20080868 Ver 2_Section II D Q4 Managed Fish 2020 PCS Creeks Report_20210701

D. Question 4- Has mining altered the use of the creek by managed fish?
Multivariate cluster analysis using a similarity profile test (SIMPROF) of managed fish
species composition and abundance from all creeks and all years revealed seven distinct
clusters among the 159 total creek/years (Figure II-D1); to date 13 managed fish species have
been collected (Table II-D1). (Note: included in previous reports as a managed species,
Crevalle jack (Caranx hippos) was removed from the snapper grouper complex in April 2012 by
the South Atlantic Fishery Management Council; therefore, as of this report, this fish is excluded
from this report section.) To simplify discussion of the data set, colored lines on the cluster
dendrograms represent non -significant structure among factors (e.g., years, creeks) at the 1
percent level (P = 0.01). Comparison of interannual variability between clusters by means of
similarity percentages (SIMPER) revealed that dissimilarity ranged from 3 to 59 percent.
Cluster A contained all DCUT11 years except 2020 (2013-2019). Cluster B contained
Broomfield Swamp 2020 and DCUT19 2015-2016, years with low CPUE for all managed fish.
Cluster C contained two pre -Mod Alt L creek/years; Broomfield Swamp 2019 and Tooley 1999
along with two post -Mod Alt L creek/years; DCUT19 2015-2016. Cluster D contained DCUT19
2017 and 2019 along with Huddles Cut 2015. Cluster E contained three control, four pre -Mod
Alt L, and two post -Mod Alt L years across eight different creeks. Cluster F was the largest of all
clusters and contained all creeks excluding Broomfield. Cluster G contained all Huddles Cut
years excluding 2015 and 2017.
Total Number of Years by Creek Type
Cluster ID Creek/Years Pre Post Control Notes
A 7 5 2 0 DCUT11 2013-19 (all fykes)
B 3 1 0 2 DCUT19 2015-16 and Broomfield Swamp 2020
C 4 2 2 0 Broomfield Swamp 2019; Jacks 2016-17 and Tooley 1999
D 3 0 1 2 DCUT19 2017, 2019 and Huddles Cut 2015
E 16 5 2 9 Three different control creeks, four pre -Mod Alt L years, and two post -Mod Alt L years
F 111 20 33 58 All creeks except Broomfield Swamp
G 15 6 9 0 All Huddles Cut years except 2015 and 2017
For the 13 years that Jacks Creek, Tooley Creek, and Muddy Creek were sampled, they
grouped together in the same cluster for nine of the 13; two of the three creeks grouped
together in the same cluster in all 13 years. Those 13 years include 1999-2001 and 2011-2020.
Other than Tooley Creek 1999 and Jacks Creek 2016-2017, which clustered into C, all years for
Jacks, Tooley, and Muddy creeks clustered within D or F. Drinkwater and Jacobs creeks were in
the same cluster as Long Creek (control) for all but one of their matched sample years. All years
for Porter Creek and its control, Duck Creek, clustered in F (Figure II-D1).
Comparison of interannual variability between clusters by means of similarity
percentages (SIMPER) was used to determine which managed fish species drove cluster
formation. A summary of average CPUE of the 13 managed fish species in the seven clusters
is shown in Table II-D1. Cluster A had the lowest total abundance of managed fish of any
cluster (31) and contained only four different species (Table II-D1). Similar to cluster A, cluster
B had a low total abundance (54) and also contained four different species. Cluster C was
characterized by moderate abundances of spot, Atlantic croaker, and Atlantic menhaden and
low numbers of two other managed species (Table II-D1). Cluster D had moderate abundances
of spot with low totals for six other managed species (Table II-D1). Cluster E contained 11 of
the 13 species and higher totals for Atlantic menhaden and spot but a lower abundance of
Atlantic croaker (Table II-D1). Cluster F was the largest cluster with 10 of the 13 managed
species high species richness and high abundance of spot, Atlantic croaker, and Atlantic
menhaden (Table II-D1). Cluster G was the most diverse cluster and included all 13 managed
species with Atlantic menhaden, spot, and striped mullet as the most abundant (Table II-D1).
