Jökull - 01.12.1980, Blaðsíða 50

NUMBER
Fjöldi
Fig. 5. Orientation of all fractures inside the Vogar fissure swarm. N is number of
measurements, X is the mean and S.D. is the standard deviation. — Mynd 5. Stefna allra
sprungna á Vogasvœðinu. N. er fjöldi mælinga, X er meðalgildið og S.D. er staðalfrávikið.
depends on how the limits between the classes
are chosen, and what rules are applied to
values that lie on the borderline between two
classes (columns). In this paper I have fol-
lowed the rules given by Kreyzsig (1972); a
value on the borderline of two classes is always
included in the higher class.
To test if a significant change in the orien-
tation of the fractures occurred along the fiss-
ure swarm, I measured the orientation of all
the fractures, and parts of fractures, inside
columns 1 and 5 (Fig. 3). Then I put forward
the null hypothesis that the mean orientation
was the same in both cases, i.e. HO: U5 = U1,
and the alternative hypothesis Hl: US^Ul.
The 0.05 level of significance was selected, and
a t-test used, for small samples (n<30), accord-
ing to Spiegel (1975). Admittedlý, column 1
includes 34 fractures, but as the test for- small
samples is more accurate than for large ones, I
decided to use the former.
The calculations give T = 0.54, which is well
within the limits -2.01 to 2.01 (Spiegel 1975),
hence H0 is not rejected. This means that
there is no significant difference in the mean
orientation of the fractures in column 5 and
column 1. Accordingly, I conclude that there
is no significant change in the orientation of
the fractures along the fissure swarm.
Length of fractures
The average length of those 120 fractures,
that lie completely inside the map (Fig. 2), is
611 m. The histogram in Fig. 7 shows the
length distribution. As said before, the length
is always linear; i.e. the distance between the
end points of each fracture is measured. In
some cases it is debatable what should be
regarded as one and the same fracture. I used
the following rule: A fracture is counted as one
as far as it can be followed as a continuous
fissure and/or fault. All minor fractures,
whether or not associated with major frac-
tures, are taken as separate fractures.
The linear correlation between length and
other parameters is usually small (Table 2).
There are only two reasonable correlations;
between length and maximum dilation
48 JÖKULL 30. ÁR