HomeMy WebLinkAboutSuWa304
Alaska Resources Library & Information Services
Susitna‐Watana Hydroelectric Project Document
ARLIS Uniform Cover Page
TK
1425
.S8
S92
no.304
Title:
SuWa 304
Letter from Wayne Dyok to Socheata Lor, October 8, 2014, in response to
U.S. Fish and Wildlife Service's comments on Initial Study Report on
Susitna-Watana Hydroelectric Project
[Title devised by cataloger.]
Author(s) – Personal:
Wayne Dyok (writer of cover letter)
Author(s) – Corporate:
Alaska Energy Authority
AEA‐identified category, if specified:
AEA‐identified series, if specified:
Series (ARLIS‐assigned report number): Existing numbers on document:
Susitna-Watana Hydroelectric Project document number 304
Published by: Date published:
Alaska Energy Authority October 8, 2014
Published for: Date or date range of report:
Socheata Lor ; U.S. Fish and Wildlife Service, Anchorage Fish and
Wildlife Field Office
Volume and/or Part numbers: Final or Draft status, as indicated:
Document type: Pagination:
Letter with attachments 42, 20, 4, 2 pages
Related work(s): Pages added/changed by ARLIS:
Response to: Letter from Socheata Lor to Wayne Dyok, October 8,
2014, providing U.S. Fish and Wildlife Service's comments on the
Initial Study Report for the Susitna-Watana Hydroelectric Project.
(SuWa 303)
SuWa 303 was a comment to: Initial Study Report. (SuWa 223)
Notes:
All reports in the Susitna‐Watana Hydroelectric Project Document series include an ARLIS‐
produced cover page and an ARLIS‐assigned number for uniformity and citability. All reports
are posted online at http://www.arlis.org/susitnadocfinder/
.)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)
)·
)
)
)
)
)
)
)
)
)
October 8, 2014
Socheata Lor
Anchorage Field Supervisor
United State Fish and Wildlife Servic e
605 West 4th Avenue, Room G-61
Anchorage, Alaska 99501-2250
Re: Susitna-Watana Hydroelectric Project, FERC Project No. 14241-000
Dear Ms. Lor:
The Alaska Energy Authority (AEA) is in receipt of a letter from the United States
Fish and Wildlife Service (USFWS) dated September 22, 2014,1 in which you provide
comments on portions of the Initial Study Report (June 3, 2014) (ISR) for the proposed
Susitna-Watana Hydroelectric Project, Federal Energy Regulatory Commission (FER C)
Project No. 14241 (Project).
Your letter raises three topics of concern: 1) data collection and reporting, 2)
effective model integration, and 3) development of a decision support system (DSS).
Your letter states that it is important that these issues be resolved prior to conducting
additional field studies.
We respectfully disagree with your comments. As documented in the ISR, AEA
was largely successful in implementing the FERC-approved study plan in 2013. This
effort included, among many other studies, a large-scale field effort for fishery studies
with a suite of 10 studies covering more than 200 sampling sites across more than 200
miles of river, with sampling occurring during not only the open water period but also
during winter and spring periods. Your letter, however, focuses on the limited exceptions
in which AEA' s data collection varied from FERC-approved study plan methods during
the 2013 field season. These variances, as we all know, occurred mostly due to private
land access issues, and conditions in the field such as the late ice breakup in the spring of
2013. The ISR includes a detailed description of proposed modifications to the study
plan to account for these variances.
Letter from Socheata Lor, United States Fish and Wildlife Service, to Wayne Dyok, Alaska Energy Authority,
Project No. 14241-000.
Attached to this letter is a comment-response table that addresses in detail each of
the concerns and comments in your September 22 letter. I think you will agree, on
careful review of our responses, that these responses address your concerns and that the
2013 study program provides a solid foundation of data upon which we can continue to
build. We look forward to continuing this dialogue in the upcoming ISR meetings.
ABA remains committed to implementing the comprehensive suite of studies
proposed in the PERC-approved study plan and encourages USFWS to continue working
with us in studying the feasibility of and potential effects associated with an undertaking
that is critically important to Alaskans. If you have questions or comments concerning
this matter, please feel free to contact me directly at (907) 771-3955.
Attachment
Cc: Distribution List
Ellen Lance
Betsy McCracken
Phil Brna
Jeff Wright
Ann Miles
Vince Yearick
Dr. Jennifer Hill
Nick Jayjack
Sincerely,
IJ~~22Z~:f/4}~
WayneDyok
Project Manager
Alaska Energy Authority
2
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Page and Paragraph Numbering
• Partial sentences at the top of a page are considered Sentence 1.
• Partial paragraphs at the top of a page are considered Paragraph 1.
• Paragraphs are numbered by their position on a page, not within a Section.
• Paragraphs are blocks of text separated by hard returns; each heading, bullet, and item in a numbered list is considered one
paragraph.
Comment Comment
Page, Number Comment
Para
PageS 1 • As currently planned, some two-year
Para4 studies cannot be completed because access
to all Focus Areas (FAs) was not granted
until after the first study year (e.g., ISRs
8.S, 9.6, 9.7, 9.9). For example, a fish
wheel was not installed and fish were not
tagged near the entrance to Devil' s Canyon
(e.g., ISR 9.7).
PageS 2 • Anomalous weather conditions prevented
ParaS or delayed fieldwork on aquatic studies
(e.g., ISR 8.S), resulted in late installation
of migrant traps, which were likely
influenced by environmental conditions
associated with late breakup (e.g., ISR 9.6).
Moreover, juvenile salmon distribution and
abundance measured in 2013 were likely
affected by the record fall floods in 2012
(e.g., ISR 9.6).
PageS 3 • Sampling has not been temporally
Para6 adequate across all seasons. ISR 9.6 reports
winter fish sampling did not occur across all
F As as proposed; early spring sampling
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
As stated in the ISRs for Studies 9.S, 9.6, 9.7, and 9.9 all of the site where access re&tricted sampling
in 2013 were sampled in 2014. As indicated in ISR Sections 9.S.7 and 9.6.7, the second year of data
for these studies that require two years of study will be conducted in 201S. The Salmon Escapement
Study 9.7 was successful at collecting sufficient data to address study objective for three consecutive
years 2012, 2013, and 2014.
This comment ignores the data and analysis presented in the ISR. The variance for not installing a
fish wheel at the entrance to Devils Canyon is described in Study 9. 7 ISR Section 4.1.8.1. This
change in tagging location was compensated for by increased :fishwheel effort and an increase in
tagging targets at the Curry fishwheels.
Downstream migrant traps were installed and operated as indicated in the Study 9.5 ISR Section
9.S.4.4.10 and Study 9.6 ISR Section 9.6.4.4.10: "flow conditions permitting, traps will be fished on
a cycle of 48 hours on, 72 hours off throughout the ice-free period." As soon as break-up and flow
conditions allowed in mid-June 2013 traps were installed and fished immediately upon installation in
June through mid-October 2013. In 2014 breakup occurred earlier and migrant trap installations
occurred in mid-May with traps operated immediately after installation (the Proposed 201S
Modifications to Fish Distribution and Abundance Study Plan Implementation Technical
Memorandum filed withFERC on September 17, 2014).
ABA agrees that floods can affect juvenile salmonid distribution. While the Fall2012 floods did not
approach the magnitude of the flood of record, they potentially distributed juvenile salmonids into
lateral habitats that may not otherwise be occupied during a low water year. ABA believes that the
range of hydrologic events that occur over the multi-year study period provide opportunities to better
understand the response of aquatic resources to flow fluctuations.
Fish sampling followed the sampling plan. RSP Section 9.6.4.1 states that "winter sites will be
selected based on information gathered during 2012-2013 pilot studies ... attempts will be made to
sample all Focus Areas." The winter pilot study was conducted in Winter 2013 at two Focus Areas
as described in Study 9.6 RSP Section 9.6.4.S. ABA made recommendations based upon the winter
Page 1
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
occurred only in three F As; initial sampling
following breakup and installation of
migrant traps did not occur until the middle
of June, and therefore, spring sampling for
fish distribution and abundance was not
conducted (e.g., ISRs 7.5, 8.5, 8.6). The
extent to which fishes move must be
described through sampling; multiple
sampling days across all seasons are
required to capture the full seasonality of a
fish's life-history strategy, which varies
considerably within a single season. A
single-day of sampling is insufficient to
understand the habitat associations of
different fish species with differing mobility
and life-stages.
Page 5 4 • Sample site selections for integrated
Para 7 studies were inconsistently co-located. For
example, invertebrate sampling locations
(ISR 9.8) were not co-located with fish
sampling locations (ISR 9.6). Failure to co-
locate sampling sites risks the magnification
of data discrepancies, and because the data
will be used as inputs for predictive models,
may jeopardize the validity of the models.
Page 5 5 • Detection arrays did not cover the entire
Para 8 channel and tagging efforts did not allow
for detection of fish migrating upstream,
therefore the data were biased and
Sus1tna-Watana Hydroelectnc ProJect
FERC Project No. 14241
Response
pilot study for sampling sites, as stated in the Study 9.6 ISR Appendix C Section 6.1.1, and the 2014
Winter Study was expanded to three Focus Areas and opportunistic sampling at accessible sites
outside of the Focus Areas. Results of the first year ofthe winter study for fish are presented in the
Study 9.5 Winter Study Technical Memorandum filed with FERC on September 17, 2014.
