HomeMy WebLinkAbout091 - Applicant SubmittalDonna Frostholm
From: Ken Sheppard <KSheppard@sksp.com>.. I�LM
Sent: Wednesday, July 13, 2022 5:35 PM
To: Philip Hunsucker p
Cc: Donna Frostholm; 'Brad Nelson'
Subject: RE: Army Corps Permit Withdrawal
Attachments: BDN Eelgrass Report_2022.pdf, otters and eelgrass.pdf; Sea Otters Increase Eelgrass
Diversity.pdf; Eelgrass in Squamish Bay.pdf
ALERT: BE CAUTIOUS This email originated outside the organization. Do not open attachments or click on links if you
are not expecting them.
Hello Phil/Donna —
I am attaching the recently completed eelgrass study on the Smersh parcel. As you can see, it shows that the native
eelgrass beds on the site have generally expanded since 2018, and have modestly expanded overall since 2016.
I am also attaching materials previously submitted to the Corps of Engineers indicating that physical disturbance in
eelgrass beds actually increases genetic diversity of the eelgrass beds, and that BDN planting operations on its existing
nearby tract have actually increased native eelgrass overall on that site.
This shows the complete inaccuracy of assertions by Marilyn Showalter and others that geoduck cultivation anywhere in
Squamish Bay harms native eelgrass throughout the bay. ("Geoduck cultivation, particularly the way BDN practices it, is
incompatible with the urgent need to foster and enhance native eelgrass and forage fish.", Comments of Marilyn
Showalter BDN-LLC NWS-2013-1268 January 20, 2022, Page 67, ) Once again, actual project specific scientific studies do
not support the biased and false arguments set forth by Ms. Showalter and others in their efforts to stop all aquaculture
in Squamish Bay by any means necessary.
When Judge Lasnik issued his 2020 Order vacating the nationwide NWP Corps of Engineers aquaculture permitting, he
ruled that the blanket conclusion by the Corps that all aquaculture activities of any kind did not cause significant
environmental impacts anywhere in the US was not supportable based on the general studies performed by the
Corps. In geoduck aquaculture he was particularly concerned about the effects of area netting and pesticides on
eelgrass, and ruled that individual permits should be issued by the Corps for each project based on analysis of the effects
of that particular project. A new 2021 case filed by the Coalition To Protect Puget sound now argues that Corps
permitting under the new NWP via "Letters of Permission" (LOP) is improper.
BDN is striving to do obtain is permits for this new project, (and on the renewal of its existing project) in full compliance
with Judge Lasnik's ruling, The Corps has been proceeding under a full individual permit process for this project, not an
LOP. No area netting or pesticides are being used or proposed. Detailed individual data, such as the new eelgrass study,
is being submitted to show the absence of any net significant negative environmental effect of this specific project. We
are doing all of the applications connected with this project "the right way", and believe we have now submitted
exhaustive project specific information showing that the project will not have significant negative environmental
impacts.
The Corps of Engineers approval of the Smersh project requires a prior Section 401 Water Quality Certification, and that
Certification has been awaiting both this new eelgrass study, and the County's SEPA Threshold Determination. It is my
understanding that we need the Threshold Determination before we can resubmit to the Corps and the DOE for this
project.
Can you let me know if there is anything else we need to do to move this matter towards that threshold determination,
and what is the expected timeline for that determination?
Thanks,
Ken Sheppard
I KETTER
\SHEPPARD
Kenneth A. Sheppard
Ketter, Sheppard & Jackson, LLP
50 116th Ave. S.E., Suite 201
Bellevue, WA, 98004
(206) 382-2600
Fax: (206) 223 3929
- CONFIDENTIALITY NOTE -
LOG ITEM
Pap o f61
The information contained in this electronic file is confidential information intended only for the use of the individual or entity named above and
may be legally privileged. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or
copy of this facsimile is strictly prohibited. If you have received this email in error, please immediately notify us by telephone or return
email. Thank you.
From: Ken Sheppard
Sent: Thursday, May 19, 2022 6:40 PM
To: Philip Hunsucker <PHunsucker@co.jefferson.wa.us>
Cc: Donna Frostholm <DFrostholm@co.jefferson.wa.us>; Brad Nelson <Brad@sea producks.com>
Subject: RE: Army Corps Permit Withdrawal
Hi Phil —
The withdrawal resulted because the DOE suddenly decided in late April that they wanted an all new eelgrass study
before they would do the 401 certification. The Corps won't give their individual permit without the 401 certification,
and we could not get the eelgrass study re -done (due to there being insufficient active eelgrass growth) before the
Corps time period to keep their file open ran out on May 11th . The Corps also apparently wants the County's SEPA
threshold determination before they will issue their permit.
So the plan is to get the new eelgrass study, provide it to everyone (county included) and then immediately re -apply for
the 401 certification and the Corps permit.
It is my understanding that Donna had everything she wanted to issue the initial SEPA determination, and I am assuming
that the temporary withdrawal of the Corps application would not affect that process. But, of course, we will provide
the County with the new eelgrass study as soon as we have it.
FYI, Brad Nelson is just now removing the first set of tubes from the 2020 planting he did on his existing farm using 3
seeds per tube vs. 4. He is observing 60% survival, no tube pushout on that planting, and no tube escapement from it
over the past two years, so that may be useful information. I am sending that additional information directly to Donna
for the file.
Thanks,
Ken
\� KETTER
SHEPPARD
Kenneth A. Sheppard
Ketter, Sheppard & Jackson, LLP
50 1161h Ave. S.E., Suite 201
Bellevue, WA, 98004
(206) 382-2600
Fax: (206) 223 3929
- CONFIDENTIALITY NOTE -
LOG UEAA
Page. off
The information contained in this electronic file is confidential information intended only for the use of the individual or entity named above and
may be legally privileged. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or
copy of this facsimile is strictly prohibited. If you have received this email in error, please immediately notify us by telephone or return
email. Thank you.
From: Philip Hunsucker <PHunsucker co.'efferson.wa.us>
Sent: Thursday, May 19, 2022 1:03 PM
To: Ken Sheppard <KShe and sks .com>
Cc: Donna Frostholm <DFrostholm@co.iefferson.wa.us>
Subject: Army Corps Permit Withdrawal
Ken:
We obtained the attached letter written by the Army Corps, confirming that BDN has withdrawn its Army Corps
permit.
Just wanted to check in to see what impact this has on the Jefferson County permitting process.
Donna is out most of this month, so I guess by default I am the primary contact in the meantime.
Philip
pronouns: he/him/his
Philip C. Hunsucker
Chief Civil Deputy Prosecuting Attorney
leffersian Co u nty Prosecuting Attorne 's Office
P.O. Box 1220, Port Townsend, WA 98368
Ph: 360-385-9219 (direct)
All e-mail sent to this address has been received by the Jefferson County e-mail system and is therefore subject to the Public Records Act, a state law
found at RCW 42.56. Under the Public Records law the County must release this e-mail and its contents to any person who asks to obtain a copy (or for
inspection) of this e-mail unless it is also exempt from production to the requester according to state law, including RCW 42.56 and other state laws.
LOG ITEM
#9) _ ,- -
E_Oc ITEM
Pais - .� m
CONFLUENCE
ENVIRONMENTAL COMPANY
3.46 N Canal St, Suite iii • Seattle, WA 98103 0
www.confenv.com
BDN Eelgrass Bed Delineation - 20i6
FINAL REPORT
Prepared for:
BDN, Inc.
Attn: Brad Nelson
Prepared by:
Confluence Environmental Company
Contact: Grant Novak
July 2022
LOG ITEM
--V/
Page Eo
146 N Canal St, Suite iii a Seattle, WA 98103 • www.confenv.com
BDN Eelqrass Bed Delineation — 2022
TABLE OF CONTENTS
CONFLUENCE
ENVIRONMENTAL COMPANY
1.0 INTRODUCTION..............................................................................................................................................1
2.0 METHODS........................................................................................................................................................2
3.0 RESULTS.........................................................................................................................................................3
4.0 REFERENCES.................................................................................................................................................4
FIGURES
Figure 1. Eelgrass Delineation Study Area and Vicinity.............................................................................................1
Figure 2. BDN West, Mocean, and Former WA Shellfish Parcels..................................................................................2
Figure3. Surveyed Eelgrass Beds.................................................................................................................................3
LOG ITEM
# 91
page -7_oF--
JUIY2022 Page i
BDN Eelgrass Bed Delineation - 2022
s.o INTRODUCTION
.w
! i
CONFLUENCE
ENVIRONMENTAL COMPANY
This document summarizes work conducted on June 1, 2022 to accurately survey the distribution of
existing native eelgrass (Zostera marina) beds on aquatic parcels in Squamish Harbor, Hood Canal,
WA. Surveys were conducted on parcel 721031007 (Smersh parcel) (Figure 1).
The purpose of the eelgrass surveys outlined in this document was to determine the extent of
eelgrass beds in relationship to aquaculture activities proposed by BDN, Inc on the Smersh parcel.
Figure s. Eelgrass Delineation Study Area and Vicinity.
JUIy 2022
LOG ITEM
D$ _—of
BDN Eelgrass Bed Delineation - 2022_ - - ■
CONFLUENCE
0 ENVIRONMENTAL COMPANY
2.0 METHODS
A transect-based survey method was employed to map the extent of eelgrass beds at the Smersh
parcel following the United States Army Corps of Engineers' (Corps') eelgrass bed delineation
guidance (Corps 2016). Surveys were conducted by walking and wading at low tide on June 1, 2022.
Following the Corps' eelgrass bed delineation guidance (Corps 2018), transects were aligned
perpendicular to the shoreline and spaced 35-feet (ft) apart across the parcel (Figure 2).
Figure 2. Survey Transects at Smersh Parcel.
