Jökull - 01.12.1994, Qupperneq 55

KOLMOGOROV-SMIRNOV TWO SAMPLE TEST OF CIRCULARITY
1156 1157 1161 1162 1200 1203 1204 1207 1267 1270 1277 1287 1418 1420 1422
MAXIMUM DIFFERENCES FOR PAIRS OF VARIABLES
1156 0.000
1157 0.244 0.000
1161 0.175 0.300 0.000
1162 0.075 0.238 0.156 0.000
1200 0.469 0.281 0.538 0.500 0.000
1203 0.563 0.369 0.613 0.575 0.131 0.000
1204 0.262 0.069 0.331 0.262 0.269 0.369 0.000
1207 0.475 0.275 0.538 0.488 0.125 0.125 0.256 0.000
1267 0.831 0.688 0.869 0.831 0.469 0.544 0.688 0.538 0.000
1270 0.450 0.294 0.500 0.481 0.094 0.150 0.294 0.162 0.419 0.000
1277 0.375 0.206 0.419 0.406 0.150 0.206 0.200 0.131 0.538 0.138 0.000
1287 0.569 0.381 0.606 0.581 0.131 0.100 0.381 0.150 0.469 0.119 0.200 0.000
1418 0.113 0.169 0.144 0.125 0.412 0.506 0.219 0.406 0.744 0.394 0.325 0.488 0.000
1420 0.163 0.263 0.063 0.144 0.519 0.594 0.300 0.513 0.850 0.494 0.419 0.594 0.131 0.000
1422 0.281 0.094 0.363 0.269 0.269 0.375 0.069 0.263 0.700 0.300 0.206 0.387 0.238 0.344 0.000
TWO-SIDED PROBABILITIES
1156 1.000
1157 0.000 1.000
1161 0.015 0.000 1.000
1162 0.759 0.000 0.040 1.000
1200 0.000 0.000 0.000 0.000 1.000
1203 0.000 0.000 0.000 0.000 0.127 1.000
1204 0.000 0.844 0.000 0.000 0.000 0.000 1.000
1207 0.000 0.000 0.000 0.000 0.164 0.164 0.000 1.000
1267 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000
1270 0.000 0.000 0.000 0.000 0.483 0.055 0.000 0.029 0.000 1.000
1277 0.000 0.002 0.000 0.000 0.055 0.002 0.003 0.127 0.000 0.097 1.000
1287 0.000 0.000 0.000 0.000 0.127 0.400 0.000 0.055 0.000 0.209 0.003 1.000
1418 0.263 0.021 0.073 0.164 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.000 1.000
1420 0.029 0.000 0.913 0.073 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.127 1.000
1422 0.000 0.483 0.000 0.000 0.000 0.000 0.844 0.000 0.000 0.000 0.002 0.000 0.000 0.000 1.000
Table 7. Kolmogorov-Smirnov two sample test of circularity. Values below 0.05 indicate significant difference between
respective samples at 95 % level of confidence.
7. tala. Kolmogorov-Smirnov prófun á samanburði meðaltalshringlögunar allra sýna. Lœgri gildi en 0,05 sýna marktœkan
mun á hrjúfleika við 95 % öryggismörk.
cant variation in the grain shape within the dataset.
Very irregular shape and jagged tephra grain outlines
result when fragmentation takes place under highly
evolved degassing and marked frothing coupled with
the expansion of magma during eruption. The strom-
bolian Laki sample (Sl) and the brown layer samples
from R-3 on Reykjanes are good examples of this.
The S1 sample from Laki lies at one end of the spec-
trum in this respect, but the P1 sample from Laki and
the brown layer samples represent intermediate val-
ues, indicating considerable expansion before quench-
ing upon contact with water.
The tephra classification scheme which is based
on grain size parameters (Walker and Croasdale,
1972) is of limited value in interpreting details in the
origin of tephra units. As an example it does not indi-
cate the influence of water on the initial phase of the
Laki eruption (Pl). Neither does it indicate the degree
of degassing of surtseyan tephra at the disintegration
stage (cf. P1 and the brown layer of R-3), which is in-
dicated by the image analysis of grain shape as well
as by the SEM images. The comparison of samples
from RG-2 indicates a change from surtseyan to
strombolian activity.
We conclude that grain shape studies supported by
SEM images are important complementary tools in
evaluating the origin of tephra and for monitoring
changing conditions at the site of eruption.
ACKNOWLEDGEMENTS
The Geological Survey of Denmark kindly allowed
the use of the Image Analysis System. Dr. K. L. Knud-
sen at the Geological Institute of the University of
Aarhus, Denmark, is thanked for making the SEM fa-
cilities available to the authors, and Svend Meldgaard
at the Geological Institute of the University of Aarhus,
Denmark, is thanked for assisting with photographic
work. Dr. Þ. Þórðarson is thanked for making the Lak-
agígar samples and grain size data available.
JÖKULL, No. 44
53