CHAPTER 3
CHICAGO AVENUE AND STATE STREET SUBWAY RENOVATION PROJECT OVERVIEW

Back to Table of Contents

3.1 GENERAL PROJECT DESCRIPTION

This chapter presents support system performance observations and building response data collected during the excavation and construction the Chicago Avenue and State Street Subway renovation project in Chicago, IL. The Chicago Department of Transportation (CDOT) authorized the renovation and expansion of the existing subway station. The renovations included adding a mezzanine section for office and vendor spaces and making the station handicap accessible. The station renovations began in June 1999 and were completed by using the cut and cover technique of construction. The construction included excavating 12.2 m of soft to medium clay to expose the existing subway station and tunnels and to allow for capital improvements. CDOT contracted with Wiss, Janney, Elstner Associates, Inc. (WJE) to monitor ground and structural movements associated with the excavation and the subsequent renovations, to monitor the vertical movements of the adjacent school, and to assess the potential for structural damage to the adjacent buildings. The adjacent buildings of most concern were the Frances Xavier Warde School and the Holy Name Cathedral. The structural response of the Warde School to the excavation was of particular interest because of its close proximity to the excavation.

Northwestern University was subcontracted by WJE to perform the portion of the contract involving monitoring and predicting the ground movements that resulted from the excavation activities.

Figure 3-1 shows the Chicago and State subway renovation project site as viewed from two different directions. Figure 3-2 shows a plan view of the Chicago and State subway renovation project site including instrument locations, temporary wall types and strut locations. The temporary wall support along State Street consisted of tieback anchors and cross-lot braces.

Monitoring and performance data collected during the construction included lateral soil deformations, pore water pressures, vertical building movements and loads in several cross-lot braces and tiebacks. Inclinometers measured lateral movements at five locations around the excavation. Pore pressures were measured using pneumatic piezometers at two locations on opposite sides of the excavation. Data from the inclinometers and piezometers were obtained daily during wall installation and excavation, and at least on a weekly basis after the excavation had reached its final depth. Strut loads were determined from measurements of strain in the struts for three of the struts. Loads in the tieback anchors were determined using load cells. The load cells were installed on four anchors, two on either side of the excavation. Strain gauge and load cell data were obtained daily during excavation, and weekly after the excavation was completed. Inclination and elongation of the foundation wall were measured using tiltmeters and extensometers, respectively. Data from these devices were obtained daily by remote access during construction. Lastly, building movements were surveyed optically once per week during the project.

3.2 SUBSURFACE CONDITIONS

3.2.1 Geology and Site Stratigraphy

The soils in the Chicago area are primarily glacially derived sediments. Four major geologic events contributed to the formation of the subsurface conditions in this area. The geologic events included: (i) Marine sediment deposition—these deposits formed the bedrock in the vast majority of the Chicago area; (ii) A long period of erosion, which occurred prior to the Pleistocene glaciation; (iii) Successive advances and retreats of a continental glacier during the Pleistocene epoch—the majority of the area soils were deposited during the Wisconsin stage of this epoch; and (iv) The variation of elevation of the glacial Lake Chicago, which included water levels from 18 m above to 30 m below the current levels of Lake Michigan (about 177 m above mean sea level).

Much of the subsoil in the Chicago area was deposited during advances and retreats of the Wisconsin Stage glacier. These subsoils consisted of debris that was aggregated and deposited directly by glacier ice without disaggregation by other agents, such as melt water. The advance and retreat process was marked by terminal moraines, which created readily identifiable strata. The geologic names of the more prominent strata in order of deposition are the Valparaiso, Tinley, Park Ridge, Deerfield, Blodgett and Highland Park. The two older strata, Valparaiso and Tinley, are comprised of hard clays and silts. The Park Ridge, Deerfield, Blodgett and Highland Park strata are comprised primarily of silty clays. These younger strata tend to be softer and more compressible than the older strata. The varying stiffness of the different strata is attributed to the changes in the weight of the overlying glaciers during deposition.