In 2013 Early Life History (ELH) sampling began two weeks after winter sampling was stopped and
continued bi-weekly through June with the exception that no sampling was conducted for two weeks
during the dynamic break up in mid-May 2013 (Study 9.6 ISR Section 4.6). As stated in Study 9.6
ISR Section 4.6.2, ELH sampling included 6 Focus Areas identified to have both spawning and
rearing habitat as well as additional sites in the Upper (Study 9.5 ISR 4.6.2), Middle and Lower River
(Study 9.6 ISR 4.6.5). Sample sites for these various fish study components were visited multiple
times during the Winter Study (1-3 times), Early Life History Study (3 times), and Fish Distribution
and Abundance Study (3 times). Some sites were visited during all three seasonal study components
and ended up being sampled more than eight times in 2013.
As an initial matter, the RSPs never specified the co-location of sample sites across study disciplines.
It did specify the location of 10 specific Focus Areas that would be evaluated relative to the different
resource disciplines and study sites across disciplines were co-located within the Focus Areas
Furthermore, this comment ignores the data and analysis presented in the Study Plan. AEA's
selection of sampling sites was consistent with the River Productivity Implementation Plan. As
presented in the River Productivity Implementation Plan Section 2.1: "All stations established within
the Middle River Segment will be located at Focus Areas established by the Instream Flow Study
(AEA 2012, Section 8.5.4.2.1.2), in an attempt to correlate macro invertebrate data with additional
environmental data (:flow, substrates, temperature, water quality, riparian habitat, etc.) collected by
other studies (e.g., AEA 2012, Section 5.5, Baseline Water Quality), for uses in statistical analyses,
and HSC/HSI development. Furthermore sites for Fish Distribution and Abundance, Habitat
Suitability Criteria, and River Productivity were all co-located within Middle River Focus Areas. In
2013, private land access restrictions prevented fish sampling in some desired locations, yet River
Productivity sampling was able to be conducted because the sites for that study were located in
mainstem and within ordinary high water. Maps depicting the co-locations of sampling sites among
these three studies will be presented in the October 15, 2014 ISR meeting.
This comment reflects a fundamental lack of understanding of the methodologies being relied upon
by the FERC-approved study plan. As stated in RSP Sections 9.5.4.4.1.2 and 9.6.4.4.1.2, remote
telemetry techniques were "intended to provide detailed information on relatively few individual
fish." PIT tags were used to "document relatively localized movements offish as well as growth
Page2
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
efficiency estimates cannot be calculated.
Detection rate and recovery of passive
integrated transponder (PIT) tags is
insufficient to yield useful data to meet
study goals and objectives (ISR 9.6).
Page5 6 • Fish targets for fish Habitat Suitability
Para9 Curve (HSC) sampling were not met (e.g.,
ISR 8.5), therefore, power to assess fish
habitat-preferences and relationships is
reduced.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
information from tagged individuals." Due to the large size of rivers in the study area, the necessity
for installing arrays across split channels, side-channels and/or as partial coverage array across a
portion of the main channel was described in the Fish Distribution and Abundance Implementation
Plan Section 5.6.5. Furthermore, the PIT tag arrays spanned the entire channels located in FA-104
(Whiskers Slough) and FA-128 (Slough 8A).
Data from PIT tag arrays provided limited,but valuable information on fish movements. As indicated
in Study 9.5 ISR Section 5.2.2.2 and Study 9.6 ISR Section 5.2.2.2, antenna arrays recorded 29,047
detections of33 fish in the Upper River and 126,351 detections of664 fish at Middle River arrays.
These resightings provided information on local and inter-stream movements of individuals for six
species in the Upper River and 11 species in the Middle River as well as site-specific growth rates for
individuals of several species (Study 9.6 ISR Section 5.5 .1 ).
This comment ignores the data and analysis presented in the ISR and reflects a fundamental lack of
understanding of the methodologies being relied upon by the PERC-approved study plan. AEA notes
that absolute target numbers were never established for HSC data collection (see RSP 8.5.4.5.1.1.5).
What was noted was that "If possible, a minimum of 100 habitat use observations will be collected
for each target species life stage. However, the actual number of measurements will be based on a
statistical analysis that considers variability and uncertainty. While information will be collected on
all species and life stages encountered, the locations, timing, and methods of sampling efforts may
target key species and life stages identified in consultation with the TWG." This was discussed
during several TWG meetings where it was emphasized that the approach AEA is taking in
developing HSC curves will include several components, including collection of new site specific
data, which is AEA's and agencies preferred approach, as well as other approaches for species or life
stages infrequently encountered. AEA listed those in RSP 8.5.4.5.1.1 and included use of existing
site specific data collected during the 1980s studies, use of site specific data from other similar
Alaska systems, as well as professional opinion.
A summary ofHSC collection efforts to date is provided below. As noted, there are a number of
species for which the numbers of observations have exceeded 100, including those for Chinook
juvenile, chum fry and spawning, coho fry, sockeye fry and spawning, Arctic grayling fry, and
whitefish fry. These species and life stage mixes reflect the majority of the target species and life
stages that are central to the habitat-flow modeling for evaluating Project effects.
Page3
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
Species
Chinook Salmon
Chum Salmon
Coho Salmon
Pink Salmon
Sockeye Salmon
Arctic Grayling
Burbot
Page4
Lifestage 2013
Fry 54
Juvenile 38
Fry 14
Spawning 348
Fry 99
Juvenile 56
Fry 0
Spawning 59
Fry 79
Spawning 181
Fry 113
Juvenile 43
Adult 4
Juvenile 2
2014 Project
Through July Total
164 218
25 63
258 272
348
181 280
28 84
39 39
0 59
299 378
181
7 120
9 52
4 8
4 6
Alaska Energy Authority
October 2014
1980s
Total
333
NR
81
140
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
Adult 17 3 20 18
Dolly Varden Fry 20 20
Adult 1 1 2 2
Longnose Sucker Fry 41 46 87
Juvenile 52 27 79
Adult 70 3 73 157
Rainbow Trout Juvenile 5 2 7
Adult 6 1 7 143
Whitefish Fry 39 73 112
Juvenile 39 15 54
Adult 29 4 33 384
For some species and life stages, the 2013 site-specific data discussed above may be used in
development of final curves. Additional HSCIHSI sampling is planned for the next year of study and
it is anticipated that most HSC relationships will be updated. However, for species and life stages
that are rarely observed, final HSC curves may be based on additional data, including utilization data
from 2012 and the 1980s studies on the Susitna River. Even then, there may still be some species
where few or no empirical HSCIHSI data were able to be collected. In those cases, AEA will
consider other methods for developing curves. This may include the use ofliterature based curves,
developing envelope curves (see, for example, Jowett et al. 1991, and GSA BBEST 2011), guilding
(e.g., creating a combined HSCIHSI curve representing multiple species and/or life stages; see, for
example, Vadas, Jr. and Orth 2001, GSA BBEST 2011), developing curves based on expert
PageS
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Page6 7 • Data collected on fish habitat for the Fish
Para 1 Passage Barrier Study (ISR 9 .12) and the
HSIIHSC component of the fish and aquatic
Instream Flow Study (ISR 8.5) were
gathered at incompatible spatial scales to
meet the study objectives.
Page 6 8 • Water quality samples were qualified as
Para3 either estimated or rejected by the analytical
laboratory due to quality-related failures
(ISR 5.5). Issues included failure to deliver
samples to the laboratories within the
method-specified temperature range; failure
to meet procedure specified holding times;
contaminated or missing field, trip, and
method blanks; and Chain of Custody and
bottle labeling discrepancies. AEA
proposed to apply a correction factor to the
2013 data to render it useable, but provided
no details on how that would be done.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
opinion/round table discussions) and the use of Bayesian statistical methods for updating data
distributions (see, for example, Hightower 2012).
This comment reflects a lack of understanding of the methodologies being relied upon by the FERC-
approved study plan. The data collected by HSC is not needed for analysis of fish barriers.
However, Fish Barriers and IFS studies are collaborating in a number of other ways including:
evaluating target species, in the development of passage criteria that meet model outputs, and to
ensure overlap in sampling locations. This collaboration will ensure that the model outputs from IFS
are applicable to the analysis of depth and velocity passage barriers. This comment ignores the data
and analysis presented in the ISR.
Water quality samples with those "qualifiers" identified by USFWS were not the reason why select
parameters were rejected and required re-sampling in 2014. The primary reasons why select water
quality parameter results were rejected was due to: matrix interference in turbid waters, recovery of
matrix spike much higher than acceptance limits, laboratory split sample results exceeded acceptance
limits, and potential for sample preservative appearing as target analyte among other criteria used to
validate and verify quality of results. A link to a table describing qualifiers applied to specific
analytes is in the companion GINA document provided to Licensing Participants in June (as cited in
ISR Study 5.5; Part C, Section 7.1.2).
