Ted Lewis ([email protected]); Raymond. S. Bradley; Pierre
Francus University of Massachusetts, Amherst. Climate System
Research Center
LACUSTRINE SEDIMENTARY PROCESSES NEAR A HIGH ARCTIC DELTA, LAKE
TUBORG, ELLESMERE ISLAND
ACKNOWLEDGMENTS
See http://www.geo.umass.edu/gradstud/lewist/lewist.htmand
http://www.paleoclimate.org for more details.
ABOVE: The major watersheds of Lake Tuborg (~81oN 75oW). The
southwest basin is meromictic (about 155 m deep); the northeast
basin is entirely freshwater (about 80 m deep). Watershed A is
nivally fed, B is mostly nivally fed, C and D are mostly glacially
fed. Inset shows location of Lake Tuborg on Ellesmere Island.
ABOVE: (A) 1959 aerial photograph of Lake Tuborg highlighting
Delta A. (B) Bathymetric map of the study area near Delta A. Black
symbols are points where bathymetry was measured. Red symbols are
CTD cast sites. The red grid illustrates the sampling area. White
labels are depth (m) below lake level.
Lake T
uborg
500
1000
1500
2500
3000
2500
3500
4500
2500
2000
1500
2500
3000
+
+480000 E8990000 N
500000 E8970000 N
+500000 E9010000 N
Agassiz Ice Cap
D
C
A
B
Alert
L. Tuborg
Eureka
Grise Fd.
0 100 200
Scale (Km)
SCALE (Km)
0 5 10
489400 489500 489600 489700 489800 489900 490000
8989600
8989700
8989800
8989900
8990000
8990100
8990200
T4-A T4-A
T4-B T4-B
T4-C T4-C
T4-D T4-D
T4-E T4-E
T4-F T4-F
T4-G T4-G
T4-H T4-H
T4-I T4-I
T4-J T4-J
T4-K T4-K
T4-L T4-L
T4-M T4-M
T4-N T4-N
T4-O T4-O
T4-P T4-P
RAIN30
7073
74
75767779
80
81
82
83
85
86
87
88
89
92
95
96
97
98
99100
101102
103104
105107108
109
110111
112
130
-105
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
UT
M N
OR
TH
ING
UTM EASTING
B
A
The images below were produced in an open source software
package called OpenDX. Data were collected with a Sea Bird
Instruments SBE19 CTD (Conductivity Temperature/Transmissivity
Depth) profiler with a 660 nm, 25 cm path length SeaTech
transmissometer.
OpenDX (http://www.opendx.org) is a visualization and analysis
application. OpenDX provides tools for manipulating, transforming,
processing, realizing, rendering and animating data and allows for
visualization and analysis methods. OpenDX was developed by the
Deep Computing Institute at IBM, and was released as open source on
May 24, 1999. It runs on UNIX-based machines, but has been ported
to Windows 95/NT.
Most importantly for this project, interpolated data can be
exported and analyzed. It is hoped that the SeaTech transmissometer
can be calibrated to suspended sediment concentration. The mass of
sediment within each water volume could then be calculated.
Table 1: CTD AND TRANSMISSOMETER SPECIFICATIONS Temperature
Conductivity Pressure Transmissivity Range -5 to +35 oC 0-7 S/m 50
to 10,000 psia 0-5 V (~0.004-20
mg/L) Accuracy 0.01 oC /6 months 0.001 S/m/month 0.5% of full
scale ? Resolution 0.001 oC 0.0001 S/m 0.03% of full scale 0.001
V
R
Cast Locations,105 m apart
Overflow plumes (attenuation increases)adjacent to
tributaryinputs.
Conductivity (us/cm). Sharp and strong chemocline at
approximately 55 m.
ATTENUATION ON JUNE 16, 2001 CONDUCTIVITY ON JUNE 19, 2001
A B C
At left: Colorbars for OpenDX images.A represents temperature
(*C),B represents conductivity (us/cm)C represents attenuation (%,
100-transmissivity).