The Park Ridge, Deerfield, and Blodgett strata define the Lake Border morainic system, which partially bounds the Lake Chicago lake plain. The Chicago and State project site is located within the lake plain. The Blodgett, Deerfield and Park Ridge strata are ice margin deposits. Deposition of these strata occurred under water, when Lake Chicago was at its highest elevation of approximately 195 m above mean sea level. Each morainic stratum can be identified by their sediment properties that resulted from deposition in a particular sub-environment. The depositional environment determined the fundamental geotechnical properties of the glacial deposits including texture, particle shape, fabric, density and stress history. Chung and Finno (1992) found that the Blodgett, Deerfield, and Park Ridge strata are distinguished by water content and undrained shear strength and that the geotechnical parameters reflect the variability of the sediment properties.

Figure 3-3 shows the geologic subsurface conditions for the Chicago and State project site. The figure presents the geologic units and the generalized site stratigraphy. The elevations in Figure 3-3 are given in terms of Chicago City Datum (CCD). An elevation of 0 m CCD corresponds to the mean average level of Lake Michigan. The geologic strata encountered at the Chicago and State project site include:

• The Blodgett stratum is supraglacial in origin, which implies deposition from the upper surface of the ice. A relatively wide range of water contents and liquid limits characterizes supraglacial deposits. The Blodgett stratum is subdivided into two sub-strata based on the undrained shear strength. The Upper Blodgett stratum consists of a desiccated crust and underlying soft clays with undrained shear strengths that increase with depth. The desiccated clay crust was formed as a result of a drop in the water level after deposition. The Lower Blodgett stratum consists soft clays with slightly higher undrained shear strengths that continue to increase with depth. The average thickness of the Upper Blodgett stratum is 3.4 m while that of the Lower Blodgett stratum is 2.4 m.
  
  
• The Deerfield stratum consists of medium stiff clay and is characterized by uniform water contents. The Deerfield stratum is about 3.7 m thick. The excavation bottoms out in the Deerfield stratum at elevation -7.9 m CCD.
  
  
• The Park Ridge stratum is a stiff to very stiff clay with lower water contents than the Deerfield stratum. The Park Ridge stratum is about 4.6 m thick. The Deerfield and Park Ridge strata are either basal melt-out tills or waterlain paratills.
  
  
• The Tinley stratum, a lodgement till, underlies the ice margin deposits and consists of very stiff to hard clays and silts. The hard soils encountered below elevation -18.3 m CCD are known locally as "hardpan." The lodgement tills are often characterized as being very dense to hard and relatively incompressible soils.

With the exception of the clay crust in the Upper Blodgett stratum, the Lake Border deposits are lightly overconsolidated as a result of lowered groundwater levels after deposition and aging.

From Figure 3-3, it is seen that a fill deposit overlies the glacial clays. The fill is mostly medium dense sand, but also contains occasional construction debris. The fill deposit is approximately 4.3 m thick.

3.2.2 Laboratory and Field Testing

Several researchers had developed stratigraphic correlations for the geologic strata in the Chicago area based on index tests and undrained shear strengths (Peck and Reed, 1954; Otto, 1963; and Chung and Finno, 1992). A profile showing the variation of undrained shear strength and index properties with depth is presented in Figure 3-4. The field vane data, unconfined compression data, pocket penetrometer data, and much of the index properties data were obtained from samples of the Blodgett, Deerfield, and Park Ridge strata at or adjacent to the Chicago and State project site. The triaxial test data and some additional index testing data were obtained from test specimens of the geologic strata found at the Chicago and State site. Kawamura (1999) and Roboski (2001) performed extensive laboratory testing on specimens from the Blodgett, Deerfield, and Park Ridge strata including index property, consolidation, and triaxial tests. The index properties included natural water content, Atterberg limits, and specific gravity at various depths. The triaxial testing included drained triaxial compression (TXC) and reduced triaxial extension (RTXE) testing and undrained-consolidated triaxial compression (TXC CIU) testing. Table 3-1 presents the summary of the index tests. Table 3-2 summarizes the consolidation properties. The summary of the drained and undrained triaxial tests is given in Table 3-3.

