20080868 Ver 2_Section III A Salinity 2020 PCS Creeks Report_20210701

III. ADDITIONAL DISCUSSION OF SUMMARY RESULTS BY PARAMETER
A. SALINITY AND ENVIRONMENTAL VARIABLES
For summary information on rainfall, other weather related events, and Tar River
discharge refer to Sections I-D through I-G. All salinity monitoring locations are associated with
South Creek and its tributaries, Pamlico River and its tributaries, or Durham Creek and its
tributaries. Both South Creek and Durham Creek are large tributaries to the Pamlico River (the
Tar River becomes the Pamlico River at Washington NC, approximately 16.9 miles upstream
from Durham Creek). The Pamlico River salinity monitor (PS1) was removed prior to Hurricane
Florence in September of 2018 for safety. Due to storm damage to the breakwater and pier,
PS1 was not re -deployed until early 2020 when repairs were completed. Data from salinity
monitors at HS1 and HS2 have endured the most loss over recent sample years due to damage
caused by unstable and dry conditions. Because of this, 2020 will represent the last full year of
salinity data for these two stations.
Table III -Al shows the 2020 monthly mean, maximum and minimum, and annual salinity
for each salinity monitor location; salinity distributions at most locations were the highest in
January, September, and October. Appendix C contains a figure showing Tar River discharge
for 2020, as well as graphs depicting the semi -continuous salinity and depth data from each
Aqua TROLL monitor (Appendices B — I are included only on the CD/DVD that accompanies
this report). On the Appendix C graphs, salinity data omitted from analysis, or missing data,
have been categorized as one of the following: insufficient water, equipment malfunction,
service error, or other.
1.0 Environmental Variables
Salinity of coastal streams can be affected by a variety of biotic and abiotic
factors. This section will determine how salinity within the monitored creeks was affected by
four environmental variables: Tar River discharge, water depth, rainfall, and wind (wind data
were supplied by PCS Aurora NOAA Station 6N).
a. Salinity and Tar River Discharge
Because the Tar River carries mostly freshwater, it is expected that
increases in discharge should result in decreases in salinity at all monitored PCS creeks (i.e.,
they should be negatively correlated). Figure III -Al shows daily Tar River discharge from 1998
to 2020 (with the exception of 2006 when no creeks were monitored per the 1998 plan) and
Figure III-A2 shows discharge at impact creeks during pre- and post -Mod Alt L periods. The
median and the mean Tar River discharge in 2020 were the second and third highest,
respectively, since 1998.
As in last year's report, relationships between salinity and Tar River
discharge were analyzed using a cross -correlation function which incorporated daily time lags.
For example, correlations were computed between salinity on one day and the discharge of Tar
River one day prior, two days prior, three days prior, and so on. Discharge and salinity data
were log -transformed prior to analyses. Table III-A2 shows the time lag that yielded the
strongest negative correlation between discharge and salinity for each Aqua TROLL, as well as
the value of Pearson's r, the correlation coefficient used in the analysis (Pearson's r ranges from
-1 to 1, where -1 indicates a strong, negative relationship, 0 indicates no relationship, and +1
indicates a strong, positive relationship). All Aqua TROLLs showed weak to strong negative
correlations between salinity and Tar River discharge, ranging from -0.176 (SC1S1) to -0.661
III-A-1

(PS1). Regarding Tar River discharge, all but one Aqua TROLL (HS1) had their most significant
time lag occur between zero and eight days.
Salinity/discharge correlations for impact creeks were similar to those for
control creeks, suggesting that mine activities have not altered how salinity in these creeks
responds to changes in Tar River discharge.
b. Salinity and Water Depth
Water depths for 2020 at each Aqua TROLL are displayed in the figures
presented in Appendix C. Because both the depth and salinity measurements are taken from
the Aqua TROLL, no time lag is expected between these two variables. Thus, a simple
correlation between the daily means of salinity and depth was conducted.
Only four Aqua TROLLs out of 31 displayed a negative correlation
between water depth and salinity ranging from -0.104 (DCS1) to -0.201 (SC1S3) (Table III-A2).
The nine significantly positive relationships ranged from +0.126 (LCS2) to +0.405 (SC1S1)
(Table III-A2). Since 2015, while most correlations have been significant statistically, the
relationship between salinity and water depth was mostly weak to intermediate with a range of
significant correlations from —0.61 (DWS2 2018) to +0.67 (DWS1 2017).
There does not appear to be any difference between impact and control creeks in how salinity
correlated with water depth.
c. Salinity and Rainfall
Rainfall is summarized in Section I-F (Figure I-F1, Table I-F1, and Table
I-F2). Figure III-A3 shows yearly rainfall at all impact creeks with pre- and post -Mod Alt L data.
Rainfall is expected to decrease salinity, as the rain introduces freshwater into the creeks.
There may also be some temporal lag in the relationship between rainfall and salinity, as it
would take time for precipitation on land to filter into the creeks and to the monitor locations.
Thus, cross -correlation functions were conducted between daily mean rainfall and daily mean
salinity at each Aqua TROLL. For each Aqua TROLL, rainfall data were obtained from the
nearest installed rain gauge.
Excluding HS1, all Aqua TROLLs showed weak, significant negative
correlations between salinity and rainfall in 2020 (Table III-A2). Significant correlations ranged
from -0.131 (HS3 and PA2S1) to -0.296 (BSCS2). Significant time lags for salinity/rainfall
relationships ranged from one day to 11 days, with 26 of 30 Aqua TROLLs representing one to
five day lags.
The salinity in both impact and control creeks seems to have responded
similarly to changes in rainfall, both in terms of strength of the relationship as well as the time
lag with the strongest response.
d. Salinity and Wind
There are two important components of wind: speed and direction.
Salinity in the Pamlico River increases on a gradient from west to east, and strong easterly and
northeasterly winds are expected to push more saline waters into the PCS creeks study area.
Other factors which interact to some degree include duration of wind event, distance of monitor
from mouth of creek, width of mouth of creek, which side of creek/river the tributary occupies,
and orientation of creek axis relative to larger water body and to the wind. Summary salinity
III-A-2

wind rose graphs for PCS Aurora 6N for 1998 through 2015 were included in the 2015 creeks
report and quarterly and annual salinity rose graphs for each Aqua TROLL were included in the
2015-2018 reports. The salinity rose graphs showed some variability in salinity with percent
frequency of counts by wind direction but also corroborated the following presumed factors:
■ overall predominant wind direction was northeast or southwest
■ salinity at the most upstream stations was less affected than those
closer to mouths of the study creeks
■ salinity variation for the downstream stations was dependent on
orientation of creek to the wind and creek mouth proximity to
Pamlico River, or South Creek, or Durham Creek.
The wind data (direction and speed) were used in the 2020 analysis of correlation as usual but
the salinity rose graphs were not prepared for this report due to the long term consistent nature
of the predominant winds at this locale.
Two sets of correlations were used to quantify the relationship between
wind and salinity at the 31 Aqua TROLLs. First, a Pearson's r correlation was found between
daily mean salinity and daily mean wind speed (mph). The results of these correlations are
given in Table III-A2. Correlations were weakly significant for 16 Aqua TROLLs, only two of
which were negative (DKCS1 and HS1). Positive correlation values ranged from 0.110 (JS2) to
0.237 (PS1).
The second set of correlations involved daily mean salinity and daily
mean wind direction. Because wind is a circular variable (i.e., values range from 1 to 360 and
the numerical distance between 360 and 1 is the same as the distance between 1 and 2), vector
averaging (Jammalamadaka and Lund 2006) was used to obtain daily mean wind direction.
Then, the correlation between salinity and wind direction was calculated using Mardia's r
statistic (Mardia 1976, and further described in Jammalamadaka and Lund 2006). This statistic
ranges from 0 to 1, in which 0 indicates no relationship and 1 indicates a strong relationship
between the two variables (Pewsey et al. 2013). Significance of Mardia's r was determined
using a randomization test outlined in Pewsey et al. (2013).
Fourteen Aqua TROLLs exhibited a significant relationship between daily
mean salinity and daily mean wind direction in 2020 (Table III-A2). Mardia's r correlations
ranged from 0.145 (JCBS1) to 0.369 (SC1S2). Since 2015 the majority of the correlations have
been significant although all have been weak with a range of 0.134 (LOCS2 in 2016) to 0.448
(JBS1 in 2018).
Salinity within the creeks in the PCS study seems to be driven by both
freshwater inputs (e.g., Tar River discharge and rainfall) and saline inputs (e.g., east winds
move more saline water from lower in the Pamlico estuary into the study area). Of these, Tar
River discharge showed the strongest relationships with salinity: salinity decreased with greater
Tar River discharge.
There did not seem to be any apparent effect of mine activities on the
relationships between salinity and the examined environmental variables. Salinity within post -
Mod Alt L creeks (Jacks, Jacobs, Drinkwater, Tooley, Huddles Cut, Porter, and DCUT11)
showed similar relationships with all environmental variables as the control and pre -Mod Alt L
periods. Thus, it does not seem that mine activity has affected how creek salinity is altered by
external environmental factors unrelated to the mine.
III-A-3

