Liquefaction at Finley Quarry? A critique of Sherrod et al. (2016)
June 13, 2019
"Our current understanding of the Western Cordillera was built from equal parts science and personality. The Geology and the Geologist cannot be separated."
Opening comment from Professor Art Snoke on the first day of his graduate course
on Cordilleran Tectonics, University of Wyoming, 1998
Two reports by two different teams of geologists describe the same outcrop, sediments, and features at Finley Quarry, WA (Coppersmith et al., 2014; Sherrod et al., 2016). I visited this site several times in years prior to those teams' visits, as have numerous geologists in decades past (Jones and Deacon, 1966; Jahns, 1967; Brown, 1968; Bingham et al., 1970; Rockwell, 1979; Farooqui, 1977; Farooqui and Thoms, 1980; Foundation Sciences, 1980). When the Sherrod/USGS article was published, I was a bit surprised by some of their descriptions and interpretations. I consulted my own field notes and photographs. I found a few apparent discrepancies and emailed my questions to the lead author. Receiving no response from him, I contacted his senior colleague at U.S. Geological Survey (Department of Interior) to ask if he might be willing to forward my questions. He did, but again, no response. My reply to the Sherrod et al. (2016) article is below.
A sketch of Finley Quarry. The area of concern lies at the far left.
A "clastic dike" shown in Figure 7.15 of Coppersmith et al. (2014) caught my attention. There are several conspicuous breccia-filled fractures in the high wall of the basalt quarry (fault gouge), so my first thought was that might be describing those. But the clastic dike was not one of those. It was clearly enclosed by sediment and labeled a "liquefaction" feature.
The fault gouge was determined to be middle Pleistocene, dated by U-series method (Paces, 2014):
"Carbonate‐rich gouge present along the fault exposed at Finley quarry shows complex layering and cementation paralleling the attitude of the fault...ages range from 700 +270/‐160 ka to 224 ±36 ka...These ages are consistent with repeated faulting over a several hundred thousand year time frame in the middle Pleistocene. If younger fracturing and cementation events are present (for instance, the event that formed the colluvial fault wedge dated at 17 ka), they were not sampled in this block of gouge."
I believe this feature is misinterpreted by Sherrod et al. (2016) as a Holocene liquefaction feature created by seismic shaking along the Olympic-Wallowa Lineament (OWL). While other information at the site, documented by these two teams and by others in the past, supports faulting and seismicity, the clastic dike could be interpreted differently.
A simpler explanation is that the dike is Pleistocene in age (not Holocene). Thousands of such dikes intrude the Touchet Beds in the area. The quarry lies below the maximum level of Missoula floods (Lake Lewis shoreline = 380m elevation) in which the Touchet Beds were deposited.
I interpret the "liquefaction dike/sill (E4)" to be older than the modern soil (Units, 17, 18 in Fig. 7.15), thus does not cross-cut it. Rather, like many other clastic dikes in the region, the feature predates the modern soil (Holocene "Humic-stained L1 loess"). The top of the dike is likely overprinted (truncated) by the modern soil. The tops of many clastic dikes in the vicinity similarly appear to fade upward over a distance of a few centimeters into the base of the soil profile (pseudo-truncation). Pedogenic processes gradually obliterated the structure of the dike.
If the dike is Pleistocene in age, there is a good chance it has a non-seismic origin. Liquefaction features are not common in Eastern Washington (no reports to date) despite abundant unconsolidated sediments with grainsize distributions that make them susceptible to seismic shaking. There is no source bed is identified and no taper direction was established in the report. The dike-sill feature could just as easily be a typical downward-injected feature, like thousands of others in the Touchet Beds.
It appears the Sherrod team was overenthusiastic in attributing the "clastic dike" to earthquake-caused liquefaction.
Figure 7.15 in Coppersmith et al. (2014) shows the outcrop at Finley Quarry examined by the Sherrod/USGS team and the Coppersmith/Quaternary Studies Team. The dike is circled in red.
Geologists from Foundation Sciences, Inc. drew a sketch of the Finley Quarry exposure some 36 years prior to arrival of the Sherrod/USGS team. Their caption reads, "Sketch of Finley Quarry Fault. Note unfaulted post-flood colluvium and loess overlying older faulted materials." (Foundation Sciences, 1980, Figure 4).
