Keywords: loess, neotectonics, joints, undergraduate research, structural geology, Omaha, Nebraska
Possible neotectonic origin of regional loess fracture sets in the Omaha, Nebraska region (draft 1 - 7/28/97)
MAHER, H. D.*, Jr., ALDER, T.**, HEIM, K.**, NIHSEN, M.**, PERSING, D.**, WELCH, J.**, Dept. of Geography & Geology, Univ. of Nebraska, Omaha, NE 68182-0199, *-instructor, **-undergraduate students.
Report audience: This report is written mostly at the undergraduate geology major level. Those with more geologic training should still find quite interesting information , but may want to skim more explanatory sections. Some sections may be hard for the non-geologist to follow.
This project was initiated as a class project in our undergraduate structural geology course. Due to a serendipitous early spring snow storm that canceled the normal course field trip to Baraboo, Wisconsin, we turned our attention to fractures in loess exposures of the immediate area. Initial results were quite intriquing, establishing that regional fracture sets exist and Alder, Welch and Maher are presently continuing the investigation.
Sub-vertical and planar to sub-planar joints occur in the loess of the Omaha region, but they are not well studied or understood. Exposures include river bluffs, and construction and quarry cuts. Multiple working hypotheses for these joints include slope related forces, volume changes (e.g. drying-wetting), and neotectonic forces. Influence of particle orientation on jointing may also be important. These joints are significant in considerations of drainage patterns, slope stability, piping, and fracture permeability, and they may also be paleostress indicators. Given their occurrence in Loveland loess, with basal deposits carbon-14 dated at 20,000 years old or less (Mandel & Bettis, 1995), some joints are very young.
Stereonet analysis of joint measurements taken from ten sites over a 30 by 15 mile area indicate that two regional joint sets of consistent orientation exist. A well developed NE trending set is sub-parallel to the present day maximum horizontal stress traction for the midcontinent region, and thus is a candidate for a neotectonic set. A second set trends NNW to SSE, a prominent direction in older rocks of the midcontinent region. Multiple populations of joints at a site suggests the joints are polygenetic and/or diachronous. Hybrid extension-shear joint populations, as defined by Hancock (1991), may exist. Well developed joints in a recent quarry cut suggests they occur within the loess, and are not only surface or slope related. Locally well-developed preferred directions in drainage patterns in loess may be in part controlled by the joints. Continued work will focus on expanding regional coverage, documenting the geomorphic expression of these joints, investigation of the relationship between fractures systems in the underlying bedrock and those in the loess, and a comparison of sets seen in loess of different ages.
Loess in the Omaha region
Loess is the most common surface deposit in the Omaha and Council Bluffs area, excluding the river-lain (alluvial) flood plain deposits. It is composed of silt-sized particles that were windblown and then settled on grass-covered surfaces. Protected from subsequent winds at the base of the grass 'mat', the loess deposit grew upwards with time, locally reaching thickness of several hundred feet in the area. A weak calcareous cement, and interlocking angular small grains give loess enough strength that it can maintain steep slopes and fracture, even though it can be easily broken up into the constiuent silt particles by hand.
Deposition of the loess occurred sporadically so that different age loess units can be defined. They are correlated with times of continental glaciation in the region. They are often separated by buried, preserved soil profiles that developed during times of very low loess depositon, surface stability and prolonged weathering of the near surface material.Volcanic ash layers, charcoal fragments, and fossils permit dating of the loess units. An older and widespread loess unit, the Loveland was deposited about 140,000 years ago, while the youngest loess, the Peorian loess was deposited about 13,000-23000 years ago (on the basis of C-14 dating, Mandel & Bettis, 1995). This later unit is important because fractures within it must have formed quite recently.
The loess is underlain by a thin glacial till, or Cretaceous sandstones of the Dakota Fm., or Pennsylvanian sediments. Dissection by tributories of the Missouri and Platte Rivers produced the present landscape. Since the loess has not been significantly buried, the fractures within the loess formed essentially at the earth's surface.
