several Tijuana Estuary stations
feet and inches in US publications, the with data from a nearby tide recording
metric equivalents should also be reported station (San Diego Bay at Broadway
for international comparisons. Pier). They found differences in tidal
amplitudes and in the times of minimum
Water Column Stratification. and maximum water levels. Data from
Impaired tidal flushing can also be three stations are replotted in Figure 1. 1.
detected through measurements of water The greatest differences between stations
temperature and salinity. In sluggish within Tijuana Estuary and the Bay are in
channels, water columns become the amplitude between higher high and
stratified, with surface and bottom water lower low water and the lag in outflow
differing in either temperature or salinity. following. the higher water. The large
In . winter, following rainfall and volume of water flowing into the estuary
strearnflow into the estuary, fresher (and

Sampling methods and comparative data from natural wetlands

at higher high tide was held for a longer
time than the inflows of the lower high Additional hydrographs are presented
tide. Flows out of Tijuana Estuary were in Williams and Swanson (1987) to show
constrained by the shallow ocean inlet. It other lags within Oneonta Slough in
was also clear that flows out of Oneonta 1986. About 1.5 km (5,000 ft) inland
Slough were constrained by the along Oneonta Slough, higher high water
narrowness of that channel's inlet, which lagged 2 hours behind that at its inlet and
was due to overwash fans (ibid.). These lower low water was a foot higher,
overwash constrictions to tidal flushing indicating that inflows and outflows were
were removed by dredging in 1987 and both constrained upstream along the tidal
an improvement in flow to Oneonta channel.
Slough was documented by resurveys of
channel cross-sections (Florsheim and
Williams 1990).

Figure 1.1. Hydrographs for two locations at Tijuana Estuary compared to uncon-
strained tidal flows at San Diego Bay. Data redrawn from Williams and Swanson (1987).

6-
Tijuana Estuary
(Oneonta Slough inlet)
> 4-
a Tijuana Estuary
z
4) (inside estuary inlet)
>
2-

CR -2-
LU San Diego Bay
(Broadway Pier)

0 1000 2000 3000 4000
Hours, 8:00am Aug. 19 - 1 0:00arn Aug. 20

Sampling methods and comparative data from natural wetlands

nontidal period, channel water salinity
2. Water quality reached 60 ppt 7 months after closure.
Estuary-wide die-offs of invertebrates
(e.g., crabs and hornsnails) were noted.
Objectives. Water quality mea- More severe changes in water quality
surements are needed to document develop when inlet closure is combined
problems, such as sewage spills, and to with a sewage spill or rainfall and runoff
predict biological impacts, such as fish event. Over the past several years, Los
kills. The most thorough water quality Pefiasquitos Lagoon has closed to tidal
investigations, i.e., those including tests flushing during the warm summer and
of heavy metal concentrations and fall months. On two occasions, major
presence of organic toxins, will help to freshwater inflows and salinity-dilution
predict the potential and risks for events occurred. The first was a major
invertebrates, fishes, and birds. In sewage spill; the second, an early rainfall.
southern California, the greatest risks to Without tidal flushing, the non-saline
marine animals develop when the mouth waters were impounded, and extreme
of an estuary or lagoon closes to tidal salinity dilution occurred, followed by
flushing. Initially, water temperature major fish and invertebrate kills.
rises, so that saturation concentrations for Additional data are provided by
dissolved oxygen decrease. Microbial Nordby and Zedler (in press), including
activity increases, along with metabolism evidence from Tijuana Estuary and Los
rates of all animals and plants; thus, Pefiasquitos Lagoon that lowered water
overall oxygen consumption increases. salinity reduces the species richness of
While algae may become more both fishes and benthic: invertebrates.
productive, high biomass will accumulate
in the surface water, shading plants in the
channel and creek bottoms, so that less
oxygen production will occur in the What to measure. Several factors
habitats occupied by benthos. Later, such as nutrient concentration, light
decay of algal cells and fronds by attenuation, dissolved oxygen, salinity,
decomposers will increase oxygen and temperature are recorded to assess
demand, and bottom waters will become water quality. The aquatic organisms,
hypoxic (low in oxygen). In the absence i.e., zooplankton and algae, also help to
of rainfall, evaporation will concentrate characterize water quality. There are
the saline water trapped in the estuary, limited data on planktonic algae, but no
and channels and creeks will become reference data for zooplankton. Sampling
hypersaline. Thus, with closure alone, before and after restoration measures take
substantial changes in water quality will place is important to determine the
develop. The changes in Los changes in ecosystem condition.
Pefiasquitos Lagoon, during the 1979
closure, illustrate these patterns (Figure Permanent sampling stations are
2.1-2.3). Comparative data are shown chosen in a stratified random manner, to
for sample dates before and after closure, assure similar representation for all
as well as inside and outside the inleL aquatic habitat types. Habitat types may
Much higher salinities were include deep channels, tidal creeks, deep
documented at Los Pefiasquitos Lagoon saline ponds, brackish and fresh sections
in the 1950's, when the lagoon remained of the incoming creeks, and freshwater
closed to tidal flushin for several ponds. Stations should be sampled
9 biweekly, or at least monthly, to detect
consecutive years. Channel waters and account for seasonal as well as yearly
exceeded 60 ppt in 1959 (Carpelan patterns.
1969). At Tijuana Estuary, during a long

Sampling methods and comparative data from natural wetlands

Figure 2.1. Changes in water temperature at Los Peffasquitos Lagoon, before and after
closure of the inlet on May 12, 1979. The "Stagnant Arm" parallels the barrier dune and has
reduced tidal influence year round; the sampling station under the railroad (RR) bridge is the
deepest site in the lagoon. Data are single measurements from the water surface (PERL,
unpubl.).

30-
Stagnant Arm
0
0

20-

C13
Outside the Inlet
CL
E io-
Under RR Bridge
Lagoon Inlet Closed

0
0 100 200 3@O
Julian Date

Figure 2.2. Reference data for lagoon water salinity at the surface, comparing sites before
and after inlet closure, as in Fig. 2. 1.

40-

Outside the Inlet
30- Stagnant Arm
CL
CL

7= 20- Un er RR Bridge

cc
Cl)

10-
Lagoon Inlet Closed

0-
0 100 200 300
Julian Date
d

Sampling methods and comparative data from natural wetlands

accumulate to bloom proportion,
Water temperature and dis. anaerobic conditions can develop at the
solved oxygen are measured using a channel bottom during the night. In tidal
dissolved oxygen -temperature meter channels, the . highest algal biomass
(e.g., Yellowsprings Instrument Model would be measurable at low tide at the
YSI 5 1 B). end of a neap tide series, when channels
would not have been greatly diluted by
Water salinity is measured to the seawater.
nearest part per thousand using an
American Optical (or Reichert) salinity In each of the aquatic habitat types,
refractometer. Alternatively, a YSI water samples for phytoplankton analysis
Salinity Meter and probe can be used for can be collected in a simple plexiglass
ease in sampling vertical salinity profiles. tube of 2-cm internal diameter, which is
Both the refractometer and salinity meter the length of the water column being
read salinity in parts per thousand. sampled (e.g., I m). This tube is
lowered vertically into the water column
Light attenuation is measured to ensure sampling from all strata, then
using a submersible light meter (e.g., Li- sealed at the top and drained into the
Cor), sampling at the surface and at the sample bottle. This sampling can be
bottom, with the depth between concurrent with the quarterly zooplankton
measurements recorded. Several sampling, utilizing the same stations.
different units can be used; the
measurement of interest is % attenuation. Phytoplankton biomass is estimated
The extinction coefficient is k = [log I0 - as chlorophyll a concentration using
log Iz]/z, where 10 is the amount of -light either the fluorometric technique or
at the water surface and Iz is the amount extraction and measurement of
of light at depth z. A simpler measure, absorbancy using a spectrophotometer.
for deeper water bodies is to lower a The data in Figure 2.4 were obtained with
Secchi disk (a 20-cm diameter disk) from the latter method, extracting the pigment
the shady side of a boat or pier. and in 90% acetone, measuring absorbancy at
determine the depth at which it is no 663, 645, and. 630 nm (subtracting
longer visible. absorbancy at 750 nm from each
reading), and applying the formula, Chl a
Nutrients are collected in water 11.64Abs663 - 2.16Abs645 +
samples using the same techniques as for O.lAbs630 (Strickland and Parsons
phytoplankton. Samples can be frozen 1972). Results (mg pigment per liter of
and analysed for nutrient concentration at acetone), divided by liters of seawater
a later time, but more reliable filtered, will yield micrograms of pigment
measurements are obtained on fresh per liter of seawater milligrams per
samples that have been kept on ice cubic meter).
between the site and the lab. Wet
chemistry techniques for nutrient analysis At present, there are no exhaustive
are time consuming and routine analyses data sets for coastal wetland channels.
are best conducted on an autoanalyzer However, the San Diego Regional Water
(e.g., Technicon Instruments Inc.), Quality Control Board has comparative
which automates the process. data sets for several lagoon sampling
stations over a limited time period.
Algae. In aquatic ecosystems, algae Experimental studies (Fong 1986, Fong
are the base of the food chain. While et al. 1987) have addressed the role of
measurements of algal populations are not nutrient additions to coastal water bodies.
very good estimators of primary
productivity, they are useful indicators of Visual estimates of the percent of the
eutrophication and tidal flushing (cf. water surface covered by macroalgae
Figure 2.3). When phytoplankton should be made at the same stations, and

Sampling methods and comparative data from natural wetlands

the genus noted (usually members of the these communities should be sampled at
Chlorophyta, either Enteromorpha or high tide, all of the habitat types can be
Ulva). Permanent plots are marked for included.
cover estimation, using a rectangular
shape to reduce variance, and a size that Plankton nets with a mesh size of 35
is appropriate for the habitat type (smaller microns are appropriate for collection of
for tidal creeks than main channels). zooplankton samples. As most of the
habitats are relatively shallow, oblique
Zooplankton. Sampling for zoo- tows from channel or lagoon bottom to
plankton can be done quarterly at the the surface are made behind a boat that is
same stations as the fish and travelling parallel to the shoreline.
invertebrates, although prior to or on a Samples should be fixed in the field in
different day, as the seining often formalin, and quantified microscopically
resuspends benthic particles, obscuring using Sedgwick-Rafter counting
the samples. Experience has shown that chambers. Zooplankton densities and
sampling for plankton is most efficiently community composition can be assessed
accomplished by boat as the water is not spatially and temporally. Sampling of
disturbed as much as when walking. plankton should be done seasonally under
This is true for the zooplankton, algae, the same tidal condition (e.g., end of a
and physical/chemical sampling, and all neap tide series), in order to reduce the
can be accomplished at the same time. As effect of seawater dilution.

Figure 2.3. Comparisons of phytoplankton (as relative amounts of chlorophyll a ) at Los
Peflasquitos Lagoon. One sample was taken prior to inlet closure. Phytoplankton
concentrations buiJt up in the poorly circulated "Stagnant Ann" following closure. Note the
log scale. Due to questionable calibration of the spectrophotometer, absolute values for
concentrations are not valid; however, relative comparisons are permissible.

101
Stagnant Arm

IM 100
E
cts Under RR Bridge
>% 10'1 Outside the Inlet
CL
0
1- 1072-
0

3' inlet closed on May 12,1979
10' - I I
100 Julian Date 300

Sampling methods and comparative data from natural wetlands

quently, on the marsh productivity. To
3. Soils: date, there are only a few sites for which
we have obtained detailed data on soil
Substrate qualities conditions. The theses of Swift (1988),
0 Cantilli (1989), and Zalejko (1989), and
and nutrient dynamics the studies of Langis et al. (in press)
provide a basis for understanding below-
ground dynamics of the coastal salt
Objectives. Soil conditions have a marshes.
major influence on vegetation growth and
on organisms that inhabit the rhizosphere Sampling strategy. All soil
of plants (e.g., amphipods, nematodes, sampling is done randomly along pre-
microbes). Four variables are especially determined 10-12 m transects. The tran-
helpful in predicting the ability of a site to sects should be along selected elevations,
support a functional salt marsh: soil so that replicate samples are under a simi-
salinity, nitrogen dynamics, organic lar tidal regime. Samples subject to sea-
matter concentration, and redox potential. sonal variation (soil or pore water
Soil salinities control seed germination nitrogen concentration, nitrogen-fixation,
and seedling establishment in the coastal denitrification, soil or pore water salinity,
wetlands (Zedler and Beare 1986). redox potential, water content) should be
Concentrations that are either too high or taken quarterly.
too low will alter vegetation composition
by restricting growth of even the most High variances among replicate
tolerant halophytes, in the case of extreme samples are more often the rule than the
hypersalinity, or by allowing the invasion exception for chemical parameters of
of cattails, bulrushes, and other glyco- soils. This variability can be reduced by
phytes, in the case of prolonged periods, compositing several samples before
of hyposalinity. Nutrient dynamics, analysis (Binkley and Vitousek 1989). In
organic matter, and redox conditions all general, one should increase sampling as
interact to control- plant growth rates, much as is practiable to improve the
which in turn affect the consumers that estimate of the mean. Soil analysis time
live among the plant roots. Soils with is the usual limiting factor. Lloyd and
low organic matter will have low McKee (1983) have suggested a statistical
nitrogen-fixation rates and low supplies procedure for determining the optimal
of the main nutrient that limits plant number of subsamples required by the
growth. Soils with high organic matter level of confidence desired. The number
will develop very negative redox of subsamples will depend on the degree
potential, which may restrict the growth of variability of the measurements for the
of some marsh plants; Cantilli (1989) sediments of the particular area. For
showed that low redox affected the example, Langis et al. (in press) obtained
growth of pickleweed, but probably not acceptable levels of variation with 4-6 soil
that of cordgrass. cores randomly selected along pre-
established transects of 10- 12 m.
The patterns of salinity, nitrogen
dynamics, organic matter accumulation, Bulk density. This parameter rep-
and redox potential vary in space and resents the weight of soil per unit volume
Aime. Wetland hydrology determines the including pore spaces. It is useful when
chemical and physical nature of salt the density differences of different soils
marsh substrate to a great extent (Mitsch must be accounted for. In which case,
and Gosselink 1986). For instance, the properties such as percent organic matter,
aerobic- anaerobic conditions resulting total nitrogen or moisture can be
from regular tidal flooding of these soils overestimated in organic rich sediments
have a profound effect on the biogeo- when expressed on a dry weight basis
chemical cycles of nutrients, and conse-

Sampling methods and comparative data from natural wetlands

because of the relative low weight of all cases, measurements are made from
organic matter (Allison 1973). replicate sampling stations. Soil salinity
is measured monthly, near the water
Bulk density is obtained by pressing salinity sampling locations. Soil salinities
into the soil a thin-walled soil can with a are also measured at the permanent vege-
cutting edge. The soil is then sliced tation transects at the time of plant
smooth at the open end and oven-dried to censusing, and in sites of special interest,
constant weight at 105'C. Bulk density = such as planting locations for the salt
mass of soil in tube/volume, expressed as marsh bird's beak and areas where weedy
g dry-wt/cm3 (Richards 1954). vegetation is invading.

Water content is measured as Soil cores are easily removed with a
weight loss upon oven drying divided by 2.5-cm-diameter soil tube (available from
dry weight of the soil sample. The soil Forestry Suppliers), which is inserted
sample is dried to constant weight at 100- about 20 cm into the soil. A subsample
110'C (Gardner 1986). Organic soils of the core is then analyzed for soil
have higher water-holding capacities than salinity. A 5-cm segment from the 0-10-
mineral soils (except for clays). cm depth characterizes the upper root
zone; and 20-30 cm the deeper root zone.
Particle size of marsh soils If the marsh soil is saturated with water,
should be assessed in the traditional the salinity measurement can be made
method used by terrestrial ecologists, immediately in the field, by expressing a
identifying sand, silt, and clay percent- drop of soil water onto a salinity refrac-
ages using a hydrometer (Gee and Bauder tometer (use one that is temperature-
1986), rather than the detailed size differ- compensated, with a range of 0-150 ppt;
entiation used by marine ecologists available from most scientific supply
(Emery settling tube and phi values). houses). Soil water is expressed by load-
Soils are characterized by type, with ing a 10 cc plastic syringe (without ne ,ed-
natural salt marshes likely falling in the le) with 2 layers of #2 Whatman filter
clay to clay-loam types. At Tijuana paper that has been cut into 12-mm-
Estuary, lower marsh soils were gener- diameter circles (punch from larger sheets
ally clayey, with 2-8% sand, 23-32% silt, using a half-inch die). The wet soil is
and 46-58% clay (Zedler et al. 1980). loaded by hand; the plunger is inserted,
and a drop of water is forced onto the
Swift (1988) compared soil texture in refractometer. Readings are in ppt.
man-made and natural marshes at Sweet-
water River Wetland Complex and found If the soil is too dry, the subsample is
coarser texture in soils of the marsh that stored in a plastic bag (WhirlpakTm bags
had been graded to expose lower- are handy) and returned to the lab for
elevation strata. Paradise Creek (natural artificial saturation with deionized water.
marsh) soils were clay lo"ams, while those The standard method for preparing
of Caltrans Connector Marsh (man-made) saturated soil pastes should be followed
ranged from loam to sandy loam. These (refer to Richards 1954) and salinity
differences in texture, along with dif- measured with a conductivity meter (e.g.,
ferences in soil organic matter (lower in Lab-Line mho meter). This method is
the man-made marsh) appeared to be time-consuming, but soil samples can be
important to soil functioning, as there refrigerated and processed in batches.
were also lower rates of nitrogen fixation The conductivity results are not directly
and less evidence for sulfate reduction. comparable with those from saturated
soils, but salinities from all pastes are
Soil salinities are assessed to help comparable with one another.
explain vegetation patterns and to track
the influence of freshwater inflows. In For comparison of the salinity of
saturated soil pastes with other refrac-

Sampling methods and comparative data from natural wetlands

tometer measurements taken in the field, Pore water chemistry can be evaluated in
water from the soil paste can be a convenient way by the use of wells that
expressed through filter paper and can be permanently installed in the sedi-
measured with the refractometer. Results ments. Water should be sampled as soon
can be viewed as the soil salinity that as possible after a high tide so that sedi-
would occur following rainfall. ments are sufficiently saturated. To
obtain a fresh sample, each well must be
Reference data on soil salinity emptied with a syringe fitted with a vinyl
are available for several wetlands in San tube (long enough to get to the bottom of
Diego County. A major factor that influ- the well) and allowed to refill before
ences soil salinity is closure of the ocean collecting the pore water sample.
inlet, as illustrated by comparing data
from Los Pehasquitos Lagoon (Figure Parameters such as pH, redox poten-
3. 1), which is often a closed lagoon, with tial and salinity can be measured directly
data from Sweetwater River Estuary in the well and samples can be taken in
(Figure 3.2) and Upper Newport'Bay the laboratory for chemical analyses.
(Figure 3.3), which are always open to Ammonium concentrations can then be
tidal flushing. Los Pefiasquitos Lagoon measured with the hypochlorite-
soils become very hypersaline following nitroprusside method and N03-+NO2-
evaporation and are measurably diluted by the cadmium reduction method
following rainfall. The effects are (APHA 1986). The depth of sampling is
stronger for surface soils (0-5 cm) than adjusted through the location of the slits
d'eeper in the profile (45-50 cm). In and by inserting the wells at different
contrast, soils of tidally flushed wetlands depths. It is possible to collect water
are mildly hypersaline for most of the from a narrow or broad range of depths
year. With a nontidal system, surface using such wells.- Water samples for
soil salinities are more variable than with chemical comparisons must be collected
the ameliorating influence of tidal waters. at similar tidal levels since Agosta (1985)
Long-term data on soil salinity have found that NH4+ concentrations tend to
been recorded at Tijuana Estuary, as part decrease as the water table goes down
of the salt marsh monitoring program of and to increase as the water table comes
PERL (Figure 3.4). Soil salinity has back up. This would especially apply to
been measured in April and in September creek banks where the level of the water
for 10 years, including changes following table is strongly affected by the tidal
a major flood year (1980) and estuarine action.
closure during a drought year (1984). The wells are constructed from plastic
Measurements to be done on (PVC) pipes (30-cm long, 2-cm inside
pore water. Since dissolved nutrients diameter, 2.5-cm outside diameter), in
are readily available to plants, it is which a series of vertical slits is cut at 2-
advantageous to sample pore water. cm intervals, from the bottom to approx-
These samples are representative of imately 17 cm. The slits are covered with
chemical conditions in the sediments. In a piece of nylon screen (NitexTm, 100 to
anaerobic sediments, dissolved nitrogen 200 @=) held in place with a sleeve made
is mostly under the NH4+ and dissolved from a thin PVC tube of dimensions
similar to the well. Since small particles
organic forms, N03- being a transient will get through the screen, particles
species. Ammonium and SRP (soluble should be allowed to settle, or it might be
reactive phosphorus) concentrations necessary to centrifuge or filter the
reflect exchange equilibria between dis- sample with WhatmanTm GF/C filter
solved and solid phases, since NH4+ and paper before analysis.
P04-2 are adsorbed on fine organic and
clay particles (Avnimelech et al. 1983).

