Chesapeake Bay Program Water Clarity Criteria Team Final Responses to Comments on the July 3, 2001 Working Draft Chesapeake Bay Water Clarity Criteria Document

December 19, 2001
Chesapeake Bay Program Water Clarity Criteria Team
Final Responses to Comments on the July 3,2001
Working Draft Chesapeake Bay Water Clarity Criteria Document

From Peter Bergstrom, USFWS

1. Minor point (but a source of confusion I'm sure): Table IV-8 is unlabeled in my copy.

Response: Table IV-8 has been labeled in the revised draft.

2. For the existing use depth you used: Greater of: 1971-1999 deepest mapped SAV depth
(which of the 2 estimates?), and 1985-1999 MLR attainment depth (or did you use WCLR?)
(Please reply to clarify which of the two estimates you used in each case.)

Response: One can get this information by looking at the respective columns of data in
Table IV-7.

3. For the application depth you used: Greater of: 1940's-1960's photography maximum
depths, and model run depths. I'll have to think about it some more to see if I agree with
these choices, and we need to discuss this. I think it yields application depths that will
restore SAV, but some are so much deeper than existing use depths, it may be hard to get
any compliance.

Response: The methodology for selecting the criteria application depths has been
expanded and edited to be more clear on how the available information was used in
setting Bay Program segment specific depths.

4. If it was this hard for me to figure this out, the reviewers will have even more trouble. I
think perhaps we need to add a table explaining how we derived these two estimates, and
how they differ.

From WASH-COG:

8. The proposed water clarity criteria represent years of development and research, and could
be precedent-setting for helping to evaluate other estuaries around the world.

Response: Restatement of information in the working draft criteria document; no
change made to text.

9. The draft criteria are based on the percent of ambient light that is available at the leaf
surface (PLL). Because it is not possible to measure this empirically, an algorithm is used to
calculate PLL as a function of several water quality parameters. Thus, assessment of
attainment is entirely based on the accuracy and applicability of the algorithm. There is no
quantification of the uncertainty associated with the algorithm, and no direct validation in
the field. Because of the regulatory significance of water quality criteria and their
attainment/non-attainment, it is important to ensure that there is no ambiguity in that

-------
determination.

Response: The determination of Bay water clarity criteria attainment is based on
observed concentrations of light (Kd or Secchi depth), dissolved inorganic nitrogen,
dissolved inorganic phosphorus, and total suspended solids, analyzed through the
algorithm. Each component of the algorithm was developed and in many cases,
validated using field derived data. The algorithm itself was verified through
application to Chesapeake Bay water quality monitoring data. The results of these
field data verifications are fully documented within chapter V Epiphyte Contributions
to Light Attenuation at the Leaf Surface and Chapter VII. Setting, Applying and
Evaluating Minimum Light Requirements for Chesapeake Bay SAV in Batiuk et al.
(2000) and summarized in the scientific peer reviewed literature paper by Kemp et al.
(in review). The only other way to attempt any validation beyond the validations
already published would be to conduct a Monty Carlo simulation and sensitivity
analysis, although we are unsure of exactly what new insights such an analysis would
provide.

The revised draft water clarity criteria document proposes publishing both the percent
light at the leaf and the percent light through the water criteria values for both salinity
regimes as Bay criteria. The implementation guidelines provide guidance as to
where/when to apply PLL vs. PLW criteria.

10. The distribution of submerged aquatic vegetation (SAV) in the Bay is much wider than
would be predicted by the water clarity criteria. There is need for further validation of the
relationship between PLL values generated by the algorithm and the distribution of SAV in
the Bay.

Response: As described in the response to comment 9, the Bay criteria have been
verified through application of the minimum light requirements and the supporting
algorithm to Chesapeake Bay water quality monitoring data. See Figure VIE-4 and the
accompanying text on page 111 in Batiuk et al. (2000). There were only 2 of 68
segment (PAXTF and PAXOH) that had over 35 hectares of SAV and the minimum
light requirement failed at 0.25 meters plus half the tidal range. If one was to expand
this to include segments with over 35 hectares of SAV that had a minimum light
requirement which failed at 0.5 meters plus half the tidal range, then POTTF, POTOH
and GUNOH are added to the list. In these few select cases where the distribution of
SAV was wider than would be predicted by the draft water clarity criteria, the SAV
beds in the Patuxent were found to be within the intertidal zone, therefore overcoming
still very poor light conditions. In the Potomac, SAV beds exist in spite of poor water
clarity due to the canopy forming species milfoil and hydrilla and the ability of these
beds to improve local water quality conditions once they are dense enough. However,
during high flow years, we often lose significant acres of grasses in the tidal Potomac,
reflecting these grass beds on right on the edge of having sufficient water quality to
survive years with poor water quality.

Results from the recent application of the draft water clarity criteria at the draft
established application depths to the last three years of Bay monitoring data available

-------
on-line-1989,1999, and 2000-have also been incorporated into the revised water
clarity criteria document as additional validation documentation.

11. The linear regression model for the water column attenuation coefficient yields values for Kd
that are used in the derivation of both PLW and PLL. However, the predicted values of Kd
show a poor correlation with observed values. For the tidal tributaries the r 2 is only 0.37.

This poor fit hinders the predictive ability of the PLW and PLL algorithms, and may yield
results that suggest non-attainment when adequate light is actually available.

Response: The linear regression model for the water column attenuation coefficient
was used to derive the diagnostic tool for examination of management options based on
PLW, but it is not used in the algorithm for calculating PLL. Calculation of PLL,
which determines criteria attainment or non-attainment, is based on Kd which is either
directly measured or calculated from Secchi depth.

The diagnostic tool based on PLW is offered as a useful procedure for examining
management options to achieve criterion attainment, when non-attainment is
observed. Site specific adjustment of the regressions will improve the utility of the
tool, but application of the water clarity criteria based on PLL need not await these
refinements to the diagnostic tool.

12. At the present time we recommend that the proposed criteria be applied as guidelines on a
"trial" basis at several key segments representing the different designated uses. In this way,
more intensive water quality monitoring and SAV mapping can be conducted to help
validate the approach.

Response: The proposed water clarity criteria are based on two comprehensive
synthesizes of Chesapeake Bay and worldwide science spanning 25 years backed up by
3 decades of data on Bay SAV distribution and abundance and 2 decades of intensive
baywide water quality monitoring data. All aspects of the criteria and its supporting
documentation have been published in the scientific peer reviewed literature.
Recognizing science and scientific understanding is continually evolving, the proposed
criteria are well beyond the "trial" stage in through scientific underpinning as well as
their application to date.

Overview of Technical Approach Used

13. The reduced abundance of submerged aquatic vegetation (SAV) from shallow waters of the
Chesapeake Bay is a well-documented problem. SAV beds provide critically important
habitat for fish and invertebrates, and also provide food for fish, shellfish and birds. The
plants also are important in nutrient cycling and sediment stabilization. The primary causes
of the loss of SAV are believed to be eutrophication and the associated loss of light
availability.

Response: Restatement of information in the working draft criteria document; no
change made to text.

14. Incident light that penetrates the water surface may be attenuated in one of two ways

-------
(Figure 4-1). First, light is attenuated in the water column primarily by particulate matter
(chlorophyll a and total suspended solids (TSS)). This attenuation (icd) can be measured
directly or calculated from Secchi depth. The amount of light penetrating the water column
is defined as percent light in water or PLW. Further attenuation occurs due to epiphytic
material (algae, bacteria, detritus, and sediment) that accumulates on the leaves of SAV. It
is difficult to measure this attenuation (ice) directly, but an algorithm was developed to
compute the biomass and attenuation of the epiphytic material based on four parameters:
TSS, dissolved inorganic nitrogen (DIN), dissolved inorganic phosphorus (DIP), and Kd
(Figure 4-2). Using this model algorithm, the amount of light reaching the leaf surface
(percent light at leaf or PLL) can be calculated using these commonly-monitored
parameters.

Response: Restatement of information in the working draft criteria document; no
change made to text.

15. The process of derivation of water clarity criteria followed a series of successive stages. The
first stage was the determination of water-column based light requirements for SAV
survival and growth. The authors found that statistical regression models that quantify the
relationship of light availability to depth of SAV growth were the most useful tool for
developing minimum light requirements for the Chesapeake Bay. Based on modeling results
as well as Bay monitoring data, minimum water-column light requirements of 22 percent for
mesohaline and polyhaline regions and 13 percent for fresh tidal and oligohaline regions
were established.

Response: Restatement of information in the working draft criteria document; no
change made to text.

16. The next stage was to determine the minimum light requirement at the leaf surface. Three
lines of evidence were considered in this determination:

1. Calculation applying 1992 SAV habitat requirements for Kd, DIN, DIP, and TSS (Batiuk
etal. 1992) into the PLL algorithm;

2. Adjustment of water-column light requirements using literature values for epiphytic
attenuation; and

3. Comparison of PLL determinations using monitoring data with SAV distribution data in
the Choptank and York Rivers.

Response: Restatement of information in the working draft criteria document; no
change made to text.