Data rejected from 2013 results will be corrected following evaluation of multiple strategies for
determining the nature of the difference between 2014 results and 2013 results. The strategy for
identifying individual correction factors may vary among the water quality parameters rejected from
the 2013 sampling effort. Specific methods that will be evaluated for each rejected parameter will be
similar to the approach described by Stuart (2002) where independent surrogate variables
(e.g., periphyton biomass, total suspended solids, nutrient concentration, flow and solar radiation)
related to the target analyte are used to examine the time series of data collected during 2013. The
approach will first test for normality in the surrogate parameter data distribution (i.e., application of
Chi-square goodness of fit test) on transformed and non-transformed data. A linear regression will be
developed between surrogate parameter and target analyte from the 2014 collection effort to
determine significant f2 predictive relationships. It is expected that surrogate parameters could
include established ratios (e.g., SRP : TP) or differences between total and dissolved metals samples
from select portions of the data set (e.g., those from low turbidity sample sites where matrix
interference does not occur in the Susitna Basin or from published literature describing the same).
The surrogate parameter will have a known synchronous (direct or indirect) relationship with the
target analyte as a test in suitability for use. A correction factor may also be derived through simple
comparison of multiple paired differences between 2013 and 2014 results for each water quality
parameter. The tests for identifying correction factors for each 2013 parameter will not begin until
the 2014 data have undergone the complete data validation/verification procedure as required by
Page6
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment
Page,
Para
Page6
Para4
Comment
Number
9
Comment
• There is evidence that juvenile salmon
may have been misidentified. A
comparison of juvenile fish collections from
the Susitna River in the 1980s (Alaska
Department ofFish and Game 1983 as cited
by R2 Consultants in the Fish Population
Summary Document), local Alaskan rivers
(Alaska Department ofFish and Game,
unpublished data; Davis et al. 2013), recent
studies on the Susitna River (Kirsch et al.
2014 ), and nearby tributaries (Miller et al.
2011 ), signal substantial differences in total
fork length distribution and habitat
associations among juvenile salmon from
that which is expected. Large numbers of
unidentified salmonidjuveniles (some of
which were PIT tagged), anomalous length
distributions and questionable habitat
associations decrease our confidence in the
accuracy of species identification. For
example, juvenile Chinook salmon
measuring 150 mm fork-length were
reported, juvenile Chinook salmon were
reportedly most abundant in beaver ponds,
there was absence of pink salmon in any
samples, and a disappearance of sockeye
salmon from Indian River between the
February draft ISR and the June draft ISR
We have strong reservations about the
identification of these juvenile fish, and
suspect many juvenile salmons identified as
Chinook salmon may be coho salmon.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
ADEC in compliance with the credible data policy.
In Study 9.6 ISR Table 5.1-2: 865 undifferentiated Pacific salmon Juveniles in MR., five percent of
all juvenile salmon,-half from Slough 6A. 436 fish have been identified after photo review and
classified to species. Resulting in a total of 429 undifferentiated Pacific salmon remaining in
database, 2.5 percent of total.
In Study 9.46 ISR Table 5.1-3: 78 undifferentiated Pacific salmon juveniles in LR, two percent of
total. AEA is in the process of reviewing photos from the Lower River, which should reduce the
number of unidentified juvenile salmonids.
In 2013, 11 undifferentiated pacific salmon were PIT-tagged (67 reported in ISR but photo review
resulted in identification of 56 ofthe 67); 4 ofthese 11 tagged tmidentified pacific salmon met length
criteria to be two-year-olds. Ten of these 11 fish have photos that are under review. In total1,872
Chinook salmon and 2,793 Coho salmon were PIT-tagged in 2013 and Winter 2014.
Pink salmon were caught during winter sampling and ELH sampling. Winter data are provided in
Study 9.6 ISR Appendix C Tables C2.2-5 and C2.2-5 and Figure C A1-17. ELH data are provided in
Study 9.6 ISR Tables 5.3-1, 5.3-2, and 5.3-3.
Summary oflarge juvenile Chinook and coho salmon. Based on growth modeling, juvenile Chinook
and coho salmon> 1 OOmm in May and June were presumed to be two-year-old fish and> 120mm
from July-April were presumed to be two years of age. These data are not consistent with data from
the 1980s and are undergoin additional analysis.
Pacific
Location PRM Habitat Chinook Coho salmon, Total salmon salmon undifferen
tiated
DMT-Talkeetna
Station 106.9 MS Susitna River 72 8 3 83
Indian River
DMT 142.1 Tributary 70 4 74
FA-141-Slough Upland Slough
17 142.3 Beaver Complex 70 16 1 87
Montana Creek
DMT 80.8 Tributary 37 4 41
FA-104-Slough Upland Slough
3A 105.7 Beaver Complex 15 25 1 41
Alaska Energy Authority
Page7 October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 ISR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
FA-104-SS 105 Side Slough 14 2 16
Upland Slough
PRM-63.5-US 62.5 Beaver Complex 9 11 20
FA-115-Slough Upland Slough
6A 116.2 Beaver Complex 6 31 37
Genetics samples were collected from 37 age-2 Chinook and four coho salmon in 2013. An
additional29 samples were collected from Chinook salmon :;::100 mm collected July 2013-April
2014. Analysis ofthese samples is currently underway. A total of approximately 600 Chinook
salmon tissue samples have been delivered to ADF&G for analysis and can be used to determine
Chinook salmon ID error rate if desired.
Approximately 11 voucher specimens have been collected for Chinook and coho salmon. These fish
will be used for meristic counts to determine species ID. The ADF&G permit limited voucher
specimen collection to 10 per species but was recently modified to up to 20 Chinook and coho
salmon.
Thirty-one photos of these larger Chinook salmon juveniles are also available for review. Review is
complete for photos gathered by one contractor but those collected by other contractors remain to be
reviewed.
Habitats where Chinook salmon were collected in 2013 and winter 2014: 681 juvenile Chinook
salmon were collected from upland slough beaver complexes compared to 3,414 coho salmon.
Approximately 14 percent of Chinook salmon were associated with upland slough beaver complexes.
The highest habitat supporting collection was tributaries, over 21 percent of total collections. Of
larger Chinook salmon juveniles, roughly one third, 100 of313 were associated with upland slough
beaver complexes.
Macro Habitat
Additional Open
Water
Backwater
PageS
Chinook salmon
All
Sizes Larger
1
31
Coho salmon
All
Sizes Lager
32 1
107
Pacific salmon,
undifferentiated
All
Sizes Larger
3
Alaska Energy Authonty
October 2014
Total
33
141
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Page6 10 • Information used to describe fish/habitat
ParaS preferences were gathered using
professional best judgment, literature, and
limited field data, but were not confirmed
with an adequate sample from the Susitna
River system (ISR 8.5). Fish/habitat data
gathered from the Susitna River is
necessary to identify preferential use of the
habitats. It is vital that these data are
accurate as they will be used to: 1) develop
Habitat Suitability Indices (HSI) and
Habitat Suitability Criteria (HSC); 2)
describe fish-macrohabitat relationships,
which may be used to evaluate project
effects; 3) validate the Instream Flow Study
(8.5) habitat model predictions; and 4)
extrapolate results from F As to geomorphic
reaches and river segments. Ultimately the
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
ClearWater 69 2 144 14 227 Plume
Main Channel 1,038 74 1,210 23 79 3 2,327
Side Channel 176 12 291 1 42 509
Para Side 11 1 3 14
Channel Complex
Side Slough 177 3 554 147 878
Side Slough 76 1 221 11 25 322
Beaver Complex
Tributary 1,875 43 1,411 6 53 3,339
Tributary Mouth 615 70 2,123 7 28 2,766
Upland Slough 108 6 378 19 1 487
Upland Slough 681 100 3,414 65 131 1 4,226
Beaver Complex
Grand Total 4,858 313 9,885 133 526 4 15,269
AEA is confused by and disagrees with the first sentence of this comment. AEA has repeatedly
described the methods being used for developing HSC related data and has noted that the preferred
method and the one that AEA has been following involves the collection of site specific data.
However, for species and life stages that are rarely observed, final HSC curves may be based on
additional data, including utilization of data from 2012 and the 1980s studies on the Susitna River.
However, there may still be some species where few or no empirical HSC/HSI data were able to be
collected. In those cases, AEA will consider other methods for developing curves. This may include
the use ofliterature based curves, developing envelope curves (see, for example, Jowett et al. 1991,
and GSA BBEST 2011), guilding (e.g., creating a combined HSC/HSI curve representing multiple
species and/or life stages; see, for example, Vadas, Jr. and Orth 2001, GSA BBEST 2011),
developing curves based on expert opinion/round table discussions) and the use ofBayesian
statistical methods for updating data distributions (see, for example, Hightower 2012). Bootstrapping
may be used as one technique for estimating variability around these types of combined curves.
Page 9
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
data will be used to develop protection and
mitigation measures and to provide a basis
for post-project monitoring.
Page6 11 • The Service is concerned about AEA's
Para 6-proposal to "scale up", and requests
Page7 rationale for its implementation (Riverine
Para 1 Model Integration Meeting 2013). "Scaling
up" is only appropriate when the sampling
is conducted accurately, in a random
fashion throughout the population, and at a
scale relevant to resource concerns. To
assess impacts from the Project on fish
resources, sampling effort must be at a scale
relevant to Susitna River fish species at
various life stages in order to adequately
quantify baseline conditions with the
accuracy required for accurate
extrapolation. For example, incorrect fish
identification and would lead to imprecise
and inaccurate extrapolation of species-
specific habitat associations.
Page7 12 • Standard error was not reported for stated
Para3 relationships between species of juvenile
salmonids at various life stages and their
habitat (e.g., ISRs 9.5, 9.6). A robust
assessment of statistical results must include
calculations for standard error.