Sampling methods and comparative data from natural wetlands

Figure 3.1. Substrate salinity at Los Pefiasquitos Lagoon, a system that is often closed
to tidal flushing. Evaporation of sea water leads to hypersaline soils, while impounded
rainfall and strearnflow reduces salinity. Data are means (and �1 s.e.) of samples at the
surface (0-5 cm) and at depth (45 -50 cm) for n=21 stations. Data are from Eilers (198 1).

100-

80-

CL
60-
45-50cm
'E 40-

C13
CO 20-
0-5cm

0-
@ov Jan Mar @ay J'ul
Month (1977-78)

Figure 3.2. Substrate salinity at Sweetwater River Estuary, a fully tidal salt marsh
adjacent to San Diego Bay. Data are means L+1 s.e.), for n=31 stations. Data are from
Eflers (1981).

100-

80-
0-5cm
CL
60-

40- 45-50cm
C13
20-

0
Nov Jan Mar May Jul
Month (1977-78)
@45-50cr

Sampling methods and comparative data from natural wetlands

Figure 3.3. Substrate salinity at Upper Newport Beach, a fully tidal salt marsh. Data
are means L+ 1 s.e.), for n=25 stations. Data are from Eilers (198 1).

100-

80-

CL 0-5cm
CL
60-

40-
cis 45-50cm
CO
20

0-
@ov Jan Mar @ay J1U1
Month (1977-78)

Figure 3.4. Interstitial soil salinity (0-10 cm) at Tijuana, Estuary. Data are from the
lower marsh, at 102 sampling stations within the 1979 distribution of cordgrass, for April
and September sampling periods. Error bars (�1 s.e.) are generally too narrow to show.

(104)

100-

80-

CL
CL
60

CO
CO 40

0
U)
20-

0
1978 1980 1982 1984 1986 1988
-@J-@@45-50cr

Year

Sampling methods and comparative data from natural wetlands

Soil pH can be measured directly in Broome (1987) noted no suvival of
pore water wells, using a pH meter and a vegetation planted in soils of pH<3.
combination electrode (cf. Figure 3.5). If
pore water wells are not used, the Redox potential. This measure-
electrode can be inserted gently into a ment is important because of it plays an
moist soil sample. If soil samples are too important role in the biogeochemical cycle
dry, they could be sealed in a plastic bag of nitrogen and sulfur, and affects the
and returned to the laboratory where this mobility of heavy metals. As for pH, it
measurement can be made on soil paste can be easily measured by inserting a
(see section on soil salinities). Soil pH redox probe directly in the pore water
could become a major concern in the case sampling well.
of acid sulfate soils. These soils can
become extremely acidic following the Values obtained with the redox elec-
oxidation of sulfides to sulphates and trode must be standardized to measure-
sulfuric acids. This situation occurs ments obtained with the standard hydro-
when tidal inundation is stopped and gen electrode by adding the appropriate
leaching of sulfates and sulfuric acid by correction factor. The correction factor is
rain is impeded by a high clay content the difference between values measured
(Linton 1969). Soil pH can vary by as with ZoBell's solution (0.003 M potas-
much of 2 units within a tidal cycle, sium ferricyanide, 0.003 M potassium
because of water infiltration or benthic ferrocyanide and 0.1 M potassium
biological activity (Wolaver et al. 1986). chloride) and its theoritical Fh of +430
Values lower than pH 4 are detrimental mV at 25C (ZoBell 1946).
to salt marsh plant establishment, since

Figure 3.5. Soil pH measured in bore holes at Los Pefiasquitos Lagoon (LPL, n=21
stations), Sweetwater River Estuary (SRE, n=31 stations), and Upper Newport Beach
(UNB, n=25 stations). Data are means L+1 s.e.). Data from Eilers (198 1).

7.4-

LPL
7.2-

7.0-

3: 6.8
CL
UNB
0 6.6-
CI)

6.4-

SFE
6.2-

6.0
Nov Jan Mar May Jul
Month (1977-78)

-Sampling methods and comparative data from natural wetlands

Extractable NH4+ and N03- 1000G) for 15 min. Absorption of the
+N02-. These inorganic forms of supernatant is measured at 600 nm on a
nitrogen give an estimate of potentially spectrophotometer. Absorption values
available nitrogen in the soil. To are then compared to a series of standards
measure, put 10 g (wet wt) of fresh soil (1.5 to 7% O.M.). Standards are pre-
sample of in a 125 ml Erlenmeyer flask; pared by adding calculated amounts of
add 100 ml 2 M KCI; set on a wrist- sucrose to ignited sediments (750'C, 4
action shaker for 1 h; filter through h). The main drawback of the method is
WhatmanTm no. I (the filtrate can be kept that some of the refractory organic
frozen until analysis); analyze for NH4+ substances will resist the digestion.
with the automated phenate meth-od and Organic carbon. The refractory
for N03-+NO2- by the cadmium- organic carbon fraction could better be
reduction method (APHA 1986). accounted for with a second technique,
Artificial seawater should be used to the modified Mebius procedure (see
prepare blanks and standards. In the Yeomans and Bremner 1988), where a
characteristically anaerobic sediments of more thorough oxidation is obtained by
salt marshes most of the extractable N digesting the sample at 1700C with
will be under the form of NH4+, N03- potassium dichromate and values of %
contributing very little to the N pool. organic carbon are obtained directly by
titration.
Total nitrogen. This is a mea-
surement of the total nitrogen in the soil A CHN analyzer could also be used
and includes the organic and inorganic to get organic carbon measurements
forms. Nitrogen is either measured as simultaneously with total nitrogen on
NH4+ after Kjeldahl digestion (APHA samples with unmeasurable levels of
1986) or directly with a CHN analyzer. inorganic carbon. If inorganic carbon is
present at a significant level (see Nelson
Percent organic matter. Both and Sommers 1982 for a discussion of
combustion and chemical oxidation meth- methods) it is advisable to use the modi-
ods have been used to measure soil fied Mebius procedure (Yeomans-and
organic matter (0M.) content. Weight Bremner 1988). The Sims and Haby
loss upon combustion in a muffle furnace procedure could also be used provided
overestimates O.M. when clay minerals that standards are expressed as % O.C.
and/or carbonates are destroyed at high instead of as % O.M.
temperatures. Swift (1988) measured
O.M. or wetland soils by combusting Nitrogen fixation could represent
samples at 7000C for 1 h and 4000C for 3 an important source of available nitrogen
h. Values obtained at 7000C were for the generally nitrogen-limited salt
consistently higher (Table 3.1). marsh vegetation. The most used and
most straightforward method involves the
Two wet digestion techniques are use of the acetylene reduction reaction,
recommended to avoid the problems of where the enzyme dinitrogenase is
carbonate or clay degradation. The rapid capable of reducing C2112 as well as
oxidation technique (Sims and Haby dinitrogen (Hardy et al. 1968; Casselman
197 1) has provided satisfactory results. et al. 1981). In this technique, C2112 is
Briefly, 10 ml of IN K2Cr2O7, followed added to incubation vessels and dinitro-
by 20 ml of concentrated H2SO4 are genase activity is monitored. Although
added to a 1-9 (dry wt) soil sample in a this method must be calibrated with an
125 ml Erlenmeyer flask. The slurry is 15N tracer to obtain absolute values,
swirled and allowed to react for 20 min, comparisons can be made on the basis of
brought to a volume of 100 ml with nmoles C2H2 reduced per unit soil.
distilled H20 and centrifuged (at ca.500-

Sampling methods and comparative data from natural wetlands

Table 3.1. Soil organic matter in the natural wetland remnant, Paradise Creek Marsh,
and the man-made Connector Marsh at San Diego Bay. Data are for soils seived with a 2-
mm mesh screen to remove roots (from Swift 1988).

4000C for 3 hr 7000C for 1 hr
ngth 0-5 cm 5-10 cm 0-5 cm 5-10 cm

Paradise Creek mean 4.90 4.86 8.14 9.24
n=3 s.e. (0.20) (0.48) (0.24) (0.48)

Connector Marsh mean 2.21 1.61 4.34 4.17
n=6 s.e. (0.16) (0.13) (0.34) (0.27)

For measuring nitrogen fixation, we
recommend the following procedure Nitrogen mineralization. Esti-
adapted from Zalejko (1989): In the mates of nitrogen mineralization rates in
field, soil cores are taken for root zone N- the field can be obtained with the buried
fixation (10-cm deep by 8-cm diameter), polyethylene bag technique (Eno 1960).
and for surface N-fixation (1 cm deep by Because plant uptake and leaching of
3 cm in diameter for surface N-fixation). nitrogen is prevented, net mineralization
Cores are placed in a I-L mason jar and rates can be estimated as the increase in
tightly capped with a lid (equipped with a NH4+ and N03 -+NO 2- (Pastor et al.
serum stopper). It is important to clean 1984). Sediment cores (8-cm depth x 5-
dirt off rims since jars must be air-tight. cm diameter) are placed into 0.03 mm
thick polyethylene bags; the bags are tied
The cores are incubated overnight in and replaced in their respective holes and
the dark at 220C, to allow microbial popu- covered with approximately 2 cm of
lations to adapt. The next morning, sediment. At the same time, another soil
acetylene and ethane (as an internal stan- core, adjacent to the plastic-enclosed
dard to account for leakage) are added to core, is collected and immediately taken
jars with a syringe, yielding an atmo- to the laboratory for analysis of
sphere of 10% and 0.01%, respectively. extractable NH4+ and N03-+NO2-.
Concentrations Of C2H4 and C2116 are These polyethylene bags are impermeable
measured with a flame ionization detector to water and permeable to gases such as
on a gas chromatograph. The headspace C02 so that constant moisture content can
gases are mixed by pumping the syringe be assured during the incubation while
plunger several times. Gas samples are permitting gas exchange (Gordon et al.
withdrawn from the jar head space on 1987). After 14 days, the bags are col-
several occasions (after I to 5 h of incu- lected and taken back to the laboratory for
bation). Samples (1 ml) are injected analysis. Increases in concentrations of
(injector temperature: 85'Q onto a 2 m extractable NH4+ and N03-+NO2- (an
PorapakTm N column. Pure nitrogen is estimate of net nitrification) are then
used as the carrier gas at a flow-rate of 30 calculated to estimate rates of
ml-min-1, the oven temperature is 50'C mineralization.
and the detector temperature 200'C.
Measurements are done on peak areas.

Sampling methods and comparative data from natural wetlands

4. Vegetation Vegetation maps. Aerial
photographs are obtained to determine
composition and overall vegetation coverage in each
growth marsh. General attributes to obtain are
the total area of each constructed marsh,
the amount of open (unvegetated space),
Objectives. Wetland vegetation Iis and the area of each type of vegetation
the most obvious and straightforward (e.g., lower intertidal marsh dominated
indicator of habitat condition. Surveys of by cordgrass, mid-intertidal marsh
vegetation are needed to document the dominated by mixed succulents, high-
success of plant growth as well as to intertidal marsh dominated by various
assess the site's potential for supporting species and with potential habitat for salt
mal populations. marsh bird's beak). Next, representative
subunits of each vegetation type in both
The species composition of habitats the constructed and reference wetlands
within an area allows one to read the are selected for detailed sampling to
history of the site. Changes in vascular determine if the plant communities are
plant distributions lag behind similar.
environmental changes, because most
species are limited in their ability to Approximate elevations are
become established even when the habitat determined from a topographic map (one-
is appropriate. Thus, the vegetation is an foot contour map needed) and used to
integrator of long-term conditions, more locate and set the length of each transect
than a measure of current events. The (50-m transects: are useful, but shorter
presence of cordgrass indicates a long lengths may be necessary in variable
history of good tidal flushing, but the topography). To characterize the high
species may persist for many months or marsh-upland transition area, the transect
even years after such conditions have should parallel the elevation band that
changed. The presence of cattails in a salt contains that habitat. About 40-50
marsh indicates that excess freshwater stations should be established in each area
inflows occurred at some time in the past, to be compared. Each transect should be
but does not prove. that flows are indicated on an aerial photo overlay.
currently being augmented. If a site has Elevations should then be surveyed at
few species present, there are two each sampling station. Each transect
possible interpretations--the habitat may should have a number, and the number of
be poorly suited for salt marsh sampling stations on each transect should
development, or the site may provide be recorded on the overlay.
suitable habitat but not yet support a
diverse vegetation due to isolation from Sampling stations should be located at
propagules or insufficient time for regular (5-m) intervals along permanent
establishment. transects. Sturdy wooden stakes should
be numbered with marine paint and
The plant species composition dictates placed so they barely protrude above the
the suitability of sites for a wide variety canopy. Tall stakes attract raptors that
of animals. Insects are among the most prey on water-associated birds (especially
host-specific species, but several wetland young chicks), and it would not be
birds are also restricted to areas desirable to place predator roosts in a
dominated by specific plants. Additional marsh used by clapper rails.
linkages between vascular plants and
animals, especially those in the soils and Quantitative sampling fo r
and root zone (rhizosphere), still await vascular plant species composi-
discovery. tion. A long-term sampling program at
Tijuana Estuary (1979-present) provides

Sampling methods and comparative data from natural wetlands

the most extensive data set for salt marsh because of their trailing and branched
species composition. Comparable habit and the difficulty of deciding what
sampling methods are thus constitutes an individual. For this
-ed. Because
recommended. The sampling program reason, cover data are prefer7
records species presence (for frequency the sum of cover of individual species
of occurrence data), visual cover may exceed 100%, a separate visual
estimates for all species, and more estimate must be made for total vegetative
intensive analysis of cordgrass, which is cover.
often a restoration target species. The
most important feature for comparing Six cover classes have been used in
occurrence data is quadrat size; data from the Tijuana Estuary monitoring program.
different size quadrats are not Frequency histograms of cover classes
comparable, as larger quadrats encounter are readily compared with the
more species. Quarter-square meter Kolmogorov-Smirnov two-sample test.
quadrats are suitable for salt marsh Mean cover can be estimated by using the
vegetation, because several individuals midpoints of each cover class. Because
can be found within that area. Quadrat of the imprecision involved in estimating
shape should also be held constant. cover, differences of less than 25%
Circular quadrats were chosen for Tijuana between two sites are not meaningful.
Estuary, because an initial objective was
to understand interspecific: interactions.
The species occurring together in a 0.25- Cover Midpoint of
m2circular quadrat have intermingling class cover class
roots. Thus, data on species that occur >0-1% 0.5%
together are useful in evaluating
interactions. Additional useful 1-5 3.0
measurements are canopy height and data 6-25 18.5
on flowening and/or fruiting by species. 26-50 38.5
51-75 63.5
To determine species composition, 76-100 88.5
cover, and canopy heights, permanent
sampling locations (quadrats. along
transects) are established and marked for Cover of vegetation is compared more
elevation. Thes6 are sampled in late precisely by sampling canopy intercept
August or September for presence and along the transect lines. The intercepts of
cover of each species and for heights of each species and of bare space are
cordgrass stems, within 0.25 m2 circular recorded within 4- or 5-m segments of
quadrats. each transect line. The smallest unit of
intercept recorded is 1 dm. Lines are
Additional data are useful for placed along elevation contours, and the
cordgrass, which has a growth form that meters of cover within 4-m segments of
allows easy measurement of culm. height. the line are recorded to the nearest
In dense stands, 0.10 m2 quadrats are decimeter. A simple comparison of man-
adequate to give reliable estimates of the made and natural wetlands uses a
mean cordgrass growth. Heights are cumulative frequency histogram of the
summarized to provide total stem length number of segments with 0, 0.1-1.0,
(sum of all heights, which is a good 1.1-2.0, 2.1-3.0, and 3.1-4.0 meters of
estimator of aboveground biomass), cordgrass cover (Figure 4.1, data from
average height, and density (number of Swift 1988). In the cumulative
stems). histogram, the tally for the first cover
class is recorded, then the tally for the
Other species are less amenable to second class is added to the first, etc.,
height measurement or density counts, until all tallies are included in the last
point of the graph. From this summary

Sampling methods and comparative data from natural wetlands

of the data, it is clear that the man-made marsh plants, based on Carpinteria Salt
marshes have far more area of zero or Marsh. His cumulative list of salt marsh,
low cordgrass cover than the reference brackish marsh, and transition to upland
wetland, Paradise Creek. The man-made includes 38 species (Table 4.1)..
site that most closely approximates
cordgrass cover in the natural marsh is The regional distribution of a subset
Nursery 1, which was in its third of these species is given in Table 4.2.
growing season at the time of Swift's This regional comparison shows that
survey. species richness relates to the degree of
tidal flushing--wetlands with a long
Censuses of target species history of tidal flushing have most of
(desirable and undesirable) should be these species.
made during the period of time when they
are most conspicuous. High marsh is
examined in April to locate and census
salt marsh bird's beak - patches
(Cordylanthus maritimus ssp. maritimus)
and associated (potential) host species.
The size of each patch is measured
(maximum diameter and the diameter
perpendicular to it), and counts of
individuals are made. Soil salinities at
10.- and 30-cm depth are measured at each
patch. If rainfall has been late, April may
be too soon to census patches, and
searches may need to be repeated later.

Patches of weedy species are located
and sampled in April, as many of the
exotics are short-lived annuals. The area
of weedy vegetation is measured as
above, and soil salinities recorded for 10-
and 30-cm depths.

An additional plant population to be
monitored more closely is the rare annual
goldfields, Lasthenia glabrata, which may
be found at the edges of seasonal pools or
salt pannes that impound rainfall. The
population at Los Pefiasquitos Lagoon MG. JIG
has been monitored for several years
(Nordby, SDSU, unpub. data), by
locating and measuring the size- of
patches, plus obtaining density estimates
using the nearest-neighbor method and b
direct counts of densities using 0.25 rn@
quadrats and 0. 10 m2 quadrats.

Salt marsh bird's beak
Cordylanthus maritimus
Reference data: Tidal marsh ssp. maritimus
community composition: Ferren
(1985) gives the most complete list of salt

Sampling methods and comparative data from natural wetlands

Figure 4.1 The cumulative percent of 4-m intervals with 0-to-4 meters of cordgrass
cover near Sweetwater Marsh, San Diego Bay in summer 1987. Each transect followed the
elevation of maximum cordgrass establishment. The Connector Marsh sites were graded
and opened to tidal flow in fall 1984 and planted with cordgrass in Jan.-Mar. 1985.
Nursery I was graded and planted in July-Nov.1983. Paradise Creek is a natural marsh
remnant just upstmam of the Connector Marsh.

Nine Sites in the Connector Marsh

100-
0-0-

80-

60-
Cr

LL
40-
> k
cc 20- Nursery 1
E Paradise Creek

0 0
0 1 2 3 4
Meters of Cover Within 4-m Interval

Table 4.1. Native plant species list for "estuarine wetlands" of the Carpinteria salt marsh
(from Ferren 1985). *=not seen by Ferren; possibly extirpated from this wetland.

