-------
Chapter V - Epiphyte Contributions to Light Attenuation at the Leaf Surface 69
Maryland Chesapeake Bay
and Tributaries
10 20 30 40 50 60
Virginia Chesapeake Bay
and Tributaries
10 20 30 40 50
Calculated PAR at SAV Canopy
(PLW, %l )
60
FIGURE V-12. Percent Light at the Leaf vs. Percent
Light Through the Water Column, by State.
Comparing values for percent surface light (PAR) at SAV
leaf surface (PLL) and percent surface light in water just
above SAV leaf (PLW) for all monitored sites in the main-
stem, tidal tributaries and embayments of Chesapeake
Bay during 1985-1996 grouped into upper (Maryland)
and lower (Virginia) estuary regions. Values of PLL and
PLW were calculated for water depth of 1 m using the
model described in this report (Table V-1) with input
monitoring data (total suspended solids, dissolved
inorganic nitrogen, dissolved inorganic phosphorus,
KCJ) for the SAV growing season of each year. Lines
indicate position of points where epiphyte attenuation
reduced ambient light levels at the leaf surface by 0,
25, 50 and 75 percent.
CONCLUSIONS
The model developed in this chapter to calculate con-
tributions of water-column and epiphytic materials to
light attenuation under different water quality condi-
tions works well for sites throughout Chesapeake Bay,
including its tidal tributaries across all salinity regimes.
Values for PLL calculated from water quality data vary
widely among sites throughout the Bay. The model
relies on a combination of empirical relationships
derived from field studies and experimental systems
and numerical computations from a well-calibrated
ecosystem process model. Much of the information on
which the model is based comes from the measure-
ments and analyses done in the mesohaline and poly-
haline regions of Chesapeake Bay; particularly studies
of two SAV species—Potamogeton perfoliatus and Z.
marina (e.g., Staver 1984; Twilley et al. 1985; Golds-
borough and Kemp 1988; Neckles 1990; Moore 1996;
Sturgis and Murray 1997). This is due to limited com-
parable data from lower salinity tidal habitats.
The model is easily used and is amenable to simple
spreadsheet computations on diverse platforms. It has
substantial utility as a screening tool to assess trends in
SAV habitat conditions at individual sites, based on
changes in water quality variables. In ecosystems such
as Chesapeake Bay, where a broad monitoring pro-
gram exists to support efforts to improve water quality
for restoring SAV to degraded habitats, this model
provides an additional important tool to guide man-
agement efforts.
-------
CHAPTER V|
Beyond Light: Physical, Geological
and Chemical Habitat Requirements
Light availability has been identified as the major
factor controlling the distribution and abundance
of SAV in Chesapeake Bay (e.g., Dennison et al. 1993).
Therefore, parameters that can affect the light avail-
ability in an environment (total suspended solids,
chlorophyll a concentration, epiphyte biomass) are
commonly included in predictions of the suitability of
certain areas for SAV growth. Several other parame-
ters that have the potential to override the light
requirements of the plants are not often considered
when determining the suitability of a site for SAV
growth (Livingston et al. 1998). For example, very high
wave energy may prevent SAV from becoming estab-
lished (due to the drag exerted on the plants and/or
the constant sediment motion), even when the light
requirements are met (Clarke 1987).
This chapter discusses physical, geological and chemi-
cal factors that affect the suitability of a site for SAV
growth. These factors differ from those described in
chapters III, IV and V in that the parameters consid-
ered there modify the light requirements of SAV. The
parameters discussed in the present chapter override
the established SAV light requirements. The parame-
ters addressed here (waves, currents, tides, sediment
organic content, grain size and contaminants) can
influence the presence/absence of SAV in a certain
area, independently of light levels. Figure VI-1 shows
how previously established SAV habitat requirements
(light attenuation coefficient, dissolved inorganic
nutrients, chlorophyll a, total suspended solids and
epiphytes) as well as the parameters discussed in this
chapter (waves, currents, tides, sediment characteris-
tics and chemical contaminants) can affect the distri-
bution of SAV.
FEEDBACK BETWEEN SAV AND THE
PHYSICAL, GEOLOGICAL AND CHEMICAL
ENVIRONMENTS
SAV beds can reduce current velocity (Fonseca et al.
1982; Fonseca and Fisher 1986; Gambi et al. 1990;
Koch and Gust 1999; Sand-Jensen and Mebus 1996;
Rybicki et al. 1997), attenuate waves (Fonseca and
Cahalan 1992; Koch 1996), change the sediment char-
acteristics (Scoffin 1970; Wanless 1981; Almasi et al.
1987; Wigand et al 1997) and even change the height
of the water column (Powell and Schaffner 1991;
Rybicki et al. 1997). In turn, these SAV-induced
changes can affect the productivity of the plants.
Therefore, a complex feedback mechanism exists
between SAV and the abiotic conditions of the habitat
they colonize, making it difficult to attribute the distri-
bution of SAV to only one factor, such as light.
By reducing current velocity and attenuating waves,
SAV beds create conditions that lead to the deposition
of small (inorganic) and low-density (organic) parti-
cles within meadows or canopies (Grady 1981; Kemp
et al. 1984; Newell et al. 1986). This in turn can affect
the availability of light (Moore et al 1994), nutrients
(Kenworthy et al. 1982) and compounds that can be
toxic to SAV (phytotoxins), such as sulfide in the sedi-
ments (Carlson et al. 1994; Holmer and Nielsen 1997).
Therefore, all these parameters contribute, to some
degree, to the regulation of SAV growth.
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 71
-------
72 SAV TECHNICAL SYNTHESIS II
Light
Transmission
Waves
Plankton
Chlorophyll a
Total
Suspended
Solids
Tides
Settling
of
Organic
Matter
Biogeo-
chemical
processes
FIGURE VI-1. Interaction between Light-Based, Physical, Geological and Chemical SAV Habitat
Requirements. Interaction between previously established SAV habitat requirements, such
as light attenuation, dissolved inorganic nitrogen (DIN), dissolved inorganic phosphorus
(DIP), chlorophyll a, total suspended solids (TSS) and other physical/chemical parameters
discussed in this chapter (waves, currents, tides, sediment organic matter, biogeochemical
processes). P = phosphorus; N = nitrogen; PLW = percent light through water; PLL =
percent light at the leaf.
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 73
When plant density is low, the attenuation of current
velocity and wave energy is also low. This results in lit-
tle accumulation of organic matter and, subsequently,
little change in sediment nutrient and phytotoxin con-
centrations. Light availability may also be low, due to
the resuspension of sediments. Oxygen demand of the
roots (to counteract the detrimental effect of the phy-
totoxins) may also be low, due to reduced photosyn-
thetic biomass.
In SAV beds with high shoot density, water flow is
reduced and more particles are deposited, leading to
higher light, nutrient and phytotoxin availability than
in less dense beds. The SAV density may reach a point
where so much organic matter is trapped that the
resulting high phytotoxin levels are no longer tolerated
by the plants, and they may start dying back (Roblee et
al. 1991; Carlson et al. 1994; Holmer and Nielsen
1997). At that point, the reduction in density may lead
to higher water flow, reduced organic matter ac-
cumulation and reduced phytotoxin levels in the sedi-
ment. This feedback mechanism (hydrodynamics —>
sediment characteristics —* plant biomass —> hydrody-
namics) may assure the health of marine and higher
salinity estuarine SAV populations over time. The
above mechanism may be less applicable for SAV col-
onizing low-salinity estuarine areas, because sulfide
concentrations do not reach levels as toxic as those in
marine environments. However, low-salinity species
can also be susceptible to sulfide, as shown by the
decreased growth rates of Potamogeton pectinatus
when sulfide was added to the sediments (van Wijk
etal. 1992).
SAV AND CURRENT VELOCITY
SAV beds reduce current velocity by extracting
momentum from the moving water (Madsen and
Warnke 1983). The magnitude of this process depends
on the density of the SAV bed (Carter et al. 1991; van
Keulen 1997), the hydrodynamic conditions of the
area (stronger reduction in flow in tide-dominated vs.
wave-dominated areas; Koch and Gust 1999) and the
depth of the water column above the plants (Fonseca
and Fisher 1986). The highest reduction in current
velocity occurs in dense, shallow beds exposed to tide-
dominated conditions (unidirectional flow). Currents
in SAV beds can be 2 to 10 times slower than in adja-
cent unvegetated areas (Ackerman 1983; Madsen and
Warncke 1983; Carter et al. 1988; Gambi et al. 1990;
Rybicki et al. 1997).
Positive Effects of Reduced Current Velocity
The advantages of reduced water flow in SAV beds
include the following:
1) Reduced self-shading due to the more vertical
position of the blades in beds resulting from
reduced drag on SAV leaves (Fonseca et al 1982);
2) Increased settlement of organic and inorganic par-
ticles, increasing the light availability and the sedi-
ment nutrient concentration (Kenworthy et al.
1982; Ward et al. 1984; references in the review by
Fonseca 1996). Notice that this can also lead to a
disadvantage due to increasing sulfide concentra-
tion in marine/higher salinity estuarine habitats
(Koch 1999) (see "Negative Effects of Reduced
Current Velocity").
3) Lower friction velocities at the sediment surface
than in unvegetated areas (Fonseca and Fisher
1986) reducing sediment resuspension and total
suspended solids concentrations and increasing
light availability (references in the review by
Fonseca 1996).
4) High residence time, allowing molecules of dis-
solved nutrients to stay in contact with SAV leaves
and epiphytes for longer periods of time, therefore
increasing the likelihood of being taken up. As a
result, high residence time reduces the nutrient
concentration in the water column (Bulthuis et al.
1984), perhaps limiting epiphytic growth, which
would otherwise lead to further reduction in light
attenuation (see Chapter V). Epiphytes and SAV
will compete for nutrients in the water column,
while only SAV can remove nutrients from the sed-
iments. Therefore, the SAV should not become as
nutrient-limited as epiphytes since they primarily
use nutrients from the sediments.
5) Increased settlement of spores of algae and larvae
of a variety of organisms, resulting in higher
species diversity of invertebrates and algae in SAV
canopies than in adjacent unvegetated areas
(Homziak et al. 1982).
-------
74 SAV TECHNICAL SYNTHESIS II
Negative Effects of Reduced Current Velocity
The points listed above illustrate the potential positive
effects of reduced current velocity in SAV beds. Alter-
nately, reduced water flow can also have detrimental
effects:
1) Concentration of phytotoxins will increase in estu-
arine/marine sediments (Koch 1999). The concen-
tration of phytotoxins in the sediment leads to an
increased oxygen demand by the roots which, if not
met due to poor light availability, has the potential
to kill the plants (Robblee et al. 1991; Carlson et al.
1994; Nepf and Koch 1999).
2) Thicker blade diffusion boundary layers will form
under reduced current velocity in SAV beds (Koch
1994). The diffusion boundary layer is a thin (a few
/am) layer of water on the surface of any submersed
object (including plants) where the transport of
solutes (e.g. carbon needed for photosynthesis or
oxygen produced by photosynthesis) is dominated
by viscous forces (i.e., by diffusion).
Increases in the thickness of this diffusion boundary
layer lead to a longer diffusional path (or thick diffu-
sion boundary layer) for carbon molecules to move
from the water column to the SAV leaf, where they are
used for photosynthesis. As the current velocity
decreases, a critical maximum diffusion boundary
layer thickness, where the flux of carbon to the plant is
slower than the flux needed for the plant to support
maximum photosynthesis, can be reached. The critical
diffusion boundary layer thickness was estimated to be
280 jim for Thalassia testudinum and 98 /urn for
Cymodocea nodosa (Koch 1994).
If a plant is exposed for long periods of time to diffu-
sion boundary layer thicknesses greater than the criti-
cal diffusion boundary layer thickness, growth can
decline, due to carbon limitation independent of the
light levels at the site. The length of time that a plant
can survive under such conditions depends on the
internal carbon reserves in the plant tissue and how
fast these reserves can be accessed (Koch 1993). This
has not yet been determined for most SAV species and
has the potential to be important in areas where mari-
nas and other structures may cause stagnant condi-
tions in SAV habitats.
Some estuarine and freshwater SAV species, such as
Potamogeton pectinatus, are capable of colonizing rela-
tively stagnant waters (like those found in ponds) due
to a physiological adaptation: the release of H+ on one
side of the blade (polar leaves) reduces the pH in the
diffusion boundary layer (Prins et al. 1982). This
decrease in pH shifts the carbon balance toward car-
bon dioxide (CO2), increasing local diffusion boundary
layer CO2 concentration and, therefore, increasing the
flux of CO2 into the plant. Other SAV can also incor-
porate CO2 from sediment porewater, where dissolved
inorganic carbon concentrations are usually much
higher than open-water concentrations (Sondergaard
and Sand-Jensen 1979; Madsen 1987). The CO2 incor-
porated by the roots is then transported to the photo-
synthetic tissue via the lacunae system (Madsen and
Sand-Jensen 1991). For a detailed discussion on mech-
anisms aquatic plants developed to deal with reduced
carbon fluxes due to thick diffusion boundary layers,
see Jumars et al. (accepted).
Epiphytes and Current Velocity
Although epiphytes are usually seen as organisms that
are detrimental to SAV growth, the very early stages of
epiphytic colonization on SAV leaves have the poten-
tial to be beneficial for the plants (Koch 1994). Very
low densities of epiphytes (only visible under a micro-
scope) may disrupt the diffusion boundary layer
enhancing the flux of carbon to the blade (Koch 1994).
As epiphytes compete for light, nutrients and carbon,
later stages of epiphytic colonization (when the epi-
phytic layer is too dense to disrupt the diffusion
boundary layer) become detrimental to SAV growth.
At the community level, epiphytes will contribute to
the reduction in current velocity (due to increased leaf
drag) which leads to the positive aspects listed above
(see "Positive Effects of Reduced Current Velocity").
Therefore, in some aspects, epiphytes can have posi-
tive effects on SAV communities, although the nega-
tive effects of epiphytes on SAV leaves (light
attenuation and increased drag on the leaves, poten-
tially dislodging them at high flows) also need to be
kept in perspective.
Epiphytes on SAV leaves may also respond to water
flow independently of the SAV response to flow. In a
study using acrylic plates, maximum periphyton bio-
mass was observed at intermediate current velocities.
Diatoms were dominant under high current conditions
while a green alga was dominant at lower current
velocities (Horner et al. 1990). In a mesocosm experi-
ment, epiphyte biomass on Vallisneria americana
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 75
increased with current velocity (Merrell 1996). There-
fore, a second order of complexity (water flow) needs
be added to future refinements of the model evaluat-
ing the effect of epiphytes on light available to SAV
leaves (see Chapter V).
Current Velocity SAV Habitat Requirements
From the positive and negative effects of the reduced
current velocities found in SAV beds, it can be con-
cluded that these plants could benefit from intermedi-
ate current velocities (Boeger 1992; Koch and Gust
1999; Merrell 1996; Koch 1999). Extremely low water
flows could increase the blade diffusion boundary
layer thickness as well as the accumulation of organic
matter in the sediment leading to carbon starvation or
death due to high phytotoxin concentrations in the
sediment, respectively. In contrast, extremely high
water flow has the potential to 1) increase drag above
a critical value where erosion of the sediment and
plants occurs, 2) reduce light availability through
resuspension of sediment and self-shading and 3)
decrease the accumulation of organic matter, leading
to reduced nutrient concentration in the sediments.
A literature review revealed that 1) the range of cur-
rent velocities tolerated by marine SAV species lies
between approximately 5 and 100 cm s"1 (physiological
and mechanical limits, respectively); 2) the range of
current velocities tolerated by freshwater SAV species
seems generally to be lower than that for the marine
species; and 3) some freshwater SAV species can tol-
erate extremely low current velocities (Table VI-1).
This may be due to alternative mechanisms of carbon
acquisition present in these freshwater plants but not
in marine plants (see "Negative Effects of Reduced
Current Velocity").
Survival of SAV in high current velocity environments
may be possible if the development of seedlings
occurred under conditions of slow current velocity in
space (e.g., a protected cove) or time (e.g., a low water
discharge period). Once a bed is established under
such conditions, it can expand into adjacent areas with
higher currents due to the reduced currents at the
edge of the bed or persist during times of higher water
flow. Therefore, the stage of the plants (for example,
seeds, seedlings, vegetative shoots, reproductive
shoots) also needs to be taken into account when con-
sidering if current velocity is above or below the estab-
lished requirement for growth and distribution. Based
on the literature review presented here, no data are
available on the current velocity requirements of
plants other than those found in well-established beds.
In summary, intermediate current velocities between
10 and 100 cm s"1 are needed to support the growth
and distribution of healthy marine SAV beds. These
requirements are lower for freshwater/estuarine SAV
species—between 1 and 50 cm s"1—especially for those
with polar leaves. If currents are above or below these
critical levels, the feedback mechanisms in the system
may become imbalanced and possibly lead to the
decline or even complete loss of the vegetation.
Although some of the feedback mechanisms between
SAV beds and current velocity involve light availability
through the effects of resuspension of sediments, self-
shading and epiphytic growth, extreme currents alone
can limit the growth of SAV. Therefore, current
velocity should be considered as a key SAV habitat
requirement.
SAV AND WAVES
As waves propagate over SAV beds, wave energy is lost
(Fonseca and Cahalan 1992; Koch 1996). This is due to
the same mechanism that causes SAV beds to reduce
current velocities-loss of momentum (Kobayashi et al.
1993). The efficiency with which waves are attenuated
by SAV beds depends on the water depth (Ward et al.
1984; Mork 1996), the current velocity (Stewart et al.
1997), leaf length (Fonseca and Cahalan 1992) and the
type of vegetation (canopy or meadow) (Elwany et al.
1995; Mork 1996; Stewart et al 1997).
Wave attenuation is strongest in dense SAV beds due
to increased drag and in meadows (where most of the
biomass is found close to the sediment surface) colo-
nizing shallow waters, where plant biomass takes up a
large portion of the water column. Canopy-forming
species that have long stems and concentrate most of
their biomass, and consequently drag, on the water
surface of a relatively deep water body have the ten-
dency to oscillate with the waves. Acting as though
imbedded in the waves, canopy-forming species
impose little drag on them (Seymour 1996) and, there-
fore, have little effect on wave attenuation.
The effect the constant motion waves imposes on
plants may lead to the breakage of the plants
(Idestam-Almquist and Kautsky 1995; Stewart et al.
-------
76 SAV TECHNICAL SYNTHESIS II
TABLE VI-1. Minimum and maximum current velocities required for SAV growth and distribution.
Minimum current velocities required to saturate photosynthesis
Current
> 0.04 cm s"1
> 0.08 cm s'1
> 0.5 cm s"1
> 3 cm s"1
> 5 cm s"1
> 5 cm s"1
> 13 cms"1
> 16 cm s"1
Current
< 7cms"1
< 45 cm s"1
< 50 cm s"1
< 50 cm s"1
< 120 cm s"1
< 150 cms"1
< 180 cms'1
Species
Potamogeton pectinatus *
Callitriche stagnalis
Ranunculus pseudofluitans
Zostera marina
Ranunculus pencillatus
Thalassia testudinum
Cymodocea nodosa
Z. marina
Source
Madsen and Sondergaard
1983
Westlake 1967
Westlake 1967
Koehl and Worcester 1991
Werner and Wise 1982
Koch 1994
Koch 1994
Fonseca and Kenworthy 1987
Maximum currents at which the following species occur
Species Source
Vallisneria americana
Ranunculus pencillatus
Zannichellia palustris
Z. marina
Z. marina
Z. marina
Z. marina
Merrell 1996
Werner and Wise 1982
Sculthorpe 1967
Conover 1964
Scoffin 1970
Fonseca etal. 1982
Phillips 1974
* Indicates species for which leaf polarity has been confirmed.
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 77
1997). This effect has been observed to be more severe
for a canopy-forming species (Myriophyllum spp.) than
for a meadow-forming species (Vallisneria spp.; Stew-
art et al. 1997). Breakage of underwater plants exposed
to waves is inversely related to current velocity. As cur-
rent velocity increases, the plants lie closer to the sed-
iment surface and are, therefore, less affected by the
orbital motion of the waves (Stewart et al. 1997).
Table VI-2 summarizes the capacity of marine SAV
species to attenuate waves under a variety of field and
laboratory conditions. The values obtained in the lab
are much higher than those obtained in the field,
because the meadow-forming plants used in the lab
experiments occupied the entire water column,
and wave attenuation is positively correlated with the
percentage of the water column occupied by the
vegetation.
Effects of High Wave Energy
The impact of high wave energy on SAV can be direct
or indirect. The direct impact of waves on SAV can be
seen when waves (in combination with currents) erode
the edges of an SAV bed (Clarke 1987) or when por-
tions of the plants are removed by storm-generated
(Thomas et al. 1961; Eleuterius and Miller 1976;
Rodriguez et al. 1994; Dan et al. 1998) or boat-
generated waves (Stewart et al. 1997). Indirect conse-
quences of wave energy in SAV beds include sediment
resuspension, changes in sediment grain size, mixing
of the water column and epiphytic growth. If the
capacity of an SAV bed to attenuate waves is reduced,
for example, due to a reduction in shoot density
because of clam dredging or eutrophication, the
underlying sediment will become more vulnerable to
erosion, and higher concentrations of suspended
TABLE VI-2. Attenuation of wave energy in meadow-forming marine SAV beds.
Attenuation of Wave
Wave Energy Period
(seconds)
Site Species
Comments Source
1.6%
7.7 %
43%
43%
44%
44%
0.35
0.35
0.4 and 0.7
0.4 and 0.7
0.4 and 0.7
0.4 and 0.7
Field
Field
Flume*
Flume*
Flume*
Flume*
Thalassia
Thalassia
Zostera
Syringodium
Halodule
Thalassia
Within
SAV bed
Edge of
SAV bed
1 minto
SAV bed
1 minto
SAV bed
1 minto
SAV bed
1 minto
Koch 1996
Koch 1996
Fonseca and Cahalan
1992
Fonseca and Cahalan
1992
Fonseca and Cahalan
1992
Fonseca and Cahalan
SAV bed 1992
Plants (meadow-formers) occupied the entire water column.
