Final Report
Pilot Project for Fixed Segmentation of the Pavement Network
UCPRC– RR- 2005/ 11
Part of PPRC Strategic Plan Item 3.2.4
By:
Erwin Kohler
Nick Santero
John Harvey
Prepared for
California Department of Transportation
Division of Research and Innovation
Office of Roadway Research
UC Pavement Research Center
University of California, Davis and Berkeley
December 22, 2005
i
Executive Summary
This report presents the results of the study, “ Pilot Project for Fixed Segmentation
of the Pavement Network.”
The goal of this pilot project was to study a small sample of the California
Department of Transportation ( Caltrans) network to determine the feasibility of
expanding the pilot approach to the entire pavement network. The project’s work
included evaluating the effectiveness of ground penetrating radar ( GPR) and limited
coring for measuring pavement layer thicknesses and types, application of an algorithm
for determining “ fixed” segmentation of the pilot network, population of a database for
the pilot network, then assessing costs of these activities.
Fixed segmentation for use in the Pavement Management System ( PMS) is
required to develop the capability to do pavement performance modeling, which is
essential for the following pavement management tasks:
• Predicting future performance of segments of the network, and
• Identifying the most cost- effective maintenance and rehabilitation ( M& R)
strategies based on life- cycle costs.
Pavement layer- type and thickness data are also needed to develop effective
pavement performance models and to conduct effective condition surveys of composite
pavements ( asphalt overlays of PCC pavement). The data are also useful for project- level
engineering.
Background information summarizing the experiences of several other states in
using GPR for pavement work is also presented.
The pilot network consisted of a total of eight roadways: three interstate highways
( I- 5, I- 505, and I- 80), four state routes ( SR- 16, SR- 45, SR- 99, and SR- 113), and one U. S.
highway ( US- 50). The roadways chosen are mostly in District 3, except for the I- 80
section and part of the I- 505 section, which are both in District 4. GPR data was collected
on 681 lane- miles of the pilot network and analyzed for 305 lane- miles. Traffic data was
obtained from Caltrans. Climate regions were determined from a recent map developed
by Caltrans and the University of California Pavement Research Center ( UCPRC).
The UCPRC collected coring data for some of the locations on the pilot network.
Some of the cores were provided to Infrasense, Inc., for GPR calibration and some were
ii
retained by the UCPRC for checking the accuracy of the layer thicknesses and types that
Infrasense determined from the GPR data. The UCPRC also collected available as- built
information and maintenance records, and the most recent Pavement Condition Survey
( PCS) data from Caltrans.
The UCPRC then used the data collected to develop fixed segmentation for the
pilot network, resulting in 236 segments for the 305 lane- miles analyzed, with an average
segment length of 1.27 miles.
Comparison of the cores retained by the UCPRC with the layer types and
thicknesses identified by the GPR showed that the GPR data was reliable, especially for
the top two layers of the pavement.
Extrapolation of the costs on the pilot network for data collection and analysis and
segmentation results in an estimate of approximately $ 7.0 million of contracted field
work consisting of GPR use and coring ( including collection and analysis), plus 12.3
person- years of additional analysis work to complete the segmentation for the entire
Caltrans 49,000 lane- mile network.
If Caltrans moves ahead with collection of pavement structure data and fixed
segmentation, it will be important to document as- built information in the structural
database as future maintenance, rehabilitation, and reconstruction work occurs, in order
to keep the database accurate.
Work beyond this pilot study is underway to determine:
• Whether PMS performance data can be used with the fixed segments to develop
reasonable performance histories for the segment, and
• Whether the performance models developed by the UCPRC from Washington
State Department of Transportation ( WSDOT) PMS data can be verified with
Caltrans PMS performance histories using the fixed network segments and other
necessary data developed in this pilot project.
iii
Table of Contents
Executive Summary ............................................................................................................. i
List of Figures .................................................................................................................... iv
List of Tables ....................................................................................................................... v
1.0 Introduction............................................................................................................. 1
1.1 Purpose of Work ................................................................................................... 1
1.2 Pilot Project........................................................................................................... 2
1.3 Scope, Schedule, and Status of Project Tasks....................................................... 3
2.0 Background............................................................................................................. 7
2.1 Network Segmentation for Pavement Management ............................................. 7
2.1.1 Performance Modeling................................................................................ 8
2.1.2 Challenges to Pavement Modeling Using the Current PMS Studied in This
Project…………………………………………………………………………….. 8
2.2 GPR Technology................................................................................................. 10
2.2.1 Brief Description of the Technology ........................................................ 10
2.2.2 Recent Experience with GPR by Caltrans and Other State DOTs............ 11
3.0 Draft Segmentation ............................................................................................... 15
3.1 Administrative Boundaries ................................................................................. 19
3.2 Traffic ................................................................................................................. 19
3.3 Pavement Structure ............................................................................................. 19
3.4 Climate Region ................................................................................................... 21
3.5 Condition Survey ................................................................................................ 22
4.0 Data Requirements and Resources Involved ........................................................ 23
4.1 Pilot Study........................................................................................................... 23
4.1.1 Traffic Database........................................................................................ 23
4.1.2 GPR Data .................................................................................................. 23
4.1.3 Coring ....................................................................................................... 26
4.1.4 As- builts.................................................................................................... 27
4.1.5 Climate...................................................................................................... 27
4.1.6 Condition Survey and IRI ......................................................................... 27
4.1.7 Database.................................................................................................... 28
4.1.8 Summed Effort.......................................................................................... 28
4.2 Extrapolated Cost and Effort to Whole Network................................................ 28
5.0 Discussion of results from the pilot project and remaining work......................... 30
iv
5.1 Utilization of Coring Data .................................................................................. 30
5.1.1 Core Sites.................................................................................................. 30
5.1.2 Determining Exact Core Locations........................................................... 31
5.1.3 Core Results .............................................................................................. 32
5.2 Comparison of Current Caltrans Maintenance Dynamic
Segmentation versus Fixed Length Segmentation.............................................. 35
5.3 Collection of Pavement Condition Indicators for Performance
Modeling in PMS................................................................................................ 37
6.0 Conclusions, Future Work, and Recommendations.............................................. 38
6.1 Conclusions......................................................................................................... 38
6.2 Recommendations............................................................................................... 38
7.0 References............................................................................................................. 40
Appendix A: Minutes - 8/ 30/ 04 Meeting on Developing Objectives for the Highway
Network Segmentation & Data Collection in D- 3 Using GPS……………………….... A- 1
Appendix B: GPR Survey Summary………………………………………………….. B- 1
Appendix C: Charts with GPR Structure Results and Data from the 2003 Pavement
Condition Survey……………………………………………………………………..... C- 1
Appendix D: Segmentation Results……………………………………………………. D- 1
Appendix E: GPR Data and UCPRC Core Comparison: Plots…………………...……. E- 1
Appendix F: GPR Data and UCPRC Core Comparison: Tables……………………….. F- 1
Appendix G: Recommendation for Changes to Caltrans Pavement Condition Survey.. G- 1
List of Figures
Figure 1. Location of roadways in relation to the entire network.................................... 17
Figure 2. Location of roadways in North Central, California.......................................... 18
Figure 3. Example of GPR data taken on a GPR section of I- 5 in Sacramento County.. 21
Figure 4. GPR equipment used on this project. ............................................................... 24
Figure 5. GPR versus core ( UCPRC) thicknesses. .......................................................... 34
Figure 6. Histograms of sections versus length with Caltrans segmentation. ................. 36
Figure 7. Histogram of sections versus length with fixed- length segmentation.............. 36
v
List of Tables
Table 1. Initial Sections for Segmentation Pilot Project............................................. 16
Table 2. Personnel Needed and Direct Cost for Segmentation of Pilot Study and
Estimated for Entire Caltrans Network..................................................................... 29
Table 3. Final List of GPR Coring Locations ............................................................. 30
Table 4. Selected Coring Locations and Physical Reference Points........................... 31
Table 5. Average Thickness Differences ( Absolute) and Standard Deviations.......... 33
1
1.0 INTRODUCTION
This report presents the findings of the study, “ Pilot Project for Fixed Segmentation of Pavement
Network.”
The goal of this pilot project was to study a small sample of the California Department of
Transportation ( Caltrans) network to determine the feasibility of expanding the pilot approach to
the entire pavement network. The project’s work included evaluating the effectiveness of ground
penetrating radar ( GPR) and defining “ fixed” pavement segments, then assessing costs of these
activities.
The work was conducted as part of the Partnered Pavement Research Center ( PPRC)
Strategic Plan Item 3.2.4 (“ Development of Integrated Databases to Make Pavement
Preservation Decisions”) for the following objectives.
• To populate databases with existing data, and perform preliminary analyses.
• To develop recommendations for ongoing collection and database management
procedures to be implemented and operated by Caltrans functional units.
Work underway on analysis of the data collected is part of, or coordinated with, activities
in PPRC Strategic Plan Items 3.2.5 (“ Documentation of Pavement Performance Data for
Pavement Preservation Strategies and Evaluation of Cost- Effectiveness of Such Strategies”) and
4.5 (“ Calibration of Mechanistic- Empirical Design Models”).
1.1 Purpose of Work
The general purpose of the work presented in this report is to support Caltrans Maintenance in its
development of an improved Pavement Management System ( PMS). Specific objectives
focus on helping Caltrans Maintenance develop the capability to do pavement performance
modeling, which is essential for the following pavement management tasks:
• Predicting future performance of segments of the network.
• Identifying the most cost- effective maintenance and rehabilitation ( M& R) strategies
based on life- cycle costs.
More specific purposes of the work address three key challenges to performance
modeling using the current Caltrans PMS.
2
1. The use of “ dynamic segmentation,” which has logistical benefits but masks the true
performance of fixed segments and confounds performance modeling. The current
system uses a “ dynamic segmentation” procedure in which the pavement is not
evaluated over fixed lengths but is divided into segments that have similar distress at
the time of each assessment. Consequently, both the segment’s length and its starting
and ending points change from year to year, and a given pavement section is
identified as appearing in a different segment from one year to the next.
2. The PMS database lacks subsurface pavement structure data, which is a key variable
in explaining pavement performance. Pavement structure cross- section data is not
available in any central or district Caltrans database and it is not routinely updated
when rehabilitation and maintenance activities are performed.
3. The nonexistence of quantification ( severity and/ or extent) for some pavement
distresses, which means that these distresses are observed and identified in the
Pavement Condition Survey ( PCS), but they are not measured.
1.2 Pilot Project
To meet the stated objectives, a pilot project was developed in which a small representative
sample of the Caltrans network in Districts 3 and 4 was selected for field testing and other data
collection and analyses. These efforts aimed at evaluating:
• The effectiveness of using ground penetrating radar ( GPR) data, limited coring, and
available collected office data to provide an uninterrupted measurement of pavement
thickness and layer type on a variety of pavement types.
• The effectiveness of establishing static, well- defined ( fixed) network segments using
the GPR and other data collected on the pavement structures — combined with
traffic, climate, and condition survey, and roughness data.
• The costs of collecting the data and performing the segmentation and extrapolation of
those costs to the entire pavement network.