The coordinates (i.e., latitude and longitude) of the transect start points, and the magnetic bearing
of transects were determined using a geographic information system (GIS) in the office and
uploaded onto a Trimble GeoXH differential GPS unit with sub -meter, real-time accuracy. At the
site, surveyors navigated to, and marked with a pin flag, the start points of each transect. Surveyors
knowledgeable in the submerged aquatic vegetation and marine nearshore ecology of Hood Canal
walked each transect while unrolling a tape measure and following the magnetic bearing of the
transects using a compass.
The bed edge was determined using the definition of an eelgrass bed described by the Corps'
Method A which states, "An eelgrass bed is defined as a minimum of 3 shoots per 0.25 mz within 1
meter of any adjacent shoots. To identify the bed boundary, proceed in a linear direction and find
the last shoot that is within 1 meter of an adjacent shoot along that transect. The bed boundary
(edge) is defined as the point 0.5 meter past that last shoot, in recognition of the average length of
JUIy 2022
BDN Eelgrass Bed Delineation-2o22�$ Q CONFLUENCE 1/9
ENVIRONMENTAL COMPANY
the roots and rhizomes extending from an individual shoot." Pin flags were placed 0.5 meters
landward of the last eelgrass shoot.
Non-native dwarf eelgrass (Zostera japonica) had been previously observed at this site interspersed
with native eelgrass. In order to avoid mischaracterizing the extent of native eelgrass by
inadvertently including non-native eelgrass in the surveys, information regarding identification
methods were reviewed by the survey team immediately prior to conducting the survey. Eelgrass
shoots were occasionally sampled along each transect by pulling from the substrate to look for
diagnostic indicators confirming whether the sampled vegetation was native or non-native eelgrass.
Only native eelgrass (e.g., Zostera marina) was mapped during this survey.
The location of each pin flag marking the native eelgrass bed edge was recorded using a sub -meter
accurate differential GPS. Once the eelgrass bed edge was marked along each transect with pin
flags, the bed edge between transects became clearly evident and a surveyor walked the entirety of
the bed edge, accurately collecting its location with the GPS.
3.o RESULTS
Surveys indicate a patchy bed of native eelgrass located between approximately -1 ft MLLW and -2
ft MLLW. Below -2 ft MLLW, a very dense bed of native eelgrass extends waterward beyond the
survey area (Figure 3). No native eelgrass was identified landward of the upper edge of the patchy
eelgrass bed.
y
750'—
Q
Oro
200
— 60'
250 ' —
I +1' — 100' PrpJ4o! Vlcini
300
350 ' --
0' [AdLt� w 200'
00, --
---260'
460'--
300'
600 ' -- .1'
— 360'
660 ' —
00'
soo ' � - a
j1 Smersh Parcel Dense Eelgrass (Zostera marina) N
/'/ Elevation Contour (MLLW) Patchy Eelgrass (Zostera marina) A
�f U1/rs 2022 Landward Zostera marina Extent , Eelgrass Protection Buffer (5m) 0 75 150 225 300
CONFLUENCE Proposed Geoduck Planting Area Peet
FVVIRONMrVTAI CJWPAN\' =Mef Wr8
0 10 2D 3D 40
Figure 3. Surveyed Eelgrass Beds.
JU1y 2022
BDN Eelgrass Bed Delineation - 2022
4.o REFERENCES
CONFLUENCE
ENVIRONMENTAL COMPANY
United States Army Corps of Engineers — Seattle District. 2018. Components of a Complete Eelgrass
Delineation and Characterization Report.
LOG ITEM
Page
My 2022
.._:)G, ITEM
ell
Page /,�_ a#_
Appendix A
Prior Eelgrass Surveys and Bed -
edge Re -verifications
JUIY 2022
7
00
elf
cr) r
�
/
3
q I
o
(L
m
It
R
/
g
0
S b
n
o
\
| e
% LO
|
% o
k
\
b
LO
�./
, •
k � \_
k k E
N N
r
N
f
2 2 2
.
(D 00 2
k k q
/
�
.
u
�
k
�
0- k
$ 2
>
$ u
| |
U )
|
\
~ to
m
_ ■ o
. |
� � ;
S o
§
to
2
O/
U!
JUIY _2
LOG ITEM
Page / Of
LOG 1I II - i V e
Pc{ _
Science
WAAAS
Supplementary Materials for
Physical disturbance by recovering sea otter populations increases eelgrass genetic
diversity
Erin Foster et al.
Corresponding author: Erin Foster, erin.foster@hakai.org
Science 374, 333 (2021)
DOI: 10. 1 126/science.abf2343
The PDF file includes:
Materials and Methods
Figs. S1 to S7
Tables S 1 to S 10
References
Other Supplementary Material for this manuscript includes the following:
MDAR Reproducibility Checklist
Materials and Methods
Study design and sample collection
p64g$ 1 �{
Sea otters (Enhydra lutris) were ecologically extinct in BC by 1820s-1850 (31, 32),
extirpated by the early 1900s, and re -introduced in 1969-72 (33). Once existing across the North
Pacific rim, sea otters were hunted to near -extinction in the maritime fur trade. Recovering from
extirpation in British Columbia (BC), Canada, sea otters occur over gradients of occupancy
ranging from long -absent (i.e., >100 years) to long -re-established (i.e., several decades). To test
the hypothesis that genetic diversity of eelgrass (Zostera marina) varied between areas where
otters were established (20-30 yrs.) and areas where they have been absent for over a century
(34), we collected 50 eelgrass blades from each of 15 eelgrass meadows (Fig. 2). Our
experimental design used a space -for -time approach (35), and was replicated in two regions: the
Central Coast of BC, and the west coast of Vancouver Island, BC (Fig. 2). On the Central Coast
we were able to test for the potential additional effect of recently arrived (<10 yrs) otters on the
genetic diversity of eelgrass. Meadows were selected where we knew the history of sea otter
occupancy. We avoided meadows with freshwater inputs, or areas where anthropogenic impacts
were likely (e.g. potential anchorages).
We collected samples by SCUBA, under BC Parks permit #107190 (Hakai Institute). In
each meadow, we laid 3-5 transects up to 200 in long, perpendicular to shore from the intertidal
to subtidal meadow edge. Transects were spaced equidistantly across the shoreward margin of
the meadows. We measured the proportion of each transect line that was disturbed by sea otter
digging, which was used to complement rather than replace our sea otter occupancy categories
because of potential inconsistencies in pit visibility and persistence (12). We collected eelgrass
blades from 50 evenly spaced locations within the meadow (mean spacing distance = 5.47 m;
range = 2.5-10 m). For each sample, we recorded location on the transect and in situ depth, the
latter standardized to chart datum. We removed the newest leaf from each eelgrass shoot and
dried it in silica within 12 hours of collection for later genetic analysis.
Seagrass allelic richness and genotypic diversity can be affected by disturbance (36, 37),
meadow size (38), temperature (18, 39), latitude (40), and depth (41, 42), each of which provide
alternative hypotheses to explain differences in genetic diversity. We limited our suite of
explanatory factors to these five fixed effects, based on the literature and our a priori knowledge
of eelgrass systems in our study areas (43). We used HOBO Pendant ® loggers to record
temperature every 30 min at 5 m below chart datum for one year and calculated the temperature
range for each meadow (mean 5.9 °C, range 3.5-7.2 °C). One logger was absent upon attempted
retrieval (Dodger) so we averaged temperature readings from the two nearest loggers (Grappler
and Wizard; Fig. 2) to impute temperature data for Dodger. To examine meadow depth, we
calculated the median depth of samples collected from each meadow (median = 0.66 m below
chart datum, mean = 0.78 m, range = 5.13 below datum to 1.6 m above datum, across all
meadows). To examine meadow size, we mapped eelgrass meadows; this required three different
approaches. The eelgrass meadows sampled in Kyuquot were mapped using SCUBA and
snorkeling. Snorkelers followed the shoreward edge and divers swam the seaward perimeter of
each meadow. A 5.5 m boat followed the snorkelers or divers and recorded GPS waypoints using
Arc Collector (ESRI 2017). To delineate the seaward edge of the meadow, divers surfaced every
3 minutes or each time the bed changed direction, and the points were plotted. On the Central
Coast, six of the eelgrass meadows we sampled (McMullin, McNaughton, Choked, Ward, Nalau
and Stirling) were mapped using drones (DJI Phantom 3/4 Pro). Drone survey flights occurred
during the morning low tide when weather was suitable for flying (winds <20 kn, no rain, tide <1
ITEM
0
PcAgG�_L'?_017
m). Drone operators used a grid pattern survey design to ensure adequate coverage.
Orthomosaics of each site were created using a Structure from Motion Multi -View Stereo (SfM-
MVS) workflow within Pix4Dmapper software (Version 2.1.61, Pix4D) in Windows 10. The
extent of eelgrass from each orthomosaic was delineated using object -based image analysis
(OBIA) with eCognition Developer software (eCognition Developer 9, 2014). Depending on site
and water conditions, it can be difficult to discern the subtidal edge of eelgrass from RPAS
imagery (44). Accordingly, georeferenced towed underwater video (SplashCam Pro) data were
used to provide ground -truth the aerial analysis. When reviewing the video, one analyst
classified substrate and vegetation over a 4 second period, which represents approximately 2 in
across the seafloor using a hierarchical habitat classification scheme (45). We obtained meadow
size for the remaining three seagrass meadows on the Central Coast and in Barkley Sound from
maps produced by British Columbia Marine Conservation Analysis (https.Hbcmca.ca]. We found
the mean meadow area was 86,000 m2 (range 2,108-389,329 m2)
In addition to environmental influences, agents of biotic disturbance not related to sea otter
foraging may occur. In the closely related Zostera noltii, digging by clam harvesters disturbs
rhizome mats and induces flowering, thereby increasing sexual reproduction (37). Traditional
harvesting of Z. marina rhizomes by BC Indigenous peoples may have also promoted flowering
and genetic diversity in eelgrass (24), a practice that declined with the arrival of Europeans.
Elsewhere, Z. marina is subjected to shoot loss through grazing by waterfowl (27). We did not
see evidence of biotic disturbances (i.e., harvests, disease, vertebrate grazers), other than sea
otter digging, at any of the meadows we examined, suggesting that where alternative biotic
agents of disturbance occur, the present-day effects are minimal.