Anemopsis californica Hordewn depressm
Arthrocnemum (=Salicornia) subterndnale Hymenolobus procumbens
Aster subulatus var. ligulatus Isocom veneta var. vernonidides
Atriplex californica Jawnea carnosa
A. lentiformis ssp. breweri *Juncus acutus var. sphaerocappus
A. patuld ssp. hastata J. bufonius
A. watsond Lasthenia glabrata ssp. coulteri
Baccharis douglasii Limonium californicum
B. pilularis ssp. consanguinea Monanthochloe littoralis
*Carexpraegracifis Salicornia iftinica
*Chenopodium macrospermum ssp.farinosum Scirpus californicus
C. strictum S. maritimus
Cordylanthus nwritimus ssp. maritimus *S. pungens
Cressa trad1lensis ssp. vallicola Spergulww mcrotheca var mcrotheca
Cuscuta salina S. mrina
Distichlis spicata ssp. spicata Suaeda calceoliformis
Euthanda occidentaUs S. californica var. pubescens
Frankenia gran&folia ssp. gran&folia Triglochin concinna
Heliotropium curassavicum ssp. occulatum Typha doiningensis

Sampling methods and comparative data from natural wetlands

Table 4.2. Presence of selected native plants in southern California coastal salt marshes.
X=extant, based on literature or personal observations; *=reintroduced, long-term status
uncertain; #--recently declined from greater abundance. Bold type indicates fully tidal
systems (ocean inlets are rarely closed). [Please forward any new information for these
species lists to PERL.]

Species codes: Fp=Frankenia palmeri; Cm=Cordylanthus maritimus ssp. maritimus;
Bm=Batis maritima; Sb=Salicornia bigelovii; Sf=Spartinafoliosa; Lg=Lasthenia glabrata;
Tc=Triglochin concinnum; Aw=Atriplex watsonil; Lc=Limonium californicum, Ct=Cressa
.truxillensis; MI=Monanthochloe littoralis; Se=Suaeda esteroa; Cs=Cuscuta salina;
Ja=Juncus acutus; Sv=Salicornia virginica; Fg=Frankenia grandifolia; Ds=Distichlis
spicata.

Fp Bm Sf Tc Lc MI Cs Ss Sv Ds
Coastal Salt Marsh Cm Sb Lg Aw Ct Se Ja JC Fg

Sweetwater Marsh x x x x x x x x x x x x x x x x x x 18
Tijuana Estuary X X # X X X X X X X 4 X X X X X X X 18
Mugu Lagoon X X X X X X X X X X X X X X X X X 17
Anaheim Bay * x x x x x x X x x x x x x x x x x 16
Upper Newport Bay X X X X X X X X X X X X X X X X 16
Santa Margarita Estuary x x x X'X x x x x x x x x x 14
Bolsa Chica Wetland x x x X X X X X X X X X X X 14
Mission Bay Reserve X X X X X X X X X X X X X 13
Carpinteria x X X X X X X X X X X X X 13
Goleta Slough x x # x x x # x x x x x x 13
Los Pefiasquitos Lagoon X X X X X X X X X X X X X 13
San Dieguito Lagoon X X X X X X X X X X X 12
Batiquitos Lagoon x x x x x x x x x x 10
Agua Hedionda Lagoon x x x x x x x x x x 10
Ballona Wedand x X x x x x 6
San Elijo Lagoon x x x x x x 6
Deveraux Lagoon x x x x x x 6
Santa Clara R. Estuary x x ' x x x x 6
San Luis R. Mouth x x x x x 5
Las Flores Marsh x x x x x 5
McGrath Lake x x x x x 5
Malibu Creek x x x x x 5
San Mateo Marsh x x x x 4

Sampling methods and comparative data from natural wetlands

For a few salt marshes in southern useful in showing transect- to- tran sect
California and northern Baja California, variability in species abundance.
descriptions are available in the published Species composition within salt
literature. See Ferren (1985) for data on marshes is related to elevation, which in
Carpinteria Marsh, Onuf (1987) for turn indicates differences in inundation,
Mugu Lagoon, Schreiber (1981) for salinity, and a host of other environ-
Ballona Wetland, Vogl (1966) for Upper mental factors. The following data were
Newport Bay; Zedler and Beare (1986) obtained in the September 1988 annual
for San Diego River Marsh, Zedler census of the Tijuana Estuary salt marsh,
(1977) for Tijuana Estuary, and including 207 quadrats between 5-18 dm
Neuenschwander et al. (1979) for San MSL (approximately 1.5-6 ft MSL, or
Quintin Bay, Baja California, Broader 4.5-9 ft NGVD). Since most of the
discussions of the California salt marshes quadrats occur on the marsh plain, these
appear in Macdonald and Barbour (1974) data characterize the lower- and middle-
and Macdonald (1977, 1988). Data for marsh habitats best. The % frequencies
additional wetlands appear in less widely- are provided by elevation for all species
circulated reports. The US Fish and that occurred in more than 10% of the
Wildlife Service has provided preliminary quadrats. Occurrences of the less
results for Santa Margarita Estuary common species are listed. Cover (mean
(Hollis et al. 1988), and the Topanga-Las %, based on cover classes) is for
Virgenes Resource Conservation District quadrats of occurrence only (excluding
has released a baseline survey for Malibu O's), so quadrat n is not constant.
Lagoon (Manion and Dillingham 1989). Elevation class 5-6 ranges from 5.0 to
Additional wetlands under study by 6.9 dm, etc.
PERL/SDSU researchers include Ballona
Wetland, Los Pefiasquitos Lagoon, and The summary data for occurrence and
Sweetwater River Wetland Complex. cover indicate different attributes of these
General descriptions of many of the species. Pickleweed (Salicornia
region's coastal wetlands have been virginica) is the most abundant, and it
published by the California Dept. of Fish generally has high cover. Saltwort (Batis
and Game, as part of a Wetland maritima) is widespread at lower
Resources Inventory, carried out in the elevations, but never has high cover.
1960's and 70's. Shoregrass (Monanthochloe littoralis) is
restricted to the high marsh, but this mat-
The detailed studies of marsh forming grass generally has high cover.
vegetation are useful in showing how From the 1974 distributions at Tijuana
different species respond to reductions in Estuary, we have identified the elevation
tidal flushing. The composition of Santa where percent occurrences are greatest
Margarita Estuary (Table 4.3), which is and indicated the average of the maximum
often closed to tidal flow in summer, is in heights these species achieved within all
strong contrast with the species-rich quadrats of occurrence (Table 4.5).
marsh at Tijuana Estuary (Table 4.4).
There are fewer species overall, and Upper marsh and transition habitats
Salicornia virginica has very high have not been studied extensively. There
dominance. Tijuana Estuary, in turn, lost are few areas of undisturbed transition-to-
some of its richness during the 1984 upland habitat . At Tijuana Estuary, one
nontidal period of 8 months. At Santa gradual slope adjacent to the salt marsh
Margarita Estuary, three pickleweed was sampled for species composition and
habitat types were distin-guished by results combined with the 1974 census
Hollis et al. (1988), but the tabular data data. The data indicate in general how far
show considerable overlap in species upslope many of the marsh species go
composition among sites that appeared (Table 4.6).
different to the eye. The results are

Sampling methods and comparative data from natural wetlands

Table 4.3. Salt marsh vegetation at Santa Margarita Estuary from Hollis et al. (1988).
Data in each column are meters of cover from one 100-m-long transect. Vegetation types
are as named by Hollis et al. (1988). Species with less than 1% cover are indicated with an
X.

Salicornial Salicornial Salicornial Maritime
Frankenia Distichlis U Scrub
@pland

Bare ground 1 0 21 30 26 9 0 6 7 22
Mean Height (cm) 47 41 28 23 38 24 34 31 55 124

Salicornia virginica 80 54 58 28 60 5 2 27 11
Jaumea carnosa 4 4
Suaeda esteroa 3
Frankenia grandifolia 18 23 x x 16
Distichlis spicata 21 21 42 14 35 x
A triplex watsonii x
Salicornia subterminalis 54 24
Cressa truxillensis x 2 4
Atriplexpatuld 3
Croton californicus 3
Isocoma venetus x 16 36
Opuntia sp. 7
Heliotropium curassavicum x
Baccharus glutinosa 13
Ambrosia chamissonis x
Weedy exotics x 32 69 4 66 13

Table 4.4. Species Composition of the Tijuana Estuary salt marsh, 1988.
Mean
Elevation class (dm) 5-6 7-8 9-10 11-12 13-18 overall Cover

Quadrat n 43 101 36 14 13 207 Varies

Species and Frequency % % % % % %. %

Spartinafoliosa 74 39 22 0 0 38 49
Batis maritima 33 63 42 0 0 45 20
Jaumea carnosa 12 32 8 0 0 19 16
Salicornia virginica 86 -96 86 39 0 82 52
Frankenia grani*folia 9 34 44 43 _@L4 32 22
Monanthochloe littoralis 0 6 44 93 46 20 47
Salicornia subterminalis 0 0 14 64 85 12 44
Triglochin concinnum 7 10
Distichlis spicara 7 17
Limonium californicum 2 6
Cressa traxillensis 5 10
Additional species present in the marsh but not encountered in the 1988 sample
Salicornia bigelovii
Suaeda esteroa
Atriplex watsonii
Cordylanthus maritimus ssp. maritimus

Sampling methods and comparative data from natural wetlands

Table 4.5. Summary data for peak firquency of occurrence and the average of maximum
heights observed at Tijuana Estuary.

Elevation of Ave. max.
peak frequency height
Species (dm MSL) in cm (s.e.)

Spartinafoliosa 3 60 (10)
Salicornia bigelovii 6 38 (10)
Bafis mafitbw 6 24 (5)
Salicornia virginica 5 36 (9)
Jawnea carnosa 7 21 (4)
Suaeda esteroa 9 29 (8)
Frankenia granXfoha 10 23 (6)
Monanthochloe littoralis .11 24 (6)
Distichlis spicata 9 25 (6)
Linwniwn californicwn 10 no data
Triglochin maritima 7 no data
Salicornia subterrdnalis 11 31 (6)
Cuscuta salina 9 no data
Cressa truxillensis 10 no data

Table 4.6. Elevation ranges of salt marsh species. Occurrences are for 2-dm elevation
classes, combining the 1974 survey and the analysis of Site 1. Class 3 indicates 3.0-4.9
dm NGVD or 1.0-1.6 ft NGVD. Metric units are useful because the vegetation responds
to elevation differences as small as 1 dm (10 cm).

Elevation class (relative to NGVD)
decimeters 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31
feet 1.0 2.3 3.6 4.9 6.2 7.5 8.9 10.0

Spartinafoliosa x- x
Batis rwritirm x x x
Salicornia bigelovii X x x X
Jaumea carnosa x X X x
Suaeda esteroa x x x x
Salicornia virginica x x x X x X x x X x
Frankenia grandifolia x x x x X x x x x x x x
Monanthochloe liuoralis x x x x x x x x x x
Distichlis spicata x x x x X X x x x x x x x X
Triglochin mantirw x x
Cuscuta salina x x x
Limonium californicwn x x x 1
Salicornia subterminalis x x X x x x x
Atriplex watsonii x x x x x x x
Cressa truxillensis x x x X X x x x x X X

Sampling methods and comparative data from natural wetlands

No salt marsh is stable; composition production of the region's dominant
changes annually. From the long-term species (Onuf 1987).
monitoring program, we know that cord-
grass responds to varying environmental By comparing data from tagged
conditions (Zedler 1983, Zedler and branches with harvest data, Onuf deter-
Nordby 1986); it declined with hyper- mined that large amounts of leaf and stem
saline drought (1984) and recovered material are lost from plants between
slowly thereafter. Annual pickleweed harvests. Thus, estimates of NAPP from
(Salicornia bigelovii), which was repeated harvests were too low by a fac-
extremely abundant at the same elevations tor of 2.3 for Salicornia virginica, 3.7 for
as saltwort in 1974 (Zedler 1977), was Jawnea carnosa, 1.89 for Limonium cali-
nearly extirpated during the 1984 fornicum, and 2.9 for Batis maritima
drought; it may persist in one small patch. (Onuf 1987, p. 7 1):. Additional errors are
Sea blite (Suaeda esteroa) also declined in no doubt present due to the high hetero-
1984, but persists more widely, although geneity of aboveground biomass that is
only as scattered individuals. In the present in the region's species-rich
1984-1988 record, three species have marshes. Unlike monotypic Spartina
been consistently common (S. virginica, alterniflora marshes of the Atlantic and
B. maritima, F. grandifolia), 2 have not Gulf of Mexico Coasts, a quarter-square
yet recovered from post-1984 declines Q. meter of southern California marsh vege-
carnosa, T. concinnum), and 4 species tation may include 8 or 10 different
have been consistently uncommon (M. species. Thus, obtaining representative
littoralis, S. subterminalis, D. spicata, L. samples of the biomass of individual
californicum). These latter four species species is nearly impossible. Combining
are not rare at Tijuana Estuary; their low biomass of all species for a total NAPP
frequencies of occurrence simply reflect estimate is not recommended, because
the low numbers of quadrats sampled at individual species reach peak biomass at
their "preferred" elevations. The different times; thus such harvest data
strongest differences are seen by would further underestimate productivity.
comparing the 1974 samples with those In common with harvest studies in all
of 1988 (Table 4.7). regions and habitat types, net productivity
estimates based on standing crops do not
Even if composition changes little, the take into account losses due to herbivory
height and vigor of plants can be highly between harvests.
variable fi-om year to year. The dynamics
of Spartina foliosa and Salicornia vir- Earlier studies (Winfield 1990, Zedler
ginica have been studied in detail and et al. 1980, Eilers 198 1) erred in assum-
results published elsewhere (Zedler 1983, ing that southern California harvest data
Zedler et al. 1986). Such information could provide accurate estimates of salt
indicates that single samples of the vege- marsh vascular plant productivity. Their
tation do not fully characterize its data are best used as descriptions of the
condition. aboveground standing crop. Eilers' har-
vest data for 1977-78 have been
reanalyzed for that purpose; sampling
stations for the entire marsh have been
Biomass and net aerial primary pooled, and data for all species com-
productivity (NAPP) of vascular bined. The results for live and dead
plants. Harvesting vegetation to biomass (Figures 4.2-4.3) show that
estimate standing crops or primary biomass accumulates in the absence of
productivity is not recommended for tidal flushing (i.e., in Los Pefiasquitos
southern California coastal marshes. The Lagoon).
harvest method is too destructive for our
remnant wetlands. Furthermore, the
standing crops grossly underestimate

Sampling methods and comparative data from natural wetlands

Table 4.7. Frequency of occurrence (% of quarter-square-meter quadrats sampled)
at Tijuana Estuary in 1974 and 1988, comparing different sampling locations, but
using the same dm elevation classes (6 6.0-6.9 dm. NGVD). The larger %
ftequencies are in bold.

Elevation class (dm) Year

n 1974 113 95 35 30 35
n 1988 37 77 24 25 11

Occurrence

Spartinafoliosa 1974. 24 0 0 0 0
1988 78 48 8 32 0

Bads ntafifi?w 1974 86 71 31 0 0
1988 30 61 71 44 36

Salicornia bigelovii 1974 90 83 46 17 14
1988 0 0 0 0 0

Jawnea carnosa 1974 42 88 60 53 51
1988 14 31 33 12 0

Suaeda esteroa 1974 26 43 63 63 34
1988 0 0 0 0 0

Salicornia virginica 1974 69 66 80 70 46
1988 86 96 96 100 54

Triglochin concinnwn 1974 6 12 3 10 0
1988 5 12 8 0 0

Frankenia grandifolia 1974 23 54 77 87 91
1988 11 30 46 40 54

Limoniwn californicwn 1974 7 12 11 10 20
1988 0 1 0 8 0

Distichlis spicata 1974 2 9 43 23 0
1988 0 4 4 4 9

Salicornia subtenninalis 1974 6 6 29 30 69
1988 0 0 0 0 46

Monanthochloe Huoralis 1974 19 32 63 90 97
1988 0 3 17 36 64

Cressa truxillensis 1974 0 0 0 0 0
1988 0 0 0 8 27

Sampling methods and comparative data from natural wetlands

Figure 4.2. Summary of live material harvested by Eilers (1981). Data are means L+1
s.e.) obtained from 0. 10 m2 rectangular quadrats. LPL-- Los Pefiasquitos Lagoon, n=21
stations; SRE= Sweetwater River Estuary, n=3 I stations; UNB= Upper Newport Beach,
n=25 stations.

1750-
LPL
> -, 1500-
04
E 1250-

1000-
0
h- SFE
CD ca 750-
4) E
> 0 500-
0 -
UNB
250

0
Nov Jan Mar May Jul
Month 0 977-78)

Figure 4.3. Summary of dead material harvested by Eilers (1981). Data are explained in
Figure 4.2.

2000-
001%
C*41 1750- LPL
E
C" 1500-
U) 1250-
U)
0%
E 1000-
0
750- SFE

V 500-
Cto
4)
250- UNB
0- @ar May
Nov Jan Jul
LPL

UNB

Month (1977-78)

Sampling methods and comparative data from natural wetlands

creeks, especially where tidal flushing is
The large difference in the amount of minimal (Rudnicki 1986, Fong 1986).
dead biomass at nontidal LPL suggests Thick mats of filamentous blue-green
that tidal flushing leads to faster decom- algae and diatoms grow on the moist
position and/or export, both of which are intertidal soils of the salt marsh and are
consistent with the findings of Winfield especially productive in summer (Zedler
(1980). Additional historic data on 1980,1982b).
standing crops appear in Winfield (1980)
for Tijuana Estuary and in Zedler et al. Estimating algal productivity rates is
(1980) for Tijuana Estuary, Los most easily accomplished by measuring
Peftasquitos Lagoon, and San Diego the amount of oxygen evolved by fronds
River Marsh. or mats placed in aquatic chambers (e.g.,
as carried out in the field by Zedler
There is a method for measuring vas- 1980). The oxygen concentration is mea-
cular plant productivity using individual sured initially and after about an hour's
leaves or branches and assessing carbon incubation time. Comparison of changes
dioxide uptake over short intervals of in light and dark chambers yields esti-
time (minutes). Portable systems are mates of gross primary productivity.
available (ca. $15,000) for measuring These short-term rates can then be used to
photosynthesis using infrared gas analy- estimate longer-term contributions of
sis to indicate changing concentrations of algae by relating gross photosynthetic
carbon dioxide in chambers that enclose rates to light regimes. The methods pro-
intact leaves. However, photosynthetic vide valuable comparisons of different
rates for small portions of the plant must algal types and different habitats, but sev-
be combined with estimates of total leaf eral errors develop when calculations of
area to obtain net productivity rates for an annual productivity are attempted (Zedler
area of marsh. Thus, some measure of 1980). In addition, sampling is. very
biomass is still needed, along with fre- destructive, especially if one attempts to
quent measurements to account for sea- characterize a large number of habitats.
sonal changes in rates and regressions to Sampling must be frequent, as algal
determine how rates vary with daily biomass and photosynthetic rates differ
inundation and temperature. Such mea- from week to week (with spring- and
sures are best used to compare functional neap-tide inundation regimes), and a large
differences of different plants or habitats number of replicates (for both light and
at single points in time. dark chambers) is needed to obtain
reliable results.
In our opinion, it is not practicable to
measure vascular plant productivity. Results for Tijuana Estuary (Zedler
While various methods can be employed, 1980) for a year of good tidal flushing
they are too destructive and do not yield (1977) indicated that algal mats beneath
data of sufficient accuracy to justify the the marsh canopy were highly produc-
damages to the site. Data on cover and tive, compared to Atlantic Coast wet-
height are more appropriate. lands, where taller, denser vascular plant
canopies are present. While good data
Algal productivity. Measure- relating productivity rates to algal
ments of algal productivity are likewise biomass or chlorophyll content are lack-
problematical. Yet this component is ing, it is conservative to assume that
probably very important to the food base, where algal mats are thick and
not only because of high growth rates, widespread, they will make an important
but also because of high digestibility contribution to wetland primary
(Zedler 1980, 1982a). Several kinds of productivity.
algae contribute to wetland productivity.
Phytoplankton and macroalgal mats To provide a general characterization
become abundant in channels and tidal of potential algal productivity, we

Sampling methods and comparative data from natural wetlands

recommend that the presence and general submerged macrophytes (primarily
abundance be noted on at least a quarterly macroalgae) having the highest per-area
basis. Dominant r rates (but covering a small area) and
ypes of algae should be
indicated, using the following categories: phytoplankton having the lowest per-area
Floating macroalgal mats (Enteromorpha, and whole-system productivity. This
Ulva); green epibenthic mats generalization is consistent with results
(Enteromorpha); blue-green epibenthic from the -salt marsh at Tijuana Estuary,
mats, and diatom films. There is no easy where benthic algal mats appear to be as
way to obtain accurate estimates of algal important to ecosystem productivity as
abundance. Cover estimates (classes as are the vascular plants (Zedler 1980).
for vascular plants) can be made within
specified areas, e.g., 1-5 m2 sampling
stations that follow the shape of channels
or creeks; square-meter segments of the
salt marsh, etc.

