Use of Geophysics for Transportation Projects (2006)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

100 Successful and unsuccessful case histories presented in this appendix were prepared by the respective transportation agencies; the synthesis consultant has made only minor mod- ifications or comments. These four case histories were se- lected for this report because they cover a multitude of geo- physical methods and techniques, as well as geotechnical applications. The authors have submitted them in a similar format to make for easy comparisons. For the purpose of this synthesis, successful or unsuc- cessful case histories (as presented in Table 4) are based solely on meeting technical objectives, not whether the proj- ect met budget or timeline constraints. It should be noted that an unsuccessful case history is not a criticism of geophysics—the methods or the tech- niques. As the authors indicate, poor application, instru- mentation, or data interpretation software may often be the reason a geophysical investigation does not meet the objectives. CASE HISTORY 1 How not to do it—A summary of a reflection seismic project to delineate potential faulting across structures proposed for a major Interstate project in southern Cal- ifornia An Unsuccessful Geophysics Project William Owen, Chief, Geophysics and Geology Branch, California Department of Transportation Objective and Purpose In the summer of 2002, a staff geophysicist was consulted to propose investigations for Interstate improvements in the San Bernardino Valley in southern California. A number of bridges and overheads were proposed for this project, which is located in an area of significant seismic hazards. The pur- pose of the studies would be to evaluate possible fault traces beneath the proposed structures. Factors influencing the se- lection and sequencing of the geophysical surveys were: (1) the project area is in a deep alluvial basin, with estimated sediment deposition in places on the order of 2 km; (2) the area is heavily urbanized; (3) vibration noise is significant; (4) the client had deadline pressures (less than 90 days) and wanted a quick turnaround; and (5) there was no identified stable funding source to pay for the work. Options Considered and Method Selected When discussing the client’s needs with the staff geologists, the consulting geophysicist outlined some options. Both sides agreed that reflection seismic was the primary method reasonably expected to provide the client’s desired resolu- tion. However, staff geologists wanted imaging of the basin to bedrock (up to 2 km). The geophysicist described the pros and cons of compressional (P) wave and shear wave reflec- tion: P-wave could give them the desired depth, with reduced resolution; shear-wave could yield the desired resolution, with the potential for shallower investigation depth and a guarantee of increased cost. For either method, the noisy en- vironment remained a problem. Because the state department of transportation (DOT) had no in-house capability for this highly specialized work, the job required an outside contract. The opinion of the geophysicist was that high-resolution, shear-wave reflec- tion seismic, using a vibratory source, was the best option; arguing that even if a deep P-wave survey was performed, the shear-wave survey would still be needed for accurate delineation of shallow offsets posing the greatest risk of surface rupture. Owing to cost issues, the staff geologists opted for a deep P-wave survey, using impact sources, with a subsequent shear-wave survey, once stable funding was secured. The geophysicist, in consultation with a qualified consul- tant, developed scopes of work and initial cost estimates. Ini- tial estimates were under budget, but the consultant’s com- mitment to another project raised questions about the deadline. That uncertainty, despite assurance from the con- tractor, led the staff geologists to pursue a different consul- tant. The consulting geophysicist was eventually removed from the project as a result of disagreements with the staff ge- ologists over consultant selection and project scope and cost. Ultimately, none of the original consultants was selected. Six months after the original deadline, staff geologists began work under agreement with a federal research agency, at a cost $30,000 more than the geophysicist’s original scope of work. Four months after work had begun the consulting geo- physicist was again contacted regarding the project. The geophysicist was invited to rejoin the technical team and was informed that the original geology team was no longer working on the project. At that point, fieldwork was nearly complete. The geophysicist requested a status briefing from the federal research agency’s lead investigator. A response was not received. APPENDIX D Case Histories