Genetic methods
LOG ITEM
; q1_
Page
We extracted Genomic DNA from —8 mg of dried leaf tissue per sample, using a DNeasy
Plant DNA extraction kit (Qiagen, USA). A subset of approximately 30 (range 28-33) eelgrass
samples was randomly selected from each meadow and genotyped at 13 microsatellite loci
(Table S1) previously identified in Z. marina (46-49). We included 6-7 samples from each
seagrass meadow on each 96-well extraction plate.
We amplified DNA using polymerase chain reactions (PCR) targeting 13 loci with a Bio-
Rad T100 Thermal Cycler (Bio-Rad Laboratories Inc. Hercules, CA, USA) in 10 µl reaction
volumes consisting of 2 pmol dNTP (New England Biolabs, Ipswich, MA, USA), Ix PCR
buffer, 1.5 units of Pag5000 (Stratagene, La Jolla, CA, USA), 1 pmol each forward and reverse
primer (Eurofins MWG Operon, Louisville, KY, USA), 0.3 pmol M13 IRDye® labelled primer
(Integrated DNA Technologies, Skokie, IL, USA), with 5-15 ng DNA template (15). PCR
conditions were as follows: 94°C for 3 min, 27 cycles of 94°C for 40 sec, 56°C for 40 sec
(except for primer GA2 = 58°C), 72°C for 1 min, with a final extension of 72°C for 30 min.
Amplified products were denatured and run on a 5% polyacrylamide gel electrophoresed for
2.5 hours on a LI-COR 4300 automated sequencer (LICOR Inc., Lincoln, NE, USA) with a
minimum of four size standards (50-350 by or 50-700 by LICOR) per 64-well gel. We scored
gels using SAGA 3.3 Microsatellite Analysis software (LICOR Inc., Lincoln, NE, USA). The
scorer was blind to sample location and sea otter occupancy.
Data quality
Prior to analysis, we took several steps to ensure data quality. Eelgrass individuals (genets)
are composed of many shoots (ramets), which are genetically identical clones that can extend up
to 1 Os of meters (50). We used the R package poppy (51) to identify identical multi -locus
genotypes (MLGs; Table S10) and assess the probability of identical genotypes arising by
LC_) ITEM
G7
P� Z' ,
T
chance due to sexual reproduction, Psex (52). The mean Pse, for all meadows was 0.00009 (range
0-0.007), suggesting all identical genotypes in our samples as probable clonemates (15). We
recorded the minimum distance (m) between each identical genotype (herein referred to as clone
length; Table S10) and removed all but one of each identical genotype to prevent resampling
clones for subsequent analyses (36, 53).
After removing clones, the remaining data were screened for null alleles, stuttering, and
allelic dropout using Microchecker (version 2.2.3; 54). For three of the markers (ZMC 12075,
ZME02125 and GA2) possible null alleles were suggested at four meadows (Wizard, Kamils,
Dodger and Ward). We checked the data for possible technical errors but did not find any signs
of excess homozygotes or excessive stuttering causing artifacts. Thus, we did not remove any of
these data (15).
We assessed deviations from Hardy -Weinberg Equilibrium (HWE) using the exact test (55)
with GenePop (56), and investigated all loci for linkage disequilibrium (LD). To mitigate the
occurrence of Type 1 errors, potential deviations from HWE and LD were tested with a
sequential Bonferonni adjustment (57). Loci at Dodger (n = 4), Wizard (n = 1), Stirling (n = 1),
Stryker (n = 1), and Triquet (n = 1) had significant probabilities of deviation from HWE (Table
S2). However, there was variability in which loci contributed to potential deviations from HWE.
Owing to this inconsistency, we included all meadows in our analyses. Of the 13 loci we
analyzed for linkage disequilibrium, one locus, CL853 Contigl, deviated in 9 meadows and was
removed from all analyses. GA23/GA12 showed linkage disequilibrium in 3 populations, and
several other loci showed disequilibrium in one or two populations (Table S3). As no other
patterns emerged in LD occurrences, we used the remaining 12 loci in all analyses.
�.� _ qPage
We used the R (version 3.2.4, 2018) package NEXT (58, 59) to interpolate allele discovery
curves for each seagrass meadow, which indicated the total number of alleles in each meadow,
and evaluated if sample sizes were sufficient to assess variation in genetic diversity among
seagrass meadows in areas with different sea otter occupancy times (Fig. 3A; 15).
Prior to testing our hypothesis that sea otter disturbance affects genetic diversity, we
evaluated the assumption that broad -scale genetic population structure of eelgrass meadows on
the Central Coast and Vancouver Island was not correlated with sea otter occupancy. We used a
Bayesian population assignment model (STRUCTURE; 60) to estimate likelihood of varying
numbers of population clusters in our data. We set the STRUCTURE parameters with a
LOCPRIOR of Vancouver Island and Central Coast, as LOCPRIOR = 1, and the remaining
parameters set to: USEPOPINFO = 0, MAXPOPS = 20, NOADMIX = 0, ALPHA = 1. We ran 5
iterations of 100,000 burn -ins and 500,000 MCMC repetitions at each value of K. We then used
delta K to select the best approximation of K (61) in Structure Harvester (62). We repeated the
STRUCTURE analysis with all of the same parameters but without setting prior locations of VI
and CC (i.e. with LOCPRIOR = 0). To corroborate STRUCTURE results, we used a spatial
principle component analysis (SPCA) with the R package adegenet (63) with a Delauney
triangulation, eigenvalues chosen interactively, and data not scaled to unit variance. We
calculated Fst values for each pair -wise comparison of seagrass meadows and between VI and
CC in Arlequin. The Arlequin program generates the P-value of the Fst test as the proportion of
permutations leading to a FST value larger or equal to the observed one. We tested Fst values
using a Bonferonni adjustment (57). We calculated the presence of private alleles in each
seagrass bed using GenAlEx (64).
qY_
Paps W Of 5"?
Both the Bayesian and frequentist methods we used to assess population structure supported
the inference that distinct populations were not confounded with sea otter occupancy. Results
from both of our STRUCTURE analyses identified two populations, one comprised of the
eelgrass meadows on Vancouver Island (VI), and another consisting of the eelgrass populations
on the Central Coast (CC; Figs. 2, S1). When LOCPRIOR was included, all VI meadows were
assigned to one population and all CC meadows to a second population; mean assignment
probabilities were 0.39, range 0.003-0.98 (pop 1, largely VI individuals; Fig. 2), and 0.61, range
0.02-0.99 (pop 2, largely CC individuals; Fig. 2). When a LOCPRIOR term was not included,
one meadow on VI was assigned to the same population as all meadows from CC (Fig. S5), and
mean assignment probabilities were 0.57, range 0.01-0.99 (pop 1; Fig. S1) and 0.43, range 0.01-
0.99 (pop 2, Fig S5). Results from the sPCA generally corroborated the STRUCTURE results
and showed population differentiation was strongest among CC and northern VI (Kyuquot), with
southern VI (Barkley Sound) showing admixture from both the CC and Kyuquot populations
(Fig. S2). Eigenvalues show strong global structure and weak local structure (Fig. S3). Ft values
among all except one pairwise comparison were significant; Fst between CC and VI samples was
low (0.025) and significant (Table S4). Private alleles occurred in 12/15 meadows (Table S5).
Genetic diversity
We calculated genetic diversity using allelic richness (AR), heterozygosity (expected [He]
and observed [Ho]), and genotypic diversity (R). AR was calculated at the locus and population
level with the R (version 3.2.4, 2018) package popgenreport (65) using methods by El Mousadik
and Petit (66), which employ rarefaction to account for differences in sample sizes and
genotyping success, standardizes to the smallest sample size of any site, and uses the smallest
number of alleles encountered in any sample across all populations and loci. We calculated He
LOr-- iIEsN4
and Ho, using GenAlEx (64). We calculated genotypic diversity, the proportion of different
multi -locus genotypes in each population, as (NG—1)/(N-1), where NG is number of different
multi -locus genotypes detected, and N is number of samples collected (67).
We examined the potential for correlation among different measures of genetic diversity
used in our study. Allelic richness and genotypic diversity were moderately correlated (r = 0.62,
p<0.01; Fig. S4A; 15). This relationship, however, was influenced by a single meadow with the
lowest genotypic diversity, as predicted by (20). When this meadow was excluded, the
correlation was lower and not significant (r = 0.49, p = 0.08; Fig. S413). We did not test for
correlations between AR and heterozygosity, or between GD and heterozygosity because Figs.
313-E show results of each of these parameters individually and there was no indication of
correlation. The discrepancy between measures of allele diversity and heterozygosity is not
surprising, as similar discrepancies have been noted in terrestrial angiosperms (68) and have
been found in Z. marina elsewhere (69). Such discrepancies can be common in clonal organisms
(21). While heterozygosity and genotypic diversity measures act on the individual level (i.e., an
individual plant is homo- or heterozygotic at each locus), allelic richness considers the total
number of alleles at each locus across the population (69), making it especially useful to capture
variation in clonal organisms (20). Genetic diversity may be particularly consequential to Z.
marina, given that meadows are typically monospecific (70, 71).
Statistical analysis
We used an information theoretic approach to assess the relative importance of the putative
fixed effects (sea otter occupancy, depth, latitude, meadow size, and temperature) in predicting
genetic diversity as measured by allelic richness or genotypic diversity. At low genotypic
diversity levels, common in experimental studies, allelic richness can be a hidden treatment
LOG ITEI�vr'
explaining genotypic diversity (20). But correlation was moderate to not significant in our
samples. Because of the lack of clear relationship between allelic richness and genotypic
diversity in clonal seagrasses more generally (20), and our hypotheses that sea otter disturbance
could either enhance allelic richness, or alternatively reduce genotypic diversity, we built a suite
of candidate models for each response variable.