-------
78 SAV TECHNICAL SYNTHESIS II
sediment particles can be expected in the water. This
is especially true for SAV beds in which fine particles
have accumulated over time. These sediments may be
resuspended at lower wave energy than the coarser
sediments outside the SAV bed (Posey et al. 1993).
Wave attenuation and sediment resuspension in vege-
tated areas depend on the water levels above the
plants. At low tide, wave energy is reduced to a greater
extent than during high tide (Ward et al. 1984). Resus-
pension of fine particles will alter the grain size distri-
bution of the sediment. In areas of high wave
exposure, sediments are coarser, which leads to lower
nutrient concentration in the sediment and, conse-
quently, lower root biomass (Idestam-Almquist and
Kautsky 1995). By contrast, the above-ground biomass
of Potamogeton pectinatus depends directly on wave
exposure; shoots are shorter in areas with high wave
exposure than in areas with low wave exposure
(Idestam-Almquist and Kautsky 1995).
In Chesapeake Bay, shore erosion (caused by wave
action) contributes 13 percent of the total suspended
matter in the upper Bay and 52 percent in the middle
Bay (Biggs 1970). Perhaps, before the decline of SAV
in this area, SAV protected the coastlines from the
direct impact of waves. Ward et al. (1984) observed
that in shallow (< 2 meters) unvegetated areas in the
Choptank River, total suspended solids concentrations
increased tenfold when the wind came from the direc-
tion of highest fetch and was >25 km h"1, but the
increased total suspended solids concentrations dissi-
pated within 24 hours after the storm. As the wind
intensifies, wave period and wave length increase lead-
ing to deeper wave mixing depths (Chambers 1987).
The small grain size sediments are the first to become
resuspended. Therefore, if wave energy increases in an
SAV bed, a shift toward coarser sediments will occur.
The consequences of this shift are addressed below, in
"SAV and the Sediments It Colonizes."
SAV growth and distribution seems to be limited by
high, but not low wave energy (Dan et al. 1998; Table
VI-3). However, high wave exposure can also benefit
the plants by reducing the epiphytic biomass (Strand
and Weisner 1996; Kendrick and Burt 1997; Weisner et
al. 1997). In high wave exposure areas, where sedi-
ments are constantly being shifted and grain size may
be skewed toward coarser particles, SAV may not be
able to become established due to the balance between
the anchoring capacity of the roots and the drag
exerted on the leaves. High wave exposure also leads
TABLE VI-3. Quantitative and qualitative descriptions of wave tolerance for Chesapeake Bay species.
Species
Canopy formers
Myriophyllum spicatwn
Wave Tolerance
Source
Wave limited
Rawls (1975), Stewart et al 1997
Zannichellia palustris
Wave limited
Flowering structures ofRuppia Wave sensitive
maritima
Meadow formers
Zoster a marina
Potamogeton pectinatus
Vallisneria americana
2 m waves
Wave tolerant
Stevenson and Confer 1978
Joanen and Glasgow 1965
Dan et al. 1998
Hannan 1967
More wave tolerant Stewart et al. 1997
than Myriophyllum
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 79
to reduced light availability due to increased sediment
resuspension, but this may be compensated by the
lower epiphytic biomass on the leaves of wave-exposed
plants. The mechanism that allows for reduced epi-
phytic biomass on plants exposed to high wave energy
is not well understood. It could be due to the rubbing
of the blades against each other.
Wave Mixing Depth Effects on
SAV Minimum Depth Distributions
The minimum depth of distribution of aquatic plants
in lakes with good water quality has been correlated to
the resuspension of sediments by waves resulting in
the scouring of sediments, uprooting of plants and
increased turbidity (Figure VI-2). Consequently,
Chambers (1987) suggested that the minimum depth
distribution (Zmin) of aquatic plants can be determined
by the wave mixing depth (Zwave), which extends to a
depth equal to half the wavelength (L),
Z =Z =-
mm wave o
(vi-iy
L can be calculated from the wave period (T) using the
following equation:
L = 4^ (VI-2)
where g is acceleration of gravity (9.805 m s"2). Equa-
tion VI-2 is a standard equation for waves propagating
over depths larger than half the wavelength (i.e.,
before they "feel the bottom" and define Zmin; see
Equation VI-1). The wave period (T) for these waves
can be predicted according to the following equation:
T =
.0.46W.
g
1.28
(VI-3)
where W is the wind velocity (m s"1) and F is the effec-
tive fetch (m). These equations allow for the predic-
tion of the wave mixing depth (Zwave) in shallow
SAV habitats.
In relatively exposed areas at the mouth of Chesa-
peake Bay (Timble Shoal Entrance and Timble Shoal
Light), wave periods (T) range between 4 and 13 sec-
onds (Boon et al. 1996). These values are typical for
wind- generated waves. Marine SAV can colonize such
areas, as seen in the Thalassia testudinum beds off-
shore of the Florida Keys (Koch 1996), the Cymodocea
nodosa beds in the Mediterranean (Koch 1994) and
the Posidonia oceanica beds in many areas in Aus-
tralia. In the shallower, more protected SAV habitats
in Chesapeake Bay, waves with smaller periods
(ripples) can be expected. Figure VI-3a shows the
wave mixing depth for ripples and wind-generated
waves while Figure VI-3b shows more details for rip-
ples, typical for shallow, semi-enclosed SAV habitats.
Low wave energy
High wave energy
Wave length
Wave mixing dent
*««*••* W**»»»«n*>«
SAV
FIGURE VI-2. Wave Energy Effects on SAV Vertical Depth Distributions. The vertical distribution of SAV beds
can be shifted into deeper waters due to wave energy. Waves can constantly shift sediments preventing the
colonization of the area or resuspend sediments, contributing to increased total suspended solid (TSS)
concentrations, which leads to reduced light levels. The zone where waves do not allow SAV to colonize is
defined as the depth equivalent to half the wavelength.
-------
80 SAV TECHNICAL SYNTHESIS II
4 6
Wave period (s)
10
B
0.2
0.4 0.6
Wave period (s)
FIGURE VI-3. Wave Period/Depth of Wave Mixing
Relationship. Depth of wave mixing (or minimum
depth of distribution of SAV) under a variety of wave
conditions. Wind-generated waves are typical for
relatively exposed areas (A) while ripples are typical for
shallow coastal areas. SAV occur in habitats character-
ized by both wave types. Depth of wave mixing for
waves periods more characteristic of SAV habitats in
Chesapeake Bay are illustrated in B.
Wave Exposure Habitat Requirements
Wave mixing depth and tides have a potentially con-
founding effect on the minimum depth of SAV distri-
bution. As will be discussed below (see "SAV and
Tides"), due to the lack of resistance of most SAV
species to dessication, tides tend to force SAV to colo-
nize deeper waters, where exposure to the air is less
likely. If waves also force the SAV to inhabit deeper
waters due to sediment resuspension in areas
shallower than the wave mixing depth (Figure VI-2),
then the minimum depth of distribution of SAV beds
should be determined by the mean low water (tide)
plus the wave mixing depth (see "SAV and Tides").
This can be visualized by imagining the water level in
Figure VI-2 fluctuating vertically with the tides as the
waves continue to propagate onshore. No data are
available to verify this hypothesis.
Although waves have the potential to force SAV to
colonize deeper areas (Figure VI-2) and tides may
enhance this minimum limit of colonization, no 'wave
exposure habitat requirement' can yet be established
for SAV in Chesapeake Bay. Wave exposure indexes
(based on fetch and wind intensity) have been sug-
gested for lake environments (Keddy 1982) and
recently also for an estuarine area (Murphey and Fon-
seca 1995; Fonseca and Bell 1998), but further
research is needed to understand the effect of waves
on the ecology and distribution of SAV beds. As high
wave exposure has the potential to override other light
requirements (Kd and epiphyte biomass) or to com-
pletely eliminate SAV from high wave exposure areas,
it should be addressed as a SAV habitat requirement in
the near future. Fonseca and Bell (1998) were able to
determine the maximum wave exposure tolerated by
SAV in Pamlico Sound, North Carolina, but when the
same methodology was applied to Chesapeake Bay, no
conclusive results could be obtained (Chiscano, in
preparation). This difference could be due to the
higher fetches in Chesapeake Bay than in Pamlico
Sound, the presence of sand bars offshore from SAV
beds (the model used does not take into account
bottom bathymetry) or the erosion of the marshes
which changes the sediment characteristics in SAV
habitats, potentially limiting the distribution of SAV
(see "SAV and the Sediments It Colonizes").
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 81
SAV AND TURBULENCE
Turbulence consists of temporally and spatially irregu-
lar water motion superimposed on the larger flow pat-
tern. It forms at such boundaries as the sediment
surface or the surface of SAV leaves. It is then trans-
ferred from larger to smaller scales (eddy sizes). In
SAV beds, the distance between SAV shoots deter-
mines the size of the turbulence scale/eddies (Ander-
son and Charters 1982; Nowel and Jumars 1984;
Ackerman and Okubo 1993). Turbulence in these
plant communities can be generated and rescaled, i.e.,
shifting the scale of the eddies formed in a SAV bed
(Anderson and Charters 1982; Gambi et al 1990; Ack-
erman and Okubo 1993; Koch 1996). Since mass trans-
fer of nutrients and carbon in the water takes place by
eddy diffusion in turbulent flows (Sanford 1997), tur-
bulence has the potential to be of extreme ecological
importance in SAV beds. Turbulence also may affect
the dispersion of particles such as pollen, larvae, seeds
and spores in the beds. Unfortunately, the effect of
turbulence on these plants is poorly understood.
The observations of turbulence in SAV beds may seem
contradictory. A region of high turbulence levels can
be observed at the canopy-water interface (Gambi
etal. 1990). Increased turbulence within the vegetation
has also been observed during a monami, high-
amplitude blade waving (Grizzle et al. 1996). By con-
trast, reduced turbulent mixing also during a monami
in a Zostera marina bed has been reported by Acker-
man and Okubo (1993). Since turbulence depends on
the current velocity and the structure of the SAV bed
(Koch and Gust 1999), at low current velocities the
turbulence levels are expected to be low. As the cur-
rent velocity increases, turbulence levels also increase.
At the point where the vegetation begins to bend over
due to the current velocity, the water flow is redirected
over the vegetation and turbulence levels among the
plants may decrease again (Nepf et al. 1997).
Since mass transport of nutrients and carbon in SAV
beds depends on turbulence levels, it can be predicted
that SAV can benefit from turbulence in the water.
The optimal turbulence levels for SAV is yet unknown.
What is known is that SAV beds rescale turbulent
energy from larger to smaller eddies. This process
depends on the architecture of the SAV bed (Koch
1996; Koch and Gust 1999). Epiphytes colonizing SAV
blades decrease the distance between "obstructions to
the flow" (like blades and shoots) and rescale turbu-
lence to even smaller eddies than those found among
blades without epiphytic growth (Koch 1994,1996).
Rescaling of turbulence occurs at the individual plant
level (Anderson and Charters 1982) and at the canopy
level (Gambi et al. 1990; Koch 1996) and may be a
mechanism for creating mixing lengths of biological
importance (i.e., mixing of the water that results in
increased productivity). Until turbulence in SAV beds
is better understood, few predictions regarding the
importance of turbulence for the health and distribu-
tion of SAV can be made.
SAV AND TIDES
Most SAV species are not tolerant of dessication
because they lack the waxy cuticle found in terrestrial
plants and, thus cannot grow in the intertidal zone.
Small SAV species that are found in intertidal pools
(like plants from the genus Halophila) and SAV beds
that retain water between their leaves at low tide, can
colonize the intertidal area. These are the exceptions.
Additionally, plants that colonize the intertidal area in
temperate zones often are removed by shifting ice dur-
ing the winter. Consequently, the minimum depth of
distribution of most SAV species is limited to the mean
low water level while the maximum depth of distribu-
tion is limited by the light availability (Figure VI-4).
As mentioned above, waves may also limit the mini-
mum depth of SAV distribution (see Figure VI-2),
therefore, tides and waves need to be considered as
confounding factors when analyzing the vertical distri-
bution of SAV. As waves and tides co-occur in many
SAV habitats, tides will constantly change the wave
mixing depth (see above). Therefore, theoretically, the
minimum depth of distribution should be at a depth
below the mean low water (MLW) line-the MLW level
plus the wave mixing depth (Figure VI-5).
Minimum Depth of Distribution
The minimum depth of distribution based on tides
alone can be defined as half the tidal amplitude (A in
m) below mean tide level (MTL) (see Figure VI-4). In
areas with diurnal tidal cycles, this will be {MHW-
-------
82 SAV TECHNICAL SYNTHESIS II
A. Small tidal range
B. Large tidal range
MHW
MTL
LW
FIGURE VI-4. Tidal Range Influence on Vertical SAV Depth Distribution. The vertical range of distribution of SAV
beds can be reduced with increased tidal range. The minimum depth of SAV distribution (Zmin) is limited by the low
tide (T), while the maximum depth of SAV distribution (Zmax) is limited by light (L). The SAV fringe (arrow) decreases
as tidal range increases. A small tidal range results in a large SAV depth distribution (A), whereas a large tidal range
results in a small SAV depth distribution (B). Mean high water (MHW), mean tide level (MTL) and mean low water
(MLW) are all illustrated.
S°
S2
faj
ea
X 3
£
w ,
TIDAL RANGE (m)
123456
increased light attenuation
MAXIMUM DEPTH OF SAV DISTRIBUTION (m)
FIGURE VI-5. Minimum and Maximum Depth of SAV
Distributions as a Function of Tidal Range. The
minimum depth of SAV distribution defined as the mean
low water (MLW) line decreases with wave exposure
(see Figure VI-2), while the maximum depth of SAV
distribution decreases with increasing light attenuation
coefficients. MTL is the mean tide level.
MLW}/2, while in areas with semi-diurnal tides it will
be {MHHW-MLLW}/2, where MHW is mean high
water, MHHW is mean higher high water and MLLW
is mean lower low water. A method to calculate the
minimum depth of distribution (Zmin) including the
wave mixing depth has been suggested by Chambers
(1987) and is described above. It should theoretically
be defined as:
rr
Z. =: •
mm ,,
(VI-4)
2 2
where the first term of the equation refers to the tidal
amplitude and the second term refers to the wave mix-
ing depth (see Equation VI-2). Equation VI-4 suggests
that in areas of high tidal amplitude and high wave
exposure, SAV will be forced to colonize relatively
deep waters. Its success in colonizing such areas will
depend on their maximum depth of distribution.
Maximum and Vertical Distributions
The maximum depth of distribution of SAV depends
on the light attenuation in the water column (Kd) as
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 83
well as on the water depth (which is a function of
tides). Therefore, tides and the maximum depth of dis-
tribution of SAV are confounding factors (Carter and
Rybicki 1990; Koch and Beer 1996). In areas with high
tidal amplitude: 1) SAV is forced into deeper areas
due to dessication and freezing (Figure VI-4); and 2)
the water column is deeper during high tide than in an
area with a smaller tidal amplitude (i.e., there is more
water to attenuate light). This will reduce the light
available to the plants as well as the number of hours
of saturating light levels (Koch and Beer 1996). There-
fore, the SAV bed is limited by the upper (determined
by tides and waves) and lower (determined by light
penetration) depths of distribution (Figure VI-5).
The maximum depth of distribution (Zmax) can be cal-
culated based on the Lambert-Beer equation:
Z=.
max
IP
(VI-5)
Kd
where Iz /I0 is the percent light required by the species
under consideration or the percent light at the maxi-
mum depth of distribution of the plants. From Equa-
tion VI-5, it is evident that, as Kd increases, the
maximum depth of distribution becomes shallower,
which further restricts the vertical distribution of the
plants (Figure VI-6).
No SAV species can survive if
(VI-6).
This shifts the focus from considering how deep SAV
can grow to how narrow their depth distribution can
be in order to sustain healthy beds. For Z. marina to
successfully colonize an area in Long Island Sound,
Koch and Beer (1996) found that
(VI-7)
was a necessary condition for the existence of this SAV
species. This 1-meter potential vertical depth distribu-
tion below Zmin is necessary as a buffer when, during
storm events, the shallower portion of the SAV bed is
exposed to air, rain or ice. The deeper portions of this
fringe can provide the necessary energy to allow the
shallower portion to recover from the stress of expo-
sure (Koch and Beer 1996).
TIDAL RANGE (m)
23456
W
09
a 3
£
MAXIMUM DEPTH OF SAV DISTRIBUTION (m)
FIGURE VI-6. Effect of Increased Light Attenuation
on Maximum Depth of SAV Distribution. The magni-
tude of the effect of increased light attenuation (Kd) on
the maximum depth of SAV distribution as determined
based on the equations presented in the text, assuming
a SAV minimum light requirement of 13 percent. MIL is
the mean tide level.
For the mixture of estuarine species in Chesapeake
Bay, the vertical depth of distribution seems to be
smaller than that found for Long Island Sound, but
this value still needs to be defined.
The management implication of not only considering
the maximum depth of distribution for SAV but the
vertical depth range that they can colonize is that, in
areas with high tidal ranges, testing attainment of min-
imum light requirements needs to be adjusted to
account for tidal ranges. The reason for this is that if
the tidal range is large (i.e., Zmin is relatively deep) and
the light availability is low (i.e., Zmax is relatively shal-
low), SAV may be restricted to such a narrow vertical
depth that its long-term survival is not viable (Koch
and Beer 1996).
Figure VI-7 indicates how Kd in combination with tidal
range and depth can be used to predict the vertical dis-
tribution of SAV in an area. In Figure VI-7, a tidal
range of 0.8 meters (see x-axis), a minimum light
requirement of 14 percent (Iflj and a Kd=1.5 m'1
(see horizontal dashed lines) are assumed. A line is
drawn vertically from the 0.8-meter tidal range. The
depth at which it intersects the diagonal line deter-
mines Zmin while the depth at which it intersects the
horizontal dashed line for the selected K, value
-------
84 SAV TECHNICAL SYNTHESIS II
TIDAL RANGE (m)
0.0 0.2 0.4 0.6 0.8 1J) 1.2 1.4 1.6 1.8 2.0
MAXIMUM DEPTH OF DISTRIBUTION (m)
FIGURE VI-7. Area-Specific Prediction of Vertical
SAV Depth Distribution. An example of how this type
of graph can be used to predict the vertical distribution
of SAV in a certain area. A tidal range of 0.8 m (see
x-axis), a minimum light requirement of 13 percent
(part of the equation to determine Zmax) and a Kd =
1.5 m~1 (see horizontal dashed lines) are assumed.
A line is drawn vertically from the 0.8 m tidal range.
The depth at which it intersects the diagonal line
determines Zmin while the depth at which it intersects
the horizontal clashed line for the selected Kd value
determines Zmax. In this case, SAV has the potential
to grow in a fringe between 0.4 and 1.3 m deep.
MTL is the mean tide level.
0.0-
0.4-
Q.
«
O
1.3-
Tide
SAV
MTL
linimum
depth of SAV
distribution
based a tidal
range of 0.8 m
Maximum depth of SAV distribution based on
a KJ = 1.5 m-1 and a minimum light requirement
of 13 percent of the light Incident on the surface
FIGURE VI-8. Illustration of Tidal Range Influences
on Vertical SAV Depth Distribution. Illustration of the
SAV vertical depth distribution fringe determined in
Figure VI-7. The fringe will occur in a 0.9 m depth
interval (1.3-0.4 m) due to tidal limitations at the top
and light limitations at the bottom.
determines Zmax. In this case, SAV has the potential to
grow in a fringe between 0.4 and 1.3 m depth (vertical
bar in Figure VI-7 and SAV arrow in Figure VI-8) . In
this case, the SAV fringe will be 0.9 meters deep
(1.3 meters-0.4 meters).
Tidal amplitudes in Chesapeake Bay (MHHW-
MLLW) range from nontidal in some dammed tribu-
taries to 50 cm in the upper Potomac River,
Washington, DC, 60 cm in the Patuxent River, Mary-
land and 64 cm in the Nansemond River, Virginia.
SAV light requirements of Kd 1.5 and 2 m'1 were estab-
lished in 1992 for Chesapeake Bay, based on a 1-meter
restoration depth (Batiuk et al 1992). With this rela-
tively high light attenuation, SAV would probably not
exist if tidal ranges were higher than 1 meter (this
allows SAV to colonize a fringe between 75 and 40 cm
in depth, respectively). From Figure VI-9a, it can be
seen that, when the low tide at the Nansemond River
occurs at noon, the maximum light intensity resembles
that of full sunlight but, the hours of saturating light
(Hsat) decrease with increasing Kd. Figure VI-9b shows
the dramatic effect that a high tide at noon has on the
light availability to SAV in the Nansemond River. The
higher the Kd, the less light is available to the SAV
when the high tide is at noon. Figures VI-9a and VI-9b
are based on a clear sunny day. If the day is cloudy, the
effect of a high tide at noon could lead the SAV to be
exposed to light levels below saturation while a low
tide at noon could reduce the number of hours of sat-
urating light that the plants receives even on a sunny
day by reducing the light available to the plants in the
early morning and late afternoon (high tides).
Tides have a significant effect on the light available to
SAV in Chesapeake Bay, where tidal amplitudes are
relatively small but Kd values are relatively high. In
areas with lower Kd values in the past, an increase in
Kd combined with high tidal amplitudes, can jointly
contribute to the decline of SAV distribution (Koch
and Beer 1996).
The SAV light requirements presented in the first SAV
technical synthesis (Batiuk et al. 1992) were estab-
lished based on a restoration depth of 1 meter MLW
(mean low water) without considering tidal range lev-
els. This overestimated Zmeiv and underestimated the
lUdA
water quality necessary to allow SAV to recolonize
areas down to a depth of 1 meter (Zmax = 1 meter).