Page7 13 • Assumptions for the estimating numbers
Para4 of Chinook salmon migrating above Devils
Canyon were not clearly specified and the
Sus1tna-Watana Hydroelectnc ProJect
FERC Project No. 14241
Response
Several points of clarification are warranted regarding the topic of scaling up. Firstly, it is important
to note that AEA is not developing fish/habitat associations so they can be extrapolated. Rather,
AEA is developing HSC curve sets that reflect fish species and life stage preferred habitat use that
will be used in the habitat-flow models for defining how Project operations may influence fish
habitats (target species and life stages) within different habitat types. The scaling up that AEA would
use is associated with the extrapolation of the habitat-flow modeling results from one location to
other, unmeasured locations.
In addition, the specific concern cited related to potential fish identification issues of juvenile fish in
selected lateral habitats has no bearing on the outcome of the habitat-modeling studies and
extrapolation of results since that analysis will consider all five of the Pacific salmon species.
And finally, AEA has identified and discussed several approaches for extrapolating the results of this
type of analysis to other areas of the Middle River during the April 15-17,2014 Proof of Concept
meetings (see http://www.susitna-watanahydro.org/wn-
content/ui!loads/2014/04/2014 04 17TT Riverine SJ.1atialExtraJ.1olation.J.1df) but has not selected a
specific approach pending further review with licensing participants.
Statistical error associated with relative abundance fish data was not in ISRs 9.5 and 9.6. As these
catch-per-unit-effort (CPUE) data were preliminary and were subject to additional post-ISR QAQC,
no error metric was calculated for the ISR. The data are available to generate Standard Error or
another measurement of error around CPUE or density estimates. Such error will be reported for
these data in the USR.
As stated in Study 9.5 ISR Section 5.1.3 and Study 9.6 ISR Section 5.1.3, data presented on habitat
associations was preliminary and based only on counts and therefore have no standard error
associated with these data. Once QAQC has been completed on the fish data, the analysis of fish-
habitat associations will be completed with additional inputs including relative abundance, species
richness, and life stages supported. As stated in RSP Section 9.6.4.3.1, Study 9.5 ISR Section 5.1.3,
and Study 9.6 ISR Section 5 .1.3 fish-habitat associations will be evaluated at the meso-habitat level.
These data will not be used to validate the instream flow model.
This comment reflects a fundamental lack of understanding of the methodologies being relied upon
by the PERC-approved study plan. As described in RSP Section 9.7.4.1.5 (Objective 1) and Section
9.7.4.6 (Objective 6), AEA planned to examine fish on selected spawning grounds (e.g., Indian
Page 10
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 ISR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
standard error of that estimate was not
reported (e.g., ISRs 9.6, 9.7).
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
River) in part to establish mark rates (proportion offish tagged) so that inferences could be made
about the representativeness of tagging across stocks. In addition, AEA stated that mark rates from
these areas can be used to estimate the abundance passing the tagging sites (but not the abundance at
the recovery site). If sufficient sampling can be obtained and some assumptions met, some inference
can be made about relative abundance among recovery locations using the estimates of mark rates
and the number of radio-tagged fish present. However, it was not an objective of this study to
produce a mark-recapture estimate of the number of Chinook salmon migrating above Devils Canyon
(or above the proposed dam site).
In the FERC Study Plan Determination (SPD) (page B-13), NMFS and the USFWS requested that
AEA add the additional goal of estimating the numbers of fish above Devils Canyon (and the
proposed dam site) to the study. FERC did not recommend this additional goal be included in the
study. Instead, FERC recommended the study be modified to require AEA to include in the 2013
ISR an evaluation of the feasibility of putting in a weir or sonar counting station at or near the dam
site during the 2014 study season to count anadromous fish.
In ISR Section 5.6.4, AEA used two different approaches to estimate of the number of Chinook
salmon that migrated above Devils Canyon in 2013. The first approach involved expanding the peak
aerial spawner count in tributaries above Devils Canyon (29 fish) by the estimated observer
efficiency (46.3 percent, as observed in the Indian River; 26/0.463 = 63 fish). This expanded count
should be considered a minimum number since only fish counted on the July 25-27 survey were
included. Chinook salmon were also observed in tributaries above Devils Canyon on four other
surveys, so it is possible that some of these fish were not present during the July 25-27 survey. Also,
this· approach assumed that the observer efficiency in tributaries above Devils Canyon was similar to
that in the Indian River (which was 'ground-truthed' with weir counts in 2013).
The second approach involved expanding the number of radio-tagged Chinook salmon detected
above Devils Canyon (3 fish) by the marked fraction of Chinook salmon in the Middle River (6.3
percent; 3/0.063 = 48 fish). It was highly unlikely that more than three fish migrated above Devils
Canyon. This approach assumed that the mark rate of fish above Devils Canyon was the same as the
mark rate of fish sampled in the Indian River. Sensitivity analyses were included in ISR Section
5.6.4 and Section 6.6 to illustrate how extreme, but unlikely, parameter values affected the expanded
counts derived from both approaches.
In summary, too few tagged and untagged fish were observed above Devils Canyon to derive a
statistically valid estimate of the number of Chinook salmon that passed Impediment 3 (or the
proposed dam site). Regardless, the study was not designed to produce such estimates. As proposed
Page 11
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
in the RSP, AEA used available data to make inferences about the abundance of Chinook salmon
above Devils Canyon. Although lacking statistical rigor, these estimates provided insight into the
order of magnitude of Chinook salmon abundance above Devils Canyon (e.g., 50-65 fish above
Devils Canyon in2013 was likely, but 100 or more was unlikely). These estimates also illustrate
how difficult it would be to achieve sufficient sample sizes to derive a reasonably accurate and
precise mark-recapture estimate for Chinook salmon above Devils Canyon.
Summary of passage events for large Chinook salmon (MEF 2: 50 em) released in the Middle
River, 2012-2014. Small Chinook salmon, a,nd large Chinook salmon released in the Lower
River, were not included in this table.
Page 12
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
.
Page7 14 • Sampling and non-sampling errors were
Para5 not clearly stated (e.g., ISR 9.7). Sampling
error is the error resulting from sampling
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
Tags Released at Curry
Number ofTags Delecled Above:
Galeway
lmpedirrent 1
lmpedirrent2
lmpedirrent 3
Proposed Dam Sile
PercentofTags Released Delecled Above:
Galeway
lmpedirrent 1
lmpedirrent 2
lmpedirrent 3
Proposed Dam Sile
Percent ofT ags Past Galewa~ Delecled Above:
lmpedirrent 1
lmpedirrent 2
lmpedirrent3
Proposed Dam Sile
Number of Tags That Approached lmpedirrent 1 (wilhin 1 km)
PercentofTags Released That Approached lmpedirrent 1
Percent ofT ags Past Galeway That Approached lmpedirrent 1
See Response to Comment 13.
Page 13
2012 2013 2014
352 536 590
313 445 491
23 17 11
20 13 8
10 3 2
6 2 1
88.9 83.0 83.2
6.5 3.2 1.9
5.7 2.4 1.4
2.8 0.6 0.3
1.7 0.4 0.2
7.3 3.8 2.2
6.4 2.9 1.6
3.2 0.7 0.4
1.9 0.4 0.2
34 60 32
9.7 11.2 5.4
10.9 13.5 6.5
Alaska Energy Authority
October 2014
Total
1,478
1,249
51
41
15
9
84.5
3.5
2.8
1.0
0.6
4.1
3.3
1.2
0.7
126
8.5
10.1
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 ISR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
only a part of the population and not the
whole population. Non-sampling errors are
those errors resulting from selection bias,
systematic non-representativeness of
samples, and transcription or recording
errors. Sampling error is usually quantified
and reported with confidence intervals or
standard errors and related to precision of
the estimates.· Non-sampling errors are
harder to recognize, yet very important, and
more closely related to the accuracy of the
estimates. Sampling errors must be clearly
accounted for in statistical analyses to
assess data reliability and interpret results.
Page7 15 • Consistent fish sampling methods were
Para6 not applied (i.e., different gear types used,
different effort was applied within and
across sampling units, concurrent use of
non-compatible gear types within a
sampling unit). This resulted in inability to
estimate sampling error because (e.g., ISR
9.6) inconsistent sampling methods resulted
in individual datasets that are not
comparable.
Page 7 16 • No power analysis was reported (ISR
Para 7 9.14), and it is unclear how sample size for
both adult and juvenile Chinook salmon
was determined. Based on the number of
genetic markers sampled and the magnitude
of genetic divergence measured in the
population documented thus far, a power
analysis would inform determination of the
number of samples needed to provide a
robust estimate of genetic diversity.
Furthermore, three years of samples may
not be adequate to characterize genetic
diversity among a species with a life cycle
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
The use of different gears consistent with habitat characteristics was implemented as proposed in the
Fish Distribution and Abundance Implementation Plan filed with FERC on March 1, 2013 with
modification described in Study 9.5 ISR Section 4.4.4 and Study 9.6 ISR Section 4.4.4.
AEA respectfully disagrees that sampling error will impact AEA' s ability to meet objectives of fish
distribution and abundance sampling for Studies 9.5 and 9.6. The fish distribution and relative
abundance methods were implemented consistent with Studies 9.5 and 9.6 RSPs, the Fish
Distribution and Abundance Implementation Plan, and FERC's SPD.