I The overall primary produc-
tivity function. Although primary
productivity is a basic ecosystem
function, there are major problems and
errors in measuring productivity rates and
calculating the contributions of different
producer components for different wet-
land areas are great. A further concern is
the destructiveness of the sampling.
Thus, we recommend that productivity
studies be considered in research
programs, rather than as monitoring
objectives.

Given the unreliability of harvest data
for estimating NAPP, we are left with
only general remarks concerning the pri-
mary productivity of southern California
coastal wetlands. Onuf (1987) estimated
productivity of phytoplankton, benthic
microflora, submerged macrophytes and
emergent macrophytes for Mugu Lagoon,
but only after offering several precautions
about the data (ibid., p. 72): "The main
caveat about the productivity estimates for
the salt marsh vascular plants is the
uncertainty arising from the measurement
techniques and calculations. The monthly
estimates are imprecise because of the
high spatial heterogeneity of the marsh.
These errors are propagated in the math-
ematical manipulations used to generate
the annual estimates." His general results
for the eastern arm of Mugu Lagoon Nest of the
suggest that the benthic microflora, and light-footed clapper rail
emergent macrophytes are both principal Rallus longirostris levipes
contributors to wetland productivity, with

Sampling methods and comparative data from natural wetlands

(and occasional tiger beetle burrows)
5. Marsh insects: suggests great importance.
Pollinators, predators, These and other functions of the
and prey hundreds of coastal wetland insects merit
assessment of the insect fauna. We
recommend general surveys of selected
Objectives. The insects of south- habitat types to characterize the insect
ern California are responsible for several communities along with specific censuses
important salt marsh functions, including of sensitive species (globose dune beetle,
pollination, seed dispersal, aerating soils, wandering skipper, and tiger beetles).
controlling herbivorous insects, and pro-
viding food for birds, small mammals, Community sampling methods.
and other carnivores. Pan traps are used in nontidal areas to
trap crawling insects and many flying
While many of the plants are wind insects. The method can be used in tidal
pollinated, there are several species that areas during neap tides, if traps are
rely on insects for pollination and seed revisited before a high tide inundates
production. The endangered salt marsh them. Replicate traps (6-10 per site) are
bird's beak (Cordylanthus maritimus ssp. randomly placed within areas of relatively
maritimus) and the regionally rare uniform plant canopy. Sampling in the
goldfields (Lasthenia glabrata) are warmer parts of the year produces more
important examples. Members of the insects and different species than
Coleoptera, Diptera, Hymenoptera, and sampling in cool seasons, according to
Lepidoptera are important in pollination. K. Williams and K. Johnson (SDSU,
Pollinators link the upper salt marsh to pers. comm.). At Tijuana Estuary,
the adjacent coastal scrub-dominated September samples had abundant beetles
upland, where alternative nectar- and ants, while January samples were
producing plants are found. Thus, a fully dominated by flies. Early spring
functional marsh has nearby transitional sampling is suitable for many butterfly
and upland habitats that maintain an species; pollinators would be best
abundance of pollinators. sampled in May; late summer is
The herbivore-control function was appropriate for tiger beetles.
shown to be critical at the dredge spoil The pan traps are small cake pans
island in south San Diego Bay. Four (approx. 23x33x5 cm deep) painted
years after cordgrass was planted, a yellow on the inside and placed in a
native scale insect (Heliaspis spartina) shallow excavation under the plant
reached epidemic densities and severely canopy, so that the edge of the pan is
reduced the@ aboveground biomass of flush with the ground and vegetation.
cordgrass. Kathy Williams (SDSU, Enough propylene glycol (anti-freeze) is
unpub. data) determined that the native poured into each pan so that the bottom is
predators (e.g., the beetle, Coleomegilla) covered to a depth of about 1 cm. Insects
were lacking, and an experimental control fall into the pans and cannot escape.
program involving the release of the After about 48 hours, the traps are
predator is now underway. strained through a sieve and the insects
are collected in 70% ethanol in labeled
Burrowing insects are an integral part jars. These can be stored for later sorting
of the soils of salt pannes and higher and identification.
intertidal marsh areas. While the role of
insects in soil aeration and organic matter Sweep sampling is necessary to
incorporation remains unquantified, the characterize insects in tidal areas and to
great abundance of rove beetle burrows capture more of the flying insects.
Standard butterfly nets are used for this

Sampling methods and comparative data from natural wetlands

sampling. Uniform numbers of sweeps at Tijuana Estuary from areas of
through vegetation are made along cordgrass (Spartina foliosa), tidal and
transects of similar length. To sample a nontidal pickleweed (Salicornia
marsh area dominated by Spartina virginica), dune, and transition to upland
foliosa, Williams et al. (unpub. report) (dominated by flat-topped buckwheat,
used 10 transects of 20 m length to Eriogonwn fasciculatum). Six orders of
sample, with 30 sweeps made along each insects comprised the bulk of the insects
transect. Transects were parallel and 10 sampled in September, January, and
in apart. April: Coleoptera, Collembola, Diptera,
Homoptera, Hymenoptera, and
Visual surveys are suitable for tiger Lepidoptera (Table 5. 1). In all, 22 orders
beetles in mudflat, sandy beach, and salt of arthropods were identified.
panne habitats. The numbers seen while
walking transects of known length are Patterns of spatial and temporal
recorded. Dune beetles may need to be variability were found, with upland and
seived from sand near and under dune transition habitats having the largest
vegetation. Chris Nagano (US FWS numbers of orders (10-11) present and
Endangered Species Office, Sacramento) winter generally having the fewest orders
indicates that mark-recapture techniques represented (:@7, except for the transition
can be used with these species. His habitat, which had 11 orders). Overall,
advice should be sought for work with all the habitats dominated by cordgrass and
rare and sensitive insects. pickleweed had the fewest orders (2-8)
represented. The summaries of arthropod
Neither the pan trap nor the sweep numbers (Figure 5.1-5.3) indicate
methods provide information that can be substantial differences between habitat
used to estimate densities, because of the types and suggest interactions with
potential for pan or net avoidance. The season. Ants were primarily in the
counts of insects are best interpreted as upland, in September; leafhoppers and
relative densities for the habitats and planthoppers were most abundant in
times sampled. Such data are valuable Sparrina in January, but not in April; and
indicators of general abundance; flies were particularly numerous in
information on the abundance or absence Salicornia.
of certain functional groups (e.g.,
pollinators, predators) and of sensitive Williams and several students are
species is essential for characterizing continuing to investigate patterns of insect
wetland ecosystems. abundance and distribution at local
wetlands, and a better understanding of
Species identification is the biggest both structural and functional aspects of
problem with characterizing the insect the arthropod community should be
community, and it may not be possible to available in the next few years.
identify many taxa beyond the family Additional species lists, but no
level. However, this is often very useful quantitative data, are available for Ballona
for examining functional groups. Even Wetland, north of Los Angeles airport
general information on size and habit (Nagano, in Schreiber 1981).
(flying or crawling) will be helpful in
characterizing insects as potential food
items for consumers such as Belding's
Savannah sparrows. Species of special
concern, such as the tiger beetles,
globose dune beetle, and wandering
skipper, need to be identified to species.

Reference data on coastal insects
(Williams et al. 1989) have been obtained

Sampling methods and comparative data from natural wetlands

Table 5.1. Orders of Arthropoda at Tijuana Estuary, in order of decreasing abundance in
pan traps (from Williams, et al., unpub. report). In all, 17,627 individuals were captured.

Order Common name % of total

Diptera flies 28.1
Hymenoptera bees, wasps 10.4
Formicidae (ants) 21.9
Homoptera, lea&oppers, planthoppers 19.7
Coleoptera beetles 6.4
Lepidoptera butterflies, moths 1.6
Collembola springtails 1.1
Araneida spiders 0.8
Isopods pillbugs 0.6
Amphipods amphipods 0.5
Dermaptera earwigs 0.4
Heteroptera true bugs 0.4
Orthoptera, grasshoppers, crickets 0.4
Acarina mites, ticks 0.3
Solpugida wind scorpions 0.1
Thysanoptera, thrips 0.1
Tricoptera caddisflies 0.1
Odonata dragorx/damselflies <0. 1
Thysanura silverfish <0. 1
Isoptera termite <0. I
Neuroptera lacewing <0. 1
Phalangida, harvestmen <0. I
Psocoptera psocids, book lice <0. 1

Figure 5.1. Relative abundance of selected orders among five habitats sampled at
Tijuana Estuary during September 1988 (from Williams et al., unpub. report).
30- Distribution of Arthropods Sept. 1988

Araneida
Coleoptera
Collernbola
20- Diptera
0 Homoptera
0
Hymenoptera
Hym:Formicidae
0 Lepidoptera
--'0 to- Others
0

0
Spartina Salicornia Dune Transition Upland

Sampling methods and comparative data from natural wetlands

Figure 5.2. Relative abundance of selected orders among five habitats sampled at
Tijuana Estuary during January 1989 (from Williams et al., unpub. report).

30- Distribution of Arthropods Jan. 1989
Araneida
Coleoptera
Collembola
Diptera
20- Homoptera
0 Hyrnenoptera
Hym:Formicidae
0 Lepidoptera
E3 Others
10-
0 MAE -' HIM
Spartina Salicornia Dune Transition Upland

Figure 5.3. Relative abundance of selected Arthropod orders among five habitats
sampled at Tijuana Estuary during April 1989 (from Williams et al., unpUb. report).

Distribution of Arthropods April 1989
30-

Araneida
Coleoptera
Collembola
20- Diptera
0 Hornoptera
0 Hyrnenoptera
Hym: Formicidae
0 El Lepidoptera
10 0 Others

IL r7i
0-- L-Mal
Spartina Salicornia Dune Transition Upland

Sampling methods and comparative data from natural wetlands

samples needed per station varies; more
6. Aquatic samples are needed to characterize areas
that are highly heterogeneous. A species-
invertebrates: Food area curve (cumulative number of species
chain support encountered, plotted against cumulative
area sampled, cf. Figure 6.1) for each
station will indicate whether or not the
Objectives. Benthic macroinverte- first few samples have encountered most
brates are good indicators of habitat of the species present.
quality and food chain support. Fully
tidal wetlands support a large number of The method proposed will capture
benthic invertebrate species, many of clams and ghost shrimp; worms and
which have long life spans and grow to some insects will also be obtained. Sieve
large size. In areas subjected to frequent size is the most important factor to keep
disturbances, such as dredging and constant, and a 1-mm screen is required
excess freshwater inflows, bivalves may to compare results with reference data
persist, but the species composition will sets. Samples are seived and all obvious
shift from longer-lived to annual species. animals identified in the field and
In the absence of good tidal flushing, released. Less obvious animals are taken
clams and ghost shrimp become replaced to the laboratory for identification.
by species of polychaetes that are tolerant Polychaetes are very difficult to identify,
of stagnant water and other environmental but representatives can be preserved for
extremes. Polychaetes also invade future reference. The numbers of
rapidly following catastrophic events, individuals by species should be
such as rapid salinity reduction (Nordby, compared temporally and between
SDSU, pers. comm.). stations. Additional information on size
of clams for common species is very
Methods.. Two methods are rec- useful; even the simple measure of the
ommended for sampling aquatic inverte- largest individual of each species at each
brates: sediment cores to assess benthic station or notes concerning the relative
burrowing forms and litterbag traps for abundance of large and small individuals
mobile invertebrates on the marsh surface would help to document large changes in
(small crabs, amphipods, and insect population structure. Further details
larvae). In addition, direct counts of (size-frequency distributions) would help
snails and of crab burrows are sometimes describe the food base.
used on mudflats. Each is discussed with
reference data below. - Benthic macro- Reference data for. benthic
invertebrates sampling stations are best cores. Our data from Tijuana Estuary
located near the fish sampling sites (see (Table 6.1-6.3) are useful for describing
below), where channel morphometry a community under disturbance stress,
(width, depth, bank characteristics) is rather than a restoration target. Historical
described. data from Tijuana Estuary and Mugu
Lagoon are used to indicate the "healthy"
Sediment cores. Benthic macroin- benthic community that should be
vertebrates are collected using a stainless expected for sites without dredging,
steel cylindrical "clam gun" 45 cm in sewage inflows, excessive street
length and 15 cm in diameter. This drainage, or wastewater discharges.
device is pressed into the sediment to a
depth of 20 cm. Replicate cores are taken
within a sampling station. A convenient
size station is a circular area of about 2 rn
in diameter. Usually, five cores per
station will include all of the species
present (Figure 6. 1). The number of

Sampling methods and comparative data from natural wetlands

Figure 6.1. Cumulative number of benthic invertebrate species for stations in the
southern and northem arms of Tijuana Estuary. Each sample was the sediment
collected with a clam gun and then sieved with a 1-mm screen.

30-

North arm, station 5
Z
I

CL 20-

0 South arm, station1

10-
E
3
z

0
0 1 2 3 4 5 6
Sample

Sampling methods and comparative data from natural wetlands

Table 6.1 Mean number of macroinvertebrate species/core sample in the north and south
channels of Tijuana Estuary (and s.e.). Data are for all species, and for polychaetes and
bivalves separately. Capitellid spp. were not identified to species and are not included in
the polychaete data. Data are from 6 "clarn gun" cores per station; each core had a diameter
of 15 cm (area 177 cm2) and was 20 cm in depth.

Station North South

All macroinvertebrates 1 (near mouth) 15.17 (1.1) 10.83(0.9)
3 (upstream of 1) 4.17(0.4) 3.67(0.56)
5 (upstream of 3) 10.83 (0.65) 1.67(0.5)
7 (upstream of 5) 3.00(0.37) 2.3(0.42)

No. of Polychaete spp. 1 6.17(0.79) 6.17(0.47)
3 3.00(0.26) 0.67(0.33)
5 2.83(0.60) 0.67(0.33)
7 1.50(0.56) 0.00

No. of Bivalve spp. 1 3.67(0.61) 2.33(0.56)
3 0.50(0.22) 1.83(0.40)
5 3.33(2.45) 0.33(0.21)
7 0.67(0.33) 0.17(0.17)

Table 6.2. Densities of macroinvertebrates in the north and south channels of Tijuana
Estuary in November 1988. Data are mean numbers per core sample (n = 6 cores/station)
and standard errors (s.e.). Stations are as in Table 6.1.

Station North South

Capitellid worms 1 2.5(0.67) 114.7 (2.11)
3 0.17(0.41) 1.5 (0.81)
5 8.83 (2.61) 0.00
7 0.33 (0.33) 0.00

Other Polychaetes 1 19.00 (5.52) 13.17 (1.22)
3 6.83(2.45) 1.00(0.37)
5 5.17 (1.74) 1.33 (2.45)
7 5.67(2.80) 0.00

Bivalves 1 10.17 (1.35) 3.67(l.05)
3 0.67(0.33) 3.5 (0.62)
5 8.33(0.99) 0.33(0.21)
7 0.83(0.4) 0.17(0.17)

Sampling methods and comparative data from natural wetlands

Table 6.3. Comparison of numbers of individuals of benthic invertebrates collected at
Tijuana Estuary (tidal conditions, but with sewage inflows) and Los Pefiasquitos Lagoon
(primarily nontidal). Taxa comprising less than 5% are listed as "others." Note that
sampling efforts differed for the two wetlands. Data are from Nordby and Zedler (in
press).
Tijuana Estuary Pefiasquitos Lagoon
Taxon

Sipunculid worms
Themiste sp. 17
Echinoderms
Dendraster excentdcus 6 3
Nemertean worms 93 3
Polychaete worms
Capitellidae 814 521
Spionidae
Boccardia spp. 68 (5 spp) 183 (4 spp)
Polydora comuta 267 110
Polydora spp. 63 (2 spp) 210 (2 spp)
Spiophanes missionensis 117
Unidentified spionid 163

Other taxa combined 437 181

Total polychaetes collected 1698 1368
Total families collected 13 11
Total polychaete species collected 35 20

Bivalve molluscs
Tagelus californianus 797 40
Protothaca standnea 554 4
Macorm nasuta . -221 6
Laevicardiwn subsViawn 30 8
Spisuld sp. 17
Other species combined 221

Total bivalves collected 1799 95
Total bivalve species -collected 18 12

Decapod crustaceans
Callianassa californiensis 234 3
Phoronida
Phoronis sp. 1 114
Brachiopoda.
Glouida albidia 1
Total sampling area (cumulative area in m2)
bivalves 4.77 m2 3.34 m2
other taxa 3.34 m2 3.34 m2
Total number of quarterly samples
bivalves 10 7
other taxa 7 7

Sampling methods and comparative data from natural wetlands

Macroinvertebrates in the north and Litterbag traps
south main channels of Tijuana Estuary
illustrate differences that relate to tidal We have found litterbag traps to be
flushing (Tables 6.1-6.2). In November very useful in collecting small
1988, the north channel (Oneonta invertebrates from marsh habitats. The
Slough) was fully tidal, while the south placement of such traps in the wetland for
channel had very little tidal influence, a month-long period allows a variety of
except at the station nearest the mouth. small invertebrates (flies, amphipods,
Mouth stations of both channels were beach hoppers, crabs, and snails) to
somewhat influenced by sewage inflows colonize the litter and be collected easily
from the Tijuana River. without damage to the habitat. However,
the animals trapped are an indicator of
Data from long-term sampling of the what is available to colonize substrates,
northern arm of Tijuana Estuary (Table and are not necessarily representative of
6.3) indicate that even though the estuary densities of organisms living outside the
is stressed by wastewater inflows, the traps. Thus, the method provides an
benthic community is still more diverse index of this invertebrate community,
than that of Los Pefiasquitos Lagoon, which is especially useful for
which is primarily nontidal. simultaneous comparisons of natural and
man-made wetlands.
Historic data from Tijuana Estuary,
before major floods and sewage inflows, Stations in the upper and lower
serve to characterize a "healthy" benthic intertidal marsh will attract different
community. Peterson (1972, 1975, species and densities of organisms. The
1977) sampled both Tijuana Estuary and method was modifed from Levings'
Mugu Lagoon (340N, 1190W) in the (1976) basket traps. Litterbags are made
1970's. His data for live bivalves in from 1-cm nylon mesh (black is
saline habitats show that Nuttallia preferred). Dried plant material weighing
(Sanguinolaria) nunaffli and Protothaca 40 g is placed inside each bag and the
staminea were the most abundant bivalves ends of the bag are folded over and
at both study sites. Tagelus stapled. The resulting bag is about 15 x
californianus, Cryptomya californica, 25 cm. For litter as filler, Rutherford
Macoma nasuta, and Laevicardium (1989) used dried cordgrass, (Spartina
substriatum were also present but in foliosa) left over from harvests of
lower numbers. experimental material at PERL. In
comparisons using native cordgrass and
Hosmer (1977) sampled bivalve commercial straw, it was clear that straw
composition and measured the sizes of trapped fewer animals. Comparisons
representatives of each species. Large between native cordgrass and a mixture
individuals were abundant. The mean of cattails and bulrushes gave similar
sizes for the dominant bivalves were: 71 results (Table 6.4). Because cordgrass
mm for Nuttallia nuttaffli ; 22 mm for should not be taken from native stands,
Protothaca staminea ; and, 27 mm for we recommend the use of cattails. and
Tagelus californianus. His results bulrushes from man-made wetlands.
contrast with recent data, i.e., Nuttallia
nuttallii no longer exists at Tijuana Three replicate litterbags should be
Estuary, and P. staminea is, on the placed at.each sampling station. Stations
average, half as large. of about 2 m in diameter are convenient.
Bags are secured at the soil surface with
wire flagging stakes. Seasonal sampling
produces different dominants, with the
largest numbers of individuals occurring
in winter. Litterbags should be left in the

Sampling methods and comparative data from natural wetlands

marsh for one month, then collected and constructed wetland. Rather, simul-
placed in zip-lock storage bags containing taneous sampling in natural and man-
a 10% buffered formalin seawater made marshes is recommended. At the
solution. Litterbags may be stored in Sweetwater River Wetland Complex, a
formalin for up to 5 days before gently comparison of natural and constructed
washing the litter over a 0.5 mm sieve to habitats (Rutherford 1989) shows that
collect the organisms. The catch is about 3 times as many individuals and
preserved in 70% isopropyl alcohol, and consistently more species were collected
identification requires microscopic in the natural marsh than in the mitigation
(dissecting scope) examination. marsh, which was assessed 4 years after
construction.