TABLE 3-1. SUMMARY OF AVERAGE INDEX PROPERTIES (AFTER ROBOSKI, 2001)

Definitions:
wn: Natural moisture content
LL: Liquid limit
PL: Plastic limit
SG: Specific gravity

TABLE 3-2. SUMMARY OF CONSOLIDATION PROPERTIES (AFTER ROBOSKI, 2001)

Definitions:
Cc: Compression index sp’: Pre-consolidation pressure
Cr: Recompression index OCR: Overconsolidation ratio
svo': Effective overburden pressure eo: Initial void ratio

TABLE 3-3. SUMMARY OF TRIAXIAL TEST DATA (AFTER ROBOSKI, 2001)

Drained Triaxial Test Data

Undrained Triaxial Test Data

3.2.3 Engineering Properties of the Soil

It is apparent from Figure 3-4 that the natural water content of the Deerfield stratum is relatively constant. The range of values is from roughly 20 percent to about 22 percent. The natural water content of the Blodgett stratum is more erratic and higher than the Deerfield. These values range from about 18 percent to about 32 percent. The erratic behavior of the Blodgett stratum water content values reflects the fact that these sediments resulted from supraglacial deposition. The natural water content continues to decrease with depth below the Deerfield, such that the water content of the Park Ridge stratum varies between 16 percent and 18 percent. Figure 3-4 shows that the natural water contents of the Blodgett stratum are near the liquid limits, suggesting that the layer is normally to lightly overconsolidated. Conversely, the natural water contents of the Deerfield and Park Ridge are closer to the plastic limits, suggesting these layers are slightly more overconsolidated than the Blodgett.

From the index properties, the glacial clays at the Chicago and State project site were classified as lean clay (CL) using the United Soil Classification System (USCS). In general, the index properties and the consolidation results presented in Table 3-1 and Table 3-2 reflect that the strata become less compressible as the depth increases. The results presented in Table 3-3 indicate the shear strength increases with depth. The specific trends obtained from the three tables are as follows:

• The initial void ratio (e_o) decreased with depth from about 0.79 to about 0.43.
• The specific gravity (S_G) increased slightly with depth from 2.63 to 2.76.
• The maximum preconsolidation stress (σ_p') increased with depth from about 85 kPa at 6 m to about 200 kPa at 18 m.
• The compression index (C_c) decreased with depth from about 0.25 to about 0.106.
• The TXC drained friction angle (φ^') increased with depth from about 24 degrees to about 32 degrees.
• The RTXE drained friction angle (φ^') decreased with depth from about 42 degrees to about 35 degrees.
• The normalized undrained shear strength (S_u/σ_vc') increased with depth from 0.34 to 0.54.

It is observed that there are some occasional variances in the overall trend of stiffer and less compressible strata with increasing depth. Roboski (2001) noted that this was possibly due to sample disturbance.

3.3 ADJACENT STRUCTURES

The existing structures adjacent to the project site influenced many of the design decisions relating to the selection of the excavation support system. The potential for damaging these structures was estimated based on the proximity of these structures to the excavation, the building type, the foundation system, and the expected magnitude and distribution of the ground movements. The Frances Xavier Warde School was of primary concern due to its proximity to the excavation.

The school was built in the late sixties and is a 3-story reinforced concrete frame structure. When viewed in plan, the building has an "L" shape. The floor system at each level consists of a reinforced concrete pan-joist system, supported by reinforced concrete beams. The beams are supported on concrete columns at interior locations and masonry bearing walls and columns around the perimeter. The bearing walls rest on a reinforced concrete basement wall. The basement wall is nominally 2.8 m tall and 400 mm thick and is supported by an approximately 1.2 m wide continuous footing. The interior columns are supported on reinforced concrete spread footings that are nominally 760 mm thick and vary in size from 4 m by 4 m to 5 m by 5 m. The average depth bottom of the footings is approximately 3.7 m below ground surface. The building façade on the north, south, and west elevations of the building consists of windows and coarse limestone with a concrete masonry unit (CMU) backup wall. The east elevation of the building consist of a Chicago common brick veneer wall and a CMU backup wall. The school was located approximately 2 m from the excavation. The proximity of the school relative to the excavation is shown in Figure 3-2.

Also adjacent to the excavation, is the 125-year-old Holy Name Cathedral. In plan view, the cathedral is a cruciform. The structure is primarily a masonry bearing wall system with the exception of the central spire, which is supported on drill piers that extend to bedrock. The façade of the cathedral consists of limestone masonry and ornate masonry features. The steep roofs are cover with slate shingles. The cathedral was located approximately 15 m southeast of the excavation. Although of initial concern, measurements recorded throughout the project indicated that the excavation-related deformations at the cathedral were insignificant.