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 85
A (Low tide at noon)
B (High tide at noon)
t/5
C
0}
~O
E
13
I
CD
2000
1800
1600
1400
1200
1000
800
600
400
200
0
V)
ex/
03
o
E
X
CD
2000
1800
1600
1400
1200
1000
800
600
400
200
0
0 2 4 6 4 10 12 14 16 1» 20 22 24
light at surface
kd=-2.Dm-1
kd=-1.Sm-1 *
kd=-1.Dm-1
6 8 10- 12 1* 16 14 20 22 24
TIME
TIME
FIGURE VI-9. Simulated Diurnal Light Curves for Different Light Attenuation Coefficients. Simulated diurnal
light curves for different light attenuation coefficients (Kd) in the Nansemond River, Virginia, with a 64 cm tidal range
during a clear, cloudless day, when the low tide occurs at noon (A) and when the high tide occurs at noon (B). The
horizontal line indicates the light level that saturates photosynthesis. The number of hours of saturating light (Hsat)
decrease with increasing Kd values when low tide occurs at noon (A) Although the number of hours of saturating
light are not affected when high tide occurs at noon (B), the higher the Kd, the lower the light levels available to the
SAV (B).
Table VI-4 summarizes the differences between the
1992 SAV light requirements based on MLW and the
revised light requirements taking the tidal amplitudes
into account. These values were obtained by changing
Z in Lambert-Beer's equation to Z 4- A/2, where A is
the tidal amplitude:
species. From Figure VI-7, it can be concluded that
SAV will not occur in areas where Kd=2 m"1 if the tidal
range is equal or higher than 1.8 meters. Under these
conditions Zmax=Zmin. The true tidal requirement can
only be determined once the minimum fringe of distri-
bution for the SAV species in question is determined.
Kd = -
IP
(VI-8).
In summary, tidal amplitude and Kd have a strong con-
founding effect on the distribution of SAV. This has
been conclusively demonstrated in the literature, and
simple equations exist to predict the SAV distribution
based on the interaction between tides and Kd (Koch
and Beer 1996). In order to incorporate tidal ampli-
tude as an SAV habitat requirement in Chesapeake
Bay, it is necessary to determine the minimum vertical
distribution for marine, estuarine and freshwater SAV
SAV AND THE SEDIMENTS IT COLONIZES
Sediments are important in determining the growth,
morphology and distribution of SAV (Short 1987; Liv-
ingston et al. 1998) due to the availability of nutrients
and phytotoxins as well as erosional/deposition
processes (Marba et al. 1994; Dan et al. 1998; Koch
1999). Extreme events causing massive erosion or dep-
osition of sediments can cause the death of entire SAV
populations. The sediment underlying an SAV bed in
Florida was completely eroded away and redeposited
elsewhere (Hine et al. 1987). The massive destabiliza-
tion of this population may have been caused by heavy
grazing of the plants.
-------
86 SAV TECHNICAL SYNTHESIS II
TABLE VI-4. Light attenuation requirements neces-
sary for the recovery of SAV down to the 1 meter
depth contour, taking tides into account. l^ (PLW)
is assumed to be 13 percent for tidal fresh and
oligohaline plants and 22 percent for mesohaline
and polyhaline plants.
Salinity
Regime
Tidal fresh
and
oligohaline
Mesohaline
and
polyhaline
Mean
Tidal
Range
0.0m
O.lm
0.2m
0.3m
0.4m
0.5m
0.6m
0.0m
O.lm
0.2m
0.3m
0.4m
0.5m
0.6m
SAV K,, (m1) habitat
requirement
for 1 m restoration target1
based on MLW2 + tidal range
1.97
1.87
1.79
1.71
1.64
1.57
1.51
1.51
1.44
1.37
1.32
1.26
1.21
1.16
SAV habitat requirements from Batiuk et al. 1992.
Mean low water.
On the other extreme, high sedimentation rates can
also be responsible for the decline of SAV populations.
Moderate depositional rates can stimulate the growth
of Thalassia testudinum (Gallegos et al. 1993) and
Cymodocea nodosa (Marba and Duarte 1994), but
high depositional rates can lead to the disappearance
of these plants.
Seedlings are more susceptible to high burial rates
than established SAV beds (Marba and Duarte 1994).
Therefore, the season of depositional events is impor-
tant in determining the chances of survival of SAV
beds. The deposition of more than 10 cm of sediment
on top of V americana tubers reduced their chances of
becoming mature plants and establishing a meadow
(Rybicki and Carter 1986). Such high depositional
rates can occur during severe storms. In contrast, Z.
marina seeds need to be buried at least 0.5 cm, where
conditions are anoxic, to promote germination (Moore
etal 1993).
During less extreme conditions, SAV can modify the
characteristics of the sediment it colonizes by reducing
current velocity and attenuating waves within its beds
(see the review by Fonseca, 1996). This leads to the
deposition of small inorganic and light organic parti-
cles (Kenworthy et al. 1982). The suitability of fine sed-
iments and sediments with high organic content for
SAV growth are addressed below.
Grain Size Distribution
Sediments within SAV beds are finer than those in
adjacent unvegetated areas (Scoffin 1970; Wanless
1981; Almasi et al. 1987). As SAV density increases,
the ability to accumulate fine particles is also
enhanced (due to the reduction in current velocity and
wave energy). As grain size distribution becomes
skewed toward silt and clay, the porewater exchange
with the overlying water column decreases. This may
result in increased nutrient concentrations (Ken-
worthy et al. 1982) and phytotoxins such as sulfide in
marine sediments (Holmer and Nielsen 1997). At the
other extreme, if SAV colonizes coarse sand, the
exchange of porewater with the overlying water col-
umn will be enhanced and nutrient availability in the
sediment may be lower than that of finer sediments.
In an experiment using different grain sizes of ground
glass (to avoid adsorbed nutrients), Ruppia maritima
was found capable of colonizing sediments from a
silt/clay mixture to coarse sand. Maximum growth was
observed in fine and medium sand particles (Seeliger
and Koch, unpublished data).
Table VI-5 lists quantitative and qualitative data on
the silt and clay amount found in healthy SAV beds.
The values range from 0.4 to 90.1 percent. The highest
values seem to be associated with beds in lower salin-
ity environments with the exception of a Zostera muel-
leri bed. Perhaps, in such environments, the plants are
able to colonize sediment with reduced porewater
exchange with the water column because sulfide does
not occur at the same levels as in marine/higher salin-
ity estuarine systems. In higher salinity environments,
it appears that plants need sediments that are more
oxygenated and in which sulfide levels can be reduced
via higher porewater advection rates. Therefore,
SAV growth may be limited by the physical and
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 87
TABLE VI-5. Percent fine sediment (< 63 ^urn) or sediment type found in healthy SAV beds1
Percent fines/sediment type Species
Source
1.9
4.8
0.4 - 9.0
1.8-9.2
6-10
14
14.6
0.8 - 14.7
15
8.1-28.8
2.8 - 30.9
1-34
2 - 39 % clay
40
4.3 - 47
58
4 - 62 % silt
48 % sUt and 14% clay
0.5 - 72
4.7-90.1
"Sat loving"
Syringodium filiforme
Thalassia testudinum
Posidonia oceanica
T. testudinum, S. filiforme and
Halodule wrightii
Chesapeake Bay SAV
Zostera marina
T. testudinum
Thalassia and Halodule
Z. marina
Halodule and Zostera
Heterozostera tasmanica
T. testudinum
Tidal Potomac River SAV
Hydrilla and Vallisneria
No SAV (Thalassia,
Syringodium and Halodule)
Hydrilla verticillata
Tidal Potomac SAV
Vallisneria americana
Zostera muetteri
Mixture of Vallisneria,
Potamogeton perfoliatus,
P. pectinatus, Ruppia maritima
P. pectinatus
Woodetal. 1969
Vfoodetal. 1969
Edgar and Shaw 1995
Livingston et al. 1998
Batiuke/a/. 1992
Marshall and Lucas 1970
Scoffin 1970
Grady 1981
Orth 1977
Murphey and Fonseca 1995
Edgar and Shaw 1995
Burrell and Schubel 1977
Carter etal 1985
Poseyetal. 1993
Livingston et al 1998
Poseyetal. 1993
Carter etal. 1985
Hutchinson 1975
Edgar and Shaw 1995
Pascals al. 1982
Sculthrope 1967;
Hasiam 1978
Saty substrate P. perfoliatus Hasiam 1978
1 Arranged in ascending order of maximum percentage of fine sediments.
continued
-------
88 SAV TECHNICAL SYNTHESIS II
TABLE VI-5. Percent fine sediment (< 63 (Jim) or sediment type found in healthy SAV beds1 (continued)
Percent fines/sediment type Species
Source
Silty substrate
Mud
Organic ooze
Soft or sandy mud
Muds or sand
Clay and sand
Medium-grained substrate
Medium-grained substrate
Sandy
Sagittaria sagittifolia
Ceratophyllum demersus
Myriophyllum spicatum
Myriophyllum spicatum
Ruppia maritima
Zannichellia palustris
Myriophyllum spicatum
Ranunculus
Najas guadalupenses
Haslam 1978
Hutchinson 1975
Patten 1956
Springer 1959
Anderson 1972
Stevenson and Confer 1978
Haslam 1978
Haslam 1978
Martin and Uhler 1939
1 Arranged in ascending order of maximum percentage of fine sediments.
geochemical processes associated with a certain sedi-
ment type and not by the grain size per se (Barko and
Smart 1986). Data in Table VI-5 are not sufficient to
establish the 'best' sediment types for SAV growth at
this time.
Sediment Organic Content
SAV tends to accumulate organic particles due to a
reduction in current velocity and wave energy within
the meadows and canopies. Organic matter can also be
accumulated in SAV colonized sediments through the
burial of rhizomes and roots produced over time. The
age of the organic deposits beneath a Posidonia ocean-
ica bed were found to be up to 3,370 years old (Mateo
et al. 1997). High burial rates and/or low decomposi-
tion rates may account for the accumulation of organic
matter over such long periods.
The organic carbon content of sediments from the
mainstem Chesapeake Bay has increased two to three
times over the last 80-100 years (Cornwell et al. 1996).
This was attributed to changes in inorganic matter
deposition due to increased phytoplankton biomass
(Harding and Perry 1997) as well as time-dependent
changes in organic matter decomposition. If the
organic content of the sediments in shallow areas in
Chesapeake Bay SAV habitats are also increasing, they
could limit the distribution of SAV. Barko and Smart
(1983) and Koch (unpublished data) concluded that
the growth of SAV is limited to sediments containing
less than 5 percent (dry weight) organic matter.
This is also supported by other data summarized in
Table VI-6.
The mechanism behind this limitation of high sedi-
ment organic content on SAV growth is not well
understood. It may be due to nutrient limitation in
very fine sediments associated with high organic
deposits (Barko and Smart 1986) or due to high sulfide
concentrations in marine sediments, known to be toxic
to high salinity SAV (Carlson et al. 1994).
The data in Table VI-6 lists organic contents of less
than 12 percent for SAV colonized sediments. The
higher values (6.5 to 12 percent) are mostly associated
with SAV species that have large leaves. Perhaps these
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 89
TABLE VI-6. Sediment organic matter as percent of dry weight in healthy SAV beds.1
Percent Organics
Species
Source
1.25
1.25
0.41 -1.38
<2
2.5
3.25
0.77 - 3.62
3.5-4.9
<5
<5
1-5.3
2.6 - 5.3
<6.5
0.8 - 7.3
8
1.6-12
<26mgCg'1
Zoster a marina
Z. marina
Z. marina
Ruppia maritima
Syringodiumfiliforme
R. maritima
Halodule and Zostera
Thalassia testudinum
Z. marina
Hydrilla and Potamogeton nodosus
Chesapeake Bay SAV
Heterozostera tasmanica
Vallisneria americana
Zostera muelleri
Z. marina
Posidonia spp.
Potamogeton pectinatus
Marshall and Lukas 1970
Orth 1977
Dan etal. 1998
Ward etal 1984
Wood et al 1969
Kemp etal. 1984
Murphey and Fonseca 1995
Wood era/. 1969
Koch 1999
Barko and Smart 1983
Batiuketal. 1992
Edgar and Shaw 1995
Hutchinson 1975
Edgar and Shaw 1995
Short (personal communication)
Edgar and Shaw 1995
vanWijkefa/. 1992
1 Arranged in ascending order of maximum percentage of organic matter.
plants can colonize sediments with higher organic con-
tent due to a large oxygen production in the leaves
and, consequently, also a higher transport of oxygen to
the roots. If the rhizosphere is well-oxygenated, the
detrimental effects associated with high organic con-
tent in the sediments may be neutralized. The distri-
bution of Potamogeton spp. in English Lakes (Pearsall
1920; summarized in Hutchinson 1975) was directly
correlated with sediment organic content (X) and min-
imum light requirement (Y) where Y=0.70 + 0.65 X
(r2=0.90). Therefore, plants growing in more organic
sediments with higher concentrations of phytotoxic
metabolites require more light to support greater
release of oxygen from their roots to the rhizosphere.
This mechanism has been used to explain the decline
in abundance of SAV populations in eutrophic regions
that have experienced an increase in sediment organic
content (Nienhuis, 1983).
Due to the large numbers of studies that observed
the percent organic matter in SAV beds to be below
-------
90 SAV TECHNICAL SYNTHESIS I
5 percent, it is recommended that caution should be
taken when transplanting SAV into areas where the
sediment organic content is higher than that value.
Additional studies are needed to define the SAV habi-
tat requirement for organic matter for different SAV
species in Chesapeake Bay.
SAV AND SEDIMENT GEOCHEMISTRY
Nutrients in Sediments
Nutrients in the sediment can be limiting to the
growth of SAV (Short 1987; Agawin et al. 1996) but do
not seem to eliminate it from colonizing certain areas.
In marine siliceous sediments, nitrogen may limit SAV
growth (Short 1987; Alcoverro et al. 1997) while in
marine carbonate sediments, phosphorus may be lim-
iting to SAV growth (Wigand and Stevenson 1994).
Potassium has been suggested to be limiting to the
growth of freshwater SAV (Anderson and Kalf 1988).
Mycorrhizae have been found to facilitate the phos-
phorus assimilation in V americana (Wigand and
Stevenson 1997), but little or no information is avail-
able on mycorrhizae in the rhizosphere of marine SAV
(Wigand and Stevenson 1994).
Although light seems to be more limiting to SAV
growth than sediment nutrient concentrations, excep-
tions can be found. In tropical SAV beds, light and
temperature are limiting in the winter while nutrients
are limiting in the summer (Alcoverro et al. 1997b).
Additionally, ammonium concentrations as low as
25 /im (in the seawater) can be toxic to Z. marina
and ultimately lead to its decimation (van Katwijk
et al 1997).
Microbial-Based Phytotoxins
A wide variety of potentially phytotoxic substances is
produced by bacterial metabolism in anaerobic sedi-
ments, including phenols and organic acids, reduced
iron and manganese and hydrogen sulfide (Yoshida
1975; Gambrell and Patrick 1978). In many aquatic
environments, sulfide probably constitutes the most
important of these toxic bacterial metabolites and has
been shown to be toxic to estuarine and marine SAV
species (van Wijck et al 1992; Carlson et al 1994).
Sulfide is generated by sulfate reducing bacteria dur-
ing organic carbon oxidation and nutrient remineral-
ization in anoxic sediments (Howarth 1984; Pollard
and Moriarty 1991). A high remineralization rate leads
to high nutrient availability and favors plant growth
but can also lead to the accumulation of sulfide, which
is detrimental to plant growth. Sulfate remineraliza-
tion depends on the temperature and amount of
organic matter in the sediment. In freshwater sedi-
ments, sulfate reduction is less important than
methanogenesis due to the lower sulfate availability.
As SAV tends to accumulate more organic and inor-
ganic particles than unvegetated areas, sulfate reduc-
tion rates can be expected to be higher within the
vegetation than outside it (Isaksen and Finster 1996;
Holmer and Nielsen 1997). This difference could also
be due to the excretion of organic compounds through
the roots (Blackburn et al. 1994).
Table. VI-7 summarizes sulfide levels observed in
healthy as well as deteriorating SAV beds, and Table
VI-8 lists sulfate reduction rates in healthy SAV beds.
While eutrophication can fuel sulfide production via
increased organic matter in the sediments, sulfide pro-
duction can also fuel eutrophication. The sulfide pro-
duced inhibits nitrification (NH4+ -> NO3") and,
consequently, increases ammonium fluxes to the water
column, which could act as a positive feedback to
eutrophication (Joye and Hollibaugh 1995).
A parallel mechanism also applies to the release of
phosphorus from the sediments. When sulfides react
with iron in the sediment, iron sulfide is formed and
phosphorus is released (Lamers et al. 1998). If the
phosphorus is not taken up by the plants, it may be
released back into the water column and, as stated
above for nitrogen, may be a positive feedback to
eutrophication.
The toxicity of sulfide to plants can also be enhanced
by eutrophication. Oxygen released from SAV roots is
needed to oxidize the sulfide and reduce its toxic
effects (Armstrong 1978). The release of oxygen by the
roots depends on the photosynthetic rates of the plant.
Therefore, if eutrophication leads to a reduction in
light availability, photosynthetic rates will be lower,
the amount of oxygen released from the roots will also
be reduced and sulfide toxicity may be enhanced
(Goodman et al. 1995).
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 91
TABLE VI-7. Sulfide levels in the sediments of healthy and dying SAV beds.
Sulfide concentration 1 Species
Plant Status
Source
350 to 1,000
88 m mol m'2 AVS
0.48 to l
> 400
> 2,000
^ote the different units.
Thalassia testudinum Healthy
Halodule beaudetti Healthy
Halophila engeltnanii Growth
Carlsonef a/. 1994
Blackburn etal. 1994
Pulich 1983
Potamogeton pectinatus Declining growth van Wijck et al. 1992
Zostera marina
T. testudinum
Reduced
photosynthesis
Dead
Goodman etal. 1995
Carlson etal. 1994
TABLE VI-8. Sulfate reduction in healthy SAV beds.
Sulfate reduction
Species
Source
20 n mol ml"1 fr1
87 to 445 n mol cm'3 d"1
7 to 7.9 m mol m'2 d'1
9.4 m mol m"2 d'1
12 to 16.8 m mol m"2 d'1
16.3 m mol m'2 d'1
27.9 to 33. 4m mol m-'d
59.1mmolnV2d-1
Zostera noltii
Z. noltii
Syringodium isoetifolius
Halophila
Zostera capricorni
Halodule uninervis
Halodule beaudetti
Cymodocea serrulata
Zostera marina
Welsh etal. 1996
Isaksen and Finster 1996
Perry and Dennison submitted
Perry and Dennison submitted
Perry and Dennison submitted
Perry and Dennison submitted
Blackburn et al. 1994
Perry and Dennison submitted
Holmer and Nielsen 1997
'Note the different units.
-------
92 SAV TECHNICAL SYNTHESIS II
Sulfide concentration in the sediment is an important
SAV habitat requirement. Correlations between pore-
water sulfide concentrations and the growth of several
SAV species have indicated that concentrations above
1 mM may be toxic (Pulich 1989; Carlson et al. 1994).
Direct manipulations of sulfide concentrations
revealed a negative effect on photosynthesis (Good-
man et al. 1995) and growth (Kuhn 1992) when levels
were higher than 1 to 2 mM. Sulfide thresholds for dif-
ferent SAV species (in combination with different light
levels) still need to be determined. Until such data are
available, critical sulfide concentrations cannot be
specified as an SAV habitat requirement.
CHEMICAL CONTAMINANTS
This review is intentionally confined to the broad
issues of the potential roles contaminants may have in
limiting the size, density and distribution of SAV pop-
ulations. Literature values pertaining to the relation-
ships between SAV and chemical contaminants are
derived from three diverse lines of inquiry: contami-
nant studies, phytoremediation efforts (Garg et al
1997; Peterson et al. 1996; Ramanathan and Burks
1996; Salt et al. 1995), and recommendations for
aquatic weed control (Anderson and Dechoretz 1982;
Anderson 1989; Nilson and Klaassen 1988). Appendix
B summarizes some of the more recent work devoted
to contaminant issues.
Most of the chemical contaminant studies have evalu-
ated the effects of herbicides on SAV growth (Fleming
et al. 1993). A few have examined other pesticides or
heavy metals (Garg et al. 1997; Gupta and Chandra
1994; Gupta et al. 1995). Thus, the vast majority of com-
pounds known to have toxic effects on biological sys-
tems remain untested (Van Wijngaarden et al 1996)
and only a few efforts have been made to systematically
evaluate additive, cumulative and synergistic effects of
multiple contaminants (Fairchild et al. 1994; Huebert
and Shay 1992; Sprenger and Mclntosh 1989).
Nonetheless, the following conclusions can be drawn
from these accumulated data.
• Herbicides have been shown to be phytotoxic to
SAV. Toxicity is somewhat species-dependent and
chemical-specific. Table VI-9 depicts the toxicity
range of the most widely used herbicides in the
TABLE VI-9. Relative effects of herbicides on net photosynthesis in Potemogeton pectinatus. The IC50 is
the predicted concentration that inhibits photosynthesis by 50%. Photosynthesis was determined by
measuring O2 production by plants over 3 hours at 20-22°C and about 58% p10,000
>1,000;<10,000
29
32
>10,000
70
240
>10,000
8
164
>10,000
95% Confidence
Interval
20-42
21-48
43-112
130 - 420
5-12
82 - 327
Slope
-50.97
-42.91
-59.14
-36.04
-41.25
-88.80
R2
0.81
0.88
0.64
0.68
0.86
0.58
-------
Chapter VI - Beyond Light: Physical, Geological and Chemical Habitat Requirements 93
United States onPotamogetonpectinatus. Inhibit-
ing concentrations range from 8 ppb to 10,000
ppb. These concentrations are consistent with
those observed when aquatic weed control is the
management objective, as well as in environ-
ments where the protection of aquatic plants is
the management objective.
• Pesticides other than herbicides have been shown
to have a phytotoxic effect on SAV, although only
a few have been evaluated.
• Heavy metals at levels corresponding to some
ambient conditions have inhibiting effects on
SAV in test systems where the variety of essential
plant nutrients has been experimentally factored.