3
This report presents the data collected and the results of the analyses performed to
complete these three evaluations. Work will continue outside this pilot project, with additional
analyses to be performed to definitively conclude:
• Whether PMS performance data can be used with the fixed segments to develop
reasonable performance histories for the segment, and
• Whether the performance models developed by the UCPRC from Washington State
Department of Transportation ( WSDOT) PMS data can be verified with Caltrans
PMS performance histories using the fixed network segments and other necessary
data developed in this pilot project.
1.3 Scope, Schedule, and Status of Project Tasks
Specific tasks to be completed for this pilot project were identified in Meeting Minutes from
August 30, 2004, “ On Developing Objectives for the Highway Network Segmentation & Data
Collection in District 3 Using GPR” ( Appendix A). The initial project scope was shown as
follows:
A. Collect GPR data on identified sections in Districts 3 and 4
• Collect approximately 1,000 lane- miles of data, analyze approximately 300 lane-miles,
and retain the remaining raw data for potential analysis later.
• Include ( a) low- volume and high- volume traffic segments, and ( b) rigid, flexible, and
composite pavement structures.
􀃆 Tasks completed and items are presented in this report. Routes identified to include
1,000 lane- miles actually consisted of about 681 lane- miles when measured ( see Appendix
B).
B. Collect other data, including:
• The Caltrans Office of Pavement Rehabilitation’s studies of deflections,
• Project as- builts [ headquarters ( HQ) data, retrieval ( intranet) of District data],
• Data from moisture sensitivity studies, and
4
• Data from the Pavement Performance Evaluation Phase I ( Stantec Project1
􀃆 Tasks completed and utilized in this report.
• Coring Data
· Some samples are to be provided to the GPR contractor to calibrate the GPR data,
and others are to be held by the PPRC to verify GPR measurements
· Perform coring at only few locations, and only in sections where the GPR data
has been analyzed.
􀃆 Tasks completed for selected sites in Districts 3 and 4.
C. Perform analyses
• Analyze GPR data for thickness and layer type,
• Map the structures,
1 Stantec, Inc. was awarded a research project by the Caltrans Division of Research and Innovation to evaluate the
performance of in- service pavements in California and hence, the success of Caltrans' pavement design and
rehabilitation procedures. As part of this project, a large number of sections distributed throughout the state of
California covering different districts and environmental zones are being tested and many pavement related data
attributes are being collected. The test sections include rigid pavements, composite pavements and new and
rehabilitated flexible pavements. Phase II of this project is currently underway and is expected to be completed in
the summer of 2006.
4
5
• Revise GPR structures results based on coring data in areas where GPR identification
is questionable,
• Compare verification data with analyzed GPR data, and
• Analyze the costs.
􀃆 Task completed and results are included in this report.
Tasks A to C were scheduled to be completed in June 2005 and to be followed by:
D. Segment the 300 analyzed lane- miles by following a procedure ( described in the minutes)
that accounts for administrative units, pavement structure, climate region, traffic loading,
and condition survey, and ride quality data. Complete this item in August 2005.
􀃆 Task completed. PPRC performed a preliminary segmentation of the network based on
traffic, climate, pavement structure ( based on GPR data and verified by selected as- builts
and GPR core data), condition survey, and International Roughness Index ( IRI) data. The
results are included in this report.
Additional scope added to the project later by Caltrans Maintenance includes the
following tasks.
E. Extract historical condition survey and IRI data from the Caltrans PMS database for the
300 analyzed lane- miles.
􀃆 This task has been completed using the last available Caltrans Pavement Condition
Survey ( 2003– 2004) based on the fixed segmentation completed as part of Task D and
included in this report. Additional condition survey and IRI data are being extracted as part
of the PPRC Strategic Plan Item 3.2.5 from previous years and maintenance and
rehabilitation histories. A separate report will be delivered.
F. Check the accuracy of performance prediction models being developed as part of
Item 4.5 of the Strategic Plan for asphalt overlays on asphalt pavement, and IRI of
flexible and rigid pavement against extracted condition survey and IRI performance
histories.
6
􀃆 Completion of this task is not guaranteed because of the dynamic segmentation present in
the California PMS condition survey data. The data collected under Task E as part of the
PPRC Strategic Plan item 3.2.5 will be used in the attempt to complete this task, which is
scheduled to be completed in March 2006.
7
2.0 BACKGROUND
Adequate segmentation of a highway network is fundamental for the successful utilization of a
Pavement Management System ( PMS), in particular for the use of pavement deterioration
models. The homogeneous segments resulting from the segmentation process need to have a
consistent traffic level and a comparable pavement structure, and need to correspond to a single
climate region. ( Section 2.1 presents a detailed discussion of pavement segmentation.)
A key part of the segmentation process is the pavement structure, in terms of materials
and layer thicknesses. Since Caltrans does not presently have adequate inventory information
about the pavement structure throughout the network, the feasibility of using ground- penetrating
radar ( GPR) for this purpose is being evaluated. A brief literature review on GPR is presented in
Section 2.2.
2.1 Network Segmentation for Pavement Management
Long pavement segments in a PMS will generally be less uniform in composition ( i. e., there will
be more variation in pavement structure, condition, and other attributes within a segment) than
short segments. However, short segments require more data storage space because of the
increased number of segments. The final decision on size and method of segmenting should be
based on selecting pavement segments that Caltrans will consider as single entities when
planning maintenance and rehabilitation. The smallest number of segments that can adequately
define the road network will be the most economical and easiest to maintain.
As outlined in a previous report ( Lea and Harvey 2002), Caltrans first implemented a
PMS in 1977, when the concept of pavement management was relatively new and computers
were not as powerful as they are today. Over the subsequent twenty- five years, advances in
computer technology and significant changes in the theory and practice of pavement
management have changed the way pavements are maintained by Caltrans. These changes have
led to the slow evolution of the Caltrans PMS database and its use within the agency. In today’s
PMS literature, the Caltrans system would be referred to as a maintenance management system
because it is geared toward providing information for short- term maintenance activities rather
than long- term pavement performance assessment and modeling as well as optimization of
expenditures for the pavement network.
8
2.1.1 Performance Modeling
Performance modeling using PMS field data is essential for continuous improvement of two key
Caltrans pavement management tasks at the network level:
• Predicting future performance of segments of the network. “ Performance” refers to
pavement surface distress in the annual condition survey and ride quality ( IRI).
Future performance is predicted using models of distress and ride quality as functions
of existing condition, structure, traffic, and climate, and maintenance and
rehabilitation strategy selection.
• Identifying the most cost- effective maintenance and rehabilitation strategies based on
life- cycle costs. Life- cycle costs can be calculated for different conditions across the
state network, but the calculation requires the models described above to predict
performance at the network level plus cost data for each strategy.
At the network level, performance models derived from observations are “ empirical.” A
pavement performance model becomes “ empirical- mechanistic” when the explanatory variables
are selected based on the mechanics of pavement damage. To make these models useful for
Caltrans management, they must be calibrated using PMS field data. Compared to project- level
design, inputs for network performance modeling ( structure, traffic, and climate) need a lower
level of detail. Collecting data across the network with project- level detail would be cost-prohibitive.
Project- level PMS data for specific segments of the network is needed for calibrating
“ mechanistic- empirical” design procedures, which rely more heavily on pavement damage
mechanics theory. Detailed data for pavement structure, traffic, climate, materials, and
construction quality must be collected from the segments in order to predict their performance.
Those models must then be calibrated using historical PMS condition survey and ride quality
data.
2.1.2 Challenges to Pavement Modeling Using the Current PMS Studied in This Project
As mentioned at the beginning of this report, three crucial aspects of the current PMS are
addressed in this project to enable performance modeling ( at the network level) and to calibrate
design procedures ( at the project level).
9
1. The use of “ dynamic segmentation,” which constantly shifts frame of reference;
2. Lack of inputs needed for modeling, because the PMS does not contain data about
subsurface pavement structure; and
3. Inadequate quantification of pavement distresses, as some parameters in the
Pavement Condition Survey ( PCS) do not relate to pavement distress mechanisms or,
if observed, are only identified as present but are not measured.
The current Caltrans PMS staff inherited a “ dynamic segmentation” procedure
established in 1977 in which pavement is not evaluated over fixed lengths. Instead, the pavement
is divided into segments that have similar distress at the time of each assessment. Consequently,
both the segment’s length and its starting point and its ending point change from year to year. As
a result, a given pavement section is often identified as appearing in a different segment from
one year to the next. Often, segmentation from year to year changes based solely on the effects
of short- lived maintenance treatments that do not change the pavement cross section. Therefore
measured distresses and ride quality in the PMS database can vary as a function of segmentation,
depending on which sections of pavement are grouped together within the segment. Although
this is a good approach for scheduling maintenance, it does not lend itself to statistical sampling
of observed performance data or to predicting performance over time. It may also result in
inefficiencies for scheduling the rehabilitation of parts of a section as they fail over several years;
in reality, the entire section of which they are a part might be failing. Effective performance
modeling requires a network of “ fixed segments,” reasonably consistent pavement variables
( e. g., structure, traffic, climate), and similar maintenance and rehabilitation history.
The second challenge arises from the biggest problem with extracting pavement
performance information from the database: the database contains little information regarding
pavement structure. In some cases it contains data specifying whether the pavement surface is
flexible ( asphalt concrete) or rigid ( portland cement concrete). In other cases, the database
contains a generic description of apparent mix type, such as open- graded or dense- graded
asphalt. Missing are data about the true materials and layer thicknesses beneath the surface,
which are among the most important variables that explain pavement performance. Without
these, pavement performance models often give useless results or incorrect results.
10
The third challenge arising from the current PCS that Caltrans uses comes from its
inclusion of several variables whose presence or absence is noted but not measured, and from
others that have no meaning in terms of pavement distress mechanisms. This challenge can be
met by making some relatively minor changes in the PCS.
2.2 GPR Technology
2.2.1 Brief Description of the Technology
Ground- penetrating radar ( GPR) pavement- related technology, which was developed during the
Strategic Highway Research Program ( SHRP), operates by transmitting short pulses of
electromagnetic energy into the pavement. These pulses are reflected back to the radar antenna
with an amplitude and arrival time that is related to the thickness and material properties
( dielectric constant) of the pavement layers.
GPR technology has the potential of being extremely useful for pavement management,
allowing highway agencies to quickly collect inventory data on all pavements under their
jurisdiction. Because GPR data collection is nondestructive, it substantially reduces the need for
frequent full- depth pavement coring. Thickness determination of existing pavement layers
employing GPR is standardized in ASTM D4748.
GPR is a high- resolution geophysical technique that utilizes electromagnetic radar waves
to scan shallow subsurfaces, to provide information on pavement layer thickness or to locate
targets. The frequency of the GPR antenna affects the depth of penetration into the pavement.
Lower- frequency antennas penetrate further than higher frequency ones do, but the latter type
yield higher resolution. To successfully provide pavement thickness information or to scan an
interface, the following conditions have to be present ( Noureldin 2003): ( 1) The physical
properties of the pavement layers must allow for penetration of the radar wave, ( 2) the interface
between pavement layers must reflect the radar wave with sufficient energy for it to be recorded,
and ( 3) there must be a significant difference in the physical properties of the layers separated by
interfaces.