Results from bags collected in
Reference data for litterbag January 1989, after one month in the
traps. Data from Tijuana Estuary, Tijuana Estuary, are provided to indicate
Sweetwater Marsh, and Paradise Creek the kinds of species collected in litterbag
all indicate that sampling time and traps. The data are from the lower marsh
location are important variables. No set and adjacent mudflats. Only the results
of data can be recommended as a for bags made using Spartina are
reference point with which to compare a presented.

Table 6.4. Relative abundance (% of total) of invertebrates trapped in litter bags that
were filled with dried Spartina, Typha, or Scirpus. Litter bags were placed in Paradise
Creek Marsh for the month of March 1989. Each site had 4 replicate bags. The %
similarity of invertebrate composition (based on abundance) comparing Spartind and Typha
litter bags was 78%, comparing Typha and Scirpus was 75%, and comparing Scirpw and
Spartina was 86%.

S Sdrj2US
Pa Typ h a

Pericoma spp. (fly larva) 78.4 58.4 83.1
Collembola (springtails) 5.8 6.2 <1
Tromboidea (???) 5.8 4.5 2.4
Assiminea californica (small snail) 3.9 5.1 1.3
Unknown larvae 1.2 8.0 4.7
Capitellid worms (polychaete) 1.6 6.6 5.1
Orchestia traskiana (amphipod) <1 7.6 2.0
Ligia occidentalis (isopod) <1 2.4 <1
Saldidae (fly) 1.8 <1 <1
Other <1 <1 <1

Sampling methods and comparative data from natural wetlands

Table 6.5. Invertebrates collected at three stations, 2 intertidal elevations per stations, in
the north channel (Oneonta Slough) of Tijuana Estuary. Station 1 (S 1) was near the ocean
inlet; station 7 was a tidal creek near the northern limit of the wetland. Spartina was used
as filler for all bags.. Four taxa with only 1 individual in n=18 bags were included in the
total but not listed.

S1 Sl S6 S6 S7 S7
Oreanis lower upp lower uppe lower uppe Mean s. e,

Capitellidae 6 0 2 0 5 9 3.7 1.5
Spionidae 61 0 0 0 0 1 10.3 10.1
Polydora ligni 0 0 0 0 7 0 1.2 1.2
Streblospio benedicti 0 0 0 0 2 0 0.3 0.3
Assiminea californica 0 27 0 206 2 72 51.2 33.0
Melampus olivaceus 0 0 0 0 0. 4 0.7 0.7
Hemigrapsus oregonensis 26 0 9 0 32 10 12.8 5.5
Pachigrapsis crassipes 23 14 19 5 3 0 10.7 3.8
Juvenile crab I 1 0 8 2 7 5 5.5 1.6
Orchestia traskiana 4 17 4 6 12 45 14.7 6.4
Other amphipods 0 0 0 0 12 0 2.0 2.0
Ligia occidentalis 0 18 0 0 22 76 19.3 12.1
Diptera. 2 45 0 1 0 0 8.0 7.4
Dipteran larvae 1 2 0 5 5 35 8.0 5.5
Dipteran larvae B 0 0 0 0 3 0 0.5 0.5
Pupae 3 25 0 3 0 2 5.5 3.9
Unknown larvae 0 1 0 1 0 0 0.3 0.2
Unknown insect 0 0 0 3 0 3 1.0 0.6

Totals 137 149 42 235 112 263 156.3 33.2

Sampling methods and comparative data from natural wetlands

released outside the blocking nets.
7. Fishes: Community Repeated tows must be made and
continued until the total number of fishes
dynamics, controlling collected declines' in two successive
factors seifiings. The first and second seinings
capture the water-column species, while
later seinings collect bottom-dwelling
Objectives. Fishes are valuable species. Stomping on the bottom scares
indicators of ecosystem health. up benthic species (e.g., burrowing
Generally, the presence of few species gobies) that would otherwise allow the
(low species richness) indicates stressful net to drag over them. The blocking nets
environmental conditions. Gobies and are then closed by drawing the two
mullet are relatively tolerant of together in a serni-circle and then pulling
environmental extremes and are among them to shore, thereby capturing those
the last species to disappear., Fishes are fishes that managed to escape the bag
valued as foods for birds that use the seine. The successive seining results can
estuary. A few species are of recreational be used to estimate total population size,
or commercial interest, e.g., California using the standard catch-per-unit-effort
halibut and longjaw mudsuckers (as technique, plotting number per catch
baitfish). against cumulative catch and extrapolating
the line to obtain the cumulative catch
Methods. Fishes are sampled when catch is zero. The area seined
quarterly (March, June, September, and (length of segment and channel width)
December) at replicate stations. The should be measured so that densities can
habitat types to be sampled for fishes be calculated on a per-surface-area basis.
include deep channels, tidal creeks, deep Maximum depths are recorded.
saline ponds, brackish creeks, and
freshwater ponds. Fish sampling is Reference data comparing. two
carried out at moderate tide levels, when fish sampling techniques. To
stations are accessible by fb9t. Sampling demonstrate the need for repeated seining
at times of moderate tidal amplitude also to obtain representative data for fish
avoids tidal transport of transient species. densities and species composition, a fish
sampling exercise was carried out by
Blocking nets prevent fish from Chris Nordby At the "east-west channel"
escaping the sampling area. Standard at Tijuana Estuary on March 11, 1987. A
mesh size allows comparisons among 10-m segment of the channel was blocked
seining efforts. Adult and juvenile fishes at each end with seines of 3-mm, mesh,
should be collected using 3-mm mesh and the blocked area was sampled with a
blocking nets and bag seine. The 3-mm third 3-mm. seine in four repeated efforts.
size mesh ensure the capture of small yet Each of the four seining efforts was equal
ecologically important.goby species. A in area and seine length. Then, the two
linear distance (e.g., 10-15 m) parallel to blocking nets were "closed" by pulling
the creek or channel sampled should be each toward the center. The combined
measured and the channel nets deployed fish catch in this fifth seining effort was a
to confine all fishes within the two nets. double effort, so the numbers of fish
The bag seine is then drawn in a circle caught were halved to compare with the
within the blocking nets and pulled to first 4 seining efforts. Water depth was
shore. For pond habitats, a semicircular approximately 1 ni on the outgoing tide.
area should be enclosed with a net seine,
and the seine drawn toward the shore. Using the catch-per unit effort method
The species composition and numbers (Figure 7.1), it was determined that the
collected per tow are recorded, a total fish population within the blocked
subsample measured (for length- portion of the channel was 1470 fish
frequency distributions), and fishes (Y=159 -0.108X; for Y=O, X=1470; i.e.,

Sampling methods and comparative data from natural wetlands

when no more fish would have been Additional sampling showed that
present, 1470 would have been caught). gobies were the dominant members of the
This catch-per-unit-effort calculation fish community. Gobies are bottom-
provides a good estimate of total dwelling species that respond to the
population when sequential catches repeated disturbances of seining; their
decline smoothly, as in this examnple. numbers increased progressively in the
However, in many cases, a catch with first three seinings (Figure 7.2), and
few fish can be followed by one with declined slightly in the last two efforts.
many fish, as gobies move into the water Two mullet were caught in the later
column. We recommend seining seinings, but were not included in these
sequentially until the catch is clearly data summaries; it is well known that
diminished, then summing the numbers seines undersample mullet because they
caught to estimate the number of fish per swim or jump out of the seines.
area.
We conclude from this comparison
In the test case (Table 7.1), the first of seining methods that: A single seining
seining yielded 157 fish; which was only can underestimate fish density by as
about 10% of the estimated fish that could much as an order of magnitude. A single
have been caught there. For the 5 com- seining can misrepresent the fish
parable seining efforts, a total of 643 fish community; the most catchable species
were recorded. Of these, 395 were will appear to be the most abundant ones,
gobies, 160 topsmelt, 64 sculpin, 11 while the least catchable species will be
killifish, and 13 flatfish. absent or undersampled. In this data set,
repeated sampling produced only one
additional species (mullet), one that is
generally undersampled due to its net
Table 7.1. Comparison of single and avoidance behavior.
repeated seining efforts at Tijuana
Estuary. Data are relative abundances,
obtained by C. Nordby, SDSU, on
March 11, 1987. Species that appeared
to be dominant are shown in bold type.

lst seining Total in 5
effort

Topsmelt 70.1% 24.9%
Gobies 6.4 61.4
Sculpin 17.8 .10.0
Killifish 4.5 1.7
Flatfish 1.3 2.0

The first seining suggested that
topsmelt were dominant; however, as
repeated seinings show, they are merely
the most catchable species. In fact, 69%
4091- P__

of the topsmelt found were caught on the Staghorn sculpin
first seining. Topsmelt are water-column (Leptocottus armatus)
species that do not readily avoid seines.

-Sampling methods and comparative data from natural wetlands

Figure 7.1. Repeated seining of a blocked channel plotted against the prior cumulative
catch. Data of C. Nordby, PERL.

160- A
y 158.66 - 0.10657x R 2 0.862

140-
.C

120-

100-

80-
0 100 200 300 400 500 600
Prior Cumulative Catch

Figure 7.2. Species composition of repeated seinings. Data of C. Nordby, PERL.

175-

150-

125-
0 Total fish
100- Goby
Topsmelt
ILL Flatfish
75- Killifish
a Sculpin
so-

0
0 1 2 3 4 5 6

Seine Number

Sampling methods and comparative data from natural wetlands

Reference data for fishes of wetland. However, the differences
wetland channels. Nordby has used between the two systems have helped to
the same sampling methods to assess show patterns of decline that follow
populations of fishes at Tijuana Estuary hydrologic disturbances. Low numbers
and Los Pefiasquitos Lagoon for several of species and single-species dominance
years (Table 7.2, Nordby and Zedler in indicate low-quality habitat; absence of
press). The sampling program indicates longjaw mudsuckers may indicate a
large differences in species composition recent period of hypersalinity; the
and density for the two wetlands and for presence of mosquitofish indicates high
stations within Tijuana Estuary. or persistent fi-eshwater inflows.
Differences are related to the history of
hydrologic disturbances at the two The total species lists (Table 7.3) for
wetlands, with Tijuana Estuary being Tijuana Estuary and Los Peffasquitos
open to tidal flushing at all times during Lagoon indicates fewer species in the
the sampling and Los Pehasquitos latter wetland, but note that the sampling
Lagoon closing annually for much of the effort was also less.
warm season. During the closures, there
were floods in both October 1987 and Ichthyoplankton sampling. If
December 1988, the fresh water was juvenile or adult fishes are not found
impounded, and large numbers of fish using seines, the habitat might still be
were killed after salinity and oxygen suitable but larvae may not be available
levels dropped. At Tijuana Estuary, raw for settling. In this case, ichthyoplankton
sewage inflows were persistent each sampling should be considered to
year, and the sampling station closest to determine if young are available for
the Tijuana River was most depleted of its colonization. The lack of ichthyo-
fish community. plankton would indicate that a basic
ecosystem function is missing. Two
Both types of hydrologic modifi- periods are recommended for sampling
cations are human impacts--the sewage ichthyoplankton--sampling in March will
flows from Mexico have changed the capture nearshore species that move into
river from an intermittent stream to a the estuary with tidal waters, and,
year-round, nutrient-rich river that is sampling in April will assess availability
laden with toxic materials. The pro- of larvae of resident species such as
longed closure to tidal flushing has topsmelt and gobies. Nordby (1982)
resulted from the construction of roads gives detailed information on the year-
and a railroad (and associated filling and round distribution of ichthyoplankton at
reduced tidal prism). Thus, none of the Tijuana Estuary.
fish communities we describe serves as a
reference datum for a "healthy" tidal

Table 7.2. Fishes in two southern California coastal wetlands (from Nordby and Zedler
in press). Data are relative abundances (% of total caught in multiple censuses each year).

Tijuana Estuary Peffasquitos Lagoon
Species 1986-87 1987-88 1988-89 1987-88 1988-89

Atherinops affinis 52 14 7 38 36
Clevelandid ios 41 58 90 22 14
Fundulus parvipinnis 4 .19 1 4 2
Giffichthys mirabilis <1 <1 28 14
Gambusia affinis <1 24
Others 3 -8 -1 -7 10

Sampling methods and comearative data from natural wetlands

Table 7.3. Fish species collected at Tijuana Estuary and Los Peiiasquitos Lagoon (modified
from Nordby and Zedler in press). X = less than 1% of annual catch.

Tijuana Peilasquitos
Estuary Lagoon
Taxon Common name 1986-88 1987-88

Atherinidae Atherinops affinis topsmelt 15,437 1,875

Blennidae Hypsoblennius gentilis bay blenny x
Hypsoblennius gilberd rockpool blenny x
Hypsoblenniusjenkinsi mussel blenny x
Bothidae Paralichthys califomicus California halibut x x

Cottidae Leptocottus arniaw staghom sculpin 1,431 346
ArteWus sp. sculpin x

Cyprinodontidae Fundulus paryipinnis California killifish 2,367 107

Engraulidae Anchoa contpressa deepbody anchovy x 67

Girellidae Girella nigricans opaleye x x

Gobiidae Clevelandia ios arrow goby 60,097 816
Giffichthys mirabi& longjaw mudsucker 275 877
Ilypnus gilberd cheekspot goby x x
Lepidogobius lepidw bay goby x
Quietuld y-cauda shadow goby x
Mugilidae Mugil cephalus striped mullet x x

Plueronectidae Hypsopsew gumlata diamond turbot x x
Plueronichthys ritwi spotted turbot x

Poecilfidae Garnbusid affinis mosquitofish 937

Rhinobatidae Rhinobatus productus shovelnose guitarfish x

Serranidae Paralabrax clathraw kelp bass x

Sciaenidae Seriphus politus queenfish x

Syngnathidae Syngnathus leptorynchus bay pipefish x x

Total fishes collected 80,165 5,087
Total species encountered 21 13
Total sampling effort (cumulative area in M2) 4,795 1,985
Number of quarterly samples 12 8

Sampling methods and comparative data from natural wetlands

Aerial photos should be used to select
8. Birds areas of relatively homogeneous
vegetation and topography. Several short
Objectives. The most obvious and transects should prove more useful than a
few long transects in comparing mean
widely appreciated animals of the coastal densities among habitat types, as the
wetlands are the birds. In recent years, standard errors are likely to be lower
several studies have quantified coastal (Hanowski et al. 1990). Census dates
wetland birds (Boland 1981, 1988) and and times are then selected, keeping in.
identified their functional roles (Boland mind that coastal bird communities
1988, Quarnmen 1984). Additional work change with season, with inclement
on endangered species has been done, weather, and with tidal condition. Patrice
and some sites are monitored yearly for Ashfield and Barbara Kus (see below)
target species (US FWS). However, sample weekly during most of the year
long-term monitoring programs of the (biweekly in summer) in order to
total bird community are still needed. compare bird communities at high tide
Because birds are a major linkage and low tide, and in order to follow
between wetlands, coordinated monitor- changes with migration.
ing programs of all water-associated birds
throughout the Pacific flyway are needed. Reference data. A detailed, year-
In addition, we need to determine how long sampling program was initiated at
different habitats attract birds and the Tijuana Estuary during 1988-89 (B. Kus
length of time it takes target species to and P. Ashfield, SDSU, pers. comm.).
find and use constructed wetlands. The program had weekly censusing of
Bird community sampling. shorebirds and waterfowl (density by
Several methods have been developed for species) during months of high activity
sampling terrestrial birds (Ralph and (Oct.-Dec.) to include weeks with both
Scott 1981), which pose special problems spring and neap tides. Thereafter (Jan.-
because of their high mobility and low June), censuses were biweekly, and only
visibility in areas of dense vegetation. In at low tide. In this sampling program,
coastal wetlands, many of the species of counts were recorded by habitat type
interest are fairly large, occur in groups, within three main transects. The two
and feed in the open. Flocks of transects with greatest bird activity were
shorebirds are highly visible. Still, there censused simultaneously by qualified
are difficulties in selecting consistent observers, to allow comparison under the
habitat areas, since tides expose different same tidal conditions. The weekly
widths of shoreline each day. Observers surveys alternately characterized bird use
need to be skilled in estimating distances, under low and high tides. Censuses
since counts of birds per area could easily began one hour before the tide condition
be biased when transects; of unmeasured that was being characterized. Observers
width are censused. Walking transects carried binoculars, a spotting scope, and
along shorelines allows counts to be a two-way, radio to maintain contact with
expressed as number per length of one another and reduce duplicate counting
transect, even if transect width varies. of individuals flying between transects.
However, data from transects of fixed Counts were made by species, noting the
specific habitat type (Table 8. 1),
width and length are easiest to analyze.
inundation level (with or without standing
Boland (1988) used multiple transects of water) and noting behavior within three
fixed length (50 m) to obtain densities classes: foraging, roosting, or flying.
(means and standard errors) for each During each visit, additional notes were
habitat. made on selected species (raptors,
Habitat types are first delineated, and Belding's Savannah sparrows, and light-
transects chosen to represent each area. footed clapper rails) that were seen
outside the regular transect.

-Sampling methods and comparative data from natural wetlands

Table 8.1. Habitat types used in censusing birds at Tijuana Estuary (B. Kus, SDSU,
pers. comm.) .

Habitat Subhabitats

Beach Lower (surf line) and Kelp (wrack)
Dune (sandy beach in estuary) Distance from channel: within 3 m and >3 m
Unvegetated intertidal Inlet, Cobbles, Channel
Vegetated intertidal Distance from channel: 0-5 m, >5 m
Salt panne (bare flats, often dry)

Table 8.2. Summary data for birds censused along dime transects within Tijuana Estuary
in 1988-89. Data from Kus and Ashfield (SDSU, unpub. data). Data for biweekly low-
tide censuses are for the period of October 7, 1988, through April 8, 1989.

Group No. of % of individuals*

Small waders 14 37.31
Gulls and terns, 13 27.84
Large waders 8 18.24
Waterfowl 16 14.45
Herons and egrets 7 0.47
Raptors 9 0.39
Grebes and pelicans -5 0.14
Land birds 18 not included

Total number of species 90

Total, sightings in low tide censuses on 14 dates 25@572

Table 8.3. Percent of total observations in each of 5 habitats during low tide surveys on
14 dates of the northern half (the most well-flushed portion) of Tijuana Estuary. There
were 23,298 bird sightings within these groups in the two northern transects.

Small Gulls, Large Water- Herons Grebes,
Habitat waders tems waders fowl egmts; pelicans

Unvegetated intertidal 77.2 22.3 72.7 98.7 92.9 89.5
Vegetated intertidal 0.5 0.0 2.1 0.7 6.1 0.0
Beach (open coast) 1.3 5.4 2.6 0.0 .0.0 0.0
Dune (beach in estuary 14.8 6.0 17.9 0.6 1.0 0.0
Island 6.2 66.3 4.8 0.1 0.0 10.5

Total sightings 8,221 7,159 4,458 3,343 98 19

8 2

Sampling methods and comparative data from natural wetlands

Table 8.4. Species list for 7 groups of birds at Tijuana Estuary (Kus and Ashfield, unpub. data).
The list of land birds includes only species seen at the three census locations.