To assess the relative importance of predictive variables on allelic richness, we built a suite
of Bayesian Generalized Linear Mixed -Effects Models (GLMER) in R (version 3.2.4, 2018),
with a gamma distribution and log link, and using the package rstanarm (72). We constructed 31
models that encompassed all pair -wise comparisons of our five fixed effects, as well as a null
model (Table S7; 15). All models included the random effect of locus. We used un-informative
priors for all estimated parameters, and evaluated r-hat statistics to ensure model convergence (r-
hat <1.1). We calculated and compared Leave One Out Information Criterion (LOOIC) scores
for each model, and LOOIC weights were used to calculate the Relative Variable Influence
(RVI) for each fixed effect, over the full suite of 31 models. To evaluate goodness of fit of the
best -supported model, we used the R packages bayesplot (73) and ggplot2 (74) to conduct
posterior predictive checks, examining kernel density curves of observed and simulated data, and
histograms of statistical skew from the posterior predictive distribution (75); the model fit the
data well. Finally, we used our top model to calculate means, standard errors, posterior density
distributions and 95% credible intervals of random effects and fixed effect coefficients. To
compare the relative effects of each predictor variable, we use the fitted model to calculate the
mean expected value of the response variable at specified values of each predictor variable of
interest, while holding all other predictor variables constant at their average value (i.e., marginal
means). By drawing from the joint posterior of all variables for this calculation, we also
LOG RE►;;
Pogo Z!5-a
{
computed the uncertainty associated with the mean expected values, which we report as 95%
credible intervals.
The above methods were replicated for a second suite of models with genotypic diversity as
the response variable, with a few differences. Because genotypic diversity is calculated based on
the individual plant, rather than across loci (i.e., plants are either a unique multilocus genotype,
or a clone), these models did not include any random effects. Also, unlike allelic richness, the
response variable (genotypic diversity) was constrained to a 0-1 range, and thus not well
described by a gamma distribution. We therefore analyzed the observed data using a Bayesian
Generalized Linear Model (GLM) with a beta distribution and a logit link function, also known
as beta regression. As with our analysis of allelic richness, we accounted for model uncertainty
by constructing 31 models that encompassed all pair -wise comparisons of our five fixed effects,
as well as a null model (Table S8). We report means, standard errors, posterior density
distributions and 95% credible intervals of fixed effect coefficients, as well as marginal mean
expected values (and associated 95% CI) of genotypic diversity at differing values of each
predictor variable.
Comparison of Bayesian GLMER models identified sea otter occupancy as the strongest
predictor of eelgrass allelic richness. Model weight (0.27) gave support for a single top -model,
which included the effects of sea otter occupancy and depth (Table S6). Relative Variable
Importance (RVI) was 1.00 for sea otter occupancy, followed by depth (0.99). A comparison of
posterior distributions showed that the effect of depth was weak but significant (the 95%
quantiles of the distributions did not overlap 0), whereas the effect of an established sea otter
population was significant and much stronger: the increase in allelic diversity associated with a
change from no otters to established otters was twice as great as the increase associated with an
increase in per in depth. To help interpret the effect of depth, we used data from (19) to quantify
the proportion of sea otter foraging dives made at subtidal vs. intertidal depths, and found that,
when foraging over eelgrass meadows, sea otters feed 68% of the time in the subtidal as opposed
to 32% in the intertidal zone. Thus, it is possible that the depth effect we found in our model is
amplified by the effect of otter disturbance, which is greater in deeper water.
Comparison of Bayesian GLM models identified meadow size, sea otter occupancy, and
temperature as the strongest influencers of genotypic diversity (Fig. 4C). Model weight (0.24)
supported a single top -model (Table S8). RVIs were highest for meadow size (0.78) and sea otter
occupancy (0.74), followed by temperature range (0.58). A comparison of posterior distributions
showed that the effects of both size and latitude were weak but significant, whereas the effect of
an established sea otter population was significant and much stronger: the increase in genotypic
diversity associated with a change from no otters to established otters was twice as great as the
increase associated with an increase in temperature range of 1 °C or an increase in size of 1 km2.
Our field methods, designed to sample the entire bed to assess bed -level allelic richness, and rely
on similar sample sizes among beds, resulted in larger beds having larger spacing between
samples, decreasing the likelihood of sampling clones.
�� ��
�..-,--
Pa$ 0#
L0 sTDli
a 0
Vancouver Island Central Coast
C
O C
0) d cc Q f0 Y 7 7 Y N
Op Unr NmM lla � 0
Y A Z p
)
J
Fig. S1.
Eelgrass populations identified with genetic data. Population assignment results from
STRUCTURE where LOCPRIOR = 0 and K = 2. Pale- and dark- blue bars show the probability
of individual population assignment at K = 2, where K is the number of populations present.
114: , (A) 52 Central Coast (B)
52 " Central Coast
1• G
e fit•
51
? 51 ?
(D a)
fy►;• K 4
Ji ,
50 •"+.~ Vancouver Island 50
t•"y": Vancouver Island „
49
-129 -128 -127 -126 -125 -128 -127 -126
Longitude 'W Longitude 'W
Fig. S2.
Spatial principle components analysis showing genetic differences in eelgrass populations.
Representation of the first global score of the sPCA analysis including geographic coordinates,
where each point represents an individual seagrass genotype and its sampling location (A) or
meadow via an interpolated map of individual sPCA scores, showing genetic clines (B).
(A)
(B)
1•.
(C)
(D)
(�1 Elgemeluea
/
(
............... .... ... ...................
Y
.......................................
s to n .a.....
Fig. S3.
sPCA results. Connection network used to define spatial weighting (A), representations of
scoring of spatial entities where white symbols are negative and black symbols are positive (B-
D). Local interpolation of global scores are shown in (B), where closer contour lines denote
strongest genetic differentiation. Sizes of squares (C) show different absolute values (i.e., large
black squares are very different from large white squares, but small squares are more similar),
with (D) being similar to (C) but using a grey scale to denote differences. Panel (E) shows
eigenvalues, with positive values (bars to the left) showing global differentiation and negative
values (bars to the right) showing local differentiation. Panel (F) shows eigenvalues of SPCA
(denoted ki with i = 1, ... , r, where kl is the highest positive eigenvalue, and kr is the highest
negative eigenvalue). Here, X1 is the largest value and shows that the first global structure is the
best interpretation.
(A)
R=0.62,p=0.013
1.0
0.81 ,
2.5 3.0
Fig. S4.
3.5 4.0 4.5 2.5
Allelic richness
LOG !TS M
� 9I
3.0 3.5 4.0
Allelic richness
4.5
The relationship between allelic richness and genotypic diversity. Correlations in the 15
Zostera marina meadows we examined (A) and with the meadow with low genotypic diversity
removed (B). Shaded area represents 95% confidence interval.
LC)G I
5
184
m
23
—0.5 0.0 0.5 1.0 1.5 2.0
Median depth (m)
Fig. S5.
Bayesian GLMER output. The relationship between median meadow depth and mean allelic
richness.
a: 1.0
V
0.9
c
c
20.8
1.0
0.9
is
0.8
0.7
c
20.6
Fig. 5b.
(A)
0 1 2 3 4
Mean meadow size (km2)
3.5 4.5 5.5 6.5 7.5
Temperature range (deg C)
LOG ITS M
# q
page of
Bayesian GLM output. The relationships between (A) mean meadow size and (B) temperature
range on mean genotypic diversity of Z. marina.
0
'8
-4
kl�
L
0 10 20 30
Years otter occupancy
Fig. S7.
LOG 3 EiVi
Relationship between otter occupancy and disturbance. The number of years since sea otter
arrival to each eelgrass meadow area, and the percent disturbance, measured as a proportion of
the transect line that was dug. Shaded area represents 95% confidence interval.
�_.
ITEM
a
N
a. b
N
a b
N
a b
N
a b
N
a b
N
a b
N
b
b
b
vN
c
N
T
r•
'n
N c'a
M a
ld
7�
Vl a
�c c�
r- Z�
00 a
N
T is
N
O a
N
�--� a
N Cam❑
M a
«S
7 a
�/.�c
PCA
,�
a
Qi
Y
ON
O!
kC
WO
tD
�n
�n
F•1 p�
V'1
V1
V)
W)
V'1
V)
V1
V1
V1
V1
W)
V'1
V'1
Vl
w
v
Q�
7-i40
01
00
d,
O
�•'� y y
O M
V7 [�
M 7
M V7
N O\
00
N
00
00
y
b1)
amp
N
a
o
a
W
O
�
.r
00
00
I�
DD
iF
M
N
N
N
N
N
It
N
M
N
M
N
N
a
Q
cd
d
Q
U
C7
Q
C7
F
H
H
Q
F
F
U
.
F
U
H
d
C
E
H
¢
H
H
d
0
E
0
U
F
F"
U
U
C7
F"
0
0
C'
b
U
H
U
d
Q
V
v
V
F�
CQ7
F
QuQ
CCd7
<
U
U
Q
U
d
F~
U
Q.
w
o
�
�
U
F~
v
�
d
v
C7
F"
d
Q
F•
H
F
d
U
U
F
d
W
Q
H
H
d
Q
H
U
U
H
•
d
c�i
c
CL
o
Q
C7
U
C7
F�
C7
c7
C7
F-
Q
d
F"
Q
C7
F~
U
C7
C7
U
U
d
U
F~
C7
C
F
W
F�
C7
U
U
d
U
U
U
U
U
C
y
i
w �
C
n
n
q
b
\O
b
b
b
00
00
m
z
a
a
�.
o
o
o
o
O
U
U
\C
o
U
M
U
c
Cd
V
N
N
U
U
Ukrl
Lra
L
N
N
N
N
H
W
0
0
C7
0
C7
U
U
U
U
c
E' L
oT•`�bD �o ��o� o 0 0
E'er a a a I
IW) W)
W) W) k, tn tn kn
N N
C7
F-�
F-�
U
v v
vl M
00
eY
00
N
*
x
U
U
F"'
H
�
d
F-
U
U
C7
U�
d
C7
d
H
U
H
CUJ
Q
U
H
U U
N
U
F�
Q
U
d
d
CQ7 U
H
C7
q
Q
C7 C7
v
C7
U
FU- H
C7
U
C7
H F
C7
C7
C7
v v
v
v
00v
a
b�0
N
00
OOi N
�
U
O
O
N
V
O
N N
N
'.
