Faulting processes during early-stage rifting: seismic and geodetic analysis of the 2009–2010 Northern Malawi earthquake sequence

Abstract

SUMMARY In December, 2009, a rare sequence of earthquakes initiated within the weakly extended Western Rift of the East African Rift system in the Karonga province of northern Malawi, providing a unique opportunity to characterize active deformation associated with intrabasinal faults in an early-stage rift. We combine teleseismic and regional seismic recordings of the largest events, InSAR imagery of the primary sequence, and recordings of aftershocks from a temporary (4-month) local network of six seismometers to delineate the extent and geometry of faulting. The locations of ∼1900 aftershocks recorded between January and May 2010 are largely consistent with a west-dipping normal fault directly beneath Karonga as constrained by InSAR and CMT fault solutions. However, a substantial number of epicentres cluster in an east-dipping geometry in the central part of the study area, and additional west-dipping clusters can be discerned near the shore of Lake Malawi, particularly in the southern part of the study area. Given the extensive network of hanging wall faults mapped in the Karonga region on the surface and in seismic reflection images, the distribution of events is strongly suggestive of multiple faults interacting to produce the observed deformation, and the InSAR data permit this but do not require it. We propose that fault interaction contributed to the seismic moment release as a series of Mw 5-to-6 events instead of a normal main shock–aftershock sequence. We find the depth of fault slip during the main shocks constrained by InSAR peaks at less than 6 km, while the majority of recorded aftershocks are deeper than 6 km. This depth discrepancy appears to be robust and may be explained by fault interaction. Structural complexities associated with fault interaction may have limited the extent of coseismic slip during the main shocks, which increased stress deeper than the coseismic slip zone on the primary fault and synthetic faults to the east, causing the energetic aftershock series. There is no evidence of deformation at the Rungwe volcanic province ∼50 km north of the earthquake sequence between 2007 and 2010, consistent with previous interpretations of no significant magmatic contribution during the sequence.

Controls on Rift Faulting in the North Basin of the Malawi (Nyasa) Rift, East Africa

“…In the Malawi rift, intrarift fault activity has also been proposed to diminish with time based on stratigraphic mapping in seismic reflection data (Mortimer et al, ). However, seismic reflection data do indicate continued activity on some intrarift faults based on offsets of shallow, young horizons (Mortimer et al, ), consistent with seismicity associated with intrarift faults (Biggs et al, ; Ebinger et al, ; Gaherty et al, ). In the southernmost Malawi Rift, intrarift faults accommodate ~50% of recent extension (Wedmore et al, ).…”

“…Onshore mapping and focal mechanisms imply relatively steep dips of the Livingstone Fault and other border faults in the Western Rift of 60–70° (Ebinger, ; Foster & Jackson, ; Wheeler & Rosendahl, ). Focal mechanisms and modeling of interferometric synthetic aperture radar data from the 2009 Karonga earthquake series that ruptured one or more intrarift faults in 2009 yield relatively shallow dips of ~40° (Biggs et al, ; Gaherty et al, ), though aftershocks from these events appear to delineate features with steeper dips of ~60° (Gaherty et al, ). To estimate the total extension accommodated on faults, we assume a dip of 60°, a relatively high value that results in conservative (low) estimates of heave.…”

How Do Variably Striking Faults Reactivate During Rifting? Insights From Southern Malawi

“…Given that these rocks have been dehydrated during one or more episodes of high grade metamorphism (Fagereng, ), fluid ingress into these fault zones is likely to be limited (Hollinsworth et al, ). Indeed, even in northern Malawi where sediments are ~500‐m thick (Kolawole et al, ), there is little fluid involvement in active faulting (Gaherty et al, ). Low fluid pressures around these faults is also illustrated by the lack of extensive vein networks observed in their damage zones (Figure ; cf.…”

“…Analysis assumes Φ = 0.43 and density model in Table S2. limited (Hollinsworth et al, 2019). Indeed, even in northern Malawi where sediments are~500-m thick (Kolawole et al, 2018), there is little fluid involvement in active faulting (Gaherty et al, 2019). Low fluid pressures around these faults is also illustrated by the lack of extensive vein networks observed in their damage zones ( Figure 5; cf.…”

Active Fault Scarps in Southern Malawi and Their Implications for the Distribution of Strain in Incipient Continental Rifts

“…Faults have been mapped in Lake Malawi by a number of seismic reflection surveys carried out in the lake (Lyons et al, ; Mortimer et al, ; Scholz, ; Scholz & Rosendahl, ; Shillington et al, ) as well as a recent network of lake bottom and onshore seismometers in northern Malawi and Tanzania (Accardo et al, ; Shillington et al, ). These seismic surveys, along with the 2009 Karonga earthquake sequence that occurred within the hanging wall of the rift‐bounding Livingstone fault, indicate that both rift border and intrabasin faults are currently active at the northern end of the Malawi Rift (Biggs et al, ; Gaherty et al, ; McCartney & Scholz, ). Active faulting is also occurring at the southern end of the lake, as demonstrated by the M w 6.1 1989 Salima earthquake (Jackson & Blenkinsop, ).…”