17. The proposed water clarity criteria are expressed as minimum light requirements for SAV
survival and growth. For tidal fresh and oligohaline waters, the criterion is 9 percent of
available light. For mesohaline and polyhaline waters, the criterion is 15 percent of
available light. Attainment of the criteria is assessed through calculation of PLL at a given
site. Thus, the water quality criteria for water clarity are aggregate criteria for TSS,
nitrogen, and phosphorus. The criteria are applicable in the SAV growing season (April -
October in most waters) within the "shallow water" use designation (i.e., out to the 2 meter
depth contour).

Response: Restatement of information in the working draft criteria document; no

-------
change made to text.

18. The protection and restoration of SAV is a critically important part of the overall
management strategy for the Chesapeake Bay. Many desired uses cannot be fully attained
without the presence of greatly expanded ranges of submerged vegetation in all parts of the
watershed. While water clarity criteria themselves are new, the concepts underlying them
are not. Efforts to develop a technically-based approach to improving conditions for SAV
have been underway for more than 10 years (Batiuk el al. 1992, 2000). Early approaches
focused on light attenuation of the water column, while more recent efforts have also
included the role of epiphytic material in light attenuation. To the extent possible, the
authors have used data from the Bay itself rather than relying on data from fresh water lakes
or the open ocean. Extensive work has been done at all stages of criteria development to test
the models and assumptions against field data for water quality and SAV distribution. In
short, the proposed water clarity criteria represent years of development and research, and
will likely be precedent-setting for helping to evaluate other estuaries around the world.
Response: Restatement of information in the working draft criteria document; no
change made to text. Work on the underlying scientific basis for the proposed water
clarity criteria has been in development really for more than past two decades,
beginning with research funded during the Chesapeake Bay Program's research phase
starting in the mid-1970s.

Technical Comments

While the approach for developing water clarity requirements is sound, it is premature to apply

them as water quality criteria.

Response: See responses to comment 20.

19. If the water clarity requirements are established as criteria, they will fall under Section
303(d) of the Clean Water Act. Waters in which the criteria are not met will have to be
included in a list to be submitted to US EPA. Then, Total Maximum Daily Loads (TMDLs)
will be required to establish the necessary load reductions to attain the criteria. This will be
particularly difficult for water clarity, which is not a parameter that can be measured
directly (see Section 4.2.2 below). Instead, load reductions would have to be defined for the
contributing factors (i.e., nitrogen, phosphorus, TSS). In turn, these load reductions would
lead to NPDES permit limits and other management tools.

Response: Agree. No changes made to the text.

20. For these reasons, it is critically important that potential water quality criteria undergo
intense scrutiny and validation before achieving such legal status. As discussed in
subsequent comments below, there are a number of technical concerns and data gaps that
must be addressed before the water clarity requirements be established as water quality
criteria. Most significantly, there is a great need for further validation of the relationship
between PLL values generated by the algorithm and the distribution of SAV in the bay. In
the interim, we recommend that the criteria be applied as guidelines on a "trial" basis at
several key segments representing the different habitats of the Bay. In this way, intensive

-------
water quality monitoring and SAV mapping can be conducted in an attempt to validate the
approach. Further, the Chesapeake Bay Water Quality Model, in combination with the PLL
algorithm, could be used to "practice" management approaches for attaining the guidelines.
Finally, because DIN and DIP are key variables in the PLL algorithm, the voluntary
reduction agreement for nitrogen and phosphorus will likely lead to improvement in SAV
survival and growth in the Bay even without the implementation of water clarity criteria.
Response: The Bay criteria have been verified through application of the minimum
light requirements and the supporting algorithm to Chesapeake Bay water quality
monitoring data. The results of these field data verifications are fully documented
within chapter V Epiphyte Contributions to Light Attenuation at the Leaf Surface and
Chapter VII. Setting, Applying and Evaluating Minimum Light Requirements for
Chesapeake Bay SAV in Batiuk et al. (2000) and summarized in the scientific peer
reviewed literature paper by Kemp et al. (in review). Since publication of the second
technical synthesis, attainment of the working draft Bay water clarity criteria has been
extensively tested using the full Bay water quality monitoring data base along with
direct comparison with SAV distribution and abundance time series since 1985 (e.g.,
Spring 2001 Water Clarity Criteria workshop proceeding report). There is no reason
for application of the draft criteria during a trial period given the past and existing
extensive application to decades of data.

There is no need to "practice" application of the Chesapeake Bay Water Quality
Model to in support of using determination of simulated attainment of the Bay water
clarity criteria. The necessary programming necessary to determine attainment of the
Bay water clarity criteria using model simulated output has been built into the Bay
Water Quality Model post processor. The Bay watershed partners will have months
and months of "practice" they work together in the interpretation of results for many
management scenarios.

The statement "Finally, because DIN and DIP are key variables in the PLL algorithm,
the voluntary reduction agreement for nitrogen and phosphorus will likely lead to
improvement in SAV survival and growth in the Bay even without the implementation
of water clarity criteria" needs to be addressed separately. The Chesapeake 2000
Agreement and the follow-on six-state memorandum of understanding committed the
watershed state jurisdictions and EPA to "achieve and maintain the water quality
necessary to support the aquatic living resources of the Bay and its tributaries and to
protect human health". The Agreement goes further to commit to defining water
quality conditions necessary to protect living resources and adopt water quality
standard consistent with these water quality conditions. Nutrient and sediment
reductions are not going to happen; they are going to be driven in an effort to restore
these water quality conditions. Without adoption of the water clarity criteria, there is
no definition of necessary water quality conditions to drive reductions in sediment and
nutrients as well in areas where dissolved oxygen is sufficient, but light isn't. The
Chesapeake 2000 Agreement also commits to setting a new SAV restoration goal tied
directly with the water clarity criteria and the underlying application depth. The
public is not going to be satisfied with a "likely" improvement in SAV. To achieve the

-------
desire restoration of SAV, there must be a set of water clarity criteria in place.

Attainment of the proposed water clarity criteria cannot be measured directly.

Response: This statement is not true. See responses to comment 21.

21. The criteria are based on the percent of ambient light that is available at the surface of the
leaves of submerged plants (PLL). Because the light requirement accounts for the
attenuation of light by epiphytic material on the leaf surface, it is not possible to measure
the available light empirically. Instead, an algorithm is used to calculate PLL as a function

of several water quality parameters. (See Section 4.1 above). Thus, assessment of attainment
or non-attainment is entirely based on the accuracy and applicability of the algorithm itself.

There is no quantification of the uncertainty associated with the algorithm, and no direct
validation in the field. Because of the regulatory significance of water quality criteria and
their attainment/non-attainment, it is important to ensure that there is no ambiguity in that
determination.

Response: Attainment of the proposed Bay water clarity criteria is measured directly
through monitoring derived data on water column light penetration (measured as Kd
or Secchi depth), dissolved inorganic nitrogen, dissolved inorganic phosphorus, and
total suspended solids assessed through the algorithm. The assessment of attainment is
based on a host of factors beyond just the algorithm. All the factors going into
monitoring the four input variables have a direct effect on criteria attainment. As
discussed in previous responses, the algorithm and the proposed Bay criteria has been
extensively verified through direct application to decades of Bay water quality and
SAV distribution monitoring data.

22. An additional problem is associated with the use of the algorithm as a criterion. Even if the
algorithm itself is appropriate, non-attainment at a particular site is due to the contribution
of several independent variables. If a TMDL were required based on such non-attainment, it
would not be possible to determine load reductions of a single parameter that would ensure
attainment in the future. Instead, a management decision would need to be made as to
whether to require reductions of TSS, nitrogen, phosphorus, or some combination of the
three. Because reductions of different parameters would impact different sources, there is
likely to be contentious debate on all TMDLs.

Response: Development of a TMDL for a 200 mile long stratified semi-temperate
estuary whose watershed drains all or parts of seven state jurisdictions is a complex
problem in its simplest form. The fact that deriving a TMDL will be made complex
because of all the factors to be considered in determining and allocating load
responsibilities should not be a factor in deriving a criteria that clearly defines part of
the water quality necessary to support aquatic living resources in the Chesapeake Bay.

The light requirements are based largely on models of light availability rather than on field or

experimental data.

Response: This broad statement is incorrect. See responses to comments 23-25.