N
U
U
LLCOG TEN-
b
3
I
N
N
F
M b0 n � bD to bD N 00
NC Oo+ O O O
a N N a N a N N
U U U
LOG�TEvi
�19 4L-
01 LOB' ITEM
O
z
U
VI n h
� C9 �
N N ;-�
Vl V)
a � O
U 7 N N
U
W)
r
O
N
O� ^
Vn ,V1. kn
N
M
W)
kn
M
ti
tr)
v�
vi
vn
vi
In
in
M O bD m — m bD N o0
O •�, '. M
U C7 C7
M_ q N N
U U U U U U O w
a a a
U U U
LOG II-EiM,
pa$_
xxxxxxxtlxxxxxx_
xxxxxxxxxxxxx_�
0
O O C
0 0 00
O O O O O O O
O O O O O O O O
w o M oo tn O o
x x x x x N M
O O O O O Cl O O O
O O O O O O O O O O
10 n .Nr O�
O
O O O O O O O O O O O
r— � N
N M N
O C C C O C O O O C O O
x �
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
A `� 3 � •`�� `a z � 3 � H � � �'' o
Table S5.
Private alleles identified in each eelgrass meadow.
Meadow
Locus
Allele (bp)
Freq
Dodger
GA12
132
0.02
Dodger
ZME02125
109
0.04
Grappler
GA23
173
0.02
Grappler
CL734
84
0.02
Wizard
GA2
128
0.02
Wizard
GA20
154
0.12
Wizard
CL679
78
0.02
Wizard
ZMC12075
94
0.33
Barter
ZMF02381
162
0.03
Kamils
GA12
148
0.02
Kamils
ZMF02381
147
0.05
Spring
GA12
146
0.02
Spring
ZMC13053
108
0.02
Spring
CL559
175
0.02
Spring
ZMF02381
138
0.02
Nalau
ZMC12075
114
0.08
Stirling
GA23
155
0.04
Stirling
GA20
146
0.02
Stirling
ZMC13053
92
0.02
Ward
na
na
na
Choked
ZME02125
119
0.02
Tri uet
na
na
na
McNaughton
GA20
136
0..06
McMullin
na
na
na
Stryker
GA2
134
0.04
Stryker
GA2
136
0.07
Stryker
CL734
129
0.10
Stryker
CL734
132
0.10
Stryker
CL734
135
0.02
Stryker
CL734
138
0.04
Stryker
CL734
141
0.02
Louisa
GA12
150
0.07
Louisa
GA20
140
0.16
Louisa
GA20
144
0.07
# 9l
Pogo ijo
Table S6.
LOG iTENI
PCA
Bayesian GLMER model weights, LOOIC differences, and parameters for all
models run where allelic richness was the response variable. Locus was included as a
random effect in all models whereas sea otter occupancy (absent, recent, established),
depth, size, latitude, and temperature were included as fixed effects.
LOO
weight
A LOO
Model parameters
0.269
0.0
(1 Locus) + Otter occupancy + Depth
0.196
0.6
(1ILocus) + Otter occupancy + Depth + Size
0.165
1.0
(1 Locus) + Otter occupancy + Depth + Temp
0.132
1.4
(1 Locus) + Otter occupancy + Depth + Latitude
0.075
2.6
(1 Locus) + Otter occupancy + Depth + Size + Temp
0.075
2.6
(1 Locus) + Otter occupancy + Depth + Latitude + Size
0.047
3.5
(1 Locus) + Otter occupancy + Depth + Latitude + Temp
0.034
4.1
(1 (Locus) + Otter occupancy + Depth + Latitude + Size + Temp
0.003
9.3
(1ILocus) + Otter occupancy + Size
0.001
11.0
(1ILocus) + Otter occupancy
0.001
11.0
(1 Locus) + Otter occupancy + Size + Temp
0.001
11.5
(1 Locus) + Otter occupancy + Latitude + Size
0.001
11.9
(1 Locus) + Otter occupancy + Latitude
0.000
12.8
(1 Locus) + Otter occupancy + Temp
0.000
13.0
(1 Locus) + Otter occupancy + Latitude + Size + Temp
0.000
13.3
(1 Locus) + Otter occupancy + Latitude + Temp
0.000
19.7
(1 Locus) + Latitude + Temp
0.000
20.7
(1 Locus) + Depth + Latitude + Temp
0.000
21.7
(1 Locus) + Latitude + Size + Temp
0.000
23.0
(1 Locus) + Depth + Latitude + Size + Temp
0.000
25.3
(1 ILocus) + Size
0.000
26.9
(1ILocus) + Size + Temp
0.000
27.0
(1 Locus) + Depth + Size
0.000
27.2
(1 Locus) + Latitude + Size
0.000
28.0
0.000
28.3
(1 Locus) + Depth + Size + Temp
0.000
28.6
(1 Locus) + Temp
0.000
28.6
(1 ILocus) + Latitude
0.000
28.7
(1ILocus) + Depth + Latitude + Size
0.000
29.2
(1 ILocus) + Depth
0.000
30.0
(1 Locus) + Depth + Temp
0.000
30.2
(1 Locus) + Depth + Latitude
�L LOG ITEavi
Table S7.
Parameter estimates for best -fitting model describing allelic richness.
Parameter estimates mean mcse sd 5% 50% 95% Rhat
Intercept
1.12
0.003
0.12
0.93
1.12
1.31
1.006
Otters Recent
-0.06
0.009
0.06
-0.17
-0.06
0.04
0.999
Otters Established
0.26
0.001
0.06
0.17
0.26
0.36
0.999
Depth
0.11
0.000
0.03
0.06
0.11
0.16
1.001
Locus CL559
-0.27
0.003
0.13
-0.49
-0.27
-0.05
1.005
Locus CL679
-0.20
0.003
0.13
-0.43
-0.21
0.00
1.005
Locus CL734
0.08
0.003
0.13
0.57
0.78
1.01
1.004
GA12
-0.01
0.003
0.13
-0.22
-0.01
0.21
1.004
GA20
-0.14
0.003
0.13
-0.36
-0.14
0.07
1.005
GA23
0.01
0.003
0.13
-0.14
0.07
0.29
1.004
GA2
0.30
0.004
0.13
0.09
0.30
0.52
1.004
ZMC06073
-0.34
0.003
0.13
-0.55
-0.33
-0.12
1.004
ZMC12075
-0.10
0.003
0.13
-0.32
-0.10
0.12
1.005
ZMC13053
-0.21
0.003
0.13
-0.43
-0.21
0.01
1.004
ZME02125
-0.09
0.003
0.13
-0.30
-0.09
0.13
1.004
ZMF02381
0.18
0.003
0.13
-0.04
0.18
0.39
1.005
LOB ITI N(I
Table S8.
Bayesian GLM model weights, LOOIC differences, and parameters for all models
run where genotypic diversity was the response variable. Sea otter occupancy (absent,
recent, established), depth, size, latitude, and temperature were included as fixed effects.
LOO A LOO Model parameters
weight
0.244
0.0
Otter occupancy + Size + Temp
0.116
1.5
Otter occupancy + Latitude + Size + Temp
0.063
2.7
Otter occupancy + Latitude + Size
0.058
2.9
Otter occupancy + Depth + Size + Temp
0.053
3.0
Otter occupancy
0.052
3.1
Size
0.052
3.1
Otter occupancy + Size
0.035
3.9
Otter occupancy + Temp + Depth + Size + Latitude
0.034
3.9
Depth + Size
0.029
4.2
Otter occupancy + Temp
0.024
4.7
null
0.022
4.8
Latitude + Size
0.020
5.0
Otter occupancy + Latitude
0.020
5.0
Otter occupancy + Depth + Latitude + Size
0.018
5.2
Depth + Latitude + Size
0.015
5.5
Size + Temp
0.013
5.8
Depth + Latitude + Size + Temp
0.013
5.8
Otter occupancy + Latitude + Temp
0.013
5.9
Latitude + Size + Temp
0.012
6.0
Depth + Size + Temp
0.012
6.0
Otter occupancy + Depth
0.012
6.0
Depth
0.011
6.2
Latitude
0.011
6.2
Otter occupancy + Depth + Size
0.008
6.8
Latitude + Temp
0.008
7.0
Temp
0.007
7.0
Depth + Latitude
0.007
7.2
Otter occupancy + Depth + Temp
0.006
7.5
Otter occupancy + Depth + Latitude
0.004
8.1
Otter occupancy + Depth + Latitude + Temp
0.004
8.2
Depth + Temp
0.003
8.5
Depth + Latitude + Temp
_qt _
PCA
Table S9.
Parameter estimates for best -fitting model describing eelgrass genotypic diversity.
Parameter mean mcse sd 5% 50% 95% Rhat
estimates
Intercept
3.56
0.018
0.92
2.12
3.53
5.14
1.002
Otters Recent
0.03
0.012
0.57
-0.85
0.00
1.01
1.001
Otters
Established
1.04
0.009
0.47
0.28
1.05
1.78
1.001
Meadow Size
-0.26
0.004
0.17
-0.54
-0.26
0.01
1.002
Temperature
0.39
0.004
0.23
0.03
0.38
0.76
1.001
LOG. FEM
Table 51O.`$ �.
Eelgrass (Zostera marina) samples collected from Vancouver Island (VI) and the
Central Coast (CC) where otters were absent, recent, or established. Horizontal lines
through the four right-hand columns denote identical genotypes (i.e. genetic clones).
Total
Sample
Metre
Est.
