Heaving-Bedrock Hazards, Mitigation,
and Land-Use Policy:
Front Range Piedmont, Colorado

ABSTRACT

Heaving bedrock is a geological hazard that is related to expansive soils, but it is more complex in terms of its uplift morphologies, deformation mechanisms, and regional distribution. It is common along Colorado's Front Range piedmont where steeply dipping sedimentary bedrock containing zones of expansive claystone is encountered near to the ground surface. It occurs in the Pierre Shale and other Upper Cretaceous formations. The heave features associated with heaving bedrock are distinctly linear and are caused by differential swelling and/or rebound movements within the bedrock. Heaving bedrock has caused exceptional damage to houses, roads, and utilities along the Front Range piedmont since suburban-type development began in the early 1970s. Much of this damage may be attributed to the long-standing tendency to assume that the bedrock may be treated, for site-exploration and design purposes, as an expansive soil having essentially uniform properties. This approach ignores the strong heterogeneity that is often present in the bedrock. In particular, drill-hole exploration surveys and drilled pier foundations, which are generally appropriate for expansive soil hazards, have proven to be inappropriate for recognizing and mitigating heaving-bedrock hazards.

This article presents a summary of heaving bedrock as a distinct geological hazard and describes the technological and policy advances that have been made in recent years to promote understanding and effectively mitigate the problem. The Colorado Geological Survey has played a key role in these advances by introducing the term "heaving bedrock" to differentiate the problem from expansive soils; leading stakeholder field trips and conferences; investigating the physical characteristics, mechanics, causes, and distribution of heaving bedrock; publishing the investigation results; assisting county governments in creating new land-use regulations; and reviewing site investigation reports for actual subdivision projects. From this experience, we conclude that a state geological survey must be active in numerous arenas - scientific, practical, and political - to assist effectively in addressing potential hazards that impact the general public.

Key Words: Heaving bedrock; expansive soils; geological hazard; differential heaving; steeply dipping bedrock; bentonite; Pierre Shale; damage; site exploration; mitigative designs; policy; Colorado Geological Survey; land-use regulations; Front Range; Colorado.

INTRODUCTION

A high incidence of damage to roads, utilities, lightly loaded residences, schools, and commercial buildings has occurred where steeply dipping beds of expansive claystone bedrock are encountered at shallow depth along Colorado's Front Range piedmont (Figure 1 and Figure 2). The total cost of this damage amounts to tens of millions of dollars and includes maintenance and repair costs incurred by homeowners, warranty insurers, water and sanitation districts, school districts, county public works departments, and taxpayers; litigation costs; and damage that has not been repaired. Damage typically begins within 10 years after construction. Some suburban areas have experienced recurring ground deformations and damage for nearly 20 years. Site exploration and mitigation practices based on widely used expansive soils models have proven to be unsuccessful in this particular region.

The damage described above is associated with distinctive, highly differential ground deformations that form elongate heave features. Called "speed bumps" by area residents, these heave features may attain sizes as large as 0.65 m high, several tens of meters wide, and several hundreds of meters long (Figure 3). The heave features exhibit a variety of cross-sectional shapes, ranging from symmetrical to strongly asymmetrical. This type of heaving is common within 1.6 to 4.8 km (1 to 3 miles) of the mountain front along the piedmont but is atypical of other areas of Colorado that are underlain by relatively flat-lying expansive soils and bedrock. A majority of the heave features are associated with the Upper Cretaceous Pierre Shale, but heaving is observed in other Upper Cretaceous formations as well (Figure 4).

The Colorado Geological Survey (CGS) has introduced the term "heaving bedrock" to describe the geological hazard responsible for this style of ground heaving. This paper summarizes important distinctions between heaving bedrock and expansive soils, and describes the technological and policy advances that have been made in recent years to promote understanding and effectively mitigate the heaving-bedrock problem in Colorado.

Problem History

Colorado's Front Range urban corridor has experienced significant and costly soil problems since the 1940s, when large-scale suburban development moved out of the central river valleys and onto surrounding low plateaus underlain by expansive soils and bedrock. Much research was conducted in the area (and worldwide) as a result of such problems, and many advances were made with respect to site exploration and structural design in areas underlain by expansive soils (see Chen, 1988; Nelson and Miller, 1992). By the early 1970s, when suburban development first began in the Front Range piedmont southwest of Denver, expansive soils were known as a geological hazard and were addressed with special mitigative designs such as drilled-pier and grade-beam foundations and floating-slab floors. In practice, the relatively soft, expansive bedrock was treated essentially the same as expansive soils. Site exploration consisted of widely spaced drill holes that were sampled and tested for material composition, grain-size distribution, moisture content, dry density, swell potential, and/or Atterberg limits. The subsequent design of houses and other facilities was based on the assumption that the underlying materials would have somewhat uniform properties within the building footprint. This approach achieved relative success when applied to flat-lying soil and bedrock in the Denver area.

