Jökull - 01.01.2004, Page 42
Jessica Black et al.
coordinates). To assist our study of bathymetry, we
also scanned and georectified the bathymetric map
made with echo sounder measurements by the Hy-
drological Service Division of the Energy Authority
in Iceland (Rist, 1975). The positions of the ter-
minal LIA moraines from Norðurjökull and Suður-
jökull (Figure 3) were digitized based on field obser-
vations, aerial photographs, bathymetry, and measure-
ments from the seismic survey.
The bathymetric map of Hvítárvatn (Figure 3) was
created by interpolation of available point datasets: a
bathymetric map made with echo sounder measure-
ments, GPS-based water depths recorded during the
seismic survey, and measurements made with an in-
tegrated fathometer and GPS Lowrance LC X-15 MT
system in 2003. The northern part of the lake was
well represented by the field measurements in 2003
(n=261) and 2004 (n=20), as well as water depth cal-
culations from the seismic transects (n=438). In pe-
ripheral areas and in the southern part of the lake, we
extracted 555 points from contours digitized from the
older bathymetric map (Rist, 1975). Bathymetry near
the edge of the lake was further constrained by 719
points taken from the lake outline, with water depth
set at zero. The five point shapefiles were merged and
then interpolated, using a tension spline masked to the
lake area, with resultant 5-m grid cell spacing. To as-
sist with visualization, we then created a shaded relief
map.
Isopach maps were created by interpolation of
the 235 point measurements from seismic lines lo-
cated distal to the LIA moraines. Maps of sediment
thickness were created for the uppermost four seismic
units, as well as for total sediment thickness encom-
passing the entire Holocene (Figure 4). Interpolation,
using Inverse Square Distance Weighting, was limited
to areas within 300 m of data points, with resulting
5-m grid-cell spacing. Contour shapefiles were then
created using an interval of 2 m or 5 m.
The sediment volume for each seismic unit was
calculated by multiplying the area of each grid cell
(25 m2) by its thickness. Each cell value within a
specific layer was summed and a volume calculated.
The volume was converted to mass, using the average
density of sediment in cores recovered from Hvítár-
vatn (1.7 g cm−3). This value is derived from gamma
density measurements made at the University of Min-
nesota, reduced by 0.35 units to compensate for an
overestimation of the sensor (Anders Noren, written
communication, July 2004).
The total sediment volume in Hvítárvatn was de-
rived by adding the estimated volume of sediment in
areas outside the seismic survey (LIA Scour areas and
the Southern Shallows, see below) to the measure-
ments from the seismic survey. Based on limited seis-
mic lines outside the main basin, a representative sed-
iment thickness of 30 m was used for the LIA Scour
zones and a sediment thickness of 10 m was applied
to the Southern Shallows.
In March 2002 we recovered four sediment cores
using the lake ice as a coring platform (Figure 3).
Each core recovered 5 to 6 m of sediment in a single
drive using a 7 cm diameter percussion-driven piston
corer (Nesje et al., 1987). The core sites were placed
at specific positions along the seismic profiles, using
GPS coordinates, to evaluate the lithostratigraphy of
as much of the sediment fill as possible, and to en-
able a correlation between the lithostratigraphy and
the acoustic stratigraphy derived from the seismic pro-
files. The coring was designed to assist in the interpre-
tation of the seismic data, and to derive a basin-wide
subdivision of the seismic records that had meaning
in a lithostratigraphic sense, in preparation for a ma-
jor coring effort in 2003.
RESULTS
Sediment character of depocenters derived from
seismic survey
The seismic profiles are used to define six charac-
teristic sediment depocenters in Hvítárvatn: for con-
venience we name these the Southern Shallows, the
LIA Scour, the Central Deep, the Prograding Delta,
the Bedrock Ridge, and the Northern Flats (Figure 3).
Seismic profiles that illustrate characteristic features
of the sediment fill in these depocenters are shown as
Figures 5, 6, 7, and 8.
The Southern Shallows comprises almost half of
the lake area. The region is shallow, increasing regu-
larly from 1 m depth at the southern end of the lake,
42 JÖKULL No. 54