I-33_BeaverDamReportFinal20190409

Ellerbe Creek Beaver Dam Study Report
City of Durham, Public Works Department, Stormwater & GIS Services Division, Water Quality Unit
Project # 13-002
April 9, 2019
Introduction
The City of Durham regularly monitors the water quality of its streams to evaluate their overall
condition. A common measurement of water quality monitored by the City of Durham is Dissolved
Oxygen or DO. DO is important for all aquatic life within the stream and affects the chemical and
biochemical processes that occur in the water as well. Therefore, low DO can have a negative impact on
the water quality of streams. A possible cause of low DO concentrations may be the impoundment of
water caused by dams constructed by beavers (Castor canadensis). In relation to another study
conducted in the Northeast Creek watershed, a contractor for the City of Durham recommended for a
device called a beaver pond leveler to be installed in order to help reduce the impoundment of water
caused by beaver dams. The contractor believed the pond leveler would increase the flow of water
below the beaver dam, thus increasing the downstream DO. However, a pond leveler was never
installed in Northeast Creek due to the fact that the selected beaver dam was abandoned and no longer
actively maintained by beavers. The Ellerbe Creek Beaver Dam Study (project 13-002) was initiated in
order to determine if installing a beaver dam pond leveler could be used to increase downstream DO
concentrations. If the leveler is successful, it may be used as a possible remediation option for other
locations in the City of Durham with issues involving the presence of a beaver dam and low DO.
The main objectives for the project were:
1. Determine the current DO concentrations directly downstream of a beaver dam in a tributary of
Ellerbe Creek.
2. Assess if installing a beaver pond leveler will have any effect on DO concentrations directly
downstream of the beaver pond.
3. Assess the short-term behavior of the beavers to determine if they will migrate and construct a
new dam as a result of the beaver pond leveler and if it would affect any benefits that might be
provided by the leveler.
To meet these objectives, a pond leveler was installed in a selected beaver dam in the Ellerbe Creek
Watershed on 3/14/2018. A pond leveler is a low-cost device that allows water to flow through a beaver
pond while maintaining the natural surroundings of the beaver dam and managing the water level of the
beaver pond upstream of the dam. The pond leveler is designed in such a way that the beavers are not
alerted to the sound of falling or trickling water and therefore, are not prompted to plug or destroy the
leveler.
DO was continuously monitored while the pond leveler was in place and active (treatment phase). The
leveler was then sealed in order to replicate conditions prior to the installation of the device (i.e., water
Page 1of11

only flowed over or through the beaver dam). DO was continuously monitored again to provide baseline
conditions (control phase) for comparison of the DO concentrations with the leveler in place and active.
This report summarizes and analyzes results from project data that were collected between 1/12/2018 —
9/20/2018.
Methods
The project study design is summarized here, for more details refer to the project's Quality Assurance
Project Plan (Durham SW, 2018).
Site Description
Data for the project were collected at a monitoring location on an unnamed tributary of Ellerbe Creek
just off Midland Terrace in Northeast Durham (Figure 1). The monitoring location was directly
downstream of a beaver dam (Main Beaver Dam) that was part of an approximately 800,000 square foot
beaver pond. The monitoring well that contained a Dissolved Oxygen logger (DO logger) was located
approximately 20 ft. downstream of the beaver dam directly next to the standpipe of the pond leveler
that was installed in the Main Beaver Dam. About 50 ft. downstream of the Main Beaver Dam was a
smaller beaver dam (Lower Beaver Dam). This Lower Beaver Dam was removed during installation of the
pond leveler; however, the beavers repaired and maintained this lower dam during the course of the
study. The repair of the lower dam caused the water between the dams, where the DO logger was
located, to be stagnant and pond -like. Although this was not an ideal location for measuring DO, if the
logger had been placed below the Lower Beaver Dam it was very likely that any effects of the pond
leveler would have been negated and unmeasurable due to the presence of the lower dam.
Page 2of11

Figure 15ite map of the project area. Insert shows location of project area in Durham County.
Page 3 of 11