The Chicago and State Street subway tunnel and station were the belowground structures adjacent to the excavation. The subway station and tunnel were constructed between 1939 and 1941. Excavation was performed using the liner-plate tunneling method. The tunnel consists of twin subway tubes and passenger platforms and is symmetrical about its centerline. The tunnel travels in the north and south directions. Each tube is approximately 5 m wide and 6 m tall in the interior and each passenger platform is 2 m wide and 5 m tall in the interior. The bottom elevation of the tunnel is 9 m CCD. The tunnel was located approximately 4.5 m west of the Warde School. The existing subway tunnel increased the overall stability of the excavation because of its mass and stiffness.

3.4 EXCAVATION SUPPORT SYSTEM DESIGN METHODOLOGY

The principal focus of the excavation support system design was to protect the Warde School from excavation-related damage. Given the proximity of the school to the excavation, a major design issue was the movement associated with construction. The continuous wall footing, along the west side of the Warde School, was about 1.5 m from the centerline of the secant pile wall. Because of this close proximity, it would have been extremely difficult to perform the excavation without causing any damage to the school. Thus during the design phase of the project, the owner of the school and the designers agreed that the support system would prevent structural damage to the Warde School while permitting minor architectural damage. The intent was that the architectural damage would be repaired at the end of the project. It became apparent that a stiff excavation support system was needed given the soft clays at the site and the requirement of minimizing movements.

A complicating factor faced by the designers was that the CDOT contractually dictates that all temporary support is the responsibility of the contractor. Consequently, the designers were reluctant to specify excavation procedures and design details in the bid documents. Rather, the bid documents contained a performance specification, a conceptual design, a lateral earth pressure loading diagram, and several deformation limits that, if exceeded, would trigger specific responses from the contractor.

The deformation limits were defined by:

1. Computing maximum horizontal wall deformation using the charts developed by Clough et al. (1989). This method relates system stiffness (defined as EI/h⁴γ_w where EI is the bending stiffness of the wall, h is the average spacing between support levels and γ_w is the unit weight of water) and factor of safety against basal heave to the maximum horizontal wall movement. This value was reduced to account for the 3-dimensional nature of the excavation along State Street using recommendations of Ou et al. (1996);

2. Assuming the maximum horizontal movement equals the maximum settlement and computing the vertical settlement distribution behind the wall by the procedure recommended by Hsieh and Ou (1998), and;

3. Using the expected settlement profile as input to a finite element model of the Warde School to evaluate the extent of damage to the building. The potential extent of damage was defined and evaluated in terms of the computed tensile stresses.

The potential damage obtained from the finite element simulation was compared to the range of damage deemed acceptable by the owner. If the potential damage exceeded the acceptable range, a stiffer wall support system was used and the evaluation process was re-started at step 1. The process was iterated until the estimated level of damage was within an acceptable range. Deformation action limits and corresponding actions were established for the project based on the results of this process and the condition survey of the building. These limits and corresponding actions are given in Table 3-4. The movements shown in the table could be either lateral movement or settlement. A number of aspects of this project were not specifically considered when predicting the maximum horizontal movement, and therefore created uncertainties in the prediction. These aspects and their effect on the predicted movements are summarized in Table 3-5.

Table 3-4. Project Performance Specifications Related to Movements

Table 3-5. Project Details and Expected Effects on Predicted Movements

While the effect of some of these aspects could have been evaluated with a finite element analyses, time constraints precluded such analyses. Furthermore, even if such analyses had been conducted, significant uncertainty would have remained. No guidance could be found from local practice because this was the first time this type of support system was used in the soft Chicago clays. These uncertainties and the lack of local experience were the reason why the instrumentation program was implemented.

3.5 EXCAVATION SUPPORT SYSTEM DESCRIPTION

The primary excavation support system used for the project consisted of a secant pile wall with three levels of support. Figure 3-5 presents a section view of the excavation support system. The plan view showing the excavation support system was presented previously in Figure 3-2. The wall consisted of overlapping 915-mm-diameter drilled shafts installed along the east and west sides of the excavation. Each shaft nominally overlapped adjacent shafts by 150 mm. A W24x55 section was placed in alternating shafts to provide additional stiffness to the wall. The top level of supports consisted of cross-lot pipe struts, which were shimmed against the upper waler. The steel pipe struts had a diameter of 610 mm and a nominal wall thickness of 17 mm. The center-to-center horizontal spacing between the struts was approximately 6.1 m. Tieback anchors were used for the second and third levels of support. The regroutable tiebacks were bundled steel tendons. The steel tendons consisted of four to five, 15-mm, 1860 MPa steel strands. The 150-mm-diameter tieback anchors were installed at a 1.5-m center-to-center spacing. They were installed at a 45-degree angle and had bond lengths of 9.1 m to 10.7 m. Unbonded lengths were at least 9.1 m. The bond zone was located in the stiff clay below elevation -12.5 m CCD. The bond length was estimated using an allowable interface strength of 125 kPa.