Small waders Gulls and terns
Actitis macularia Spotted sandpiper Larus phdadelphia Bonaparte's gull
Pluvialis squatarola Black-bellied plover L. heermanni Heermann's gull
Chara&ius vociferus KiUdeer L. delawarensis Ring-billed gull
C. sendpalmatus Semipalmated plover L. canus Mew gull
C. alexamWnus Snowy plover L. californicus California gull
Arenaria interpres Ruddy turnstone L. argentatus Herring gull
A. melanocephala Black turnstone L. occidentalis Western gull
Calidris canutus Red knot Sterna caspia Caspian tern
C. mauri Western sandpiper S. maxima Royal tern
C. minutilla Least sandpiper S. elegans Elegant tern
C. alpina Dunlin S. hirundo Common tern
C. alba Sanderling S. forsteri Forster's tern
Gallinago gallinago Common snipe Rynchops niger Black skimmer
Rallus longirostris levipes Light-footed clapper rail
Pelicans, grebes, and cormorants
Large waders PoXlymbuspoeiiceps Pied-billed grebe
Himantopus mexicanus Black-necked stilt PoiUceps nigricollis Eared grebe
RecurWrostra americana American avocet Aechmophorus occidentalis Western grebe
Catoptrophorus semipalmatus Willet Phalacrocorax auritus Double-crested
Tringa sp. Yellowlegs cormorant
Numenius phaeopus Whimbrel Pelecanus occidentalis Brown pelican
N. americanus Long-billed curlew
Limosafedoa Marbled godwit Herons and egrets
Limnodromus sp. Dowitchers Ardea herodias Great blue heron
Casmerodius albus Great egret
Waterfowl Egretta caendea Little blue heron
Branta canadensis Canada goose E. rufescens Reddish egret
Anas amencana American wigeon E. thula Snowy egret
A. strepera Gadwall Nycticorar nyclicorax Black-crowned
A. crecca Green-winged teal night heron
A. platyrhynchos Mallard Phoenicopterus sp. Flamingo
A. acuta Northern pintail
A. discors Blue-winged teal "Land birds"
A. cyanoptera Cinnamon teal Ceryle alcyon Belted kingfisher
A. clypeata Northern shoveler Sayorms nigricans Black phoebe
Aythya affinis Lesser scaup Sayornis saya Say's phoebe
Melanittaperspicillata Surf scoter Eremophila alpestris Horned lark
Bucephala clangula Common goldeneye Hirundo rustica Barn swallow
B. albeola Bufflehead Corvus corax Common raven
Mergus serrator Red-breasted merganser Lanius ludovicianus Loggerhead shrike
Oxyurajamaicensis Ruddy duck Cistothoruspalustris Marsh wren
Fulica americana American coot Thyromanes bewickii Bewick's wren
Chamaeafasciata Wrentit
Raptors Anthus spinoletta Water pipit
Asio flammeus Short-eared owl Carpodacus me.)dcanus House finch
Pandion haliaetus Osprey Geothlypis trichas Common yellow-
Elanus caeruleus Black-shouldered late throat
Circus cyaneus Northern harrier Passerculus sandiiich- Belding's Savannah
Accipiter cooperii Cooper's hawk ensis beldingi sparrow
Buteojamaicensis Red tailed hawk Melospiza melodiaa Song sparrow
Falco sparverius American kestrel Sturnella neglecta Western meadowlark
F. columbarius Merlin Columba livia Rock dove
F. peregrinus Peregrine falcon Mimus polyglottos Northern mockingbird

Sampling methods and comparative data from natural wetlands

Figure U. Temporal variability in numbers of sight ings for the most common groups
of birds using Tijuana Estuary (Kus and Ashfield, unpub. data).

IM
C
150O.J
CM Gulls and tems

1000- 0
Small waders
P

0
-J 500-
0 - Large waders
0 Waterfowl
W
04t 7 Jan I" Mar 10
E 200
1988-1989
Z

Sweetwater River Wetland Santa Margarita Estuary. Birds
Complex (SRWC). For her M.S. were censused biweekly at SME for one
thesis, Ashfield (in prep.) censused birds year, beginning April 1986 (Hollis et al.
along several permanent transects along 1988). During late winter/early spring of
San Diego Bay. Weekly sampling for 1987, the ocean inlet closed, and the
most of the year (biweekly in summer) reduced abundance and diversity of birds
provided sufficient data to compare bird was attributed to reduced tidal flushing.
use. during both high and low tides. Coots increased and ruddy ducks were
Some of her results have been higher than the previous year, while other
summarized and used in evaluating the water birds had reduced numbers.
bird community of the constructed marsh Shorebird use was characterized as
(Section IH); the relative abundances for especially low in recent years. Habitats
the two natural wedand sites are repeated sampled included willow woodland, river
here (Table 8.5), with rare species channel, marsh, ponds, salt flat, and
included. When completed, Ashfield's sewer ponds. Densities were lowest
thesis should be consulted for details of from May-July and highest from late fall
other sampling locations and seasonal to early spring.
patterns of abunda nce.

Sampling methods and comparative data from natural wetlands

Table 8.5. Water- as soci ated birds at Other bird counts. Local sources
SRWC. Data of Ashfield (in prep.) are of information on bird use may provide
for 13 months in 1989-90, for low-tide important historical records. The
censuses, for all species (excepting gulls) Audubon Christmas Bird Count data
that made up >0. 1 % of the sightings. PC from wetland sites should be obtained on
= Paradise Creek and BS = Bay an annual basis (published in special
Shoreline off Gunpowder Pt. editions of American Birds). In addition,
the Pt. Reyes Bird Observatory (4990
% of total sighting-S PC BS Shoreline Highway, Stinson Beach, CA
94970) has information on shorebirds for
Western sandpiper 15.5 40.6 selected coastal water bodies. A recent
Dowitcher spp. 25.0 11.7 effort of the Observatory was to count
Willet 20.0 7.2 shorebirds in all wetlands between Baja
Marbled godwit 4.9 8.5 California and British Columbia on a
Red knot 0 8.7 single date, April 22, 1989. This is a
Dunlin 5.7 1.7 very useful comparative data base. For
Bufflehead 6.6 0 San Diego County, the census resulted in
Killdeer 4.3 0.1 over 12,000 shorebird sightings, of
Great blue heron 4.0 0.1 which approximately 75% were small
Surf scoter 0 3.8 waders (over 50% were western
Lesser scaup 0.3 3.3 sandpipers). The largest concentration of
Western grebe 0 3.1 shorebirds was in south San Diego Bay,
Least sandpiper 2.1 0.7 with the next largest group in the Flood
Pied-billed grebe 2.3 <. 1 Control Channel of the San Diego River.
Snowy egret 1.2 1.0
Long-billed curlew 1.6 0.5 Endangered bird populations.
Brant 0 1.9 Clapper rails are censused during the
Great egret 1.7 0.1 breeding and nesting season by listening
Black-bellied plover 0.2 1.6 to their calls. Observers stand quietly at a
Black-necked stilt 1.4 0 vantage point. Calls are most frequent at
Spotted sandpiper 1.2 0 dusk, and surveys begin 0.5 hr before
Greater yellowlegs 0.5 0.6 dusk and last 0.5-1.0 hr after dark,
Forster's tern 0.7 0.3 depending on calling activity. Less
Northern pintail 0 0.9 calling leads to longer census periods.
Sernipalmated plover 0 0.8 Experts can often distinguish paired and
Ruddy turnstone 0 0.5 unpaired birds by their call: a "clatter"
Sanderling 0 0.4 call usually indicates a pair especially if
Mallard 0.3 0 the male and female call in a duet; a "kek"
American widgeon 0 0.3 call is made by a searching male or
Northern shoveler 0 0.3 female. Unpaired females searching for a
Black-crowned night-heron 0.2 0 mate is a third type of call.
Red-breasted merganser 0.2 0
American avocet 0.2 0 In addition, the presence of rails can
Least tern 0 0.2 be identified using recorded calls (during
Black skimmer 0 0.2 the territorial season), following the
Homed grebe 0 0.2 techniques developed by Barbara Massey
Brown pelican 0 0.1 (Endangered species consultant, Long
Green-winged teal 0 0.1 Beach), Dick Zembal (US FWS, Laguna
Snowy plover 0 0.1 Niguel), and Paul Jorgensen (Calif. Dept.
Whimbrel 0 0.1 of Parks and Recreation).

Total number of species 23 39

Sampling methods and comparative data from natural wetlands

The recovery plan for the light-footed Table 8.6. Historical data for light-
clapper rail (LFCR Recovery Team 1977, footed clapper rails in southern California
p. 15) states the objective of increasing (R. Zembal and B. Massey, US FWS,
the breeding population to "at least 400 unpub. report). Not all 35 wetlands were
pairs by preserving and restoring censused every year.
approximately 4,000 acres of wetland
habitat in at least 15 marshes..." Data for No. of marshes
several years (Table 8.6) show high Year No. of pairs with LFCR
variation in the rail population, with all
values well below the target of 400 pairs. 1980 203 11
1981 173 15
1982 221 18
1983 249 18
1984 277 19
1985 142 14
1987 178 ---

Figure 8.2. Importance of selected southern California wetlands to the region's
population of light-footed clapper rails. Data are percent of total birds censused, by year.

C 100%_
0

cis
:3 so%-
CL Carpinteria Marsh
0
6001.- 0 Anaheim Bay
E3 Mission Bay (K-F Res.)
0 Upper Newport Bay
40%-
Sweetwater R. Wetlands
0 20%- Tijuana Estuary

4)
0
0%_
04 CV) WIWI
CL OD CO CO CO CO CD CD
0) 0) 0) 0)

Sampling methods and comparative data from natural wetlands

Figure 8.3. Dynamics of the light-footed clapper rail populations at three wetlands in
southern California. The population at Tijuana Estuary crashed while the ocean inlet was
closed to tidal flow (April to December 1984).

50-

- Tijuana Estuary
40- - Sweetwater R. Wetlands

Mission Bay
LL 30-

0
2o-
W

10-

0
1978 1980 '1982 1984 1986 1988

California least terns are Table 8.7. Historical data for California
monitored for type and extent of habitat least tern breeding population (for state of
use (when present, April-Nov.) in both California), as summarized in the US
constructed and reference channel FWS Biological Opinion (Steucke 1988).
habitats. Feeding rates are recorded, and
type and quantity of food use are Census
estimated. The California Dept. of Fish
and Game should be consulted to advise Early 1970's 623-763
on censuses and to exchange information 1982 1015-1245
on the abundance of these birds 1983 1196-1321
elsewhere. 1984 887-997
1985 954-1084
The recovery plan for this endangered 1986 936-1016
species (CLT Recovery Team 1980, p.
13) states that it at least 1,200 pairs
distributed among colonies in at least 20
coastal wetland ecosystems throughout California least tern
their 1977 breeding range" would be (Sterna albifrons browni)
required to restore and maintain the
breeding population. Historical data
(Table 8.7) show that the state-wide
population fluctuates, with no clear trend
for an increasing density.

Sampling methods and comparative data from natural wetlands

Figure 8.4. ffistorical data for California least terns at selected sites in San Diego County
(E. Copper, ornithologist and least tern authority, unpub. data).

0 - Tijuana Estuary
Saltworks, San Diego Bay
120. 111- Sweetwater River
U) Chula Vista Wildlife Reserve
L.
rd 100-

80
0
d 60-
Z
E
3 40-
E
E 20

0
1960 1970 1980 1990
Year

Belding's Savannah sparrows to Santa Barbara County, sununarizing
are censused during their spring territorial censuses from 1973, 1977, and 1986.
nesting period. Territories are defended White (1986) gives additional data for
by singing males, which are best Los Pefiasquitos Lagoon and Tijuana
censused between 0600 and 0900 h Estuary.
(White 1986). Observers carry binocu-
lars, walk transects through the entire
marsh, and count singing males separ-
ately from non-singing birds (which Table 8.8. Comparative data for the
could be either males or females). The Belding's Savannah sparrow in southern
counts of singing males estimate the California (A. White, PERL, unpub.
number of breeding pairs. To carry out data). All censuses were in March 1989
detailed studies of habitat use, territory and used the same sampling methods.
sizes should be estimated and mapped Numbers are for singing males, which
according to the method of White (1986). indicate breeding pairs.
Estimated pairs
There are several sources of data
comparing populations of Belding's Tijuana Estuary 299-320
Savannah sparrow. A recent survey Sweetwater River Marsh
(Table 8.8) included 4 wetlands. Zembal Complex 160-183
et al. (in press) provided data for 31 Los Pefiasquitos Lagoon 56
coastal wetlands from San Diego County Ballona. Wetland 31

Sampling methods and comparative data from natural wetlands

9. Reptiles and sand or shredded newspaper to provide
cover for trapped animals and to reduce
amphibians fighting and predation among captives.
Each bucket had an oversized masonite
Objectives. Censuses are made to lid, which was propped open during
determine the species present in various times when traps could be checked. The
habitats. The functions performed by the pitfall traps were placed 14.4 rn apart,
herpetofauna are not well studied. with 15 buckets for each sampling site.
Species distributions and abundances are Drift fences were used in conjunction
also poorly known. with the pitfall traps, to help direct ani-
mals into the traps. Espinoza used alu-
Methods. Pitfall traps were used minum flashing, supported with wooden
by Hollis et al. (1988) to characterize stakes, and added sand at the base to act
reptiles and amphibians at Santa as a horizontal barrier. These "fences"
Margarita Estuary. These traps were "5- were 36 cm high and constructed in a
gallon buckets sunk in the ground with an widened "H" shape, with parallel legs of
elevated wood cover. Natural debris was 21 m and a 30.5-m cross link. Five traps
used to construct drift fences to direct the were buried along each 21 -m leg (the
animals toward the traps." (ibid., p. 45). middle trap also encountered the cross
No indication of the numbers of traps or link), and 5 along the cross-linking fence.
amount of searching was provided. Only Fencing passed over the middle of each
a few species were were found (Table bucket, so that animals moving along
9.1). either side of the fence would encounter
the trap. Espinoza also searched sites,
Similar methods were employed by looking under vegetation, debris, rocks
R. Espinoza (SDSU, unpub. data) in and other litter. Animals found were
1989, with sampling in ten sites ranging measured 'from snout to vent, and the
from wet to upland areas (Table 9.1). condition of their tails noted. Direct
His traps were 5-gal. plastic buckets with observations provide underestimates of
holes drilled in the bottom to insure nocturnal fauna, so are best used- in
drainage in case of rain. Buckets held conjunction with pitfall traps.

Table 9.1. Reference data for herpetofauna at Tijuana Estuary (1) and Santa Margarita
Estuary (SM), including willow woodland, coastal strand, maritime scrub and grassy habitats.

Order Farnily Common narn SW&ies SM i

Anura - Hylidae Pacific Uwfrog Hyla regilla x X
Toads, Ranidae Bullfrog Rana catesbiana X X
frogs Pipidae African clawed frog (exotic) Xenopus laevis X
]3ufonidae California toad Bufo boreas halopidlus X

Squarnata - Iguanidae Great Basin fence lizard S. occidentalis biseriatus X X
Lizards, Side-blotched lizard Uta stansburiana X X
snakes San Diego coast lizard Phrynosorna coronatum blainvillei X
Scincidae Coronado (Island) skink Eunwces skiltonianus interparietalis x
Anguidae San Diego alligator lizard Gerrhonotus multicarinatus webbi x x
Silvery legless lizard Anniella pulchra pulchra X
Colubridae San Diego gopher snake Pitouphis melanoleucus annectens x x
California king snake Lampropellis getulus californiae X X
Harnmond two-striped garter snake Thamnophis hammondi hamnwndi X
Viperidae Southern Pacific rattlesnake Crotalus viridus helleri X
Testudenes Emydidae Southwestern pond turde Clemmys rwrmorata palfida X
Turtles

Sampling methods and comparative data from natural wetlands

10. Mammals At Santa Margarita River Estuary
(Hollis et al. 1988), 30 trap lines were
Objective. Our knowledge of the used in eight habitat types (dense
role of mammals in coastal wetlands is Salicornia, Salicornia / Distichlis, Sali-
rudimentary. Part of their role in the food cornia / salt panne, UplandlSalicornia,
chain is feeding on plants and being coastal strand, maritime scrub, cattails,
preyed upon by raptors. Some seed and willow woodland. Each of the lines
dispersal is carried out by mammals. included 10 traps, which were maintained
Their burrowing activities disturb the soil for four consecutive evenings every 3-4
surface and open space for seedling months.
establishment (Cox and Zedler 1986). In Reference data. At Tijuana
constructed and restored wetlands, Estuary, there were 114 captures and 8
herbivores can decimate transplants, recaptures during the 7-month study
especially if the soil is not very wet or (Table 10.1). Only three rodent species
tidal inundation is lacking. Ground occurred in the marsh. These were the
squirrels and rabbits are probably the western harvest mouse (Reithrodontomys
main grazers on transplants, and a pre- megalods), the deer mouse (Peromyscus
planting survey of the project site would maniculatus), and the house mouse (Mus
help determine if fencing or herbivore musculus). Five additional species were
exclusion cages (Zedler 1984) will be trapped in Goat Canyon, an upland area.
necessary. Additional studies are needed The species list was extended with
to quantify mammal populations in evidence from sightings, scat, skeletons,
coastal wetlands and to determine their and pellets.
unique roles in the ecosystem.
Methods. Small mammals are At Santa Margarita River Estuary,
trapped using Sherman traps. A con- Hollis et al. (1988) and previous surveys
venient way to express the results is as recorded 22 species. Hollis et al. (ibid.)
found the highest number of species per
new captures per 100 trap nights. Mark- habitat (9) in the maritime scrub habitat,
recapture methods provide more accurate while the largest number trapped was in
estimates of population size than single the willow woodland (Table 10.2).
trappings. There were seasonal changes in numbers
At Tijuana Estuary, Taylor and trapped. Low numbers were associated
Tiszler (1989, unpub.) sampled small with inlet closure and marsh flooding; in
mammals in seven habitat types, of which 1987, high densities in Feb. and May
four were in the salt marsh (Salicornia were followed by low densities in
i August.
Salicornia / salt panne, salt panne, and
salt grass), and the remainder' in
peripheral habitats (border, upland,
upland transition). Each was sampled
monthly with 2 trapping grids. A grid
consisted of 10 Sherman traps located at
points 10 in apart in a 20-point grid.
Trapping was done from November
through May, for a total of 1440 traps.
Although animals were marked, the low
numbers of recaptures precluded density
estimates for any of the species.

Sampling methods and comparative data from natural wetlands

Table 10.1. Mammals known from Santa Margarita River Estuary (SM; from Hollis et
al. 1988) and Tijuana Estuary (TE; Taylor, SDSU, unpub. data).

Family Common namo Scientific name

Marsupiala Didelphiidae Opossum Didelphis marsupialis x x
Insectivora Soricidae Ornate shrew Sorex ornatus x
Lagomorpha Leporidae Blacktail jackrabbit Lepus californicus x x
Audubon cottontail Sylvilagus auduboni x x
Rodentia Sciuridae Calif. ground squirrel Spermophilus beecheyi x x
Geomyidae Botta pocket gopher Thomomys bottae x
Heteromyidae San Diego pocket mouse Perognathusfallax x x
Agile kangaroo rat Dipodomys agilis x
Cricetidae Western harvest mouse Reithrodontomys megalotis x x
Deer mouse Peromyscus maniculatus x x
Cactus mouse Peromyscus eremicus x
California mouse Peromyscus cal#'ornicus x
Brush mouse P. boy1ii . x x
Desert wood rat Neotoma lepida x
Dusky footed wood rat N. fuscipes x
California meadow mouse Microtus californicus x x
Muridae Norway rat Rattus norvegicus x
House mouse MUS musculus x x
Carnivora Procyonidae Raccoon Procyon lotor x
Mustilidae Long-tailed weasel Mustelafrenata x x
Striped skunk Mephitis mephitis x
Canidae Coyote Canis latrans x
Artiodactyla Cervidae Mule deer Odocoileus hemionus x

Table 10.2. Results of small mammal trapping at Santa Margarita River Estuary
(summarized from Hollis et al. 1988).

