Jökull - 01.12.1982, Síða 119

data and methods
In this study we have followed three lines of
approach. We have searched through written re-
cords about the earthquakes, interviewed inhabit-
ants of the earthquake zone, and studied aerial
photographs.
Historical accounts mention surface breakage as-
sociated with a few earthquakes, but the descrip-
tions are generally short, and only exceptionally
suffice to locate faults in the field. Most accounts
prior to 1900 were compiled by Thoroddsen (1899 and
1905), who mentions faulting in association with
earthquakes in the years 1294, 1308, 1339, 1391,
1630, 1784, and 1896. The 1896faultingisdescribed
in considerable detail, but in most of the other cases
faultingis only mentioned briefly. In addition, fairly
thorough descriptions of fáulting are available in
newspaper reports of the earthquake of 1912.
Interviewing the inhabitants turned out to be a
very effective way ofsearching for earthquake fract-
ures. Farmers generally know their property well,
and could guide us to structures that fitted our
descriptions, even though in some instances they
did not realize that these were related to past earth-
quakes.
Studying aerial photographs is the most fruitful
method. Many of the larger fractures and fissures
show up on low and intermediate altitude photo-
graphs available at the Icelandic Geodetic Survey.
After a systematic inspection of the photographs,
the likely localities were checked in the field. When
a fracture had been located, one usually found it to
be a part of a fracture system that could be traced
over some distances. The most prominent systems
were then mapped by students using compass and
tape measure.
MAIN CHARACTERISTICS
OF THE FRACTURES
The appearance of the fractures depends strongly
on the nature of the surfaccmaterial where they are
exposed. Most of the area studied here is covered by
lavas, moraines and glaciofluvial deposits. The
fractures are most conspicuous where lava is cover-
ed by a thin layer of soil and vegetation. The faults
become obscured where the loose material is very
thick. Surface conditions also determine how well
the fractures are preserved. It is, for example, a
hopeless task to trace the faults in the sand filled aa
lavas that cover large partsofthe Land district. Wet
marshes also tend to obliterate the fault traces.
Fig. 3. Major faults and fracture systems of the
earthquake of May 6th 1912.
Mynd 3. Helstu sprungur og sprungukerfi frájarbskjálft-
anum 6. maí 1912.
In spite of their variable appearance, there are a
few features characteristic of most of the faults, that
make them identifiable in the field. Cases of doubt
JÖKULL 32. ÁR 115