Area Otters Meadow identical
Transect
samples
ID
mark
length
genotypes
b015
3 60
10
b017
3 70
b031
1 17.5
Wizard 30 6
5
VI
b037 1 10
7.5
b041 1 2.5
Absent b105 1 5
b149 1 40 35
Dodger 31 6
b131 2 50
b146 1 30 25
b109 3 55
Grappler 31 0 na
Barter 32 0 na
Established Kamils 31 0 na
Spring 32 0 na
h088 1 27.5
25
h126 2 2.5
Stirling 31 6
h113 2 10
CC Absent h117 2 15 15
h120 2 25
h080 3 20
Nalau 28 4 hl36 2 30 50
h152 2 80
PC-1
Area Otters Meadow Total identical Sample Transect Metre Est.
samples ID mark length
genotypes
Recent
h149 2 85 5
h145 2 90
h409 3 0 10
h381 5 10
Ward 31 7 IA 1 10
Triquet 31
h417 1 12.5
h420 1 15 10
h407 1 5
h460 2 50 5
h462 2 45
T h474 2 0 15
h477 2 15
h291 1 12.5
7.5
h289 4 20
h329 2 7.5
2.5
h294 4 10
McNaughton 30 10 40
Choked 32 0
h295
4 45
42.5
h318
5 2.5
h308
1 7.5
2.5
h319
5 10
na
Stryker 29 2 h219 1 2.5 25
h231 l 27.5
Louisa 33 2 h238 1 90 10
Established h247 1 80
h333 1 20
McMullins 31 3 h337 1 2.5 17.5
h341 1 10
References and Notes:
1. J. A. Estes, M. Heithaus, D. J. McCauley, D. B. Rasher, B.
structure and function of ocean ecosystems. Annu. Re
(2016). doi: 10. 1 146/annurev-environ- 110615-085622
V --
PC-age_.'
v
Worm, Megafaunal impacts on
. Environ. Resour. 41, 83-116
2. J. A. Estes, J. Terborgh, J. S. Brashares, M. E. Power, J. Berger, W. J. Bond, S. R. Carpenter,
T. E. Essington, R. D. Holt, J. B. C. Jackson, R. J. Marquis, L. Oksanen, T. Oksanen, R.
T. Paine, E. K. Pikitch, W. J. Ripple, S. A. Sandin, M. Scheffer, T. W. Schoener, J. B.
Shurin, A. R. E. Sinclair, M. E. Soule, R. Virtanen, D. A. Wardle, Trophic downgrading
of planet Earth. Science 333, 301-306 (2011). doi:10.1 126/science. 1205106 Medline
3. J. A. Estes, P. D. Steinberg, Predation, herbivory, and kelp evolution. Paleobiology 14, 19-36
(1988). doi:10.1017/50094837300011775
4. D. H. Janzen, P. S. Martin, Neotropical anachronisms: The fruits the gomphotheres ate.
Science 215, 19-27 (1982). doi: 10. 1 126/science.215.4528.19 Medline
5. S. C. Banks, G. J. Cary, A. L. Smith, I. D. Davies, D. A. Driscoll, A. M. Gill, D. B.
Lindenmayer, R. Peakall, How does ecological disturbance influence genetic diversity?
Trends Ecol. Evol. 28, 670-679 (2013). doi:10.1016/j.tree.2013.08.005 Medline
6. J. S. Oliver, P. N. Slattery, Destruction and opportunity on the sea floor: Effects of gray whale
feeding. Ecology 66, 1965-1975 (1985). doi:10.2307/2937392
7. A. R. Templeton, H. Brazeal, J. L. Neuwald, The transition from isolated patches to a
metapopulation in the eastern collared lizard in response to prescribed fires. Ecology 92,
1736-1747 (2011). doi:10.1890/10-1994.1 Medline
8. D. A. Potvin, K. M. Parris, K. L. Smith Date, C. C. Keely, R. D. Bray, J. Hale, S. Hunjan, J. J.
Austin, J. Melville, Genetic erosion and escalating extinction risk in frogs with increasing
wildfire frequency. J. Appl. Ecol. 54, 945-954 (2017). doi:10.1111/1365-2664.12809
9. A. A. Agrawal, A. P. Hastings, M. T. J. Johnson, J. L. Maron, J. P. Salminen, Insect
herbivores drive real-time ecological and evolutionary change in plant populations.
Science 338, 113-116 (2012). doi: 10. 1 126/science. 1225977 Medline
10. J. A. Estes, J. S. Brashares, M. E. Power, Predicting and detecting reciprocity between
indirect ecological interactions and evolution. Am. Nat. 181 (suppl. 1), S76-S99 (2013).
doi:10.1086/668120 Medline
11. M. Hessing-Lewis, E. U. Rechsteiner, B. B. Hughes, M. Tim Tinker, Z. L. Monteith, A. M.
Olson, M. M. Henderson, J. C. Watson, Ecosystem features determine seagrass
community response to sea otter foraging. Mar. Pollut. Bull. 134, 134-144 (2018).
doi:10.1016/'.ma olbul.2017.09.047 Medline
12. S. B. Traiger, B. Konar, A. Doroff, L. McCaslin, Sea otters versus sea stars as major clam
predators: Evidence from foraging pits and shell litter. Mar. Ecol. Prog. Ser. 560, 73-86
(2016). doi: 10.3354/mepsl 1871
13. S. E. Blok, B. Olesen, D. Krause -Jensen, Life history events of eelgrass Zostera marina L.
populations across gradients of latitude and temperature. Mar. Ecol. Prog. Ser. 590, 79-
93 (2018). doi:10.3354/meps 12479
LOG JTUIv-
nA
14. S. Cabago, R. Santos, Seagrass reproductive effort as an ecological indicator of disturbance.
Ecol. Indic. 23, 116-122 (2012). do i: 10. 10 1 6/i.ecolind.2012.03.022
15. Materials and methods are available as supplementary materials.
16. E. Foster, J. Watson, M. A. Lemay, M. T. Tinker, J. A. Estes, R. Piercey, L. Henson, C.
Ritland, A. Miscampbell, L. Nichol, M. Hessing-Lewis, A. K. Salomon, C. T. Darimont,
Zostera marina microsatellite and environmental data, Dryad (2021);
doi:10.5061/dryad. ns 1 rn8pt4
17. A. Caballero, A. Garcia -Dorado, Allelic diversity and its implications for the rate of
adaptation. Genetics 195, 1373-1384 (2013). doi:10.1534/genetics.113.15841 d Medline
18. J. M. Ruiz, L. Marin-Guirao, R. Garcia -Munoz, A. Ramos -Segura, J. Bernardeau-Esteller, M.
Perez, N. Sanmarti, Y. Ontoria, J. Romero, R. Arthur, T. Alcoverro, G. Procaccini,
Experimental evidence of warming -induced flowering in the Mediterranean seagrass
Posidonia oceanica. Mar. Pollut. Bull. 134, 49-54 (2018).
doi:10,1016/j.maMolbul.2017.10.037 Medline
19. E. U. Rechsteiner, J. C. Watson, M. T. Tinker, L. M. Nichol, M. J. Morgan Henderson, C. J.
McMillan, M. DeRoos, M. C. Fournier, A. K. Salomon, L. D. Honka, C. T. Darimont,
Sex and occupation time influence niche space of a recovering keystone predator. Ecol.
Evol. 9, 3321-3334 (2019). doi:10.1002/ece3.4953 Medline
20. S. I. Massa, C. M. Paulino, E. A. Serrao, C. M. Duarte, S. Arnaud-Haond, Entangled effects
of allelic and clonal (genotypic) richness in the resistance and resilience of experimental
populations of the seagrass Zostera noltii to diatom invasion. BMC Ecol. 13, 39 (2013).
doi:10.1186/1472-6785-13-39 Medline
21. S. Arnaud-Haond, C. M. Duarte, F. Alberto, E. A. Serrao, Standardizing methods to address
clonality in population studies. Mol. Ecol. 16, 5115-5139 (2007). doi:10.1 I 11/a.1365-
294X.2007.03535.x Medline
22. R. W. Boessenecker, A middle pleistocene sea otter from northern California and the
antiquity of Enhydra in the Pacific Basin. J. Mamm. Evol. 25, 27-35 (2016).
doi: 10. 1007/s 10914-016-9373-6
23. J. A. Coyer, G. Hoarau, J. Kuo, A. Tronholm, J. Veldsink, J. L. Olsen, Phylogeny and
temporal divergence of the seagrass family Zosteraceae using one nuclear and three
chloroplast loci. Syst. Biodivers. 11, 271-284 (2013).
doi :10.1080/ 14772000.2013.821187
24. S. Cullis-Suzuki, S. Wyllie-Echeverria, K. A. Dick, M. L. D. Sewid-Smith, O. K. Recalma-
Clutesi, N. J. Turner, Tending the meadows of the sea: A disturbance experiment based
on traditional indigenous harvesting of Zostera marina L. (Zosteraceae) the southern
region of Canada's west coast. Aquat. Bot. 127, 26-34 (2015).
doi :10.1016/j.aquabot.20 I 5.07.001
25. R. Felger, M. B. Moser, Eelgrass (Zostera marina L.) in the Gulf of California: Discovery of
its nutritional value by the Seri Indians. Science 181, 355-356 (1973).
doi: 10. 1 126/science. 181.4097.355 Medline
.._-._ 91
26. T. B. Reusch, A. Ehlers, A. Hammerli, B. Worm, Ecosystem recovery after climatic extremes
enhanced by genotypic diversity. Proc. Natl. Acad. Sci. U.S.A. 102, 2826-2831 (2005).
doi: 10. 1 0731pnas.050000 8102 Medline
27. A. R. Hughes, J. J. Stachowicz, Genetic diversity enhances the resistance of a seagrass
ecosystem to disturbance. Proc. Natl. Acad. Sci. U.S.A. 101, 8998-9002 (2004).
doi: 10. 1 073/pnas.0402642 101 Medline
28. B. B. Hughes, R. Eby, E. Van Dyke, M. T. Tinker, C. I. Marks, K. S. Johnson, K. Wasson,
Recovery of a top predator mediates negative eutrophic effects on seagrass. Proc. Natl.