AEA agrees that power analysis will be needed, but disagree that we have the information at hand to
perform a meaningful analysis at this time to determine adequate sample sizes. NMFS also
commented that a power analysis will be critical to determine iflack of detection of effect is due to a
lack of samples or truly a lack of effect. In response we added a power analysis section (Section
4.6.9) in the Final2014 Implementation Plan (Study 9.14 ISRPart B). This section added a power
analysis so that we can quantify the level of effect necessary for detection. As for using a power
analysis to determine the number of samples needed, we could use programs such as POWSIM, but
we would have to make assumptions regarding the relatedness of individuals sampled,
representativeness of samples of underlying populations, and variation among years. Results from
such an analysis using the limited information we have, could provide misleading results.
AEA also agrees that additional years may be needed, especially if we do not detect differences due
to a lack of statistical power (small samples sizes, small number of years). However, if we are able to
Page 14
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 /SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
offive to seven years; this limitation must
be addressed in the study results.
Page 8 17 • Samples from presumed siblings were
Para1 proposed for removal from the genetic
analyses (ISR 9.14). Only if the samples
have been collected in a non-random way
may this method be justified. Purging
related animals as proposed will bias the
results. Furthermore, ISR 9.14 proposes to
exclude samples from juvenile Chinook
salmon if they show significant differences
in allele frequency from adult Chinook
salmon. Using all data will produce a more
robust estimate of allelic frequencies across
the entire population.
Page 8 18 • Using a Bonferroni adjustment on the
Para2 tests for Hardy-Weinberg Equilibrium (ISR
9.14) will increase the risk of a Type-2 error
and reduce the statistical power of the test
to detect a difference. Furthermore,
estimates of genetic distance using F51 must
include a correction for sample size
otherwise small samples tend to look like
outliers (ISR 9.14).
Page9 19 Model integration is the manner in which all
Para2 of the physical studies interact to assess
baselines and Project impacts on the Susitna
River. Within the ISRs, methodologies for
model integration are not transparent and it
is not possible to determine if model
integration will identify project impacts
with any degree of certainty.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
collect an adequate number of samples, we may have adequate power to distinguish among
hypotheses with adequate certainty.
AEA agrees that samples from presumed siblings should not be removed. Removal of juvenile
collections based on comparisons to adults was added to the 2014 Genetics Implementation Plan
(Study 9.14 ISRPartB) based on recommendations from USFWS and NMFS. We are in agreement
about keeping all samples to provide the most robust overall estimate of population allele
frequencies. Section 4.6.2 (Study 9.14 ISR Part B) reflects this change.
AEA agrees with not using the Bonferroni adjustment on tests for HWE. Section 4.6.3 of the 2014
Genetics Implementation Plan (Study 9.14 ISR Part B) was revised to reflect this recommendation as
it was received in NMFS's written comments to the Draft 2014 Genetics Implementation Plan.
We agree that estimates ofF"1 will need to be corrected for sample size. Section 4.6.8 of the 2014
Genetics Implementation Plan (Study 9.14 ISR Part B) was revised to reflect this recommendation as
it was received in NMFS's written comments to the Draft 2014 Genetics Implementation Plan.
AEA disagrees. This comment reflects a fundamental lack of understanding of the methodologies
being relied upon by the FERC-approved study plan. The two Riverine Modelers Meetings held in
November 2013 and April 2014 respectively were specifically held in response to licensing
participant concerns about model integration. Review of the presentations from both of these
meetings which are available on AEA's website (htm://www.susitna-watanahydro.orglmeetings/Qast-
meetings£) clearly demonstrate the linkages between the models and how individual model outputs
will be used in evaluating Project effects for each resource discipline, with an emphasis on effects on
fish habitats. The meeting notes for the two meetings provide a clear record of the major topics
discussed and licensing participants' questions pertaining to model integration. Indeed, one of the
comments provided at the end of the April meeting by a USGS representative suggested that the
modeling and model integration efforts were moving in the right direction-" .... thought it was a
great meeting and that the studies are making good progress. Feels that there has been tremendous
amount of focus on where the problem areas are and are a lot further along than in November 2013."
Page 15
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 ISR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Page9 20 As previously stated by the Service
Para3 (USFWS 2013), we are concerned that time
allotted to develop methods for model
integration is inadequate. Prior to the
release ofthe June 3, 2014, ISRs, a three-
day Riverine Modeling Integration Meeting
(RMIM) was held (November 13-15, 2013).
The goal of this meeting was to provide a
forum to review and discuss various
riverine-related modeling and study
integration efforts (AEA Instream Flow
Study-Technical Team [ISF-TT] Riverine
Modeling Integration Meeting Agenda,
2013). A collaborative meeting such as this
one was a good effort toward developing
meaningful model integration methods and
the Service encourages AEA to continue
this type of cooperative work.
Page 9 21 During the RMIM, 25 and 50-year scenarios
Para4 for predicting project impacts to the
physical river channel and habitats were
proposed. While those timelines are
consistent with what is specified in RSP and
may present a manageable timeframe for
the modeling work (B. Fullerton, Personal
Communication, November, 2013), they
may not be sufficient to assess impacts to
fish and wildlife resources in a biologically
meaningful way.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
Since then, the resource modelers have continued working in a collaborative fashion on each of the
respective models.
AEA notes that there have been two three-day Riverine Modelers meetings, and one two-day
Riparian Modeling meeting designed to provide Licensing Participants with updates on model
development and integration and to solicit feedback and suggestions on model refinements. The first
of these was held from November 13-15, 2013, the second April15-17, 2014 which involved a Proof
of Concept discussion to demonstrate the integration of the different models by highlighting model
outputs within a single Focus Area (FA-128) (Slough SA). The third was held April29-30, 2014 to
discuss various riparian I riverine-related modeling and study integration efforts, and present and
discuss proposed metrics. During these meetings, each of the resource modelers explained first the
specific models they were working on and the model dependencies on other models or data sources,
as well as the model outputs to other models.
AEA disagrees. The time frames ofO, 25, and 50 years were selected because they represent time
intervals that span the potential length of the FERC license, and as well are reasonable increments
from which to gauge and compare changes in channel morphology (RSP 6.6, Section 6.6.4.2.2.1) that
may translate into changes in fish habitat. Having time intervals at shorter increments of
geomorphological modeling would be less likely to elicit substantive changes in channel
morphologies and would therefore be less likely to elicit changes in the results of the habitat-flow
modeling.
However, the greatest potential effects of Project operations on fish and fish habitats are on the actual
regulation of flows that would occur over much shorter time intervals (annual, seasonal, weekly,
daily, and hourly) and for which the habitat-flow models are being developed to evaluate. As
described in RSP 8.5, Section 8.5.7.4.1.1, the "Temporal analysis will involve the integration of
hydrology, Project operations, the Mainstem Open-water Flow Routing Model, and the various
habitat-flow response models to project spatially explicit habitat changes over time. Several
analytical tools will be utilized for evaluating Project effects on a temporal basis. This will include
development and completion of habitat-time series that represent habitat amounts resulting from flow
conditions occurring over different time steps (e.g., daily, weekly, monthly), as well as separate
Page 16
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
Page9 22 The Service is concerned the modeling
Para5 capability to answer biological questions is
not sensitive enough to detect biologically
meaningful changes to species and habitats
likely to be affected by project operations.
We recommend that modelling capabilities
be developed that incorporate biological
inputs and deliver outputs that are validated
under an appropriate range of operational
scenarios (e.g., base load, ecological flows,
load-following, run-of-river). The temporal
scales (e.g., 25, 50-year) must have
biological relevance. For example, 5, 10 and
15 year operational scenarios should be
Susitna-Watana Hydroelectric Project
FERC Project No. 14241
Response
analysis that address effects of rapidly changing flows (e.g., hourly) on habitat availability and
suitability. The Mainstem Open-water Flow Routing Model and habitat models will be used to
process output from the Project operations model. This will be done for different operating
scenarios, hydrologic time periods (e.g., ice free periods: spring, summer, fall; ice-covered period:
winter [will rely on Ice Processes Model-Section 7.6]), Water Year types (wet, dry, normal}, and
biologically sensitive periods (e.g., migration, spawning, incubation, rearing) and will allow for the
quantification of Project operation effects on the following:
• Habitat areas (for each habitat type-main channel, side channel, slough, etc.) by
species and life stage; this will also allow for an evaluation of the effects of breaching
flows on these respective habitat areas and biologically sensitive periods (e.g.,
breaching flows in side channels during egg incubation period resulting in temperature
change).
• Varial zone area (i.e., the area that may become periodically dewatered due to Project
operations, subjecting fish to potential stranding and trapping and resulting in reduced
potential invertebrate production).
• Effective spawning areas for fish species of interest (i.e., spawning sites that remain
wetted through egg incubation and hatching).
• Other riverine processes."
These shorter time intervals (hourly, daily, weekly, monthly) represent those that are the most
biologically meaningful in the sense that they would have the most direct and immediate effect on
fish and fish habitats. If warranted, it will also be possible to evaluate effects over longer time steps
that encompass Project operations over several different water years.
See AEA's response to Comment 21.
Page 17
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014/SR COMMENT LETTER
Comment Comment
Page, Number Comment
Para
considered to demonstrate the model's
ability to detect generational impacts to fish
populations and habitat persistence (e.g.,
Susitna River Chinook salmon; five to
seven years).
Page9 23 Data collected for some studies do not
Para6 provide the information needed for the
proposed integrated modeling efforts.