6/1
7
6/2
1
6/1
9
6/2
3
6/2
5
6/2
7
0
20
40
60
80
100
0
5
10
15
20
25
30
AT
TE
NU
AT
ION
(%
)
DATE
An NSF grant to R. Bradley and a GSA award to T. Lewis supported
research. The Polar Continental Shelf Project provided field
logistics. Joe Rogers and Chloe Stuart provided field assistance.
Dr. Chris Duncan, Frank Keimig, and Celeste Asikainen.
A B
C D
At left: (A) Attenuation on June 16, 2001. Note the large
attenuation increase at the chemocline (~55 m). This attenuation
spike is seen throughout the meromictic basin even before the onset
of summer snowmelt. This could be a result of algal mats, sediment
of insufficient density to pass into the monimolimnion, or
precipitation from the dissolved load when anoxic water (in the
monimolimnion) comes in contact with oxic water (in the
mixolimnion). (B) Attenuation on June 28. The attenuation spike is
thicker, less well defined, and has shallowed somewhat. Meltwater
may have caused some mixing, and suspended sediment may have
“ramped” on the density difference at the chemocline. (C) Water
temperature on June 19. Note the relatively warm water of the
monimolimnion and mixing near the delta face. (D) Attenuation on
June 19. Turbid water near the delta face.
6/16 6/236/19 6/28At left: (E) Attenuation on June 19. (F) Water
temperature on June 19. The overflow plume is composed of water
slightly warmer than its surroundings.
At right: Top: Each 3-D visualization is composed of volumetric
pixels (voxels). Each voxel has an interpolated value. Histograms
of attenuation for each date were produced. Bottom: A surface plot
combining the four histograms (cf. Beierle et al., 2002). Each
voxel has a known volume, and field calibration of the
transmissometer to suspended sediment concentration should be
possible. Therefore, the mass of sediment within each visualization
(or a portion of each visualization) could be calculated
INTRODUCTION: Lake Tuborg is a large glacially fed and glacially
dammed lake on west-central Ellesmere Island. It has two
sub-basins. The southwest sub-basin is meromictic, and has a
maximum depth of 155 m. The northeast sub-basin is freshwater and
is shallower (Zmax, ~80 m).
This poster shows processes of sedimentation recorded near a
large nival delta (watershed A at left, photo at right) located at
the north-central portion of the lake. Processes were recorded
during peak nival melt in 2001 with sediment traps, a datalogging
flow meter, and a CTD.
Annual suspended sediment discharge from this nival delta is
likely much less than glacially-fed deltas at Lake Tuborg. However,
the nival tributary flows into the lake in its meromictic basin,
and the delta face is steep. Therefore, this is an ideal location
to study how sedimentary processes are affected by meromixis.
19-May 2-Jun 16-Jun 30-Jun 14-Jul 28-Jul 11-Aug 25-Aug
30
20
10
0
19-May 2-Jun 16-Jun 30-Jun 14-Jul 28-Jul 11-Aug 25-Aug
-15
-10
-5
0
5
10
0
4
8
12
0
4
8
12
19-May 2-Jun 16-Jun 30-Jun 14-Jul 28-Jul 11-Aug 25-Aug
0
.40
.80
1.20
0
.40
.80
1.20
Nu
mb
er
of D
ays
Ave
rag
e ‘4
7-’0
1 P
reci
p (
mm
)Te
mp
era
ture
(*C
)
Pre
cip
itatio
n (
mm
)
La
ke L
eve
l (m
)
A
B
C
19-May 2-Jun 16-Jun 30-Jun 14-Jul 28-Jul 11-Aug 25-Aug
-20
-10
0
10
-20
-10
0
10
2001 Regional Near Surface Temperatures
Resolute
Eureka
Lake Tuborg (~10 m ASL)
Air T
em
pe
ratu
re (
20
01
)
(10 m ASL)
(67 m ASL)
D
Above: Weather and climate at Lake Tuborg and regional weather
stations. (A) Precipitation at Eureka, 1948-2001. Lower graph
(black) shows mean precipitation for each day. Upper graph (red)
shows the number of days used for to calculate mean
precipitation.