• The environments holding the greatest potential
for pesticide suppression of SAV populations are
headwaters and shallow waters immediately adja-
cent to the urban, forest and agricultural areas
where pesticides are most widely used and acute
concentration level exposures are most likely
to occur.
• The environments holding the greatest potential
for adverse effects of heavy metals are those
where clay and organic sediments chemically
concentrate both metals and plant nutrients for
extended periods.
• The utility of ambient testing of contaminant
concentrations is highly controversial. For pesti-
cides, the constraint of monitoring frequency and
location are limiting factors for accurate ambient
assessment of contaminant presence. Assessment
of heavy metals and other contaminants is con-
founded by the difficulty of distinguishing
between the concentration of biologically active
forms and total concentration (Liang and Schoe-
nau 1996).
PHYSICAL AND GEOLOGICAL SAV HABITAT
REQUIREMENTS
In order to fully define the SAV habitat requirements
in Chesapeake Bay, parameters other than light and its
modifiers need to be taken into consideration. Some
physical, geological and geochemical parameters have
the potential to override established SAV light
requirements. Where field and laboratory experimen-
tal data were sufficient, physical and geological SAV
habitat requirements were identified (Table VI-10).
-------
94 SAV TECHNICAL SYNTHESIS II
TABLE VI-10. Summary of physical and geological SAV habitat requirements for Chesapeake Bay.
Parameter
SAV Habitat
Requirement
Observations
Current velocity
Oligohaline Habitats
Polyhaline Habitats
Minimum depth of
SAV distribution
More data are needed (specially for canopies and
0 to ? cm s'1 meadows; plants with polar/non polar leaves) to
>10 and < 100 cm s"1 further define this SAV habitat requirement.
A is the tidal amplitude ({MHW-MLWJ/2 for
diurnal tides and (MHHW-MLLW}/2 for semi-
diurnal tides); g is the acceleration of gravity (9.805
m s"1); and T is the wave period.
Z . =4+J*.
mm f\ />
Maximum depth of
SAV distribution
Should be defined by IJ10 is the minimum percent light required by the
species under consideration. Kd is the light
-ln(—) attenuation coefficient. This calculation should use
I0 MWL (mean water level) as a reference. Using
- MHW (mean high water) may underestimate
d and using MLW (mean low water) may
overestimate Zm,T.
max
Tides
Sediment grain size
Oligohaline Habitats
Polyhaline Habitats
Sediment organic
content
SAV can be expected
to successfully
colonize areas where
No restrictions?
<20% silt and clay
0.5 m has been chosen as a conservative value of
the smallest vertical depth of colonization observed
for SAV in Chesapeake Bay. This value was found
to be 1 m for eelgrass in Long Island Sound and
needs to be adjusted for the species and sites in
question. Ztt6a is the minimum depth of distribution
and depends on the tidal range and wave mixing
depth (see Equation VI-4).
These preliminary SAV habitat requirements are
based on compilations of data from the literature.
Specific studies are needed to confirm these
preliminary requirements.
This SAV habitat requirement is based on a
compilation of data from the literature. Although all
literature values converge to the suggested value,
specific studies are needed to further verify the
requirement.
-------
CHAPTER V||
Setting, Applying and Evaluating
Minimum Light Requirements for
Chesapeake Bay SAV
"Phis chapter defines the two types of minimum light
I requirements for Chesapeake Bay SAV, and
explains how to test their attainment using two new
percent-light parameters that can be calculated from
water quality monitoring data. It compares the five
1992 habitat requirements from the first SAV technical
synthesis (Batiuk et al. 1992, Dennison et al. 1993) to
the new minimum light requirements, in terms of
attainment of habitat requirements and their ability to
predict SAV presence. The chapter also describes the
adjustment of the new percent-light parameters to
account for tidal range, and examines the relationships
between the two percent-light parameters and the aver-
age depth at which SAV is growing in Chesapeake Bay.
DEFINING AND APPLYING THE
MINIMUM LIGHT REQUIREMENTS
AND WATER-COLUMN LIGHT REQUIREMENTS
Evaluation of ambient water quality conditions sup-
porting the required amount of light reaching SAV can
be viewed in terms of two percent-light parameters.
• Percent light through water (PLW) is the amount
of ambient sub-surface light absorbed or
reflected by water itself, water color caused by
dissolved organic materials, suspended organic
and sediment particles, and phytoplankton in the
water column down to the sediment surface at
the restoration depth selected.
• Percent light at the leaf (PLL) is the amount of
ambient light that actually reaches the SAV leaf
after penetrating the overlying water column and
that is further reduced after being absorbed or
reflected by epiphytic material growing (attached
algae) or settled (organic and inorganic solids) on
the SAV leaf surface at the restoration depth
selected.
The percent light through water parameter is a
component of the percent light at the leaf parameter
(Figure VII-1). To reflect these two ways of evaluating
percent light, we define a "minimum light require-
ment" with attainment tested by the percent light at
the leaf, or PLL parameter (Table VII-1). We also
define a related "water-column light requirement"
with attainment tested, as described above, by the per-
cent light through the water or PLW parameter. Note
that in the original SAV technical synthesis, published
by Batiuk et al. (1992), the habitat requirements had
the same names as the parameters used to test their
attainment, but now they have different names.
The 1992 water quality-based habitat requirements for
Chesapeake Bay SAV were applied separately to test
their attainment (Batiuk et al. 1992). The light attenu-
ation coefficient (Kd) habitat requirement, applied
alone, was roughly the equivalent to the water-column
light requirement defined here and evaluated by the
percent light through the water parameter. Evaluated
collectively, the attainment of the five 1992 SAV habi-
tat requirements provided the best estimate at the
time for defining the water column conditions neces-
sary to achieve sufficient light at the SAV leaf surface.
This estimate is replaced with the attainment of the
minimum light requirements defined here.
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 95
-------
96 SAV TECHNICAL SYNTHESIS II
Percent
Light at
the Leaf
(PLL)
100% Ambient Light of Water Surface
Percent Light
Through Water
(PLW)
SAV Leaf
I
3
o
g_
c
3
CT
O
S
r+
O
V)
00
»
a
Suspended Solids Settled on Epiphytes
Epiphytes Growing on SAV Leaf
FIGURE VIM. Two Percent Light Parameters for Evaluating Ambient Conditions. Illustration of the relationship of
the two percent light parameters and the water quality conditions influencing both of them.
Water-Column Light Requirements
The water-column light requirements are the same as
the two light requirements derived from the in-depth
review and analysis of a wide variety of data modeling
and research findings documented in Chapter III: 13
percent for tidal fresh and oligohaline areas and 22
percent for mesohaline and polyhaline areas. Since
most of the SAV light requirement studies summa-
rized in Chapter III had epiphytes on the SAV, but the
light measurements in those studies did not estimate
light attenuation due to epiphytes, we used these light
requirements to set the water-column light require-
ments. The attainment of the water-column light
requirements is tested using the percent light through
the water (PLW) parameter.
Minimum Light Requirements
Minimum light requirements were determined by
comparing the results of three lines of evidence.
1. Calculation using the 1992 SAY habitat
requirements.
One line of evidence was derived by applying the
salinity regime-based values for the 1992 SAV habi-
tat requirements for Kd, dissolved inorganic nitro-
gen, dissolved inorganic phosphorus and total
suspended solids (Table VII-1) into the equation
for determining PLL (Equation V-l) (Table V-l),
PLL= [e-^X^fe-^WjlOO.
Using this equation, a PLL value of 8.3 percent was
calculated for tidal fresh and oligohaline salinity
regimes. In mesohaline regimes, the calculated
PLL value was 17.3 percent, while it was 13.5 per-
cent in polyhaline regimes. The mesohaline and
polyhaline PLL values differed, even though they
had the same 1992 Kd and dissolved inorganic
nitrogen SAV habitat requirements, because they
had different dissolved inorganic phosphorus SAV
habitat requirements. This difference influenced
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 97
CO
CD
I
JD
"CO
TJ
CO
>,
co
GQ
CD
Jsc
CO
CD
Q.
CO
CO
CD
6
CD
E
CD
C^
CO
co
CO
T3
£=
O
O
CD
CO
-a
co
CO
E
T3
CD
T3
C
CD
E
o
o
CD
cc
LU
m
I
?
es
s s
wi
V
s
I-H
A
5?
^ D,
J &
V-
o
I
•d
B
S
"
(U
«
W
|
^^
r/l Cu
.-r S
0 T3
£
00
B
p
I
2"
&<
a
e
*•" ™ r-<
^ .S3 .S g
o
3
l^s
g § s»
||
6 "§
'5
c/3
C
ca
o
_o
-5 'H 43
-------
98 SAV TECHNICAL SYNTHESIS II
the limiting nutrient and, therefore, the resulting
calculated PLL value. From application of the 1992
SAV habitat requirements, minimum light require-
ments of 8 percent for tidal fresh/oligohaline
regimes and 15 percent (the average of 17.3 and
13.5) were derived from this line of evidence.
2. Accounting for Epiphytic Light Attenuation
As discussed in Chapter III, the scientific studies
used to derive the water-column light targets did
not factor in the shading effects of epiphytes, which
grow on SAV leaves at all depths and on experi-
mentally shaded plants in the field. Several studies
in various estuarine habitats indicate that light
attenuation by epiphytic communities tends to con-
tribute an additional 15 to 50 percent shading on
SAV (e.g., Borum and Wium-Andersen 1980,
Bulthuis and Woelkerling 1983, van Dijk 1993).
One recent detailed study of turtlegrass beds in
Florida coastal waters (Dixon 2000) showed that,
while light levels at the maximum depth of seagrass
colonization averaged about 22 percent of surface
irradiance, epiphytic attenuation reduced this to
approximately 14 percent of the surface light that is
actually available for plant photosynthesis. This
represents an average of approximately 35 percent
additional shading by epiphytes.
Light attenuation by epiphytic material appears to
be generally important throughout Chesapeake
Bay, contributing 20 to 60 percent additional atten-
uation (beyond PLW) in the tidal fresh and oligo-
haline regions, where nutrient and total suspended
solids concentrations were highest, and 10 to 50
percent in the less turbid mesohaline and polyha-
line regions (Figure V-ll). These calculated contri-
butions of epiphyte shading are consistent with the
values derived for PLW and PLL by applying the
1992 SAV habitat requirement values in equations
II-1 and V-l, respectively, where PLL represents
approximately 30 percent additional light reduction
from PLW (Table VII-1).
Based on literature values for seagrass minimum
light requirements, where epiphyte effects were
either avoided with experimental manipulation
(e.g., Czerny and Dunton 1995) or taken into
account with direct measurement (e.g., Dixon
2000), and results from analysis of Chesapeake Bay
data, epiphytic material was assumed to make a 30
percent additional contribution to light attenuation
throughout Chesapeake Bay shallow water habi-
tats. Accounting for the epiphytic contribution to
light attenuation, minimum light requirements for
mesohaline/polyhaline and tidal fresh/oligohaline
habitats, respectively, were calculated to be 15 per-
cent and 9 percent of surface irradiance. These val-
ues, which represent the actual minimum light
needed to support SAV growth at the leaf surface,
include the additional 30 percent epiphytic light
attenuation beyond the respective water-column
light targets derived in Chapter III. For mesoha-
line/polyhaline habitats, factoring the additional 30
percent epiphytic light attenuation into the 22 per-
cent water-column light target yields a 15 percent
minimum light requirement as 30% = 100(22-
15)/22. A 9 percent minimum light requirement for
tidal fresh/oligohaline habitats was derived by fac-
toring the additional 30 percent epiphytic light
attenuation into the 13 percent water-column light
target, as 30% = 100(13-9)/13.
The derived SAV water-column light requirement
and minimum light requirement for Chesapeake
Bay's mesohaline and polyhaline habitats, 22 per-
cent and 15 percent surface light, respectively
(Table VII-1), are remarkably close to the respec-
tive values of 22 percent and 14 percent surface
light derived through field experimentation for
turtlegrass in Florida (Dixon 2000) through this
second line of evidence.
3. Comparisons of Field Conditions and
SAV Growth Gradients
Medians of nearshore water quality data (from the
Choptank and York rivers) and Chesapeake Bay
Monitoring Program midchannel data were
assessed for relationships between calculated PLL
values, SAV growth categories and the proposed
mesohaline/polyhaline and tidal fresh/oligohaline
minimum light requirements of 15 percent and 9
percent, respectively. As described in detail in the
section "Comparing Water Quality Medians over
Categories of SAV Growth," the calculated PLL
values from observed water quality conditions asso-
ciated with "persistent" and "fluctuating" SAV beds
were either all very close to or well above the mini-
mum light requirement, or the limited set of devia-
tions could be readily explained, confirming the
proposed values through the third line of evidence.
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 99
From these three lines of evidence, minimum light
requirements of 15 percent surface light for mesoha-
line/polyhaline habitats and 9 percent surface light for
tidal fresh/oligohaline habitats were established (Table
VII-1). The attainment of the minimum light require-
ment is tested using the PLL parameter.
Primary and Secondary Habitat Requirements
The minimum light requirement is considered the
"primary habitat requirement" (Table VII-1). All the
other requirements, including the water-column light
requirement, are called "secondary habitat require-
ments." This nomenclature was chosen because testing
the attainment of the minimum light requirement is
the primary means for assessing whether an area of
shallow water has water quality adequate to support
SAV growth, whenever the data needed to calculate
PLL are available: light attenuation coefficient (Kd) or
Secchi depth, dissolved inorganic nitrogen, dissolved
inorganic phosphorus and total suspended solids
measurements. SAV habitat quality should be evalu-
ated using the water-column light requirement as a
substitute for the minimum light requirement only if
the data needed to calculate PLW are available (light
attenuation coefficent or Secchi depth) and the
parameters needed to calculate PLL, dissolved
inorganic nutrients and total suspended solids, are
unavailable.
Once the attainment of the minimum light require-
ment has been tested, the attainment of the secondary
habitat requirements should be tested only if the min-
imum light requirement is not met. Testing attainment
of the secondary requirements can suggest possible
reasons for non-attainment of the minimum light
requirement. The secondary requirements should only
be used as diagnostic tools for research or manage-
ment purposes. See Chapter V for a description of a
diagnostic tool based on the total suspended solids and
chlorophyll a secondary habitat requirements.
If the minimum light requirement is met, but SAV is
absent from or sparse in the nearby area, a review of
the many factors that can prevent SAV growth should
be undertaken (see Chapter VI).
Calculating Percent Light Parameters
Building on their initial descriptions in Chapter V,
those applying the minimum light requirements need
to understand the following terms.
Z: SAV restoration depth. This is measured from
just below the water surface to the sediment-water
interface, which is where a submerged plant must
start growing. Z is used in formulae for PLW and
PLL to specify the path length for light passing
through water. This depth is referenced to MLLW
or MTL (see below).
PLW: percent light through water. The percent of
the light level measured just below the surface of
the water that reaches the restoration depth Z, after
passing through the overlying water column but not
through any epiphytes or associated material on an
SAV leaf surface. When Z = 1 meter below MLLW
with no tidal range adjustment, PLW is equivalent
to Kd (light attenuation coefficient) in the 1992 SAV
habitat requirements (Batiuk et al. 1992). In this
document, both of those conditions are relaxed for
PLW as well as PLL: here Z is varied from 0 to
1 meters, and is referenced to mean tidal level with
a tidal range adjustment.
PLL: percent light at the leaf. This refers to the per-
cent of light measured just below the surface of the
water that reaches the surface of an SAV leaf grow-
ing at restoration depth Z (at the sediment-water
interface), after passing through the water column
and any epiphytes and associated material on an
SAV leaf.
MLLW: mean lower low water. This is the mean ele-
vation over time of the lower of the two daily low
tides, where there are mixed tides, as in most of
Chesapeake Bay. Mixed tides occur when the two
high and two low tides each day occur at different
elevations. MLLW is used as the reference for the
bathymetry on nautical charts to minimize the risk
of boats running aground, but it does not estimate
the average depth of water (and thus the average
path length for light attenuation) above a sub-
merged plant through the day.
MTL: mean tidal level. The midpoint between high
and low tides; where there are mixed tides, the mid-
point between MHHW (mean higher high water)
and MLLW. MTL estimates the average depth of
-------
100 SAV TECHNICAL SYNTHESIS II
water above a submerged plant (see Chapter VI for
details).
PLL calculations require measured values of light
attenuation coefficient, total suspended solids, dis-
solved inorganic nitrogen and dissolved inorganic
phosphorus (Figure VII-2). The value for Kd can be
based on a direct measure of light attenuation calcu-
lated by lowering a light meter down through the water
column or converted from Secchi depth data using the
conversion factor Kd = 1.45/Secchi depth (see Chapter
III). However, in some cases there will not be enough
monitoring data to calculate PLL, but there will be Kd
or Secchi depth data that can be used to calculate
PLW. PLW, Kd and PLL are related as follows.
Light measurements (of photosynthetically active radi-
ation, PAR, using a flat cosine sensor) that are needed
for PLL and PLW calculations include:
I0 = light level just below surface of the water
(usually at 0.1-meter depth); and
Iz = light level at depth Z (often measured at
1.1-meter depth for SAV monitoring; in this
example,
z = 1.1-0.1 = 1.0).
Note that when calculating Kd or PLW from pairs of
light measurements, Z represents the difference
between the depth of the subsurface light measure-
ment (Io) and the depth of the deeper light measure-
ment (Iz). Otherwise it represents the restoration
depth chosen.
Another light value needed to calculate PLL, Ize
(which is not measured directly, but is calculated from
Kd, Z, total suspended solids, dissolved inorganic
phosphorus and dissolved inorganic nitrogen) is light
level at the leaf surface (after passing through the
water column as well as any epiphytes and associated
material).
Percent Light through Water (PLW)
Percent Light at the Leaf (PLL)
100% Ambient Light of Water Surface
Inputs
•Kd measured directly
or
•Kd calculated from
Secchi depth
Calculation
•Total suspended solids
•Dissolved inorganic
nitrogen
•Dissolved inorganic
phosphorus
Calculation
Evaluation
PLW vs. Water-Column
Light Requirement
•Ke = Epiphyte attenuation
•Be " Epiphyte biomass
Evaluation
PLL vs. Minimum Light
Requirement
Approach followed when
only Secchi depth/direct
light attenuation data are
available
Recommended approach for
best determination of the
amount of light reaching SAV
leaves
FIGURE VII-2. Calculation of PLW and PLL and Comparisons with their Respective Light Requirements.
Illustration of the inputs, calculation, and evaluation of the two percent light parameters: percent light through water
and percent light at the leaf.
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 101
PLW and PLL can then be expressed as:
PLW = IJ10 x 100
PLL = Ize/Io x 100
or
PLL = [e -] [e
from Chapter V, Table V-l.
x 100
(VIM)
(VII-2)
(VII-3)
See Appendix C for the SAS code used to calculate
PLL from Kd, total suspended solids, dissolved inor-
ganic nitrogen and dissolved inorganic phosphorus. Kd
can be converted to PLW, and PLW to Kd as follows:
yi0 (PLW/100) =
(VII-4)
where Z is in meters and the units for Kd are m1
(Batiuk et al. 1992, page 17; Equation VII-2). This
would be written in SAS or other programming lan-
guage to calculate PLW from Kd as:
PLW = exp(-Kd*Z) * 100
(VII-5).
Taking the natural log of both sides of this equation,
Kd can also be calculated from PLW:
In(PLW) = -Kd*Z + ln(100), or
Kd = -ln(PLW/100)/Z
(VII-6).
Because the calculation of PLL includes total sus-
pended solids and dissolved inorganic nutrient data, it
cannot be converted to an equivalent Kd value.
Adjusting Percent Light Parameters for
Tidal Range and Different Restoration Depths
All of the percent light formulae used in this chapter
were adjusted to account for the tidal range. The 1992
Kd requirements (2 or 1.5 nr1 depending on salinity,
Batiuk et al. 1992) used Z = 1 with no adjustment
because Batiuk et al. (1992) assumed a new plant was
growing at the sediment surface with 1 meter of water
above it. They assumed that there was (on average) 1
meter of water above the sediment surface at the 1-
meter mean lower low water (MLLW) contour, which
was called "MLW" in that document. This assumption
was incorrect. At the 1-meter MLLW contour there
will be 1 meter of water above the plant only once a
day, when the lower low tide occurs (assuming mixed
tides). Thus, on average, there will be more than 1
meter of water above this point, and the light reaching
the sediment surface will be less than what was
expected when the 1992 requirements were set.
This greater depth of water above the plant would be
offset somewhat as the plant grew and the upper parts
were closer to the surface, but the intent of the Chesa-
peake Bay SAV habitat requirements has always been
to predict conditions needed to establish new plants,
which must start growing at the sediment surface. The
habitat requirements were also intended to predict the
conditions needed to restore SAV to a particular shal-
low water area of the tidal Chesapeake, as defined by
the Tier II and Tier III restoration targets (see Chap-
ter VIII). These are the areas of potential SAV habitat
to the 1- and 2-meter depth contours, respectively
(Batiuk et al. 1992). Because those contours were
mapped relative to MLLW, the restoration depth (Z)
in the percent light formulae must be adjusted with the
tidal range to reflect the light conditions at the sedi-
ment surface at those depths more accurately. This
essentially changes the tidal elevation reference to the
mean tidal level (MTL) instead of MLLW. However,
NOAA does not provide contours referenced to MTL,
since they would be useless for navigation. If the user
of these requirements is not interested in predicting
water quality conditions needed to restore SAV to a
particular restoration depth, they could ignore the
tidal range adjustment. For this reason, any users of
these percent-light parameters should state in their
methods section what Z value(s) were used, and
whether any adjustment to Z was made for tidal range.