Beaver Pond Leveler
Construction
The beaver pond leveler used for this study was based on The Clemson Beaver Pond Leveler design
(Figure 2; http://www.dnr.sc.gov/wildlife/publications/pdf/ClemsonBeaverPondLeveler.pdf) and was
constructed and installed on 3/14/2018 by Carolina Environmental Contracting, Inc. The pond leveler
was constructed of 8" and 10" diameter polyvinyl chloride (PVC) pipe and consisted of two main parts:
the intake device and the outlet pipe.
Elbow and stand pipe are optional.
Needed only to manage water level
if maintaining pond is an objective
8" dia. 46 pvc pipe Intake
Device
Beaver
Dam
zap
T - joint fitted with a drain
plug may replace elbow
Pond Side
1' Re -bar
6' long
Figure 2 Clemson Beaver Pond Leveler Source: The Clemson University Extension Service
The intake device (Figure 3) consisted of a 10' section of 10" diameter PVC pipe with 2" holes drilled
intermittently along its length. The PVC pipe of the intake device was surrounded by a galvanized
welded wire cage to prevent the beavers from plugging the pipe. The outlet pipe (Figure 4) consisted of
an approximately 20' section of 8" diameter PVC pipe with an elbow and standpipe section connected to
the end. The elbow and standpipe at the end of the outlet pipe allowed for the beaver pond behind the
dam to be maintained as the water level of the pond would only drop to the height of the top of the
standpipe.
Figure 3 Beaver pond leveler intake device.
Page 4 of 11

Figure 4 Beaver pond leveler outlet pipe with standpipe.
Installation
The pond leveler was installed by first using a backhoe to open a notch in the beaver dam that was large
enough for the outlet pipe to fit in. While the water level in the beaver pond behind the main dam was
being lowered, the intake device was connected to the outlet pipe using PVC primer and glue and the
lower beaver dam was being removed. The pond leveler was then placed through the notch in the dam
so the intake device was on the upstream side of the dam and resting on the bottom of the pond, and
the outlet pipe was on the downstream side of the dam where the standpipe was approximately 20'
from the dam (Figure 5). After the pond leveler was in place and all parts were properly attached, the
notch in the beaver dam was filled back in. It was anticipated that the beavers would complete the
repairs to the dam and further stabilize the pond leveler.
Figure 5 Pond leveler installed and active in the main beaver dam.
Page 5of11

Standpipe Capping
After data were collected while the pond leveler was active, the standpipe of the pond leveler was
sealed (Figure 6) to replicate conditions prior to the installation of the device. This was done by using
PVC primer and quick -dry PVC cement for wet conditions to attach an 8" diameter PVC slip cap to the
end of the standpipe.
Figure 6 Standpipe capped (right). Capped PVC with holes is the well that contains the DO logger (left).
Dissolved Oxygen Logger
Data collection
Continuous dissolved oxygen data were collected using an Onset HOBO U26 Dissolved Oxygen Data
Logger. Before deployment, the DO logger was calibrated and launched in the Stormwater Lab (located
in the Annex building of City Hall) following the recommendations of the manufacturer. The DO logger
was deployed in a well 1 ft. downstream from the standpipe of the pond leveler. The well was
constructed of a 4" PVC pipe with 1" holes drilled approximately 6" apart to allow for appropriate mixing
with the surrounding stream water (Figure 6). The DO logger was attached to a foam block inside the
PVC well, which allowed the logger to move up and down as water levels fluctuated to consistently
collect DO measurements at a fixed depth below the surface.
The DO logger was programmed to collect DO concentration (mg/L) and temperature (°C)
measurements every 30 minutes. Data were downloaded from the DO logger on a weekly basis. During
the field visits a calibrated YSI ProODO field meter was used to collect DO saturation (%) and
concentration (mg/L), temperature (°C), and barometric pressure (inHg) readings both inside and
Page 6 of 11