During the RMlM, for example, it was
revealed the Water Quality Modeling study
(ISR 5.6) would require data collected on
the spatial distribution of groundwater
discharge to surface water bodies.
Analytical or numerical groundwater flow
simulation would be one (of several) ways
to satisfy this input requirement. However,
the Groundwater Study (ISR 7 .5) does not
explicitly state analytical or numerical
groundwater flow simulations would be
undertaken in support of the other physical
process models.
Page9 24 As a follow up to the RMlM, a Proof of
Para 7-Concept (POC) meeting was held April 15-
Page 10 17, 2014. This meeting was to: 1) confirm
Para 1 successful integration of models and
associated metrics in a single FA (Slough
128); 2) examine the modeling process
rather than focus on the actual POC results;
and 3) clarify many questions related to the
integration of multiple models. The
discussions of modeling processes at the
POC meeting was considered valuable by
the Service, but not fully effective in
demonstrating successful model
development and integration; many
questions regarding model development and
integration were unanswered. To develop
Susrtna-Watana Hydroelectnc ProJect
FERC Project No. 14241
Response
AEA disagrees, Review of the November Riverine Modelers Meeting notes (http://www.susitna-
watanahydro.orgL:I:YQ-content/uQloads/20 14/02/2013 .11.13Modelers Notes.Qdf) indicates questions
did occur related to the Water Quality model that pertained to the integration of groundwater, that
were addressed by noting that data from targeted grab samples as well as data from groundwater
wells would be used, as well as data from other locations. Additional information was provided on
the groundwater study during the April Proof of Concept meetings (http://www.susitna-
watanahydro.org/w-content/uQloads/2014/04/2014 04 15TT Riverine Presentation-
Grotmdwater.Qdf), and more recently in two Technical Memorandums (GWS and R2 2014a,
http://www.susitna-watanahydro.org/wp-
content/uploads/20 14/09/07.5 _ GW _ GWS _ T6 _ TM _Aquatic_ Hydro _Final_ Draft_ 20140925 .pdf;
GWS and R2 2014b, http://www.susitna-watanahydro.org/wp-
content/uploads/20 14/09/07.5 _ GW _ GWS _ T5 _ TM _ Riparian_Final_Draft_20 140926.pdf) which
describe some of the analysis leading to development of preliminary groundwater/surface water
relationships in selected Focus Areas.
AEA will take this under advisement.
Page 18
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 ISR COMMENT LETTER
Comment Comment
Page, Number Comment Response
Para
greater stakeholder confidence in the
models, the Service recommends
conducting a formal model integration
meeting to: 1) establish a model
development process, 2) develop an
understanding of inputs and outputs, 3)
demonstrate conceptual linkages, 4)
demonstrate the predictive capabilities of
the models, and 4) conduct sensitivity
analyses to better understand model
limitations and reduce uncertainty.
REFERENCES
AEA (Alaska Energy Authority). 2012. Revised Study Plan: Susitna-Watana Hydroelectric Project FERC Project No. 14241.
December 2012. Prepared for the Federal Energy Regulatory Commission by the Alaska Energy Authority, Anchorage,
Alaska. http://www.susitna-watanahydro.org/study-plan.
Geo-Watersheds Scientific and R2 Resource Consultants, Inc. 2014a. Preliminary Groundwater and Surface-Water Relationships in
Lateral Aquatic Habitats within Focus Areas FA-128 (Slough 8A) and FA-138 (Gold Creek) in the Middle Susitna River,
Technical Memorandum, Study 7.5. Susitna-Watana Hydroelectric Project, FERC No. P-14241. Prepared for Alaska Energy
Authority, Anchorage, Alaska. September 2014. 136 pp. http://www.susitna-watanahydro.org/wp-
content/uploads/2014/09/07.5_GW_GWS_T6_TM_Aquatic_Hydro_Final_Draft_20140925.pdf.
Geo-Watersheds Scientific and R2 Resource Consultants, Inc. 2014b. Groundwater and Surface-Water Relationships in Support of
Riparian Vegetation Modeling, Technical Memorandum, Study 7.5. Susitna-Watana Hydroelectric Project, FERC No. P-
14241. Prepared for Alaska Energy Authority, Anchorage, Alaska. September 2014. 58 pp. http://www.susitna-
watanahydro.org/wp-content/uploads/2014/09/07.5_GW _GWS_T5_TM_Riparian_Final_Draft_20140926.pdf.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241 Page 19
Alaska Energy Authority
October 2014
AEA RESPONSE TO USFWS SEPTEMBER 22, 2014 ISR COMMENT LETTER
GSA BBEST (Guadalup~, San Antonio, Mission, and Aransas Rivers and Mission, Copano, Aransas, and San Antonio Bays Basin and
Bay Expert Science Team). 2011. Environmental flows recommendations report. Final submission to the Guadalupe, San
Antonio, Mission, and Aransas Rivers and Mission, Copano, Aransas, and San Antonio Bays Basin and Bay Area Stakeholder
Committee, Environmental Flows Advisory Group, and Texas Commission on Environmental Quality. March 1, 2011.
Unpublished report available online http://www.tceq.texas.gov/permitting/water_rights/eflows/guadalupe-sanantonio-bbsc.
Hightower, J.E., J.E. Harris, J.K. Raabe, P. Brownell, and C.A. Drew. 2012. A Bayesian Spawning Habitat Suitability Model for
American Shad in Southeastern United States Rivers. Journal ofFish and Wildlife Management, 2(3): 184-198.
http:/ /scholarworks. umass.edu/fishpassage journal_ articles/2046.
Jowett, I. G., J. Richardson, B.J.F. Biggs, C.W. Hickey and J.M. Quinn. 1991. Microhabitat preferences of benthic invertebrates and
the development of generalised Deleatidium spp. habitat suitability curves, applied to four New Zealand rivers. New Zealand
Journal ofMarine and Freshwater Research 25(2):187-199.
Stuart, D.L. 2002. A study ofperiphyton induced pH spikes on the White River, Washington. MS Thesis, University of Washington.
Vadas, Jr., R.L., and D.J. Orth. 2001. Formulation of habitat suitability models for stream fish guilds: Do the standard methods
work? Transactions ofthe American Fisheries Society 130:217-235.
Susitna-Watana Hydroelectric Project
FERC Project No. 14241 Page 20
Alaska Energy Authority
October 2014
United States Department of the Interior
FISH AND WILDLIFE SERVICE
Anchorage Field Office
605 W. 4th Avenue, Room G-61
Anchorage, Alaska 9950 I -2250
In Reply Refer To:
FWS/ AFES/AFWFO
Mr. WayneDyok
Susitna-Watana Project Manager
Alaska Energy Authority
813 West Northern Lights Boulevard
Anchorage, Alaska 99503
Dear Mr. Dyok:
SEP 2 2 2014
FERC Project P-14241, Susitna-Watana Hydropower
The U. S. Fish and Wildlife Service (Service) is providing comments on the Alaska Energy
Authority's (AEA) June 3, 2014, Initial Study Report (ISR) for the proposed Susitna-Watana
Hydropower project (Project). We provide AEA with our preliminary findings of concern so
that they may be meaningfully considered prior to and discussed at the October, 2014 ISR
meeting. The Service intends to provide full and detailed comments on these and other topics
by the November 30,2014, Federal Energy Regulatory Commission's (FER C) filing deadline.
As per the FERC Integrated Licensing Process (ILP; 18 CFR 5.15 (c)(2)), the ISR meeting
scheduled in October, 2014, provides an opportunity for AEA and licensing participants to
discuss the 2013 studies and identify potential modifications to study designs based on the first
year's data collection. The process allows for review and recommendation of changes to
sampling methodologies implemented by first year studies to ensure study objectives, as
specified in the PERC-approved Revised Study Plans (RSP), are met. Our filing to FERC by
November 30, 2014, will formalize our comprehensive comments and recommendations after
AEA has had the opportunity to address our concerns during the October, 2014 ISR meeting.
The Service has identified three topics of significant concern: 1) data collection and reporting,
2) effective model integration, and 3) development of decision support systems (DSS). These
three topics are closely tied together because precise and accurate data provide inputs to models
that are used to support Project decision-making.
In these preliminary comments, the Service identifies data collection and reporting concerns
(Attachment I) and recommends the data issues be resolved as soon as possible. Without robust
data from individual studies, we are concerned the data do not meet study objectives, that model
validation will be hindered, and model integration may lead to incorrect conclusions. Given the
magnitude of our concerns related to data collection and reporting, we believe it may not be
Mr. Wayne Dyok
possible to yield plausible model predictions describing baseline conditions or to predict
potential impacts. It is important that these issues be resolved prior to conducting additional
field studies.
2
Much of the data collected under FERC approved study plans are proposed for use in fish habitat
models, and the development of those models are based on changes to cha1mel geomorphology
and hydrology. Relationships among hydrologic models should be validated and models
calibrated for the Susitna River system before their use in fish habitat models. Likewise,
relationships among fish habitat models should be validated, and models calibrated for the
Susitna River system prior to their use in estimating Project effects under various operational
scenarios. To our knowledge there is currently no specific model integration process proposed
that will ensure sound relationships among models and their accurate calibration for the Susitna
River system. The Service believes that development and implementation of rigorous model
integration procedures is critical to our review of this project and we discuss our preliminary
concerns in detail (Attachment II).