Uncommon s1mcies Ornate shrew Brush mouse
(averaged <5 new Botta pocket gopher Desert wood rat
captures per 100 trap Pocket gopher Dusky-footed wood rat
nights in any habitat): California mouse Audubon's cottontail
Common Wecies: Western harvest mouse (WHM)
Deer mouse (DM)
California meadow mouse (CMM)
House mouse (HM)

Mean new captures 12er 100 tW nights (by habitat Z=) for the 4 common Wecies:

Habitat WHM DM CMM HM

Dense Salicornia 13.9 4.6 11.9 7.8
SalicornialDistichlis 9.7 4.4 3.5 2.5
Salicornialsaltpan 1.6 12.6 0.3 0.2
UplandlSalicornia 9.5 9.4 1.6 4.1
Coastal strand 3.3 19.5 0 5.0
Maritime scrub 3.8 14.6 3.8 0.6
Cattails 4.3 18.6 7.7. 9.3
Willow woodland 2.3 19.7 4.0 6.2

Minimal Monitoring Program

needed; 2 = desirable; 3 = worthwhile).
V. Recommendations It should be noted that our priority
designations are tentative; as we come to
for minimum understand more about wetland
monitoring ecosystem functioning, we are more able
to select indicators of function. In the
future, it may become clear that some
We recognize that most assessment priority 3 variables are essential
and monitoring programs will be con- measurements, or vice versa.
strained by funding and by the availability
of personnel who are qualified to sample Sampling frequency. As dis-
such things as nitrogen fixation. Local cussed earlier, ecosystem attributes differ
resource agencies have requested that we in their temporal variability. Birds come
recommend minimal requirements for use and go daily, and large changes occur
in setting permit conditions for wedand with the fall and winter migratory
restoration or construction projects. periods. In contrast, plant invasions or
Since the main purpose of monitoring is local extinctions usually become obvious
to characterize the structure and only after a year or two. Thus, we
functioning of the metland, the sampling recommend that some attributes be
program should be able to withstand the measured as often as weekly, others
review of field ecologists. Thus, the seasonally or annually, and some only
program should identify the habitats after major events are noted.
being characterized; it should have
replicate sampling stations within each Not all wetlands will have the same
habitat; and it should provide data that temporal variability,, so it is difficult to
document ecologically meaningful suggest a single program that can fit all
changes when they occur. General systems. Soil salinities, for example,
analyses of the data should indicate that will* show greater extremes and more
the sampling program is encountering the sudden changes in lagoon wetlands (often
bulk of the species present, and that nontidal) than in fully tidal marshes.
variances among replicate sampling Monitoring programs should be tailored
stations are not excessively high. The to the needs of the system being
recent text on "Ecological Methodology"' monitored, beginning with frequent
(Krebs 1989) provides further discussion measurements and reducing sampling if
of these issues. experience suggests that reducing the
frequency will not'significantly reduce
Monitoring programs can be information about the system. Monitors
expanded or reduced in different ways, should be prepared to increase sampling
by varying the number of attributes exam- frequency in response to events such as
ined, the frequency of examination, and floods, wastewater spills, algal blooms,
the number of sampling stations. or inlet closure.
Additional cuts or additions could include
the detail of examination within sampling Numbers of sampling stations.
stations (e.g., depths at which soil Field monitoring programs should
salinity is measured) and back at the provide an adequate sample of the area to
laboratory (e.g., determination of inver- which results will be generalized.
tebrates to family or to species; chemical Experienced field ecologists can usually
analysis of pooled or individual soil walk through a site and delimit habitat
samples from each sampling station). areas that are "relatively homogeneous,"
but aerial photos are a great aid. Within
Priority attributes. The attributes each habitat area, replicate samples are
can be prioritized based on what we need taken at no fewer than three stations.
to know and how much information is Initial sampling will provide estimates of
provided by -the data (priority I = most heterogeneity (variance of each attribute

Minimal Monitoring Program

measured); if initial replicate stations give should the monitoring period include
high variance (e.g., if the standard error years of unusual events. Such was the
exceeds 10% of the mean), additional case at Tijuana Estuary, where salt marsh
replicate samples are needed to monitoring began in 1979, two major
characterize the attribute adequately. floods occurred in winter 1980,
Because the system's variability dictates prolonged flooding occurred through
the number of replicate samples needed, April 1983, the tidal inlet closed for 8
we cannot prescribe the number of months in 1984, and raw sewage inflows
sampling stations needed. Our advice is became substantial and continuous in
to plan for a large number of replicate about 1986.
stations and cut back if variances are low.
Krebs (1989) discusses the numbers of Long-term monitoring allows one to
replicate samples needed to provide distinguish directional changes (e.g.,
ecologically meaningful data. Results can expansion of cordgrass, declines of
be summarized to test for differences endangered bird populations) from short-
between different locations (e.g., restored term shifts (e.g., annual variability in
and natural wetlands) or differences with shorebird use). Permits for projects in
time (e.g., year-to-year changes). wetlands sometimes require 20 years of
monitoring (Phil Williams, Phil Williams
An alternative approach to replicate Assoc., San Francisco, pers. comm.) and
sampling within habitat areas is often require up to 10 years of
appropriate where gradients of assessment (with annual measures in the
environmental conditions are present. earlier years).
For estuarine channels that range from
high salinity at the inlet to low salinity Paul Zedler analyzed San Diego
inland, it is more useful to position rainfall data as a measure of the variability
sampling stations along the gradient and of vernal pool environments (Zedler and
to plot water quality characteristics Black 1989). For the 138 years of
against distance. Instead of clumping record, the range was 57 cm, which is
sampling stations within homogeneous about twice the average annual rainfall.
sampling areas, one would distribute the He calculated that monitoring would need
stations at intervals proceeding upstream to take place for 7-8 years just to include
from the ocean inlet. Stations should be half that range. Studies of 2-3 years
closer together where environmental would typically cover only 20-30% of the
changes are likely to be greatest. Results historic range. Along a similar vein,
can be summarized as graphs of each Westeby (in press) states, "Studies
attribute against distance from inlet, lasting 10-50 years do not provide
looking for spatial trends and evidence of enough replicate years to reliably detect
shifts through time. any but the strongest correlations between
variables at one site, nor do they estimate
How long to monitor. From the frequencies of rare events very well."
standpoint of the biota, a twenty-year
monitoring period is not unreasonable. It Conclusion. Choosing a sampling
may take even longer for the restored program that can provide ecologically
marsh to develop its full potential as meaningful data through a ten-to-twenty-
habitat for rare species, such as year period is not easy. Dozens of
endangered birds. It may take longer for decisions need to be made, and most
the soil organic matter to increase to require careful judgment based on
natural levels. It may take longer for preliminary data from the system in
herbivory problems to become controlled question. The assistance of experienced
by native predators. Finally, for a region field workers will be needed to tailor any
that has highly variable rainfall, it may 11 generic monitoring program" to the
take 20 years to characterize the mode, or system in question.
most usual condition of the wetland,

Minimal Monitoring Program

Table 4. 1. Priorities for wetland attributes to be monitored. Priority 1 = most needed.

Frequency of
Attribute and measures Priority measurement

Hydrology
Salinity of water 1 monthly
Salinity of interstitial soil water I seasonally (at least Apr. & Sept.)
Water levels at various fidal cycles 2 spring tide cycle, e.g., in Nov.
Tidal flow rates at distances frDrn inlet 3 spring tide cycle, e.g., in. Nov.
Topography
Elevation 1 initially and after storms or floods
Slope of channel banks 3 initially; annually thereafter at
permanent cross-sections
Soils
Texture 2 initially
Organic matter 1 pre-planting; plan for amendments
Toxic substances 3 pre-dredging; costly analysis
Redox potential 2 useful in diagnosing cause of
plant mortality
Sulfides and pH 3 redox potential is easier
Nutrient dynamics
Nitrogen fixation rates 3 seasonal, as research study
Inorganic nitrogen in sediments and pore water 1 initially to plan for amendments;
repeat if plant growth is poor
Litter decomposition 3 seasonal, as research study
Nitrogen mineralization rates 3 seasonal, as research study
Foliar nitrogen of dominant plant species 2 annually in September
Algae
Cover by dominant type 1 monthly with salinity samples
Vascular plants
Aerial photos of plant cover and habitat types I annually
Heights and total stem length of cordgrass 1 annually in September
Cover of vascular plants 1 annually in September
Patch size of rare annual plants 1 annually in spring
Density of annual plants 3 annually in spring
Consumers
Decomposers and shredders 3 seasonal, as research study
Aquatic insects 2 seasonally
Terrestrial insects, 3 seasonal, as research study
especially pollinators I important where annual plants are
required; census in spring
and predatory insects 1 important where insect
herbivory is obvious; census in
warm season (May-Oct.)
Fishes 1 at least in June and September
Benthic invertebrates I seasonally
Birds I weekly in fall-spring;
biweekly in summer
Reptiles and amphibians '3 summer
Mammals 3 (high priority in regions with
salt marsh harvest mouse)

Minimal Monitoring Program

Table 4.2. A minimum monitoring program for a wetland with aquatic and marsh habitats.

Before Annual monitoring schedule
Mj J F M A M J J A S 0 N D
Hydrology
Salinity of water x x x x x x x x x x x x
Salinity of interstitial
soil water x x x x
Topography
Elevation x and after storms or floods
Soils
Organic matter x repeat 2-3 months after any soil amendments
Nutrient. dynamics
Inorganic nitrogen in sediments
and pore water x repeat 2-3 months after any soil amendments
Algae
Cover by dominant type x x x x x x x x x x x x
Vascular plants x
Aerial photo analysis
Heights and total stem length
of cordgrass x
Cover of vascular plants x
Patch size of. rare annual plants x
Consumers
Fishes x x x x
Benthic invertebrates x x x x
Birds W WWWWB B B B WWW

X = sample once during month; x = sample to omit if funds are insufficient;
W = sample weekly through month; B = sample biweekly.

Literature cited

their possible relationships with the
Literature cited latitudinal distribution of shorebird
species. M.S. Thesis. San Diego
State University. San Diego. 78 p.
Adamus, P. R., and L. T. Stockwell Boland, J. 1988. The ecology of North
1983. A method for wetland func- American shorebirds: Latitudinal dis-
tional assessment, Vol. 1. Federal tributions, community structure, for-
Highway Administration Report No. aging behaviors, and interspecific
WWA-EP-82-23. U.S. Department competition. Ph.D. Dissertation.
of Transportation. Washington, D.C. University of California. Los
Angeles. 256 p.
Adamus, P. R., E. J. Clairain, Jr., R. D. Broome, S.W. 1987. Creation and
Smith, and R. E. Young. 1987. restoration of tidal wetlands of the
Wetland evaluation technique (WET); Southeastern United States. Pp. 37-
Vol. H; Methodology. Tech. Rep. Y- 72 in J. Kusler and M. E. Kentula,
87. Waterways Experiment Station, eds. Wetland creation and restora-
Corps of Engineers. Vicksburg, tion: The status of the science. Vol. 1:
Miss. Regional Reviews. EPA/600/3-89/
Agosta, K. 1985. The effect of tidally 038a, Environmental Research
induced changes in the creekbank Laboratory. Corvallis, Oregon.
water table on pore water chemistry- Broome, S., W. Seneca, and W.
Estuarine Coastal Shelf Sci. 21: 389- Woodhouse, Jr. 1986. Long-term
400. growth and development of trans-
Allison, F. E. 1973. Soil organic matter plants of the salt marsh grass,
and its role in crop production. Sparlina alterniflora. Estuaries 9:63-
Elsevier. New York. 637 p. 74.
APHA. 1985. Standard methods for the Broome, S. W., C. B. Craft, and E. D.
examination of water and wastewater Seneca- ' 1987. Creation and devel-
(16th edition). American Public opment of brackish-water marsh
habitat. Pp. 197-205 in J. Zelazny
Health Association, American Water
Works Association, Water Pollution and J. S. Feierabend, eds. Increasing
Control Federation. Washington our wetland resources, Corporate
D.C. 1268 p. Conservation Council Proceedings.
National Wildlife Federation.
Avnimelech, Y; M. Yamamoto and R.G. Washington, D.C.
Menzel. 1983. Evaluating the release Cantilli, J. 1989 Sulfide phytotoxicity
of soluble components from sedi-
ment. J. Environ. Qual. 12: 86-91. in tidal salt marshes. M.S. Thesis.
San Diego State University. San
Binkley, D and P. Vitousek. 1989. Soil Diego. 115 p.
nutrient availability. Pp. 75-96 in R. Cantilli, J., K. Swift, and J. B. Zedler.
W. Pearcy, J. Ehleringer, H. A. 1989. Free sulfide in natural and
Mooney and P.W. Rundel, eds. man-made salt marshes. Poster pre-
Chapman Hall. New York. 457 p. sentation, Jan. 1989 Annual Meeting,
Boland, J. 1981. Seasonal abundances, Society of Ecological Restoration and
habitat utilization, feeding strategies Management. Oakland, Calif.
and interspecific competition within a Carpelan, L. H. 1969. Physical charac-
wintering shorebird community and teristics of southern California coastal

Literature cited

lagoons. Pp. 319-334 in A. Ayala University of California. Santa
Castahares and F. B. Phleger, eds. Barbara. 300 p.
Lagunas Costeras, Un Simposio.
Mem. Simp. Intern. Lagunas Cos- Florsheirn, J., and P. B. Williams.
teras UNAM-UNESCO. UNAM. 1990. Geomorphic and hydrologic
Mexico, D. F. changes to the Tijuana Estuary: 1986-
1990. Philip Williams& Assoc. San
Casselman, M. E., W. H., Patrick, Jr., Francisco. 14 p.
and R. D. DeLaune. 1981. Nitrogen
fixation in a Gulf Coast salt marsh. Fong, P. 1986. Monitoring and manipu-
Soil Sci. Soc. Am. J. 45:51-56. lation of phytoplankton dynamics in a
southern California estuary. M.S.
CLT Recovery Team. 1980. California Thesis. San Diego State University.
least tern recovery plan. U. S. Fish San Diego. 106 p.
and Wildlife Service, Endangered
Species Program, Region 1. Fong, P., R. Rudnicki, and J. B. Zedler.
Portland, Oregon. 58 p. 1987. Algal community response to
nitrogen and phosphorus loadings in
Ciupek, R. B. 1986. Protecting wet- experimental mesocosms: Manage-
lands under the Clean Water Act 404: ment recommendations for southern
EPA's Conservative Policy on California lagoons. San Diego
Mitigation. National Wetlands News- Association of Governments. San
letter 8:12-14. Diego. 102 p.

Covin, J. 1984. The role of inorganic Gardner, W.H. 1986. Water Content.
nitrogen in the growth and distribu- Pp.493-544 in Methods of soil anal-
tion of.Spartina foliosa at Tijuana ysis, Part 1: Physical and mineralogi-
Estuary, California. M.S. Thesis. cal methods (2nd edition). Am Soc.
San Diego State University. San of Agronomy--Soil Sci, Soc. of
Diego. 60 p. America. Madison, Wisconsin.

Covin, J., and J. B. Zedler. 1988. Gee, G. W., and J. W. Bauder. 1986.
Nitrogen effects on Spartina foliosa Particle-size analysis. Pp. 399-409 in
and Salicornia virginica in the salt Methods of soil analysis, Part 1:
marsh at Tijuana Estuary, California. Physical and mineralogical methods
Wetlands 8:51-65. (2nd edition). Am. Soc. of
Agronomy-Soil Sci. Soc. of America.
Eilers, P. 1981. Production in coastal Madison, Wisconsin.
salt marshes of southern California.
Grant No. R805438-01-1, Environ- Golet, F. C. 1986. Critical Issues in
mental Research Laboratory, U.S. wetland mitigation: A scientific per-
Environmental Protection Agency. spective. National Wetlands News-
Corvallis, Oregon. 88 p. letter 8:3-6.

Eno, C.F. 1960. Nitrate production in Good, J. W. 1987. Mitigation estuarine
the field by incubating the soil in development impacts in the Pacific
polyethylene bags. Soil Sci. Soc. Northwest: From concept to practice.
Am. Proc. 24:277-279. The Northwest Environmental Jour-
nal 3:93-112.
Ferren, W. R., Jr. 1985. Carpinteria
Salt Marsh: Environment, history, Gordon, A.M., M. Tallas and K. Van
and botanical resources of a southern Cleve. 1987. Soil incubations in
California estuary. Publ. No. 4., The polyethylene bags: effect of bag
Herbarium, Dept. of Biol. Sciences. thickness and temperature on nitrogen

Literature cited

transformations and C02 Permeabil- and restoration: The status of the
ity. Canadian Journal of Soil Science science. Vol. 1: Regional Reviews.
67:65-75. EPA/600/3-89/038a, Environmental
Research Laboratory. Corvallis,
Grubb, P. J. 1977. The maintenance of Oregon..
species richness in plant communities:
The importance of regeneration niche. Krebs, C. J. 1989. Ecological method-
Biological Review 52:107-145. ology. Harper & Row. New York.

Gwin, S. E., and M. E. Kentula., 1990. Kusler, J. A., and M. E. Kentula, eds.
Evaluating design and verifying 1989. Wetland creation and restora-
compliance of wetlands created under tion: The status of the science.
Section 404 of the Clean Water Act in Volume 1: Regional Reviews. Vol-
Oregon. EPA/600/3-90/061. Envi- ume II: Perspectives. EPA/600/3-
ronmental Research Laboratory. 89/038a,b. Environmental Research
Corvallis, Oregon. Laboratory. Corvallis, Oregon.

Hanowski, J. M., G. Niernei, and J. G. Langis, R., M. Zalejko, and J. B. Zedler.
Blake. 1990. Statistical perspectives In press. Nitrogen assessments in a
and experimental design when constructed and a natural salt marsh
counting birds on line transects. The of San Diego Bay, California.
Condor 92:326-335. Ecological Applications (a new jour-
nal of the Ecological Society of
Hardy, R.W.F., R.D. Holsten, E.K. America).
Jackson and R.C. Bums. 1968. The
acetylene - ethylene assay for N2 Langis, Ren6, and J. B. Zedler. 1989.
fixation: Laboratory and field evalua- Some aspects of nutrient dynamics in
tion. Plant Physiol. 43:1185-1207. natural vs. man-made salt marshes.
Poster presentation, Jan. 1989
Hollis, H., D. Zembal, K. Foerster, M. Annual Meeting. Society - of
Kenney, R. Bransfield, B. Harper, Ecological Restoration and Man-
L. Hays, and N. Gilbert. 1988. agement. Oakland, California.
Preliminary results of the resources
inventory at the Santa Margarita River Levings, C. D. 1976. Basket traps for
Estuary. U.S. Fish and Wildlife surveys of a gammarid amphipod,
Service. Laguna Niguel, California. Anisogammarus confervicolus
128 p. (Stimpson), at two British Columbia
estuaries. J. Fish. Res. Board Can.
Holmberg, N. D., and R. Misso. 1986. 33:2066-2069.
Mitigation: Determining the need.
National Wetlands Newsletter 8:10- LFCR Recovery Team. 1977. Light-
11. footed clapper rail recovery plan.. U.
S. Fish and Wildlife Service,
Hosmer, S. C. 1977. Pelecypod- Endangered Species Program, Region
sediment relationships at the Tijuana 1. Portland, Oregon. 33 p.
Estuary. M.S. Thesis. San Diego
State University. San Diego. 119 p. Lincoln, P. G. 1985. Pollinator effec-
tiveness and ecology of seed set in
Josselyn, M., Zedler, J., and T. Cordylanthus maritimus subsp.
Griswold. 1989. Wetland mitigation maritimus at Point Mugu, California.
along the Pacific Coast of the United Final Report to U.S. Fish and
States. Pp. 1-35 -in J. Kusler and M. Wildlife Service Endangered Species
E. Kentula, eds. Wetland creation Office, Order No. 10181-9750.
Sacramento, California. 31 p.