23. Typically, water quality criteria are based on a statistical evaluation of data from carefully-

-------
controlled laboratory experiments under rigorously-defined conditions. This ensures that the
resulting level of protection and experimental uncertainty are defined and understood. In the
case of the proposed water clarity criteria, four lines of evidence were evaluated (Batiuk el
al. 2000):

1. Physiological studies of photosynthesis/irradiance relationships;

2. Results of field observations of the maximum depth of SAV colonization and available
light at that depth;

3. Experiments involving the artificial or natural manipulation of light levels during long- or
short term growth studies; and

4. Statistical models intended to generalize light requirements.

Response: The vast majority of criteria published in the past 30 years by EPA has been
focused on chemical contaminants which by the very nature of their mode of impacting
aquatic organism, could be derived strictly through laboratory tests. However,
impairments due to nutrient and sediment over enrichment are caused by a whole
different set of mechanisms. Past methodologies and approaches to criteria derivation
can not just applied as is to the derivation of criteria protecting against the adverse
effect of eutrophication.

24. The authors concluded that the models represented the best source of information for
determining light requirements for the Chesapeake Bay. However, they conceded that most
of the models were based on either fresh water lakes or polyhaline marine environments,
with little data available on the fresh water tidal, oligohaline or mesohaline environments
that characterize much of the Bay watershed (Batiuk el al. 2000, p. 27).

Response: The models reviewed as part of the Bay water clarity criteria derivation
process were based on extensive set of data. Yes, most of the models focused on fresh
water and marine habitats, but the extensive review of worldwide scientific literature
yield extensive Bay specific species information on minimum light requirements,
particularly for species in mesohaline and polyhaline habitats. The selected minimum
light requirements were derived from the collective set of information derived from
the four sets of information summarized in comment 23 as well as the extensive
analysis of Chesapeake Bay field data supporting the original set of 1992 SAV habitat
requirements (Batiuk et al. 1992). The focus was on the convergence minimum light
requirements from several different approaches. The revised draft criteria document
has been edited to more clearly reflect what remains problematic is addressing
minimum light requirements in turbid habitats inhabited by canopy forming species.

25. The use of models is not inherently inappropriate, because they are based on field data.

However, this further supports the position that it is premature to use the resulting values as
guidelines rather than water quality criteria at this point in time.

Response: The scientific/technical basis for statement "the use of models is not
inherently inappropriate" is unclear. The models were based on field data-one can
not recreate all the natural factors influencing the derivation of minimum light
requirements within a laboratory. This does not make the application of those models
in support of criteria derivation "inherently inappropriate." The limitations of using
only laboratory-based data to determine minimum light requirements was

-------
documented in detail in Batiuk et al. 2000 (see Chapter III Light Requirement for SAV
Survival and Growth).

Other environmental factors may preclude the growth and restoration of SAV even if the
proposed criteria are attained.

Response: True statement, but not relevant to the determination of attainment of the
Bay water clarity criteria.

26. Implicit in the derivation of water clarity criteria is the assumption that light availability is
the primary environmental factor that limits SAV survival and growth (Batiuk et al. 2000).
However, the authors also concede that, in some cases, other environmental factors may
override the established light requirements. In other words, these factors may preclude SAV
from particular sites even when the minimum light requirements are met. These factors
include:

• Wave action;

• Currents;

• Tides;

• Sediment organic content;

• Sediment grain size; and

• Toxic chemicals

Response: The fact that there are clearly other factors beyond water clarity that
influence the restoration of Bay SAV has nothing to do with attainment of the water
clarity criteria. It has everything to do with taking other actions as needed to help
prompt SAV restoration. Factor that absolutely prevent any future SAV restoration
are addressed through the application depths, the new SAV restoration goal, and
through implementation of the water clarity criteria.

27. In some regions of the Bay, natural conditions of wave action, currents, tides, or sediment
may prevent the establishment of SAV, rendering the water clarity criteria unnecessary or
inappropriate for those locations. The authors conclude that, "Although we recognize that
factors other than light (including waves, tidal currents, sediments and toxic chemicals)
also limit SAV distribution in both pristine and perturbed coastal habitats, we have not yet
devised a scheme to explicitly account for them. " (Batiuk et al. 2000, p 8).

Response: Restatement of text from Batiuk et al. 2000; no changes made to text.

28. While this in itself is not sufficient reason not to establish criteria for water clarity, it
illustrates the need to further study all factors that may inhibit SAV in the Bay, as well as
the need to derive and apply these values on a more site-specific basis.

Response: Unless these other factors have a direct effect on the plants minimum light
requirements, then these factors do not need to be factored into the Bay water clarity
criteria.

The distribution of SAV in the Chesapeake Bay is much wider than would be predicted by the
water clarity criteria.

Response: See responses to comment 10.

-------
29. Batiuk el al. (2000) attempted to validate the light requirements by relating calculated
values of PLL to field data on SAV presence (over a 10-year record) in areas adjacent to
water quality monitoring stations. SAV presence was categorized as: always abundant
(AA), always some (AS), sometimes none (SN), usually none (UN) and always none (AN).
It was assumed that PLL would exceed the minimum requirement in the AA areas, and
would be approximately equal to the requirement in the AS and SN areas. In fact, in fresh
tidal and oligohaline waters, the median values of PLL at the 0.5 meter and 1.0 meter depths
were 5-8 percent and 1-3 percent in AS and SN areas, far below the minimum light
requirement of 9 percent. Thus, SAV were found to be present in areas with light levels
predicted to be well below the minimum requirement. The authors noted these
discrepancies, but concluded that they were "easily explained." Similar results were found

in relating PLW to changes in SAV coverage from year to year in tidal fresh and oligohaline
waters (Batiuk el al. 2000). Positive increases in SAV coverage occurred even when the
median PLW was considerably less than the minimum requirement. Finally, the authors
noted that SAV are often found at depths greater than the maximum that would be predicted
based on light requirements alone.

Response: See response to comment 10.

30. All of these observations suggest that the proposed water clarity criteria may be more
stringent than necessary to support SAV survival and growth, particularly in tidal fresh and
oligohaline environments such as the Potomac River and its tributaries. It may also
demonstrate the uncertainty associated with the criteria and the algorithm. Further
validation work should be conducted in these environments before the minimum light
requirements are applied formally as water quality criteria.

Response: The proposed Bay water clarity criteria are reflective of a set of minimum
light requirements protective of the collective set of species within the two set of
salinity regimes. Levels of light required to support existing, healthy SAV beds are
likely less than levels required to restore unvegetated regions. Levels of light required
during the time periods when the grasses are just emerging from the bottom sediments
and when the plants are directing their carbon resources to production of reproductive
structures (seed, tubers, etc.) are also likely to be more then those required for typical
growth.

The minimum light requirements for the Potomac River are considerably lower than the value

applied to tidal fresh and oligohaline environments throughout the Chesapeake Bay watershed.
Response: This statement is not correct when more recent published findings are taken
into account. See response to comment 31.

31. The proposed water clarity criteria are based on a minimum light requirement for the water
column (PLW) of 13 percent for fresh tidal and oligohaline environments. However,
previous research (Batiuk el al. 1992) suggested minimum light requirements of 11 percent
in the tidal fresh and 7 percent in the oligohaline reaches of the Potomac River. Thus, SAV
may thrive in waters that do not attain the proposed water clarity criteria in the Potomac and
its tributaries. This observation further supports the need for more site-specific guidelines,
rather than Bay-wide criteria. It may be most appropriate to establish individual guidelines

-------
for each segment. Setting bay-wide criteria and then relying on established procedures for
developing site-specific criteria or variances would lead to a wasteful use of resources and
lengthy legal challenges.

Response: There is no basis for deriving site-specific criteria when we are looking to
restore the same community of SAV species to the same salinity regimes across all Bay
tidal habitats. There is no reason why Susquehanna Flats wild celery plants have
different minimum light requirements from the same species of plants in the tidal
fresh Potomac or tidal fresh James rivers. That doesn't mean there are not different
factors (e.g., chlorophyll vs. suspended solids) are influencing light levels in different
habitats. In the 1992 technical synthesis, early Potomac field data indicated a set of
water column light requirements that were less than the set of water column light
requirements published for tidal fresh and oligohaline waters. More recent scientific
investigations focused on the tidal Potomac published in the scientific peer reviewed
literature by Carter et al. (2000) reported a range of water column light requirements
from 11 to 14.5 percent for the tidal fresh and oligohaline portions of the tidal Potomac
River. The 13 percent water column-based light requirement, used as part of the basis
for the low salinity Bay water clarity criteria, falls right in the middle of the Potomac
derived range of percentages of ambient light.

The calculated water column attenuation coefficient (t:d) exhibits a poor correlation with field-

measured values.

32. The linear regression model for the water column attenuation coefficient yields values for Kd
that are used in the derivation of both PLW and PLL. However, the predicted values of Kd
show a poor correlation with observed values (Batiuk et al. 2000). For the mainstem
Chesapeake Bay, the r 2 is 0.61, while for tidal tributaries the r 2 is only 0.37. This poor fit
hinders the predictive ability of the PLW and PLL algorithms, and may yield results that
suggest non-attainment when adequate light is actually available. The authors concede that
site-specific coefficients will likely be needed as a "refinement" in the future. This is yet
another reason why it is premature to establish the light requirements as water quality
criteria.

