LEVELING SURVEY OF KOA’E FAULT SYSTEM DURING THE 2021 KILAUEA DIKE INTRUSION

The Koa’e fault system is located south of the summit caldera between the southwest and east rift zones of Kilauea volcano. A series of line-leveling and crack measurement stations were established by the United States Geological Survey - Hawaii Volcanoes Observatory (USGS-HVO) scientists in the 1960s and have been monitored periodically since, using the same methodology. We measured elevation differences preceding the September 2021 summit eruption and compared them to similar measurements made in 2019 in order to refine the amount and extent of surface deformation associated with the dike emplacement.

Three months after Kilauea volcano stopped erupting into a lava lake in the Halema’uma’u Crater, earthquake activity and ground deformation increased in the southern summit caldera between August 23rd and August 25th, 2021. This activity indicated a small magmatic intrusion south of the summit caldera. 478 earthquakes were detected, most between M1-2, at approximately 1-2km below Earth’s surface. For weeks following the dike intrusion, no eruption occurred. On September 29th, 2021, Kilauea volcano resumed its eruption within Halema’uma’u Crater.

We measured a maximum of 15 centimeters of differential uplift since the 2019 survey. This deformation is likely mainly the result of dike emplacement in August 2021. The amount of uplift decreased away from the inferred axis of the intrusion, and an abrupt end to the deformation was observed at a pre-existing fault located approximately 6000ft (1800m) to the south. Although we did not measure any differential motion along pre-existing faults in the area, their location seemingly affected dike propagation. The style of deformation observed in 2021 is in stark contrast to the deformation style associated with the 2018 Lower East Rift Zone eruption and partial summit caldera collapse, during which there was significant differential motion along faults. These measurements should provide additional constraints for models of intrusion size.