Others may agree. I found some curious comments from the authors of the Coppersmith et al. article in a table near the end of their article (see Table 7.2). The comments suggest there were disagreements about the stratigraphy at the site, including the clastic dike-sill feature,
The postulated liquefaction feature (dike and sill d) that appears to post-date the deposition of [~11,000 year old Glacier Peak G] tephra is the strongest evidence for a strong shaking earthquake in the general vicinity of Finley quarry (which may or may not be on the RAW-Rattles fault source). The best expressed part of the mapped feature cross cuts Unit 13. The USGS log interprets Unit 13 as L1 loess. However, this unit shares characteristics similar to massive to poorly bedded slackwater deposits observed elsewhere (e.g., at the MCBONEs locality) (i.e., it contains scattered clasts and irregular blocks of pre-flood ice-rafted or ripped up material). Although the mapped extent of the feature up into the modern soil suggests a post-flooding event, additional work or review is needed to preclude a flood-related liquefaction phenomenon.
I disagree with some of the interpretations made at Finley Quarry by the Sherrod/USGS team, mostly because they provide an abbreviated description of the liquefaction feature, a key element in their young-seismicity narrative. In his first few years with USGS, Sherrod has recalculated the 1936 Stateline/Milton-Freewater Earthquake magnitude (no change) and endeavored to find the 1872 Chelan Earthquake epicenter using lidar (relocated epicenter several miles south; renamed it North Cascades Earthquake). His work at Finley Quarry is consistent with those efforts.
Answers to the following questions concerning the geology at Finley Quarry would be helpful:
1.) Is the "liquefaction dike-sill" in Unit 19 something different than the vertically-sheeted clastic dikes ubiquitous in the Touchet Beds?
If the same, the feature would properly be called a Pleistocene clastic dike deformed by Holocene mass wasting associated with the local hillslope or other injection process. Unit 19, "gravelly silt to fine-medium well-bedded sand interbedded with silt laminae", appears to describe Missoula flood rhythmites. If not, are similar dike-sill features present in other Eastern Washington locations?
2.) Was a source bed for the liquefaction dike-sill feature identified?
To date, no clastic dikes of Pleistocene age in Eastern Washington have been shown to ascend from a buried source bed. I am unaware of any reports describing Holocene liquefaction dikes in the Cascade Range or east of the Cascades. Perhaps recent trenching by USGS has uncovered some (i.e., Bennett's work at Burbank, WA).
3.) Could Units 16 and 13 be diamicts similar to those described by Spencer & Jaffee (2002) and Bader et al. (2016) at locations in the northern Walla Walla Valley?
These authors interpret massive to crudely bedded, silty deposits that contain anomalous pebbles to ancient floods sweeping loess-covered hillslopes clean. In the western WWV, eastern Pasco Basin, and southern Palouse Slope (areas located just north of Finley Quarry), these silty beds are relatively common. They are associated with Palouse loess below 380m elevation and contain calcic paleosols, mammal fossils, and several tephras. Bedding is very difficult to resolve in water-lain silt. The presence of pebbles suspended in the otherwise massive matrix is the clue to a water-lain origin, thus diamict. These would not be Touchet Beds, as indicated on p. 1652 and in Fig. 12, but older flood deposits (probably pre-late Wisconsin).
An undeformed clastic dike in the Touchet Beds of south-central Washington State.
Sherrod et al., 2016, Active Faulting on the Wallula Fault within the Olympic-Wallowa Lineament, Washington State, USA, Geological Society of America Bulletin v. 128
Coppersmith et al., 2014, Appendix E: Structural analysis and Quaternary investigations in support of the Hanford PSHA in Hanford Sitewide Probabilistic Seismic Hazard Analysis, Pacific Northwest National Lab Report #23361
Foundation Sciences, Inc., 1980, Geologic reconnaissance of parts of the Walla Walla and Pullman, Washington, and Pendleton, Oregon 1 degree x 2 degree AMS quadrangles , a consultant's report to U.S. Army Corps of Engineers-Seattle District, Contract #DACW67-80-C-0125, 144 pgs.
Bingham et al., 1970 , Geologic investigation of faulting in the Hanford region, Washington with a section on the occurrence of microearthquakes , U.S. Atomic Energy Commission, Division of Reactor Development and Technology Report #90-27, 126 pgs.
Camp et al., 2017, Field-Trip Guide to the vents, dikes, stratigraphy, and structure of the Columbia River Basalt Group, eastern Oregon and southeastern Washington, USGS/DOI Scientific Investigations Report 2017–5022–N, p. 65-66