Response: See response to comment 11.

The biomass of epiphyton may be significantly overestimated by the algorithm for PLL.

33. While a significant regression was found between predicted and observed epiphytic
biomass, Batiuk et al. (2000) cited several environmental factors that can lead to
significantly lower biomass than predicted. The most important may be grazing by
invertebrates. Heavy grazing pressure may preclude epiphytic algal responses to nutrients.

Wave stress and water flushing may also affect epiphytic accumulation. When the
accumulation of epiphyton is significantly lower than that predicted by the algorithm, the
actual available light is higher than that predicted by PLL. Therefore, non-attainment may
be indicated even when adequate light is available. Further refinement of the model to
address these environmental factors is needed.

Response: The statement that "The biomass of epiphyton may be significantly

-------
overestimated by the algorithm for PLL" is not true at all times. There are some cases
where epiphyte biomass is overestimated and other cases where the biomass is
underestimated. In tidal fresh and oligohaline systems, the algorithm tends to
overestimate epiphyte biomass. The existing published algorithm is based on the best
available information both in the scientific literature as well as the most recent
applicable Bay specific laboratory and field data. Further refinement of the
algorithm, as is the case with all criteria and methodologies, is possible but must await
the next generation of new data. The algorithm has been validated for application to
Chesapeake Bay habitats as the best measure of water clarity.

Correlations between SAV area and median water quality were weak in tidal fresh water

segments.

34. Correlations were calculated between SAV area, by year, and median water quality from the
Chesapeake Bay Program water quality monitoring stations, by year (Batiuk el al. 2000). In
tidal fresh segments, there was significant correlation between some measurement
parameters of SAV coverage areas and TSS and dissolved inorganic phosphorus, but not for
dissolved inorganic nitrogen or chlorophyll a. Further, PLL did not have significant
correlations with SAV area parameters in tidal fresh segments. These findings cast further
question on the applicability of the water clarity criteria as currently derived in tidal fresh
areas such as the Potomac River and its tributaries.

Response: As described in response to previous comments, the PLL algorithm tends to
overestimate epiphytic biomass in tidal fresh to low salinity habitats. In the Potomac
River in particular, the dominant presence of canopy forming species-milfoil and
hydrilla-enable these SAV beds to exist in spite of poor water clarity. However,
during high flow years, we often lose significant acres of grasses in the tidal Potomac,
reflecting these grass beds on right on the edge of having sufficient water quality to
survive years with poor water quality. The current water quality conditions in the
Potomac, which fall below the draft water clarity criteria, are insufficient to support
sustained growth of the full array of SAV species that can inhabit these tidal fresh and
oligohaline habitats particularly during years of high spring runoff.

There is relatively poor agreement between midchannel and near shore water quality monitoring

data, particularly in the Potomac River.

Response: This broadly sweeping comment is not true in all tidal systems.

35. In the Chesapeake Bay and its tributaries, most long-term water quality monitoring stations
are located in midchannel for ease of sampling. However, since SAV grow in the shallow
depths near shore, it is important to know whether the midchannel data is similar to the
nearby nearshore data when available. Batiuk el al. (2000) conducted an extensive review of
monitoring data to examine this relationship. They found poor agreement for parameters
such as dissolved inorganic nitrogen and phosphorus in many locations, including the
Potomac River. However, they noted that the PLL values derived from the monitoring data
usually yielded the same conclusions regarding attainment vs. non-attainment. Even so, it is
clear that future monitoring must include additional nearshore stations.

-------
Response: The Chesapeake Bay Program partners are in the process of developing a
revised design for monitoring across the tidal waters from water quality up through
the lower trophic levels (plankton, benthos, SAV). Addressing the need to better
characterize water quality conditions in shallow water habitats is a top objective of
this monitoring network design work.

There is a great need for additional data to refine the derivation of the water clarity criteria.
Response: See response to comment 36.

36. While there is a solid scientific basis to the conceptual approach used in the derivation of
the water clarity criteria, a review of Chapter 4 and Batiuk el al. (2000) indicates that this
is very much a work-in-process. Batiuk el al. (2000) present a number of areas in which
additional research or monitoring data are needed, including the following key
components:

• Field and laboratory experiments on minimum light requirements for the particular SAV
species that are most prevalent in the Chesapeake Bay;

• Additional data on water column attenuation and the environmental factors that
contribute to it;

• Field and laboratory studies to further investigate the dynamics of epiphytic biomass;

• Further research on the role of environmental factors other than light that impact SAV
distribution (e.g., wave action, sediment grain size, and sediment chemistry).

Response: The second technical synthesis document did document areas requiring
further research and analysis to continue to better understand the SAV and habitat
interrelationships. Our understandings are infinitely perfectible given new data and
scientific findings. However, there are times when we must use the wealth of scientific
information at hand to reach a conclusion, make a decision, derive a criteria to
support policies and commitments. It will be many years before a significant enough
body of new scientific findings become available to warrant sweeping enhancement to
the minimum light requirements, algorithm, models, diagnostic tools, physical habitat
requirements, published in the 2000 technical synthesis and embodied in the proposed
Bay water clarity criteria.

37. While it is true that derivation of any water quality criterion is a dynamic process and
regular updates may be needed to account for new data, it is very important that new criteria
not be published before the level of uncertainty is reduced to an acceptable amount. This is
particularly true for the proposed water clarity criteria, because the derivation process is so
unique and has not been applied elsewhere.

Response: The comment refers to the need to reduce the level of uncertainty in the
proposed criteria to an acceptable amount, but does not define what is an acceptable
amount.

The use of the Chesapeake Bay Water Quality Model to calculate the water clarity expected
under various management strategies shows that the clarity criteria cannot be attained at
the 2-meter depth through much of the Bay.

Response: The working draft water clarity criteria document does not state the

-------
criteria must be attained throughout the Bay at the 2 meter depth.

38. The Water Quality Model was used to predict nutrient and TSS levels under several
different potential nutrient reduction management strategies. In turn, the algorithms were
used to predict the maximum depth at which the minimum light requirements would be met.
The results indicated that, even under the "All Forest Watershed" scenario, the light
requirements could only be met at depths significantly less than the 2-meter application
depth in many regions of the Bay. In the Middle Potomac River, for example, the maximum
depth at which the light requirements would be met was 1.39 meters. Although the authors
indicated that these simulations will be re-run using the updated Water Quality Model, these
preliminary results indicate that the clarity criteria are not attainable under any nutrient-
reduction scenario at the proposed application depth.

Response: The working draft document clearly spells out the proposed process for
determining the application depths (pages 13-16) and includes a first cut at the
segment by segment application depths. At this stage, comments concluding that the
"clarity criteria are not attainable under any nutrient-reduction scenario" are
premature given the need to incorporate new model scenario results and work in all
the newly available historical SAV bed information.

There are a number of additional issues related to the implementation of the water clarity
criteria.

39. While the language for implementation of the criteria has yet to be developed, there are
several issues that will certainly need to be addressed. These include the following:

• Application depth. The criteria are intended to apply out to the 2-meter depth in waters
classified as "shallow water" designated use. However, as discussed previously, the
criteria may not be attainable at this depth in many parts of the bay. Specific application
depths for important segments of the Bay may be the most significant implementation
issue. The criteria document discusses "existing use depths" and "potential use depths",
but it is unclear how these would be applied.

Response: The working draft criteria document provided the draft application depths
in Table IV-8. The revised draft document has been edited to make this whole section
more clear as to how the application depths were derived and how they would be
applied.

• Season. The criteria will apply during the SAV growing season (April-October in tidal
fresh, oligohaline, and mesohaline segments; March-May and September-November in
polyhaline segments). The seasonal nature of these criteria, and particularly the
difference in seasons between segments, will make management by the States difficult.
For example, development of TMDLs would be extremely complex.

Response: The draft criteria apply to the stated seasons because they are the growing
seasons for the underwater grasses. We have models which help us translate the
required ambient light, dissolved nutrient and suspended solids conditions back into
nutrient and sediment loading reductions. The time variable nature of these models

-------
addresses the concerns about seasons. Simply applying the criteria year round would
not be reflective of the water quality conditions needed for underwater grasses.

• Exceptions. The criteria will apply out to the 2-meter depth except "in circumstances
where natural conditions and other information indicate sufficient light could not reach
that depth during the SA Vgrowing season. " While this flexibility is welcome, it is not
clear how it would be applied. Would different application depths be used on a segment-
by-segment basis, as discussed above? Or would "variances" be applied to those
segments, with alternative or no clarity criteria applied? Because of the many problems
associated with obtaining water quality variances, it is inappropriate to issue new criteria
with the full knowledge that widespread variances would be necessary. (See Section
4.2.6.)