Before Z can be adjusted for tidal range, value (s) for
Z must be chosen. In Batiuk et al. (1992), Z = 1 meter
was used for the 1-meter restoration requirements and
all the Kd calculations, and Z = 2 meters was used only
for the 2-meter restoration requirement (Kd < 0.8 m"1),
(Table 1, page iii, and Table I V-l, page 27). SAV
growth to waters 1-meter deep was thought possible
under current light conditions, while growth to the 2-
meter depth was judged likely only under greatly
improved light conditions, thought to have existed in
the 1950s and before, based on the documented
deeper growth of SAV in Chesapeake Bay in these
time frames.
In this chapter, Z was set to 2,1, 0.5, 0.25 and 0 meters
MLLW, before adjusting it for the tidal range. Some
analyses used only some of these depths. The Z value
was varied to as low as 0 meters (representing
the intertidal zone) because in some segments,
-------
102 SAV TECHNICAL SYNTHESIS II
particularly the upper tidal Patuxent River, SAV cur-
rently grows intertidally. Intertidal growth enables
SAV to grow in areas where light conditions are cur-
rently too poor to allow SAV growth in deeper water
(see below). In other estuaries, Z could be set to any
other value where there are different light or tidal con-
ditions or different restoration targets.
Accounting for the tidal range was done by adding half
of the tidal range to the value used for Z above (see
Appendix D for details on tidal range estimation). It is
recognized that this is not the most accurate method to
estimate the average light over a tidal cycle, but was
selected for simplicity, and any introduced errors are
small. Where there are mixed tides as in most of
Chesapeake Bay, the tidal range needed is the "diur-
nal" or "greater tropic" range, defined as (MHHW-
MLLW). To show that Z has been varied and has had
half the tidal range added to it, the variable names
used for calculated values of PLW and PLL are:
PLW(2+) and PLL(2+): Z is set to 2 meters and
half the tidal range is added to it;
PLW(1+) and PLL(1+): Z is set to 1 meter and
half the tidal range is added to it;
PLW(0.5+) and PLL(0.5+): Z is set to 0.5 meters
and half the tidal range is added to it;
PLW(0.25+) and PLL(0.25+): Z is set to 0.25
meters and half the tidal range is added to it; and
PLW(0+) and PLL(0+): Z is set to 0 meters and
half the tidal range is added to it.
Although Kd values could also be adjusted using the
method described here for PLW and PLL, this was not
done to maintain a clear distinction between the 1992
light requirements described here (using Kd assuming
Z = 1 meter and no tidal adjustment) and the minimum
and water-column light requirements described here
(using PLL or PLW respectively, to test their attainment,
with a range of Z values and a tidal range adjustment).
All PLL or PLW values are compared to the same
respective light requirements (listed in Table VII-1)
regardless of which depth Z is used and whether or not
they have been adjusted to account for tidal range.
That is because the light requirements represent how
much light is needed by the plants, regardless of where
they are growing. The adjustments to Z to account for
tidal range are done to better predict how deep SAV
are likely to be able to grow under current light condi-
tions averaged over tidal depth.
Changing the Z values compared to what was used in
Batiuk et al. (1992) raises the question of what Z value
with tidal range adjustment is most comparable to the
Z = 1 meter with no tidal range adjustment estab-
lished in the original SAV technical synthesis in 1992.
Since half the tidal range is close to or above
0.5 meters in many segments (see Appendix D, Table
D-4), setting Z to 0.5 meters is roughly equivalent to
analyses using Z = 1 with no adjustment. This
assumption was verified by the analysis of medians
over SAV growth categories (see tables VII-2 and VII-
4). Thus, PLW(0.5 + ) will give medians and attain-
ment results most similar to the equivalent Kd values
using the 1992 requirements, while PLW(1+) will give
a more accurate estimate of the light reaching the 1-
meter MLLW depth contour, or a leaf of SAV at that
depth contour.
Recommendations for Applying Percent Light
Variables and Other Habitat Requirements
Calculate the percent-light variables (PLL or PLW)
for a chosen restoration depth Z. If Z is referenced to
a depth contour and thus to a depth below MLLW, add
half the diurnal tidal range to Z. If Z is referenced to
mean tidal level (MTL), do not add half the tidal range
to Z. (This assumes that Kd has already been calcu-
lated, or Secchi depth has been used to estimate Kd;
see above for how to do this.)
Choose the restoration depth(s) Z based on local con-
ditions. In Chesapeake Bay we have targets to restore
SAV to shallow water areas to 1- and 2-meter depths
below MLLW (see Chapter VIII), but in some cases
we have set Z to shallower depths (see above).
Use (or collect) Kd data rather than Secchi depth if
both are available. Kd is based on the light wave-
lengths used in photosynthesis and Secchi depth uses
visual light. Also, Kd can be measured more accurately
in clear, shallow waters where the Secchi disk is still
visible on the bottom.
Calculate PLL whenever the data needed for it are
available (Kd or Secchi depth, total suspended solids,
dissolved inorganic nitrogen and dissolved inorganic
phosphorus). Since the minimum light requirement
(which is tested with PLL data) is the primary or most
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 103
useful single habitat requirement, PLL data are needed
to test it. Calculate PLW when Kd or Secchi data are
available, but one or more of the other parameters
needed for PLL calculations are not available.
Use water quality measurements from the "surface"
layer (usually 0.1- to 0.5-meters deep, sometimes 1-
meter deep). These measurements, along with Kd or
Secchi depth data, should come from water quality
monitoring station(s) as close as possible to the poten-
tial habitat or actual SAV bed of interest.
Collect water quality data for calculating parameters
used to test minimum light requirement attainment
during the SAV growing season only. In Chesapeake
Bay this is April-October (7 months) for tidal fresh,
oligohaline and mesohaline areas. In polyhaline areas,
where eelgrass (Zostera marina) is dominant,
the growing season is March-May and September-
November (6 months), due to summer dieback of eel-
grass. Sampling frequency should be at least monthly.
If there are missing data, no tests of attainment should
be done unless there are data from at least four differ-
ent months during the SAV growing season.
Where eelgrass is present in mesohaline areas, or
extensive widgeongrass (R. maritimd) is present in
polyhaline areas, it may be informative to calculate
medians using both growing seasons. In other estuar-
ies, the local growing season should be determined for
each of the dominant species from SAV biomass and
water temperature measurements over at least one
year.
Test the attainment of PLL values by comparing their
growing season medians to the minimum light
requirement for the applicable salinity regime (9 per-
cent or 15 percent; see Table VII-1). Attainment of
PLW values is tested by comparing them to the water-
column light requirement for the applicable salinity
regime (13 percent or 22 percent; see Table VII-1).
Test the attainment of the minimum light requirement
or the water-column light requirement in one of two
ways: by calculating PLL or PLW medians and deter-
mining if they are greater than or less than the mini-
mum light requirement or water-column light
requirement, respectively, or by a nonparametric
statistical test. If software to perform the statistical
test is available, it should be used, rather than the
median comparison, since it uses more of the informa-
tion in the data. The two methods are:
1. To compare median PLL or PLW values over the
SAV growing season to the minimum light
requirement or the water-column light require-
ment. If the medians are greater than the mini-
mum light requirement or water-column light
requirement, the requirement is "Met"; if the
median is less than or equal to the minimum
light requirement or water-column light require-
ment, the requirement is "Not Met."
2. To perform a statistical test by calculating the
difference between the individual measurements
of the percent light parameter used, PLL or
PLW, and the appropriate minimum light
requirement or water-column light requirement,
respectively, and running a nonparametric sign
test on the difference variable. This tests the null
hypothesis that the difference is zero, or that
there was no difference between the measured
data, PLL or PLW, and the minimum light
requirement or water-column light requirement,
respectively. See Appendix D for details; the
three outcomes are called "Met," "Borderline"
or "Not Met."
EVALUATING MINIMUM LIGHT REQUIREMENTS
USING CHESAPEAKE BAY WATER QUALITY
MONITORING DATA AND SAV SURVEY DATA
The next four sections of this chapter use Chesapeake
Bay water quality monitoring data and SAV distribu-
tion data to evaluate the minimum light requirements
that were set above and to see how useful the percent
light at the leaf parameter is in testing their attainment.
These evaluations are not attempting to test the model
used to develop the percent light at the leaf calcula-
tions, since most of the monitoring data (both for water
quality and SAV area) available were collected over too
broad a spatial and temporal scale to be used for that
purpose. The goal of the next four sections was to see
how well the results of analyses of baywide monitoring
data fit with expectations in four different areas:
1. Was there better median water quality where
there was more SAV growth, and worse median
water quality where there was less or no SAV
growth?
2. Are there segments where the minimum light
requirements consistently failed, that had SAV
-------
104 SAV TECHNICAL SYNTHESIS I
growing in them? If so, can we determine rea-
sons for this apparent paradox?
3. Were the percent light parameters and other
SAV habitat requirements significantly corre-
lated with measures of SAV area by Chesapeake
Bay Program segment and year, regardless of the
depth at which it was growing?
4. Were the percent light parameters and other
SAV habitat requirements significantly corre-
lated with measures of SAV area by Chesapeake
Bay Program segment and year over four depth
categories?
Where there are discrepancies, these may be produc-
tive areas for future research, which usually would
involve more detailed monitoring of both water qual-
ity and SAV abundance. There are many other similar
questions that could be asked. The water quality and
SAV area data are available through the Chesapeake
Bay Program and Virginia Institute of Marine
Science web sites, (www.chesapeakebay.net and
www.vims.edu/bio/sav, respectively) for researchers
and managers who want to explore other questions.
The two key fields linking the two data sets are Chesa-
peake Bay Program segment and year. Data for both
water quality and SAV area are available for each year
from 1985 onward, except for 1988, when the SAV sur-
vey was not conducted due to budget constraints.
There are 78 Chesapeake Bay Program segments, but
only 69 of them have water quality data available.
These segments vary in overall size and in the extent of
shallow water habitat. For that reason, in many of the
analyses described below, the SAV area measured in a
segment was divided by the extent of shallow water
habitat in that segment, measured as the "Tier II" area
(see Chapter VIII for a definition). Most of the water
quality data were measured at midchannel stations,
often in fairly deep water, which may limit their use-
fulness in these types of analyses (see Chapter IX).
COMPARING WATER QUALITY MEDIANS OVER
CATEGORIES OF SAV GROWTH
For this section, medians of nearshore water quality
data (from the Choptank and York rivers) and Chesa-
peake Bay Monitoring Program midchannel data were
assessed for empirical relationships with SAV growth
categories. It was expected that medians would be bet-
ter where there was more SAV growth and worse
where there was less growth. "Better" here means
lower levels for Kd, total suspended solids, chlorophyll
a, dissolved inorganic phosphorus, dissolved inorganic
nitrogen and higher levels (more light) for PLW and
PLL. The point at which the habitat requirements fell
among the medians over the different growth cate-
gories was used as an empirical check on the values
chosen for the habitat requirements. The comparisons
using Chesapeake Bay Program midchannel water
quality data are a less rigorous test of the habitat
requirements than those using nearshore data, since
the Bay water quality monitoring stations may be sev-
eral kilometers or more from shallow water SAV habi-
tat, and Bay water quality monitoring stations were not
placed along gradients of SAV growth. The 1992 SAV
habitat requirements were based on the nearshore
data analyzed here, but have not been checked with
Bay water quality monitoring data using this method.
Methods
Three growth categories were used for analyses using
nearshore water quality data and five growth cate-
gories for analyses using midchannel water quality
data; see Appendix D for definitions of the categories.
The tidal fresh and polyhaline salinity regimes had
segments falling in only four of the five growth
categories.
The ranges of annual seasonal median nearshore
water quality over SAV growth categories were used in
Batiuk et al. (1992) to help determine the habitat
requirements. The authors of that document examined
the maximum medians at monitoring stations near
healthy or fluctuating SAV beds and used those to help
set the habitat requirement (it was not always set at
the maximum). The assumption was that if some SAV
were growing near the station with the maximum
median, then SAV should be able to grow at similar
sites where that median water quality occurs. This
approach was used here as an empirical check or veri-
fication of the 1992 habitat requirements, rather than
as a way to derive the requirements.
For the nearshore data, maxima of the annual growing
season medians by station were used, as in Batiuk et al.
(1992) (or minima for PLL/PLW), while for the
Chesapeake Bay Program midchannel monitoring
data, medians of the annual growing season medians
by segment were used. The midchannel stations are
not as close to SAV beds as the nearshore stations, and
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 105
the SAV growth categories for CBP midchannel data
are based on aerial survey data, not on transplant suc-
cess nearby. The more general nature of these data
argued for using medians instead of maxima. Maxima
could be used for the nearshore data because the sta-
tions were near SAV beds. In the midchannel data, if
maxima are used, they are generally worse than the
habitat requirements, even in segments with SAV. This
is probably because midchannel stations tend to be far-
ther from SAV than the nearshore stations.
Kruskal-Wallis nonparametric ANOVA was used to
compare differences among the median water quality
values over different categories of SAV growth, using
the NPAR1WAY procedure in SAS. This analysis tests
the null hypothesis that all the SAV growth categories
had the same median water quality. Statistically signif-
icant differences (ANOVA P' < 0.05 in the following
tables, shown in bold) show that water quality differed
among segments in the different SAV growth cate-
gories, which is the expected outcome for water qual-
ity parameters that affect SAV growth.
Results and Discussion
Minima of annual growing season medians of nearshore
monitoring data from the Choptank and York rivers
(mesohaline and polyhaline, respectively) for the per-
cent light habitat requirement parameters are shown in
Table VII-2. Maxima of annual growing season medians
for the secondary habitat requirement parameters other
than PLW are shown in Table VII-3. Both tables group
the data using three categories of SAV growth based on
the persistence of SAV transplants near the monitoring
stations. Data from 1986-1989 were used in both tables;
these are the same data that were analyzed by Batiuk et
al. (1992) to set the original SAV habitat requirements
for Chesapeake Bay.
Medians of annual growing season median values from
Chesapeake Bay Program midchannel water quality
monitoring stations by salinity regime for the percent
light habitat requirement parameters are shown in
Table VII-4. Medians of annual growing season
median values for the secondary habitat requirement
parameters other than PLW from the same stations
TABLE VII-2. Minima of annual SAV growing season medians of percent-light parameters from Choptank
and York River nearshore monitoring stations by salinity regime and nearby SAV growth category, com-
pared to the minimum light requirement and water column light requirement values shown, and Kruskal-
Wallis ANOVA P for differences among categories, using 1986-1989 data and adding half the tidal range
to the restoration depth Z value listed for the percent light through water (PLW) and percent light at the
leaf (PLL).
Salinity
Mesohaline
(Choptank)
ANOVA P
Polyhaline
(York)
ANOVA P
SAV
Growth
Persistent
Fluctuating
None
Persistent
Fluctuating
None
PLW
3.2%
1.5%
0.0%
0.0001
5.2%
3.5%
0.5%
0.0001
PLW
(1+)
WCLR
14.5%
9.3%
0.4%
0.0001
WCLR
17.8%
14.0%
4.1%
0.0001
PLW
= 22%
31.1%
22 9%
3.3%
0.0001
= 22%
32.3%
oc oo/
t f fl
12.2%
0.0001
PLW
45.4%
36 1%
9.3%
0.0001
44.7%
40 0%
21.2%
0.0001
PLL
3.0%
1.4%
0.0%
0.0001
4.0%
2.4%
0.4%
0.0001
PLL
MLR
13.5%
7.9%
PLL
= 15%
28.8%
16 1%
0.4% 2.0%
0.0001 0.0001
MLR =15%
12.8%
9.1%
3.1%
0.0001
22.2%
161%
8.2%
0.0001
PLL
42.1%
259%
4.6%
0.0001
30.7%
207%
13.4%
0.0001
N
11
19
18
11
7
7
Note: Double lines separate values above and below the habitat requirement. For all of these parameters, less is
worse (less light). P values from Kruskal-Wallis nonparametric one-way ANOVA showing significant differences
among growth categories (at P < 0.05) are in bold. WCLR = water-column light requirement. MLR = minimum
light requirement. N is the number of station x year combinations in that category.
-------
106 SAV TECHNICAL SYNTHESIS II
TABLE VII-3. Maximum of annual SAV growing season medians compared to secondary habitat require-
ments, other than PLW, from Choptank and York River nearshore monitoring stations by salinity regime
and SAV growth category, and Kruskal-Wallis ANOVA results, using 1986-1989 water quality data for total
suspended solids (TSS), chlorophyll a (CHLA), dissolved inorganic phosphorus (DIP) and dissolved
inorganic nitrogen (DIN).
Salinity
Reefime
Mesohaline
(Choptank)
ANOVA P
Polyhaline
(York)
ANOVA P
SAV TSS CHLA DIP DIN
Growth
Persistent 140 11.5 00055 0167
Fluctuating 22.0 13.2 00136 0.223
None 36.5 38.5 0.0498 1.55
0.0001 0.0038 0.0001 0.01
Persistent HR no 00260 0128
Fluctuating 21.1 134 00369 0.173
None 25.6 20.6 0.0499 0.182
0.0011 0.42 0.0004 0.101
N
11
19
26
11
7
7
Note: Double lines separate values above and below the habitat requirement. For all of these parameters, more is
worse. P values from Kruskal-Wallis nonparametric one-way ANOVA showing significant differences among
growth categories (at P < 0.05) are in bold. N is the number of station x year combinations in that category.
are shown in Table VII-5. Both tables group the data
using five categories of SAV growth based on the
median, minimum and maximum SAV area mapped in
the same CBP segment from 1978-1997 (see Appendix
D for methods). Water quality data from 1985-1998
were used in both tables.
ANOVA significance levels (P values) show that almost
all of the medians of these parameters differed signifi-
cantly among the categories of SAV growth, which was
expected. The exceptions were chlorophyll a and dis-
solved inorganic nitrogen in the York River (Table VII-
3) and total suspended solids and chlorophyll a in
polyhaline areas (Table VII-5). Significant differences
among growth categories were not expected in two of
these four cases because all the medians for total sus-
pended solids and chlorophyll a in polyhaline areas in
Table VII-5 were better than the total suspended solids
and chlorophyll a habitat requirements set in Batiuk et
al. (1992). The polyhaline regime also has fewer seg-
ments and thus had among the smallest sample sizes in
Table VII-5, making it more likely that ANOVA results
will not be significant.
The point at which the established SAV habitat
requirements fell among the medians over growth cat-
egories is shown with a double line under or over those
cells in tables VII-2 through VII-5 and as horizontal
lines for the minimum light requirement in Figure
VII-3. In most salinity regimes, segments with median
water quality better than the habitat requirements had
some SAV, while those with median water quality
worse than the habitat requirements had less or no
SAV. This was the expected pattern, which can also be
seen in Figure VII-3 for PLL wherever at least one bar
crosses the dashed line representing the minimum
light requirements. Note that the y-axis scales are
different in low- and high-salinity regimes in Figure
VII-3. The median of PLL(0.5+) was greater (better)
than the minimum light requirement for the Always
Abundant growth category in all salinity regimes
except oligohaline, but the PLL(1+) median was
greater (better) than the minimum light requirement
for the Always Abundant category in mesohaline and
polyhaline regimes only. The PLL(1+) median was
also near or above the minimum light requirement for
the Sometimes None and Always Some categories in
mesohaline and polyhaline regimes.
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 107
TABLE VII-4. Medians of annual SAV growing season medians of percent light parameters from
Chesapeake Bay Program midchannel water quality monitoring stations, by salinity regime and SAV
growth category compared to the minimum light requirement and water column light requirement values
shown, and Kruskal-Wallis ANOVA P for differences among categories, using 1985-1998 data and adding
half of the tidal range to Z for the percent light through water (PLW) and percent light
at the leaf (PLL) parameters.
Salinity
Regime
Tidal fresh
ANOVA P
Oligohaliiu
ANOVA P
Mesohaline
ANOVA P
Polyhaline
ANOVA P
SAV
Growth
AA
SN
UN
AN
AA
AS
SN
UN
AN
AA
AS
SN
UN
AN
AA
AS
SN
AN
PLW
(2+)
1.4%
0.1%
0.0%
0.1%
0.0001
0.5%
0.3%
0.0%
0.0%
0.0%
0.0001
10.9%
6.4%
3.9%
1.7%
0.1%
0.0001
11.1%
6.7%
5.7%
2.9%
0.0001
PLW
fl+*
WCLR
8.5%
1.7%
0.3%
2.5%
0.0001
WCLR
5.4%
3.3%
0.9%
0.7%
0.5%
0.0001
WCLR
"294%
21.6%
15.9%
10.3%
1.9%
0.0001
_ WCLR
"285%
20.5%
18.4%
12.2%
0.0001
PLW
(O.S+\
= 13%
PLW
rt).25+*
209% 330%
7.5% 1 5 4%
2.2% 5.7%
9.8%
0.0001
= 13%
18 2%
11.1%
5.4%
4.5%
3.3%
0.0001
= 22%
47.8%
39.5%
32.9%
24 9%
8.0%
0.0001
= 22%
44.8%
36.1%
33.3%
25.1%
0.0001
20.1%
0.0001
=33.2%
21.3%
134%
11.1%
9.3%
0.0001
60.9%
53.5%
47.7%
39 9%
16.1%
0.0001
57.4%
47.8%
44.8%
36.1%
0.0001
PLL
(2+)
PLL
n+\
PLL
rt).5+*
PLL
f0.25+*
N
MLR =9%
1.3%
0.1%
0.0%
0.1%
0.0001
7.8%
1.4%
0.3%
1.9%
0.0001
181% 273%
5.6% 111%
1.3% 3.0%
6.6%
0.0001
12.9%
0.0001
14
=53
14
81
MLR =9%
0.4%
0.3%
0.0%
0.0%
0.0%
0.0001
3.2%
2.3%
0.8%
0.6%
0.3%
0.0001
8.5%
7.1%
4.3%
3.8%
2.2%
0.0001
14.1%
11.6%
13
42
101% 56
7.5% 25
5.5%
0.0001
92
MLR =15%
10.1%
5.8%
3.5%
1.5%
0.1%
0.0001
"25.8%
41.3%
184% 327%
14.1% 27.9%
78% 194%
1.5%
0.0001
5.3%
0.0001
53.0%
44.4%
38.8%
307%
10.3%
0.0001
28
125
95
98
47
MLR =15%
««•••«•••••«
9.8%
5.0%
4.4%
2.1%
0.0001
"24 8% 40 1%
13.2% 22.4%
13.1%
8.1%
0.0001
21.9%
15.0%
0.0001
50.9%
29.6%
28.5%
20.5%
0.0001
42
14
14
14
Note: Double lines under a cell separate values above and below the water-column light requirement (WCLR) or
minimum light requirement (MLR) shown. Numbers in parentheses are the restoration depth (Z) to which half
the tidal range was added before calculations. For all parameters, less is worse (less light). P values from
Kruskal-Wallis nonparametric one-way ANOVA showing significant differences among growth categories (at P
< 0.05) are in bold. N is the number of segment x year combinations in that category. AA = always abundant;
AS = always some; SN = sometimes none; UN = usually none; and AN = always none.