2.0 Interannual and/or Monthly Comparisons
Annual salinity for each Aqua TROLL in the eight impact creeks, medians for pre -
and post -Mod Alt L years, and annual salinity for nearby control creeks are shown in boxplots
(Figure III-A4 through Figure III-A10). Figure III-A11 through Figure III-A13 display box plots of
monthly salinity at each impact creek Aqua TROLL with data combined into pre -Mod Alt L, post -
Mod Alt L, and 2020 year categories. Figure III-A14 and Figure III-A15 display box plots of
monthly salinity at each control creek Aqua TROLL for all years combined and for 2020. These
data and figures are discussed in more detail below for each creek.
The majority of the Aqua TROLLS in 2020 had a median salinity level that was
higher than in 2019. Exclusions included SCSI , SCS2, BSCS1, DKCS1, HS2, and LCS1.
In 2020 despite high Tar River discharge and rainfall totals, median salinity
appeared to be higher in most months when compared to previous years; however, November
and December 2020 represented lower median months for most stations.
a. Broomfield Swamp Creek and SCUT1 (control)
The second year of pre -Mod Alt L salinity and depth data was collected in
2020 from three Aqua TROLLs in Broomfield Swamp Creek at the locations shown in Figure I-
B1 and from three Aqua TROLLs in SCUT1 (control) as shown in Figure I-B2. The two creeks
are the most upstream tributaries to South Creek in the study and the Aqua TROLL locations
are the most distant from the Pamlico River in the study. Monthly and annual summary data for
2020 are shown in Table III-A1. Although less saline, these creeks followed a similar seasonal
trend to the other creeks in 2020 with increased salinity distributions in January and the summer
to early fall months (Table III-A1). Salinity in Broomfield Swamp Creek in 2020 ranged from 0.0
to 15.2 and from 0.0 to 16.6 in SCUT1. Mean salinity in Broomfield Swamp Creek at the three
monitors ranged from 1.4 to 3.8 while the SCUT1 mean ranged from 0.5 to 3.9 (Table III-A1). All
six stations across both creeks in 2019 displayed similar trends to those in 2020 as both years
represented low salinity distributions with little variation between years (Figure III-A4).
3.0 Pre -/Post- Mod Alt L Comparisons
Two statistical tests were used to determine potential effects of Mod Alt L
activities on impact creeks. Welch's t-tests determined if mean post -Mod Alt L salinity
significantly differed than mean pre -Mod Alt L salinity. Concordance correlations determined if
the similarity in salinity between impact and control creeks significantly differed after Mod -Alt L
activities. A concordance correlation is similar to a normal correlation, but examines the
similarity in values between two variables, instead of the presence of a linear trend (Lin 1989).
The concordance correlation coefficient, r, ranges from -1 to +1, in which -1 indicates no
similarity between the variables and +1 indicates the two variables are identical. For each
impact creek, two concordance correlations were conducted: one between salinity values of the
impact and control creek for all samples taken in pre -Mod Alt L years (and the corresponding
years for the control creek) and one for the post -Mod Alt L years. If Mod Alt L activities affected
salinity, the concordance correlation between impact and control should be lower for the post -
Mod Alt L years than the pre -Mod Alt L years (i.e., salinity in the impacted creek should be less
similar to the control creek). For some impacted creeks, multiple control creeks were used.
mean ± 1 SD.
Salinity measurements for pre- and post -Mod Alt L time periods are reported as
III-A-4

a. Jacks Creek
There are two Aqua TROLLs located in Jacks Creek-JS2 approximately
400 feet upstream of the mouth and JS1 approximately 2,475 feet (0.46 mile) upstream of the
mouth where the main channel divides into two narrower channels (Figure I-B3). Monitors were
re -installed 20 July 2011 after CAMA-approved pier construction was completed. The first post -
Mod Alt L year was 2015. The historic 645-acre drainage basin was reduced over many years
to approximately 317 acres pre -Mod Alt L and then to 150 acres by Mod- Alt L at the end of
2015; no further Mod -Alt L drainage basin reduction is permitted to occur in Jacks Creek. In
preparation of this report, it was noted that JS1 and JS2 salinity data from several early Jacks
Creek years (2003-2005) had not been included in pre- and post -Mod Alt L analyses for the past
several years; the data gap appears to have occurred when analysis switched from Sigma Plot
to R. Archived raw data were used where retrievable and measures taken from the Appendix C
salinity graphs or the monthly min/max/avg values taken from the tables in the 2003-2005
reports were used for periods of absent or corrupted archive data.
The JS1 and JS2 Aqua TROLLs showed a pattern of increased yearly
salinity distributions (2000-2002) before distributions decreased in 2003 (Figure III-A5). Salinity
distributions steadily decreased annually from 2011 to 2016 at both JS1 and JS2 (Figure III-A5).
Despite the third highest mean Tar River discharge and the second highest rainfall total at Jacks
Creek (Figure III -Al and Figure III-A3), the salinity distributions were higher than all post -Mod
Alt L years except 2017 (Figure III-A5). Pre -Mod Alt L monthly salinity means and medians
were generally higher than post -Mod Alt L means and medians for both Jacks Creek monitors.
Monthly salinity means and medians for 2020 were higher than all post -Mod Alt L values except
for November and December for both Aqua TROLLs and in August for JS2 (Figure III -Al 1).
Salinity at Jacks Creek was compared to both SS1 (a nearby control
Aqua TROLL in South Creek with data from 1999) and Little Creek (LCS1 and LCS2), the
nearest control creek with two monitors since 2011 (Figure III-A5). At JS1, pre -Mod Alt L
salinity (8.02 ± 4.29 psu) was significantly higher than post -Mod Alt L salinity (6.50 ± 3.61 psu; t
= 14.12, P < 0.001). At SS1, salinity levels were significantly higher during corresponding pre -
Mod Alt L years (8.79 ± 4.38 psu) and decreased during post -Mod Alt L years (7.03 ± 3.60 psu; t
= 17.09, P < 0 .001). At LCS1, salinity during corresponding pre -Mod Alt L years (5.44 ± 4.06
psu) was also significantly higher than salinity during post -Mod Alt L years (2.88 ± 3.58 psu; t =
18.80, P < 0.001).
The concordance correlation analysis between JS1 and SS1 showed that
salinity values were closely associated post -Mod Alt L (pre rr = 0.87 and post rr = 0.95). The
concordance correlation between JS1 and LCS1 showed that salinity values did not differ post -
Mod Alt L (pre rr = -0.16 and post rr = 0.47) (Table III-A2).
At JS2, pre -Mod Alt L salinity (8.62 ± 3.81 psu) was significantly higher
than post Mod Alt L salinity (7.12 ± 3.66 psu; t = 14.747, P < 0.001). Corresponding pre- and
post -Mod Alt L statistics for SS1 are given above. At LCS2, salinity during corresponding pre -
Mod Alt L years (10.18 ± 3.35 psu) was also significantly higher than salinity during post -Mod Alt
L (6.56 ± 3.73 psu; t = 29.42, P < 0.001) (Table III-A2).
The concordance correlation analysis between JS2 and SS1 showed that
salinity values were more similar post -Mod Alt L (pre rr = 0.92 and post rr = 0.98); values
between JS2 and LCS2 did not differ post -Mod Alt L (pre rr = -0.26 and post rr = 0.92) (Table III-
A2).
III-A-5

b. South Creek (SS1) (control)
The South Creek Aqua TROLL is located approximately 2,600 feet (0.5
mile) downstream from the mouth of Jacks Creek on the north bank of South Creek (Figure I-
B4). As noted for Jacks Creek, some SS1 salinity data (2004 only in this case) also had been
inadvertently omitted in the recent pre- and post -Mod Alt L analyses; the same approach was
used for SS1 data recovery as was done for Jacks Creek data (raw archived data where
possible, otherwise values from 2004 report Appendix C graphs, or min/max/avg values from
2004 report table). Salinity at SS1 was variable from the study years 1999-2002, with an
increase in salinity distributions between 2003 and 2008 and then a decrease in salinity
distributions from 2012 to 2016. Salinity distributions fluctuated in the last four years with 2020
representing another increase. In 2020, the median salinity at SS1 was slightly below the
median for 1999-2014 (includes Jacks Creek pre -Mod Alt L years. (Figure III-A5), slightly further
below the 2011-2013 median (Jacobs Creek pre -Mod Alt years. Figure III-A6), several psu
below the 2011-2012 median (Drinkwater Creek pre -Mod Alt L years; Figure III-A7), and also
below the 1999-2002/2010-2011 median (Tooley Creek pre -Mod Alt L years; Figure III-A8).
Similarly to most stations at other creeks, SS1 saw peaks in salinity distributions in January and
late summer to early fall (Figure III-A14).
c. Little Creek (control)
There are two Aqua TROLLs located in Little Creek, which drains into
South Creek from the south and has a large mostly agricultural drainage basin (- 1,822 acres).
The downstream Aqua TROLL, LCS2, is approximately 1,000 feet upstream from the mouth and
LCS1 is approximately 4,800 feet (0.9 mile) upstream from the mouth (Figure I-B5). Aqua
TROLLs were installed 22 June 2011 after CAMA-approved pier construction was completed.
The salinity distribution for 2020 was slightly lower than the distribution in
2019 at LCS1 and continued the up one year and down the next trend apparent since 2016,
while LCS2 saw its first increase since 2017 (Figure III-A5, Figure III-A6, and Figure III-A7).
Salinities at both Aqua TROLLs in 2020 saw similar trends to other creeks with peaks in
January, October, and November (Figure III-A14).
d. Jacobs Creek
There are two Aqua TROLLs in Jacobs Creek, which is located on the
northwest side of South Creek. The downstream Aqua TROLL, JCBS2, is approximately 400
feet upstream from the mouth and JCBS1 is approximately 3,306 feet (0.6 mile) upstream of the
mouth (Figure I-B6). Monitors were installed 20 July 2011 after CAMA-approved pier
construction was completed. The first post -Mod Alt L year was 2014. The historic 751-acre
drainage basin was reduced over many years to approximately 393 acres by the end of 2014;
no further drainage basin reduction from Mod Alt L is permitted in Jacobs Creek.
The salinity distribution, in 2020, at both Aqua TROLLs increased for the
first time since 2017 (Figure III-A6) even though 2020 Tar River discharge and rainfall totals at
all rain gauges were some of the highest throughout the study (Figure III -Al and Figure III-A3).
The two Aqua TROLLs have similar monthly salinity distributions over the years, with JCBS2
distributions usually slightly higher than the more variable upstream JCBS1 (Figure III -Al 1).
Excluding September, all post -Mod -Alt L monthly means and medians were lower than pre -Mod
Alt L values at both Aqua TROLLs.
III-A-6