In NCHRP Synthesis 255 ( see References) the capabilities of GPR systems are listed as:
• Asphalt layer thickness determination: GPR results are used to estimate thickness to
within 10 percent; GPR accurately measures thicknesses of up to 0.5 m.
11
• Base thickness determination: Thicknesses are estimated, provided that a dielectric
contrast between the base and subgrade exists. ( The best results occur when the
subgrade is made up of clay soils, which are highly conductive compared to sands or
gravels.)
• Concrete thickness determination: Depth constraints and accuracy are not yet well
defined. This is because portland cement concrete attenuates GPR signals more than
asphalt does; PCC conductivity changes as the cement hydrates; reinforcing steel
contained in slabs makes interpretation difficult; and the dielectric contrast between
PCC and the base may not be adequate for reflection detection.
• Void detection: Although GPR has detected air- filled voids as thin as 6 mm, the
detection of water- filled voids is more problematic.
2.2.2 Recent Experience with GPR by Caltrans and Other State DOTs
As nondestructive testing has become an integral part of pavement evaluation and rehabilitation
strategies in recent years, Caltrans and other state highway agencies have looked into GPR
technology for network inventory and at the project level.
2.2.2.1 Caltrans
An evaluation of GPR and other non- destructive techniques for pavement thickness
evaluation was carried out for Caltrans by Infrasense, Inc ( 2003). The work focused on
determining quality control accuracy in newly constructed asphalt and concrete pavements. The
work involved theoretical analysis, laboratory testing on small slabs and simulated pavement
materials, testing at full- scale testing facilities, and actual testing on recently constructed
pavement sections in California. The actual testing was carried out on eleven selected pavement
sections, six of asphalt and five of concrete. Test sections were 305 meters ( 1,000 feet) long. The
asphalt sites were selected to represent three main conditions: ( a) thick and thin asphalt on
aggregate base; ( b) asphalt on concrete; and ( c) thick and thin asphalt overlays. The concrete
sites were selected to represent variations in concrete thickness and age. Age was selected as a
variable because of its influence on GPR penetration and on the mechanical wave velocity. The
asphalt sites were tested with the horn antenna ( typical GPR test) method and the common
12
midpoint, ( or CMP, a semi- static GPR) method. The concrete pavements were evaluated with
two different impact- echo devices, along with the CMP method. After this evaluation, cores were
taken for comparison with the test data. Twenty cores were taken at each asphalt site and ten at
each concrete site. The thickness values determined from the various test methods were
compared to the core values. The comparison showed generally good correlation, but at each site
a calibration was also needed. One core location per site was selected for calibration.
For asphalt pavement, the GPR was found capable of measuring the average section
thickness to within 2.5 mm ( 0.1 inches) of the average core value. This level of accuracy was not
achieved on concrete pavements.
2.2.2.2 Indiana
In 2001 the Indiana DOT ( Noureldin et al 2003) conducted experimental evaluation of the GPR
for network inventory by taking measurements at sections in five interstate highways ( I- 64, I- 65,
I- 69, I- 70, and I- 74), five U. S. highways, and nine state routes. GPR was used to test the truck
lane for both directions of traffic ( east- west or north- south) of each selected roadway at highway
speed. Although GPR can display pavement layer thickness continuously, it was decided to
collect thickness data at only five incremental locations ( every 1,000 ft, or 300 m) of each mile.
As part of the study, the researchers also obtained an estimate of the total pavement thickness
using FWD testing, which complemented data from the GPR tests regarding the thickness of the
top surface portion of the combined surface layers. Top surface portion thickness information is
very important for situations in which mill- and- fill operations are needed. The GPR estimates of
concrete pavement thickness, of hot mix asphalt ( HMA) thickness of flexible pavements, and
HMA thickness of composite pavements matched almost perfectly. GPR thickness estimate of
pavement layers underneath these layers was not as accurate and needs adjustment through very
limited coring. GPR did not provide any estimate of unbound pavement layers or of total
pavement thickness.
The relevant conclusions of the study are the following:
• Network- level testing employing the FWD and GPR is a worthwhile, technically
sound program that provides a baseline of the structural capacities of in- service
pavements.
13
• GPR is not expected to completely eliminate the need for coring, although GPR can
be used to establish the coring requirements, fill the gaps in thickness estimation, and
verify thickness results.
2.2.2.3 Virginia
Al- Qadi et al. ( 2005) report that GPR was used to evaluate the layer thicknesses of seventeen
pavement sites of different types ( flexible, continuously reinforced, and jointed plain concrete)
and different pavement ages ( up to five years old, between ten and fifteen years old, older than
twenty years with a surface less than ten years old; and older than twenty years with a surface
older than ten years). The sites were located in different parts of Virginia on major interstates
and high traffic- volume roads.
Analysis of the GPR data collected from all sites showed that for flexible pavements, the
GPR thickness error increased with pavement age ( 4.4 percent error for pavements up to five
years old to 5.8 percent error for pavements older than twenty years with surfaces older than ten
years). Comparison of sites of the same age but with different pavement types showed that
flexible pavements had a relatively high thickness error, while the jointed plain concrete
pavement ( JPCP) had the lowest thickness error. This could be mainly due to the presence of thin
HMA layers in flexible pavements ( these layers are significantly smaller than the GPR signal’s
wavelength) as well HMA layers of different ages. GPR considers layers with the same dielectric
constant as one homogeneous layer, thus sometimes introducing an error in the thickness
computation.
The study concluded that the error produced in predicting the thickness of HMA and
concrete is very reasonable, and that GPR accuracy in predicting pavement layer thicknesses
surpasses other available techniques — with the exception of coring, which is time- consuming,
has a very low coverage area, and is considered a destructive technique that requires traffic
closures.
2.2.2.4 North Carolina
In North Carolina ( Corley- Lay and Morrison 2001), thirteen LTPP ( Long Term Pavement
Performance) sites were tested one or more times with GPR to obtain layer thickness variability
14
over 152.4- m ( 500- ft) test sites. Duplicate runs were made on the same day on one of the sites,
and these paired tests were compared after the GPR data were processed. Five of the sites
showed good agreement with a Student’s t- test. Asphalt layers for the sites varied in average
thickness between 89 mm and 292 mm ( 3.5 and 11.5 in.). Thinner asphalt layers tended to have
lower coefficient of variation when the asphalt thickness was less than 152 mm ( 6 in.). The
standard deviation was generally less than 25 mm ( 1 in.).
2.2.2.5 Other DOT agencies
Other DOT agencies recently involved in verification of GPR technology are New Jersey
( Gucunski 2004), Missouri ( Cardimona 2003), and Kentucky ( Willet 2002). They report good
results for thickness determination. The Florida and Texas Departments of Transportation both
own GPR equipment. The Florida DOT uses GPR primarily to establish pavement thicknesses.
In Texas, the Materials Division of the Texas Transportation Institute ( TTI) has developed
performance specifications and test procedures for GPR systems. TTI has also developed a GPR
training program that has been used to train Texas DOT personnel in the two state districts that
own and use GPR.
15
3.0 DRAFT SEGMENTATION
The initial step in the segmentation pilot project was to identify highways and routes to be
analyzed in the study. A total of eight roadways were selected: three interstate highways ( I- 5, I-
505, and I- 80), four state routes ( SR- 16, SR- 45, SR- 99, and SR- 113), and one U. S. highway
( US- 50). The roadways chosen are mostly in District 3, except for the the I- 80 section and the
sourthern part of the I- 505 section, which are in District 4. The research team selected these
roads, which span five counties, for the pilot program because they believed that the extent and
diversity of their pavement sections fairly represented the entire state network. Only one lane per
selected route was chosen. For GPR purposes, the eight routes were converted into the twelve
sections — totaling 305 lane- miles — listed in Table 1. Figure 1 shows the pilot sections with
respect to the whole state network; Figure 2 shows the exact testing locations on a partial map of
the state overlaid with a GPS generated map ( using GPS coordinates obtained during the GPR
testing).
The segmentation process consisted of dividing the pilot network into homogeneous
segments based on administrative boundaries, traffic load, climate, pavement structure, and
pavement condition.
16
Table 1. Initial Sections for Segmentation Pilot Project
ID Route County Description Dir. Lane Length
( mi)
CAL009 SR- 99 Sacramento US- 50 to San
Joaquin Co line SB Out 25
CAL011 I- 5 Sacramento US- 50 to San
Joaquin Co line SB Out 24
CAL013 SR- 99 Sacramento/ Sutter From I- 80 to
SR- 70 split NB Out 16
CAL015 SR- 113 Yolo From Davis to
Woodland NB Out 11
CAL017 I- 5 Sacramento Yolo/ Colusa
line SR- 113 SB In 21
CAL031 I- 80 Solano Solano County WB Out 45
CAL033 I- 5 Sacramento/ Yolo SR- 113 to SR-
99 split NB Out 13
CAL035 SR- 16 Colusa/ Yolo Woodland to
SR- 20 WB Out 48
CAL041 I- 80 Solano Solano County WB In 45
CAL047 US- 50 Sacramento Sunrise Blvd. to
El Dor. Co line EB Out 11
CAL049 SR- 45 Yolo Yolo County SB Out 13
CAL050 I- 505 Solano/ Yolo I- 5 to I- 80 SB Out 33
17
Figure 1. Location of roadways in relation to the entire network.
18
Figure 2. Location of roadways in North Central, California.
The segmentation procedure consisted of five consecutive passes through the network to
progressively break down the entire length of the roadways into segments that share common
attributes. Section 0 presents the details of the data utilized in the segmentation of the pilot
project and the effort involved completing the tasks. [ Note: The segmentation process was
modified with respect to the minutes of the meeting on August 30, 2004 ( see Appendix A)].
Review of the data collected resulted in a change in the order of the segmentation passes and in
the decision not to use condition survey data in the process. Condition survey data was not used
because it was found to be less important for segmentation than originally thought due to its
temporary nature. The segmentation process was as follows.
19
3.1 Administrative Boundaries
The first segmentation pass consists of dividing the roadways into units based on district and
county boundaries. This step is based on past Caltrans practice of programming rehabilitation at
the district and county levels, which has resulted in different structures on each side of
boundaries.
In the pilot project, this pass meant dividing I- 505 at the line between District 4 and
District 3, and dividing SR- 99 and I- 5 at the Sacramento/ Sutter and Sacramento/ Yolo county
lines, respectively. This step increased the number of pavement segments to fifteen from twelve.
3.2 Traffic
The researchers divided segments within counties if there was a significant change in traffic
loading between them, hence major intersections served as natural boundaries between sections.
Intersections are permanent physical reference points that also help in locating the sections in the
field and can be used for assigning names to the sections they separate. Traffic data is also
required for assignment of priorities during the selection of rehabilitation projects. The current
Caltrans highway traffic database was used.
Dividing the network according to changes in traffic increased the number of sections
from 15 to 173. The process included intersections that do not currently affect traffic in the route
but that could eventually grow and become significant. The length of the new segments ranged
from 0.10 miles to 7.12 miles, with approximately 50 percent of them being less than 1.25 mile.