-------
108 SAV TECHNICAL SYNTHESIS I
TABLE VII-5. Medians of annual SAV growing season medians of parameters with secondary SAV habitat
requirements other than PLW, from Chesapeake Bay Program midchannel water quality monitoring
stations by salinity regime and SAV growth category, and Kruskal-Wallis ANOVA P for differences among
categories, using 1985-1998 water quality data for total suspended solids (TSS), chlorophyll a (CHLA), dis-
solved inorganic phosphorus (DIP) and dissolved inorganic nitrogen (DIN).
Salinity
Regime
Tidal fresh
ANOVA P
Oligohaline
ANOVA P
Mesohaline
ANOVA P
Polyhaline
ANOVA P
SAV
Growth
AA
SN
UN
AN
AA
AS
SN
UN
AN
AA
AS
SN
UN
AN
AA
AS
SN
AN
TSS CHLA DIP
9 975 8 825 0 006
DIN N
0.9409 14
20 23.8 001 5 06643 53
24.025 19.375 0.0328
17 7.6942 0.02
0.0001 0.0001 0.0001
17 4.65 0.0465
18.5 8175 00139
25 28.65 0005
27.3 17.37 0.0234
32.75 13.03 0.02
0.0001 0.0001 0.0001
7.95 8.1 0.004
10.5 9.15 0.007
11 10 0005
15 15.15 0.01
27 11.89 0.015
0.0001 0.0001 0.0001
10 6.3479 0.003
9.75 5.8613 0.01
1105 71324 00147
11.5 5.95 0.0245
0.50 0.59 0.0001
1.1708 14
0.37 81
0.0001
0.859 13
0.636 42
^0.115 56
0.146 28
0.2255 98
0.0001
0.079 28
0.105 125
0.082 95
0091 98
0.1765 47
0.0001
0.045 42
0.1175 14
0 1403 14
0.21 12
0.0001
Note: Double lines under a cell separate values above and below the respective habitat requirement; see Table
Vn-1 for values. For all parameters, more is worse. P values showing significant differences among growth
categories (at P < 0.05) are in bold. N is the number of segment x year combinations in that category. AA =
always abundant; AS = always some; SN = sometimes none; UN = usually none; and AN = always none.
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 109
CO
inomomomoino
•*'*CO .Q
CD .5 p
'5)S < 3
££ II C
|f 31
~ 03 0) 0)
(0 -* c C
«|i =
««> ®
(O
(D
! r:
s
A
C
8 to "3 F
P s> g-g
E co
'•s"0
w oo E ra
•C o o J2
S.o> co re
£" V n "o
58ii-
O ? CO ~ D)
"5 g ® = 2
o » o w -jg
to 3 Z 05 "d
S T3 >, ? '•*=
DC S TH -I o
8
co Sr T3
II ~ ^
• "
0 c
< E.Q
0 o)
.2 = £
"f
IS
O g
o^
co .<5
-L T3
0
catego
ht requ
e resto
es
an
m
V
m
nd
m
ica
URE
FIG
sea
tive abu
ble mini
PLL indi
-------
110 SAV TECHNICAL SYNTHESIS II
There were a few deviations, however, from the
expected pattern in some segments.
• PLW(2+) and PLL(2+) medians were worse
than their applicable water-column and mini-
mum light requirements, respectively, in all cate-
gories (Table VII-4, Figure VII-3).
• The oligohaline midchannel medians for several
parameters—total suspended solids, PLW(1+),
PLL(1+) and PLL(0.5+)—in tables VII-4 and
VII-5 were worse than the habitat requirements
in all growth categories, even in the one segment
(POTOH) in the highest SAV growth category
(Always Abundant). However, the median for
PLW(0.5+) was better than the water-column
light target in the Always Abundant category, and
thus fit the expected pattern.
• The tidal fresh midchannel medians for
PLW(1+) and PLL(1+) were also worse than the
light requirements and targets in all growth cate-
gories, although the medians for PLW(0.5+) and
PLL(0.5+) were better than the respective water-
column and minimum light requirements in the
Always Abundant category, and thus fit the
expected pattern (Table VII-4).
• In the polyhaline nearshore data the maximum
median for dissolved inorganic phosphorus in the
'Persistent' category, 0.026 mg/1, was just above
the dissolved inorganic phosphorus habitat
equirement, 0.02 mg/1 (Table VII-3).
• In polyhaline segments using the Chesapeake
Bay Program water quality monitoring midchan-
nel data, the total suspended solids and chloro-
phyll a medians were all better (lower) than the
habitat requirements (Table VII-5).
• In the nearshore data, the minima of mesohaline
and polyhaline medians for PLW(1+) and
PLL(1+) were all worse than the light require-
ments (Table VII-2).
Of the six deviations described above, only the last one
appears to require further scrutiny, for the following
reasons:
• In the first deviation, we do not expect restora-
tion for 2 meters to be possible under current
conditions.
• The next two deviations are in tidal fresh and
oligohaline segments, where the match between
habitat requirement attainment and the presence
of SAV is not as close as in higher salinity seg-
ments, due partly to the very shallow depth dis-
tribution of SAV in some low salinity segments.
• The fourth deviation is minor, affecting only one
parameter-dissolved inorganic phosphorus.
• The fifth deviation was found in midchannel
monitoring data, which is often collected rela-
tively far from the SAV beds in polyhaline seg-
ments, given the relatively large size of these
higher salinity segments.
This leaves as a problem to be explained, the sixth
deviation, the fact that all the PLW(1+) and PLL(1+)
medians from nearshore data from the Choptank and
York rivers were worse than the respective light
requirements, including those from stations near "Per-
sistent" SAV beds (Table VII-2). Likely reasons for
these deviations are as follows.
As noted above, half the tidal range is close to or
above 0.5 meters in many segments (0.3 meters and 0.4
meters, respectively, in these Choptank and York seg-
ments), so setting Z to 0.5+ in these analyses (rather
than 1+) is closer to the analyses in Batiuk et al. (1992)
that were used to derive the 1992 Kd requirements, in
which Z = 1 with no tidal range adjustment. For
PLW(0.5+) and PLL(0.5+) in Table VII-2, the mini-
mum and water-column light requirements fell
between the medians for "Fluctuating" and "None"
categories, exactly as would be expected based on what
the requirements mean. Most of the PLW(0.5+) and
PLL(0.5+) medians in tidal fresh and oligohaline seg-
ments were also consistent with expectations (except
PLL(0.5+) in oligohaline segments) (Table VII-4).
In both rivers, the extreme values that set the minima for
PLW and PLL either came from a single station and year,
and the next larger median was consistent with the
requirement or target, or the minima were not that much
lower than the requirement. In the Choptank River
mesohaline data, many of the extreme values came from
a single station and year, Chapel Creek in 1986. Without
these anomalous data, in the "Persistent" category the
minimum PLW(1+) median was 22.1 percent instead of
14.5 percent, and the PLL(1+) median was 21.3 percent
instead of 13.5 percent, both of which are better than the
water-column light requirement and minimum light
requirement for that salinity regime (22 percent and 15
percent, respectively).
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 111
In the York River polyhaline data, the minimum for
PLW(1+) for "Persistent," 17.8 percent, was from a
single station and year (Guinea Marsh in 1987); the
next higher PLW(1+) values, 19.6 percent and 21.3
percent, were closer to the water column light require-
ment of 22 percent. The minimum for PLL(1+) in that
segment, 12.8 percent, was close to the minimum light
requirement for that salinity regime (15 percent); the
two next higher values in the "Persistent" category
were 13.3 percent and 14.1 percent.
Thus, it appears that the minimum and water-column
light requirements do not need to be adjusted to
account for any of the discrepancies noted above,
when comparing them to the medians over growth
categories from midchannel and nearshore water qual-
ity monitoring data.
Another more general comparison was done to check
if segments with better water quality tended to have
more SAV One would generally expect the best water
quality where there was the most SAV, and the worst
water quality where there is the least SAV, in this
sequence: Persistent > Fluctuating > None for
nearshore data, or Always Abundant (AA) > Always
Some (AS) > Sometimes None (SN) > Usually None
(UN) > Always None (AN) for midchannel data.
The maximum, minimum and median levels in tables
VII-2 through VII-5 were usually, but not always,
directly proportional to the amount of SAV present.
This pattern was seen most clearly in the nearshore
data (tables VII-2 and VII-3) and in the mesohaline
and polyhaline segments in tables VII-4 and VII-5.
IDENTIFYING SEGMENTS WITH PERSISTENT
FAILURE OF THE MINIMUM LIGHT
REQUIREMENTS AND CHECKING
THEM FOR SAV GROWTH
Each Chesapeake Bay Program segment with water
quality data was compared to determine whether con-
sistent failure of the minimum light requirement pre-
dicted a lack of SAV growth. This was done by looking
for any Chesapeake Bay Program segments with per-
sistent failure of the minimum light requirement that
contained appreciable amounts of SAV (over 35
hectares). In these segments, possible reasons for the
mismatch were examined. The corresponding analysis,
checking for segments with the minimum light
requirement usually met where there was little or no
SAV, would not be useful because SAV can be lacking
for a variety of reasons (lack of propagules, high wave
action, etc.).
Methods
Chesapeake Bay Program segments with PLL medians
failing the minimum light requirements more than half
of the years from 1992-1997 were tabulated using the
sign test at P = 0.05 and the FREQ procedure in SAS.
The restoration depth (Z) was varied over 1, 0.5, 0.25
and 0 meters plus half the tidal range. Segments were
identified as persistently failing the minimum light
requirements at the lowest (worst) depth if they had
failed half or more of the years. The 1997 SAV area
was checked for each of the segments with persistent
failure at 0.5 meters or less, and the segment was
flagged as a problem segment if the SAV area was over
35 hectares (86 acres). The expectation was that SAV
would not grow where the minimum light requirement
was failed persistently. This SAV hectare cutoff was
somewhat arbitrary, but was chosen to leave out some
small segments (such as the Northeast and Bohemia
rivers) that contain very small amounts of SAV,
sometimes a single bed. For each of the problem
segments, possible reasons for the presence of SAV
were examined.
Results and Discussion
Figure VII-4 shows the Chesapeake Bay Program seg-
ments with median PLL values failing the minimum
light requirement at different Z values half of the
years or more between 1992 and 1997. Those segments
failing at Z = 0.5 meters or less are identified by name.
Of that group, there were only two problem segments,
Patuxent tidal fresh and oligohaline. Both segments
failed the minimum light requirement at Z = 0.25
meters plus half the tidal range and had more than
35 hectares of SAV in 1997. It is likely that they con-
tain SAV in spite of the poor light conditions because
most of the SAV present in both segments is growing
in very shallow water, including some growing in the
intertidal zone.
COMPARING DIFFERENT SAV HABITAT
REQUIREMENTS AS PREDICTORS
OF SAV AREA
Correlations of annual estimated SAV area with annual
median water quality were used to test the ability of
each of the 1992 SAV habitat requirements and the two
-------
112 SAV TECHNICAL SYNTHESIS II
Northeast (C),
w
Bohemia (B)
(Q
Potomac
Tidal Fresh (C),
Oligohaline(C)
Gunpowder (
Back(B)
Chester
Oligohaline(A)
Choptank
.Oligohaline(B)
Patuxent
Tidal Fresh (B),
Oligohaline(B)
Nanbcoke
Tidal Fresh (C)
Mesohaline(B)
Wicomico
MesohalinefC)
Mattaponi
Oigohaline(C
Pamunkey
Qigohaline(B) -"^-
York
Mesohaline(B)~
Pocomoke
TWal Fresh (C)
FAILED AT LEAST HALF OF YEARS
A: Failed at 0.0 m MLLW + half TR
B: Failed at 0.25 m + half TR
C: Failed at 0.5 m + half TR
D: Failed at 1.0 m + half TR
E: Not Failed at any depth
East Branch Elizabeth (C)
South Branch Elizabeth (B)
FIGURE VII-4. Segments Failing PLL Requirements Half of the Years of More, 1992-97. Chesapeake Bay
Program segments with SAV growing season median PLL values failing the minimum light requirement half the
years or more between 1982 and 1997. Only the Chesapeake Bay Program segments in categores A, B, and C
are labeled. TR = diurnal tidal range. Segment names in bold indicate both the Patuxent tidal fresh and oligohaline
segments had PLL failed at 0.25 meter + half TR (category B) or less, AND appreciable amounts (>35 ha) of SAV.
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 113
percent-light parameters to predict SAV area. This abil-
ity is a desirable feature of any SAV habitat require-
ments, since it justifies using the habitat requirements
to set nutrient and sediment reduction targets and to
target the best areas for SAV restoration. However, any
lack of significant correlations in this analysis does not
invalidate the model used to derive PLL. Most of the
data used were collected in large-scale monitoring pro-
grams, and not over the small spatial and temporal
scales needed for a research project.
In this analysis it was expected that light parameters
(Kd, PLW and PLL) would have stronger and more
significant correlations with SAV area than other habi-
tat requirement parameters, since ecological models
show that light is the primary factor limiting SAV dis-
tribution in Chesapeake Bay (Batiuk et al. 1992; chap-
ters III and V).
Methods
Correlations were calculated with the SAS procedure
CORK between SAV area in hectares, by year, and
median water quality from the Chesapeake Bay Pro-
gram water quality monitoring stations, by year.
Nearshore water quality data were not used because
they do not have associated SAV area data. Spearman
rank correlation (nonparametric) was used rather than
Pearson (parametric) correlation, because SAV area is
not normally distributed, even with transformations,
due largely to the large number of zeroes. The zeroes
make SAV area a difficult variable to use in correla-
tions, even Spearman correlations. Another problem
is that in many salinity regimes most or all of the seg-
ments and years with high SAV area are in only one or
two segments, so the results may reflect conditions in
those segments rather than in the all the segments in
the salinity regime. In Spearman correlations, if the
two data sets are ranked exactly the same way, rs =
+ 1; if they have opposite ranks, rs = -1, and 0 means
no association. For example, to be ranked the same
way, the segment and year with the highest PLL
median would have the highest SAV area; the second
highest PLL would have the second-highest SAV area;
and so on. When there are ties (as with the many
zeroes) they all receive the same rank, making it
harder to find a significant correlation. When dis-
cussing correlations the terms "stronger" and
"weaker" are used to mean "larger (or smaller)
absolute value of rs" since for some water quality
parameters the expected correlations with SAV area
are positive (PLW and PLL, more light means more
SAV), and for others they are negative (all other
parameters, more pollution means less SAV).
The spatial units used for this analysis were the 69
Chesapeake Bay Program segments that have water
quality data, grouped into salinity regimes. The time
periods over which water quality medians were calcu-
lated were the SAV growing season (April-October
except in polyhaline where it is March-May and Sep-
tember-November), or in spring (April-June, except
March-May in polyhaline segments). Spring data were
tested separately to see if spring water quality was
more important than water quality over the whole
growing season. After spring, many species are grow-
ing at or near the surface, and thus their survival and
growth might be less sensitive to water quality condi-
tions in the summer and fall.
Three different measures were used for SAV area
measured in the same year as the water quality data:
SAV hectares (SAVH), SAVH as a percent of Tier II
area (PCT_T2), and SAVH as a percent of Tier III
area (PCT_T3). The latter two measures were calcu-
lated to correct for the differing amounts of potential
SAV habitat in different CBP segments, which differ
greatly in size.
Lagged effects were tested two ways, by replacing SAV
area from the current year (SAVH) with SAV area
from the following year (LAGSAVH) or replacing it
with the change in SAV area from this year to the fol-
lowing year (CHGSAVH). The expected correlations
were negative for Kd, total suspended solids, chloro-
phyll a, dissolved inorganic phosphorus and dissolved
inorganic nitrogen, since for these parameters higher
levels should lead to a reduction in SAV area, and pos-
itive for PLW and PLL. Both Kd and PLW were
included because PLW calculations included the tidal
range adjustment for Z, while those for Kd did not.
Results and Discussion
Results are summarized in Table VII-6, and complete
results are given in Appendix E, tables E-l, 3, 5 and
7 over the whole growing season and in Appendix E,
tables E-2, 4, 6 and 8 over the spring portion of the
growing season. These correlations show the following
over the whole SAV growing season:
-------
114 SAV TECHNICAL SYNTHESIS I
TABLE VII-6. Salinity regimes that had statistically significant (P < 0.05) Spearman rank correlation
coefficients in expected directions, between water quality parameters from Chesapeake Bay Program
midchannel water quality stations and measures of SAV area over Chesapeake Bay Program segments
(see Appendix E, tables E-1 through E-8 for correlations, P values and sample sizes).
Water Quality
Parameter
Ka
PLW(1+)
PLL(1+)
TSS
CHLA
DIP
DIN
PCT_
OH+
MH+
PH+
OH+
MH+
PH+
OH+
MH+
PH+
TF
OH+
MH+
MH+
OH#
MH+
PH+
MH#
PH
SAV Area
T2 PCT_
OH+
MH+
PH+
OH+
MH+
PH+
OH+
MH+
PH+
TF
OH+
MH+
MH+
TF+
OH#
MH+
PH+
MH+
PH
Parameter
T3 SAVH
OH+
MH+
PH+
OH+
MH+
PH+
OH+
MH+
PH+
OH+
MH+
OH
MH+
TF+
OH#
MH+
PH+
MH+
PH+
LAGSAVH CHGSAVH
OH+
MH+
PH+
OH+
MH+
PH+
OH+
MH+ MH
PH+
TF
OH+
MH+
OH
MH+
PH#
TF+
OH#
MH+
PH+
MH+
PH+ PH+
KEY. PCT_T2 = SAVH/Tier H area* 100, PCT_T3 = SAVH/Tier m area* 100, SAVH=SAV hectares for same year as
water quality data; LAGSAVH=SAV hectares for following year; CHGSAVH=change in SAV hectares from that year to
next. TF = tidal fresh, OH = oligohaline, MH = mesohaline, and PH= polyhaline (see Table VII-1 for saUnities).
Light attenuation coefficient = Kj; percent light through water = PLW; percent light at the leaf = PLL; total suspended
solids = TSS; chlorophyll a = CHLA; dissolved inorganic phosphorus = DIP; and dissolved inorganic nitrogen = DIN.
Regimes in bold had correlations >+/- 0.5 over the whole growing season; some were > 0.5 over the spring also.
+ Also had significant correlation over spring part of growing season (if no symbol, significant over whole year).
# Significant correlation over spring part of growing season only (April-June except March-May in polyhaline).
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 115
In oligohaline, mesohaline and polyhaline segments,
four of the five SAV area parameters (all but
CHGSAVH) showed the expected statistically signifi-
cant correlations (P < 0.05) with the three light
parameters: more light meant more SAV (negative
correlations for Kd, positive for PLW and PLL). The rs
values were near 0.35 for oligohaline and 0.55 for
mesohaline and polyhaline segments.
Most of the other habitat requirement parameters
tested had weaker but significant correlations in these
three salinity regimes, but the correlations were not
significant for dissolved inorganic nitrogen in oligoha-
line or total suspended solids and chlorophyll a in
polyhaline segments (except that chlorophyll a had a
significant correlation with LAGSAVH in the spring).
Oligohaline dissolved inorganic nitrogen was not
expected to correlate with SAV area because there is
no habitat requirement for dissolved inorganic nitro-
gen in tidal fresh and oligohaline segments.
Tidal fresh segments showed weaker correlations but
they were statistically significant for most area param-
eters for total suspended solids and dissolved inor-
ganic phosphorus.
The expected significant relationships were stronger
for PLL than for PLW in polyhaline segments, and
almost the same in other segments where they were
significant (oligohaline and mesohaline segments).
Correlations for Kd differed from those for PLW
because Kd was not adjusted for tidal range.
The three different measures of SAV area in the same
year as the water quality data (PCT_T2, PCT_T3, and
SAVH) usually had similar, significant correlations
with water quality variables over the whole growing
season, with a few exceptions. In tidal fresh segments,
SAVH did not have significant correlations with total
suspended solids while the other measures did, and
PCT_T2 lacked significant correlations with dissolved
inorganic phosphorus (Table VII-6). In oligohaline
segments, PCT_T2 and PCT_T3 lacked any significant
correlations with chlorophyll a, while these were found
for SAVH. In all segments with significant correla-
tions, the correlations were slightly stronger with
SAVH than with the other two parameters. Thus,
based on the correlations, there did not seem to be a
compelling reason to correct the SAVH variable with
the Tier II or Tier III areas, since correlations with
SAVH were slightly stronger (Appendix E).
Lagged SAV hectares (the area mapped the following
year, LAGSAVH) showed similar correlations with the
SAV habitat requirements as were found for SAV area
mapped the same year, except for chlorophyll a in
polyhaline habitats, which was significantly correlated
with lagged SAV area but not correlated with SAV
area the same year. Most of the significant polyhaline
correlations with SAV habitat requirements were
slightly higher for lagged SAV area (except for dis-
solved inorganic phosphorus), compared with SAV
area mapped the same year. This tends to support the
hypothesis that water quality in the current year affects
SAV area in the next year, but the effect was not a
dramatic one. Thus, unlagged SAV area (SAVH)
seems adequate for most correlative analyses.
