Rit (Vísindafélag Íslendinga) - 01.06.1971, Síða 92
92
GUÐMUNDUR PÁLMASON
coefficients and differences in propagation lengths along the refractor.
This may in some cases assist in identifying the various refracted
waves in the seismograms.
The attenuation by scattering and absorption appears to be small
in most cases and difficult to detect, except where the structure is
regular and the waves can he observed over a distance of several tens
of kilometers. The main attenuation data come from the longer pro-
files (L1 to LIO). The P3-wave appears usually to be more attenuated
than the P4-wave. As an example may be mentioned the profile
Kerid-Hvítárvatn (L7), wheretheattenuation appears to be stronger
than on the other long profiles. The P3-wave appears to be attenuated
0.5 dB/km and the P4-wave 0.13 dB/km. Since the predominant fre-
quency of the refracted waves is about 10 c/sec, the attenuation per
wavelength will be 0.3 and 0.09 dB respectively.
O’Brien (1967a) has discussed attenuation of refracted waves and
divides the attenuation coefficients into three groups. These have
mean values of a few hundredths of a decibel per wavelength, several
tenths of a decibel per wavelength, and several decibels per wave-
length. The first group is, according to O’Brien, indicative of a thick,
nonporous refractor, the second group is indicative of a thick, porous
refractor, and the third group indicates a thin refractor. The terms
thick and thin are in reference to the wavelength.
Comparing the attenuation coefficients for the P3- and P4-waves
as obtained in Iceland with the above classification, it is found that
most of them fall within the first group, i.e. a thick, nonporous re-
fractor. In a few cases the P3-wave appears to have an attenuation
of a few tenths of a decibel per wavelength, suggesting “a thick,
porous refractor”.
No reliable data are available on the attenuation in the upper
layers 0, 1 and 2. The attenuation in these is probably larger than in
the deeper layers 3 and 4.
In many cases the attenuation coefficient appears to be negative,
i.e. the reduced amplitude increases with distance. In some cases this
may be caused by the effects of a structure, but it is also likely that
in some cases this may be caused by a small positive velocity gradient
in the refractor and a reinforcement of the head wave by a refracted
wave which is bent upwards.