Jacobs Creek was compared to four nearby control creeks (SS1, PA2,
Little, and Long) to determine potential post -Mod Alt L differences. At JCBS1, pre -Mod Alt L
salinity (11.18 ± 2.84 psu) was significantly higher than post -Mod Alt L salinity (7.31 ± 3.53 psu;
t = 32.45, P < 0.001). At all four control creeks, salinity during corresponding pre -Mod Alt L
years was also significantly higher than during corresponding post -Mod Alt L years (SS1 - pre:
11.81 ± 2.30 psu, post: 7.09 ± 3.43 psu, t = 48.52, P < 0 .001; PA2S1 - pre: 12.70 ± 2.11 psu,
post: 7.85 ± 3.35 psu, t = 50.23, P < 0.001; LCS1 - pre: 6.52 ± 4.10 psu, post: 2.88 ± 3.47 psu, t
= 23.70, P < 0.001; LOCS1 - pre: 12.76 ± 2.00 psu, post: 7.84 ± 3.43 psu, t = 51.45, P < 0.001).
The concordance correlation analysis showed that salinity values
between JCBS1 and the four controls (LCS1, SS1, PA2S1, and LOCS1) were similar to each
other (comparison with LCS1: pre rr = 0.25 and post rr = 0.33; comparison with SS1: pre rr =
0.73 and post rr = 0.90; comparison with PA2S1: pre rr = 0.63 and post rr = 0.94; comparison
with LOCS1: pre rr = 0.57 and post rr = 0.91) (Table III-A2).
At JCBS2, pre -Mod Alt L salinity (12.73 ± 2.27 psu) was also significantly
higher than post -Mod Alt L salinity (7.85 ± 3.39 psu, t = 48.12, P < 0.001). At all four control
creeks, salinity during corresponding pre -Mod Alt L years was also significantly higher than
salinity during corresponding post -Mod Alt L years (SS1 and PA2S1 given above; LCS2 - pre:
11.58 ± 2.70, post: 6.58 ± 3.55 psu, t = 43.99, P < 0.001; LOCS2 - pre: 12.91 ± 2.10 psu, post:
8.05 ± 3.35 psu, t = 50.27, P < 0.001) (Table III-A2).
The concordance correlation analysis showed that salinity values
between JCBS2 and all four control creeks did not significantly differ post -Mod Alt L
(comparison with PA2S1: pre rr = 0.95 and post rr = 0.98; comparison with LOCS2: pre rr = 0.95
and post rr = 0.97; comparison with LCS2: pre rr = 0.74 and post rr = 0.84; comparison with
SS1: pre rr = 0.89 and post rr = 0.94) (Table III-A2).
e. PA2 (control)
The small man-made creek and marsh PA2 drains into South Creek from
the northwest and has the smallest drainage basin (-22 acres) in the study. One Aqua TROLL
(PA2S1) is located near the middle of PA2, approximately 800 feet from the mouth; there is no
downstream salinity monitor in PA2 itself, but the downstream salinity monitor for Drinkwater
Creek is within 200 feet of the mouth of PA2 and is used for comparison/analysis (Figure I-B7).
The PA2 monitor was installed 13 July 2011 after CAMA-approved pier construction was
completed.
Salinity at this location followed similar pattern to that shown in South
Creek, Little Creek downstream, and Long Creek. The median salinity in 2020 was higher than
the median for post -years in both Jacobs and Drinkwater creeks. Like many of the creeks in the
study, the salinity distribution in 2016 at PA2S1 was the lowest in the study years thus far
(Figure III-A5, or Figure III-A6); in 2016, Tar River discharge was slightly above the mean
(Figure III-A1) and rainfall was slightly above the mean (Figure III-A3 and Table I-F2). Salinity
distributions in 2020 were very similar to those at the downstream salinity monitor for Drinkwater
Creek (DWS2) as would be expected due to proximity. In 2020, like in most other creeks,
salinity distributions were the highest in January, September, and October (Figure III-A15; Table
III-A1).
III-A-7

f. Drinkwater Creek
There are two Aqua TROLLs located in Drinkwater Creek, which is also
on the northwest side of South Creek. The downstream Aqua TROLL, DWS2, is located
approximately 400 feet upstream from the mouth and DWS1 is located approximately 3,520 feet
(0.7 mile) upstream from the mouth and just upstream of the out -of -use railroad crossing of the
creek (Figure I-B8). Monitors were installed 30 June 2011 after CAMA-approved pier
construction was completed. The first year of post -Mod Alt L data was 2013. The historic 605-
acre drainage basin was reduced over many years to approximately 152 acres by the end of
2014 and no further basin reduction from Mod Alt L is permitted.
For all years, the downstream Aqua TROLL has had higher salinity
distributions than the upstream Aqua TROLL by about 2-3 psu (Figure III-A7). Salinity
distributions for both stations in 2020 was the second highest of all post -Mod Alt L years.
Drinkwater Creek was compared to four nearby control creeks (South
Creek, PA2, Little, and Long) to determine potential post -Mod Alt L differences. At DWS1, pre -
Mod Alt L salinity (10.40 ± 3.66 psu) was significantly higher than post -Mod Alt L salinity (6.52 ±
3.74 psu, t = 22.84, P < 0.001). At all four control creeks, salinity during corresponding pre -Mod
Alt L years was also significantly higher than salinity during corresponding post -Mod Alt L years
(SS1 - pre: 12.01 ± 2.51 psu, post: 7.62 ± 3.57 psu, t = 38.49, P < 0 .001; PA2S1 - pre: 13.14 ±
2.38 psu, post: 8.37 ± 3.46 psu, t = 39.33, P < 0.001; LCS1 - pre: 7.58 ± 4.14 psu, post: 3.60 ±
3.16 psu, t = 22.72, P < 0.001; LOCS1 - pre: 13.28 ± 2.12 psu, post: 8.37± 3.54 psu, t = 43.30,
P<0.001).
The concordance correlation analysis showed that salinity values at
DWS1 was similar to the four controls (SS1, PA2S1, LCS1, and LOCS1) (comparison with SS1:
pre rr = 0.55 and post rr = 0.82; comparison with PA2S1: pre rr = 0.43 and post rr = 0.76;
comparison with LCS1: pre rr = 0.59 and post rr = 0.50; comparison with LOCS1: pre rr = 0.35
and post rr = 0.75) (Table III-A2).
At DWS2, pre -Mod Alt L salinity (13.22 ± 2.42 psu) was significantly
higher than post -Mod Alt L salinity (8.51 ± 3.51 psu, t = 38.58, P < 0.001). At all four control
creeks, salinity during corresponding pre -Mod Alt L years was also significantly higher than
salinity during corresponding post -Mod Alt L years (SS1 and PA2S1 given above; LCS2 - pre:
12.21 ± 2.84, post: 7.07 ± 3.67 psu, t = 36.45, P < 0.001; LOCS2 - pre: 13.38 ± 2.34 psu, post:
8.58 ± 3.46 psu, t = 40.07, P < 0.001) (Table III-A2).
The concordance correlation analysis showed that post -Mod Alt L salinity
values between DWS2 and all four control creeks did not significantly differ (comparison with
PA2S1: pre rr = 0.96 and post rr = 0.99; comparison with LOCS2: pre rr = 0.96 and post rr =
0.98; comparison with SS1: pre rr = 0.90 and post rr = 0.93) (Table III-A2).
g. Long Creek (control)
There are two Aqua TROLLs located in Long Creek, which is on the
southeast side of South Creek opposite Tooley Creek. The downstream Aqua TROLL, LOCS2,
is approximately 500 feet upstream from the mouth and LOCS1 is approximately 2,000 feet (0.4
mile) upstream from the mouth (Figure I-B9). The historic -630-acre basin of Long Creek was
reduced decades ago to -223 acres by previous mine reclamation activities. Units were
installed 13 July 2011 after CAMA-approved pier construction was completed.
III-A-8

Salinity distributions in Long Creek at the two monitors have followed
similar patterns over their sample years (Figures III-A5, III-A6, or III-A7). The salinity distribution
in 2020 represented the first increase in three years (Figures III-A5, III-A6, or III-A7). In 2020, at
both Aqua TROLLs, distributions were the lowest late in the year and the highest in January and
October (Figure III-A14, Table III-A1).
h. Tooley Creek
There are two Aqua TROLLs located in Tooley Creek, which is on the
northwest side of South Creek, close to where South Creek joins Pamlico River. The
downstream Aqua TROLL, TS2, is approximately 336 feet upstream of the mouth and TS1 is
approximately 2,468 feet (0.46 mile) upstream of the mouth (Figure I-B10). Monitors were re-
installed in 2010 and the first year of post -Mod Alt L data was 2012. Mod -Alt L mine activities in
the Tooley Creek drainage basin, most of which occurred in 2012, reduced the basin to
approximately 257 acres from the approximately 571-acre pre -Mod Alt L basin. Mod -Alt L
activities in Tooley Creek were completed by the end of 2013 and no further basin reductions
are permitted.
The two Aqua TROLLs have had similar salinity distributions since 1999;
both fluctuating alongside yearly discharge and rainfall (Figure III -Al and Figure III-A3). For
both Tooley Creek Aqua TROLLs, the 2020 salinity distribution was higher than either the pre -
or the post -Mod Alt L monthly values in January, March -May, July and September, while
November and December represented the two lowest months for 2020 (Figure III-Al2). The
salinity distribution for 2020 was nearly the same at both stations and slightly above the post -
Mod Alt L median (Table III -Al and Figure III-A8).
To determine potential post -Mod Alt L differences, Tooley Creek was
compared to one control creek monitor (SS1; Long Creek is a control creek closer to Tooley but
only has data from 2011 forward and therefore includes only one Tooley Creek pre -Mod Alt L
year). At TS1, pre -Mod Alt L salinity (9.53 ± 3.48 psu) was significantly higher than post -Mod
Alt L salinity (9.08 ± 3.61 psu; t = 4.20, P < 0.001). At SS1, salinity during corresponding pre -
Mod Alt L years (10.16 ± 4.36 psu) was also significantly higher than salinity during
corresponding post -Mod Alt L years (8.15 ± 3.76 psu; t = 12.19, P = < 0.001).
The concordance correlation analysis between TS1 and SS1 showed that
salinity values were more similar post -Mod Alt L (pre rr = 0.87 and post rr = 0.91) (Table III-A2).
At TS2, pre -Mod Alt L salinity (10.25 ± 3.47 psu) was significantly higher
than post -Mod Alt L salinity (9.31 ± 3.60 psu; t = 8.75, P < 0.001). Corresponding pre- and post -
Mod Alt L statistics for SS1 are given above (Table III-A2).
The concordance correlation analysis between TS2 and SS1 showed that
salinity values were more similar post -Mod Alt L (pre rr = 0.84 and post rr = 0.91) (Table III-A2).
i. Pamlico River (PS1) (control)
The Pamlico River Aqua TROLL is located on the south shore of the river
more or less equidistant (approximately 1.7 miles) between the mouths of Huddles Cut and
South Creek (Figure I-B12). Salinity distributions at PS1 declined every year from 2012-2016,
and the median salinity in 2016 was half as much as in 2012; the 2018 salinity distribution was
similar to 2001, 2007, and 2017. Due to damage from Hurricane Florence to the infrastructure
to which the Aqua TROLL was mounted, there were no salinity data for PS1 from mid-
III-A-9