3.3 Pavement Structure
The next step was to divide sections with similar traffic into units that had comparable pavement
structure. This includes the surface types and the number and thickness of the layers that
constituted the pavement. Ideally the construction history would have been used to identify the
materials and the age of the pavement sections, but this information was not available for most
sections. Sources checked included as- built records, deflection study reports, and major
maintenance archive files.
The thicknesses obtained through the GPR testing on all the roadways, combined with
some as- built drawings and existing knowledge of the pavements, permitted the research team to
20
differentiate the sections at the points of change in their structure. The method used at this stage
for identification of the point of change was visual and without a statistical analysis because in
most cases the GPR data showed a clear distinction between sections that needed to be separated.
Statistical algorithms for automatic detection of changes based on GPR pavement structure data
will be tested later when checking performance of segments. Figure 3 shows an example of GPR
thickness and material for the section on I- 5 SB ( southbound) in Sacramento County. At
postmile 21.80 there was a change in AC thickness ( thickness on one side is approximately 5.7
inches and on the other it is 8.4 inches).
The figure also shows the IRI and cracking data from the 2003– 2004 PCS. The figure
shows alligator cracking data from the PMS database, which illustrates a problem for the
Pavement Condition Survey caused by the lack of structural data in the PMS. Alligator cracking
was surveyed because the surface of the pavement is asphalt; however, alligator cracking can
never occur in this pavement because it is a composite pavement consisting of an asphalt overlay
of PCC. The pavement condition surveyor has no way of knowing that this is a composite
pavement from the information available in the PMS. There is no option in the PCS for
evaluating a pavement as a Composite pavement, only Rigid or Flexible. Composite pavements
make up a significant portion of the Caltrans network, and made up 20 percent of the lane- miles
in this pilot project. Reflection cracking, the most common distress occurring on composite
pavements, is not included in the PCS.
21
0
5
10
15
20
25
30
35
40
28
27
26
25
24
21
20
19
18
17
16
15
14
13
12
10
Post Mile
Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base
PCC
AC
IRI
AC - Alligator B
Cracking
Figure 3. Example of GPR data taken on a GPR section of I- 5 in Sacramento County.
After the segmentation by pavement structure was done, the total number of segments
increased from 173 to 236. The length of the new segments remained between 0.10 and 7.12
miles, but the average length decreased from 1.68 miles to 1.27 miles. Approximately 50 percent
of the resulting segments at this point were shorter than 1.05 miles.
3.4 Climate Region
Differing climate regions ( per the Caltrans Climate Region Map, June 2005) were used as a
segmentation pass. Most of the sections were contained within the Inland Valley ( IV) climate
region. The exception was the westernmost ten miles of I- 80 in Solano County, which is in the
Central Coast ( CC) climate region. This pass resulted in one additional segment, increasing the
total to 237..
22
3.5 Condition Survey
The last pass of the segmentation process was to divide the section into homogeneous units from
the standpoint of pavement condition. However, analysis of the condition survey data indicated
that to rationally partition segments, consistent condition survey data over several years would be
necessary, and rehabilitation and maintenance records were needed to explain changes in
observed distresses. These could not be obtained within the schedule for this project. In the end,
distress and IRI data may not be needed to further divide the segments if the performance and
histories show that the segmentation- based administrative boundaries, traffic, pavement cross
section, and climate region result in reasonably homogeneous sections with relatively uniform
performance within them. Charts with GPR structure results and data from the 2003 Pavement
Condition Survey are presented in Appendix C.
The decision not to segment based on condition survey data is further supported by the
temporariness of certain maintenance procedures that may conceal existing damage. For
example, a slurry seal on a portion of a segment that has uniform structure, traffic, and climate
region, and has alligator cracking across its entire length, may show zero alligator cracking in the
PCS data for a portion of it because of the seal. However, the cracking remains and will come
through the slurry seal after several years. Segmentation based on PCS data such as this may add
inaccuracies to the process as time wears through various temporary maintenance solutions.
The segmentation is presented in Appendix D in the form of a table containing the
postmiles and the physical references for the resulting segments.
23
4.0 DATA REQUIREMENTS AND RESOURCES INVOLVED
A variety of data were collected and analyzed for the segmentation process. Sources included
private contractors, Caltrans documents, field work data, and project records.
4.1 Pilot Study
Resources employed in the segmentation of the pilot network are as follows.
4.1.1 Traffic Database
A record of the most recent traffic counts can be found on the Caltrans website. 4 Included in the
database is the average annual daily traffic ( AADT) at certain intersections, political boundaries,
and other unique landmarks, along with the corresponding postmiles. The points defined in the
traffic log created definitive segments: in the urban areas, these segments tended to be between
0.1 mile and 4.0 miles long; in rural areas, the segments could extend over 30 miles. For the GPR
sections covered by the PPRC in this study, the traffic sections typically remained small and only
a few extended beyond 5.0 miles long. None of them was over 10 miles long.
This data was used as the second pass for the segmentation, as explained in Section 3.0.
A Microsoft Excel version of the database is available on the web so no conversions are needed
in order to manipulate the data for this project. Once downloaded, locating the desired sections is
straightforward and takes very little time. For the twelve GPR sites considered in this pilot study,
the process took about three person- hours.
4.1.2 GPR Data
4.1.2.1 GPR data collection and equipment
GPR data was collected at a density of one scan per linear foot of travel. Although this may seem
excessive for network- level work, this data rate is desirable for two reasons: ( 1) according to the
contractor, pavement type ( JPC, CRCP, AC/ PCC, etc.) is more easily identified with denser data;
and ( 2) the data will be available for future project work where the denser scan spacing might be
more desirable. The data from this project have already been used for project- level analysis of
SR20, providing thicknesses for backcalculation of foamed asphalt stiffnesses.
24
The GPS system operated concurrently with the GPR data collection. GPS coordinates
were recorded once per second with the current GPR scan number in a separate position log file.
Data was collected at speeds of up to 60 mph. Two- hundred- and- fifty- six samples were
taken during 20 nanosecond scans using 16- bit data resolution. The 20- nanosecond range
provided the potential for layer- depth information capability down to 36 inches. This depth
generally exceeds the penetration capability of the GPR equipment.
The GPR equipment used on this project included a GSSI SIR- 20 radar control and data
acquisition unit, a Model 4108 1- GHz horn antenna, mounting equipment, and an electronic
distance- measuring instrument ( DMI) attached to the vehicle wheel ( see Figure 4). The DMI had
a resolution of 500 pulses per foot. The GPR equipment was approved and licensed by the FCC.
Also included was Trimble Model AG114 GPS, or an equivalent system, for recording GPS
coordinates. This GPS system provided submeter accuracy when used in a differential mode in
conjunction with the Omnistar service. According to the manufacturer’s specifications, the GPS
data obtained with this service is in NAD83- compatible format.
Figure 4. GPR equipment used on this project.
GPR data was analyzed by the contractor at 0.1- mile intervals — based on the vehicle
DMI — beginning at the county line or other marked reference point in each test section. GPS
25
coordinates were reported for each GPR data point analyzed. When the 0.1- mile interval point
fell on a bridge deck ( this was easily identified in the GPR data), a neighboring location on either
side of the deck was selected.
The results of the GPR analysis were provided in Microsoft Excel data files, one for each
section. The data reported at each location represents 200 feet of pavement, ± l00 feet on either
side of the reported location. Exceptions to the 200- foot length occurred when there was a bridge
deck or other anomaly in the pavement structure within the ± l00- foot interval. Where this
occurred, the interval was shortened to include only the pavement representative of the local
area.
Within each file, there are five columns for each analyzed layer, and up to four layers
analyzed. The five columns for each layer are described as follows:
• Layer type ( e. g., AC, PCC, base),
• Layer thickness ( average of 200 feet, in inches),
• Layer dielectric constant ( average of 200 feet, no units),
• Layer thickness standard deviation over 200- foot length ( inches), and
• Layer confidence.
The contractor assigned a number from one to four to each analyzed data point to reflect
his degree of confidence. An explanation of the numbering code follows:
1. Layer boundary and type is clear.
2. Layer boundary is unclear – calculated thickness may be affected.
3. Layer type is unclear – best assessment, but it is possible that identified type is
incorrect. For example, assigning a “ 3” to layer 2 when it is suspected to be AC but
might be Base.
4. A combination of 2 and 3.
4.1.2.2 GPR cost
Infransense, Inc., the contractor providing the GPR information, charged $ 30,923 for 305 lane-miles
of data, which included planning, mobilization, and the collection and analysis of the raw
data. The per- mile cost of data collection was $ 16.48, while the per- mile cost of analysis was
26
$ 51.15. The cost of planning and setup was $ 1,720; the cost of mobilization and demobilization
was $ 8,575.
4.1.2.3 Plotting results for segmentation
The GPR data received from the GPR contractor were easily plotted using Microsoft Excel.
Creation of charts showing cross sections of the twelve sections took eight person- hours.
4.1.2.4 Identification and segmentation of structure changes
Depth trends in the plotted GPR data are visually evident in most cases, making
identification of major structure changes possible by inspection. Material- type recognition by
dielectric constant is not an error- proof process and therefore uncalibrated GPR results do not
always properly identify material type. Structure changes based only on material type are
difficult to distinguish.
Once structure changes were identified on the charts, the exact corresponding postmile
was located in the GPR data and recorded in the segmentation database. Visually identifying the
structure changes, locating the precise point of the structure change in the GPR database, and
segmenting based on the structure changes took approximately 30 person- hours.
4.1.3 Coring
Coring was completed for thirteen sites: Twelve in District 3 and one in District 4. The sites were
cored on nine days between July 7 and September 16, 2005. Closures were performed by
Caltrans district Maintenance personnel. The internal cost of these closures to Caltrans is not
known. From UCPRC experience, a private contractor would charge approximately $ 2,000–
$ 3,000 per day for the closures.
A crew of six people was necessary for this work. The crew was responsible for running
the coring machine, using the Dynamic Cone Penetrometer ( DCP), recording data, and
backfilling the core- holes. Including travel, setup, and breakdown, this took approximately
60 person- hours. The DCP provided data regarding layer thicknesses below the depth of cores.
Details on the coring are presented in Section 5.1.
27
4.1.4 As- builts
An attempt was made at collection of as- built information for the GPR sections. Caltrans has
provided UCPRC with a limited number of as- builts. District offices were visited to find
additional as- builts. However, many segments did not have as- built records because of age, lost
documents, and work that has been performed but not recorded. Depending on the organization
of records and the availability of the necessary documents, this task could take up to 16 person-hours.
4.1.5 Climate
Most GPR segments for this project were located in the Inland Valley climate region, with two
sections split between the Inland Valley and Central Coast regions . Segmentation based on
climate boundaries is simplified by the Caltrans Climate Map, making time- demand for this step
negligible ( zero person hours). Caltrans Maintenance has developed a map that defines the exact
postmiles that define boundaries between climatic regions on each route for nearly the entire
state.
4.1.6 Condition Survey and IRI
Condition surveys, which include the IRI, are available from the Caltrans Pavement Management
System ( PMS) database. Though the GPR sections have not been segmented based on the
condition survey data, the pavement condition has been entered into the GPR database and it has
been used to generate charts for comparison showing pavement condition alongside the GPR
results. The plotted data includes the IRI, alligator B cracking ( AC), and third stage cracking
( PCC).