outside of the well before downloading the data using a HOBO waterproof shuttle. The DO logger was
also cleaned by gently brushing the sensor with a soft -bristled toothbrush and rinsing with tap water to
reduce fouling. The relative level of fouling of the sensor was recorded (scale of 1-5). After data were
successfully downloaded, a second measurement of DO concentration and saturation, temperature, and
barometric pressure was recorded using the YSI ProODO.
HOBOWare Pro software was used in the office to transfer downloaded data from the waterproof
shuttle to the Water Quality Unit (WQU) server. The Dissolved Oxygen Assistant in the HOBOWare Pro
software was initially used to correct for sensor fouling and drift using the field measurements taken at
the beginning and end of each deployment period. The Dissolved Oxygen Assistant was only able to
correct DO readings when the measurements taken at the beginning and end of each deployment were
above 0 mg/L. Due to low streamflow, the DO readings were often at or around 0 mg/L when the field
measurements were taken. This caused the corrected data produced from the Dissolved Oxygen
Assistant to be of poor accuracy. Therefore, the data used for analysis were the uncorrected data and
were examined for relative differences in DO during the control and treatment phases of the project.
Individual files were saved in the proprietary .hobo file format and also exported as comma -delimited
(csv) files.
DO data were then compiled, checked for gaps and errors and graphed with precipitation data.
Precipitation data were not directly measured by WQU staff, but were obtained from the USGS website
(https://waterdata.usgs.gov/nwis) for the rain gage on Ellerbe Creek at West Murray Ave (USGS Station
ID 360143078540945) which is approximately 1.8 miles away from the location of the DO logger. Final
graphs were prepared using the graphing function in Microsoft Excel 2010.
Data Collection Timeline and Issues
DO data used for analysis were collected from 8/9/2018 — 9/12/2018. Treatment data, when the pond
leveler was installed and active, were collected first from 8/9/2018 — 8/29/2018. The standpipe of the
pond leveler was capped on 8/30/2018 to replicate baseline conditions. Control data were then
collected from 8/30/2018 — 9/12/2018. While it was initially intended for baseline data to be collected
prior to the installation of the pond leveler, due to unforeseen circumstances the collection of data was
delayed multiple times throughout the course of the study.
Initially, after the pond leveler was installed the beavers did not immediately return to further stabilize
and repair the damage to the dam caused by the installation process. To help draw the beavers back to
the dam, beaver attractant (Hawbaker's Beaver Lure) was applied in late March of 2018. After the
application of the attractant, the beavers returned and began to maintain the dam and the pond leveler
began to function with water overflowing from the top of the standpipe. However, due to multiple
heavy storms in the spring and early summer, the beaver dam was damaged and breached multiple
times causing water to bypass the leveler and standpipe. As there was no effect on downstream DO if
the leveler was bypassed, collection of treatment data while the leveler was functional was further
delayed.
Page 7of11

Due to these multiple delays in the collection of data, the study design was altered to collect DO
concentrations with the leveler active (open standpipe for the treatment phase) and inactive (capped
standpipe for the control phase). The DO concentrations under the two conditions were then compared.
Results and Discussion
DO data collected when the standpipe was active are shown in Figure 7 and DO data collected when the
standpipe was capped and inactive are shown in Figure 8. Data collected from 9/13/2018 — 9/20/2018
were excluded due to damage of the DO logger well which was the result of many large rain events
(including Hurricane Florence) which caused for the data collected during this time period to not be
useable .
Standpipe Active
C 961
4
3.5
3
J
m
2.5
2
C
O
U
0 1.5
0
1
0.5
0.00
0.05
0.20
0 1 ? ''I i 91 - I I - — I ' ' I— i h 4 I i - :W'. ! ''V IAA- 0.25
8/9 8/10 8/11 8/12 8/13 8/14 8/15 8/16 8/17 8/18 8/19 8/20 8/21 8/22 8/23 8/24 8/25 8/26 8/27 8/28 8/29
M Precipitation (in) —D.O. Conc (mg/L) • DO Field Observation (mg/L)
Figure 7 Dissolved oxygen and precipitation data when the standpipe was active.
Page 8 of 11