II-D-1

Comparison of interannual variability by means of ANOSIM detected no spatial
differences of significance between pre- and post -Mod Alt L managed fish assemblages within
drainage basins of Jacobs Creek, Drinkwater Creek, Huddles Cut, Porter Creek, and DCUT11;
however, spatial differences of significance were detected between pre- and post -Mod Alt L
managed fish assemblages within Jacks Creek and Tooley Creek [Global R = 0.316; P = 0.021
and Global R = 0.311; P = 0.041].
As discussed earlier in Section II-C Question 3 (forage base), low total CPUE observed
locally throughout South Creek and surrounding tributaries in both 2016-2017 was repeated for
CPUE of managed species as well. The most abundant managed species captured in the 10
trawled creeks were spot, Atlantic croaker, and Atlantic menhaden, with spot typically the most
abundant of the three. An overall decline in CPUE of spot was observed in most creeks since
2013; however, 2016-2017 represented the lowest or close to the lowest CPUE of spot
experienced within all creeks/years. Similar to total CPUE, CPUE of spot for Jacks Creek 2016-
2017 represented two of the three lowest CPUE's for spot across all trawl creeks/years; post -
Mod Alt L spot catch for Jacks Creek was most affected by this trend, as 2016-2017 comprise
two of the six post -Mod Alt L years. The decrease in CPUE of spot throughout South Creek and
surrounding tributaries along with significant increases in CPUE of Atlantic croaker in post -Mod
Alt L years of both Drinkwater and Tooley creeks [small (n) = 26, big (n) = 104; T = 1,093.0; P =
<0.001 and small(n) = 65, big(n) = 117; T = 4,851.5; P = 0.001] accounted for differences
between managed fish assemblages in pre- and post -Mod Alt L years.
As also discussed earlier in Section II-C Question 3 (forage base), low total CPUE for
Huddles Cut since 2009 was likely affected to a large degree by the formation of a sandbar at
the mouth of Huddles Cut. This sandbar limits water exchange and fish access into and out of
the creek and likely affected CPUE for managed fish as well. Spot, striped mullet, and Atlantic
menhaden were the three most abundant species of managed fish captured in Huddles Cut,
with spot typically the most abundant of the three. Along with impeded access into and out of
the mouth of Huddles Cut, there was the overall decline in CPUE of spot since 2013 mentioned
above; however, Huddles Cut was the only creek where CPUE of spot increased from 2015-
2020. Spot CPUE at Huddles Cut in 2016 was over 10 times the CPUE observed for spot in
2015, nearly doubled from 2016 to 2017, doubled from 2017 to 2018, and continued to increase
in 2019 and 2020 to the second highest CPUE of spot encountered for the 17 Huddles Cut
years. Although striped mullet catch was nearly non-existent in both 2015 (seven individuals)
and 2017 (five individuals) at Huddles Cut, CPUE of striped mullet in Huddles Cut remained
higher than in DCUT19 2013-2016 and 2018-2020 (a control creek also fyke netted and
adjacent to the Pamlico River) with a large catch increase in 2020 (251 individuals). The
continued increase in CPUE of spot at Huddles Cut in recent post -Mod Alt L years likely
resulted in no spatial differences of significance being detected between pre- and post -Mod Alt
L managed fish assemblages within the drainage basin of Huddles Cut through 2020.
The data do not specifically indicate that mine activities have altered managed fish
communities. When pre- and post -Mod Alt L years from five of seven creeks were compared,
there was no statistical indication of any detectable difference. It is also not valid to assume
reduction of the drainage basin of the other two creeks has altered fish assemblage since no
cluster contained only post -Mod Alt L creek/years and most post -Mod Alt L creek/years
clustered among other pre -Mod Alt L creeks/years for the same creek, and/or control
creeks/years (see Section III for further analysis).
As the number of creeks sampled increased since 2011, capture of managed fish
species also increased. In all, 13 managed species have been captured, with a low of five
species in 2004 (Jacks and Muddy creeks were the only creeks sampled) to a high of 11
II-D-2

species in 2001 and 2012. Huddles Cut in 2020 had the highest number of managed species
captured in an individual creek/year with nine. The most abundant managed species captured
across all creeks were spot, Atlantic croaker, and Atlantic menhaden. Of the 13 managed
species captured, four diadromous species were collected across all creeks/years: alewife
(Alosa pseudoharengus), American eel (Anguilla rostrata), American shad (Alosa sapidissima),
and striped bass (Morone saxatilis). These four diadromous species represented a total of 400
individuals; however, American eel alone accounted for 355 (89 percent).