Response: The working draft document provided a table that laid out a draft set of
Chesapeake Bay Program segment by segment application depths for the criteria.

• Averaging period. There is no discussion as to the averaging period for the proposed
criterion. Would individual grab samples for TSS, DIN, and DIP be used to assess
instantaneous attainment, or would long-term averages be used? Because of the
importance of short-term events such as storms causing sharp increases in turbidity, it
seems appropriate to use long-term averages.

Response: A comprehensive set of implementation guidelines has been incorporated
into the revised draft criteria document. These guidelines specifically address the
averaging period for the criteria and how data from the four parameters are to be
applied.

References

Batiuk, R., R. Orth, K. Moore, J. C. Stevenson, W. Dennison, L. Staver, V. Carter, N. Rybicki,
R. Hickman, S. Kollar and S. Bieber. 1992. Submerged aquatic vegetation habitat requirements
and restoration targets: a technical synthesis. CBP/TRS 83/92. USEPA Chesapeake Bay
Program, Annapolis, MD.

Batiuk, R. A., P. Bergstrom, M. Kemp, E. Koch, L. Murray, J. C. Stevenson, R. Bartleson, V.
Carter, N. B. Rybicki, J. M. Landwehr, C. Gallegos, L. Karrh, M. Naylor, D. Wilcox, K. A.

Moore, S. Ailstock and M. Teichberg. 2000. Chesapeake Bay submerged aquatic vegetation
water quality and habitat-based requirements and restoration targets: a second technical
synthesis. CBP/TRS245/00 EPA 903-R-00-014. USEPA Chesapeake Bay Program, Annapolis
MD.

From LRSC ad hoc group, in La Plata August 17
Peter Bergstrom DRAFT 7-27-01

40. The goal of the standards is to protect designated uses of shallow waters, as described on pp.
11-12 of the July 3 draft document. However, I hope we can set application depths that
would improve habitat conditions enough to enable us to meet the CBP SAV restoration

-------
goal (114,000 acres). In 2000 the mapped SAV was at 60% of that area, so the increase
needed to reach the goal is about 45,000 acres. Since the goal was not based on restoration
of SAV to a specific depth, we can't just pick the depth that was used to set the goal. The
amount of potential SAV habitat to the 1 meter contour is 409,000 acres, far more than the
goal, and to 2 meters, 619,000 acres.

Response: The revised criteria document outlines a more specific set of decision rules
describing how the revised set of proposed application depths were determined.

41. The application depths shown in the current draft (Table IV-8, starts on page 41) do NOT
include the half tidal range. Half of the tidal range for that segment was added before
analyses were done, however. The application depths in the final standards will include this
(and thus the depths will be deeper).

Response: The proposed revised application depths now factor in half the tidal range
in the revised criteria document.

Rich Batiuk and his staff have found a number of data sources with which to estimate SAV
restoration depths. They are to be commended for assembling large amounts of data and
converting it to a usable form. The main issues to be reviewed by this group are which data
sources should be used, and how they will be used. There is general agreement on many of
these issues among the ad hoc group working with Rich to set the depths, but we have not
yet agreed on all of them.

Response: The revised criteria document outlines a more specific set of decision rules
describing how the revised set of proposed application depths were determined.

42. Maximum depth of 1971-1999 mapped SAV beds overlaid on depth contours, and clipped
to determine the deepest contour (up to 2 m) that had SAV mapped deeper than it. These
probably underestimate maximum bed depth due to limited water clarity. Shown in two
columns in Table IV-7; the second column used more restrictive criteria.

For the second of these two columns, we set a minimum percentage of area to count a depth
contour for the second column. We decided that 5 % of the mapped SAV needed to be
deeper than a contour, making up at least 10% of the potential habitat in that depth contour,
for it to "count" as the maximum SAV depth. See page 13 text, Table IV-5, and Appendix
C, Tables C-2 and C-3. We have to choose which ONE of these to use in determining the
final depth (can't use both because they are two versions of the same data).

Response: The working draft criteria document included very vague documentation
describing the use of the second column-"Max. depth meeting requirements"-as part
of the information used to define the draft existing use depths. More explicit
documentation has been added to the revised draft criteria document.

43. Maximum attainment depth of Minimum Light Requirements (MLR) based on water quality
data (Secchi, DIN, DIP, TSS): the depth at which the MLR was met. Also shown in
preceding column in table IV-7: maximum depth at which Water Column Light
Requirement (WCLR) was met, which uses Secchi data only. Both are shown because there
is some debate concerning the accuracy of the equations used to estimate leaf surface light

-------
attenuation. These equations affect the attainment of the MLR but not the WCLR, since it
does not estimate leaf surface attenuation. As with the previous source, we have to choose
which ONE of these to use in determining the final depth (can't use both because they are
two versions of the same data).

Response: The working draft criteria document included documentation describing
the maximum depth meeting the minimum light requirements calculated from 1985-
1999 Bay water quality monitoring data as part of the information used to define the
draft existing use depths. More explicit documentation has been added to the revised
draft criteria document.

44. Maximum depth of 1940's-1960's SAV beds from historical photography. These are
incomplete because all the photos have not yet been interpreted. These photos were taken to
measure areas of crops so many are not ideal for seeing SAV, because they may have been
taken outside the peak SAV growing season, when the water was turbid, not at low tide,
and/or with sun glare on the water. However, they are the best estimates we have of SAV
abundance during this period.

Response: These limitations of the applicability of the historical photography have
been added to the revised draft criteria document.

45. Water quality model simulated maximum depth meeting MLR, under 33% reduction above
tributary strategy scenario.

For all estimated depths, we used the maximum depth over all years with data. We could
have used the mean or median depth instead, but wanted to know the deepest depth estimate
that was supported by past or modeled future conditions in that segment.

Response: No change made to text.

46. How to choose one depth per segment as the "existing use depth " and one as the
"application depth " (for the standard), from among the estimates of maximum depth

For the existing use depth in Table IV-8 we used:

Greater of: 1971-1999 deepest mapped SAV depth using VTMS survey data, and 1985-1999
MLR attainment depth using CBP water quality data

For the application depth in Table IV-8 we used:

Greater of: 1940's-1960's photography maximum SAV depths, and model run depths (both
were missing for some segments, in which case we used 0.5 m more than the existing use
depth).

I need to discuss this table further with Rich before our August 17 meeting to make sure I
understand it. The application depth is the end result of the whole process, so it is the
critical number.

Whatever method we choose for this, I would like to see depths chosen that are enough of
an improvement over current conditions to cause a real expansion in SAV area, without

-------
being so deep that the standards are labeled unrealistic and thus are not used. I'm not sure
yet whether the highest, mean, weighted mean, or some other option best achieves this.
Response: No change made to text.

Spatial and temporal application once depths are chosen

47. Will these depths need to be met for MEAN conditions over all stations in a segment, or for
EACH station in a segment (spatial)? Do they need to be met each year, for means over
groups of years, or for some minimum percentage of years (temporal)?

Response: The revised draft criteria document contains detailed implementation
procedures specifically addressing these issues.

48. The July 3 draft has an outline of an implementation section that covers these issues (spatial
and temporal application), as well as several other issues. Obviously if we set protective
standards but set relaxed implementation guidelines, the standards won't achieve very
much.

Response: The intent of the draft implementation guidelines is not to "relax" the
resultant criteria and their application to the shallow water designated use via
adoption as state standards. The guidelines have been drafted to provide for a
consistent baywide approach to recognizing natural processes that will effect criteria
attainment and how to determine attainment using monitoring data and model output
given recognized limitations and uncertainties.

From VAMWA and MAM WA

Clifton Bell - Malcolm Pirnie, Inc.

Will Hunley - Hampton Roads Sanitation District

49. The PLL calculation under-estimates the potential depth of SAV survival in tidal freshwater
and oligohaline segments. Mechanistically, the PLL-based water clarity criterion is best
suited to meadow-forming SAV with most of their biomass in the lower part of the water
column, or for evaluating the light needed to establish SAV where it is currently lacking. It
appears to underpredict the depth survival of existing SAV beds in tidal fresh and
oligohaline environments, where the effective depth of water over the leaves is often
significantly less than the total water column depth. For example, Batiuk and others (2000)
found that the median PLL was less than 9 (3-8) in TF/OH segments with 'always abundant'
SAV at a depth of 1.0 m (Figure VH-3 of reference). Batiuk and others (2000) also found
that positive increases in SAV coverage occurred in the TF/OH Potomac even when the
PLW at 1 m was "considerably less" than 13. Ongoing SAV transplanting studies performed
by VIMS and the Hopewell Regional Wastewater Treatment Facility (HRWTF) show that
the PLL calculation underpredicts the potential depth of SAV survival in the tidal
freshwater James River (Moore and others, 2000; Moore and others 2001).

