The geochemistry of volcanic, plutonic and turbiditic rocks from Sumba, Indonesia

Rocks that reveal the geology of Sumba for times before the Later Miocene ( similar to 16 Ma) are relatively few and are not particularly well exposed. This has led to uncertainty about the nature of the basement rocks of the island and especially about whether Sumba originated as a fragment of Austr alia, or of that part of southeastern Eurasia which many authors have calle d Sundaland. A third possibility is that Sumba is underlain by are material generated on the ocean floor and is not a fragment of either continent. We have studied the geochemistry of volcanic, plutonic and turbiditic rocks c ollected from Sumba in an attempt to provide additional insight into the is land's origin and history between Late Cretaceous (similar to 86 Ma) and Ea rly Miocene (similar to 16 Ma) times. Late Cretaceous to Early Oligocene (s imilar to 31 Ma) volcanic rocks on Sumba range compositionally from basalts to andesites, acid are of typical oceanic island-are affinity, exhibiting geochemical characteristics similar to those of high-Al basalts and their d erivatives. Compositions indicate evolution along both calc-alkaline and th oleiitic trends. Some samples show indications of possible modifications by slab-derived melts and/or related fluids and also of contamination by turb iditic sediments. Gabbros and diorites collected from the Paleocene Tanadar o intrusion are compositionally similar to the associated volcanic rocks an d. we consider. represent the plutonic equivalents of high-Al basalt. The g eochemistry of Cretaceous turbiditic sedimentary rocks on Sumba indicates c lose proximity to an intra-oceanic island-are environment. These results ar e consistent with the geochemical, sedimentological, stratigraphic, paleont ological and paleomagnetic results of other investigators which together in dicate that: (1) Late Cretaceous to Early Oligocene volcanic. plutonic and volcaniclastic rocks of Sumba are island-arc- and forearc-related; (2) the are involved appears to have been what we refer to as the Great Indonesian Volcanic Are, which had been active in this area from at least similar to 8 6 to approximately similar to 31 Ma. The Great Indonesian Volcanic Are was closely affiliated with, but may have occurred offshore Sundaland. Sumba is therefore a fragment of this oceanic (Aleutian-type) island are and not a piece of the main Sundaland continent. Thr geochemistry of the igneous rock s on Sumba is inconsistent with what we would expect from an Andean-type Co ntinental Are. No continental basement is required to explain the compositi on and origin of the igneous rocks of Sumba. (C) 2001 Elsevier Science Ltd. All rights reserved.