The portion of the excavation north of the Warde School was braced with two levels of struts. The upper level struts consisted of HP10x42 sections. The lower level consisted of HP12x53 sections. A soldier pile and lagging wall was placed along the north and south ends of the excavation.

The east-west cross-section through the excavation given in Figure 3-5 shows why the combined support was required at this site. The Warde School had a 3-m deep basement that precluded using tiebacks for the first level. Also, there was the need to place the first brace near the top of the wall to minimize movements. The presence of the tunnel precluded using cross-lot supports for the second and third levels. Tiebacks in the lower level were installed with the drilling equipment placed atop the tunnel, 3 m away from the face of the wall.

3.6 FIELD INSTRUMENTATION

The field instrumentation was divided into three groups; ground instrumentation, structural support instrumentation and instrumentation for the adjacent structure. The ground instrumentation included inclinometers and piezometers. The structural support instrumentation included load cells and strain gauges. Optical survey points, string pot extensometers, tiltmeters, and crack gauges were used to monitor the response of the Frances Xavier Warde School to the construction activities. The ground instrumentation was installed and initialized prior to any construction activities at the site. Data from the ground instrumentation were obtained daily during the wall installation and excavation activities and at least weekly during the station renovation and backfill activities. The excavation support system instrumentation was installed as soon as the particular structural element was placed. Data were collected from these instruments on the same schedule as that of the ground instrumentation. The adjacent structure instrumentation was also installed and initialized prior to any construction activities. Optical surveys were made at least weekly during the project. The tiltmeters and extensometers were queried several times per day by remote access throughout the project. The crack gauge data were collect weekly after the onset of damage was observed.

3.6.1 Ground Instrumentation

Lateral movements of the soil behind the secant pile wall were recorded using five inclinometers located around the excavation site. The locations of the five inclinometers relative to the site are shown in Figure 3-2. Lateral ground movements were monitored around the Warde School with Inclinometer 1 and Inclinometers 2 placed along the east side of the excavation and Inclinometer 5 placed around the north side of the school. The free field lateral movements were measured with Inclinometer 4 placed on the west side of the excavation. Inclinometer 3 was located on the southwest corner of the school. This inclinometer was installed to monitor deformations that might affect the cathedral.

Inclinometer 1, Inclinometer 2, and Inclinometer 3 were all installed to depth of 22.5 m (elevation -18.3 m CCD). Inclinometer 5 and Inclinometer 4 were both installed to depth of 24.4 m (elevation -20.1 m CCD). For all inclinometers, the primary direction of inclination was towards the adjacent excavation and the secondary direction was 90 degrees to the primary direction. According to the manufacturer, the accuracy of the inclinometer measurements was ± 6 mm per 25 m of casing.

Pore pressures were measured using pneumatic piezometers at two locations. The piezometers were placed at opposite sides of the excavation, adjacent to Inclinometer 1 and Inclinometer 4. The piezometers were installed in nested pairs. At both locations, the two piezometers were separated vertically by about 3 m. Figure 3-6 presents the subsurface locations of the piezometers. The piezometers adjacent to Inclinometer 1 were installed at nominal elevations of -5.8 m CCD and -8.8 m CCD within the soft to medium clay. The piezometers adjacent to Inclinometer 4 were installed at nominal elevations of -6.4 m CCD and -9.4 m CCD. Note the elevation of the bottom of the excavation was at -8 m CCD.

3.6.2 Structural Support Instrumentation

Strut loads were determined from measurements of strain in the struts. Vibrating wire strain gages were attached to Struts 3, 4, and 5. Three strain gauges were attached to each strut at 120-degree intervals around the circumference. The strain gauges on Struts 3 and 4 were placed at a distance from the east end of approximately 2/5 the length of the strut. The gauges on Strut 5 were placed in the middle of the strut. Temperature corrections were applied to all strain measurements.

