Figures S1 and S2 show the logs of each individual exposure following the northern edge of channel A up to the fault zone. Logs east of the fault associated with trench 4 are shown in Figure S1, and Figure S2 shows the logs west of the fault, associated with trench 7. In each figure, the logs are arranged in sequential order from farthest away from the fault to closest to the fault. Locations of radiocarbon samples are shown in exposures 4H, 7A, and 7J. Sample numbers are followed by ages in calibrated years B.P., with two-sigma age ranges (Table 1). In addition to the fault-parallel exposures, the fault perpendicular exposures created as the result of sequential slicing-back parallel to the fault are shown as the last log on each figure (4H and 7K), reversed so the view is northward. The faults documented in exposures 4H and 7K could be the result of ongoing creep, could represent postchannel A surface rupture, or both. The approximate location of each exposure is shown on Figure 7 of the main article, and measured distances perpendicular to the trench wall from reference points in trenches 4 and 7 are given on each log. The reference point for exposures 4A–4G is the nail at grid point AA-6 on the west wall of Trench 4, and the reference point for exposures 7B–7J is the nail at grid point AA-2. Measurements were made perpendicular to the trench walls with a measuring tape, and reference nails were surveyed by differential GPS.
We traced the southern edge of gravel III (channel B) in map view to the fault from trench 4 (Fig. S3). Our initial field interpretation suggested that the offset equivalent east of the fault would be gravel IV, exposed in trench 7, and we also traced the southern edge of this unit to the fault from trench 7 (Fig. S4). However, subsequent radiocarbon analyses show that gravels III and IV are not correlative (Table 1 in the main article). Nevertheless, channel B must have crossed the fault and been offset, and we attempted to put some constraints on its offset based on the relations exposed in trenches east of the fault. However, radiocarbon dates of sediments exposed in trench 7 show that this section is everywhere significantly younger than channel B. Samples 130, 131, 139, 140, 141, and 142, collected in trench 7 from the sediments into which gravel IV is incised, show that no sediments remain that are older than about 1000 years B.P. (Table 1 and Fig. 6 in main article). In contrast, channel B is at least 2500 years B.P. (Table 1 in main article). East of the fault, the only Holocene sediments as old as channel B are exposed in trench 3 (Table S1), approximately 100 m southeast along the fault from the point where channel B intersects the fault. The fact that only young sediments are exposed east of the fault in all other trenches means that we cannot put any constraints on the offset of channel B, and we conclude that channel B east of the fault was destroyed by erosion during and/or prior to the formation of channel A.
Each of the two methods used to trace the channel edges into the fault has advantages and disadvantages. The sequential-slices method used for channel A provides a higher degree of confidence that we are matching corresponding points across the fault because we can see the entire cross section of the feature in each slice. However, this method has the disadvantage that the feature being followed is destroyed in the process of exposing it. The map-view method used to trace channel B largely preserves the feature. However, there is greater uncertainty that corresponding parts of the channel are being matched across the fault. Which method is best to use depends on the particular geometry of the channel, as well as on other considerations. For example, since this site was slated to become a housing development shortly after our study was completed, we did not feel there was any real loss in destroying the edge of channel A as we followed it to the fault, because the feature would no longer be available for study after the construction of the housing development.
All logs of southern walls are reversed, so the view is northward in all figures. We excavated trench 1 (Fig. S5) across the projected trace of the Maacama fault in the surface into which Haehl Creek is incised. There is no surface expression of the fault at this location, and the exact position of the fault was unknown prior to excavation. During excavation, we encountered an extensive, > 3 m thick gravel unit that was unstable, and we therefore were compelled to bench the excavation in the region underlain by the gravel as a safety measure (Fig. S5a,b). This gravel was designated gravel II and is part of channel A. Because trench 1 is oriented subparallel to the channel, the gravel appears along a very wide expanse of the trench wall (more than 12 m), even though the actual width of the channel in cross section is only about 5–6 m (Fig. 6 in main article). Logs of both the north wall (Fig. S5a) and south wall (Fig. S5b) are presented here, in addition to a detailed log showing the fault zone exposed in the north wall (Fig. S5c). The entire south wall of the trench was logged, and the easternmost part of the north wall was not logged.
Trench 8 (Fig. S6) was excavated across the fault about 60 m northwest of trench 1 (Fig. 5 in main article), to re-expose the channel observed in trench SHN-1. This is channel B. Logs of both the north wall (Fig. S6a) and south wall (Fig. S6b) are presented here.