A DSS is one of the end products of the studies, where data and models from the studies are
ultimately used to help make decisions on the effects of the Project on natural resources. We
understand AEA intends to develop a DSS using a manual matlix method by early 2015 (FERC
2013). As the DSS plays such an important role in the assessment ofProject impacts, the
Service requests its development be a collaborative process so that the fundamental objectives,
assumptions, critical inputs, weighting methods, and other pruis of the model are mutually agreed
upon. Furthermore, we are concerned that the timeline for DSS development is lagging other
efforts. The ILP process is founded under the principal of early identification of potential issues
and conducting studies needed to fill information gaps (FERC 2014). Data gaps may be revealed
once the fundamental objectives for the DSS are formulated. Until the DSS development
process occurs, it is uncertain all the data needed to implement the DSS has been gathered.
Because the DSS is not scheduled for development unti12015, it is distinctly possible that crucial
new data needs may be revealed when updated study reports are filed by AEA in 2016 (as per the
ILP extension approved by FERC on January 28, 2014). However, going forward, the Service
believes the development of a collaboratively designed DSS is of great importance to this Project
and recommends that, if practicable, the timeline for its development be accelerated.
Finally, FERC established a new schedule for the proposed Susitna-Watana hydroelectric project
ILP in their January, 2014 determination. In that determination, FERC ordered AEA to submit
fmal ISRs on June 3, 2014, for stakeholder review, to hold a meeting in October, 2014, to present
results of those ISRs, and to discuss AEA proposed changes to the studies or those proposed by
other licensing participants. During a meeting with the Service and National Mmine Fisheries
Service on September 2, 2014, AEA stated its intent to re]ease reports from 21 new or
continued studies conducted in 2014, with intent to discuss results at the October 15,2014, ISR
meeting. On September 17,2014, AEA filed 10 of2l reports to FERC. Because the data were
gathered outside timelines specified by the FERC -ordered process, and given the limited review
Mr. Wayne Dyok 3
time the Service will have, we will be unable to consider and comment on those study reports in
advance of the October, 2014 ISR meeting. Furthermore, we recommend AEA dedicate the
limited time at the October, 2014, ISR meeting to discuss concerns related to 2013 studies, as
reported in the June 3, 2014, ISR. Additionally, an email on May 6, 2014, copied to the Service
by FERC, indicated that studies carried out by AEA in 2014 were conducted outside of the ILP
process and would not be considered "second year" studies. This is procedurally very important
because ~'\either the Service, nor other licensing participants (Non-Govemmental Organizations
(NGO) Participants 2014), will have the opportunity to fully review or comment on the design
and implementation of the 2014 studies. The Service will be unable to meaningfully contribute
to the discussion of the 2014 studies and urge AEA to not discuss any work conducted in 2014 at
the ISR meeting. Instead, we suggest the interim results gathered between study years (i.e., 2014
data collection) be discussed at the next quatterly Technical Workgroup meeting, once we have
had sufficient opportunities to review those additional data.
Summary
This letter describes some of the Service's concerns with studies reported in the June 3, 2014,
ISR, and we are providing them to AEA prior to the November 30, 2014, FERC filing deadline
so some issues can be discussed and resolved in a timely manner. The concerns address: 1) data
collection and reporting, 2) ability to recommend further studies under the FERC ILP licensing
process, 3) development of valid models to assess baseline conditions and effects from Project
operations on fish and wildlife resources, 4) capacity to formulate recommendations under
section 1 O(j) of the Federal Power Act for protection, mitigation, and enhancement measures
associated with the Project, and 5) formulation of informed decisions pursuant to our Section 18
Fishway Prescription authority under the Federal Power Act. We believe the modified ILP
schedule for the Project affords AEA the opportunity to make necessary changes to studies prior
to entering the second year of study. The Service believes this review process accommodates the
development and implementation of more accurate, effective, and cost-effective plans of study
for the Project.
Thank you for the opportunity to submit these comments in advance to the October, 2014 ISR
meeting. We hope they are useful to AEA and will generate valuable conversations at the
meeting. If you have questions, please contact Ellen Lance (907) 271-1467.
Sincerely,
~ Sochea~-
Anchorage Field Supervisor
Mr. Wayne Dyok
Cc: Sarah Goad, AIDEA
Betsy McGregor, AEA
Nicholas Jayjack, FERC
Joe Klein, ADFG, Sport Fish Division
Jeanne Hansen, NMFS
Sue Walker, NMFS
Mike Bethe, ADFG, Habitat Division
Matthew LaCroix, EPA
Literature Cited
[FERC] Federal Energy Regulatory Commission. 2013. Letter to Wayne Dyok. Study Plan
Detennination on 14 remaining studies for the Susitna-Watana Hydroelectric Project.
February 1, 2013.
__ 2014. Integrated Licensing Process. http://www.ferc.gov/industries/hydropower/gen-
info/licensing/ilp.asp, accessed September 18, 20 14).
[NGO Participants] Wood, M., W. Wolff, J. Konigsberg, M. Wood, R. Schryver, J. Seebach
(collectively the NGO Participants). 2014. Initial Study Report Meeting, Susitna-Watana
Hydroelectric Project (P-14241). Letter filed with FERC, September 16,2014.
4
[USFWS] U.S. Fish and Wildlife Service. 2013. Letter to FERC Re: Alaska Energy Authmity's
Revised Study Plan for the Susitna-Watana Hydroelectric Project No. 14241-000. March l 8,
2013.
u
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
:J
Attachment I. Data Issues
Below we discuss our preliminary concerns relating to deviations from study plans, quality
assurance and control, and statistical practices and procedures for the 2013 study year.
Deviations From Study Plans-Deviations from established sampling designs occurred in some
studies for various reasons, and in some cases resulted in reduced sample size or compromised
reliability of data. Below we provide examples.
1 • As currently planned, some two-year studies cannot be completed because access to all
Focus Areas (FAs) was not granted until after the first study year (e.g., ISRs 8.5, 9.6, 9.7,
9.9). For example, a fish wheel was not installed and fish were not tagged near the
entrance to Devil's Canyon (e.g., ISR 9.7).
2 • Anomalous weather conditions prevented or delayed fieldwork on aquatic studies (e.g.,
ISR 8.5), resulted in late installation of migrant traps, which were likely influenced by
environmental conditions associated with late breakup (e.g., ISR 9.6). Moreover,
juvenile salmon distribution and abundance measured in 2013 were likely affected by the
record fall floods in 2012 (e.g., ISR 9.6).
3 • Sampling has not been temporally adequate across all seasons. ISR 9.6 reports winter
fish sampling did not occur across all F As as proposed; early spring sampling occurred
only in three F As; initial sampling following breakup and installation of migrant traps did
not occur until the middle of June, and therefore, spring sampling for fish distribution and
abundance was not conducted (e.g., ISRs 7.5, 8.5, 8.6). The extent to which fishes move
must be described through sampling; multiple sampling days across all seasons are
required to capture the full seasonality of a fish's life-history strategy, which varies
considerably within a single season. A single-day of sampling is insufficient to
understand the habitat associations of different fish species with differing mobility and
life-stages.
4 • Sample site selections for integrated studies were inconsistently co-located. For example,
invertebrate sampling locations (ISR 9.8) were not co-located with fish sampling
locations (ISR 9.6). Failure to co-locate sampling sites risks the magnification of data
discrepancies, and because the data will be used as inputs for predictive models, may
jeopardize the validity of the models.
5 • Detection arrays did not cover the entire channel and tagging efforts did not allow for
detection of fish migrating upstream, therefore the data were biased and efficiency
estimates cannot be calculated. Detection rate and recovery of passive integrated
transponder (PIT) tags is insufficient to yield useful data to meet study goals and
objectives (ISR 9.6).
6 • Fish targets for fish Habitat Suitability Curve (HSC) sampling were not met (e.g., ISR
8.5), therefore, power to assess fish habitat-preferences and relationships is reduced.
1
7 • Data collected on fish habitat for the Fish Passage Barrier Study (ISR 9.12) and the
HSIIHSC component of the fish and aquatic Instream Flow Study (ISR 8.5) were
gathered at incompatible spatial scales to meet the study objectives.
Quality Assurance and Control Concerns -Below we preliminarily provide some discrete
examples where the Service has data quality concerns. Poor data quality has a rippling effect
throughout this assessment process because extrapolating inaccurate results throughout the river
would amplify errors across the river and associated habitat.
8 • Water quality samples were qualified as either estimated or rejected by the analytical
laboratory due to quality-related failures (ISR 5.5). Issues included failure to deliver
samples to the laboratories within the method-specified temperature range; failure to
meet procedure specified holding times; contaminated or missing field, trip, and method
blanks; and Chain of Custody and bottle labeling discrepancies. AEA proposed to apply
a correction factor to the 2013 data to render it useable, but provided no details on how
that would be done.