Acad. Sci. U.S.A. 110, 15313-15318 (2013). doi: 10. 1 073/pnas.1 302805110 Medline
29. A. Elson, D. C. Hartnett, Bison increase the growth and reproduction of forbs in tallgrass
prairie. Am. Midl. Nat. 178, 245-259 (2017). doi: 10. 1674/0003-0031-178.2.245
30. C. Eichberg, T. W. Donath, Sheep trampling on surface -lying seeds improves seedling
recruitment in open sand ecosystems. Restor. Ecol. 26, 211-219 (2018).
doi:10.1 I I I /rec.12650
31. J. Watson, in Workshop on Rebuilding Abalone Stocks in British Columbia, A. Campbell, Ed.
(Canadian Special Publication Fisheries and Aquaculture Science, 2000), vol. 130, pp.
123-132.
32. N. A. Sloan, L. Dick, Sea otters of Haida Gwaii: Icons in Human -Ocean Relations, (Gwaii
Haanas Archipelago Management Board and Haida Gwaii Museum, 2012).
33. M. A. Bigg, I. B. MacAskie, Sea otters reestablished in British Columbia. J. Mammal. 59,
874-876 (1978). doi: 10.2307/1380I63
34. L. M. Nichol, J. C. Watson, R. Abernethy, E. Rechsteiner, J. Towers, "Trends in the
abundance and distribution of sea otters (Enhydra lutris) in British Columbia updated
with 2013 survey results," (DFO CSAS, 2015);
35. S. T. A. Pickett, in Long-term Studies in Ecology, G. E. Likens, Ed. (Springer, 1989), chap.
5, pp. 110-135.
36. R. M. Connolly, T. M. Smith, P. S. Maxwell, A. D. Olds, P. I. Macreadie, C. D. H. Sherman,
Highly disturbed populations of seagrass show increased resilience but lower genotypic
diversity. Front. Plant Sci. 9, 894 (2018). doi:10.3389/fpis.2018.00894 Medline
37. A. Alexandre, R. Santos, E. Serrao, Effects of clam harvesting on sexual reproduction of the
seagrass Zostera noltii. Mar. Ecol. Prog. Ser. 298, 115-122 (2005).
doi: 10.3354/meps298115
38. S. H. Kim, J. H. Kim, S. R. Park, K. S. Lee, Annual and perennial life history strategies of
Zostera marina populations under different light regimes. Mar. Ecol. Prog. Ser. 509, 1-
13 (2014). doi:10.3354/me s10899
39. L. Marin-Guirao, L. Entrambasaguas, J. M. Ruiz, G. Procaccini, Heat -stress induced
flowering can be a potential adaptive response to ocean warming for the iconic seagrass
Posidonia oceanica. Mol. Ecol. 28, 2486-2501 (2019). doi:10. l 1 1 1 /mec.15089 Medline
40. J. L. Olsen, W. T. Stam, J. A. Coyer, T. B. H. Reusch, M. Billingham, C. Bostrom, E.
Calvert, H. Christie, S. Granger, R. la Lumiere, N. Milchakova, M.-P. Oudot-Le Secq, G.
Procaccini, B. Sanjabi, E. Serrao, J. Veldsink, S. Widdicombe, S. Wyllie-Echeverria,
North Atlantic phylogeography and large-scale population differentiation of the seagrass
Zostera marina L. Mol. Ecol. 13, 1923-1941 (2004). doi:10.1.1.11/j.1365-
294X.2004.02205.x Medline
41. S. R. Fain, A. DeTomaso, R. S. Alberte, Characterization of disjunct populations of Zostera
marina (eelgrass) from California: Genetic differences resolved by restriction -fragment
length polymorphisms. Mar. Biol. 112, 683-689 (1992). doi: 10. 1007/BF00346187
42. L. K. Reynolds, J. J. Stachowicz, A. R. Hughes, S. J. Kamel, B. S. Ort, R. K. Grosberg,
Temporal stability in patterns of genetic diversity and structure of a marine foundation
species (Zostera marina). Heredity 118, 404-412 (2017). doi: 10. 1038/hdy.2016.114
Medline
43. K. P. Burnham, D. R. Anderson, Model Selection and Multimodel Inference: A Practical
Information -theoretic Approach. (Springer-Verlag, 2002).
44. N. K. Nahirnick, L. Reshitnyk, M. Campbell, M. Hessing-Lewis, M. Costa, J. Yakimishyn,
L. Lee, Mapping with confidence; delineating seagrass habitats using Unoccupied Aerial
Systems (UAS). Remote Sens. Ecol. Conserv. 5, 121-135 (2018). doi: 10. 1 002/rse2.98
45. L. Reshitnyk, C. L. K. Robinson, P. Dearden, Evaluation of WorldView-2 and acoustic
remote sensing for mapping benthic habitats in temperate coastal Pacific waters. Remote
Sens. Environ. 153, 7-23 (2014). doi:10.1016/j.rse.2014.07.015
46. K. Oetjen, T. B. H. Reusch, Identification and characterization of 14 polymorphic EST -
derived microsatellites in eelgrass (Zostera marina). Mol. Ecol. Notes 7, 777-780 (2007).
doi:10.1111/6.1471-8286.2007.01694. x
47. T. B. H. Reusch, W. T. Stam, J. L. Olsen, Microsatellite loci in eelgrass Zostera marina
reveal marked polymorphism within and among populations. Mol. Ecol. 8, 317-321
(1999). doi:10.1046/'.1365-294X.1999.00531.x Medline
48. K. Oetjen, S. Ferber, I. Dankert, T. B. H. Reusch, New evidence for habitat -specific selection
in Wadden Sea Zostera marina populations revealed by genome scanning using SNP and
microsatellite markers. Mar. Biol. 157, 81-89 (2010). doi: 10. 10071s00227-009-1297-8
49. M. Jahnke, P. R. Jonsson, P.-O. Moksnes, L.-O. Loo, M. Nilsson Jacobi, J. L. Olsen,
Seascape genetics and biophysical connectivity modelling support conservation of the
seagrass Zostera marina in the Skagerrak -Kattegat region of the eastern North Sea. Evol.
Appl. 11, 645-661 (2018). doi:10.1111/eva.12589 Medline
50. J. L. Olsen, J. A. Coyer, B. Chesney, Numerous mitigation transplants of the eelgrass Zostera
marina in southern California shuffle genetic diversity and may promote hybridization
with Zostera pacifica. Biol. Conserv. 176, 133-143 (2014).
doi:10.1016/ j.biocon.2014.05.001
51. Z. N. Kamvar, J. F. Tabima, N. J. Grunwald, Poppr: An R package for genetic analysis of
populations with clonal, partially clonal, and/or sexual reproduction. PeerJ 2, e281
(2014). doi: 10.77171peeri.281 Medline
_of 5-3
52. J. C. Parks, C. R. Werth, A study of spatial features of clones in a population of bracken fern,
Pteridium aquilinum (Dennstaedtiaceae). Am. J. Bot. 80, 537-544 (1993).
doi:I0.I002Ij,.1537-2197.1993.tb13837.x Medline
53. J. H. Kim, J. H. Kang, J. E. Jang, S. K. Choi, M. J. Kim, S. R. Park, H. J. Lee, Population
genetic structure of eelgrass (Zostera marina) on the Korean coast: Current status and
conservation implications for future management. PLOS ONE 12, e0174105 (2017).
doi:10.1371/journal.pone.0174105 Medline
54. C. van Oosterhout, W. F. Hutchinson, D. P. M. Wills, P. Shipley, MICRO -CHECKER:
Software for identifying and correcting genotyping errors in microsatellite data. Mol.
Ecol. Notes 4, 535-538 (2004). doi:10.1 111 J.1471-8286.2004.00684.x
55. S. W. Guo, E. A. Thompson, Performing the exact test of Hardy -Weinberg proportion for
multiple alleles. Biometrics 48, 361-372 (1992). do!: 10.2307/2532296 Medline
56. F. Rousset, genepop'007: A complete re -implementation of the genepop software for
Windows and Linux. Mol. Ecol. Resour. 8, 103-106 (2008). doi:10.111 1/j.1471-
8286.2007.01931.x Medline
57. W. R. Rice, Analyzing tables of statistical tests. Evolution 43, 223-225 (1989).
doi:10.1111/j.1558-5646.1989.tb04220.x Medline
58.T. C. Hsieh, H. Ma, A. Chao, R package version 2.0.19, (2019).
59. A. Chao, N. J. Gotelli, T. C. Hsieh, E. L. Sander, K. H. Ma, R. K. Colwell, A. M. Ellison,
Rarefaction and extrapolation with Hill numbers: A framework for sampling and
estimation in species diversity studies. Ecol. Monogr. 84, 45-67 (2014). doi:I0.1890/i3-
0133_1
60. J. K. Pritchard, M. Stephens, P. Donnelly, Inference of population structure using multilocus
genotype data. Genetics 155, 945-959 (2000). doi:10.10931genetics/155.2.945 Medline
61. G. Evanno, S. Regnaut, J. Goudet, Detecting the number of clusters of individuals using the
software STRUCTURE: A simulation study. Mol. Ecol. 14, 2611-2620 (2005).
doi:10.1 I I I/j.1365-294X.2005.02553.x Medline
62. D. Earl, B. Vondoldt, Structure Harvester: A website and program for visualizing
STRUCTURE output and implementing the Evanno method. Conserv. Genet. 4, 359-361
(2012). doi:10.1007/s12686-011-9548-7
63. T. Jombart, adegenet: A R package for the multivariate analysis of genetic markers.
Bioinformatics 24, 1403-1405 (2008). doi: 10. 1093/bioinformatics/btn 129 Medline
64. R. Peakall, P. E. Smouse, GenAlEx 6.5: Genetic analysis in Excel. Population genetic
software for teaching and research -an update. Bioinformatics 28, 2537-2539 (2012).
doi:10.1093/bioinformatics/bts460 Medline
65. A. T. Adamack, B. Gruber, PopGenReport: Simplifying basic population genetic analyses in
R. Methods Ecol. 5, 384-387 (2014). doi:10.I 111/2041-210X.12158
66. A. El Mousadik, R. J. Petit, High level of genetic differentiation for allelic richness among
populations of the argan tree [Argania spinosa (L.) Skeels] endemic to Morocco. Theor.