Change in SAV hectares (the change from the area
this year to the area next year) did not have any signif-
icant expected correlations with water quality except
for PLL(1 + ) in mesohaline segments and dissolved
inorganic nitrogen in polyhaline segments.
Comparisons of significant correlations with water
quality over the whole growing season (Appendix E,
tables E-l, 3, 5 and 7) to significant correlations with
spring water quality (Appendix E, tables E-2, 4, 6 and
8) showed they differed in the following cases.
Spring correlations were weaker in tidal fresh seg-
ments compared to correlations over the whole grow-
ing season for total suspended solids and dissolved
inorganic phosphorus (the only two parameters that
had significant correlations over the whole year in tidal
fresh segments). Spring correlations were stronger in
oligohaline segments for Kd, PLW, PLL and dissolved
inorganic phosphorus and weaker for total suspended
solids and chlorophyll a. Whole-year correlations were
stronger or very similar in mesohaline segments for all
but one of the parameters with significant correlations
(Kd, PLW, PLL, chlorophyll a and dissolved inorganic
phosphorus). Spring correlations were slightly
stronger for total suspended solids. Spring correlations
were weaker in polyhaline segments for Kd, PLW, PLL,
dissolved inorganic phosphorus and dissolved inor-
ganic nitrogen. Spring correlations for chlorophyll a in
polyhaline segments were barely significant, but were
not significant over the whole growing season.
Spring correlations were weaker in 12 cases, and
stronger in six cases. Thus, there does not seem to be
a compelling reason to use spring water quality in SAV
-------
116 SAV TECHNICAL SYNTHESIS I
habitat requirements in place of water quality over the
whole growing season.
Now, as expected, the light parameters (PLL, PLW
and Kd) usually had the strongest correlation. In oligo-
haline segments, total suspended solids, PLW and PLL
had the strongest correlations with unlagged SAV area
over the whole growing season, while dissolved inor-
ganic phosphorus and total suspended solids had the
strongest correlations in tidal fresh segments. In meso-
haline and polyhaline segments, PLL, PLW and/or Kd
had the strongest correlations with unlagged SAV area
over the whole growing season.
Scatter plots for each salinity regime are graphed in
figures VII-5 through VII-8 for the four salinity
regimes, respectively. PLL(1+) was used on the x-axis
in all segments for consistency. PLL had the strongest
correlation with SAV area in polyhaline segments, and
the third-strongest correlation in oligohaline and
mesohaline segments. PLL did not have significant
correlations with SAV area in tidal fresh segments.
Note the large numbers of segments and years with
zero or very low SAV areas in all salinity regimes,
which make correlation analysis difficult. However,
the general pattern that can be seen is the expected
one: more SAV where PLL is higher.
CORRELATING SAV DEPTH WITH
MEDIAN WATER QUALITY FOR HABITAT
REQUIREMENT PARAMETERS
The same correlations done in the last section using
SAV area and water quality medians were repeated
using SAV depth by year, in place of SAV area by year.
The assumptions in this analysis were that SAV would
grow deeper where water quality was better, and shal-
lower where water quality was worse. In general, cor-
relations with light parameters (Kd, PLW and PLL)
were expected to be stronger than those with other
parameters, since light most directly determines the
depth at which SAV can grow. Correlations between
SAV depth and water quality were expected to be
weaker than those between SAV area and water qual-
ity because SAV depth is only available for segments
that had some SAV that year. This both reduces the
sample size for comparisons with SAV depth and
reduces the range in the associated water quality, since
generally the worst water quality occurs in segments
with no SAV. Both changes reduce the likelihood that
there will be statistically significant correlations
between SAV depth and water quality.
Factors other than water quality also affect the depth
at which SAV grows, however. Some SAV species have
a lower light requirement than others (see Chapter
III), and thus may be able to grow deeper than others,
which could increase the weighted mean SAV depth
where those species are dominant. Also, physical fac-
tors such as current, tides, sediment and wave action
could affect SAV depth distribution (see Chapter VI)
independently of water quality.
Methods
Methods are described here briefly; detailed methods
for calculating SAV depth and percent of SAV within
depth categories are given in Appendix E. SAV poly-
gons for each year were overlaid with depth contours
at 0.5, 1 and 2 meters MLLW. The area of SAV within
each Chesapeake Bay Program segment that fell
within four depth categories was calculated: less than
0.5 meters, 0.5 to 1 meter, 1 to 2 meters or greater than
2 meters deep. For this analysis, the four values of SAV
area over depth ranges were divided by the total area
to convert them to percentages of the total area in that
segment (PCT05, PCT1, PCT2 and PCTGT2) and also
to a single weighted mean depth (SAVDEP). Analysis
methods used were the same as in the previous sec-
tion, except that lagged depth and annual change in
depth were not examined, and correlations with spring
water quality were not examined due to small sample
sizes and relatively weak spring correlations in the
analyses done above.
Results and Discussion
Spearman rank correlations of percentages of SAV in
depth categories and weighted mean SAV depth with
SAV habitat requirement parameters are shown in
Appendix E, tables E-9 through E-12 for the four
salinity regimes. Results are summarized in Table VII-
7. For most parameters (Kd, total suspended solids,
chlorophyll a, dissolved inorganic phosphorus and dis-
solved inorganic nitrogen), negative correlations were
expected because more pollution should yield shal-
lower SAV, and positive correlations were expected for
PLW and PLL, since more light should yield deeper
SAV. Both expectations are reversed (positive for
most, negative for PLW/PLL) for the shallowest depth
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 117
140%
U)
5% 10% 15% 20% 25% 30% 35% 40% 45%
Growing season median PLL(H-) (% light at the leaf)
FIGURE VII-5. Tidal Fresh SAV Area vs. Percent Light
at the Leaf. Tidal fresh SAV area as a percent of the
Tier II SAV distribution restoration target by year and
Chesapeake Bay Program segment vs. median percent
light at the leaf [PLL(1 +)] by year and Chesapeake Bay
Program segment, using 1985-1998 data (no data from
1988). Spearman rs = 0.14, N = 124, P = 0.12.
100% i
0) 90%
|2 80%
jj 70%
^ 60%
°m 50%
10 40%
£
J5 30%
J 20%
5 10%
CO
0% .
0"
•
« .
• •
**
> •
/. 5% 10% 15% 20% 25% 30% 35% 40% 45%
Growing season median PLL(1+) (% light at the leaf)
FIGURE VII-6. Oligohaline SAV Area vs. Percent Light
at the Leaf. Oligohaline SAV area as a percent of the
Tier II SAV distribution restoration target by year and
Chesapeake Bay Program segment vs. median percent
light at the leaf [PLL(1 +)] by year and Chesapeake Bay
Program segment, using 1985-1998 data (no data from
1988). Spearman rs = 0.37, N = 182, P = 0.0001.
100% -,
0 90%
S>
£ 80% -
= 70% -
a
P 60%
« 50%
| 40%
3 30% -
a
•= 20% -
< 10%
0% -
0'
•
^«X *
\ » •• •
*^4 **»**^«»»* 1 t**t»»**»* *4^
*r** * -*-• ** "ftfftfvdt'ittitiiif* * «
H 5% 10% 15% 20% 25% 30% 35% 40% 45%
Growing season median PLL(1 +) (% light at the leaf)
FIGURE VII-7. Mesohaline SAV Area vs. Percent Light
at the Leaf Mesohaline SAV area as a percent of the
Tier II SAV distribution restoration target by year and
Chesapeake Bay Program segment vs. median percent
light at the leaf [PLL(1 +)] by year and Chesapeake Bay
Program segment, using 1985-1998 data (no data from
1988). Spearman rs = 0.51, N = 326, P = 0.0001.
100°'
0> 90%
ff
£ 80%
u 70%
o
^ 60%
850%
8 40%
•S 30%
£• 20%
>
< 10%-
0%
» •• • « •
• »
» ti 4, * * *
.* .*•;*•*
* A % *
** •• « «
. ^ „ „» ^ .»AA *.*^
0% 5% 10% 15% 20% 25% 30% 35% 40% 45%
Growing season median PLL(1+) (% light at the leaf)
FIGURE VI1-8. Polyhaline SAV Area vs. Percent Light
at the Leaf. Polyhaline SAV area as a percent of the
Tier II SAV distribution restoration target by year and
Chesapeake Bay Program segment vs. median percent
light at the leaf [PLL(1 +)] by year and Chesapeake Bay
Program segment, using 1985-1998 data (no data from
1988). Spearman rs = 0.50, N = 60, P = 0.0001.
-------
118 SAV TECHNICAL SYNTHESIS II
TABLE VII-7. Salinity regimes that had statistically significant (P < 0.05) Spearman rank correlation
coefficients in expected directions, between water quality parameters from Chesapeake Bay Program
mid-channel water quality stations and measures of SAV depth over Chesapeake Bay Program segments
(see Appendix E, tables E-9 through E-12 for correlations, P values and sample sizes), using data from
whole growing season.
Water quality
naraineter
K,
PLW(1+)
PLL(1+)
TSS
CHLA
DIP
DIN
SAV depth
SAVDEP PCT05
TF
OH
PH
TF
OH
PH
TF
OH
PH
TF
OH
OH
MH
MH
PH
MH
PH
TF
OH
PH
TF
OH
PH
TF
OH
PH
TF
OH
TF
OH
MH
MH
PH
MH
PH
parameter
PCT1
TF
OH
PH
TF
OH
PH
TF
OH
PH
TF
OH
TF
OH
MH
MH
PH
MH
PH
PCT2
TF
OH
MH
TF
OH
MH
TF
OH
MH
TF
OH
OH
MH
MH
PH
MH
PH
PCTGT2
TF
OH
MH
PH
TF
OH
MH
TF
OH
MH
PH
TF
OH
OH
PH
MH
PH
KEY: SAV area that fell within four depth categories: less than 0.5 meters, 0.5 to 1 meter, 1 to 2 meters or greater than 2
meters deep, was mapped, and each area was divided by the total area over all four ranges to convert them to percentages of
the total area in that segment (PCT05, PCT1, PCT2 and PCTGT2, respectively) and also to a single weighted mean depth
(SAVDEP).
TF = tidal fresh, OH = oligohaline, MH = mesohaline, and PH= polyhaline (see Table VB-1 for salinities).
Light attenuation coefficient = K,,; percent light through water = PLW; percent light at the leaf = PLL; total suspended
solids = TSS; chlorophyll a = CHLA; dissolved inorganic phosphorus = DP; and dissolved inorganic nitrogen = DIN.
Regimes in bold had correlations >+/- 0.5 over the whole growing season.
-------
Chapter VII - Setting, Applying and Evaluating Minimum Light Requirements for Chesapeake Bay SAV 119
category, PCT05, since SAV is expected to be more
common in the shallowest depths when conditions are
worse because it cannot grow in deeper water.
Table VII-7 shows that, as expected, most of the light
parameters showed significant correlations with SAV
depth parameters in the expected directions, except
with some depth parameters in mesohaline and polyha-
line areas. In mesohaline areas, significant correlations
were found with the deeper percentages only (PCT2
and PCTGT2), and in polyhaline areas, significant cor-
relations were found with the shallower depths and the
weighted mean depth only (SAVDEP, PCT05, and
PCT1). Reasons for these differences are not known.
In examining correlations with the other water quality
parameters two patterns were seen. Correlations with
total suspended solids and chlorophyll a were stronger
in tidal fresh and oligohaline segments, while correla-
tions with nutrients were stronger in mesohaline and
polyhaline segments. Most of the correlations with
total suspended solids and chlorophyll a were signifi-
cant and in the expected directions in tidal fresh and
oligohaline segments, suggesting that they may affect
depth distributions along with light in lower salinity
areas. In mesohaline segments, only chlorophyll a had
significant correlations in the expected direction, and
in polyhaline segments, there were no significant cor-
relations with total suspended solids or chlorophyll a.
In tidal fresh and oligohaline areas, all of the correla-
tions between nutrients (dissolved inorganic phospho-
rus and dissolved inorganic nitrogen) and depth
parameters were either significant but in the wrong
direction, or not significant. This was expected for dis-
solved inorganic nitrogen, since there is no habitat
requirement for dissolved inorganic nitrogen in the
two lower salinity regimes. In mesohaline and tidal
fresh segments, all but one of the correlations with dis-
solved inorganic phosphorus and dissolved inorganic
nitrogen were significant and in the expected direc-
tions.
CONCLUSIONS
Comparisons of SAV area or depth to water quality
should always be done for salinity regimes separately.
None of the detailed relationships were consistent
across all salinity regimes.
In general, segments with median water quality better
than the SAV habitat requirements had some SAV,
while those with medians worse than the requirements
had less or no SAV. SAV also tended to grow at deeper
depths where water quality was better, and at shal-
lower depths where water quality was worse. This
provides empirical confirmation of the light require-
ments that were determined from research and
ecological modeling.
Segments that failed the minimum light requirement
in half of the past six years or more were identified and
their SAV area checked. There were two segments
with more than 35 hectares of SAV in 1997 that failed
PLL at Z = 0.25 plus half the tidal range or less. In
both cases, there were reasons why SAV could be
growing there even though monitoring data showed
the minimum light requirement was usually failed.
In the polyhaline regime, PLL was a better predictor
of SAV area and SAV depth than PLW, when there was
a significant relationship with SAV area. In other salin-
ity regimes, PLL and PLW were very similar as pre-
dictors, except in oligohaline segments, where PLW
was a slightly better predictor than PLL of SAV area
and depth.
PLL or PLW were often, but not always, the strongest
predictors of SAV area among all the SAV habitat
requirements. However, given the highly skewed dis-
tribution of SAV area data and differences in percent
light levels, these results are not really a test of the use-
fulness of these parameters.
In some cases, in all four salinity regimes, water qual-
ity showed slightly stronger correlations with the SAV
area mapped in the following year, compared to cor-
relations with SAV area in the current year. However,
the improvement with lagged SAV area did not appear
to be consistent enough or large enough to warrant
using the latter when calculating correlations with
water quality data, especially since lagging drops a year
off the sample size.
Spring median water quality did not appear to be con-
sistently better than growing season median water
quality in predicting SAV area, and in many cases
(especially in polyhaline segments) it was a worse pre-
dictor. Thus, we recommend using water quality data
over the whole growing season to assess attainment of
the SAV habitat requirements.
One or more of the three light parameters (Kd, PLW
and/or PLL) usually were the best predictors of SAV
-------
120 SAV TECHNICAL SYNTHESIS II
depth over most depth categories. In oligohaline seg-
ments, chlorophyll a had slightly stronger correlations
in three depth categories, and in polyhaline segments,
dissolved inorganic nitrogen and dissolved inorganic
phosphorus had stronger correlations than light with
the percent of SAV area in the two deeper categories.
Dissolved inorganic nitrogen and dissolved inorganic
phosphorus were also the strongest predictors of SAV
area in polyhaline segments.
Chesapeake Bay Program midchannel water quality
monitoring data often show the expected patterns in
analyses that compare SAV area and depth with water
quality, even though the stations are not located next
to SAV beds. This supports the continued use of these
data to assess attainment of SAV habitat require-
ments. However, care must be taken to omit data from
some stations where the SAV is in a different water
body from the monitoring stations (such as Little
Creek in segment CB8PH). In these cases, additional
water quality monitoring data from sites near the SAV
beds are needed.
-------
CHAPTER
Chesapeake Bay SAV Distribution
Restoration Goals and Targets
"T~he original tiered SAV distribution restoration targets
I for Chesapeake Bay were first published in the 1992
SAV technical synthesis in response to commitments set
forth in the Submerged Aquatic Vegetation Policy for the
Chesapeake Bay and Tidal Tributaries (Chesapeake Exec-
utive Council 1989). The Tier I SAV distribution restora-
tion target is the restoration of SAV to areas currently or
previously inhabited by SAV as mapped through regional
and baywide aerial surveys from 1971 through 1990
(Batiuk et al. 1992; Dennison et al. 1993). The Tier II and
Tier III distribution restoration targets are the restora-
tion of SAV to all shallow water areas delineated as exist-
ing or potential SAV habitat, down to the 1- and 2-meter
depth contours, respectively.
Baywide and Chesapeake Bay Program segment-based
target acreages were published in 1992 for the Tier I
and Tier III restoration targets (Batiuk et al. 1992). The
lack of sufficient Bay bottom bathymetry data to create
a 1-meter depth contour prevented delineation of the
Tier II restoration target at that time. In 1993 the
Chesapeake Executive Council formally adopted the
Tier I SAV distribution restoration target as the Chesa-
peake Bay Program's first quantitative living resource
restoration goal (Chesapeake Executive Council 1993).
The refined baywide and regional SAV distribution
restoration goals and targets presented here are critical
in assessing the success of efforts to restore SAV in
Chesapeake Bay and its tidal tributaries.
DISTRIBUTION TARGETS DEVELOPMENT
APPROACH
The tiered Chesapeake Bay SAV distribution restor-
ation targets were originally developed in 1992 by
mapping potential SAV habitat on U.S. Geological
Survey (USGS) quadrangles; removing shallow water
habitat areas where SAV were not expected to revege-
tate; and comparing these areas with historical survey
data and the most current distribution data. Compos-
ite SAV maps were plotted by USGS quadrangles from
all available computerized digital SAV bed data from
Chesapeake Bay aerial surveys from 1971 through
1990. The 1- and 2-meter depth contours at mean low
water (MLW) were digitized from National Oceanic
Atmospheric Administration (NOAA) bathymetry
maps. Because the NOAA bathymetry maps are rela-
tively inaccurate in small tidal creeks and rivers where
depth contours generally were not present, an overes-
timate of an area within a certain depth contour can
occur. These maps were overlaid at the 1:24,000 scale
to produce composite maps of known and documented
SAV distribution since the early 1970s, with the outline
of potential SAV habitat initially defined by the 1- and
2-meter depth contours.
Potential habitat was initially defined as all shoal areas
of Chesapeake Bay and tributaries less than 2 meters.
Although historically SAV in Chesapeake Bay probably
grew down to depths of 3 meters or more, the 2-meter
depth contour was chosen because it was the best com-
promise of the anticipated maximum depth penetration
of most SAV species. For several SAV species (notably
Myriophyllum spicatum and Hydrilla verticillata), maxi-
mum depth penetration might be greater than 2 meters,
but it was felt that this would be an exception.
Areas that were highly unlikely to support SAV were
annotated on the composite maps. Criteria for exclud-
ing certain areas from the maps were based primarily
on habitat areas exposed to high wave energy and that
have undergone physical modifications that prevented
them from supporting SAV growth. The absence of
documentation on the historical presence of SAV in a
Chapter VIII - Chesapeake Bay SAV Distribution Restoration Goals and Targets 121
-------
122 SAV TECHNICAL SYNTHESIS II
certain region of a tributary, embayment or the main-
stem was not used as a reason to delineate and exclude
the shallow water habitats in these regions as unlikely
to support future SAV growth. For example, some
areas that have not supported SAV in the recent past
(such as the tidal fresh and oligohaline areas of the
James, York and Rappahannock rivers) were included
in the distribution restoration targets. This distinction
was based on the following assumption: since the
upper Potomac River near Washington, D.C. sup-
ported these stands of SAV in the early 1900s (Gum-
ming et al. 1916), there should be no reason to assume
that SAV was not present in similar areas in the tidal
fresh and oligohaline reaches of other river systems in
Chesapeake Bay. The anecdotal evidence from dis-
parate regions of the Bay, as well as aerial photo-
graphic evidence for some areas in the 1930s, indicate
the major areas where SAV grew in the early part of
the 20th century. In addition, many small tidal creeks in
tidal fresh and oligohaline areas throughout the Bay
and its tidal tributaries today contain small pockets of
a variety of SAV species. It is assumed that these are
the last remnants of what were once large expansive
stands in earlier periods in the upper sections of these
tributaries. The seed and pollen records (Brush and
Hilgartner 1989) support the line of evidence that SAV
was once significantly more abundant than it is today.
The areas annotated as highly unlikely to support SAV
were digitized and deleted from the ARC/INFO files
of potential SAV habitat delineated by the 2-meter
depth contour. A second level of habitat restriction
was considered in those areas where SAV was
presently found or had the potential to grow in the 2-
meter contour. This habitat restriction was considered
in areas where wave exposure is highly likely to pre-
vent SAV from growing down 2 meters in depth but
would be dampened enough to allow SAV to grow
closer inshore (less than 1 meter). Assessment of areas
that would fall into this category was based on the
same criteria used to generate the composite maps for
the 2-meter restricted areas. The complete, detailed
description of the original process for developing the
tiered restoration goals and targets is found on pages
109-119 in Batiuk et al. (1992).
TIERED SAV DISTRIBUTION RESTORATION
GOALS AND TARGETS
To provide incremental measures of progress, a tiered
set of SAV distribution restoration targets have been
established for Chesapeake Bay. Each target repre-
sents expansions in SAV distribution that are antici-
pated in response to improvements in water quality.
These water quality improvements will be measured as
achievement of the minimum light requirements at
1- and 2-meter restoration depths. Progress toward the
SAV distribution restoration targets will continue to be
measured through the annual Chesapeake Bay SAV
Aerial Survey Monitoring Program.
Refinements have been made to the Tier I restoration
goal as a result of a reevaluation of the historical SAV
aerial survey digital data sets, including a thorough qual-
ity assurance evaluation, which resulted in corrections to
the original data. The revised Tier I restoration goal
areas are presented by Chesapeake Bay Program seg-
ments in Table VIII-1 and illustrated in Figure VIII-1.
The Tier II SAV distribution restoration target is the
restoration of SAV to all shallow water areas delin-
eated as existing or potential SAV habitat down to the
1-meter depth contour. Building from the recent com-
pletion of a synthesis of all available Bay bathymetry
data (Chesapeake Bay Program 1997), a 1-meter
depth contour along the entire Chesapeake Bay and
tidal tributaries shoreline was developed. The Tier II
target includes all areas of past SAV habitat delineated
in the Tier I goal, as well as shallow water habitats
delineated within this 1-meter depth contour (Figure
VIII-2; Table VIII-1). Tier II excludes areas where
SAV is considered unlikely to survive and grow due to
the direct and indirect adverse effects of high wave
action. These "exclusion zones" used for the Tier II
and Tier III targets described here were the same ones
used in defining Tier III areas in Batiuk et al. (1992).