Literature cited

Lindau, C. W., and L. R. Hossner. National Wetland Technical Council.
1981. Substrate characterization of 1985. Pacific regional wetland func-
an experimental marsh and dime natu- tions. Proceedings of a workshop
ral marshes. Soil Science Society of held at Mill Valley, April 14-16,
America Journal 45:1171-1176. 1985. University of Massachusetts.
Amherst.
Linton, T. L. 1969. Physical and biolog-
ical problems of impounding salt Nelson, D. W. and L. E. Sommers.
marshes. Pp. 451-455 in A. Ayala 1982. Total carbon, organic carbon,
Castaflares and F. B. Phleger, eds. and organic matter. Pp. 539-579 in
Lagunas Costeras, Un Simposio. A. L. Page, R. H. Miller, and D. R.
Mem. Simp. Intern. Lagunas Keeney, eds. Methods of Soil
Costeras. UNAM-UNESCO. Analysis, Part 2. Chemical and
Mexico, D. F. microbiological properties- - Agron-
omy Monograph no. 9 (2nd edition).
Lloyd, F. T. and W. H. McKee, Jr. American Society of Agronomy Inc.,
1983. Replications and subsamples Soil Science Society of America.
needed to show treatment responses Madison, Wisconsin.
on forest soils of the coastal plain.
Soil. Science Society of America Neuenschwander, L. F., T. H. Thorsted,
Journal: 47:587-590. Jr., and R. J. Vogl. 1979. The salt
marsh and transitional vegetation of
Macdonald, K. B. 1977. Coastal salt Bahia de San Quintin. Bull. Southern
marsh. Pp. 263-469 in M. G. California Academy of Science
Barbour and J. Major, eds. 78:163-182.
Terrestrial vegetation of California.
John Wiley and Sons. New York-. Nordby, C. S. 1982. The comparative
ecology of ichthyoplankton within
Macdonald, K. B. 1988. Part 8. Coastal Tijuana Estuary and in adjacent
salt marsh. Pp. 1008-1010 in M. G. nearshore waters. M.S. Thesis. San
Barbour and J. Major, eds. Diego State University. San Diego.
Terrestrial vegetation of California, 101 P.
2nd ed. John Wiley and Sons.
New York. Nordby, C. S., and J. B. Zedler. In
press. Responses of fishes and ben-
Macdonald, K. B., and M. G. Barbour. thos to hydrologic disturbances in
1974. Beach and salt marsh vegeta- Tijuana Estuary and Los Pefiasquitos
tion of the North American Pacific Lagoon, California. Estuaries.
Coast. Pp. 175-233 in R. J. Reimold
and W. H. Queen, eds. Ecology of Odum, W. E. 1987. Predicting
Halophytes. AcademicPress. New Ecosystem Development Following
York. Creation and Restoration of
4 Wetlands. Pp. 67-70 in J. Zelazny
Manion, B. S., and J. H. Dillingham. and J. S. Feierabend, eds. Increasing
1989. Malibu Lagoon: A baseline our wetland resources, National
ecological survey. Topanga-Las Wildlife Federation Corporate
Virgenes Resource Conservation Conservation Council Proceedings.
District. Topanga, California. 180 p. Washington, D.C.

Mitsch, W. J., and J. G. Gosselink. Odum, W. E. 1988. Comparative ecol-
11986. Wetlands. Van Nostrand ogy of tidal freshwater and salt
Reinhold Co., New York. 539 p. marshes. 1988. Annual Review of
Ecology and Systematics 19:147-176.

Literature cited

Onuf, C. P. 1987. The ecology of Peterson, C. H. 1975. Stability of
Mugu Lagoon, California: An estuar- species and of community for the
ine profile. U.S. Fish Wildl. Serv. benthos of two lagoons. Ecology
Biol. Rep. 85(7.15). 122 p. 56:958-965.

Onuf, C. P., M. L. Quarnmen, G. P. Peterson, C. H. 1977. Competitive
Shaffer, C. H. Peterson, J. W. organization of the soft-bottom mac-
Chapman, J. Cormak, and R. W. robenthic communities of southern
Holmes. 1978. An analysis of the California lagoons. Mar. Biol. 43:
values of central and southern 343-359.
California coastal wetlands. Pp. 186-
199 in P.E. Greeson, J. E. Clark, Quarnmen, M. L. 1984. Predation by
and J. E. Clark, eds. Wetland shorebirds, fish and crabs on inverte-
functions and values: The state of our brates in intertidal mudflats: An
understanding. American Water experimental test. Ecology 65:529-
Resources Association. Minneapolis, 537.
Minnesota.
Quarnmen, M. L. 1986. Measuring the
Onuf, C. P., and. M. L. Quammen. success of wetlands mitigation.
1985. Coastal and riparian wetlands National Wetlands Newsletter 8:6-8.
of the Pacific region: The state of
knowledge about the food chain sup- Ray, D., L. Locklin, L. Strand, M.
port. Pp. 98-173 in Proceedings of a Cappelli, J. Leslie, P. Webb, S.
workshop held at Mill Valley, April Hatfield, J. Raives, J. McGrath, and
14-16, 1985. University of Mass., J. Bower. 1986. Draft report of the
Amherst. wetlands task force. California
Coastal Commission. San Francisco.
Park, R., M. Trehan, P. Mausel, R. 28 p.
Howe. 1989. The effects of sea
level rise on U.S. coastal wetlands. Ralph, C. J., and J. M. Scott, eds.
Pp. 1-52 in J. Smith and D. Tirpak. 1981. Estimating numbers of terres-
The potential effects of global climate trial birds. Studies in avian biology
change on the United States: Appen- No. 6. Cooper Ornithological Soci-
dix B - Sea level rise. EPA-230-05- ety. Allen Press. Lawrence, Kansas.
89-052, Office of Policy, Planning I
and Evaluation. U.S. Environmental Richards, L. A., ed. 1954. Diagnosis
Protection Agency. Washington, and improvement of saline and alkali
D.C. soils. Agric. Handbook No. 60,
U.S. Dept. Agriculture. Washington,
Pastor, J., J. D. Aber, C. A. D.C. 1.60 p.
McClaugherty and J. M. Melillo.
1984. Aboveground production and Rudnicki, R. 1986. Dynamics of
N and P cycling along a nitrogen macroalgae in Tijuana Estuary:
mineralization gradient on Blackhawk Response to simulated wastewater
Island, Wisconsin. Ecology 65:256- addition. M.S. Thesis. San Diego
268. State University. San Diego. 82 p.

Peterson, C. H. 1972. Species diver- Rutherford, S. 1989. Detritus produc,
sity, disturbance and time in the tion and epibenthic communities of
bivalve communities of some natural versus constructed salt
California lagoons. Ph.D. Thesis. marshes. M.S. Thesis. San Diego
University of California, Santa State University. San Diego. 79 p.
Barbara. 230 p.

100

Literature cited

example. M.S. Thesis. San Diego
Rutherford, S., and J. B. Zedler. 1989. State University. San Diego. 84 p.
Natural vs. man-made salt marshes:
Invertebrate distributions. Poster Taylor, E. W., and J. Tiszler. 1989.
presentation, Jan. 1989 Annual The mammals of Tijuana Estuary.
Meeting, Society of Ecological Report to Tijuana Estuary Tidal
Restoration and Management. Restoration and Enhancement Project.
Oakland, California. San Diego State University. San
4 SF BCDC (San Francisco Bay Diego. Unpublished.
Conservation and Development Thompson, G. Jr. 1988. Sierra Club, et
Commission). 1988. Mitigation: An al., Plaintiffs, v. John 0. Marsh,
analysis of tideland restoration pro- Secretary of the Army, et al.,
jects in San Francisco Bay. March Defendants. Civil No. 86-1942-
1988 Staff Report. 74 p. GT(IEG) Memorandum Decision and
Order. U.S. District Court. San
Schreiber, R. W. 1981. The biota of the Diego.
Ballona region, Los Angeles County.
Los Angeles County Natural History US ACE (U.S. Army Corps of
Museum Foundation. Los Angeles. Engineers). 1982. Sweetwater River
Flood Control Channel, State
Shisler, J. K., and D. J. Charette. 1984. Highway Route 54, Interstate
Evaluation of artificial salt marshes in Highway Route 5, Recreation
New Jersey. P-40502-01-84. New Facilities, and Conservation of
Jersey Agric. Exp. Station. 160 p. Marshlands, San,?@_ Diego County,
California. Environmental Impact
Sims, J.R., and V.A. Haby. 1971. Statement. District Office, U.S.
Simplified colorimetric determination Army Corps of Engineers. Los
of soil organic matter. Soil Sci. 112: Angeles.
137-141.
US FWS (U.S. Fish and Wildlife
SMBB Recovery Team. 1981. Draft Service). 1988. Biological Opinion
recovery plan for the salt-marsh 1-1-78-F-14-R2 - The combined
bird's beak (Cordylanthus maritimus Sweetwater River Flood Control and
Nuttall ssp. maritimus): An endan- Highway Project, San Diego County,
gered plant. U. S. Fish and Wildlife California. Letter by Wally Steucke,
Service, Sacramento Endangered Acting Regional Director, Portland,
Species Office. Sacramento, Cali- Oregon, to Col. T. Ono, District
fortlia. 59 p. plus appendices. Engineer, dated March 30, 1988.
District Office, U.S. Army Corps of
Strickland, R., ed. 1986. Wetland func- Engineers. Los Angeles.
tions, rehabilitation, and creation in
the Pacific Northwest. Washington Vogl, R. J. 1966. Salt-marsh vegetation
State Department of Ecology. of upper Newport Bay, California.
Olympia, Washington. Ecology 47:80-87.

Strickland, J. D. H., and T. R. Parsons. Weller, M. W. 1989. Waterfowl man-
1972. A practical handbook of sea- agement techniques for wetland
water analysis. Fisheries Research enhancement, restoration and creation
Board of Canada. Ottawa. 3 10 p. useful in mitigation procedures. Pp.
105-116 in J. Kusler and M. Kentula,
Swift, K. 1988. Saltmarsh restoration: eds. Wetland creation and restora-
Assessing a southern California tion: 'Me status of the science. Vol.
II: Perspectives. EPA/600/3-89/038a,

101

Literature cited

Environmental Research Laboratory. T'hesis. San Diego State University.
Corvallis, Oregon. San Diego. 71 p.

Westeby, Mark. In press. On long term Zalejko, M., K. Swift, and J. B. Zedler.
ecological research in Australia. 1989. Potential nitrogen inputs (N-
Proc. NSF/MAP Workshop on long- fixation) in natural and man-made salt
term ecological research. Berchtes- marshes. Poster presentation,
gaden, Germany. 26 p. Society of Ecological Restoration and
Management, Annual Meeting.
White, A. 1986. Effects of habitat type Oakland, Calif. Jan. 1989.
and human disturbance on an endan-
gered wetland bird: Belding's Zedler, J. B. 1977. Salt marsh com-
Savannah sparrow. M.S. Thesis. munity structure in the Tijuana
San Diego State University. San Estuary, Calif. Estuarine and Coastal
Diego. 73 p. Marine Science 5:39-53.

Williams, K., M. Busnardo, D. Gibson, Zedler, J. B. 1980. Algal mat produc-
K. Johnson, and S. Snover. 1989. tivity: comparisons in a salt marsh.
Terrestrial arthropods of Tijuana Estuaries 3:122-131.
Estuary. Report to Tijuana Estuary
Tidal Restoration and Enhancement Zedler, J. B. 1982a. Salt marsh algal
Project. San Diego State University. mat composition: spatial and temporal
San Diego. Unpublished. comparisons. Bull. Southern
California Acad. Sci. 81:41-50.
Williams, P. and M. Swanson. 1987.
Tijuana Estuary Enhancement: Zedler, J. B. 1982b. The ecology of
Hydrologic analysis. Final Report to southern California coastal salt
California State Coastal Conservancy, marshes: a community profile. U.S.
Oakland, California. 50 p. plus Fish and Wildlife Service, Biological
technical appendices. Services Program. Washington, D.
C. FWS/OBS-31/54. 110 p.
Winfield, T. P. 1980. Dynamics of car-
bon and nitrogen in a southern Zedler, J. B. 1983. Freshwater impacts
California salt marsh. Ph. D. in normally hypersaline marshes.
Dissertation, Univ. of Calfornia, Estuaries 6:346-355.
Riverside, and San Diego State
University. San Diego. 76 p. Zedler, J. B. 1984. Salt Marsh
Restoration: a Guidebook for South-
Wolaver, T. G., J. Zieman, and B. ern California. T-CSGCP-009.
Kjerfvei 1986. Factors affecting Calif Sea Grant College Program,
short-term variability in sediment pH Institute of Marine Resources. La
as a function of marsh elevation in a Jolla, California. 46 p.
Virginia mesohaline marsh. J. Exp.
Mar. Biol. Ecol. 101: 227-237. Zedler, J. B. 1988a. Restoring diversity
in salt marshes: Can we do it? Pp.
Yeomans, J. C., and J. M. Bremner. 317-325 in E. 0. Wilson, ed.
1988. A rapid and precise method for Biodiversity. National Academy
routine determination of organic Press. Washington, D. C.
carbon in soil. Commun. in Soil Sci.
Plant Anal. 19:1467-1476. Zedler, J. B. 1988b. Salt Marsh
Restoration: Lessons from California.
Zaleiko, M. K. 1989. 1989. Nitrogen Pp. 123-138 in J. Cairns, Jr., ed.
fixation in a natural and a constructed Rehabilitating Damaged Ecosystems.
southern California salt marsh. M.S. CRC Press. Boca Raton, Florida.

102

Literature Cited

Zedler, J. B., and P. Beare. 1986.
Temporal variability of salt marsh
vegetation: the role of low-salinity
gaps and environmental stress. Pp.
295-306 in D. Wolfe, ed. Estuarine
variability. Academic Press. New
York.

Zedler, J. B., J. D. Covin, C. Nordby,
P. Williams, and J. Boland. 1986.
Catastrophic events reveal the
dynamic nature of salt-marsh vegeta-
tion in southern California. Estuaries
9:75-80.

Zedler, J. B., and C. S. Nordby. 1986.
The Ecology of Tijuana Estuary,
California: An Estuarine Profile.
U.S. Fish and Wildlife Service
Biological Report 85 (7.5). 104 p.

Zedler, J. B., R. Langis, I Cantilli, M.
Zalejko, K. Swift, and S.
Rutherford. 1988. Assessing the
functions of mitigation marshes in
southern California. Pp. 323-330 in
J. A. Kusler, S. Dalky, and G.
Brooks, eds. Proceedings of the
National Wetland Symposium: Urban
Wetlands. Association of State
Wetland Managers. Byrne, N.Y.

Zedler, J. B., R. Langis, J. Cantilli, M.
Zalejko, and S. Rutherford. 1990.
Assessing the successful functioning
of constructed salt marshes. Pp. 311-
318 in H. G. Hughes and T. M.
Bonnicksen, eds. Proceedings of the
1 st Annual Conference 1989. Society
of Ecological Restoration and
Management. Madison, Wisconsin.

Zedler, J. B., T. P. Winfield, and P.
Williams. 1980. Salt marsh produc-
tivity with natural and altered tidal cir-
culation. Oecologia (Berl.) 44:236-
240. -

103

Additional References

Additional Day, John W., Jr., C. Hall, W. Kemp,
references and A. Ydfiez-Arancibia. Estuarine
ecology. John Wiley & Sons. New
York. 558 p.

Dept. of Fish and Game. 1989. 1988
Allen, L. 1982. Seasonal abundance, annual report on the status of
composition, and productivity of the California's state listed threatened and
littoral fish assemblages in Upper endangered plants and animals. The
Newport Bay, California. Fishery Resources Agency, Dept. of Fish and
Bull. 80:769-790. Game. Sacramento, California.
Allen, L., and M. Horn. 1975. Ferren, W., Jr., M. H.- Capelli, Anujha
Abundance, diversity and seasonality Parikh, D. L. Magney, K. Clark, and
of fishes in Colorado Lagoon, J. R. Hallar. 1990. Botanical
Alamitos Bay, California. Estuarine resources at Emma Wood State Beach
and Coastal Marine Sci. 3:371-380. and the Ventura River Estuary,
California: Inventory and manage-
Baker, J. M., and W. J. Wolff, eds. ment. Publ. No. 15, The Herbarium,
1987. Biological surveys of estuaries Dept. of Biol. Sciences. University
and coasts. Cambridge University of California. Santa Barbara. 310 p.
Press. Cambridge.
Fink, B. H., and J. B. Zedler. 1990.
Barnes, R. S. K., ed. 1977. The Endangered plant recovery: Experi-
coastline. JohnWiley&Sons. New mental approaches with Cordylanthus
York. 356 p. maritimus ssp. maritimus. Pp. 460-
468 in H. G. Hughes and T. M.
Beare, P. A., and J. B. Zedler. 1987 Bonnicksen, eds. Proceedings of the
Cattail invasion and persistence in a Ist Annual Conference 1989. Society
coastal salt marsh: the role of salinity. of Ecological Restoration and
Estuaries 10:165-170. Management. Madison, Wisconsin.
Buresh, R. J., M. E. Casselman, and Gersberg, R., F. Trindade, and C.
W. H. Patrick, Jr. 1980. Nitrogen Nordby. 1989. Heavy metals in
fixation in flooded soil systems, a sediments and fish of the Tijuana
review. Adv. Agron. 33: 149-192. Estuary. Border Health V:5-15.
Cairns, J., Jr., ed. 1988. Rehabilitating Hook, D., W. McKee, Jr., H. Smith, J.
damaged ecosystems. Vol. I-II. Gregory, V. Burrell, Jr., M. DeVoe,
CRC Press. Boca Raton, Florida. R. Sojka, S. Gilbert, R. Banks, L.
192 p., 222 p. Stolzy, C. Brooks, T. Matthews, and
T. Shear, eds. 1988. The ecology
Cox, G. W., and J. B. Zedler. 1986. and management of wetlands, Vol. I-
The influence of mima mounds on 11. Timber Press. Portland, Oregon.
vegetation patterns in the Tijuana 592 p., 394 p.
Estuary salt marsh, San Diego
County, California. Bull. Southern Jordan, W. R., Ill., M. E. Gilpin, and J.
California Acad. Sci. 85: 158-172. D. Aber. 1987. Restoration ecology.
Cambridge University Press.
Daiber, F. C. 1982. Animals of the tidal Cambridge. 342 p.
marsh. Van Nostrand Reinhold.
New York. 422 p. Josselyn, M. 1983. The ecology of San
Francisco Bay tidal marshes: A

104

Additional References

community profile. U.S. Fish Wildl.
Serv. Rep. No. FWS/OBS-83/23. Zedler, J. B., and G. W. Cox. 1984.
Washington, D.C. 102 p. Characterizing wetland boundaries: a
Pacific Coast example. Wetlands
Josselyn, M., ed. 1982. Wetland 4:43-55.
Restoration and Enhancement in
California. Proceedings of a 1982 Zedler, J. B., R. Koenigs, and W.
Workshop; Report No. T-CSGCP- Magdych. 1984. Freshwater release
007. California Sea Grant College and southern California coastal wet-
Program. La Jolla, California. 110 p. lands: Report 1, Strearnflow for the
San Diego and Tijuana Rivers;
Macdonald, K. B. 1977. Plant and Report 2, Review of salinity effects
animal communities of Pacific North and predictions of estuarine responses
American salt marshes. Pp. 167-191 to lowered salinity; Report 3,
in V. J. Chapman, ed. Wet coastal Management plan for the beneficial
ecosystems. Elsevier Scientific. use of treated wastewater in the
New York. Tijuana River and San Diego River
estuaries. San Diego Assoc. of
Neuenschwander, L., T. Thorsted, Jr., Governments. San Diego.
and R. Vogl. 1979. The salt marsh
and transitional vegetation of Bahia de Zedler, J. B., and C. P. Onuf. 1984.
San Quintin. Bull. So. Calif. Acad. Biological and physical filtering in
Sci. 78:163-182. arid-region estuaries: seasonality,
extreme events, and effects of water-
Niering, W. A. 1985. Wetlands. The shed modification. Pp. 415-432 in
Audubon Nature Series Guides. V. Kennedy, ed. The estuary as a
Alfred A. Knopf. New York. 638 p. filter. Academic Press. New York.

Pomeroy, L., and R. Wiegert. 198 1.
The ecology of a salt marsh.
Springer-Verlag. NewYork. 271p.

Purer,Edith. 1942. Plant ecology of the
coastal salt marshlands of San Diego
County. Ecol. Monogr. 12:82-111.

Sinienstad, C. A., C. D. Tanner, and R.
M. Thom. 1989. Estuarine wetland
restoration monitoring protocol.
Final Draft Report to the Region 10
Office of the U. S. Environmental
Protection Agency. Seattle. 191 p.

Stevenson, R. E., and K. 0. Emery.
1958. Marshlands at Newport Bay.
Paper No. 20, Alan Hancock
Foundation Publications. University
of Southern California Press. Los
Angeles.

Valiela, 1. 1984. Marine ecological
processes. Springer-Verlag. New
York. 546 p.

105

3 6668 00001 9325 ,