Loess fractures are common and have been noted in the literature but have received relatively little attention. Examples of exposures can be seen in figures 1 & 2 below.
Figure 1: Loess with typically developed fractures, exposed in a quarry wall near Maass Road, Bellevue. Note the reddish layer in the lower section of the cliff. This is a paleo-soil that is developed in the Loveland Loess. Above is the younger Peorian loess, which was dated at circa 21,000 years by the Carbon-14 method at this locality by plant fragments near the base of the Peorian loess (Mandel & Bettis, 1995).
Figure 2 (below): Overview of Maass Road quarry site, which is one of the 9 study sites sampled.
Multiple working hypotheses for loess fracture formation
Fractures and joints in geologic material can have a variety of causes. In case of the loess of the Omaha region the following causes seem reasonable candidates. Also, given are some expectations associated with a certain cause. These may allow us to test the various hypotheses.
We suggest the loess fractures may be of distinct importance in the following ways;
Natural or artificial loess exposures were found. At each site a traverse would be made along the loess exposure. Every accessible and relatively planar fracture was measured by Brunton. In that fractures were best exposed and developed in the upper reaches of the exposure, sampling was somewhat constrained by access. 50-150 measurements were collected per site. The direction of the local slope was also noted. The readings were plotted and contoured as poles using Allmendinger's stereonet program. A major joint population was interpreted to exist if a concentration of more than 10% of the data was found in a 1% counting circle area. A minor population was interpreted to exist if more than 5% of the data occurred in a 1% area.
At this point we have data from 10 sites that cover a 25 by 20 mile area. We believe the data is consistent enough to draw some strong initial conclusions.
Fig. 2 shows representative plots from 4 sites. While some variance can be seen, the plots show consistently oriented sets from site to site. This is also reflected in Fig. 3, a plot of most of our data to date.
Figure 3: These are color contoured equal area stereonet plots of poles to measured fractures in loess at 4 different sites. C.I. = contour interval for each plot. Small black dots are a scatter plot of individual poles. Many of these, thought, are hidden on the periphery since vertical dips were common. Note that pole concentrations plot near to the margin, indicating the overall steep dip of the joints. The great circles represent the average fracture orientation for sets as defined above. In the case of the Bellevue data a spread of pole concentrations with 2 maxima is shown by two NE-SW trending great circles. The data suggests two semi-orthogonal fracture sets are common to all the sites depicted here. Additional minor populations are unique to given locations.
Figure 4: This is composite plot, similar in format to those above, of most of the data collected to date.
From these plots we interpret the following regional fracture sets to exist. The dominant set, shown as the two heavier great circles, trends 50 to 63 degrees NE-SW. It is fairly common for this population to show a spread, with relatively closely spaced multiple maxima in excess of 10%. This could be interepreted as a hybrid extension-shear joint population as described by Hancock (1991) or as as fractures formed during a shifting stress field. Pollard and Ayidin (1988) argue that there is no such thing as shear joints. In either case, the most significant thing about this set is that it is parallel to the present day, average, maximum horizontal stress direction in the mid-continent region of N59E (Grana & Richardson, 1996). It is also well developed in the younger Peorian loess. This then is a good candidate for a neotectonic fracture set, and will be referred to as such in subsequent discussion.
A second set that is fairly consistent has an average orientation of 138-318 (NW-SE), and is orthogonal to highly oblique to the neotectonic set. This is parallel to a regional set seen throughout Nebraska and the Dakotas. There is some suggestion that it is mainly absent from the Peorian loess and may represent an older fracture set (note its absence from the Maass Road data composed of younger Peorian loess). This is also sub-parallel to the prevailing wind direction during loess deposition. On a speculative note we suggest it is an inherited set, due to strain of the anistropic basement rocks.
A third fracture set has an almost due north-south orientation. We propose no mechanism for its formation at this time.