Penaeid shrimp and blue crabs are also managed species but were not enumerated
during the creeks study from 1999 through 2010. Collection of more detailed qualitative
information on shrimp and crabs was initiated in 2011 in conjunction with fish collections (via
trawl at all creeks except Huddles Cut, DCUT11, and DCUT19 which are via fyke nets; DCUT11
and DCUT19 were sampled in 2013 for the first time). The limited data prevent detailed
evaluation; however, penaeid shrimp were most frequently captured (69 percent) from Duck
Creek in 2011 and the highest numbers of penaeid shrimp collected were from Tooley Creek in
2012. The frequency and numbers of penaeid shrimp seemed to be in decline in all creeks
since 2011-2012, until a slight increase in 2015-2020 in all creeks except for DCUT11. No
penaeid shrimp were captured in DCUT11 upstream or downstream fyke nets in any year and
were captured only in 2013, 2018, and 2020 in DCUT19 upstream fyke net. Additionally, no
penaeid shrimp were captured in 2013 from PA2 (trawl) or Huddles Cut downstream fyke net, in
2014 trawls from Jacks Creek, Little Creek, PA2, Long Creek, Porter Creek, or Huddles Cut
downstream fyke net, in 2015 from Huddles Cut upstream fyke net, in 2016 trawls from both
Jacks Creek and Porter Creek, in 2017 from Porter Creek, in 2019 from Huddles Cut
downstream fyke net, or Broomfield Swamp Creek, SCUT1 and Duck Creek trawls, or in 2020
from Broomfield Swamp Creek (Table II-C3).
The frequencies at which blue crab were captured and the numbers collected were both
highest among all creeks for Huddles Cut upstream fyke net in 2016 and 2019 and downstream
fyke nets in 2017; the upstream Huddles Cut fyke net frequency tied with Jacobs Creek in 2017
for highest. No blue crabs were captured at DCUT11 during 2013-2016, 2018, and 2020;
however, blue crabs were collected in DCUT11 downstream fyke net in 2017 and in both fyke
nets in 2019. Blue crabs were also absent in 2013 trawls from Jacks Creek and Little Creek,
and from DCUT19 downstream fyke net; furthermore, none were captured in 2014 trawls from
Long Creek, Muddy Creek, and Duck Creek, or DCUT19 upstream fyke net (Table II-C3). The
frequencies at which both penaeid shrimp and blue crab were captured and the numbers
collected were highly variable across all creeks except DCUT11 where no penaeid shrimp have
been captured and blue crab have only been collected twice. No apparent trends can be
detected from the 10 years of blue crab data.
Answer: The multivariate cluster analysis of managed fish for all creeks and all
collection years reveals some differences based on gear type (fyke net vs trawl) and also
separates some pre -Mod Alt L and post -Mod Alt L years within clusters; however, the
analysis does not reveal distinct changes in the assemblages of managed fish due to
mine activities within the drainage basins of Jacks Creek, Jacobs Creek, Drinkwater
Creek, Tooley Creek, Huddles Cut, Porter Creek, and/or DCUT11.
Comparison of interannual variability by means of ANOSIM detected spatial
differences of statistical significance between pre- and post -Mod Alt L managed fish
assemblages of Jacks Creek and Tooley Creek. Low total CPUE observed locally
throughout South Creek and surrounding tributaries in 2016-2017 trawl samples also
likely affected CPUE of all managed species in every creek; however, low CPUE of spot
II-D-3

and higher CPUE of Atlantic croaker likely contributed to separation of post -Mod Alt L
years from pre -Mod Alt L years in Tooley Creek. Lower CPUE of spot in post -Mod Alt L
years in Jacks Creek likely contributed to separation of post -Mod Alt years from pre -Mod
Alt L years as two of the six post -Mod Alt L years for Jacks Creek were 2016 and 2017.
