50thKaikoura05 -1- Kaikoura 2005 CHARACTERISATION OF NEW ...
50thKaikoura05 -1- Kaikoura 2005 CHARACTERISATION OF NEW ...
50thKaikoura05 -1- Kaikoura 2005 CHARACTERISATION OF NEW ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
TRACKING CRUSTAL DIFFERENTIATION<br />
AND ASSIMILATION PROCESSES AT ARC<br />
VOLCANOES: A URANIUM SERIES<br />
ISOTOPE PERSPECTIVE<br />
R. George 1 ,S.Turner 1 ,R.Price 2 ,C.Cook 2 &B.<br />
Finney 3<br />
1 GEMOC Key Centre, Earth and Planetary<br />
Sciences, Macquarie University, Sydney<br />
2 School of Science and Technology, University of<br />
Waikato, Private Bag 3105, Hamilton<br />
3 Department of Earth Sciences, University of<br />
Bristol, Queens Rd, Bristol, BS8 1RJ, UK<br />
(rgeorge*els.mq.edu.au)<br />
Parent-daughter isotope systems that decay on time<br />
scales appropriate to the rates of processes<br />
themselves have proved powerful tools in tracking<br />
the histories of many arc-related processes from<br />
fluid release during plate subduction to dynamic<br />
melting of the mantle wedge. However, the rates<br />
inferred for mantle processes can only be<br />
considered real if the effects of continental crustal<br />
interaction are demonstrably minimal. In order to<br />
try and deconvolve mantle and crustal processes in<br />
this context we consider two contrasting case<br />
studies.<br />
Ruapehu volcano, New Zealand: On an equiline<br />
diagram the Ruapehu samples form a subtle<br />
positive array that extends from Taupo rhyolites<br />
and is intermediate between them and Kermadec<br />
lavas. It is conceivable from this data that a simple<br />
model of mixing between mantle-derived melts and<br />
Taupo crustal melts may explain the Ruapehu<br />
disequilibria. However, other reservoirs may be<br />
involved. For example, the range of calculated<br />
disequilibria derived from elemental Th/U ratios in<br />
averaged metasedimentary xenoliths, Torlesse and<br />
Waipapa basement averages could also be involved.<br />
Thus an alternative, although speculative, model is<br />
that Ruapehu samples reflect melts sourced from<br />
the lower to middle crust (as sampled in the<br />
xenolith suite), mixed with those generated at<br />
shallower levels (represented by Taupo-like<br />
magmas). Irrespective of the exact details of likely<br />
components, the key aspect of these systematics is<br />
that for the first time, there is sufficient information<br />
available to show that an inclined array on the<br />
equiline diagram for a single volcanic suite is<br />
unequivocally the product of open system<br />
processes: mixing between mantle or lower to midcrustal<br />
melts and shallow upper crustal<br />
components.<br />
Okmok volcano, Aleutian arc: At face value, 238 U-<br />
230<br />
Th disequilibria in young volcanics from Okmok<br />
suggest time scales of 50 kyr or less for some<br />
combination of mantle-crustal processes to take<br />
place. Given good evidence in the arc as a whole<br />
for fluid transfer from the subducted plate to take<br />
less than 10 kyr to reach the arc volcano (George et<br />
al., 2003), this additional time could reasonably be<br />
supposed to relate to a crustal process. Extremely<br />
good negative correlations between Sr and O<br />
isotopes are consistent with crustal assimilation of<br />
low � 18 O, hydrothermally-altered wallrocks. While<br />
the correlation between these two parameters and<br />
( 230 Th/ 232 Th) appears to be less coherent, it is<br />
striking that these activity ratios broadly increase<br />
with increasing Sr isotope ratios and decreasing<br />
� 18 O. The assimilant could therefore be<br />
hydrothermally-altered arc crust of relatively recent<br />
age of less than 100 kyr.<br />
ORAL<br />
THE SURFACE IMAGE <strong>OF</strong> AN ACTIVELY<br />
GROWING – INVERTED? – REVERSE<br />
FAULT: THE OSTLER FAULT IN THE<br />
MACKENZIE BASIN (SOUTH ISLAND, <strong>NEW</strong><br />
ZEALAND)<br />
F. C. Ghisetti<br />
Dept. of Geology, University of Otago, PO Box 56,<br />
Dunedin, New Zealand<br />
(francesca.ghisetti*stonebow.otago.ac.nz)<br />
A reappraisal of the geometry and structural<br />
evolution of the Ostler fault within the Mackenzie<br />
basin is under way. A new geological map of the<br />
fault zone (scale 1:50,000) is almost complete, and<br />
two seismic reflection lines across the fault will be<br />
acquired by A. Gorman in the early months of 2006<br />
(Otago University Research Grant). The Ostler fault<br />
is a 50 km long, N-S reverse fault, dipping 30º-50º<br />
W. Previous studies have documented the<br />
segmented nature of the fault, average rates of<br />
deformation �1 mm/yr, and the occurrence of two<br />
seismic events in the last 6 ka. According to<br />
Kleffmann and Stern (in preparation) the Ostler<br />
fault bounds the edge of a strong gravimetric<br />
gradient of residual Bouguer anomalies, with a<br />
minimum < -200 µN/kg west of the fault. This<br />
setting is interpreted in terms of a buried marine<br />
Tertiary basin, controlled by earlier normal faults.<br />
The deep geometry of the fault is largely<br />
unresolved, but stratigraphic and structural features<br />
revealed by surface mapping need to be considered<br />
in any interpretation.<br />
A >1000 m thick sequence of terrestrial mudstones,<br />
siltstones and gravels (Pliocene?-Pleistocene?) is<br />
exposed in the fault hanging wall only. Internal<br />
architecture reveals a growth geometry of beds<br />
tilted 30º-60º W, unconformably overlain by coarse<br />
gravels (120 ka?). Above, six orders of river<br />
terraces are preserved, with the three highest orders<br />
correlating to the 35, 22-17 and 14 ka moraines.<br />
Terrace distribution indicates shifting of faulttransverse<br />
drainages, relatable to episodic uplift of<br />
the upthrown block during progressive northward<br />
propagation of segment ruptures in the late<br />
50 th <strong>Kaikoura</strong>05 -29- <strong>Kaikoura</strong> <strong>2005</strong>