9 • There is evidence that juvenile salmon may have been misidentified. A comparison of
juvenile fish collections from the Susitna River in the 1980s (Alaska Department ofFish
and Game 1983 as cited by R2 Consultants in the Fish Population Summary Document),
local Alaskan rivers (Alaska Department ofFish and Game, unpublished data; Davis et
al. 2013), recent studies on the Susitna River (Kirsch et al. 2014), and nearby tributaries
(Miller et al. 2011 ), signal substantial differences in total fork length distribution and
habitat associations among juvenile salmon from that which is expected. Large numbers
of unidentified salmonidjuveniles (some of which were PIT tagged), anomalous length
distributions and questionable habitat associations decrease our confidence in the
accuracy of species identification. For example, juvenile Chinook salmon measuring 150
mm fork-length were reported, juvenile Chinook salmon were reportedly most abundant
in beaver ponds, there was absence of pink salmon in any samples, and a disappearance
of sockeye salmon from Indian River between the February draft ISR and the June draft
ISR. We have strong reservations about the identification of these juvenile fish, and
suspect many juvenile salmons identified as Chinook salmon may be coho salmon.
1 0 • Information used to describe fish/habitat preferences were gathered using professional
best judgment, literature, and limited field data, but were not confirmed with an adequate
sample from the Susitna River system (ISR 8.5). Fish/habitat data gathered from the
Susitna River is necessary to identify preferential use of the habitats. It is vital that these
data are accurate as they will be used to: 1) develop Habitat Suitability Indices (HSI) and
Habitat Suitability Criteria (HSC); 2) describe fish-macrohabitat relationships, which
may be used to evaluate project effects; 3) validate the Instream Flow Study (8.5) habitat
model predictions; and 4) extrapolate results from F As to geomorphic reaches and river
segments. Ultimately the data will be used to develop protection and mitigation measures
and to provide a basis for post-project monitoring.
11 • The Service is concerned about AEA's proposal to "scale up", and requests rationale for
its implementation (Riverine Model Integration Meeting 2013). "Scaling up" is only
2
appropriate when the sampling is conducted accurately, in a random fashion throughout
the population, and at a scale relevant to resource concerns. To assess impacts from the
Project on fish resources, sampling effort must be at a scale relevant to Susitna River fish
species at various life stages in order to adequately quantify baseline conditions with the
accuracy required for accurate extrapolation. For example, incorrect fish identification
and would lead to imprecise and inaccurate extrapolation of species-specific habitat
associations.
Statistical Practices and Procedures -Based on our preliminary reviews, we note (below) failures
to report standard statistical procedures and calculations required for complete analyses.
12 • Standard error was not reported for stated relationships between species of juvenile
salmonids at various life stages and their habitat (e.g., ISRs 9.5, 9.6). A robust
assessment of statistical results must include calculations for standard error.
13 • Assumptions for the estimating numbers of Chinook salmon migrating above Devils
Canyon were not clearly specified and the standard error of that estimate was not reported
(e.g., ISRs 9.6, 9 .7).
14 • Sampling and non-sampling errors were not clearly stated (e.g., ISR 9. 7). Sampling error
is the error resulting from sampling only a part of the population and not the whole
population. Non-sampling errors are those errors resulting from selection bias,
systematic non-representativeness of samples, and transcription or recording errors.
Sampling error is usually quantified and reported with confidence intervals or standard
errors and related to precision of the estimates. Non-sampling errors are harder to
recognize, yet very important, and more closely related to the accuracy of the estimates.
Sampling errors must be clearly accounted for in statistical analyses to assess data
reliability and interpret results.
15 • Consistent fish sampling methods were not applied (i.e., different gear types used,
different effort was applied within and across sampling units, concurrent use of non-
compatible gear types within a sampling unit). This resulted in inability to estimate
sampling error because (e.g., ISR 9 .6) inconsistent sampling methods resulted in
individual datasets that are not comparable.
16 • No power analysis was reported (ISR 9.14), and it is unclear how sample size for both
adult and juvenile Chinook salmon was determined. Based on the number of genetic
markers sampled and the magnitude of genetic divergence measured in the population
documented thus far, a power analysis would inform determination of the number of
samples needed to provide a robust estimate of genetic diversity. Furthermore, three
years of samples may not be adequate to characterize genetic diversity among a species
with a life cycle of five to seven years; this limitation must be addressed in the study
results.
3
17 • Samples from presumed siblings were proposed for removal from the genetic analyses
(ISR 9.14). Only if the samples have been .collected in a non-random way may this
method be justified. Purging related animals as proposed will bias the results.
Furthermore, ISR 9.14 proposes to exclude samples from juvenile Chinook salmon if
they show significant differences in allele frequency from adult Chinook salmon. Using
all data will produce a more robust estimate of allelic frequencies across the entire
population.
18 • Using a Bonferroni adjustment on the tests for Hardy-Weinberg Equilibrium (ISR 9.14)
will increase the risk of a Type-2 error and r~duce the statistical power of the test to
detect a difference. Furthermore, estimates of genetic distance using Fst must include a
correction for sample size otherwise small samples .tend to look like outliers (ISR 9.14).
Literature Cited
ADFG. Unpublished data for Willow Creek and Deshka River tagging study.
Alaska Department ofFish and Game. 1983. Resident and Juvenile Anadromous Fish Studies
on the Susitna River Below Devil Canyon, 1982. Phase II, Volume 3 Basic Data Report.
ADFG/Su Hydro Aquatic Studies Program. Anchorage, Alaska
Davis, J.C., and G.A. Davis. 2013. Water quality in the Lower Little Susitna River: Cumulative
draft report, July 2007 through June 2012. Final Report for the Alaska Department of
Environmental Conservation. Aquatic Restoration and Research Institute, Talkeetna, AK.
Kirsch, J.M., J.D. Buckwalter, and D.J. Reed. 2014. Fish inventory and anadromous cataloging
in the Susitna River, Matanuska River, and Knik River basins, 2003 and 2011. Alaska
Department ofFish and Game, Fishery Data Series No 14-4, Anchorage.
Miller, E.M., J.C. Davis, and G.A. Davis. 2011. Monitoring juvenile salmon and resident fish in
the Matanuska-Susitna Basin. Final Report for the U.S. Fish and Wildlife Service, Mat-Su
Salmon Habitat Partnership. Aquatic Restoration and Research Institute, Talkeetna, AK.
Riverine Model Integration Meeting. 2013. AEA meeting minutes. November 2013.
4
Attachment II. Model Integration
19 Model integration is the manner in which all of the physical studies interact to assess baselines
and Project impacts on the Susitna River. Within the ISRs, methodologies for model integration
are not transparent and it is not possible to determine if model integration will idenjify project
impacts with any degree of certainty.
20 As previously stated by the Service (USFWS 2013), we are concerned that time allotted to
develop methods for model integration is inadequate. Prior to the release of the June 3, 2014,
ISRs, a three-day Riverine Modeling Integration Meeting (RMIM) was held
(November 13-15, 2013). The goal of this meeting was to provide a forum to review and discuss
various riverine-related modeling and study integration efforts (AEA Instream Flow Study-
Technical Team [ISF-TT] Riverine Modeling Integration Meeting Agenda, 2013). A
collaborative meeting such as this one was a good effort toward developing meaningful model
integration methods and the Service encourages AEA to continue this type of cooperative work.
21 During the RMIM, 25 and 50-year scenarios for predicting project impacts to the physical river
channel and habitats were proposed. While those timelines are .consistent with what is specified
in RSP and may present a manageable timeframe for the modeling work (B. Fullerton, Personal
Communication, November, 2013), they may not be sufficient to assess impacts to fish and
wildlife resources in a biologically meaningful way.
22 The Service is concerned the modeling capability to answer biological questions is not sensitive
enough to detect biologically meaningful changes to species and habitats likely to be affected by
project operations. We recommend that modelling capabilities be developed that incorporate
biological inputs and deliver outputs that are validated under an appropriate range of operational
scenarios (e.g., base load, ecological flows, load-following, run-of-river). The temporal scales
(e.g., 25, 50-year) must have biological relevance. For example, 5, 10 and 15 year operational
scenarios should be considered to demonstrate the model's ability to detect generational impacts
to fish populations and habitat persistence (e.g., Susitna River Chinook salmon; five to seven
years).
23 Data collected for some studies do riot provide the information needed for the proposed
integrated modeling efforts. During the RMIM, for example, it was revealed the Water Quality
Modeling study (ISR 5.6) would require data collected on the spatial distribution of groundwater
discharge to surface water bodies. Analytical or numerical groundwater ·flow simulation would
be one (of several) ways to satisfy this input requirement. However, the Groundwater Study
(ISR 7.5) does not explicitly state analytical or numerical groundwater flow simulations would
be undertaken in support of the other physical process models.
24 As a follow up to the RMlM, a Proof of Concept (POC) meeting was held April 15-17,2014.
This meeting was to: 1) confirm successful integration of models and associated metrics in a
single FA (Slough 128); 2) examine the modeling process rather than focus on the actual POC
results; and 3) clarify many questions related to the integration of multiple models. The
discussions of modeling processes at the POC meeting was considered valuable by the Service,
but not fully effective in demonstrating successful model development and integration; many
questions regarding model development and integration wete unanswered. To develop greater
stakeholder confidence in the models, the Service recommends conducting a fonnal model
integration meeting to: I) establish a model development process, 2) develop an understanding
of inputs and outputs, 3) demonstrate conceptual linkages, 4) demonstrate the predictive
capabilities of the models, and 4) conduct sensitivity analyses to better understand model
limitations and reduce uncertainty.
Literature Cited
IFS-TT: Riverine Modeling, Draft Meeting Agenda, November 13-15, 2013,
http://www.susitna-watanahydro.org/wp-contentluploads/2013/10/SuWa IFS-
TT Modeling2013Novl3-15 -Agenda.pdf
2