September 2018 through entire 2019. Once reinstalled in 2020, salinity slightly decreased
compared to 2018 (Figure III-A9).
j. Huddles Cut
There are three Aqua TROLLs located in Huddles Cut, which is on the
south side of Pamlico River. The downstream Aqua TROLL, HS3, is at the mouth just upstream
of the bridge across SR306 at the Bayview-Aurora ferry landing, HS1 is approximately 3,020
feet (0.6 mile) upstream of the mouth on the main prong, and HS2 is approximately 1,800 feet
(0.3 mile) upstream of the mouth on the west prong (Figure I-B13). Direct mine impacts to the
Huddles Cut watershed ended in 2011 and approximately 289 acres remain out of the
approximately 552-acre pre -Mod Alt L basin. No further reductions are permitted.
Salinity distributions at HS3 have been higher than either HS1 or HS2 for
almost every year, except 2015 when HS2 had a slightly higher distribution (Figure III-A9).
Despite high rainfall and discharge in 2020, salinity distributions were high at all three Aqua
TROLLS (Figure III-A1, Figure III-A3, and Figure III-A9). HS1 and HS2 represent two of the five
Aqua TROLLs in which salinity distributions increased in post -Mod Alt L years. In 2020, all three
monitors recorded the highest salinity distributions in the summer and the lowest in spring
(Figure III-A13 and Table III-A1).
To determine potential Mod Alt L differences, Huddles Cut was compared
to the Pamlico River monitor (PS1) for all years and the Duck Creek monitors (DKCS1 and
DKCS2) only for the years 2011-2020 (Figure III-A9). PS1 has been monitored since 1999,
almost the same amount of years as Huddles; however, PS1 was not monitored during the year
of 2019 due to infrastructure damage from Hurricane Florence. There were no pre -Mod Alt L
monitoring years coinciding with Huddles Cut from Duck Creek so these monitors were used to
reference patterns in upstream and downstream Huddles monitors from 2011-2020. Both sets
of upstream and downstream monitors from Duck Creek and Huddles Cut followed a similar but
not identical fluctuating trend throughout matching sample years (Figure III-A9). At HS1, pre -
Mod Alt L salinity (4.89 ± 3.43 psu) was significantly lower than post -Mod Alt L salinity (6.31 ±
3.60 psu; t = -15.611, P < 0.001). At PS1, salinity was significantly higher during corresponding
pre -Mod Alt L years (11.39 ± 6.17 psu) and decreased by 17 percent during post -Mod Alt L
years (9.48 ± 3.66 psu; t = 13.17, P < 0.001).
The concordance correlation analysis between HS1 and PS1 showed that
salinity values were more similar post -Mod Alt L (pre rr = 0.05 and post rr = 0.50) (Table III-A2).
At HS2, pre -Mod Alt L salinity (4.87 ± 3.63 psu) was significantly lower
than post -Mod Alt L salinity (6.57 ± 3.48 psu; t = -19.234, P < 0.001). Corresponding pre- and
post -Mod Alt L statistics for PS1 are given above (Table III-A2).
Like HS1, the concordance correlation analysis between HS2 and PS1
showed that salinity values were more similar post -Mod Alt L (pre rr = 0.07 and post rr = 0.52)
(Table III-A2).
At HS3, pre -Mod Alt L salinity (8.62 ± 4.47 psu) was significantly higher
than post -Mod Alt L salinity (8.22 ± 3.67 psu; t = 3.23, P = 0.001). Corresponding pre- and post -
Mod Alt L statistics for PS1 are given above (Table III-A2).
The concordance correlation analysis between HS3 and PS1 showed that
salinity values were more similar post -Mod Alt L (pre rr = 0.36 and post rr = 0.87) (Table III-A2).
III-A-10

k. Porter Creek
There are two Aqua TROLLs located in Porter Creek, which is on the
south side of Pamlico River upstream from Aurora and the PCS main facility. The downstream
Aqua TROLL, PCS2, is approximately 2,377 feet (0.45 mile) upstream from the mouth, at the
outer edge of a broad expanse of the creek and PCS1 is approximately 8,600 feet (1.6 miles)
upstream from the mouth in the middle of a narrow, deep channel (Figure I-B14). Monitors were
installed 19 and 27 July 2011. The first year of post -Mod Alt L data was 2016. The estimated
historic 3,745-acre drainage basin was reduced to approximately 1,154 acres by the end of
2016 (reductions include all mine activities, not only Mod Alt L). In 2017 the 1,154-acre Porter
Creek drainage basin was reduced by an additional 314 acres to roughly 840 acres; no
additional reduction is permitted.
While overall behavior was the same, the salinity distribution at PCS2
was higher than PCS1 every year. Both Aqua TROLLs experienced a steady decline in their
distributions from 2011-2015 then fluctuating in the years after (Figure III-A10). 2020 represents
the first increase in median yearly salinity since 2017 and was above the post -Mod Alt L median
(Figure III-A10). In 2020, like other creeks, PCS1 and PCS2 showed the highest salinity
distribution in January, August, and September while November and December represented the
lowest months (Figure III-A13 and Table III-A1).
Salinity at Porter Creek was compared to nearby control Aqua TROLLs,
one in Durham Creek (DCS1), one in a small tributary DCUT19 (DC19S1), and two in Duck
Creek (DKCS1 and DKCS2), a control creek across the Pamlico River. At PCS1, pre -Mod Alt L
salinity (6.28 ± 4.06 psu) was significantly higher than post -Mod Alt L salinity (5.62 ± 3.79 psu; t
= 4.91, P < 0.001). At DC19S1, salinity during corresponding pre -Mod Alt L years (6.34 ± 3.28
psu) was not significantly different than salinity during corresponding post -Mod Alt L years (6.10
± 3.80 psu; t = 1.88, P = 0.061). At DKCS1, salinity during corresponding pre -Mod Alt L years
(5.48 ± 4.25 psu) was significantly higher than salinity during corresponding post -Mod Alt L
years (4.06 ± 3.58 psu; t = 13.876, P < 0.001) (Table III-A2).
The concordance correlation analysis between PCS1 and DC19S1
showed that salinity values were more closely associated post -Mod Alt L (pre rr = 0.75 and post
rr = 0.94). This was not the case between PCS1 and DKCS1, which showed that salinity values
were statistically less similar post -Mod Alt L (pre rr = 0.70 and post rr = 0.67) (Table III-A2).
At PCS2, pre -Mod Alt L salinity (8.06 ± 3.73 psu) was significantly higher
than post -Mod Alt I salinity (6.49 ± 3.69 psu; t = 12.43, P < 0.001). At DCS1, salinity during
corresponding pre -Mod Alt L years (6.39 ± 3.24 psu) was also significantly higher than salinity
during post -Mod Alt L years (5.99 ± 3.80 psu; t = 2.98, P = 0.002). At DKCS2, salinity during
corresponding pre -Mod Alt L years (8.02 ± 4.25 psu) was also significantly higher than salinity
during post -Mod Alt L years (5.89 ± 3.93 psu; t = 12.23, P < 0.001) (Table III-A2).
The concordance correlation analysis between PCS2 and DCS1 showed
that salinity values were more closely associated post -Mod Alt L (pre rr = 0.95 and post rr =
0.96). The concordance correlation between PCS2 and DKCS2 showed that salinity values did
not statistically differ post -Mod Alt L (pre rr = 0.71 and post rr = 0.87) (Table III-A2).
III-A-11