101 Results More than one year after that request, a draft report was sub- mitted. The consulting geophysicist concluded in review that, as feared, the P-wave reflection data did not success- fully image shallow targets of greatest interest to the project. Also of concern were the deeper portions of the data (Figure D1). In the opinion of the consulting geophysicist, the inter- pretation of the deep seismic sections included geologic structures that were not plausible given available knowledge of faulting and geology in the area. The rebuttal from the in- vestigators essentially agreed that the data could not meet the primary objective required by the state DOT, but disagreed with the assessment of the interpretations. However, in their final draft the investigators presented a significantly differ- ent reinterpretation of the same data (Figure D2). Reasons for Failure and Lessons Learned Both practitioners and users of geophysics must be cognizant of project limitations that may affect the geophysical inves- tigation. In this case, a number of factors contributed to the failure. Project deadlines placed extreme limits on what could be done. It was apparent from the outset that this type of inves- tigation, from initialization to final report, could not be com- FIGURE D1 Initial interpretation of reflection seismic section. Grayed zone is interpreted extent of sedimentary basin. Many interpreted faults on the section appear inconsistent with the available data or are incompatible with existing knowledge of faulting in the basin. FIGURE D2 Revised interpretation of Figure D1. New interpretation may be more plausible, but spaciousness is likely the result of poor data quality. Resolution, particularly in the upper 500 m, is insufficient to reliably discern fault proximity relative to structure locations.

102 pleted by the desired deadline. As it happened, the ultimate path chosen by the geology team resulted in project duration much longer than the consulting geophysicist’s original plan. In the end, deadlines were extended to accommodate the in- vestigation. Lack of a stable funding source made it difficult to de- velop the scope of work and to obtain contractor commit- ments. Quite simply, although funding issues were the result of legislative actions beyond control of the state DOT, the DOT could not appear serious in its desire to pursue the in- vestigation at an accelerated tempo when the consultants could not be assured that funding existed to carry it out. The noisy environment significantly affected the data ob- tained using the impact sources. That and other factors lim- ited usable bandwidth to lower frequencies, resulting in a low-resolution data set of questionable interpretation that could not satisfy the state DOT’s need for detailed and accu- rate resolution of shallow features. Conclusions The main point of this case history is that a geophysicist is in- valuable in planning, coordinating, and conducting geophys- ical surveys. Although California law allows geologists to in- corporate geophysics into their practice, reflection seismic is particularly complex and was clearly beyond the experience and practice of the staff geologists, who against advice de- veloped a scope of work without input from the consulting geophysicist. In the end, the client spent $215,000 for work that could not fulfill the desired objective. The follow-up shear-wave survey was not carried out and, in this case, the client’s negative experience makes that follow-up unlikely. CASE HISTORY 2 Application of Ground Penetrating Radar at the Stony Rapids Airfield in a remote area of northern Saskatchewan A Successful Geophysics Project Saskatchewan Department of Highways and Transporta- tion (SDHT) and Pavement Scientific International P. Jorge Antunes, Principal Geotechnical Engineer Curtis Berthelot, President, Allan Widger, Executive Director Engineering Gordon King, Regional Executive Director Objective and Purpose The objective of this trial project was to use ground pene- trating radar (GPR) to measure the extent of moisture accu- mulation within the Stony Rapids Airfield substructure. The Saskatchewan Department of Highways and Trans- portation (SDHT) is responsible for maintaining and operat- ing 18 provincial airfields, including the Stony Rapids Air- field in Northern Saskatchewan. During regular airfield inspections, SDHT staff and pilots noticed the formation of two large depressions in the airstrip that were beginning to affect aircraft take-off and landing operations. Owing to its remote northern location with limited equip- ment available, a make-shift site investigation was com- pleted by drilling a few large diameter boreholes on the graded portion adjacent to the runway. The boreholes were drilled in the vicinity of the depressed areas with a large di- ameter auger on loan from the local electrical company. The boreholes identified large volumes of water contained in the substructure near the ground surface. SDHT needed to quan- tify the extent of the moisture accumulation beneath the run- way to determine whether a subsurface drainage system would be required. If this system was required, design param- eters would be necessary to determine the type and size of the subsurface drainage system needed for the Stony Rapids Airfield. Options Considered and Method Selected The first option considered to measure the extent of mois- ture accumulated in the Stony Rapids Airfield substructure was to use conventional coring, sampling, and laboratory analysis along a grid pattern on the runway. It was decided that the conventional methods were too expensive and not practical. The other method considered was the use of GPR to use nonintrusive geophysical methods to map the groundwater beneath the runway surface. It was believed that this tech- nology could be cost-effective for this project. As a result, SDHT contracted with Pavement Scientific International to perform a GPR assessment of the Stony Rapids Airfield. The logistics of transporting the portable GPR equipment were handled by using a chartered aircraft to airlift the equip- ment and crew from Saskatoon, Saskatchewan, to the remote location. Once there, the equipment was attached to a stan- dard pick-up truck for use in the geophysical survey, as shown in Figure D3. Results The GPR surveys were conducted on the runway during the intervals of aircraft inactivity. The truck-mounted GPR unit was shuttled on and off the runway throughout the day with- out having to restrict regular aircraft operations. GPR profiles were collected starting from the northeast end of the airstrip and ending 1.6 km at the southwest end. There were five radar scans for each of the seven passes.