Trench 3 (Fig. S7) was excavated within the compressional stepover, about 30 m southeast of trench 1 (Fig. 5 in main article). This trench exposed Pleistocene lakebeds between two reverse oblique faults that juxtapose the lakebeds over Holocene fluvial sediments. Only the south wall was logged. The log is reversed, so the view is northward. Radiocarbon data is given in Table S1. The westernmost fault terminates below the surface and may be related to the Willits surface rupture. Two radiocarbon samples from the unfaulted unit above the fault, samples 74 and 75, give ages of 2740–2850 cal B.P. and 920–1050 cal B.P., respectively (Table S1). These two samples were collected very close to one another, and the age discrepancy therefore most likely indicates that sample 74 is reworked and sample 75 is closer in age to the time of deposition. If so, then the age of sample 75 can be used to estimate the minimum age of the earthquake. Sample 72, collected from the faulted horizon, gives an age of 1550–1770 cal B.P. (>Table S1) and can be used to constrain the maximum age of the earthquake. Given these relations, the earthquake occurred 920–1770 cal. B.P., a range that permits this to be the Willits surface rupture (1060–1180 cal B.P.) but provides no better constraints on its age.
Trench 5 (Fig. S8) was excavated parallel to the fault zone east of the fault in order to better understand relations among units exposed in trenches 1, 6, and 2. Only the western wall was logged.
Trench 4 (Fig. S9) was excavated parallel to the fault zone west of the fault in order to find the northern edge of gravel I and to identify additional possible targets to trace across the fault zone. The entire eastern wall was logged, but only the southern end of the western wall.
Trench 7 (Fig. S10) was excavated parallel to the fault zone east of the fault in order to find the northern edge of gravel II and to search for the offset equivalent of gravel III. The entire western wall and only the southern end of the eastern wall were logged, (Fig. S2)
Figure S11 shows details of the northern wall of trench 2. Figure S11a shows the monocline developed over the western fault trace exposed in trench 2, and Figure S11b is a close-up of the western trace (expressed here as a reverse fault) and the buried fault scarp associated with the most recent surface rupture.
Trench 3 exposed reverse faults (Fig. S12a,d) that juxtapose Holocene sediments over Pleistocene lake beds. The Pleistocene bedding exposed in the trench is nearly vertical and contains several tephra units (Figs. S12b,c), indicating that the compressional stepover is a long-term feature of this part of the Maacama fault.
The fold that we document in the stepover does not have any surface expression. This, we surmise, is because when Haehl Creek was flowing across the surface, it continually removed the fold’s surface expression by erosion. The most erosion appears to have occurred in the area around T3, where the Pleistocene lake beds are nearly at the surface. Not until Haehl Creek abandoned the surface could the fold begin to emerge and be expressed in the topography. Because the surface was abandoned only in the last 500 years, no surface expression has yet developed, although this will likely change in the future as long as Haehl Creek remains incised. Our investigation of the property indicates that agricultural activities have occurred and that it was at one time used as a landing strip. However, the family who owned the property for several generations told us that no significant land surface modification had been done at the time we conducted our study. It is possible that a small surface expression of the fold existed prior to the land use of the last 100 years; however, given the limited land modification, it could not have been a large feature.
Table S1. Radiocarbon analysis of charcoal collected from south wall of trench 3 (see Fig. S3 for sample locations).
Figure S1. Logs of trench 4, slices A–H. Log A is of the upper part of the east wall of trench 4 (Fig. 6b in main article). Logs 4B–G are fault-parallel logs of incremental slices excavated by hand to follow the northern edge of channel A (gravel I) to the fault (see Fig. 7 in main article). The shaded rectangle in log 4B indicates the area excavated to create exposures 4C–H. The channel edge encounters a fault zone between logs 4D and 4E, which are 30 cm apart. Distances from the reference nail (RN, which is nail AA-6 on the western wall of trench 4; log not shown) are given for each log. Log 4H is the fault perpendicular exposure created by the southern wall of hand excavation (reversed so the view is northerly). Note radiocarbon sample 144, located on log 4H.
Figure S2. Logs of trench 7, slices A–K. Logs 7A and 7B are parts of the east and west walls of trench 7. The shaded rectangle in log 7B indicates the area excavated by hand to create exposures 7C–K. Logs 7C–J are logs of incremental slices excavated by hand to follow the northern edge of channel A (gravel II) to the fault (see Fig. 7 in main article). The channel edge encounters a fault zone between logs 7H and 7I, which are 30 cm apart. Distances from the reference nail (RN, which is nail AA-2, shown on log 7A) are given for each log. Log 7K is of the fault perpendicular exposure created by the southern wall of the hand excavation (reversed so the view is northerly). Note radiocarbon sample 102, located on log 7A, and sample 146, located on log 7J (Table 1 in main article).
Figure S3. Gravel III (channel B) is traced in map view to the fault. (a) The annotated photograph shows the northern part of trench 4 and the hand-excavated, map-view exposure of the southern edge of gravel III (channel B) between trench 4 and the fault zone. The shaded blue region indicates exposed gravel. The string grid is 0.5 m square. (b) Log of the floor of the excavation showing a map view of the gravel III channel edge.