(B) Red line graph shows mean daily 2001 air temperature
recorded at lake level with a datalogging thermistor. Temperature
was recorded with a 15-minute interval. Black line graph shows
1947-2001 air temperatures recorded at Eureka. Black bar chart
shows 2001 Lake Tuborg precipitaton (mm).
(C) Lake level fluctuations at Lake Tuborg in 2001. Zero datum
is the lowest recorded lake level.
(D) Air temperatures recorded at Lake Tuborg, Eureka and
Resolute in 2001. R-squared for Resolute and Tuborg is 0.57;
R-squared for Eureka and Tuborg is 0.90.
Mean daily temperature (5/21-8/11) for:Lake Tuborg , 2.9
*CResolute, 2.9 *C
0 200 400
-80
-60
-40
-20
0
Me
ters
Be
low
Su
rfac
e
Meters From Shore
T4-H T4-ET4-FT4-G
Mixolimnion
Monimolimnion
-85 m
55 m
-79 m
-67 m
6/2212:00
6/2300:00
6/2312:00
6/2400:00
6/2412:00
6/2500:00
6/2512:00
DATE AND TIME
B
0 40 80120
0
0.1
0.2
0.3
0.4
Flo
w V
elo
cit
y (
m/s
)
IRPM
Y=0.002679*X+0.03012r =0.99
A CAt left: Current meter results from Site G. Currents were
recorded when the meter was hung at the chemocline (55 m). Repeated
deployments were made below the chemocline, even during times of
peak nival melt.
(A) Calibration curve (B) Lake Tuborg Currents (C) Current meter
courtesy of Roger Lewis and John Sweeney.
Based on CTD profiles, the difference in concentration between
freshwater and saltwater based on salinity and temperature is about
20 g/L. During the 1995 slush flow at Delta D, SSC reached 5 g/L
(Braun, 1997), while SSC reached 2 g/L during the 2001 nival flood
at Delta A. This theoretically precludes the occurrence of
fluvially generated underflows in the meromictic sub-basin of Lake
Tuborg, even during extreme events.
0 50 100 150 200 250Particle Diameter (um)
0
2
4
6
Fre
quen
cy(%
)531
0 50 100 150 200 250Particle Diameter (um)
0
2
4
6
Fre
quen
cy(%
)
410
0 50 100 150 200 250Particle Diameter (um)
0
2
4
6
Fre
quen
cy(%
)
589
0 50 100 150 200 250Particle Diameter (um)
0
2
4
6
Fre
quen
cy(%
)
640
0 50 100 150 200 250Particle Diameter (um)
0
2
4
6
Fre
quen
cy(%
)
1032
0 50 100 150 200 250Particle Diameter (um)
0
2
4
6
Fre
quen
cy(%
)
1036
LEGEND: Numbers in red boxes represent mass accumulation rate,
and are size-scaled. Black histograms are
Results from delta-proximal sediment traps deployed from June
20-30, the period of peak nival melt. Sediment traps were deployed
to determine the effect of the monimolimnion on sedimentation. One
trap on each mooring was placed at the top of the chemocline, and
one was 1.5 m above the lake bottom. Mass accumulation rate (MAR)
units are g/m2/d. Grain size data are averages of 3-runs per
sample. Each trap was “double barreled”, and grain size and MAR
results are averages of the two barrels. MAR and grain size mode
decreased with distance from shore. Interestingly, MAR for upper
and lower traps on the same mooring are similar. This implies that
little additional sediment is added in the monimolimnion. These
results agree with CTD casts and current recordings, which imply
that stratified flows (interflows and overflows) are confined to
the mixolimnion. Gravity cores were obtained with a Glew corer.
Core T4G is mostly massive sand. Two silt clasts are present in the
upper 12 cm. These are likely slump blocks, transported from the
steep delta slope. Sandy laminae between 13-18 cm could record
daily sediment pulses. Core T4E consists mostly of sand laminae
separated by silt and clay laminae, except for a massive silty
section from 12-18 cm.
Jo
e R
og
ers
Brr
rrrr
rr!
Rain Gauges Radiation Shield
Se
aC
at S
BE
19
E
F
Page 1