103 Figure D4 is as an example of a GPR scan. This scan shows a significant increase in dielectric permittivity (an in- dication of increased relative moisture content) in the sub- base at the depressed areas, as well as the surrounding area. This indicated further expansion of the depressed areas. The moisture accumulation contours measured within the granu- lar base layer were the same as found in the subbase. This could be the result of moisture diffusion from the subgrade or moisture infiltration from the pavement cracks. The source of the moisture could not be determined because surface FIGURE D3 Truck-mounted GPR at Stony Rapids Airfield. FIGURE D4 Ground penetrating radar scan pass 1A.

104 cracking information was not available at the time of the GPR survey owing to a recent application of a bituminous surface course on the entire runway. Figure D5 illustrates the moisture accumulation con- tours measured both at the top of the subbase and within the granular base layer within the Stony Rapids Airfield. These contours verified the theory that the runway depressions were caused by moisture within the runway substructure. The contours also showed that there were other areas that were prone to settlement along the runway. This confirmed the need for a subsurface drainage system beneath the air- field runway. The entire survey was completed within 14 h, which in- cluded 6 h of flying time. In addition, one day was required to complete the data analysis and generate the results used for this study. Reasons for Success The GPR survey was an innovative solution to the prob- lems faced at the Stony Rapids Airfield. The results of the survey proved that the extensive groundwater was already extending past the existing depressions in the runway and that remedial measures needed to be taken to install the subsurface drainage system. The information from this GPR survey was used in the design process as well as in a request made to a Canadian federal government agency to fund this airport work under a federal monetary grant structure. The experience and knowledge of both the project en- gineers as well as the consultant involved made the project successful. SDHT has numerous senior engineers willing to share their knowledge and experience. Innovations and new technology are also encouraged within SDHT. This mindset within SDHT makes innovative technologies open to consideration by front-line staff and executive managers. Conclusions This case history illustrates SDHT’s positive experience with GPR technology and ultimately the use of this type of geo- physics. The ability to carry out a nonintrusive geophysical survey, in a single day, at a remote northern airport, while maintaining full airfield aircraft operations makes this proj- ect a success from a logistics point of view. In addition, the results allowed decision makers to confirm design parame- ters, as well as obtain federal government funding for the project. For these reasons, this project has helped to promote the further use of GPR and geophysics within SDHT. CASE HISTORY 3 Geophysics and Site Characterization: K-18 bridge over the Kansas River An Unsuccessful Geophysics Project Neil M. Croxton, P.G., CPG, Regional Geologist, Kansas Department of Transportation Objective and Purpose In 2001, the Kansas DOT (KDOT) decided to begin incor- porating geophysics in preliminary investigations to supple- ment drilling and sampling programs. Geologists at KDOT began looking for a place to experiment with different geo- physical methods. We wanted to find a characteristic bridge project, preferably one with good as-built elevations, so that we could easily compare the geophysical results with site- specific data. The project selected for the tests was a proposed bridge over the Kansas River on K-18 between Manhattan and Junction City (Figure D6). The bridge is to be built alongside the exist- ing structure, which was constructed recently enough so that detailed geology information is available. The riverbed itself is nearly 800 ft wide, sits 10 ft below the bank, and is braided with sand bars. Fluctuations in water flow are common. To design the foundations for Piers 2, 3, and 4 we needed top-of-bedrock elevations across the riverbed, along with some idea about the degree of weathering at the soil/rock interface. Options Considered and Method Selected Drilling at the proposed pier locations in the riverbed would have only been possible by obtaining specialized drilling equipment or by conducting extensive earthwork. Neither option was economically possible, nor were they seriously considered. The site appeared to be a good choice however as a test of modern engineering geophysics. Bedrock in the area consists of alternating layers of Permian age limestone and shales. Kansas River alluvium typically consists of quartz sand with lenses of clay and silt; the alluvium is less FIGURE D5 Stony Rapids Airfield wetted up subgrade contours.