Standpipe Inactive
4.5
0.00
a
0.05
3.5
3
J
m
2 6
DIG
2.5
A
c
o
c
-
2
c 2
m
&
n
c
O.ls Z
00 1.5
a`
1
a.20
0.5
•0.4
• 0.3
0 0.25
8/29 S/30 8/31 9/1 9/2 9/3 9/4 9/5 9/5 9/7 9/8 9/9 9/10 9/11 9/12 9/13
Precipitation (in) —DO conc (mg/L) s DO Field Observation (mg/L)
Figure 8 Dissolved oxygen and precipitation data when the standpipe was inactive.
When comparing the graphs, there appear to be more notable and frequent fluctuations in DO
concentrations in response to rain events when the standpipe was active than when the standpipe was
inactive. In between rain events, though, the leveler did not appear to provide any benefits to instream
DO concentrations and levels were consistently at or near 0 mg/L during dry weather. However, there
was one spike in DO concentration between 8/23 — 8/25 that does not line up with any rain event. This
could have been caused by a small localized rain event not captured by the rain gage or some other
disturbance to the beaver pond upstream of the pond leveler.
When the standpipe was inactive, instream DO concentrations did not increase in response to multiple
rain events (one > 1 in.), and there was only one period of time (9/2) when DO concentration levels
were above 0 mg/L. These increases were also very small (< 0.5 mg/Q.
Water temperature likely had no strong effect on DO concentrations as there was little variation in
water temperature during data collection. When the standpipe was active water temperature varied
from 21.8 — 25.1 °C and when the standpipe was inactive the water temperature varied from 23.2 — 25.2
°C. The 2 — 3 °C change in water temperature correlates to an approximately 0.6 — 0.3 mg/L change in
DO concentration. DO concentration fluctuations this small are not likely to impact the DO
concentrations in the water column.
While there is a North Carolina water quality standard for DO concentration of a daily average DO
concentration of >_ 5.0 mg/L and a minimum DO concentration of 4.0 mg/L, we did not compare results
Page 9 of 11

to the standard due to the fact that our results were uncorrected and could have possibly been affected
by drift or fouling of the DO sensor. DO field observations, measured with a YSI proODO field meter,
illustrate that the DO logger readings were biased low which further explains why our data could not be
used for comparisons.
The presence of the pond leveler appeared to have a minimal effect on the behavior of the beavers.
Initially, after the pond leveler was installed, the beavers did not return to the dam to repair the
damage. However, after beaver attractant was applied the beavers returned to the dam and continued
to maintain and repair the dam for the remainder of the study. Additionally, there were no observed
attempts by the beavers to plug or destroy the pond leveler.
In summary, the beaver pond leveler did have a small effect on DO concentrations directly downstream
of the beaver pond based on increases in DO concentrations from near 0.0 mg/L to as high as 4.2 mg/L
after rain events. However, the increases in DO were minimal and were only maintained for periods of
time directly following rain events. Additionally, the pond leveler was only operational for a small
portion of the entire study due to damage of the beaver dam caused by consistent rain events during
the spring and summer of 2018 and other unforeseen issues. However, the pond leveler was successful
at lowering and maintaining the level of the beaver pond behind the dam without disrupting the
behavior of the beavers. As a result of these observations, the beaver pond leveler appears to be more
suited for managing water levels and remediating potential flooding issues caused by beaver dams than
for increasing downstream DO concentrations, at least on small headwater drainages such as our study
site.
Page 10 of 11

References
COD SW. 2018. Ellerbe Creek Beaver Dam Study (Project #13-002) Quality Assurance Project Plan
(QAPP). Water Quality Unit. Durham, NC. Water Quality Unit.
Clemson University Cooperative Extension Service. 1994. The Clemson Beaver Pond Leveler. Available
from http://www.dnr.sc.gov/wildlife/publications/pdf/ClemsonBeaverPondLeveler.pdf (accessed
January 2019).
Acknowledgements
We would like to thank both Al Estok and Dana Hornkohl of the Stormwater Infrastructure Group in the
City of Durham's Public Works Department for their continued assistance and invaluable input on this
project. Additionally we want to give Al an especially big thank you for his many hours and photos taken
in the field analyzing the behaviors and activities of the ever elusive beaver at our monitoring site.
Beaver repairing damage to the Main Beaver Dam. Photo taken on 9/26/2018 at 5:27 AM.
Page 11 of 11