by Michael P. Poland, Hawaiian Volcano Observatory (USGS)
In the past decade, synthetic aperture radar interferometric (InSAR) has enjoyed increasing use as a tool for detecting and characterizing surface deformation associated with volcanoes, earthquakes, glaciers, and other geological processes. Though InSAR can only image deformation that occurs along the radar line-of-sight and is subject to atmospheric, orbital, and other errors that can be difficult to quantify, the method has the advantage of high spatial resolution (especially in arid, unvegetated environments) without requiring equipment on the ground. As a result, InSAR is extremely useful for mapping deformation in poorly accessible or unmonitored parts of the world.
One such area is the Kivu Basin of East Africa. Located along the border of the Democratic Republic of the Congo and Uganda/Rwanda, the Kivu Basin is part of the East African Rift System, an area of abundant volcanic and seismic activity that has formed as a result of tectonic extension between East Africa and the rest of the continent. The basin is the site of Lake Kivu and the city of Goma (population > 400,000), which lies on the northern shore of the lake. This population is at risk from two volcanoes; Nyiragongo and Nyamulagira. Nyiragongo’s summit is only 18 km north of Goma and has been the source of eruptions of unusually fluid lava that caused hundreds of fatalities. Northwest of Nyiragongo, 15 km distant, is Nyamulagira volcano, a massive shield that erupts lava flows every 3 years on average, though seldom into populated areas.
On January 17, 2002, Nyiragongo erupted with little precursory warning. Cracks opened between the summit of the volcano and Goma, and within a few hours, lava had inundated a heavily populated portion of the city, entered the lake, and formed a small delta. The eruption was over by the next day, but about 15% of Goma had been destroyed, 70-100 people were killed (mostly by indirect causes associated with the eruption), and 120,000 left homeless.
Because there was little seismic monitoring and no terrestrial geodetic data (for example, from GPS), the cause of the eruption was difficult to assess. SAR scenes from RADARSAT-1 however, captured the region before and after the eruption. A comparison of numerous pre-and post-eruption scenes shows widespread deformation in the entire Kivu Basin, not just displacements focused in the vicinity of Nyiragongo (Figure 1). The InSARderived deformation is consistent with a pattern of tectonic rifting of the entire region and a model of the eruption (based on fracture patterns, eruptive chronology, geochemistry, etc.) that proposes rifting as a trigger for the activity. Magma contained in a summit lava lake, which had been crusted over since the late 1990s, apparently drained into a set of fractures that had last been active in 1977, mixed with fresh subsurface magma, propagated the fractures down the south flank of the volcano, and erupted at several points between the volcano’s summit and Goma.
Given the importance of the InSAR results for characterizing volcano-tectonic activity in the region (especially in the absence of other geodetic data), a program of regular InSAR monitoring of both Nyiragongo and Nyamulagira was initiated using RADARSAT-1 and ENVISAT SAR acquisitions. In the years since the January 2002 eruption of Nyiragongo, no further rifting events have taken place, and no deformation has been observed at Nyiragongo, despite renewed lava lake activity and episodes of intense seismic activity.
Nyamulagira volcano has erupted twice (in August 2002 and May 2004) since that time. In both instances, no pre-eruptive deformation has been observed in InSAR data, though co-eruptive deformation localized to the vicinity of the eruptive vent is apparent in image pairs that span the eruptions (Figure 2). Such deformation is probably associated with the shallow magma feeder system. The lack of precursory deformation suggests that magma rises quickly beneath Nyamulagira and that there is little, if any, monthslong accumulation of magma in a subsurface reservoir.
Despite the fact that no deformation prior to eruptions has yet been detected in the Kivu Basin, it is important to continue the program of InSAR monitoring. The method is by far the most effective for measuring surface displacements in the region and offers the best chance of detecting any precursory deformation to future volcanism or tectonism, in addition to characterizing displacements associated with past activity. InSAR results from the Kivu Basin exemplify the utility of such data in understanding volcano-tectonic processes and argue for continued application in the Kivu Basin and elsewhere, especially where other deformation measurements are lacking.
Click here to download a copy of the newsletter featuring this article