The Tier III SAV distribution restoration target is the
restoration of SAV to all shallow water areas delin-
eated as existing or potential SAV habitat down to the
2-meter depth contour. A new 2-meter depth contour
along the entire tidal Bay shoreline was developed
through contouring the expanded Bay bottom bathym-
etry database. The revised Tier III target includes all
areas in the Tier I goal and Tier II target, as well as
shallow water habitats delineated within this new 2-
meter depth contour (Figure VIII-3; Table VIII-1).
The Tier III target excludes areas where SAV is con-
sidered unlikely to survive and grow due to the direct
and indirect adverse effects of high wave action.
Figure VIII-4 illustrates the Chesapeake Bay Program
Segmentation Scheme, and Table VIII-2 provides the
tiered SAV distribution restoration goals and targets in
terms of hectares.
-------
Chapter VIII - Chesapeake Bay SAV Distribution Restoration Goals and Targets 123
TABLE VIII-1. Chesapeake Bay SAV distribution restoration Tier I goal,
Chesapeake Bay Program segment in acres.
CBP
Segment
CB1TF
NORTF
ELKOH
C&DOH
BOHOH
SASOH
CB2OH
CB3MH
BSHOH
GUNOH
MIDOH
BACOH
PATMH
CHSMH
CHSOH
CHSTF
MAGMH
SEVMH
SOUMH
RHDMH
WSTMH
EASMH
CB4MH
CHOMH1
CHOMH2
CHOOH
CHOTF
LCHMH
PAXMH
PAXOH
PAXTF
WBRTF
HNGMH
FSBMH
NANMH
NANOH
NANTF
WICMH
MANMH
Chesapeake Bay Program
Segment Name
Northern Chesapeake Bay
Northeast River
Elk River
Chesapeake & Delaware Canal
Bohemia River
Sassafras River
Upper Chesapeake Bay
Upper Central Chesapeake Bay
Bush River
Gunpowder River
Middle River
Back River
Patapsco River
Lower Chester River
Middle Chester River
Upper Chester River
Magothy River
Severn River
South River
Rhode River
West River
Eastern Bay
Middle Central Chesapeake Bay
Mouth of the Choptank River
Lower Choptank River
Middle Choptank River
Upper Choptank River
Little Choptank River
Lower Patuxent River
Middle Patuxent River
Upper Patuxent River
Western Branch of the Patuxent River
Honga River
Fishing Bay
Lower Nanticoke River
Middle Nanticoke River
Upper Nanticoke River
Wicomico River
Manokin River
Tier I
Goal
7,690
20
1,105
2
42
408
660
1,725
57
865
860
0
124
3,751
0
0
586
465
52
15
116
6,126
376
7,388
462
0
0
1,522
356
2
15
0
3,951
32
0
0
0
0
682
, and tiers II
TierH
Target
13,714
934
2,785
124
1,132
2,614
5,439
4,702
2,068
2,896
1,431
1,302
1,984
6,990
1,663
766
1,500
1,347
1,485
623
1,092
13,091
7,949
12,968
3,771
852
0
8,102
5,155
1,436
504
32
10,645
6,939
3,188
1,522
598
3,442
6,336
and III targets by
Tierm
Target
20,401
2,743
5,028
170
1,905
3,699
9,212
5,510
4,606
7,460
2,481
2,861
3,543
11,510
2,310
870
2,177
2,108
2,288
904
1,527
20,808
9,301
18,424
6,222
1,285
0
11,799
8,829
2,073
707
32
15,481
13,633
7,714
2,056
887
6,385
9,338
continued
-------
124 SAV TECHNICAL SYNTHESIS II
TABLE VIII-1. Chesapeake Bay SAV distribution restoration Tier I goal, and tiers II and III targets by
Chesapeake Bay Program segment in acres (continued)
CBP
Segment
BIGMH
POCMH
POCOH
POCTF
TANMH
POTMH
POTOH
POTTF
MATTF
PISTF
CB5MH
RPPMH
CRRMH
RPPOH
RPPTF
PIAMH
CB6PH
CB7PH
MOBPH
YRKMH
YRKPH
MPNOH
MPNTF
PMKOH
PMKTF
JMSPH
ELIPH
LAFMH
ELIMH
EBEMH
SBEMH
WBEMH
JMSMH
JMSOH
CHKOH
JMSTF
APPTF
LYMPH
CB8PH
Chesapeake Bay Program
Segment Name
Big Annemessex River
Lower Pocomoke River
Middle Pocomoke River
Upper Pocomoke River
Tangier Sound
Lower Potomac River
Middle Potomac River
Upper Potomac River
Mattawoman Creek
Piscataway Creek
Lower Central Chesapeake Bay
Lower Rappahannock River
Corrotoman River
Middle Rappahannock River
Upper Rappahannock River
Piankatank River
Western Lower Chesapeake Bay
Eastern Lower Chesapeake Bay
MobjackBay
Lower York River
Middle York River
Lower Mattaponi River
Upper Mattaponi River
Lower Pumunkey River
Upper Pumunkey River
Mouth of the James River
Mouth of the Elizabeth River
Lafayette River
Middle Elizabeth River
Eastern Branch of the Elizabeth River
South Branch of the Elizabeth River
Western Branch of the Elizabeth River
Lower James River
Middle James River
Chickahominy River
Upper James River
Appomattox River
Lynhaven & Back Bays
Mouth of the Chesapeake Bav
Tier I
Goal
902
2,078
0
0
19,899
988
4,265
6,405
133
835
4,776
2,471
541
0
0
1,994
1,265
12,081
13,744
54
1,401
0
0
0
0
40
0
0
0
0
0
0
0
0
225
0
0
175
0
Hern
Target
3,197
14,016
1,406
581
38,874
26,069
7,188
7,794
697
588
15,021
19,770
1,819
1,653
3,190
5,668
3,936
28,510
22,978
8,387
5,088
445
996
598
2,187
1,616
0
0
0
0
0
0
17,613
6,476
3,506
10,400
1,307
3,304
697
TierHI
Target
5,068
17,969
1,515
749
58,024
45,807
15,199
17,838
1,389
914
18,691
30,035
2,612
2,511
4,515
7,789
5,130
32,575
30,554
12,666
7,139
613
1,352
860
2,654
2,266
0
0
0
0
0
0
29,138
10,954
4,505
12,842
1,604
3,961
1.053
TOTAL Chesapeake Bay
113,720 408,689
618,773
-------
Chapter VIII - Chesapeake Bay SAV Distribution Restoration Goals and Targets 125
TABLE VIII-2. Chesapeake Bay SAV distribution restoration Tier I goal,
Chesapeake Bay Program segment in hectares.
CBP
Seement
CB1TF
NORTF
ELKOH
C&DOH
BOHOH
SASOH
CB2OH
CB3MH
BSHOH
GUNOH
MIDOH
BACOH
PATMH
CHSMH
CHSOH
CHSTF
MAGMH
SEVMH
SOUMH
RHDMH
WSTMH
EASMH
CB4MH
CHOMH1
CHOMH2
CHOOH
CHOTF
LCHMH
PAXMH
PAXOH
PAXTF
WBRTF
HNGMH
FSBMH
NANMH
NANOH
NANTF
WICMH
MANMH
Chesapeake Bay Program
Seement Name
Northern Chesapeake Bay
Northeast River
Elk River
Chesapeake & Delaware Canal
Bohemia River
Sassafras River
Upper Chesapeake Bay
Upper Central Chesapeake Bay
Bush River
Gunpowder River
Middle River
Back River
Patapsco River
Lower Chester River
Middle Chester River
Upper Chester River
Magothy River
Severn River
South River
Rhode River
West River
Eastern Bay
Middle Central Chesapeake Bay
Mouth of the Choptank River
Lower Choptank River
Middle Choptank River
Upper Choptank River
Little Choptank River
Lower Patuxent River
Middle Patuxent River
Upper Patuxent River
Western Branch of the Patuxent River
Honga River
Fishing Bay
Lower Nanticoke River
Middle Nanticoke River
Upper Nanticoke River
Wicomico River
Manokin River
Tier I
Goal
3,112
8
447
1
17
165
267
698
23
350
348
0
50
1,518
0
0
237
188
21
6
47
2,479
152
2,990
187
0
0
616
144
1
6
0
1,599
13
0
0
0
0
276
and tiers II and III targets by
Hern
Target
5,550
378
1,127
50
458
1,058
2,201
1,903
837
1,172
579
527
803
2,829
673
310
607
545
601
252
442
5,298
3,217
5,248
1,526
345
0
3,279
2,086
581
204
13
4,308
2,808
1,290
616
242
1,393
2,564
Tierm
Target
8,256
1,110
2,035
69
771
1,497
3,728
2,230
1,864
3,019
1,004
1,158
1,434
4,658
935
352
881
853
926
366
618
8,421
3,764
7,456
2,518
520
0
4,775
3,573
839
286
13
6,265
5,517
3,122
832
359
2,584
3,779
continued
-------
126 SAV TECHNICAL SYNTHESIS I
TABLE VIII-2. Chesapeake Bay SAV distribution restoration Tier I goal, and tiers
Chesapeake Bay Program segment in hectares (continued)
CBP Chesapeake Bay Program
Segment Segment Name
BIGMH Big Annemessex River
POCMH Lower Pocomoke River
POCOH Middle Pocomoke River
POCTF Upper Pocomoke River
TANMH Tangier Sound
POTMH Lower Potomac River
POTOH Middle Potomac River
POTTF Upper Potomac River
MATTF Mattawoman Creek
PISTF Piscataway Creek
CB5MH Lower Central Chesapeake Bay
RPPMH Lower Rappahatmock River
CRRMH Corrotoman River
RPPOH Middle Rappahannock River
RPPTF Upper Rappahannock River
PIAMH Piankatank River
CB6PH Western Lower Chesapeake Bay
CB7PH Eastern Lower Chesapeake Bay
MOBPH MobjackBay
YRKMH Lower York River
YRKPH Middle York River
MPNOH Lower Mattaponi River
MPNTF Upper Mattaponi River
PMKOH Lower Pumunkey River
PMKTF Upper Pumunkey River
JMSPH Mouth of the James River
ELIPH Mouth of the Elizabeth River
LAFMH Lafeyette River
ELMH Middle Elizabeth River
EBEMH Eastern Branch of the Elizabeth River
SBEMH South Branch of the Elizabeth River
WBEMH Western Branch of the Elizabeth River
JMSMH Lower James River
JMSOH Middle James River
CHKOH Chickahominy River
JMSTF Upper James River
APPTF Appomattox River
LYNPH Lynhaven& Back Bays
CB8PH Mouth of the Chesapeake Bav
Tier I
Goal
365
841
0
0
8,053
400
1,726
2,592
54
338
1,933
1,000
219
0
0
807
512
4,889
5,562
22
567
0
0
0
0
16
0
0
0
0
0
0
0
0
91
0
0
71
0
TierH
Tarcet
1,294
5,672
569
235
15,732
10,550
2,909
3,154
282
238
6,079
8,001
736
669
1,291
2,294
1,593
11,538
9,299
3,394
2,059
180
403
242
885
654
0
0
0
0
0
0
7,128
2,621
1,419
4,209
529
1,337
282
II and III targets by
Tier in
Target
2,051
7,272
613
303
23,482
18,538
6,151
7,219
562
370
7,564
12,155
1,057
1,016
1,827
3,152
2,076
13,183
12,365
5,126
2,889
248
547
348
1,074
917
0
0
0
0
0
0
11,792
4,433
1,823
5,197
649
1,603
426
TOTAL Chesapeake Bay
46,022
165,394 250,414
-------
SAV Distribution
Restoration Target
Figure VIII-2. Tier II SAV Distribution Restoration Target.
-------
•|Bog UOHBJ01S9H
H-IIIA 3Jn6y
-------
SAV Distribution
Restoration Goal
Figure VIII-3. Tier III SAV Distribution Restoration Goal.
-------
'f-IIIA ajnBy
HW3HS
HW383
it '
HdNAlJ*Jg^H»\ldVn
-------
CHAPTER |X
Comparing Nearshore and Midchannel
Water Quality Conditions
I
n the Chesapeake Bay region, most governmental
I agency tidal water quality monitoring programs sam-
ple only midchannel locations to reduce sampling time
and costs. This provides crucial water quality informa-
tion for determining status and long-term trends of the
Chesapeake Bay mainstem, its tidal tributaries and
embayments. However, does midchannel monitoring
provide adequate information to characterize the sta-
tus of biologically important nearshore areas? If the
nearshore values are found to be statistically similar to
the midchannel values, then resource managers could
make better informed decisions about nearshore areas
without requiring additional water quality monitoring
locations. Conversely, differing nearshore and mid-
channel conditions might require revision of existing
monitoring programs or the initiation of new ones.
Several studies have addressed the nearshore vs.
midchannel sampling issue in Chesapeake Bay (Table
IX-1; Stevenson et al. 1991; Batiuk et al. 1992; Chesa-
peake Bay Program 1993; Ruffin 1995; Bergstrom,
unpublished data; Parham 1996). While most studies
indicate that midchannel data can be used to describe
nearshore conditions, several suggest the opposite.
There is no doubt that demonstrable differences in
water quality can occur between nearshore and mid-
channel stations over varying temporal and spatial
scales, especially when submerged aquatic vegetation
is present (Ward et al. 1984; Moore et al. 1995; Moore
1996). Other possible causes of variability between
nearshore and midchannel environments include
localized resuspension of sediments, algal patchiness,
point source effluents or sediment chemistry variabil-
ity (Goldsborough and Kemp 1988; Moore 1996).
The findings presented in this chapter result from a
comprehensive analysis of directly accessible
nearshore and midchannel data in the Chesapeake
Bay collected since 1983, to determine whether mid-
channel water quality monitoring data are applicable
for characterizing nearshore environments. The full
report describing these analyses has been published by
Karrh (1999). Data for this study were incorporated
from all over Chesapeake Bay and its tidal tributaries,
including the upper Chesapeake Bay region; the Mid-
dle, Magothy, Rhode, Chester, Choptank, Patuxent,
Potomac, Rappahanock, Poquoson, York and James
rivers; and Mobjack Bay. Data were obtained from
state monitoring efforts, academic researchers and cit-
izen monitors. Most data used in the analyses came
from unvegetated areas.
METHODS
The parameters analyzed included Secchi depth, dis-
solved inorganic nitrogen, dissolved inorganic phos-
phorous, chlorophyll a, total suspended solids and
salinity. These are the parameters most relevant to the
survival of submerged aquatic vegetation (Batiuk et al.
1992; Dennison et al. 1993). Salinity was also included
as a diagnostic parameter in assessing the comparabil-
ity of sites. The study period was limited to the SAV
growing season (April to October in tidal fresh, oligo-
haline and mesohaline areas and March to May and
September to November in polyhaline areas). In some
datasets, it was necessary to sum the nitrite, nitrate
and ammonia fractions to obtain dissolved inorganic
nitrogen and to convert data to similar units in all
datasets. Since many of the citizen monitoring datasets
Chapter IX - Comparing Nearshore and Midchannel Water Quality Conditions 131
-------
132 SAV TECHNICAL SYNTHESIS II
TABLE IX-1. Summary of previous nearshore/midchannel comparisons. Area indicates tidal tributary or
mainstem Chesapeake Bay study area. Source indicates publication that the results appear in. Under Kd
(light attenuation coefficient), DIN (dissolved inorganic nitrogen), DIP (dissolved inorganic phosphorous),
TSS (total suspended solids) and chlorophyll a, the results are shown for each study, based upon whether
the midchannel data can be used to characterize the nearshore environment. Yes = midchannel data can
be applied to the nearshore areas, No = cannot be used and ND = no data.
Area
Source
DIN DIP TSS Chlorophyll a
Upper Bay
Chester
Choptank
Choptank
Choptank
Patuxent
Patuxent
Potomac
York
Mainstem
Bay
Batiukefa/. 1992
Ruffin 1996
Batiuketal. 1992
Stevenson et al. 1991
Parham 1996
Bergstrom unpublished
Parham 1996
Batiukefa/. 1992
Batiuke/a/. 1992
Chesapeake Bay Program
1993
Yes
No
Yes
No*
Yes
ND
Yes
Yes
Yes
No
Yes
ND
Yes
No*
ND
Yes
1SID
Yes
Yes
Yes
Yes
ND
Yes
No*
ND
Yes
ND
Yes
Yes
Yes
Yes
ND
Yes
No*
ND
ND
ND
Yes
Yes
Yes
Yes
ND
Yes
No*
ND
ND
ND
Yes
Yes
Yes
* The authors of this study concluded that the wide variation in the data was masking any
potential significant statistical result, therefore they based their conclusions on their correlational
analyses.
report Secchi depth and not Kd, Kd values given in
some datasets were converted to Secchi depth using
the appropriate conversion factor (see Chapter III;
Batiuk et al. 1992).
Data Sources
Data for this study were obtained from many sources.
Only datasets with two or more years were used for the
analysis. Citizen monitoring data were from the
Alliance for the Chesapeake Bay, the Magothy River
Association and the Anne Arundel County Volunteer
Monitoring Program. Nearshore water quality data
used in this analysis were obtained from George
Mason University, the University of Maryland, the
Virginia Institute of Marine Science, the Smithsonian
Environmental Research Center and Harford Com-
munity College. Midchannel water quality data were
synthesized from monitoring programs run by the
Maryland Department of the Environment, the Mary-
land Department of Natural Resources, the Virginia
Department of Environmental Quality and the
U.S. Geological Survey. All data have been placed
into SAS datasets for analysis and are directly accessi-
ble through the Chesapeake Bay Program web site at
www.chesapeakebay.net.
Station Selection
The stations used in the comparisons were selected
through the use of Arc View® desktop GIS software to
create maps showing the positions of the stations
-------
Chapter IX - Comparing Nearshore and Midchannel Water Quality Conditions 133
(Figure IX-1). As most of the nearshore stations were
established for other purposes than a nearshore/mid-
channel comparison, stations were picked for compar-
ison based solely on their proximity to one another,
and to maximize the number of possible comparisons.
Stations compared were located less than 10 kilome-
ters apart, with most paired stations less than five kilo-
meters apart. The spatial relationship of stations to
one another was considered, so that a nearshore sta-
tion located far up a subtributary was not paired with
a midchannel station in the tributary's mainstem. In
this way, the stations did not differ dramatically in
chemical or physical nature. The distances between
paired stations were determined, allowing conclusions
to be drawn on how far away from a given midchannel
station the nearshore water quality conditions can still
be characterized using midchannel data.
Statistical Analysis
Individual nearshore/midchannel data were analyzed
using the Wilcoxon paired-sample test (Wilcoxon
1945; Zar 1984). This test examines differences
between two samples of the same observation (i.e., one
nearshore vs. one midchannel station sampled on the
same day). The actual daily values were used, not a
median. If all of the samples from the two stations
being compared had approximately the same number
and magnitude of positive and negative differences,
then the stations were considered similar in respect to
the parameter of interest. However, if one station had
a consistently higher or lower value than the other,
then the stations were considered significantly differ-
ent with respect to the parameter of interest.
The Wilcoxon paired-sample test is a nonparametric
analog to the paired-sample t-test, and is more appro-
priate to water quality data where the data cannot be
assumed to be normally distributed. The Wilcoxon
paired-sample test is 95 percent as powerful in detect-
ing differences between two sets of data as the t-test.
Significance was evaluated at an level of .05. The tests
were performed using a SAS program. For the pur-
poses of this report, the term, "comparison" refers to
a station A vs. station B statistical analysis. Figures IX-
2 and IX-3 show example box and scatter plots for the
York River, along with the results of the statistical
analyses. A complete set of similar figures for all the
nearshore/midchannel paired station comparisons are
published in Karrh (1999).
In order to perform the Wilcoxon test, the data from
two stations must have paired observations. Since
many of the stations were sampled on different dates,
the data were forced to match by date. A 10-day sam-
pling difference was used as the limit to keep temporal
differences between stations to a minimum while max-
imizing the number of paired stations for the compar-
ison. Most of the temporal differences were one to five
days.
Ideally, the stations compared would have data col-
lected within hours of one another. However, the data
used in this study were obtained from a variety of
sources, each with different sampling schedules and
protocols. In order to have sufficient observations to
perform some of the comparisons, it was necessary to
be fairly pragmatic about temporal differences. The
analyses were conducted using all available data for all
years. It has been argued that the data should be ana-
lyzed by year to account for interannual water quality
variability. However, the goal of this study was to
determine if midchannel data are applicable to
nearshore conditions overall. For example, if a six-year
dataset was analyzed by year for a parameter and there
were three significant and three nonsignificant results,
are these stations comparable or are they different?
RESULTS AND DISCUSSION
The results of this comprehensive study show that appli-
cability of midchannel data to nearshore environments
is very site-specific. There are wide variations in the
results within tributaries and between comparisons
using one midchannel station vs. multiple nearshore sta-
tions. Karrh (1999) describes the site-specific nature of
the results in more depth. Possible causes of this vari-
ability include localized resuspension of sediments,
algal patchiness, point source effluents or sediment
chemistry variability. Also, differences in sampling
schedule and protocols between midchannel and
nearshore sampling programs could contribute to
observed differences. Another confounding factor may
be the presence of SAV at certain sites, as the plants can
change total suspended solids, dissolved inorganic
nitrogen, dissolved inorganic phosphorus, chlorophyll a
concentrations and light penetration locally.
Table IX-2 summarizes the results of the statistical
analyses by tributary, expressed as the percentage of the
total number of comparisons that yielded a nonsignifi-
cant result (i.e., the nearshore and midchannel stations
-------
134 SAV TECHNICAL SYNTHESIS I
Midchannel stations
Nearshore stations
FIGURE IX-1. Nearshore and Midchannel Water Quality Monitoring Stations. Chesapeake Bay and tidal tributary
water quality monitoring stations used in the nearshore vs. midchannel analyses.