Jökull - 01.12.1989, Page 47

Fig. 5. The gravity map
after a regional trend has
been subtracted from the
Bouguer anomaly. The
lighter shade marks areas of
gravity less than -6 milli-
gals, while the heavier
shade shows areas where
gravity values are above 6
milligals. Contour lines are
2 milligals apart.
Mynd 5. Þyngdarkortið eft-
ir að bakgrunnssvið hefur
verið dregið frá Bouguer
þyngdarkortinu. A Ijósgráu
svœðunum eru þyngdargildi
lœgri en -6 milligal, en á
dökkgráu svæðunum eru
þau hœrri en +6 milligal.
the space domain with a circular Hanning window
to reduce leakage, or edge effects, and a periodo-
gram is then calculated from the Fourier transform
of the data. The periodogram, without any smooth-
ing in the frequency domain, constitutes our esti-
mate of the power spectrum.
For a map with random values the spectrum con-
sists of concentric circles but for a surface
influenced in some particular direction there appear
radial peaks in the spectrum indicating the preferred
direction of lineaments in the map (Þórarinsson and
others, 1988). This allows us to put a quantitative
estimate on the lineation in a mapped surface by
studying its two-dimensional power spectrum.
Three fairly strong peaks appear in the spectrum
depicted in Figure 62. The direction of the largest
peak is 30-40° (clockwise from the horizontal),
reflecting the rifting zone from Hengill to Langjök-
ull. The second strongest peak points 70-80° and
reflects the Hengill-Reykjanes part of the rifting
zone. These two peaks merge at lower frequencies
(closer to the center of the spectrum) and together
they constitute the signal from the gravity low.
The third peak in the power spectrum points 130-
140° and represents lineation perpendicular to the
general trend of the rifting zone. This peak is notice-
ably weak at lower frequencies but quite prominent
at higher frequencies, indicating a fairly shallow ori-
gin. Its main sources are the alignment of the gravity
highs in the northeast (Hafnarfjall and Ferstikla) and
the Hekla-Hreppafjöll gravity low.
2. A fourth peak appears at high frequencies pointing straight
north and south, but this is probably a spurious signal due to
edge effects in the Fourier Transform.
JÖKULL, No. 39, 1989 45