105 than 30 ft thick at this location. We believed that the contrast of material properties at the alluvium/bedrock contact—sand overlying limestone and hard shale—might lend itself to good quantitative geophysics data that would provide the needed information. Seismic refraction and 2D resistivity profiling were the two geophysical techniques that we believed could provide the necessary information. Geophysicists at the Kansas Ge- ological Survey were consulted; the consensus was that both resistivity and refraction had their advantages and limitations in this geologic setting. We finally decided to try both the seismic and electrical methods and learn for ourselves. KDOT has a standing contract for drilling services with a consulting firm that also performs geophysical investiga- tions. A proposal from this company was submitted and ac- cepted. KDOT would pay $33,000 for resistivity and seismic refraction field surveys, data interpretations, and their report. Results Field work took place over 4 days in early December 2004. Later that month, the consulting firm submitted its report. In it, our consultant expressed little confidence with the refrac- tion results. Background noise was blamed for the poor data—wind, road noise from the nearby bridge, and vehicle activity at the nearby Fort Riley military base. The resistiv- ity data were much better, and the geophysicists were opti- mistic about the results. The report contained interpreted top- of-bedrock elevations for both surveys. It is our experience that the alluvium/bedrock contact be- neath even the largest rivers in eastern Kansas is generally planar (Figure D7). Occasional scour holes are found, but the rock is usually too resistant to give dramatic weathering dif- ferences across a site. At the K-18 crossing, drill holes on op- posite banks (separated by about 800 ft) showed that the con- tact varied by less than 3 ft. We expected that at least one of the geophysics methods would clearly show this contact be- tween such different materials as sand and hard Permian bedrock. We were disappointed. Resultant interpretations for the bedrock contact from both refraction and resistivity data showed irregular soil/bedrock interfaces with unrealistic high and low points. In the riverbed itself, the refraction data yielded differences of up to 16 ft; the resistivity results varied by up to 20 ft. The two methods diverged by as much as 25 ft toward the north end of the channel area. The resistivity profile showed large areas of very high resistivity in locations where drill holes found only sand (Figure D8). Very low resistivity was shown in places that we know is hard bedrock. And finally, both meth- ods seemed to reflect the topography, as the interpreted bedrock contact followed the surface elevation up both banks and over sand bars between river channels. In short, the results were use- less for our design and planning work. Reasons for Failure We consulted with geophysicists at the Kansas Geological Survey to help us figure out what might have gone wrong. Seismic and electrical specialists decided that the computer FIGURE D6 K-18 bridge site located on the Kansas River between Manhattan and Junction City.

106 programs used to interpret the data are most likely responsi- ble for the poor results. There are a handful of different pro- grams available to help geophysicists handle refraction and resistivity surveys, and each has its own biases. Scientists who use these programs must be very careful in choosing which to apply to a certain situation. By this time, the bridge foundation geology report was due. There was no time to have the geophysical data reinter- preted or evaluated any further. The piers in the channel were designed using geology information from the existing bridge. Lessons Learned Companies specializing in seismic refraction and resistivity on this scale are not common in the central plains. KDOT Geology assumed that, given the planer geology and existing FIGURE D7 Conception geologic model of K-18 site. FIGURE D8 Geophysical results and geological (drilling) elevations for top-of-rock.