Response: In tidal fresh and oligohaline systems, the algorithm tends to overestimate
epiphyte biomass. The existing published algorithm is based on the best available
information both in the scientific literature as well as the most recent applicable Bay
specific laboratory and field data. Further refinement of the algorithm, as is the case

-------
with all criteria and methodologies, is possible but must await the next generation of
new data.

The Bay criteria have been verified through application of the minimum light
requirements and the supporting algorithm to Chesapeake Bay water quality
monitoring data. The results of these field data verifications are fully documented
within chapter V Epiphyte Contributions to Light Attenuation at the Leaf Surface and
Chapter VII. Setting, Applying and Evaluating Minimum Light Requirements for
Chesapeake Bay SAV in Batiuk et al. (2000) and summarized in the scientific peer
reviewed literature paper by Kemp et al. (in review). Since publication of the second
technical synthesis, attainment of the working draft Bay water clarity criteria has been
extensively tested using the full Bay water quality monitoring data base along with
direct comparison with SAV distribution and abundance time series since 1985 (e.g.,
Spring 2001 Water Clarity Criteria workshop proceeding report). The algorithm has
been validated for application to Chesapeake Bay habitats as the best measure of
water clarity.

50. Possible reasons for underprediction of the depth of SAV survival include (1) rapid leaf
growth and elongation of canopy-forming species that increases the effective PLW; (2)
overprediction of epiphytic growth; and (3) the ability of certain freshwater species to
survive on less than 9 PLL (Moore, 2000). Underestimation of the potential depth of SAV
survival can have several detrimental regulatory consequences, such as setting restoration
depths too low or overestimating the water quality needed to achieve a particular restoration
depth.

Response: Beyond the documented overestimation of epiphyte biomass, the presence of
canopy forming species can also contribute to underprediction of the depth of SAV
survival. However, in these tidal fresh and oligohaline habitats, water quality
conditions must be sufficient to support a diverse array of underwater grasses, a
number of which are not extensive canopy formers. The water clarity criteria
established for protection of tidal fresh and oligohaline habitats was set to provide
sufficient light for the full array of species, not a select set of canopy forming species
whose life strategies lend themselves to lower minimum light requirements. Finally,
light levels required to support revegetation of currently unvegetated habitats-e.g.,
upper tidal James River, are generally higher than light levels required to sustain
existing dense SAV beds. Recognizing these factors, the draft water clarity criteria can
be seen as underpredicting the depth of SAV survival for some species (canopy
formers like milfoil and hydrilla) yet possibly overpredicting SAV survival for
meadow forming species or currently unvegetated habitats. If we error on the side of
providing a bit more light then is needed to restore SAV to a depth less than 2 meters,
then we have errored on the side of a positive Bay restoration action.

-------
51. Because the PLL-based criteria do seem appropriate and reasonable for estimation of the
light needed to establish SAV beds where it is currently lacking, we support its application
even in TF/OH segments (given the ability to make site-specific modifications as
recommended in a following comment). However, we raise this issue here to encourage
additional research on how the existing models could be improved in the future to more
accurately predict the water quality needed to achieve or maintain growth at target depths in
TF/OH segments, particularly in areas where SAV is observed to grow much deeper than
the PLL calculation would indicate. Possible solutions include:

Response: The revised draft criteria document has been edited to include specific text
supporting site specific modifications where sufficient scientific data are available for a
region. Such site specific modifications to the criteria must be adopted following the
respective states' water quality standards regulations.

• Use of observed depths of SAV growth as one basis for assessment of compliance.
Response: Attainment of water quality criteria and standards must be based on water
quality conditions, not a biological resource response.

• Reduction of light requirements in areas with abundant established beds.

Response: The draft water clarity criteria are needed to support not only revegetation
of currently under vegetated habitats but also existing, established SAV beds.

• Use of the PLL calculation to calculate light at an effective depth (as opposed to total
water column depth) to consider the ability of SAV to grow up in the water column.

Response: Efforts to calculate light at an effective depth, which is interpreted here to
mean to the water column depth of the upper depth of the leaves, would be impossible
to apply. Different species of Bay grasses have very different growth strategies (leaves
elongate from the base of the plant in some and at the end of the leaves in others). In
any given grass bed, the depths of the upper most leaves would vary considerably,
making it not possible to determine the effective depth at any one time.

52. Our review of the criteria document also indicates that the existing technical uncertainties
have been documented. However, the role and impact of these referenced uncertainties on
the calculation of the criteria should receive further attention before they can be successfully
applied. We believe that the criteria uncertainties could be addressed primarily through
two concepts (1) SAV exclusion zones, and (2) site specific modifications to the criteria.

We also recognize that the criteria development process will involve additional work related
to implementation involving spatial / temporal application, naturally reduced turbidity, and
monitoring data evaluation. We are commenting on the items below related to exclusion
zones and site specific modifications because they have not been referenced in the
document thus far.

Response: See responses to comments 51 (re. site specific modifications) and 53 (re.
SAV exclusion zones).

53. Exclusion zones should be established

-------
It is recognized in the document that although light is the principle factor controlling the
distribution of SAV, other factors such as the availability of propagules, suitable sediment
characteristics, wave action, and chemical contaminants, etc. can preclude SAV even given
adequate light levels. However, the document does not yet indicate how these questions
should be addressed from a criteria application perspective. From a conceptual stand point
the criteria document should indicate that the water quality criteria are not intended for
areas which SAV are precluded due to non-water clarity based factors. This is essential
because the water clarity criteria are based on SAV as a designated use, and it is pointless to
apply the criteria in areas where SAV have no chances of establishment. The obvious
difficulty lies in determining which specific areas of the Bay littoral zone fall under this
classification. The difficulty is due to the very large amount of shoreline involved in the
Bay system coupled with a generally poor understanding of the non-water clarity factors at
work. However, a practical course of action is needed to address identification of these
exclusion zones. We recommend that the existing sources of information (historical record,
charts, existing sediment data, etc.) be used to develop a first cut series of maps that
illustrate these exclusion zones. Further, the criteria document should reference these maps
and recommend that they remain flexible and amendable to change as the level of
understanding improves. Suggested criteria document language related to this concept is
provided below:

Water clarity criteria are not intended to apply in areas where SAV are precluded by non-
water clarity related factors such as unsuitable sediment substrate, excessive wave action,
etc. These areas are denoted as SAV exclusion zones. A map of these SAV exclusion
zones developed on the best available information and professional judgement are shown in
Figure x. These exclusion zones should be periodically re-evaluated and refined as
additional information and site-specific results become available, [note: The implementation
procedures should provide guidance on the procedures and data requirements for these
studies.

Response: Areas where natural physical factors would prevent underwater grasses
from ever growing even if sufficient water clarity conditions were restored have
already been mapped out as part of the process to determine the Tier HI restoration
target first published in first SAV Technical Synthesis (Batiuk et al. 1992) and then
revised slightly and republished by Batiuk et al. 2000. The mapping of these areas will
be updated as part of ongoing work on the Chesapeake 2000 Agreement commitment
to "revise SAV restoration goals and strategies to reflect historic abundance...". A new
figure mapping out the existing delineated as no grow SAV areas has been included in
the revised draft criteria document along with additional new text.

Site-specific modifications to the criteria equations

54. The water clarity criteria do not consist of actual numbers but rather equations involving
measured water column light attenuation, TSS, DIN, and DIP. The diagnostic tools utilize
TSS and chlorophyll as a means to diagnose the causes for water column light attenuation.

This could be considered advantageous from a criteria perspective because it lends itself to
site specific refinements and improved accuracy. On the other hand, use of a generic set of

-------
equations may or may not lead to accurate conclusions and associated management actions.
Technical areas where the algorithm can be improved with site specific data include light
attenuation coefficients for background color, TSS, chlorophyll a, and epiphyte responses
associated with TSS, DIN, DIP, and grazer terms.

Response: The water clarity criteria do in fact consist of actual numbers stated in
terms of the percent ambient light at the water surface reaching the underwater grass
leaf surface at the established application depth. The draft criteria document
recognizes opportunities for further tailoring the Bay water clarity criteria to broad
regions of the tidal Bay habitats beyond salinity given additional targeted data
collection, scientific research and confirmation of improvements to the existing set of
criteria. At this time, the best available scientific research and most current Bay
monitoring program findings fully support the proposed set of salinity regime specific
water clarity criteria. Given more scientific investigations and more years of Bay
monitoring data, the underlying minimum light requirements and algorithms used in
determination of the criteria and diagnostics are infinitely improvable. Given past
experiences, it has taken about 8-10 years to generate sufficient amount of new
knowledge and understanding to invest significant resources to undertake major
revisions to the definition of underwater grasses light requirements. The Chesapeake
2000 Agreement and the six state memorandum of understanding commit the
watershed partners to use the available science and technical information to establish
the water quality conditions necessary to support the Bay's living resources now.