Appl. Genet. 92, 832-839 (1996). doi:10.1007/BF00221895 Medline
67. M. E. Dorken, C. G. Eckert, Severely reduced sexual reproduction in northern populations of
a clonal plant, Decodon verticillatus (Lythraceae). J. Ecol. 89, 339-350 (2001).
doi:10.1046/i.1365-2745.2001.00558.x
68. B. Comps, D. Gomory, J. Letouzey, B. Thiebaut, R. J. Petit, Diverging trends between
heterozygosity and allelic richness during postglacial colonization in the European beech.
Genetics 157, 389-397 (2001). doi:10.10931 enetics/157.1.389 Medline
69. A. R. Hughes, J. J. Stachowicz, Ecological impacts of genotypic diversity in the clonal
seagrass Zostera marina. Ecology 90, 1412-1419 (2009). doi:10.1890/07-2030.1 Medline
70. G. Procaccini, J. L. Olsen, T. B. H. Reusch, Contribution of genetics and genomics to
seagrass biology and conservation. J. Exp. Mar. Biol. Ecol. 350, 234-259 (2007).
doi_ 10.1016/i . i embe.2007.05.03 5
71. S. L. Talbot, G. K. Sage, J. R. Rearick, M. C. Fowler, R. Muniz-Salazar, B. Baibak, S.
Wyllie-Echeverria, A. Cabello-Pasini, D. H. Ward, The structure of genetic diversity in
eelgrass (Zostera marina L.) along the north Pacific and Bering sea coasts of Alaska.
PLOS ONE 11, e0152701 (2016). doi:10.I371Jjournal.pone.0152701 Medline
72. B. Goodrich, J. Gabry, I. Ali, S. Brilleman, rstanarm: Bayesian applied regression modeling
via Stan. R package version 2.21.1, (2020).
73. J. Gabry, T. Mahr, bayesplot: Plotting for Bayesian Models R package version 1.7.9, (2019).
74. H. Wickham, ggplot2: Elegant Graphics for Data Analysis. (Springer, 2016).
75. J. Gabry, D. Simpson, A. Vehtari, M. Betancourt, A. Gelman, Visualization in Bayesian
workflow. J. R. Stat. Soc. Ser. A 182, 389-402 (2019). doi:10.1 1 1 llrssa.12378
LOG ITS.
# , Y/
Pe-9
a 01 5�3
Physical disturbance by recovering sea otter populations increases eelgrass
genetic diversity
Erin Foster et al.
Corresponding author: Erin Foster, erin.foster@hakai.org
Science 374, 333 (2021)
DOI: 10.1126/sci ence.abf2343
Abstract
Most knowledge regarding the role of predators is ecological in nature. Here, we report how
disturbance generated by sea otters (Enhydra lutris) digging for infaunal prey in eelgrass (Zostera
marina) meadows increases genetic diversity by promoting conditions for sexual reproduction of plants.
Eelgrass allelic richness and genotypic diversity were, respectively, 30 and 6% higher in areas where
recovering sea otter populations had been established for 20 to 30 years than in areas where they had
been present <10 years or absent >100 years. The influence of sea otter occupancy on the
aforementioned measures of genetic diversity was stronger than those of depth, temperature, latitude,
or meadow size. Our findings reveal an underappreciated evolutionary process by which megafauna
may promote genetic diversity and ecological resilience.
"In addition to environmental influences, agents of biotic disturbance not related to sea otter foraging
may occur. In the closely related Zostera noltii, digging by clam harvesters disturbs rhizome mats and
induces flowering, thereby increasing sexual reproduction (37). Traditional harvesting of Z. marina
rhizomes by BC Indigenous peoples may have also promoted flowering and genetic diversity in eelgrass
(24), a practice that declined with the arrival of Europeans. Elsewhere, Z. marina is subjected to shoot
loss through grazing by waterfowl (27). We did not see evidence of biotic disturbances (i.e., harvests,
disease, vertebrate grazers), other than sea otter digging, at any of the meadows we examined,
suggesting that where alternative biotic agents of disturbance occur, the present-day effects are
minimal."
OG ITE M
pa
EELGRASS IN SQUAMISH BAY-�:-,��
Statement of Brad Nelson
NWS-2017-230 AQ
NWS-2013-1268 AQ
January 30, 2022
About Native And Invasive Eelgrass
There has been much discussion about eelgrass in Squamish Bay, and the claimed effects of
geoduck aquaculture on that eelgrass. I'd like to summarize my own experience in that regard,
all gained through direct observation.
I think everyone acknowledges that generally a dense eelgrass bed of native Zostera Marina
runs east and west along Squamish Bay, from the hood canal bridge to near the incoming river.
This eelgrass bed exists from about the -1 to -2 seawater elevation down to about the -12 ft
elevation. If you could see it all, it would be about 60 to 120 feet wide depending on beach
contour. At the lowest tides in the summer, you can only see about 12 to 20 feet of it. This is the
natural environment for eelgrass in Squamish Bay. Typically, directly below the eelgrass are
kelp beds. This may be why the eelgrass doesn't go deeper. Probably a little rocky, which kelp
needs to attach. Below the kelp beds is sand starting at about 30 feet deep.
My first introduction to Squamish Bay was in 2011 when a company called Mocean Shellfish
contacted me to help harvest and market the product from its existing geoduck aquaculture
project located at about 600 Shine Road. What I noticed at the time, besides that the geoducks
grown by Mocean were beautiful white #1 quality geoducks, was that above the natural eel grass
delineation level the beaches were totally clear of any vegetation with the exception of some
sparse Japanese Eelgrass (Zostera Japonica) at the higher levels on the nearby beaches. (There
are two common varieties of eelgrass in the bay, one called "Native Eelgrass" (Zostera Marina)
and the other "Japanese Eelgrass" (Zostera Japonica) The Native Eelgrass is protected, while
the Japanese eelgrass is considered an unwelcome invasive species.)
In the summer of 2012 we began to harvest the Mocean beach. At that time I noticed the
Japonica was actually increasing in density and expanding lower onto the beaches. I also noticed
Japonica becoming intermixed with the Marina at the higher levels where the Marina grew.
I wound up purchasing the tidelands from Mocean, along with some other tidelands to the west,
and at the time of my first plantings of new geoduck seed in this area in 2013 the main body of
these beaches was still clear of any vegetation. By the time I planted in 2014, the former Mocean
parcel was being invaded by more Japonica covering most of the beach.
Eelgrass in Squamish Bay, Nelson — Page - 1
By 2015 we were seeing some native Marina beginning to take hold in the planted areas on all
of these parcels where the invasive Japonica was expanding. Not much Marina, but a few shoots
here and there, along with a few very small patches.
The thick Marina beds that exist in the lower minus tide elevations cannot exist at higher levels
without some help. The help is coming from the invasive Japonica that is tending to stabilize the
beach enough that some Marina can remain secure at these higher elevations enough to survive
the winter storms. In addition to the Japonica, I have observed that the addition of our tubes also
contributes to stabilizing the beach even more, resulting in more Marina survival outside of its
normal subtidal areas.
The Smersh beach that I am seeking to develop into a geoduck aquaculture project is located
further to the east, at about 1200 Shine Road, This property is currently essentially devoid of
any Japonica and therefore no Marina is taking hold outside of its natural area.
The Marina eelgrass bed in the Squamish Bay area is very strong and healthy. There is concern
by a few folks that the Marina bed is receding. I assure you this is not the case. It is absolutely
normal for the Marina to expand and recede at its higher subtidal beach elevations. It naturally
expands when conditions allow. And that native Marina can also be totally dislodged and
destroyed due to adverse winter storms and wave action that can tear up the beach's top 2 to 3
inches of sand. I have personally seen this very often on my farm to the west of the Smersh
beach. It is always after heavy weather conditions during a low tide event. I think it's safe to
assume the same is occurring on the Smersh beach along with all the rest of the Squamish Bay
beaches. I'm sure I am one of the few to witness this because I am usually the only person on
the beach at 1:30 AM during the late -night winter low tides, when I am patrolling for any
dislodged geoduck tubes.
The invasive weed Japonica does not seem to get dislodged the way that the native Marina does.
I think this is because the Japonica has a stronger root system. Where the Marina root is
relatively straight, the Japonica root Y's off. I would be happy to show anyone interested.
What Effect Will GeoduckWhat Effect Will Geoduck Aguaculture Have on Eelgrass in the Area?
on Eelgrass in the Area?
I predict that if the Smersh farm is allowed to continue forward, its PVC tubes will stabilize the
beach. That will allow the invasive weed Japonica to take hold, followed by some native
Marina.
By law, I am allowed to remove Japonica eelgrass from my projects, because it is considered an
invasive toxic weed. But I don't remove it, because my geoducks seem to like it. And if I did
remove it, no Marina would exist on the beach either.
Eelgrass in Squamish Bay, Nelson — Page - 2
LOG
-ql_
1 C49 ( 57 .41)
The downside for me is that if I leave the Japonica in place and it encourages new growth of the
native Marina on my farm, there may then be some who wish to terminate all farm activity to
preserve native Marina that is only there because of my geoduck farm.
By allowing the Smersh project to proceed, it would be integrated into a bigger picture of 6
rotating crop cycles. At maturity, one area will be harvested, while another will be replanted.
Yes, it is true that much Japonica and some existing Marina will be displaced during harvest.
But in the big picture, more will be created than will be damaged.
To the naked eye, it is difficult to tell the difference between Marina and Japonica. My guess is
that any forage fish may not be able to tell the difference either. Could it be considered a net
gain of protective sea grass overall?
Again, I would like to invite any interested parties to contact me, and perhaps take an invited
tour of the beaches to learn more.
Sincerely, Brad Nelson BDN LLC
Eelgrass in Squamish Bay, Nelson — Page - 3
LOB ITEM
� 9i
���