I. Durham Creek DCS1 (control), DCUT11, and DCUT19 (control)
In 2013, three Aqua TROLLs were installed in Durham Creek basin, one
approximately 13,717 feet (2.6 miles) upstream of the mouth of Durham Creek (DCS1, a
control, Figure I-B15) one in an unnamed tributary to Durham Creek on its east side,
approximately 365 feet upstream from where it drains into Durham Creek, DCUT11 (DC11S1,
Figure I-B16), and one in a second unnamed tributary to Durham Creek on its west side
approximately 341 feet upstream from where it drains into Durham Creek, DCUT19 (DC19S1, a
control, Figure I-B17). Durham Creek drains into the south side of Pamlico River. The entire
Durham Creek drainage basin totals 31,051 acres and encompasses three of the study creeks:
Porter Creek (2,728 acres), DCUT11 (166 acres), and DCUT19 (119 acres). The first post -Mod
Alt L year for DCUT11 was 2018 when 84.50 acres were removed from the basin; no further
impacts to DCUT11 are permitted.
All three Aqua TROLLs had very similar salinity distributions in 2020 as
well as in the other years with a few slight exceptions (Figure III-A10). Salinity distributions at all
three Aqua TROLLs decreased since 2013, particularly between 2013 and 2014, until a rise in
2017 followed by another downwards trend in the next two years until it increased again in 2020
(Figure III-A10). In 2020, the salinity distribution at DC11S1 was greater than the post -Mod Alt
L distributions in the first seven months of the year and lower in the last five (Figure III-A13;
Table III-A1). DCS1 and DC19S1 salinity distributions in January and September 2020
represented the two highest months at both stations (Figure III-A15 and Table III-A1).
Salinity at DC11S1 was compared to two nearby control Aqua TROLLs,
DCS1 and DC19S1. At DC11S1, pre -Mod Alt L salinity (5.22 ± 3.59 psu) was not significantly
different than post -Mod Alt L salinity (5.40 ± 3.97 psu; t = -1.23, P = 0.218). At DC19S1, salinity
during corresponding pre -Mod Alt L years (5.97 ± 3.49 psu) was significantly lower than salinity
during corresponding post -Mod Alt L years (6.37 ± 3.81 psu; t = -2.85, P = 0.004). At DCS1,
salinity during corresponding pre -Mod Alt L years (6.06 ± 3.43 psu) was not significantly
different than salinity during post -Mod Alt L years (6.28 ± 3.90 psu; t = -1.58, P = 0.115).
The concordance correlation analysis between DC11S1 and DCS1
showed that salinity values were less similar post -Mod Alt L (pre rr = 0.94 and post rr = 0.92).
The concordance correlation between DC11S1 and DC19S1 showed that salinity values were
as similar in post -Mod Alt L years as they were in pre -Mod Alt L (pre rr = 0.90 and post rr = 0.90)
(Table III-A2).
m. Duck Creek (control)
There are two Aqua TROLLs in Duck Creek, a tributary to Pamlico River
with a confluence on the north side west of the town of Bath. The downstream Aqua TROLL,
DKS2, is approximately 1,400 feet (0.27 mile) upstream from the mouth and DKS1 is
approximately 7,780 feet (1.5 miles) upstream from the mouth (Figure I-B18). Much of the
—3,118 acre drainage basin is either in agriculture or silviculture. Aqua TROLLs were installed
26 July 2011 after the CAMA-approved pier construction was completed.
Salinity distributions declined since 2011 at DKCS1 and since 2012 at
DKCS2 until 2017 when salinity increased at both Aqua TROLLs and has fluctuated in years
since (Figure III-A9 or III-A10). With the exception of 2011 when it was slightly higher, median
salinity at DKCS1 was always lower than at DKCS2; in most years it was 2-4 psu lower. In
2020, the salinity median at both Aqua TROLLs was near the median of the 10 sample years
(Figure III-A9 and III-A10).
III-A-12

4.0 Summary and Conclusions
Salinity steadily declined in the Pamlico River and tributaries from 2011-2016 (or
2012-2016 for some creeks) until a rise in salinity in 2017 and fluctuation in years after. This
short-term trend has been observed at every Aqua TROLL regardless of location (upstream or
downstream), creek basin (within South, Durham, Porter, or Duck creeks or Pamlico River
drainages), creek type (impact or control), or size of drainage basin. In 2020, most Aqua
TROLLS saw an increase in salinity levels, except for LCS1, HS2, both Duck Aqua TROLLs,
and all three SCUT1 Aqua TROLLs.
Correlations with environmental factors showed that salinity was most strongly
associated with Tar River discharge; each Aqua TROLL showed a weak to strong negative
correlation between salinity and discharge. Water depth had weak positive correlations in nine
of the 31 Aqua TROLLs and four weak negative correlations. Rainfall and wind speed were
only weakly correlated with salinity; every station except one (HS1) had a significant negative
correlation with rainfall and 16 were significant for wind speed with only two of those being
negatively correlated. Thus, salinity within the study creeks declined mostly with increased Tar
River discharge and, to a lesser extent, increased rainfall and decreased wind speed, while
increased water depths tended to increase salinity. Tar River discharge has fluctuated
throughout sample years with 2020 representing one of the highest years recorded (Figure III -
Al ). Total rainfall for across all sample years has also seen fluctuations with 2020 representing
the highest total ever recorded for almost every creek (Figure III-A3).
Since annual salinity distributions declined at all creeks for roughly five years
until all creeks increased in 2017 followed by a decrease in 2019 and another increase in 2020,
there is little evidence to suggest that Mod Alt L activities have impacted creek salinity. Instead,
salinity seems to be more influenced by regional and climatic factors such as Tar River
discharge, precipitation, and wind. These large-scale factors, as well as others such as drought
(which is a function of precipitation), are known to be important determinants of the salinity of
the Tar/Pamlico estuary (Stanley and Nixon 1992, Kimmel 2016).
Because salinity has decreased in every Aqua TROLL since 2011 or 2012 with a
respectively high salinity year only in 2017 and 2020, it is not surprising that the creeks recently
impacted by Mod Alt L activity (Jacks, Drinkwater, Jacobs, Porter, and Tooley) had lower
salinities during post -Mod Alt L than pre -Mod Alt L. The control creeks for those five impact
creeks also showed the same pattern in their corresponding pre- and post -Mod Alt L years.
The concordance correlations showed that the similarity in salinity between
impact and control creeks was not greatly affected by Mod Alt L activities. If Mod Alt L activities
did affect salinity, it would be expected that salinity in impact and control creeks would become
less similar after a creek is considered post -Mod Alt L; however, salinity at most of the Aqua
TROLLs in impact creeks became more similar to salinity in control creeks after Mod Alt L
activities. For those Aqua TROLLs that became less similar, the concordance correlations were
still very high and significantly greater than zero, indicators of a sustained, close resemblance to
the control creeks.
Therefore, the similar salinities observed at all Aqua TROLLs regardless of mine impact, the t-
tests which showed significant declines in salinity at most impact and control creeks, and the
concordance correlations which showed that salinity between impact and control creeks was
similar even post -Mod Alt L, all suggest that mine activities have had little impact on study creek
salinity.
III-A-13

coS
®ooco
(Tomo F
1
f
F
F
7o F
®F
1-
F
F
F
GYF
csml-
H
1
l
1
00004 004
(mess 6o1 uo) oast} oigns 'Amu IIie❑
1
III-A-14

me ®e
mom
m
mew
m
mono oomoam -
mo mom*
moos oomeml,
mo �F
caM-
me�
am000 F
coo �rt
moos oemeell.-
1-
a
i 1 1
a
a
ad-
aoF
F
��y�y
I��
�
i
m0000 I- --I
m0 m 0 F IIILj 11=J-
F--4=1=-
H L��
mn co o
me F
meee eemomF
1
a
a
00001
001
(ale.Js 60l uo) uosig oigno 'mol j AIlea
- lsod L L1 OO
- aJd Llama
- lsod JalJod
— aid J81 0d
- lsod saIPPnH
Did saIPPnH
- lsod RaIOal
aJd +Col0al
- 1S0d Sg00er
Did scion('
lsod Jelem)Iu!Ja
2Jd Jalenqu!Q
lsod sef
aJd sloes
- aid W00J8
— OZOZ
- 610Z
- 8LOZ
— LLOZ W
- 910Z
- SLOZ
- PLOZ
- £LOZ
- ZLOZ
- LLOZ
010Z
600Z
000e
LOOZ
900Z
900Z
r00Z
£00Z
Z00Z
LOOZ
000Z
6661
8661
N
a)
L.L
Ln
0)
cn
0
0
c
a)
E
a)
c
D
a)
0
D
c
ci3
D
a)
cn
a)
X
O
III-A-15

- Pre
- Po91
Jacks
iwai Annual Rainfall fin)
0 10 30 50 70
Drinkwater
1999 2001 2003 2005 2013 2015 2017 2019 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Jacobs
2011 2012 2013 2014 2015 2016 2017 2016 2019 2020
Huddles
1 1 1
Tooley
1999 2001 2010 2012 2014 2016 2016 2020
Porter
1999 2001 2007 2009 201 2013 2018 2017 2019 2011 2012 2013 2014 2015 2018 2017 2018 2019 2020
DCUT11
1 i 1 1
2013 201a 2015 .201E 2011 2010 2010
Figure III-A3. Total annual rainfall at four long-term rain gauges over the years of the PCS
creeks study (early years used the PCS NOAA Station Aurora 6N for all creeks; rain gauges in
each creek were installed in 2011; Jacks, Drinkwater, and Jacobs creeks share the PA2 rain
gauge; Porter and DCUT11 share a rain gauge) ). If a particular site was missing data, data
from the closest rain gauge was used (including Aurora 6N). Black line is mean pre -Mod Alt L
total annual rainfall, violet is mean post -Mod Alt L total annual rainfall.
III-A-16

0
- Pre
- Freer
1
2019
2020
5C51
20'0
2020
65C52
2019
2020
0
0
BSCS3
2019
2020
Figure III-A4. Salinity boxplots for Broomfield Swamp Creek stations (BSCS1, BSCS2 and
BSCS3) compared to SCUT1 stations (SCSI, SCS2, and SCS3) on annual basis with both
years representing Broomfield Swamp Creek pre -Mod Alt L years. Solid lines indicate median
salinity for pre -Mod Alt L years (black). (Dots=5t"/95t" outliers, upper/lower whiskers=90t"/10t"
percentiles, upper/lower box edges=75t"/25t" percentiles, solid line in boxes=median).
III-A-17