Adoption of more site specific modifications to the PLL algorithm would also address
the concerns raised in the above comment.

55. Epiphyte issues may be particularly acute in the tidal fresh environment. Batiuk and others
(2000) point out that "much of the information on which the [leaf-surface light attenuation]
model is based comes from the.. .mesohaline and polyhaline regions of Chesapeake
Bay.. .This is due to limited comparable data from lower salinity tidal habitats." The work
of Moore and others (2000, 2001) suggests that the PLL calculation may generally
overestimate biofouling in TF segments. It is recommended that additional research be
performed on epiphytic growth in TF/OH segments to improve the PLL calculation in low-
salinity environs. In the meantime, the water clarity criteria document should explicitly
allow the epiphytic growth component of the PLL calculation to be modified (or even
turned off) in certain locations/segments based on field observations.

Response: Further research will certainly help improve how epiphytic growth is
factored into the underlying basis for the criteria. As these new findings become
available and can be factored into a refined set of criteria, then the states should adopt
them at that time into a revised as of state water quality standards. There is no
scientifically justifiable means at this time of documenting how and under what

-------
circumstances the epiphytic growth component of the water clarity criteria could be
"modified (or even turned off) in certain locations/segments based on field
observations." We don't know how site or even regional specific differences in
epiphytic growth patterns really are at this point in time to support such detailed site
specific modification of the water clarity criteria. As these data become available, the
option exists under state water quality standards regulations to develop such site
specific modifications.

Modification of effective depth of application for established beds:

56. The criteria document references the uncertainties exist does not indicate how they would
be addressed in the criteria process. We believe they should be addressed as site specific
modifications to the criteria. Provisions for site specific modifications are common in
many water quality criteria / standards. Examples include hardness adjustments to metals
criteria and water effect ratios for copper. The provisions generally operate by recognizing
that the criteria / standard may not be appropriate for application in all areas (within the
criteria document) while outlining procedures / methods to carry out the sites specific
modifications. Suggested criteria document language related to this concept is provided
below:

The water clarity criteria may not be appropriate for application in all areas. During the
development of the criteria it was recognized that uncertainty exists in elements of the
water clarity equations (i.e. PLL) and the associated diagnostic tools. Site specific
modifications to the criteria equations are recommended to address these uncertainties.

[note: The implementation procedures should provide guidance on the procedures and
data requirements for site specific studies and modifications].

Response: Text has been added to revised draft criteria document recognizing the
option to develop site specific modifications to the criteria as a part of the states' water
quality standards regulations.

Other implementation issues

57. The criteria document references that spatial / temporal application, naturally reduced
turbidity, and monitoring data evaluation are important elements to consider during the
criteria implementation development phase. We concur and look forward to additional
discussions on these topics. In the near future we will forward the results of additional
water quality studies and SAV projects to assist in these discussions.

Response: These issues and their influence on application of the water clarity criteria
and determination of criteria attainment have been addressed in the implementation
guidelines incorporated into the revised criteria document.

References

Batiuk, R. A., P. Bergstrom, M. Kemp, E. Koch, L. Murray, J. C. Stevenson, R. Bartleson, V.

Carter, N. B. Rybicki, J. M. Landwehr, C. Gallegos, L. Karrh, M. Naylor, D. Wilcox, K. A.

-------
Moore, S. Ailstock and M. Teichberg. 2000. Chesapeake Bay submerged aquatic vegetation
water quality and habitat-based requirements and restoration targets: a second technical
synthesis. CBP/TRS245/00 EPA 903-R-00-014. USEPA Chesapeake Bay Program, Annapolis
MD.

Moore, K.A., R. Orth, J. Fishman. 2000. Restoration of submerged aquatic vegetation (SAV) in
the tidal freshwater James River: A pilot study. Report submitted to the Hopewell Regional
Wastewater Treatment Facility. 21 p.

Moore, K.A., K. Segerbloom, B. Neikirk. 2001. Restoration of submerged aquatic vegetation
(SA V) in the tidal freshwater James River: Year 2. Report submitted to the Hopewell Regional
Wastewater Treatment Facility. 19 p.

From CBF

58. In earlier correspondence, we have stated our position that all Chesapeake Bay segments
should be established at 2 meter (or greater) water clarity criteria depths. At that point, we
had not seen any of the historical depth data that is being developed through VIMS and MD
DNR. After reviewing the preliminary historical information, we generally feel this data
should be used to designate these application depths, within the following guidelines:

Response: A summary of the 1940s-1960s Bay grasses distributions interpreted from
historical photography was summarized in Table IV-7 in the July 3, 2001 working
draft criteria document.

59. A recognition that some of the historical photos that are being interpreted for this purpose
may not be totally accurate (i.e. quality of photographs due to time of year or turbidity of
water, anomalies, etc.) and that some generalization must be made within each tributary
basin. Additionally, other sources of historical photographs should be sought for areas
where the USD A photographs are not adequate.

Response: The Virginia Institute of Marine Science and Maryland Department of
Natural Resource scientists working on these historical photographs have sought
photographs from a variety of different sources, given constraints on the timing of the
photography and sufficient landmarks to georeference the photographs. The revised
draft criteria document has an expanded description of the origin of the photographs,
how the information they provided was used in establishing the proposed application
depths, and limitations in their use as briefing described in the above comment.

60. There must be consistency between river systems that are similar in characteristics. For
example, we feel the Severn (SEVMH) and the South (SOUMH) in Maryland and the Upper
James (JMSTF) and the Appomattox (APPTF) in Virginia should reflect the same
application depths.

Response: The information available for each river system and each Chesapeake Bay
Program segment within the larger river systems was first analyzed separately with
the established decision rules applied separately to each segment. In cases where there
were large differences in application depths between directly adjacent river systems

-------
along the lines outlined in the above comment, reasons for proposing different
application depths were then documented.

61. Some of the segments in the Virginia section of the Bay are large in area compared to
Maryland segments (i.e. CB7PH). To best represent these large segments and other
segments that may have several distinct zones of historical SAV, we would ask that the
water clarity depths be estimated using less than the conservative "10% of segment below a
certain contour" to estimate the application depths.

Response: The large segments in the lower Chesapeake Bay mainstem all have
recommended application depths of 2 meters given as well as recent and historical bed
depth patterns. Any changes to the use of a 10 percent cutoff value to a less
conservation value would not effect the recommended application depths for these
lower Bay segments.

62. Depths greater than 2 m should be specified when documented from historical information.
Response: Those segment where historical bed depth data indicated growth down
below the 2 meter depth have been footnoted in the updated table displaying the
revised proposed application depths in the revised draft criteria document for
informational purposes only.

From Virginia DEQ

63. Efforts toward revising Virginia's water quality standards, as an outcome of this process,
will have the benefit of correcting deficiencies in the existing dissolved oxygen standard.
One such deficiency was the definition and use of natural conditions. The Commonwealth
has no interest in creating similar problems with the other criteria under development for
water clarity and chlorophyll, such as applying unobtainable clarity requirements in areas of
turbidity maximum and naturally high sediment resuspension, or stringent chlorophyll
levels in areas with historically elevated concentrations, again due to natural conditions
(e.g., in small embayments and creeks).

Response: In the case of the water clarity criteria, the publication of a set of consistent
implementation guidelines to be applied across all tidal Bay waters should address the
general concerns expressed in the above comment.

64. While use of 'Percent Light at the Leaf (PLL) seems to be the consensus of the Technical
Workgroup, it is not certain whether PLL or 'Percent Light through the Water' (PLW) is a
better indicator and representation of the criteria's purpose - SAV growth and survival in
shallow water. This is especially true in the tidal fresh environment where the nutrient -
epiphyte growth connection does not seem to be very strong.

Response: To ensure the wide variety of underwater grass species living in tidal fresh
to close to ocean salinity waters across all tidal habitats get the required amount of
light, the working draft water clarity criteria were based on the percent light reaching
the leaf surface. The role of epiphytes in blocking light from reaching the leaf surface
is well established in the scientific literature and through two decadal scale
Chesapeake Bay technical syntheses. The criteria document recognizes variability in
the overall contribution of epiphytes to reducing light reaching the leaf surface across

-------
different habitats and even salinity regimes. However, the best available scientific
information at this time fully support factoring epiphytic growth into the
determination of criteria attainment.

Given the additional parameters required to calculate the percent light to leaf and
existing and documented variability with the epiphyte contribution, the revised draft
water clarity criteria document proposes publishing both the percent light at the leaf
and the percent light through the water criteria values for both salinity regimes as Bay
criteria. The implementation guidelines provide guidance as to where/when to apply
PLL vs. PLW criteria.