Expansive soils assumptions, and the resulting designs, were largely unsuccessful in the Front Range piedmont area from the very beginning. From the early 1970s to the early 1990s, numerous subdivisions experienced exceptional and recurring damage. Paradoxically, other subdivisions in the area appeared to be relatively unaffected. Although there were advancements in the understanding of expansive soils and the state of engineering practice over this 20-year period, the style and magnitude of damage in this area remained consistently higher than for other expansive soils areas in Colorado. The piedmont area ranked near the top in terms of claims and payouts for several national home warranty corporations. Numerous lawsuits were filed, involving a wide variety of stakeholders on both sides of the disputes. Eventually, however, much of the burden of repairing the long-term, recurrent damage to private property fell upon individual homeowners. Typically, such damage is not covered by homeowners' insurance or federal disaster relief funds.

Previous Work

Published studies addressing this problem are rare, particularly those that link physical geological characteristics to the distinctive type of heaving. The role of the area's bedrock in creating a potentially distinct geological hazard was first mentioned by Gardner (1969), several years before subdivision construction began in the area. Gardner described the occurrence of bentonite beds up to 0.3 In thick in many of the steeply dipping formations and concluded that those areas underlain by near-vertical bentonite beds are unsuitable for building foundations. Hart (1974) mapped several bedrock units along the piedmont that have greatly contrasting swell potentials. Simpson and Hart (1980) warned of a high risk of differential swelling for foundations constructed over steeply dipping bedrock units that contain different compositions within adjoining layers.

The first public-record investigation of damage in the area is a master's thesis by Kline (1983). Kline investigated geologic and non-geologic variables that may have influenced damage in a subdivision. He found that the depth to bedrock was of primary importance, with more damage occurring where the bedrock was shallower. A majority of the damaged buildings were founded on drilled piers. Gipson (1988) described the geologic setting of the Front Range piedmont and described its influence on the damage that had occurred there. He noted two unusual heaving morphologies. The first is the elongate "speed-bump" morphology. The second consists of broad, gently tilting, differentially uplifted surfaces ranging up to a block (-200 m) long from their lowest to highest edges. Gipson attributed the damage to (1) steeply dipping bentonite layers being flanked on either side by less-expansive bedrock, and (2) weathering of the steeply dipping formations, which appears to be deeper than it is for flat-lying bedrock.

E. C. Weakly (personal communication, 1993) mapped thousands of heave features in the piedmont area for a national warranty company during 1989 and 1990. He noted a strong tendency for the features to be aligned longitudinally along the direction of regional bedrock strike and attributed the distinct style of heaving to the presence of individual bentonite beds. The warranty company used the results to determine areas for which they would not issue warranty insurance for new construction. Unfortunately, because of the private contract, Weakly's mapping and results have not been released for use by the general public.

Major evidence for the differences between damage from steeply dipping bedrock and flat-lying bedrock and soils was presented by Thompson (1992a, 1992b). By comparing the piedmont area with the greater Denver area, Thompson showed that the damage rate in the piedmont area underlain by steeply dipping, expansive bedrock was more than three times greater than the damage rate for flat-lying, expansive bedrock and soils around Denver. He found that the amount of overburden (i.e., the depth to bedrock) was a critical factor governing damage in the piedmont area, with essentially no damage occurring when >3.3 m of overburden soil and/ or fill was present. Thompson also showed that Subsurface moisture in the piedmont area increased to depths of as much as 10 m after development, much deeper than the 3 in of moisture penetration commonly assumed in engineering practice in the Denver area.

Nichols (1990, 1992) hypothesized that the differential, linear heave features in the Front Range piedmont area were the result of increased rates of rebound (i.e.. time-dependent release of stress) from unloading and disturbance of over-consolidated claystones, and not the result of hydration-induced swelling. Nichols et al. (1994) showed an example of ground heaving along a thrust fault surface in near-horizontally bedded Pierre Shale in South Dakota that resulted in an asymmetrical, linear heave feature (similar to those seen in Colorado's piedmont). He attributed this feature to rebound after removal of road-cut overburden.

Setting the Stage for Change

New subdivisions built in the Front Range piedmont during the 1970s, 1980s, and early 1990s continued to experience damage from bedrock heave. Most builders and engineers ignored or were unaware of the relevant technical papers listed above. They continued to use site-exploration and building-design methods based on expansive soils models that assume flat-lying, laterally uniform strata (Figure 5). The use of progressively longer drilled piers (increased from 3 to 5 m in length to 8 to 12 m) and structurally supported floors in place of floating slab floors improved structural performance to some degree. However, these expansive-soil designs were not especially successful in areas experiencing severe differential heaving. A strong prejudice existed to continue to treat the piedmont bedrock as an expansive soil because expansive-soil theories and practices were being applied successfully in expansive soil and bedrock in other parts of the Denver area. Specific designs for expansive soil were accepted as the standard of engineering practice in any expansive substrata. Deviations from the standard of practice were discouraged because of liability issues if subsequent failure occurred.

By the late 1980s, a number of major stakeholders - the engineers, builders, county planners, and especially, the homeowners, had become frustrated by the continued high incidence of damage in the piedmont area. Many of the stakeholders had some experience and knowledge of the problem, but no consensus existed for a solution. When the stakeholders met, it was most often as antagonists in litigation.