Although there is a decline in CPUE of some managed species in Huddles Cut in
post -Mod Alt L years, the continued increase in CPUE of spot at Huddles Cut in post -Mod
Alt L 2015-2020 and persistent catches of Atlantic menhaden in both pre- and post -Mod
Alt L years contributed to no detection of spatial differences of significance between pre -
and post -Mod Alt L managed fish assemblages through 2020. Both diversity and
abundance of managed fish captured at Huddles Cut remain higher than most all other
sampled creeks and higher than DCUT11 and DCUT19 for 2013-2020 (data period for the
only other creeks sampled via fyke nets).
No data set for some pre -Mod Alt L years in some creeks and variability in the frequency
and numbers of both penaeid shrimp and blue crab collected make it difficult to discern
any mine -related spatial patterns in the abundance of either species.
II-D-4

c
E
0
CD
0
N
d —
A BC DE
PCS Fish Collections -Managed Fish
Group Average
F
}III 1 1 I ., I I V I I I I N 6: II 441 INininvwi�-rina itdNnlinl L6I N N1,.,I ',�N IL I I I
{ 1 1 I I I I D , .� • �.lti Do
iflp
��"���D�
ci oar
11..S1 Q_ C] fa C] R._.Ja G]
�I LL J I J h 1 I L f.L !_ u{ !, Il a I I I I I I I I[ 1
�'�j Ap.)(Einiz,
Q� Qom_ NNNNNNNN,,DcDo
:�Sp00= [ NNN!➢NNN0767�NN
`-� Y Y37731777733,
L. A. A
Figure II-D1. Dendrogram of hierarchical clusters of similarity for fish community abundance and composition among managed fish species for all creeks and years sampled [Bray -Curtis similarity; Log(x+1)]. Black lines
within dendrogram represent statistically significant cluster structure and colored lines represent non -significant cluster structure at the 1 percent level (P = 0.01). Gray creeks/years are pre -Mod Alt L, bold creeks/years are
post -Mod Alt L, and blue creeks/years are control creeks.
II-D-5

Table II-D1. Catch-per-unit-efforta (CPUE) average for managed fish species caught across seven groups identified by cluster analysis in study collection years through 2020. Fish collections occurred in
April, May, and June each year 1999 through 2005 and 2007 through 2020.
Common name
Scientific name
CPUEa average by species per cluster
Cluster A (7)
Cluster B (3)
Cluster C (4)
Cluster D (3)
Cluster Eb (16)
Cluster Fb (111)
Cluster G (15)
Alewife
Alosa pseudoharengus
0.00
0.00
0.00
0.00
0.01
0.01
0.06
American eel
Anguilla rostrata
0.15
0.00
0.23
0.00
0.69
0.11
0.21
American shad
Alosa sapidissima
0.00
0.00
0.00
0.00
0.00
0.00
0.01
Atlantic croaker
Micropogonias undulatus
0.00
0.41
5.12
0.13
1.08
32.62
3.61
Atlantic menhaden
Brevoortia tyrannus
0.00
0.03
10.85
0.69
13.77
9.86
29.41
Bluefish
Pomatomus saltatrix
0.00
0.00
0.00
0.00
0.00
0.00
0.05
Red drum
Sciaenops ocellatus
0.00
0.00
0.00
0.03
0.00
0.01
0.78
Southern flounder
Paralichthys lethostigma
0.01
0.00
0.27
0.38
0.12
0.28
1.35
Spot
Leiostomus xanthurus
0.12
0.92
5.00
6.46
45.88
103.84
245.70
Spotted seatrout
Cynoscion nebulosus
0.00
0.00
0.00
0.00
0.04
0.05
0.11
Striped bass
Morone saxatilis
0.00
0.00
0.00
0.00
0.00
0.00
0.01
Striped mullet
Mugil cephalus
0.05
0.00
0.00
0.72
0.66
0.08
29.91
Summer flounder
Paralichthys dentatus
0.00
0.03
0.00
0.05
0.08
0.19
0.91
a Number of individuals caught during an approximate 16-hour set of fyke nets or one minute, 75-yard trawl
b Individuals of bluefish (5) and striped bass (3) were captured across Mod Alt L sample years within cluster F along with bluefish (1) and striped bass (1) in E, but CPUE remained less than 0.00