65. While the CBP Water Quality Model may be used to simulate SAV growth and assess
attainment of restoration goals, it would not provide sufficient resolution for use on small
temporal/spatial scales.

Response: The small temporal and spatial limitations of the Bay water quality model
have been documented by the Chesapeake Bay Program's Modeling Subcommittee
overall and on a CBP segment by segment basis using the calibration/verification
findings.

66. Another issue is the use of the "application depth" in reference to water clarity criteria.

Rather than implement such a complex system that changes by segment, the application
depth should be set using the existing Tier I goal for SAV restoration. This agrees with the
C2K agreement by the signatories to protect and restore 114,000 acres of SAV. We also
suggest that actual, observed SAV acreage should be considered a replacement for water
clarity as a criteria in the assessment process, and we expect further discussion on this issue
as the Bay Program partners proceed with criteria development.

Response: First, the Chesapeake 2000 Agreement commits the signatories to "revise
SAV restoration goals and strategies to reflect historic abundance...". The existing
Tier I goal is likely to be replaced with a baywide goal more reflective of Bay grass
distributions prior to the early 1970s. The Chesapeake 2000 Agreement also commits
to setting a new SAV restoration goal tied directly with the water clarity criteria and
the underlying application depth. Therefore, the application depth can not be set
using the existing Tier I goal for SAV restoration. Second, given the water clarity
criteria will be adopted by the states as water quality standards, the attainment of a
standard can not be determined based on observed SAV acreage but on actual
measurement of water clarity conditions.

From Maryland DNR

Tidal Fresh and Oligohaline Areas

67. The technical documentation is weak for tidal fresh and oligohaline areas which have
canopy-formers that do not depend upon light penetration for most of the season. It is not
clear to what extent the criteria for tidal fresh and oligohaline zones is based on the light
requirements of canopy-forming species such as hydrilla (hydrilla verticillata). Using the

-------
light requirements of canopy-forming species could result in much lower light requirements
than are necessary for non-canopy forming species, or could result in more narrow temporal
periods during which light limitation is critical.

Recommendation

Tidal fresh and oligohaline criteria should be re-examined and future drafts of this
document should include explicit information on the species that are driving the criteria in
these zones and address the specific growth requirement of both canopy and non-canopy
forming species.

Response: The underlying scientific and technical documentation for the proposed
water clarity criteria-Batiuk et al. 2000 and many scientific peer reviewed journal
papers published from the individual technical synthesis chapters-clearly states that
the minimum light requirements set for protection of tidal fresh and oligohaline
habitats were based, in part, on the requirements of the more sensitive non-canopy
forming species. The revised criteria document text has been expanded to incorporate
this documentation.

Consideration of Criteria During Critical Growth Periods

68. During high flow seasons and storm events, concentrations of total suspended solids can
reach very high levels and water clarity can be severely degraded. In addition, algal bloom
during spring and summer, including colony forming species, can greatly reduce light
transmission for periods of days to weeks. These episodic events can have serious impacts
on SAV growth and survival, especially during the critical spring emergence period, but are
unlikely to be adequately captured with season-long averages or medians.

Recommendation

Requirements of species (including all salinity zones) during short-term growth periods
when water clarity may be particularly critical (e.g. spring ) should be examined and
considered for inclusion in future drafts of these criteria in addition to season-long averages.

This would be analogous to temporally-tiered criteria for dissolved oxygen.

Response: The issue of short term responses to reduced water clarity conditions was
addressed by Batiuk et al. (2000) in the second SAV technical synthesis.
Unfortunately, there was insufficient scientific data at the required small spatial and
short term temporal scales to support add a more time period specific element to the
draft Bay water clarity criteria beyond season medians. The implementation
procedures for determining attainment of the Bay water clarity criteria includes
calculation of monthly PLL values across the respective SAV growing season to help
account for shorter term periods of poor water clarity in overall criteria attainment.

Depth for SAV Restoration in Different Areas

69. Using the 1971 to 1999 SAV coverage to establish restoration criteria could limit restoration
goals in many areas since much of the Bay's SAV community had experienced massive
declines by this time period.

Recommendation

As historical SAV coverage information becomes available from ongoing analyses in MD

-------
and VA, use these data as the primary basis for establishing restoration target depths
between 0.5 and 2.0m. When these data become available, the criteria for determining the
greatest depth of restoration can be made more robust. Currently, if 5% of mapped beds
within a segment are within the deepest zone, that zone is included for restoration. This
level is certainly within the error limits of bathymetry and SAV mapping and may lead to
spurious conclusions.

Response: The historical Bay grass bed depth distribution data being generated by
colleagues at Maryland Department of Natural Resources and the Virginia Institute of
Marine Sciences was directly used as one of two principal sets of information in
developing the draft application depths first published in the July 3, 2001 working
draft criteria document. Two decision rules were applied to establish the maximum
depth value for each Chesapeake Bay Program segment, employing 5 percent and 10
percent cut off values. See page 13 in the working draft document for details.

Attainment of Criteria

70. The degree of attainment (e.g., percent of violations allowed, statistical certainty of
violation, etc.) has yet to be defined. What statistical test(s) will be used to determine if the
criteria have been met? How many samples will be needed to determine if the criteria have
been met at a predetermined level of statistical certainty? What level of statistical certainty
will be acceptable?

Recommendation

In future drafts and the final version, include explicit guidance on the parameters that would
define attainment and non-attainment.

Response: The revised draft criteria document incorporates a comprehensive set of
implementation guidelines addressing the requested needs.

Monitoring

71. Most of the current monitoring stations in the tidal tributaries are monitored in the middle of
the channel either monthly or bimonthly, depending on the time of year. What changes will
be needed in the current monitoring program in terms of sampling frequency and location to
determine whether or not the water clarity criteria have been attained? To what extent can
mid-channel data serve to characterize shallow-water habitats?

Recommendation

Define minimum monitoring needs to provide sufficient data to assess criteria attainment
and non-attainment

Response: The issue of use of mid-channel to characterize shallow water habitats has
been addressed in the implementation guidelines incorporated into the revised criteria
document. Efforts are underway in the Chesapeake Bay Program to better address
the need for monitoring data which better characterize shallow water system through
a new tidal monitoring network design.

From Pennsylvania DEP

-------
72. Need to clearly define what is calculated by the Water Quality Model (PLL, PLW) and what
the variables are (Kd, Ke DIN, DIP??) in those calculations as the model will be the tool
used to define attainment of the water clarity criteria. PLL/PLW requirements should be
calculated using the same algorithms as the model if possible, or at least using consistent
values for parameters common to each calculation.

Response: The implementation guidelines, incorporated into the revised criteria
document, spell out in detail how the Bay water quality model output is to be used to
present simulated water clarity criteria attainment. Recognizing differences in
temporal and spatial scales between Bay monitoring data and Bay model simulated
output, every effort has been made to calculate water clarity criteria the same way
using both observed data and model simulated output for the four variables.

73. The algorithm used to compute percent light reaching SAV depths uses a Kd = 2.0 m"1 for
tidal fresh and oligohaline regions and Kd = 1,5 m~' for mesohaline and polyhaline regions.

This algorithm appears to be the one used to generate the 13 and 22% light requirements,
which are then adjusted to account for epiphytic attenuation to arrive at the 9 and 15% PLL.
Are the Kd values used to generate these requirements the same as those used in the Water
Quality Model? Is it realistic to assume that this value is constant across all tidal fresh and
oligohaline segments?

Response: The algorithm computes the percent light reaching the Bay grass leaf
surface using what ever restoration depth the user decides to apply. In the case of the
proposed water clarity criteria, the restoration depths are the proposed application
depths, which vary CBP segment by segment.

74. It is not clear from the text, table or figure where the 8 and 15% (average of 13 and 17%)
light requirement comes from in the Calculation Using the 1992 SAV Habitat Requirements
section (pg. 10). The text shows the addition of the Ke and Be terms to Equation IV-1 to get
IV-2, but fails to describe how those values were derived in order to calculate the previously
mentioned light requirement. Additionally, the text in the same section references Table
IV-1 for SAV Habitat Requirements used. This table shows a Chl-a habitat requirement of
<15 (ig/1, but the median Chi-a criterion proposed in the Chi-a document are substantially
lower than this. Should Habitat Requirements be replaced by proposed criteria when
available?

Response: The revised criteria document text has been edited to make it more clear as
to the origin of the 9 percent and 15 percent criterion values. The Bay specific
chlorophyll a criteria were derived independently of the 1992 SAV habitat
requirement for chlorophyll a.

75. Use consistent terminology when writing algorithms in text (see exponential representation

in Equations IV-1 and IV-2). Figure IV-2 has them the same but does not have the negative
required in front of the exponent.

Response: The revised criteria document text, tables, equations, and figures have all
been edited to reflect consistent terminology through the document.

-------