Numerous jökulhlaups have followed the river Jökulsá á Fjöllum in northeast Iceland during the Holocene. Some of these fall under the category of catastrophic floods that carved out the present-day Jökulsárgljúfur canyon, over 100 km north of the present-day Vatnajökull. Sedimentary beds from 16 jökulhlaups between 6.3 and 4.1 ka ago are preserved in Vesturdalur, near the north end of the canyon. Chemical composition of volcanic glass in the jökulhlaup deposits correlates to three volcanic systems beneath Vatnajökull. The characteristics of the Bárðarbunga volcanic system dominate in 12 beds, and those of Grímsvötn and Kverkfjöll in one bed each, two remain unsolved. The characteristics of the Bárðarbunga glass in the jökulhlaup sediments are mostly low TiO2 and high MgO (TiO2 \7.3 w\%). Seventeen basaltic “Low-Ti” tephra layers from Bárðarbunga have been identified in soil sections in north Iceland from this period. Grain characteristics of the tephra indicate phreatomagmatic origin. Dispersal maps confirm source area below northwest Vatnajökull and tephra volume (bulk) of the order of 1 km3 for the largest layers. As the preserved soil sections are distal (>50 km) from source, it is likely that only the largest tephra layers have been preserved. The mid-Holocene floods confirm the existence of glaciers on Bárðarbunga, Kverkfjöll, and Grímsvötn at that time. The magnitude of these jökulhlaups is not well constrained, but apparent cross sections out of Vesturdalur fit a peak discharge of order 50,000 m3 /s and likely total volume of a few km³. These repeated jökulhlaups 6.3 to 4.1 ka ago did not cause large erosion at Vesturdalur. Their source areas were most likely the calderas of the central volcanoes, which may have changed in size and form since the mid-Holocene. Eruptions within the Bárðarbunga caldera are a possible source for 12 of the floods. Bárðarbunga may have hosted a geothermal area and a subglacial caldera lake similar to present day Grímsvötn, which may explain the repeated, apparently similar-magnitude jökulhlaups over this long period.
Larsen et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: