Jökull - 01.12.1994, Blaðsíða 36

# Station n Geographic position (deg. min. sec., m) Scaled formal error (cm) Correlation
Lat. Long. Height dx dy dz dn de du n-e e-n u-n x-y y-z z-x
1 DJUP 3 64 38 48.64225 -14 17 00.93900 66.5742 0.55 0.35 1.11 0.42 0.36 1.17 -0.13 -0.46 0.15 -0.34 -0.40 0.62
2 FLAT 11 64 14 00.92255 -15 29 34.81554 63.7357 0.26 0.16 0.55 0.23 0.17 0.56 -0.21 -0.44 0.30 -0.30 -0.31 0.50
3 GILD 7 64 21 32.45916 -15 20 49.26299 74.4188 0.29 0.18 0.68 0.27 0.19 0.69 -0.22 -0.48 0.38 -0.34 -0.32 0.53
4 HEIN 6 64 18 14.06364 -15 3919.95442 149.0727 0.34 0.21 0.69 0.31 0.22 0.71 -0.18 -0.43 0.27 -0.32 -0.29 0.48
5 HOFF 9 64 25 05.23804 -15 23 40.04907 114.8160 0.33 0.20 0.64 0.28 0.21 0.66 -0.21 -0.39 0.19 -0.24 -0.31 0.51
6 HOFN 20 64 16 01.03650 -15 11 53.40999 73.1249 0.01 0.01 0.01 0.01 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00
7 HROL 2 64 07 06.15155 -16 04 32.25718 84.4616 0.48 0.28 0.91 0.42 0.29 0.94 -0.16 -0.39 0.19 -0.32 -0.27 0.49
8 HVAL 5 64 24 12.51638 -14 32 30.42675 75.8189 0.37 0.23 0.75 0.30 0.24 0.77 -0.23 -0.45 0.21 -0.28 -0.38 0.54
9 STAP 5 64 19 39.61728 -15 15 02.26818 101.0965 0.31 0.19 0.65 0.27 0.20 0.67 -0.13 -0.45 0.28 -0.36 -0.30 0.52
10 UPPS 4 64 12 29.99704 -15 43 22.41994 97.5473 0.40 0.24 0.83 0.34 0.26 0.85 -0.27 -0.49 -0.25 -0.31 -0.40 0.52
Table 2. GPS results, point positions in geographic, ellipsoidal coordinates (WGS84)
2. tafla. Niðurstöður GPS-landmœlinga. Staðsetning mœlipunkta er gefin í landfrœðilegum sporvöluhnitum (WGS84)
ambiguities were resolved. Then an L3 ambiguity-
fixed solution was determined. A network adjustment
of the solutions yields the results presented in Tables
2 and 3. The coordinates of the point HOFN were
held fixed to the values determined by Jahn (1991)
from measurements done in 1987. Errors as deter-
mined by the rms scatter of repeated measurements
around the weighted mean, and scaled by the formal
errors of the Bemese solutions, are mostly well with-
in 1 cm in the horizontal components and only slight-
ly worse for the vertical component.
The same data set was analysed independently by
Sigmundsson (1992) using the same software. In his
analysis 50% of the L1 and L2 ambiguities were re-
solved. The results are very similar, and the differ-
ences in individual components are mostly within the
error bars of the present solutions. Only one station
shows differences larger than 2 cm, the most distant
point at Djúpivogur.
GRAVITY OBSERVATIONS
Gravity was measured at all GPS points plus a
number of additional points in July 1991 and in July-
August 1992 with LaCoste-Romberg (LCR) model G
gravity meters (1991: No.G 688; 1992: No.G 445).
An additional, short survey was conducted in 1991 at
several points to test a Sodin thermostated quartz
spring gravity meter (No. 254GT). All gravity differ-
ences were measured relative to point 701 (Fig. 1,
Table 4), either directly or in steps. In 1991 the mea-
surements were mostly carried out in loops of about 1
h duration, for long distances in a symmetrical
scheme, i.e. A-B-C-...-C-B-A. The instrument G 688
showed some irregular drift, and the survey was re-
peated in 1992 with instrument G 445. The additional
survey with the quartz spring instrument 245GT fol-
lowed the scheme A-B-A-B-A-C-A-C-A-D... The
drift was remarkably low, but the individual reading
scatter was more than twice that of LCR G 445, part-
ly because of wind and sea surf at the nearby coast.
The gravity differences were evaluated prelimi-
narily in the field by constructing the instrument drift
by hand. No excessive errors were detected, in cases
of irregular drift the measurements were repeated.
After correcting for the drift the remaining scatter is
~10 pGal (LCR G 688, 1991), ~5 pGal (LCR G 445,
1992), and <10 pGal (Sodin 245GT, 1992).
Solid-earth tides were computed for the center of
the survey region (64° 15'N; 15°W). In 1991 the in-
strument drift sometimes obscured the tidal effects. In
1992 the instrument drift was generally so small that
the "observed drift" always reflected the tides.
After correcting the readings for earth tides, the
instrument drift was analyzed and the gravity differ-
ences (versus Pt. 701) were estimated again with the
aid of a gravity evaluation program by Peter Smilde,
Mainz (pers. comm. 1993). The program interpolates
34
JÖKULL, No. 44