Loads in the tieback anchors were determined using hollow-core load cells mounted directly onto the tieback anchors. Four load cells were installed at two separate locations along the east wall. For both locations, one load cell was install on an upper level tieback and one load cell was installed on a lower level tieback. The locations of the load cells corresponded to Strut 3 and Strut 4. The locations of Struts 3 and 4 relative to the excavation site are shown in Figure 3-2. The load cells measured the forces in the tieback anchors directly.

3.6.3 Adjacent Structure Instrumentation

Figure 3-7 denotes the locations of the instrumentation used to monitor the response of the Warde School. Settlement was monitored by optical survey points established on the interior columns, on the roof, and along several locations on the exterior walls of the Warde School. The interior column survey points were all placed in the basement level. There was also several survey points located on top of the secant pile wall. The survey was conducted to an accuracy of about ± 3 mm.

Tiltmeters were installed along the perimeter walls in the basement level of the Warde School to measured inclination of the foundation wall directly. The tiltmeters were mounted on baseplates that were attached to the wall and were typically placed about 1.5 m above the basement floor. Tiltmeters T1, T2, T3, and T6 measured tilt in the north-south direction. Tiltmeters T4, T5, T7, T8, and T9 measured tilt in the east-west direction.

The figure also shows that extensometers were mounted along the wall in the basement level. These devices were string pot potentiometers mounted on baseplates and measured length changes of the foundation wall directly. Extensometers S5 and S6 were horizontal to the wall and Extensometers S1 and S2 and, S3 and S4 were placed diagonally along the wall. Measurements obtained from the extensometers were not corrected for temperature effects and were exposed to variations in the temperature during the duration of the project.

Consequently, accurate estimations of cumulative elongation or shortening of the basement foundation walls could not be obtained. Although there were occasional spikes in the data, estimates of the incremental changes in length of the basement wall were negligible. The extensometer data was ultimately not use for any analyses because the uncertainty in the quality of the data.

A crack gauge was used to monitor a diagonal wall crack located on the south wall of a third level classroom. The classroom was located at about the center of the building, just north of the C5 location (refer to Figure 3-7). The crack gauge monitored the change in width of the crack in both the horizontal and vertical directions

3.7 SUMMARY

The excavation along State Street was approximately 40 m long and 24 m wide and was advanced to an average final depth of 12.2 m (elevation -7.9 m CCD). The excavation along Chicago Avenue was approximately 24 m long and 7 m wide and was advanced to a depth of 8.2 m (elevation -4 m CCD). The secant pile wall was approximately 18.3 m deep and was constructed by drilling overlapping 915-mm diameter shafts. W24x55 sections were placed in alternating shafts. Center-to-center spacing of each shaft was 750 mm. The pipe struts were installed without preload at a depth of 0.6 m below ground surface. The tiebacks were installed at a 45-degree angle, down to the stiff and hard clay. Only about one-half of the anchors on either side were regrouted. The minimum bond length for the east side tiebacks was 9.1 m. The minimum bond length for the west side tiebacks was 10.7 m. No tiebacks were used for the support system along Chicago Avenue and the secant pile wall system was only used for about half the south wall. Soldier piles and lagging were used for the rest of the wall at this location. Two levels of cross-lot struts were installed without preload along Chicago Avenue at depths of 0.9 m and 4.3 m.

The soils at the Chicago and State project site are primarily lightly overconsolidated glacial clays. The clay layers are distinguished by water content and undrained shear strength, with the strata becoming stiffer and stronger as depth increases. The Frances Xavier Warde School was located approximately 2 m from the excavation and was instrumented with settlement points located on interior basement columns and along the exterior foundation walls. Lateral movements of the soil behind the secant pile wall were recorded using two inclinometers located on the west side of the school, one inclinometer located on the north side of the school, one inclinometer located on the west side of the excavation and one inclinometer located at the southeast corner of the excavation. Pore pressures were measured using pneumatic piezometers placed at opposite sides of the excavation. Strut loads were determined from vibrating wire strain gages attached to Struts 3, 4, and 5. Loads in tiebacks were determined from load cells mounted on both upper and lower level tiebacks at the Strut 3 and Strut 4 locations.

Home | About | Research | Technology Transfer | Education | Managament & Policy | Information Resources | Contact

Last updated: June 18, 2007    © 2005 Infrastructure Technology Institute