• F
F
T
J
n�
lei
DE 9Z OZ 5L PI 5 0 9E 5Z DE 91 OL
(nsdl ZI Jes Owl) AW!IlS
1-
I
0 0E 9E 9E 51 0L 5 0
00
-F
T
I
a
-R
4 -I -�• -
[__Fp-
4 i A
— - l5
OP SZ OZ SL 01 S 0 0£ SZ 0Z S1 01 S. 0
(nsd) ADUIIes (nsd) Wuneg
(nsd) limes
00 SZ 0Z 91 0L S 0
Once .ulle$
2012 2013 2010
_1
_A
A
A
_A
y-+
O L :I)
co a)
cLE, 2
as °,O
E "=
O C
02 co
a)' o c�
•c c
J c E
- .s II
O - N
2 a) X
0 o 92 0
O - - a)
a) U
co •—
L
N U 7
a) to a) O
0 0
Rio
Q a)
0
Q H
� O
L
Z
O- . O
i
€ a)
aco .O CL
LL E 5 Q
III-A-18

JCB51
M. M 2 MU M. Mt M00 MI! MO MO NM
W�+ mn N* W. 20* 29* W, 20 0 M4 SOW
Figure III-A6. Salinity box plots for Jacobs Creek stations compared to Long Creek, Little Creek,
South Creek, and PA2 stations on annual basis with control creek data arranged to match pre -
and post -Mod Alt L years for Jacobs Creek. Solid lines indicate median salinity for pre -Mod Alt L
years (black) and post -Mod Alt L years (violet). NOTE: SS1 graph includes years when Jacobs
Creek was not monitored. (Dots=5t"/95t" outliers, upper/lower whiskers=90t"/10t" percentiles,
upper/lower box edges=75t"/25t" percentiles, solid line in boxes=median).
III-A-19

2011 2c12 2013 2014 0010 2016 2017 1010 2019 2020
2011 2012 2312 M. 2O13 1010 2011 2011 1010 2010
2p11 2312 201E M. 201E 201E 1011 2011 2013 1020
PA291
2011 2313 7013 2014 2303 2016 2017 2313 2010 2020
Figure III-A7. Salinity box plots for Drinkwater Creek stations compared to Long Creek, Little Creek,
South Creek, and PA2 stations on annual basis with control creek data arranged to match pre- and
post -Mod Alt L years for Drinkwater Creek. Solid lines indicate median salinity for pre -Mod Alt L
years (black) and post -Mod Alt L years (violet). NOTE: SS1 graph includes years when Drinkwater
Creek was not monitored. (Dots=5t"/95th outliers, upper/lower whiskers=90t"/10t" percentiles,
upper/lower box edges=75t"/25t" percentiles, solid line in boxes=median).
III-A-20

Lill
! I
NEI
F - 1 i A
QE St OL sl OI s o
ass,11 Ames
4
1
OE SL OL Si 06
osvtll wain
s o
— 1 1 1 �—
F +»
sz o¢ s Os
Insdy+lueeeg
----F
F 1 I F i
•
ve sz or sk Ok
medy suyeg
F
4
1
I
1
I
III
--+
OE 42 OL SS 0, S 0
Insn] q.o.ieg
control creek
A
-a
Y
U
c0
'
L
c EL)>
T N
N J =
N 4—, 0
Q
as O 0)
c ) 11
Q ct
c L O
O O 0
c
O
CO 0 C
N c CO
- c 0 "a
a) E a)
U o
c cn
c°)Co
- w U
c
O a)
-YcU=
cD= >.7
U 7 N
O
�W o fn
O N
O O N
a)
_ �U TLQ
co a)
-; > EoN
O
-
U L N in
N
>,
aii
a�
c"Jn c N
NJ- O
-< D).
>,ow
o ( L
1- N ,
0_
o 0 Q-
z
.(1;
O
,a_.) a
O O-•O
U 5
a)
Q
73 0
0
Q cs)Q
u
- O L
a2Y
O N N
ILa Q -
III-A-21

R
R
R
—f L L! +-
OD
i
of sE OE St Oi
mw4 Wanes
10-H
-+ 11-i
F
-L
VO Oi
(d! Woin
4 7
i
- li
4
i 1l
-ED
IIi
SL
OE St Pt
(nsd)Ry�pe$
DS se
(ce) 1uuo.5
s
SL OL St OL
ice) kom5
4
4
..1.
F
ss Da St Dw
044 Aimee
Q U
0
o 92 c
L
Co N u_ 0
Co -
CO Q- c 0
0 -0 CO
c .0
a) Co c/)
a) �c a.
U 0 = C
CO
O E 0
0 L
L
0 L o_
0 CO a)
_c a) D)
b
J E CV
N CO =
. < lf)
o
co 2 0 0)
c 2
c o_ N a)
cts
0 o c 0
N .-
c c o
o
cn c -0
-)
a)
a) .�
U E aoi
as 0 U OU ()vi
o 'a c
o ,� a) a)
2� 2
a) a) J a)
L c J D-
O- -0 0 =
—Qo
0 0
o U) Co
C = c7O)
o 0 Q u
-a C
=�
N oL y—
a) 0 L
CO a)
L p
_ as L
a)
L >1 >1 0
o "= Q
=
0Q�
x00)L
0 2
En
N
c 0 0 a)
Z
ctl =
wC o
cts
aY
ai a) p :
L
Q.5
a)�u
9 a)
aE >, o
IL °.
upper/lower
III-A-22

11]
a
0)
Pre PCS1
— Post
T
1 t - 1 1 1 1
1 I I 1 1 I
2011 2013 2015 2017 2019
PCS2
T
1 1
1 1 I f I
2011 2013 2015 2017 2019
DC11 S1
T
1
T
1
1
2013 2015 2017 2019
a
m
DKCS1
to
1
1
DC19S1
1 1
I I I I 1 I 1 1 1 1 I 1 I I 1 i i
2011 2013 2015 2017 2019 2013 2015 2017 2019
DKCS2
2011 2013 2015 2017 2019
DC19S1
T
1 1 1 1 1 1 1 1
2013 2015 2017 2019
A
DCS1
2013 2015 2017 2019
Figure III-A10. Salinity boxplots for Porter Creek stations (PCS1 and PCS2) and the DCUT11 station
(DC11S1) compared to Durham Creek stations (DC19S1 and DCS1 for DCUT11) and Duck Creek
stations (DKCS1 and DKCS2 for Porter Creek) on annual basis with control creek data arranged to
match pre- and post -Mod Alt L years. Solid lines indicate median salinity for pre -Mod Alt L years (black)
and post -Mod Alt L years (violet). (Dots=5th/95th outliers, upper/lower whiskers=90th/10th percentiles,
upper/lower box edges=75th/25th percentiles, solid line in boxes=median).
III -A-23

Salinity (PSU)
Salinity (PSU)
Jacks Creek - JS1
Jacobs Creek - JCBS1
0
0 -
0
Jan
Feb
Mar
T
I
1
I
I 1
Apr
1
4
D
May
June
i
July
14
8
R1
I
Aug
T
1
Sep
Oct
Nov
a
e
r
Dec
IT
9 1
1 1
Drinkwater Creek - DWS1
0_
D
IL1
0 -
Jan
Feb
T
frl I
I
1
r
1
T
Mar
8 8
0 0
T
I T
I I
I I
I 1
1 J.
I
1
Apr
j
1
May
1
June
T
J.
T
J.
July
T
■
Aug
i
1
Sep
T
r
1 B
Ir
Oct
Nov
Dec
Salinity (PSU)
Salinity (PSU)
Jacks Creek - JS2
Jacobs Creek - JCBS2
0
0
0_
4i -
Jan
T T.
g1
1 1
I
1
6
i
Feb
Mar
Apr
T
4
1 T
1
T
May
June
July
r s
1
II
1 I
1 I
T
Aug
T
1
Sep
Oct
1
1
1
Nov
I
1 1
Dec
0 —
Drinkwater Creek - DWS2
M —
0
N
0
0 —
Jan
Feb
Q
7
Mar
T
Apr
May
I I
June
T T
I
I I
July
I
1
i
Aug
T
1
1
I
1
Sep
Oct
T T
I
B
111
1
0
Nov
T
L41I
1
1
T
Dec
Figure III-A11. Monthly salinities at non -control Aqua TROLLs: Jacks Creek, Jacobs Creek, and Drinkwater Creek. (Dots=5t"/95t" outliers, upper/lower whiskers=90t"/10t" percentiles, upper/lower box
edges=75th/25th percentiles, solid line in boxes=median).
III-A-24

Tooley Creek - TS1
Salinity (PSU)
Tooley Creek - TS2
Salinity (PSU)
Figure III-Al2. Monthly salinities at non -control Aqua TROLLs: Tooley Creek. (Dots=5th/95th
outliers, upper/lower whiskers=90th/10th percentiles, upper/lower box edges=75th/25th
percentiles, solid line in boxes=median, dotted line=mean).
III-A-25

