Jökull - 01.01.2010, Side 95

Geothermal noise at Ölkelduháls, SW Iceland
spectral peaks in the frequency range between 6 and 7
Hz that are not stable in frequency at stations LB3 and
LB5 at the beginning of the observational period and a
peak at ∼5 Hz at stations LB7, LB8 and LB9. These
spectral peaks are generally very sharp in frequency
and therefore probably related to some resonance phe-
nomenon that is unlikely to be natural because of how
narrow they are. These peaks contain very little inte-
gral power.
LA2 0.020
LA3 0.006
LA4 0.018
LA5 0.012
LA6 0.024
LA7 0.013
LA8 0.041
LA9 0.051
2.10 2.15 2.20 2.25 2.30 2.35 2.40
a
m
p
li
tu
d
e
s
p
e
c
tr
a
l
d
e
n
s
it
y
[
a
rb
.
u
n
it
s
]
frequency [Hz]
Figure 6. A blown up figure of the sharp amplitude
spectral peak consistently detected along line A and
parts of line B (spectra along line A only shown here).
Each spectrum has had a regional polynomial trend
removed and is then normalized to its maximum,
which is indicated to the right (in arbitrary, but iden-
tical units). This spectral peak is very confined in fre-
quency and highly variable in amplitude. – Stækkuð
mynd af skörpum róftoppi sem mælist reglubundið
eftir báðum mælilínum (hér eru aðeins sýnd róf eftir
línu A). Í hverju tilviki hefur bakgrunnshneigð rófsins
verið fjarlægð. Hámarksútslag hvers topps er skráð
til hliðar.
We compute the root-mean-square amplitude of
the spectrum for each station over the frequency range
between 3 and 7 Hz in order to characterize the overall
amplitude behavior of the data. The results are plot-
ted in Figure 7. There it is evident that the amplitude
drops quite significantly and systematically with dis-
tance from the geothermal field at Ölkelduháls. The
amplitude is highest at station LB1 and falls by three
orders of magnitude to the most distant stations LA7,
LA8 and LA9 at 3 km distance to the ENE. The drop
in amplitude is also dramatic to the WSW, although
not monotonic at the most distant stations LB7, LB8
and LB9. We have labeled the deviations from mono-
tonic amplitude decay away from the centre of the
profile in the figure with letters A, B and C. We spec-
ulate that at anomaly A at station LA4 may be due
to proximity to the geothermal field under the SW
end of Hrómundartindur. It is also possible that the
geothermal area in Grændalur kicks in here. We note
that the database of surface geothermal activity taken
from Árnason has a point near station LA4, although
no activity was visible at the time of observation. This
small peak may be related to that. The anomalies at
the WSW end of the profile are more difficult to re-
late to the surface activity. Steaming activity is found
some 200 m west of stations LB7, LB8 and LB9, but
this should affect station LB8 in a similar manner to
LB7 and LB9. The geothermal activity in Klambragil
may also affect the noise. Dispite these complications
it seems safe to associate the noise in the 3–7 Hz range
with the geothermal activity on the basis that ampli-
tude generally decays away from the crossover of the
profile with a roughly linear zone of geothermal activ-
ity. The stability of this noise over time suggests that it
is well determined and that variations in space and ori-
entation are well determined complications, not due
to lack of precision. Inconsistency in spectral shape
from one station to another, as well as from one com-
ponent to another, suggests that the noise source is
distributed.
POLARIZATION
Some of the above-mentioned differences may be at-
tributed to dominance of different wave types in the
wave field or scattering effects. To address this we
JÖKULL No. 60 95