The raw data from the PMS database needed to be converted into a manageable format,
which took about 10 person- hours. This task was completed for the whole state highway
network. Loading the PMS data into the GPR database and outputting the resulting plots took
another 20 person- hours. In sum, the condition survey data took 30 person- hours.
28
4.1.7 Database
Development and population of the database for the pilot segmentation project took place at the
same time that the data was being retrieved from all the sources. A nominal one person- hour is
being accounted for database handling.
Information collected for segmentation is currently stored in Excel with location
identifiers tied to the distance measured from nearest physical reference, such as structures or
paddles, for which the exact GPS coordinates have been obtained. Soon, the data will be loaded
into a relational database ( Access) and delivered to Caltrans.
4.1.8 Summed Effort
The estimated time spent on the segmentation process for the twelve GPR sites sums to
148 person- hours. Other costs include the contract costs for the GPR ($ 30,923 for 305 lane-miles),
lane closures ( estimated to be between $ 14,000 and $ 21,000 if done by a private
contractor), materials for coring ( bits, backfill, etc.) and various travel costs.
4.2 Extrapolated Cost and Effort to Whole Network
The Caltrans 2003 State of the Pavement Report states that there are over 49,000 lane- miles of
pavement in the California highway network. If segmentation of 305 lane- miles requires
148 person- hours, then the whole network would take nearly 24,000 person- hours to complete.
This amount is approximately 12.30 PY ( assuming 1,940 hours per year). At an estimated rate of
$ 94.43 per mile, the GPR data collection, analysis, and calibration, the cost for contracting the
GPR testing over the entire network would be approximately $ 4.63 million, not including
mobilization. If mobilization is assumed to be 12 percent of the cost of testing, then the total
estimated cost of GPR testing and analysis can be considered $ 5.2 million.
The cost of lane closures needs to be added to that amount. At an assumed rate of
$ 3,000 per day, and considering about 600 days of closures to complete all the required coring,
the cost would be $ 1.8 million. This brings the total direct cost to an estimated $ 7.0 million. The
direct cost and personnel needed are shown in Table 2.
29
Table 2. Personnel Needed and Direct Cost for Segmentation of
Pilot Study and Estimated for Entire Caltrans Network
Item
Pilot study
305 lane- miles
( actual)
Caltrans network
49,000 lane- miles
( extrapolated)
Personnel 0.076 PY 12.30 PY
Direct cost $ 48,500 $ 7,000,000
The actual direct and personnel cost, both for field ( coring and GPR) and office work,
will likely be less than the figures stated above. Time spent retrieving data and segmenting based
on that data will drop significantly as personnel become increasingly proficient at the process.
Also, the cost per lane- mile of GPR measurement and analysis will decrease if a bid system to
determine the lowest price can be implemented.
It must be noted that the pavement structure database for the PMS that could be created
by a GPR project would lose its value over time unless it is routinely updated with accurate
information regarding the changes to pavement structures caused by future rehabilitation,
maintenance, and reconstruction.
30
5.0 DISCUSSION OF RESULTS FROM THE PILOT PROJECT AND REMAINING
WORK
5.1 Utilization of Coring Data
5.1.1 Core Sites
The coring for the GPR was completed on September 16, 2005. A total of 43 cores were
extracted from 13 sites in Districts 3 and 4. The original plan called for 16 sites with a total of 65
cores. The difference between these numbers is due to scheduling problems for the Caltrans
Maintenance force and time constraints that arose in the field. A summary of the coring locations
is shown in Table 3.
Table 3. Final List of GPR Coring Locations
Closure
No.
Section
ID County Route Direction Start End Coring Date
No. of
Cores
Data Given to
Infrasense?
1a CAL050 Solano 505 SB 8.10 8.40 9/ 16/ 2005 4
X
1b
CAL05
0 Solano 505 SB 5.00 5.40 Cancelled – time constraints in the field
2
CAL01
3
Sacrament
o 5 NB
27.7
0
28.2
2 Cancelled – could not get closure
3
CAL01
3 Sutter 99 NB 5.68 6.18 Cancelled – could not get closure
5 CAL015 Yolo 113 NB 2.89 3.20 8/ 22/ 2005 6 X
5a CAL015 Yolo 113 NB 8.40 8.70 8/ 25/ 2005 3
9 CAL047 Sacramento 50 EB 17.20 17.50 7/ 7/ 2005 4 X
10 CAL047 Sacramento 50 EB 20.01 20.31 7/ 11/ 2005 4
11 CAL049 Yolo 45 SB 10.80 11.10 8/ 24/ 2005 4 X
12 CAL049 Yolo 45 SB 7.82 8.12 8/ 25/ 2005 4
12a CAL049 Yolo 45 SB 9.00 9.30 8/ 24/ 2005 4
14 CAL009 Sacramento 99 SB 8.86 8.96 7/ 13/ 2005 2 X
15 CAL009 Sacramento 99 SB 6.26 6.36 7/ 21/ 2005 2
16 CAL035 Colusa 16 WB 3.04 3.14 7/ 25/ 2005 2 X
17 CAL035 Colusa 16 WB 1.84 1.94 7/ 25/ 2005 2 X
18 CAL035 Colusa 16 WB 0.64 0.74 7/ 25/ 2005 2
Most sites were chosen by Infrasense, Inc., and confirmed by the UCPRC. Sites were
chosen based on abrupt changes in the apparent pavement structure and uncertainties in the GPR
data. Two sites ( Closures 5a and 12a) were chosen strictly by the UCPRC for control purposes.
31
5.1.2 Determining Exact Core Locations
Determining exact core locations was critical to the success of the project. Cores taken in the
field needed to be matched up with the GPR results for exactly the same location so that an
accurate comparison of the two could be made.
Infrasense provided location data relative to a local physical reference at each site. This
data included a unique local reference point, distances from the reference point, and GPS
coordinates. Physical references were Caltrans postmile paddles, bridge decks, and obvious
changes in surface material ( i. e., from PCC to AC overlay). Examples of this data appear in
Table 4.
Table 4. Selected Coring Locations and Physical Reference Points
Closure
No.
Section
ID
County Route Dir. Approx
PM
Local
Physical
Reference
Dist
from
Ref. ( ft)
Latitude
( ddmm. mm)
Longitude
( dddmm. mm)
9 CAL047 SAC 50 EB 17.20 E Joint
Bridge
Deck
377.00 3838.402881 12111.69367
9 CAL047 SAC 50 EB 17.30 E Joint
Bridge
Deck
905.00 3838.416024 12111.58439
9 CAL047 SAC 50 EB 17.40 E Joint
Bridge
Deck
1433.00 3838.429634 12111.47536
9 CAL047 SAC 50 EB 17.50 E Joint
Bridge
Deck
1961.00 3838.443318 12111.36599
10 CAL047 SAC 50 EB 20.01 SAC RP
20
57.00 3838.513709 12108.50724
10 CAL047 SAC 50 EB 20.11 SAC RP
20
585.00 3838.518535 12108.39712
10 CAL047 SAC 50 EB 20.21 SAC RP
20
1113.00 3838.523859 12108.28717
10 CAL047 SAC 50 EB 20.31 SAC RP
20
1641.00 3838.528628 12108.17718
32
In the field, core locations were marked using a digital survey wheel taken from a given local
reference point. GPS measurements were taken at each core and used as a distance check once
the data was entered into the database. The database shows discrepancies between Infrasense’s
GPS coordinates and the UCPRC’s field GPS coordinates ranging from 4.7 feet to 158.6 feet.
These values can be found in Table 9 of Appendix F. Possible reasons for the discrepancies
include:
• Inherent inaccuracies in the UCPRC GPS receiver, which did not have differential
capability ( this device provided a typical error of ± 3m, with extreme error up to ± 30m
at some locations.);
• Inexact physical reference locations measured by Infrasense;
• On- the- fly measurements may be prone to inaccuracies;
• Mileposts were not necessarily in the same location in each direction;
• Equipment malfunction;
• Survey wheel was decommissioned by the UCPRC after CAL015 sites because of
malfunctions that might have affected previous sites; or
• Other human errors.
The “ Approximate Postmiles” were used as a check to ensure coring was done in the
right vicinity. They were also used to coordinate the closures with Caltrans maintenance yards.
Infrasense calculated the postmiles as a distance from certain physical features, such as paddles
or county lines. Because of the complexities in the Caltrans postmile system ( equations,
inaccuracies, etc.), coring locations were recorded independently of the approximate postmiles.
5.1.3 Core Results
Core layer thicknesses, layer material types, and DCP results were disclosed to Infrasense for the
seven sites noted in Table 3. This data was used by Infrasense to verify and calibrate both their
thickness and material- type results. After a review of the core data disclosed to them, Infrasense
33
determined that a systematic calibration was not necessary. However, the following changes
were made to CAL15- 5 by Infrasense:
• Layer 2 thickness ( PCC): Reduced by 12%;
• Revised GPR data locations to match UCPRC core locations — shifted ~ 0.1 miles to
account for GPS discrepancies in a few cases.
Minor changes were recommended by Infrasense for two sites that had not been disclosed to
them:
• CAL15- 5a: Reduce thickness of PCC layer by 12 percent
• CAL035- 18: Change layer 3 material to “ base”
After the changes were made, the GPR results were compared to the remaining cores.
DCP results were used to estimate underlying layer materials and very approximate thicknesses.
The comparisons can be found in Appendices E ( plots) and F ( tables).
Review of the results shows that the GPR technology is effective for determining layer
thicknesses for all layers. The accuracy level decreases with depth, with layers 1 and 2 being
more accurate than layers 3 and 4. The average thickness difference ( absolute percentage of total
layer) and accompanying standard deviations are presented in Table 5. A comparison of the GPR
versus core ( UCPRC) thicknesses is plotted in Figure 5.
Table 5. Average Thickness Differences ( Absolute)
and Standard Deviations
Layer 1 Layer 2 Layer 3 Layer 4
No. of Cores 31 22 15 3
Average Difference 12.62% 10.17% 27.88% 20.89%
Standard Deviation 11.2% 15.0% 23.4% 11.3%
34
0
5
10
15
20
0 5 10 15 20
GPR Thickness ( in)
Core Thickness ( in)
Layer 1
1: 1 Line
0
5
10
15
20
0 5 10 15 20
GPR Thickness ( in)
Core Thickness ( in)
Layer 2
1: 1 Line
0
5
10
15
20
0 5 10 15 20
GPR Thickness ( in)
Core Thickness ( in)
Layer 3
1: 1 Line
0
5
10
15
20
0 5 10 15 20
GPR Thickness ( in)
Core Thickness ( in)
Layer 4
1: 1 Line
Figure 5. GPR versus core ( UCPRC) thicknesses.
Some of the more extreme values in Figure 5 may be a result of the discrepancy between
the GPR readings location and core locations discussed in 5.1.2. At some sites, layer thicknesses
are highly variable over small areas, so even a small difference in between the GPR reading and
the core location can result in a large difference in layer thicknesses. These extreme values affect
the averages and standard deviations expressed in Table 5.
Layer types as indicated by the GPR reading matched up well with the UCPRC cores.
