Jökull - 01.12.1999, Blaðsíða 42

stress pattern to explain the entire fault population
of Iceland. In lack of fault and striae cross-cuttings,
their analysis is based on mechanical consistency but
not on chronological paleostress separation. Nor-
mal faults and strike-slip faults were analysed sep-
arately and interpreted with two stress regimes: ex-
tension parallel and perpendicular to the rift zone for
widespread normal faults, and compression parallel
and perpendicular to the rift zone for the sets of conju-
gate strike-slip faults. They showed that permutations
between 0\ and cr2 (maximum and intermediate com-
pressive stresses, respectively) induce changes from
normal to strike-slip faulting, as demonstrated by seis-
micity (Einarsson, 1979,1991).
From similar mesoscopic studies south of Langa-
vatn, Passerini et al. (1990) first showed that NNE
faults have dip-slip as well as dextral and sinistral
movements, and that dip-slip movements prevail on
WNW faults. After re-examination of their data
(Passerini et al, 1997) they concluded that NNE-
trending, mostly strike-slip faults predominate. They
presented two genetic schemes for the fault popula-
tions, similar to the stress regimes suggested by Berg-
evaietal. (1990).
In the present study, field analyses on large-scale
(regional) and small-scale (mesoscopic) planes are in
agreement with observations of aerial photographs.
Faults have variable strikes, but the fault pattern is
dominated by WNW and NNE faults (Figures 2 and
5a); both populations show normal and strike-slip
movements with similar length, displacements and
frequency. The dyke pattern is dominated by N-S
and NNE-trending dykes with the same frequency and
thickness. As Bergerat et al. (1990) analysed normal
faults and strike-slip faults separately, and in view of
mechanical relationships, in the present study an at-
tempt is made to interpret the paleostress conditions
of normal faults, strike-slip faults and dykes together,
and with respect to their cross-cuttings. The inter-
pretation is not based on computed stress tensors but,
similarly, the method consists of finding the best rela-
tionships between a stress field and the trends of the
mentioned fractures. The number of strike-slip faults
is, however, lower than in other studies, and dextral
and sinistral polarities were not systematically found
in the same sites. Though not all the data can be incor-
porated, four compatible subsets are suggested (Fig-
ure 9). Their order is not chronologically presented.
(1) Sinistral and dextral strike-slip faults trend
N10°-30°E and N90°-100°E, respectively, and fit
well with N140°-150°E normal faults, as well as with
NW striking dykes common below the unconformity.
They reflect a NE extension (0-3) and NW compres-
sion (cti), similar to Sæmundsson’s (1978) sugges-
tions.
(2) In this subset, sinistral and dextral strike-slip
faults trend N140°-170°E and N30°-40°E, respec-
tively, in good agreement with WNW and E-W nor-
mal faults as well as with WNW (and a few E-W)
dykes more common above the unconformity. They
support a N170°-180°E trending o3 and an E-W to
N100°E trending o\. This subset fits with N-S exten-
sion proposed by Einarsson et al. (1977).
(3) N60°-80°E sinistral and N0°-20°E dextral
strike-slip faults, along with N30°-40°E normal faults
and dykes, indicate a NW o3 and a NE o\, very near
to the rifting stress conditions.
(4) No dextral strike-slip fault is found among the
measurements for this subset which consists of N-S
dykes, a few N40°-50°E sinistral strike-slip faults,
and N-S normal faults. The directions of o3 and o\
are respectively E-W and N-S for this subset.
In first approximation, these fluctuations may be
interpreted in terms of stress field instability. An in-
teresting aspect is the mixture between rifting and
non-rifting conditions and the related fluctuations in
the direction of o3 and o-\. Part of these fractures
present cross-cuttings, as described earlier; they indi-
cate a complicated relationship between the four sub-
sets: dykes of subsets 3 and 4 are cut by dextral strike-
slip and normal faults of subsets 1 and 2. Dextral
strike-slip faults of subset 1 are younger than dykes
and normal faults of subset 3, and dextral strike-slip
faults of subset 3 are younger than WNW dykes and
normal faults of subset 2. The mutual cross-cutting of
WNW and NNE normal faults and dykes (subsets 2
and 3), is interpreted as evidence of their simultane-
ous activity, implying coexistence of inferred rifting
and non-rifting stress fields. This coexistence is also
reflected by either N-S and NNE dykes cut by other
40 JÖKULL, No. 47