ISVOLC: Deglaciation and GIA Affecting Crustal and Mantle Stresses in Iceland. Much More Magma?

The hypothesis of the ISVOLC project is that retreat of Icelandic glaciers since the end of the 19th Century has the potential to impact both volcanic and seismic activity. As volcanic activity increased significantly during (and in the <2kyr after) the late Pleistocene deglaciation, it is expected that present day deglaciation will once again effect volcanic activity in Iceland. The unloading of glaciers and subsequent rebound response of the Earth can significantly alter the state of stress in the crust and mantle. Within the ISVOLC project (https://isvolc.is) we have developed a new generation of Finite Element Glacial Isostatic Adjustment (GIA) models, using the COMSOL Multiphysics software package, which employ spaciotemporal estimates of glacier mass balance. Utilizing this new detailed ice history, we simulate GIA numerically and, once best fit earth parameters are found by utilizing InSAR and GNSS measurements, produce revised estimates for stress changes in the crust and mantle. From these we can calculate mantle decompression melting increases and shallow crustal stressing which may already be affecting volcanism and seismicity. We find that rates of total magma production beneath Iceland are enhanced by up to a factor of ~3 due to the glacier retreat induced decompression melting. However, it is highly uncertain when this additional magma will reach the surface and in what volumes. Stress changes around magma bodies at shallow level in the crust can bring such magma bodies closer to or further away from failure, depending on their geometry. Our new models predict stressing rates in the shallow crust that are comparable to those from tectonic extension for volcanic systems, seismic zones, and fissure swarms that are near or underneath Vatnajökull (Bardarbunga, Grímsvötn, northeastern Volcanic Zone, south Northern Volcanic Zone), with a unique pattern for each system. Glacially induced stressing in these areas may significantly shorten the timeline to seismic, diking, or eruptive events and alter the preferred orientations of dike propagation. Stressing rates from glacial mass loss are an order of magnitude smaller for systems beneath smaller glaciers and those not beneath ice (e.g. Askja, Katla, South Iceland Seismic Zone, Tjornes Fracture Zone), but still significant enough to consider when assessing hazard. Efforts in further improving the GIA modeling include effects of more realistic non-linear mantle rheology leading overall to somewhat higher viscosity estimates and more subdued GIA response in the far-field, as well as increases in the rate of magma production predicted by our models. Near-future work will involve the projection of glacial unloading and the subsequent earth response to evaluate effects on long-term hazards.