Deeper AC was sometimes recorded as “ Base” or “ BB” ( bituminous base). These layers
sometimes exhibited aging effects ( such as the breakdown of materials) that may have caused the
misnaming. The GPR was unable to differentiate between base types, including cemented bases
35
( LCB or CTB) or asphalt- treated permeable bases ( ATPB). Open- graded AC ( OGAC) layers
were not distinguished from other AC types and were grouped together with the underlying AC
layers. For example, if a layer consisted of 25mm OGAC and 100mm DGAC, the GPR output
would be 125mm AC.
5.2 Comparison of Current Caltrans Maintenance Dynamic Segmentation versus Fixed
Length Segmentation
The length of segments that Caltrans uses for evaluation of pavement condition was obtained
from the program, “ Pavement Condition Reporting System.”∗ Statistics were obtained for data in
the years 2000 and 2004 for about 305 miles of roadway for the pilot segmentation study.
Histograms with the number of segments in 0.1- mile intervals are presented in Figure 6 for each
of these two years. The charts show that the number of segments identified by the Caltrans
dynamic segmentation for the pilot network increased from 225 to 431 between 2000 and 2004,
and that average length dropped from 1.17 to 0.66 miles. One- mile segments seem to be the most
common survey unit.
The same chart was prepared for the fixed segmentation performed as part of this study
and is presented in Figure 7. It can be noted that the fixed- length segmentation produced
∗ Version 3.0.0 March 17, 2005.
36
segments whose lengths are spread over a wider range. Figure 7 shows segments only up to
5 miles long, but there were four additional sections between 5.0 and 7.3 miles long.
0
10
20
30
40
50
60
70
80
90
0.1 0.4 0.7 1.0 1.3 1.6 1.9 2.2 2.5
Section length ( miles)
Number of sections
Year 2004
Nr. of sections: 431
Average length: 0.66 mi
0
10
20
30
40
50
60
70
80
90
0.1 0.4 0.7 1.0 1.3 1.6 1.9 2.2 2.5
Section length ( miles)
Number of sections
Year 2000
Nr. of sections: 225
Average length: 1.17 mi
Figure 6. Histograms of sections versus length with Caltrans segmentation.
0
10
20
30
40
50
60
70
80
90
0.1 0.4 0.7 1.0 1.3 1.6 1.9 2.2 2.5 2.8 3.1 3.4 3.7 4.0 4.3 4.6 4.9
Section length ( miles)
Number of sections
PPRC Segmentation
Nr. of sections: 236
Average length: 1.27 mi
Figure 7. Histogram of sections versus length with fixed- length segmentation.
A comparison between the 2004 Caltrans segmentation versus the fixed- length
segmentation indicates that there would be roughly 200 fewer segments to survey ( 237 instead of
431) in the pilot network, and that the average segment length would be 1.27 miles rather than
0.66 miles. The validity of these conclusions is limited until segmentation by surface condition is
performed; however this indicates that fixed segments could result in fewer segments to survey,
reducing the cost of the Caltrans Pavement Condition Survey.
37
5.3 Collection of Pavement Condition Indicators for Performance Modeling in PMS
As mentioned in Section 2.1, collection of pavement condition data depends on whether the
information is going to be used for PMS or for project- level maintenance. In order to collect the
necessary data to develop or calibrate empirical models for pavement management and to
calibrate mechanistic- empirical pavement design models, some minor changes to the PCS
procedure have to be implemented. It seems that there are two possible approaches.
1. The first option is to continue with the current scheme of condition surveys, but to use
the fixed- length segments as breakpoints ( PCS hits the same ends as the PMS
segments) so the results can be tracked year after year. Since more than one PCS
segment is likely to be found within a PMS segment, a weighted average of the
condition in the segments, based on length, can be obtained to represent the condition
of the entire PMS segment.
2. The second alternative is to conduct condition surveys for PMS purposes,
independent of the Pavement Condition Survey for maintenance. Since the level of
detail in a PMS condition survey is lower, it is a common practice to report smaller
segments with an equivalent condition simply as “ same as previous” because there is
no need for extensive auscultation and to reduce the cost of the field work.
This report contains a list of recommendations regarding the type of information
necessary for PMS purposes. The recommended items are shown in the table in Appendix G, in
which data included in Caltrans current Pavement Condition Survey Method is compared with
information required for PMS purposes. In that table, recommended items to be changed are
shaded.
38
6.0 CONCLUSIONS, FUTURE WORK, AND RECOMMENDATIONS
6.1 Conclusions
The conclusions that can be drawn from the pilot segmentation study and the GPR testing are as
follows:
1. Fixed- length segmentation is a process that involves analysis of roadway information
from various sources. Once the segmentation process is completed, the resulting
segments will provide a theoretically sound frame for future pavement condition data
collection that would allow for the development of performance models.
2. Segmentation of the pilot network showed that the best approach to break down
segments of roadway is by means of the following steps: ( a) administrative
boundaries, ( b) traffic load, ( c) pavement structure, ( d) climate region, and ( e)
pavement condition ( if needed).
3. The direct cost to implement PMS segmentation and to collect GPR data for
inventory of pavement structure for the entire network is estimated — based on
extrapolation from this pilot project — at approximately $ 7 million of contracted field
work, while the approximate need for personnel for segmentation analysis is an
additional 12.3 PY.
4. Ground- penetrating radar ( GPR) pavement- related technology has been evaluated by
Caltrans and by other DOTs, and it has been found to be reliable, both for project-level
assessment, and for network- level inventory.
5. GPR testing supplemented with limited coring and DCP data to populate the
inventory database with pavement structures throughout the California highway
network appears feasible. The information provided by the GPR contractor was easy
to use and reliable, based on the coring by the UCPRC. The available data indicates
that GPR provides reasonable cross- sections.
6.2 Recommendations
The following are the recommendations based on this project.
1. A condition survey for PMS purposes needs to be implemented. It will consist of
either minor changes to the current Pavement Condition Surveys or the
39
implementation of a parallel data collection unit, focused only on the variables
needed for adequate performance modeling.
2. If funding the segmentation of the entire network is an issue, the process can be
staged, adding more roads each year to spread the costs over several years.
3. After inventory information is generated for parts of the network ( using GPR), it is
important to document as- built information in the structural database as future
maintenance, rehabilitation, and reconstruction work occurs, in order to keep the
database accurate.
40
7.0 REFERENCES
1. Al- Qadi, I., Lahouar, S., Jiang, K., MeGhee, K., and Mokarem, D. January 2005.
“ Validation of Ground Penetration Radar Accuracy for Estimating Pavement Layer
Thicknesses.” Presented at the 84th annual meeting of the Transportation Research Board,
Washington D. C.
2. ASTM International. 2005. “ ASTM D4748- 98: Standard Test Method for Determining the
Thickness of Bound Pavement Layers Using Short- Pulse Radar.”
3. NCHRP. 1998. “ Synthesis 225: Ground Penetrating Radar for Evaluating Subsurface
Conditions for Transportation Facilities.”
4. Noureldin, A. S.; Zhu, K; Li, S; Harris, D. 2003. “ Network Pavement Evaluation With
Falling- Weight Deflectometer And Ground- Penetrating Radar,” Transportation Research
Record 1860: 90– 99.
5. Corley- Lay, J; Morrison, C. S. 2001. “ Layer Thickness Variability for Flexible Pavements in
North Carolina.” Transportation Research Board 1778: 107– 112.
6. Maser, K. R. 2002. “ Use of Ground- Penetrating Radar Data for Rehabilitation of Composite
Pavements on High- Volume Roads.” Transportation Research Board 1808: 122– 126.
7. Mishalani, R. and Koutsopoulos H. 1995. “ Uniform Infrastructure Fields: Definition and
Identification.” Journal of Infrastructure Systems 1, no. 1.
8. Thomas, F. 2003. “ Statistical Approach to Road Segmentation.” Journal of Transportation
Engineering 129, no. 3.
9. Lea, J. and Harvey, J. T. August 2002 ( Revision December 2004). “ Data Mining of the
Caltrans Pavement Management System ( PMS) Database.” Draft report prepared for the
California Department of Transportation. Pavement Research Center, University of
California, Davis and Berkeley.
10. Infrasense, Inc. 2003. “ Non- Destructive Measurement of Pavement Layer Thickness.”
Caltrans Report No. 65A0074.
D R A F T
A- 1
§ 3.2.41
Minutes - 8/ 30/ 04 Meeting On Developing Objectives for the
Highway Network Segmentation & Data Collection In D- 3 Using GPS2
Attendees:
Design: bill_ farnbach@ dot. ca. gov,
Construction: chuck_ suszko@ dot. ca. gov,
Geometronics: adrian_ davis@ dot. ca. gov, Jim Brainard/ D03/ Caltrans/ CAGov
Maintenance: carole_ harris@ dot. ca. gov, pattie_ pool@ dot. ca. gov, susan_ massey@ dot. ca. gov,
Research: alfredo_ b_ rodriguez@ dot. ca. gov, james_ n_ lee@ dot. ca. gov, michael_ m_ samadian@ dot. ca. gov,
t_ joe_ holland@ dot. ca. gov, Michael Essex
PPRC/ Dynatest: jtharvey@ ucdavis. edu, Nick Coetzee
Introduction
This meeting dealt with the development of objectives for the “ Highway Network Segmentation & Data
Collection In D- 3 Using GPS” or what is being called “ The Segmentation Pilot Project.” The objectives
developed during the meeting were broken down into five key areas: A) which highways/ routes and which
lanes to collect data from, B) the types of data to collect, C) the kinds of analysis to be performed, D) the
phases of segmentation and whether all five phases can be achieved, E) deliverables, and F) lane miles
versus cost option selection.
A key issue to be resolved by this project is whether ground- penetrating radar ( GPR) for the continuous
measurement of pavement thickness can be used effectively ( i. e., is the technology sufficiently developed
such that the use of GPR hardware and software generates measurements that are reproducible and
repeatable).
Background
In the last meeting John Harvey presented a plan and costs for testing pre- selected parts of the highway
network in District 3 ( Sacramento & Yolo counties). Since that meeting it was decided to revisit the
rationale behind what and how much of the network would be sufficiently representative to meet the main
objective of understanding how segmentation, data collection, population of data bases, and the subsequent
analyses can be done in future by Caltrans resources and whether addition resources will be required. These
issues are addressed Sections A, B, C, & D with a final determination of the optimum amount of lane miles
versus cost is made in Section F.
A Data Collection Locations
The parts of the D- 3 network ( highways and route) listed below were previously identified as being good
candidates that should include a sufficiently diverse set of roadway structural sections to be representative of
the overall highway systems within California.
1 “ Development of Integrated Databases to Make Pavement Preservation Decisions” – PPRC Strategic Plan 03/ 04.
2 The segmentation of highway networks and related data collection was not originally envisioned as part of the PPRC 03/ 04
Strategic Plan Section 3.2.4. It evolved as a logical next step that will precede the development and population of the integrated
databases originally intended.