Figure S4. Gravel IV is traced in map view to the fault. (a) Photo of T7 and hand-excavated map view exposure of gravel IV. The black lines mark the edges of the excavations. Red lines show faults and are dashed lines where faults are approximated. Green shading shows gravel IV exposed in the T7W wall and in the map view. The intersection of the channel edge and fault is cut away by the inset trench. The string grid is 0.5 m square. Blue flags on the wall of T7W mark the locations of radiocarbon samples. (b) Log of the floor of the excavation, showing a map view of the gravel IV channel edge intersecting the fault zone.
Figure S5. Logs of trench 1, showing the locations of the radiocarbon samples: (a) trench 1 north wall, (b) trench 1 south wall (reversed so the view is northward), and (c) detail of the fault zone exposed in the trench 1 north wall. The excavation was benched in this area due to an unstable, thick gravel sequence. Note that fault traces are subvertical, and the fault zone is about 1 m wide.
Figure S6 (a) and (b). Logs of trench 8, showing locations of the radiocarbon samples.
Figure S7. Log of trench 3 (reversed), showing locations of the radiocarbon samples, locations of photographs shown in Figure S12, and structural measurements. The ages of radiocarbon samples are given in Table S1.
Figure S8. Log of trench 5 (see Fig. 5 in main article for location).
Figure S9. Full log of trench 4, eastern wall.
Figure S10. Full log of trench 7, western wall.
Figure S11. (a) The northern wall of trench 2 (Fig. 8 in main article), documenting the folded Holocene sand bed (unit 70) over a reverse oblique fault. Upper and lower contacts of the sand bed are marked by black lines. The red line marks the dipping fault exposed beneath the fold. Horizontal distance between the grid nails is 1 m; vertical distance is 50 cm. The green dot on the trench wall marks the same grid nail marked in (b) for reference. Circles with an X and a dot indicate the side of the fault moving away from and toward the viewer, respectively. The trench wall is oriented N70° E. (b) The northern wall of trench 2, showing a reverse oblique fault in the western part of the trench. A red line marks the fault trace juxtaposing Pleistocene clay over Holocene gravel. The fault terminates below the string grid, and overlying units are folded. The green dot marks the location of the grid nail, which is also marked in (a). An arrow points to the exposed fault plane that strikes N54° W and dips 48° NE. Slickensides on this surface trend N4° E and plunge 38° NE, showing oblique right-lateral reverse fault motion. The orange flag marks the location of radiocarbon sample 46 (Table 2 in main article).
Figure S12. Annotated photographs showing close-up views of trench 3. See Figure S7 for photo locations. Photographs are reversed to make the views northward, to be consistent with the log of trench 3. (a) Photograph shows the eastern fault, which has a component of reverse slip, bringing Pleistocene clay over Holocene sand and silt. Yellow lines mark depositional contacts; red lines mark faults. (b) Photograph showing interbed of tuff within Pleistocene clay. Green arrows mark the near-vertically dipping tuff, which is offset across two minor low-angle faults, marked by red arrows. Black dots indicate nails holding string grid, which is 1 m horizontal × 0.5 m vertical. (c) Photograph showing nearly vertical beds of Pleistocene clay displaced across minor low-angle faults. Black arrows point to representative beds; red arrows mark faults. A pocket knife is shown for scale. (d) Photograph showing the westernmost fault exposed in trench 3. Fault motion is oblique reverse, juxtaposing Pleistocene clay (gray unit) above Holocene sand and silt (brown unit). The fault plane is marked by red arrows. The black arrow shows the location of the exposed fault plane, with slickensides plunging 24° N, 40° E, indicating right-lateral, reverse slip (location 1, Fig. S7).
The radiocarbon calibration program CALIB, version 6.0, by Stuiver and Reimer (1993) is available online at http://radiocarbon.pa.qub.ac.uk/calib/calib.html (last accessed September 2013).
Reimer, P. J., M. G. L. Baillie, E. Bard, A. Bayliss, J. W. Beck, P. G. Blackwell, C. E. Buck, G. Burr, R. L. Edwards, M. Friedrich, P. M. Grootes, T. P. Guilderson, I. Hajdas, T. J. Heaton, A. G. Hogg, K. F. Hughen, K. F. Kaiser, B. Kromer, F. G. McCormac, S. W. Manning, R. W. Reimer, J. R. Richards, J.R. Southon, S. Talamo, C. S. M. Turney, J. van der Plicht, and C. E. Weyhenmeyer (2009). IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP, Radiocarbon 51, no. 4, 1111–1150.
Stuiver, M., and P. J. Reimer (1993). Extended 14C database and revised CALIB radiocarbon calibration program, Radiocarbon 35, 215–230.
[ Back ]