N
a
cts
Salinity (PSU)
In
N
r
Huddles Cut - HS1
Jan
PRE
POST
2020
Ti
-
Feb
Mar
r
1
1
1 I
I
I I
1
1
1
Apr
May
jII
June
1 I
I I '-
1 J. 0
July
aI
1
I I
I 1
1 1
Aug
Huddles Cut - HS2
I
T
1^
L
Sep
Oct
8
1
0
t
a
0
Nov
0
1
Dec
T
MEP
Jan
Feb
r
Mar
0
T
I
I T
I
1
Apr
T.
May
0
T
June
T
1
r
Aug
I
I
1
1
Sep
1 1
I 1
1
,
1
1
Oct
g T
1
T
Nov
7
1 1
1 1
I I
I
I
T
tI 1
1 1
r
Dec
L
Huddles Cut - HS3
Salinity (PSU)
Salinity (PSU)
447
Porter Creek - PCS1
1
ID -
In
Ir1
c
c
LL7
Jan
r
T
T
1
T
Feb
i
T
T
Mar
T
Apr
T
May
T
June
T
1
1
T
J.
July
Aug
T
T
I 0
1 •
Sep
Oct
T
Nov
Dec
T
T
T
Porter Creek - PCS2
Jan
r
T
T
,
Feb
Mar
Apr
T
1
•
;�
May
1
1 1
T
11
June
T.
T
July
T
1
1
Aug
Sep
Oct
Nov
T
T
T
Dec
DCUT11 - DC11S1
Jan
T
1
Feb
T
Mar
Apr
T.
1
May
r
I
T T
1
June
0
T
1T
I
I
1 1 T
July
a
11
B
1
Aug
• •
•
T I
a
1
1
T
1
•
Sep
•
n8
O
T
1
1
1
1
Oct
T
1
Nov
T
I
Dec
Figure III-A13. Monthly salinities at non -control Aqua TROLLs: Huddles Cut, Porter Creek, and DCUT11 (DC11S1). (Dots=5t"/95t" outliers, upper/lower whiskers=90t"/10t" percentiles, upper/lower box
edges=75t"/25t" percentiles, solid line in boxes=median).
III-A-26

Jan
ALL
2020
i
4l -
T
Feb
Mar
Apr
Little Creek - LCS1
May
T,
June
Go
July
1
South Creek - SS1
Aug
Long Creek - LOCS1
8
Sep
T
•
T
Oct
Nov
Dec
i
1
ry
0
a
Io
Jan
Feb
T
L.
Mar
1
Apr
T
Little Creek - LCS2
May
June
July
8
T
Aug
Sep
T
1
Oct
Nov
Dec
PA2 - PA2S1
Long Creek - LOCS2
Figure III-A14. Monthly salinities at control Aqua TROLLs: Little Creek, PA2, Long Creek, and South Creek (SS1) (Dots=5th&95th outliers, upper/lower whiskers=90th&10th percentiles,
upper/lower box edges=75th&25th percentiles, solid line in boxes=median).
III-A-27

Duck Creek - DKCS1
Jan
2oza
Salinity (PSU)
Feb
1 ■
I
I
Mar
T
Apr
1.
11
May
June
July
Aug
Sep
Oct
Nov
T
Dec
T
Durham Creek - DCS1
Pamlico River - PS1
0)
a
co
Salinity (PSU)
c,
O
N
c
Duck Creek - DKCS2
Jan
Feb
Mar
Apr
-r
May
1-
June
Aug
Sep
July
r
Oct
8
T
-8-
Nov
T
1
T
Dec
DCUT19 - DC19S1
Figure III-A15. Monthly salinities at control Aqua TROLLs: Duck Creek, Durham Creek, DCUT19, and Pamlico River (PS1) (Dots=5t"&95t" outliers, upper/lower whiskers=90t"&10t" percentiles,
upper/lower box edges=75"&25t" percentiles, solid line in boxes=median).
III-A-28

I
O J J
u a)Uct
J 2 0 H
OU
o U Q
as J 0 s
U
^ ( ca
QU).o a)
L
O (1) -
NO
N 0 o -
c T N
(/) o
_ -
a c
(B C O
NniCE
> < ^�
C • 0
>'U) Y
iI
Yas
E
c U L O
E L o a)
a)
= (j
E c U
•
c 0
E
a) i1i
c U s
as 00 U :�
a) 0 L co
'-
U
^ E
U U) _ O c°
(/) L )
r OU = (D as
Q ca U a)
- v) (1) a)
_a)-0a) _0
(U0 a)
C2
0
V
a)
0
Cam.)
0
n
N
7
T
(C
L
0.
a
aS
a)
L
as
m
C. o 0
0 0
(0
M C7 C7
0 0 0
co co v
o
CO N D 10
N co. C7 Q
o c�
N D W
tic0vrn
(io.4
N7 co N
C7 CV D
N N co
N 1� C7
a 4 (3
((7 N- N CO
cO
4aio°
(4 x
a) as .c
EEE
co
(0
M N v
C N: c;
ti -4 co
ai o .4
a) a) v
ni-4oa)
co 10 N
c co c7 M
CO 6
N co co W
nitiv�
cotiv�
aio?
N�?om
cri
(C)
N LO D V
onion
rn v v
LO
d7 M M CO
(f7 N 6 a)
CV 1O. M c)
a)La. v�
o
7 7 CO
CO La.
N C7
co co (f)
vv
CO (f)
v�v(-v
ti o
(f) [p a) co
1-: "
N
a) m
c .c
a)
EEE
M
0)
0
(1)
m
(n
.n.n ou7
o ti o ca
ti Lo CO
r` o
v v ((7
ti o
co. D. D. E
N C7 CO
C7 c C7 .4
F a) N CO
d N v
(4 x c
E EE
0)
L)
0)
(n
1� 6. L7 CO
0 0 o v
(f)
N CO - (0
coa)vv
D N C) CO
id c7 °M°
(f7 10
co a7 -
o(iv�
C
(4 x
E E
E
m
rn.ou7
(`7 co 6 00
(0
CO a) v co
CV CO v
N:
(f) CO
CO v
(f) 1� C
N CO v
N 7 c0 CO
C7 (f) p
a) c m .O
a)
E E E c
0)
L)
0)
Cr)
0„;co6 00,
(0
a7 ( ) M
co o
r ti
O co
r �
v � co. 2
co 0.)ci
M M — v
47 a7 N C4
co M (()
co co o 4
c° c° N cr.?
CO c° -
N
co -
N: d7 N
a) c m .O
a)
E E E c
0)
m
1� . N M
c •• co
Lo
N LO 10 N
id
co
ti CV
idai��
CV D. CO CO
C7 N ti
rn
N 4a) 00
N M CO
co. NtiL2
N (f) CO v
CO a)
f CO N CO
N CV c° 8
c° a) CO -4
CO c° E
co ai co
N
W N •• Lo
a) m .c
a)
EEE
0)
III-A-29

N
N
7
Z
0
Q.
as
cs)
a
a
LL
Table III -Al (continued).
10
CO
co (0 - 6
C 1 CO M
( C) CO
cvnocno
( n a)
N C7 n
L7 N co
coC!00
N 4 ti co
c° o (0
n C'
n m E
co 0)M
( n C O
CQ co co
( n o cv a`n,
I� ai M
rn
co ai co 4
( n n LC) .171)
co a) .n
( n M n (C)
I` CV d
(C) (0
4 CO O tV
C3 C
N N ._
EEE C
a) LC) N CO
c0 co - 6
M
CO 0) N C)
u� n��
CO 6. (0
N N
Cl CO LC)
a) ti M V
CC] N P.-
CO C L7 ca
N n co O
In OD O
co Ln N LC]
N- N C7Lo
LC] n (n
co ai
N n
CO c
ti N- ( N
4 CC] o
a
EEE
u)
m
v
'O oCO
a) CD CO
LC)
V C) a) N
co a) c
v o 0
C7 ('7 cfl
C7 N c()
N CO CO C]
(6 4 6 co
CO 7
coincoin
a)�n�
a) Co Co C)
co CO .n
.n 4 co in
o rn 10
N CO io
a)a)C❑F
LC)
cc; oo0)
co m a) co
o cv • .n
co in CO
V
CO
al N LO
(n
CO CO V
(.6 CO
co CO cv Co
• inoa)
n CO. n
Crj C . CV
N CO
a) n Lo
10
(
rn rn n 7
oCD. coc92
• (n (fl
CO N
cc o (ri
CO 7 10
oz;o 10
(-U 1` in 09
ai ai co f
ai• oo a)
CO N
coa)a)
4 in
ti CO
(n
I� - C7 n
no1-CO
co d
in N co CO
co c) o
CO (0• n L7
L7 N co
CO CO 6
( 4 • op
co o
▪ 4 v
n o
n Cr)
a) Co r M
ri r� c
co CO co co
coM6
( n co Co
(0 O) M
in CO
co co • n
x
M n ic)
CV 6
xr-
C7 (C] "Zr
N C
E E E C
LC"4-
`Cr
al Lfl •• CO
( C)
N. CO V CC]
co ai cri
in
m o co
CO CO CC] co
C7 ai 4
(S) Cr). cO (Ca
Ln Ln a)
N 1C) 92
ai c)
in
co co
L) . CO CO
o 2ic
CU CO CO c,
6) 6) (O t
Cr). r Co
aioma)
LC) Q N
co CO N
(n C▪ . CO
cri - o co_
CC] CO C) N
Cn0)oo
N CO 6
co N N (0
c
coc)oco
N C7 N.
N Cr 6 V
Cfl co. CC)
CO D
p CO
N n 6
co
N n t
0) co - co
• oal
( n M N (Q
CO 4 c7 V
N c
EEE
CO
U
J
C7 . N ate)
co LO o 0
10
CO CO N Co
r CO
N m
C7 LC • Co
V
•
Cr Ca. o CO
••In
V
CO Ln CO N-
CO N "
rn .n n m
LCDCOo1
N CO co n
m co o
I. CA CA C7
n a) CV
�ti ] M
n a) C [n1
a] N Cy CO
CO LO • M
(4 ( c
EEE
(C)
CO V c
CO CO
C7 a• a)
C) N CA a)
ci 4 o co
CJ 7 N (C)
a)1n�
rn• cvn6��
o m (n
o 4 c47-
CO a) M
ocoa)cr,O.
6) a) cc 4
io (n LO M
aio(ri�
N a. CO co
(C7 0�0
LM iz
6iC.C) 'CO
( C)
V n C) N
(.6 CO M
CO a) 1- C)
o
a) N (n (C)
• N
CO N
CO f
CO
(0
rn▪ 4cc;
CO ▪ (0 92
(n n
C o in T
c000(o
co 0 0
N- .(in
aiocoa)
CO 7 C)
Co N ( j
III-A-30