D R A F T
A- 2
Highways / Route
• I- 80 ( Solano Co./ Yolo Co./ Sacramento Co – 35 miles)
• I- 5/ US 99 ( Sacramento Co. -? X miles)
• SR 20 ( Lake Co. to Grass Valley -? X miles)
Lanes, Lane Directions, Measurement
• 1 to 2 lanes per direction
• 4- lane facility ( outside – 1 direction, inside – other direction
• 6- lane facility ( outermost 2 lanes in each direction)
• 2- lane facility ( 1 direction)
• Type of initial measurement using GPR:
• General thickness ( homogeneous sections)
• Changes in structural cross section ( need horizontal sub- meter precision – get
information from Surveys)
B Other Data Collection Needs
Office Data
• Office of Pavement rehabilitation deflection studies
• As- Builts ( HQ data, retrieval [ intranet] of District data)
• Data from Moisture Sensitivity studies
• Data from the Pavement Performance Evaluation Phase I ( Stantec project)
Coring Data
• Use for verification of GPR measurements
• Take in questionable areas ( visually distinct fro the surrounding pavement)
• Use to calibrate GPR units used in the pilot
• Criteria for sampling
• A few random sites
• Areas designated for analysis only
Criteria to Define Changes In Pavement Structure
• Where the average thickness changes greater than 50 mm ( between 0.1 mile sections)
• Where the order of layer type changes
• Where independent METS GPR data shows significant changes
C Analysis
1. Pick 1000 lane miles from 2850 lane miles ( narrow sections – contact Pat Kelley @ D- 3
Design)
2. Take office data and map out structures
3. Collect existing coring data
4. Analyze GPR data
5. Identify questionable areas and do coring
6. Compare verification data with GPR information from analysis
7. Do the economic analysis
D R A F T
A- 3
D Segmentation3
A successful segmentation plan will consist of five passes through the network, each one resulting in a
further segmentation:
1. In the first pass, administrative considerations will prevail. This will lead to dividing the highway
network according to districts and routes. For example, I- 5 would first be segmented according to
the Caltrans districts that it lies along.
2. In the second pass, segments within an administrative unit ( route and district) are further segmented
according to pavement structure ( AC on granular, PCC, AC on PCC, AC on LCB, AC on CTB, etc),
subgrade type, with each segment having a “ uniform” pavement structure with regard to type and
general thicknesses, and underlying subgrade type.
3. In the third pass, uniform pavement structure segments are broken if they cross a climate region
boundary.
4. In the fourth pass, segments are broken if there is a significant change in traffic loading ( which
means that major intersections are natural boundaries between sections).
5. In the fifth pass, segmentation is based on surface measurements. At this level, the objective is to
divide the highway into sections that are homogeneous in their current condition ( general state of
surface distresses and IRI).
For this pilot process the first three passes will be done and the fourth and fifth passes will be conducted
depending on time, budget, and availability of traffic data.
Deliverables ( due dates)
Pilot Project Technical Deliverables
1. ( C1) – Develop a list of the 1000/ 300 lane miles ( GPR measurements/ coring & other data collection)
[ 9/ 04]
2. ( C3) – Develop information on preliminary structures/ sections ( include available information from
databases and maps) [ 12/ 04]
3. ( C5) – Final structures/ sections information ( database information & maps) [ 3/ 04]
4. ( C6) – Write Tech Memo ( technical feasibility of segmenting highway network) [ 6/ 05]
5. ( C7) - Write Tech Memo ( Economic Analysis) [ 6/ 05]
6. ( D1) - Write Tech Memo ( Segmentation Pilot Project) [ 8/ 05]
Other Deliverables
1 Marketing plan for upper management ( with technical backup)
3 Segmentation process details are from “ A Plan for Segmentation of Highway Pavements for Use in Caltrans’ Pavement
Management System,” Samer Madanat, April 29, 2004.
D R A F T
A- 4
Lane Miles Involved Vs. Costs4
Plan
# Lane miles to be
measured with GPR
# Lane miles to collect
additional data on for analysis
Estimated cost
A 2,850 300 $ 76,000
B 1,000 1,000 $ 76,000
C 2,000 2,000 $ 147,000
D 300 300 $ 36,000
E 500 500 $ 50,000
F 1,001 300 $ 40,000
Recommendation: Go with Plan F
Post Meeting Information/ Discussion
The purpose of the Segmentation Pilot is to demonstrate the feasibility of segmenting the highway network
into homogeneous sections that will allow for accurate prediction of pavement performance and the
optimization of the Maintenance budget process. However it is not clear how actual future segmentation
activities will be performed or who will perform them. What is anticipated is the development of a data
warehouse that will incorporate a tremendous amount of data from a wide variety of sources. This will
include the following:
• Research databases including the HVS field and laboratory databases and several others.
• Databases from the Pavement Performance Evaluation project, Phases I & II – Phase II to be started
in late 2004.
• The Pavement Management System ( PMS) including the existing database and the new one to be
developed starting in 2005.
• METS database( s).
Decision/ Action Needed
The above projects need to be coordinated closely to assure that data collected is compatible in terms of
populating what could become the PMS data warehouse. This raises a number of issues that will need to be
addressed:
1. Who will be the lead to verify that the right kinds of data are being collected ( essential and helpful
variables)?
2. How will the meta data be developed and by whom?
3. How will data quality be assured?
4. Do we need a Department- wide data collection, preservation, and availability policy, i. e., should the
Districts and Headquarters be required by a Directive to actively participate in an enterprise
pavement system in which design, construction, maintenance, research, and traffic data is available
to all potential users of pavement data? This was answered previously ( more- or- less) in the
affirmative but no strategy was developed to address this issue.
Suggestion:
It has been suggested that either Research or the Pavement Standards PMS Team write a white paper for
review by the Acting Director and the Acting Deputy Director for Maintenance and Research.
4 Cost estimates are based on a consultant’s estimate to do data collection and analysis for varying lengths of roadway
( Infrasense Inc. PPRC Pilot GPR Project Ground Penetrating Radar Survey in Sacramento and Yolo Counties, August 25, 2004.
Appendix B: GPR Survey Summary
GPR
File #
Date
Collected Route
Map
Direction
Target
Lane
Type
( speed) Layout Description
Approx.
Survey
Length
( mi) GPR Data Characteristics
Recommended
for Analysis
CAL009 Mar. 7, 2005 SR- 99 SB low US- 50 to San J. Co line 25 Mostly AC/ PCC, some full depth AC, somewhat variable x
CAL010 Mar. 7, 2005 SR- 99 NB faster US- 50 to San J. Co line 25 Mostly AC/ PCC; fairly homogeneous
CAL011 Mar. 7, 2005 I- 5 SB low US- 50 to San J. Co line 24 PCC, very homogeneous, some radio interference at S end. x
CAL012 Mar. 7, 2005 I- 5 NB faster US- 50 to San J. Co line 22 PCC, very homogeneous, some radio interference at S end.
CAL013 Mar. 7, 2005 SR- 99 NB low From I- 80 to SR- 70 split 16 Mostly full depth AC, fairly homogenous x
CAL014 Mar. 7, 2005 SR- 99 SB faster From I- 80 to SR- 70 split 16 Mostly full depth AC, fairly homogenous
CAL015 Mar. 7, 2005 SR- 113 NB low From Davis to Woodland 11
Very homogeneous PCC; North section appears to have Bituminous
Base. Is this possible? x
CAL016 Mar. 7, 2005 I- 5 NB low Yolo/ Colusa line SR- 113 21
Very homogeneous AC/ PCC, with several local full depth AC patches,
especially near YOL RP 11
CAL017 Mar. 7, 2005 I- 5 SB faster Yolo/ Colusa line SR- 113 21 Very homogeneous AC/ PCC x
CAL030 Mar. 8, 2005 SR- 113 SB faster From Davis to Woodland 11
Very homogeneous PCC; North section appears to have Bituminous
Base. Is this possible?
CAL031 Mar. 8, 2005 I- 80 WB low Solano County 45
Long homogeneous sections of full depth AC and PCC, some AC/ PCC,
layer type interpretation clear except in some sections near western end.
CAL032 Mar. 8, 2005 I- 80 EB low Solano County 45
Long homogeneous sections of full depth AC and PCC, some AC/ PCC,
layer type interpretation clear except in some sections near western end.
CAL033 Mar. 8, 2005 I- 5 NB low SR- 113 to SR- 99 split 13 homogeneous, looks like AC/ PCC/ Base. Not sure about the PCC x
CAL034 Mar. 8, 2005 I- 5 SB faster SR- 113 to SR- 99 split 13 homogeneous, looks like AC/ PCC/ Base. Not sure about the PCC
CAL035 Mar. 8, 2005 SR- 16 WB low Woodland to SR- 20 48
Mostly full depth AC, extremely variable, numerous pavement layers,
may be difficult distinguishing bound from unbound layers x
CAL036 Mar. 8, 2005 SR- 20 EB low Lake Co. line Sutter RP9 47 Mostly full depth AC, somewhat variable x
CAL037 Mar. 8, 2005 SR- 20 EB low Sutter RP9 to Grass Valley 41
Mostly full depht AC, some homogeneous sections, other areas hightly
variable, may be difficult to distinguish bound from unbound layers in
some areas
CAL041 Mar. 9, 2005 I- 80 WB faster Solano County 45 same as low speed, includes CRCP section in Fairfield x
CAL042 Mar. 9, 2005 I- 80 EB faster Solano County 45 same as low speed
CAL043 Mar. 9, 2005 SR- 160 SB low From I- 80 to Rio Vista 46
Mix of full depth AC and AC/ PCC. Some long homog. sections, some
areas with high variability;
CAL047 Mar. 10, 2005US- 50 EB low Sunrise Blvd. to El Dor. Co. line 11
2 Miles of homogenous PCC; the rest full depth AC, with lots of layers,
and variable.. Maybe difficult to distinguish bound from unbound layers x
CAL048 Mar. 10, 2005US- 50 WB faster Sunrise Blvd. to El Dor. Co. line 11
2 Miles of homogenous PCC; the rest full depth AC, also mostly
homogeneous
CAL049 Mar. 10, 2005SR- 45 SB low Yolo County Sect. 13
Full depth AC. Homogeneous in the north end; extreme changes in
pavement structure in the south end. x
CAL050 Mar. 10, 2005I- 505 SB low I- 5 to I- 80 33 Mostly PCC, some AC/ PCC, very homogeneous x
CAL051 Mar. 10, 2005I- 505 NB faster I- 5 to I- 80 33 Mostly PCC, some AC/ PCC, very homogeneous
Total Lane Miles Data Collection = 681 Analysis Total = 307
Section
Analyzed
C- 1
APPENDIX C: CHARTS WITH GPR STRUCTURE RESULTS AND DATA FROM THE
2003 PAVEMENT CONDITION SURVEY
C- 2
0
5
10
15
20
25
30
35
40
24
21
20
19
18.5
16
15
13
12
11
9
8
7
5
4
3
2
1
0
Post Mile
Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base, PCC
Base, PCC,
AC
PCC, AC,
Base
AC
IRI ( test)
AC - Alligator
B Cracking
PCC - 1st
Stage
Cracking
PCC - 3rd
Stage
Cracking
Figure C1. GPR cross section with selected PCS data – Sacramento SR- 99 SB ( outside lane), CAL009.