107 borehole data, local geotechnical firms could provide the same level of service as the well-known geophysics compa- nies. In the future, KDOT will require proof of similar work with references and, through competitive bids, utilize firms that specialize in geophysical investigations. Conclusions The failure of seismic refraction and 2D electrical resistiv- ity profiling to provide reliable information at the K-18 bridge has not affected KDOT’s determination to use engi- neering geophysics during its preliminary site investiga- tions. A different stream crossing will be chosen, and we will try again. Synthesis Author’s Comment The construction phase may reveal details about the soil/bedrock interface and its surface relief and degree of weathering. It is recommended that these physical measure- ments, made during construction and/or additional site in- vestigation work, be integrated into the geophysical data, the interpretation constrained by these parameters, and the same software used to produce new sections. Typically, applying constraints to the algorithms will help the geophysical model more closely reflect the geologic model; and, it is the opin- ion of the synthesis author that “ground truth” evidence is needed for this project in the areas interpreted to be anom- alous from the conceptual model. This ground truth comment applies for all geophysical investigations. CASE HISTORY 4 Geophysical Investigation of the USH 53 Birch Street In- terchange Site, Eau Claire, Wisconsin A Successful Geophysics Project Dan Reid, Geologist, Bureau of Technical Services, Ge- otechnical Section, Wisconsin Department of Transportation Objective and Purpose As part of the realignment of U.S. Highway (USH) 53 in Eau Claire, Wisconsin, the Wisconsin DOT (WISDOT) will be constructing an interchange for the Birch Street connection to this highway on a parcel of property owned by the city of Eau Claire. An old city landfill, with little to no historical data on the nature and extent of filling, occupies a preexist- ing ravine on the property that extends beneath the limits of construction of the proposed Birch Street intersection. A geo- physical investigation was conducted at the property in De- cember 2002 to estimate the horizontal extent of landfilling within the proposed limits of construction and to focus future geotechnical and environmental work at the site. Options Considered and Method Selected During the design phase of the project, discussions between geotechnical engineering and project design staff indicated that a phased investigation approach was warranted at this site, and that a geophysical study would be appropriate as the second investigation phase (after a preliminary environmen- tal investigation). Several geophysical options were consid- ered for this study including seismic refraction, GPR, elec- trical resistivity, and electromagnetics (EM). Project design staff indicated that they had limited funds for this work and that they needed a quick turnaround on the results. Based on this information, and that soil borings from the project area indicated that native soils consisted predominantly of allu- vial sand and gravel (which should be characteristically less conductive than the waste material placed in the ravine), the EM and GPR methods were selected. An EM31 terrain con- ductivity instrument manufactured by Geonics, Ltd., and a RAMAC GPR instrument with a 200 MHz antenna manu- factured by Mala Geoscience were used for this investiga- tion. Interpretations of the EM31 data were developed using SURFER(r) (Golden Software, Inc.) to contour the data sets in 2- and 3-dimensional simulations of the study site, whereas GPR data were plotted and interpreted using the Mala, GroundVision(r) software package. The field investigation began by establishing a surveyed grid with 50-ft centers, approximately 750 ft long (north to south) and 600 ft wide (west to east) that covered the limits of proposed construction over the landfill. The terrain conductiv- ity and GPR profiles were completed by running the instru- ments across the survey grid from west to east, beginning at the northern end of the site and proceeding to the south. Three ad- ditional profiles with each instrument were also completed from south to north after the first set of data had been obtained. Results The EM31 data indicate that there are two distinct anomalies; a conductivity high located in the northwest section of the grid and a conductivity low located in the west-central sec- tion of the grid. Both of these features are likely associated with landfilling at the site. In general, waste with higher ap- parent conductivity is located in the northwest portion of the landfill grid (later identified as industrial waste), whereas the southern portion of the grid contains waste with lower over- all apparent conductivity (later identified as construction and demolition waste). Figure D9 presents 2D and 3D image maps of conductivity data on the landfill grid. GPR profile data clearly illustrate the extent of filling along the margins of the landfill. For this study, interpreta- tions made on GPR data sets were focused on the profile mar- gins, primarily because some areas of the landfill, specifi- cally the thicker fill sequences in the center of the preexisting ravine, were highly conductive to GPR signals and resolution of subsurface conditions in these areas of the profiles was

108 poor. In addition, the profile margins covered the limits of the preexisting ravine where the preexisting ground surface would be closer to the surface. The buried, preexisting ground surface was the primary target of GPR profiling to help characterize the limits of landfilling at the site. Figure D10 presents a partial GPR profile collected along the west edge of the site, near the conductivity high noted on Figure D9, and clearly shows the downward trending preexisting ground surface near the edge of the grid. Interpretations of the extent of landfilling at the site were developed using both the EM31 image maps and GPR profile data. Reasons for Success and Lessons Learned Overall, these two geophysical methods worked well to iden- tify both the horizontal extent of filling and the location of high conductive fill material within the preexisting ravine. A subsequent geotechnical investigation phase confirmed the geophysical interpretations and provided data on the charac- ter of the fill material. The geophysical methods achieved the objectives and purpose of this investigation primarily be- cause they were appropriate for the subsurface conditions en- countered. Conclusions The geophysical investigation used on this project proved to be a cost-effective, accurate, and timely method of evaluat- ing the extent of landfilling on the site. The success of this investigation will help ensure that geophysical methods are considered as practical investigative tools on other WISDOT projects in the future. FIGURE D10 Partial GPR profile along west edge of site. FIGURE D9 2D (left) and 3D (right) image maps of conductivity data on landfill grid.