0
0
U
0
sz
a)
N
a
a
L
a)
LL
Table III -Al (continued).
6 Co co
Ln
I— p N co
(0 co
Co . D
p p • Co
C• S) 4
N
•
. c) N
CD Ca 6 M
u7 co 'CO
a) ii
o u7 LL7
ti co 4 c4
o N CO 6
p N a p
p 6 CO c4
co . Lo
a CO N
u7 tir--
6 N • 111
Lc? ti p
cF co 6) co
C
N x
a) m .c
E E E c
C0
co co co Lf)
co 0) N-
00 c0 C+)
CO N p
L7 N 6i 4
CJ C7 6
M V o co
coC. C0
a) (o LC7
Co V CO ©
6 Co 6
CA 6) N C0
CO 6) 6
p 0)
p c) CO
co ti p p
p CO
L0 co. N
-4 p 6 4
C
N Cx4 .6
E E E =
u7
N 0) N ti
10
co M co c7
L Li7 0
0) D. co co
p 6
6) h 6) p
Co Liz (7l
[0 . ▪ t CO
6) co LC) F
u7 0) RI I`
Co 6 6 V
L0 .
6 M
rn rn 0
6
co L0 N C0
6 I— N
C` r--
6 - o
CO
p p L8 )
ti Ld)
N MI .6
E E E =
in
2
- ti N
6 p L7 CD
r CO p p
CO o
M ~ cu
(0 co p
I— N CO
o 6 N
p co. o
6 Co 4 co
. [0 u7
LC7 N
CAp (Ca
ti Ca - V 0)
ti N ti C7
CO Co ID c4
V V CO co
6 6) x V
ti o co 6
N Co o C0
4 Co N
r C0 r 3 CO CO 43
6) ti u7 co
C
(13 Cx4 .6
E E E =
N
in
2
0)
• 0) (7) o
copo�
• co 6) p
C0 N L7
✓ - �
co ID
CO
r- . LC7 L0
Co co 4 a)
10 Li. CO 'CO
- . 0) M
C7 N CO
LC) . ▪ N C0
6 O CU
0 c‘io V• co
•c0
N [ . r C0
co M
CO ? CO
(D CO
r CO
co. 6) co
Lci
N CV .6
E E E =
10
in
2
CO (IDLC) C7
(0 N 6 CO
CO CT)
. N (0
prcoo
co V CC1
. LC7 N
O co LC7
Co p c)
o r 6 r`
- "4—
co p
6 Ca N CO
L0 r 6) p
r- i N
CO — C0
p 6
M h ID
C6
C4 x
a) m .c
E E E a
in
d
p co N LL7
v co o
o N C7
4Cooc
C7 C7 N (0
C` 0) V
r` d
o co RI 0,7
0) . ▪ LC7 (0
6 N 6 a)
C0 C0 10
CO 0) r
N CA C7
CO l0 o c4
LC) N V co
LCi ti ID
v rn C)
u7 I� r
CO C7) CO C0
LCi I� ID
C4
a) m .c
E E E a
Cn
0
10
M'cr ti M
Co
L C) o
4 6 o
N N CO (0
I- 0) CO
ti rn co o
M Co
6) CO LC7 L0
CO L6
C0 co co 0
10 0) V
CO o r C7
t (0 10
N CA N LC7
(O N- tTr
I: Co 6
CO V V C0
I- 6i u7
L L7 h co
6 C r
D. 'Cr. r CO
✓ ^ CO
✓ r C
C3
a) CC44 .6
E E E
N
N
0
a
10
0) M o
u7o°D
LC) a7
v ti o
C` CO M
cicoov
L7 CV CO c,1
CO CO co 0
M Co CO
r 0) (0 CO
f� M N
CO C0 0) CO
ti 0) V
ti
4 10
C` CO CO
u7 N- Fj)V
0 0 10 co
Co 10
07 CO
r- 0i CO V
▪ r-
co 0)
6 N O `r
r 117 � M
r r �
C
C4
a) m .c
EEE
(I)
0
N
o a)
• N 10
a7 r C7
ci Co o
CV CO
c-i a v
6
6 Ca 10
M LC) co
ti
N
6J N
L C) C2 p
ti 10
C
co O ca
I 6 c4-
(-a N co co
u 7 I- ID
N
C0 N- V
o co N- M
ca co a
co p
M CO O
q NI
V • CO
III-A-31

�
0
0
2
0
3
\
_
L
Table III -Al (concluded).
G a m r
¥ 6 6 \
o m r
a d $
G o = ¥
¥ a 6 (42
00 o co co
w % r
G n o =
A a 6
‘.,T-
.o
1-
0 = co
/ / & %
a)§ .c
E E E c
/
0
0
0
r ¥ r r
g a 6 $
o CO o =
¥ % 6 \
CO CO CO a
a a 6 $
¥ o a o
a / d $
CO o a m
▪ 2 6 $
o a o
w e 6 7
G o G /
W R 6 t
o G G
w e 6 7
m 2 1- 1-
% / &
a)§ .c
E E E c
/
0
c
\ ni \
2 0 ¥ o
6 / 6 7
\ co
- 7
@- -32

Table III-A2. Correlations between 2020 salinity and several climatic variables: Tar River
discharge, water depth, rainfall, wind speed and wind direction. White = Pre -Mod Alt, Gray = Post -
Mod Alt L, and Blue = Control. Bold values indicate significant correlation.
Station
Sample Size
(# of days)
Pearson's r correlation
Mardia's r correlation
Wind Direction
Discharge
Water Depth
Rainfall
Wind Speed
Broomfield Swamp Creek
BSCS1
341
-0.226 (lag: 0 days)
+0.128
-0.186 (lag: 1 day)
-0.019
0.212
BSCS2
340
-0.284 (lag: 0 days)
-0.065
-0.296 (lag: 2 days)
+0.096
0.118
BSCS3
340
-0.281 (lag: 0 days)
+0.139
-0.265 (lag: 3 day)
+0.091
0.101
SCUTI
SC1S1
366
-0.176 (lag: 0 days)
+0.405
-0.134 (lag: 1 days)
-0.096
0.198
SC1S2
362
-0.269 (lag: 0 days)
+0.159
-0.181 (lag: 2 day)
+0.043
0.294
SC1S3
366
-0.365 (lag: 0 days)
-0.201
-0.243 (lag: 2 day)
+0.083
0.162
Jacks Creek
JS1
366
-0.263 (lag: 0 days)
+0.007
-0.188 (lag: 4 days)
-0.006
0.094
JS2
366
-0.281 (lag: 0 days)
-0.122
-0.171 (lag: 4 days)
+0.110
0.045
South Creek
SS1
352
-0.305 (lag: 0 days)
+0.133
-0.200 (lag: 4 days)
+0.151
0.103
Little Creek
LCS1
337
-0.253 (lag: 0 days)
+0.169
-0.258 (lag: 1 day)
_0_065
0.369
LCS2
364
-0.364 (lag: 0 days)
-0.058
-0.289 (lag: 1 day)
_0.068
0.145
Jacobs Creek
JCBS1
356
-0.248 (lag: 6 days)
+0.016
-0.175 (lag: 5 days)
+0.057
0.061
JCBS2
366
-0.321 (lag: 7 days)
-0.062
-0.169 (lag: 4 days)
+0.163
0.033
PA2 Creek
PA2S1
351
-0.268 (lag: 8 days)
+0.034
-0.131 (lag: 4 days)
+0.067
0.032
Drinkwater Creek
DWS1
366
-0.254 (lag: 0 days)
+0.038
-0.179 (lag: 1 day)
_0_052
0.152
DWS2
366
-0.330 (lag: 7 days)
-0.116
-0.160 (lag: 4 days)
+0.116
0.038
Long Creek
LDCS1
366
-0.387 (lag: 7 days)
-0.063
-0.143 (lag: 5 days)
+0.143
0.089
LDCS2
366
-0.389 (lag: 7 days)
-0.073
-0.147 (lag: 5 days)
+0.168
0.064
Tooley Creek
TS1
340
-0.345 (lag: 7 days)
+0.043
-0.152 (lag: 4 days)
+0.123
0.068
TS2
366
-0.362 (lag: 7 days)
+0.003
-0.146 (lag: 4 days)
+0.156
0.030
Pamlico River
PS1
346
-0.661 (lag: 4 days)
+0.001
-0.180 (lag: 10 days)
+0.237
0.108
Huddles Cut
HS1
345
-0.254 (lag: 35 days)
+0.241
+0.100 (lag: 27 days)
-0.115
0.264
HS2
366
-0.380 (lag: 0 days)
+0.344
-0.134 (lag: 1 days)
-0.056
0.199
HS3
339
-0.518 (lag: 5 days)
-0.082
-0.131 (lag: 1 day)
+0.161
0.120
Porter Creek
PCS1
366
-0.379 (lag: 6 days)
+0.002
-0.172 (lag: 1 day)
+0.028
0.129
PCS2
366
-0.539 (lag: 5 days)
-0.082
-0.134 (lag: 11 days)
+0.196
0.198
Durham Creek
DCS1
366
-0.441 (lag: 3 days)
-0.104
-0.163 (lag: 3 days)
+0.190
0.268
DC19S1
366
-0.458 (lag: 4 days)
-0.083
-0.151 (lag: 11 days)
+0.142
0.149
DC11S1
366
-0.336 (lag: 0 days)
+0.009
-0.153 (lag: 1 days)
+0.119
0.220
Duck Creek
DKCS1
353
-0.276 (lag: 0 days)
+0.334
-0.224 (lag: 1 day)
-0.155
0.188
DKCS2
366
-0.465 (lag: 3 days)
+0.126
-0.194 (lag: 7 day)
+0.042
0.102
III -A-33