C- 3
0
5
10
15
20
25
30
35
40
22
21
20
19
18
16
15
14
13
11
10
9
8
7
6
5
4
3
2
1
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 350 in/ mi) and 3rd Stage Cracking ( 0- 100%)
Subbase
Base
PCC
AC
IRI
PCC - 1st
Stage
Cracking
PCC - 3rd
Stage
Cracking
Figure C2. GPR cross section with selected PCS data – Sacramento I- 5 SB ( outside lane), CAL011.
C- 4
0
5
10
15
20
25
30
35
40
45
27.5
28
29
33
34
35
36
36.86 / 0
2
4
5
6
7
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
450
IRI ( 0- 450 in/ mi) and Cracking ( 0- 100%)
Subbase
Base, BB
PCC, AC,
Base, BB
AC
IRI
AC - Alligator B
Cracking
I- 5 Postmiles
Figure C3. GPR cross section with selected PCS data – Sacramento and Sutter SR- 99 NB ( outside lane), CAL013.
C- 5
0
5
10
15
20
25
30
35
40
1
2
3
5
6
7
8
9
10
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Subbase
Base
PCC
IRI
PCC - 1st Stage
Cracking
PCC - 3rd Stage
Cracking
Figure C4. GPR cross section with selected PCS data – Yolo SR- 113 NB ( outside lane), CAL015.
C- 6
0
5
10
15
20
25
30
35
40
28
27
26
25
24
21
20
19
18
17
16
15
14
13
12
10
Post Mile
Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base
PCC
AC
IRI
AC - Alligator B
Cracking
Figure C5. GPR cross section with selected PCS data – Sacramento I- 5 SB ( inside lane), CAL017.
C- 7
0
5
10
15
20
25
30
35
40
44
43
42
41
39
38
37
36
34
33
32
31
29
28
27
26
24
23
21
18
11
10
9
8
7
6
5
4
Post Miles
Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0 - 100%)
Base
Base, Subbase,
PCC, AC
AC, PCC,
Base, BB
AC, PCC
IRI
AC - Alligator B
Cracking
PCC - 1st
Stage Cracking
PCC - 3rd
Stage Cracking
Figure C6. GPR cross section with selected PCS data – Solano I- 80 WB ( outside lane), CAL031.
C- 8
0
5
10
15
20
25
30
35
31
32
34
34.66 / 0
1
2
3
4
5
6
7
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
IRI ( 0- 350 in/ mi) and Cracking ( 0- 100%)
Base ( 2)
Base ( 1)
PCC
AC
IRI
AC - Alligator B
Cracking
PCC - 1st Stage
Cracking
PCC - 3rd Stage
Cracking
Figure C7. GPR cross section with selected PCS data – Sacramento and Yolo I- 15 NB ( outside lane), CAL033.
C- 9
0
5
10
15
20
25
30
35
40
40
38
36
35
34
33
31
30
29
27
26
25
24
23
22
20
19
18
16
15
13
12
11
10
9
7
6
5
4.02
2.98
2.02
0.98
0 / 7
6
4.02
3
2
1
0
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base
Base, AC
Base, BB,
AC
AC
IRI
AC -
Alligator B
Cracking
Figure C8. GPR cross section with selected PCS data – Colusa and Yolo SR- 16 WB ( outside lane), CAL035.
C- 10
0
5
10
15
20
25
30
35
40
44
43
42
41
39
38
37
36
34
33
32
31
29
28
27
26
24
23
21
18
11
10
9
8
7
6
5
4
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base ( 2)
Base ( 1)
AC ( 2)
AC ( 1)
IRI
AC - Alligator B
Cracking
1st Stage
Cracking
PCC - 3rd Stage
Cracking
Figure C9. GPR cross section with selected PCS data – Solano I- 80 WB ( inside lane), CAL041.
C- 11
0
5
10
15
20
25
30
35
40
18
19
20
21
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base, BB
Base, BB
AC
PCC, AC
IRI
AC - Alligator B
Cracking
PCC - 1st Stage
Cracking
PCC - 3rd Stage
Cracking
Figure C10. GPR cross section with selected PCS data – Sacramento US- 50 EB ( ouside lane), CAL047.
C- 12
0
5
10
15
20
25
30
35
40
12.9
11
10
9
8
7
6
5
4
3
2
1
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
IRI ( 0- 400 in/ mi) and Cracking ( 0- 100%)
Base
AC, Base, BB
AC
IRI
AC - Alligator B
Cracking
Figure C11. GPR cross section with selected PCS data – Yolo SR- 45 SB ( outside lane), CAL049a.
C- 13
0
5
10
15
20
25
30
35
40
45
50
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
5
4
3
2
1
0 / 10.7
6
5
4
3
2
Post Mile Depth ( in)
0
50
100
150
200
250
300
350
400
450
500
IRI ( 0- 500 in/ mi) and Cracking ( 0- 100%)
Base, Subbase
Base, Subbase,
PCC
PCC, AC
AC
IRI
AC - Alligator B
Cracking
PCC - 1st Stage
Cracking
PCC - 3rd Stage
Cracking
Figure C12. GPR cross section with selected PCS data – Yolo and Solano I- 505 SB ( ouside lane), CAL050.
D- 1
APPENDIX D: SEGMENTATION RESULTS
D- 2
PM from Traffic
Data Physical Reference
Section ID Route County Direction
Climate
Region Start End Start End
CAL009 99 Sacramento SB IV 24.35 23.13 SACRAMENTO, JCT. RTE. 51, NORTH JCT.
RTE. 50; END FREEWAY
SACRAMENTO, 12TH AVENUE
IV 23.13 21.94 SACRAMENTO, 12TH AVENUE SACRAMENTO, FRUITRIDGE ROAD
IV 21.94 21.57 SACRAMENTO, FRUITRIDGE ROAD MARTIN LUTHER KING JR. BOULEVARD
IV 21.57 21.46 MARTIN LUTHER KING JR. BOULEVARD observed structure change
IV 21.46 20.86 observed structure change 47TH AVENUE
IV 20.86 19.61 47TH AVENUE FLORIN ROAD
IV 19.61 17.66 FLORIN ROAD SACRAMENTO, MACK ROAD
IV 17.66 17.46 SACRAMENTO, MACK ROAD observed structure change
IV 17.46 17.24 observed structure change SACRAMENTO, STOCKTON BOULEVARD
IV 17.24 15.90 SACRAMENTO, STOCKTON BOULEVARD COSUMNES RIVER BOULEVARD/ CALVINE
ROAD
IV 15.90 15.66 COSUMNES RIVER BOULEVARD/ CALVINE
ROAD
observed structure change
IV 15.66 15.16 observed structure change observed structure change
IV 15.16 14.87 observed structure change SHELDON ROAD
IV 14.87 13.84 SHELDON ROAD LAGUNA BOULEVARD/ BOND ROAD
IV 13.84 12.76 LAGUNA BOULEVARD/ BOND ROAD ELK GROVE BOULEVARD
IV 12.76 11.26 ELK GROVE BOULEVARD observed structure change
IV 11.26 10.07 observed structure change GRANT LINE ROAD
IV 10.07 9.26 GRANT LINE ROAD observed structure change
IV 9.26 8.96 observed structure change ESCHINGER ROAD
IV 8.96 8.46 ESCHINGER ROAD observed structure change
IV 8.46 7.96 observed structure change observed structure change
IV 7.96 7.36 observed structure change DILLARD ROAD
IV 7.36 6.01 DILLARD ROAD ARNO ROAD
IV 6.01 4.39 ARNO ROAD MINGO ROAD
IV 4.39 3.56 MINGO ROAD observed structure change
IV 3.56 3.53 observed structure change TWIN CITIES, JCT. RTE. 104 EAST
IV 3.53 3.26 TWIN CITIES, JCT. RTE. 104 EAST observed structure change
IV 3.26 3.16 observed structure change observed structure change
IV 3.16 2.70 observed structure change WALNUT STREET
IV 2.70 2.26 WALNUT STREET observed structure change
IV 2.26 2.16 observed structure change observed structure change
IV 2.16 1.88 observed structure change GALT, PRINGLE AVENUE
IV 1.88 1.57 GALT, PRINGLE AVENUE GALT, SIMMERHORN ROAD
IV 1.57 1.26 GALT, SIMMERHORN ROAD observed structure change
IV 1.26 0.79 observed structure change GALT, C STREET
IV 0.79 0.33 GALT, C STREET GALT, FRONTAGE ROAD
IV 0.33 0.12 GALT, FRONTAGE ROAD SAN JOAQUIN- SACRAMENTO COUNTY LINE
CAL011 5 Sacramento SB IV 22.57 20.53 SACRAMENTO, JCT. RTE. 50 SACRAMENTO, SUTTERVILLE ROAD
IV 20.53 19.30 SACRAMENTO, SUTTERVILLE ROAD SACRAMENTO, SEAMAS AVENUE ( FRUITRIDGE)
IV 19.30 18.65 SACRAMENTO, SEAMAS AVENUE SACRAMENTO, 43RD AVENUE
D- 3
PM from Traffic
Data Physical Reference
Section ID Route County Direction
Climate
Region Start End Start End
( FRUITRIDGE)
IV 18.65 17.19 SACRAMENTO, 43RD AVENUE SACRAMENTO, FLORIN ROAD
IV 17.19 16.15 SACRAMENTO, FLORIN ROAD SACRAMENTO, POCKET/ MEADOWVIEW
ROADS
IV 16.15 15.65 SACRAMENTO, POCKET/ MEADOWVIEW
ROADS
observed structure change
IV 15.65 13.05 observed structure change observed structure change
IV 13.05 12.04 observed structure change LAGUNA BOULEVARD
IV 12.04 10.83 LAGUNA BOULEVARD ELK GROVE BOULEVARD
IV 10.83 8.49 ELK GROVE BOULEVARD HOOD- FRANKLIN ROAD
IV 8.49 4.65 HOOD- FRANKLIN ROAD Lambert Road
IV 4.65 2.13 Lambert Road TWIN CITIES ROAD
IV 2.13 0.02 TWIN CITIES ROAD SAN JOAQUIN- SACRAMENTO COUNTY LINE
CAL013 99 Sacramento/
Sutter
NB IV 26.72 26.76 SACRAMENTO, JCT. RTE. 80 ( I- 5 Postmile) observed structure change
IV 26.76 26.96 observed structure change observed structure change
IV 26.96 29.02 observed structure change SACRAMENTO, DEL PASO ROAD ( I- 5 Postmile)
IV 29.02 29.91 SACRAMENTO, DEL PASO ROAD ( I- 5 Postmile) SACRAMENTO, JCT. RTE. 99 NORTH ( I- 5 Postmile)
- Start SR 99 Postmiles
IV 32.12 32.67 SACRAMENTO, JCT. RTE. 99 NORTH ( I- 5
Postmile) - Start SR 99 Postmiles
observed structure change
IV 32.67 33.36 observed structure change ELKHORN BOULEVARD
IV 33.36 35.37 ELKHORN BOULEVARD ELVERTA ROAD
IV 35.37 36.86 ELVERTA ROAD Sacramento- Sutter County Line
IV 0.00 0.61 Sacramento- Sutter County Line observed structure change
IV 0.61 0.95 observed structure change RIEGO ROAD
IV 0.95 4.21 RIEGO ROAD observed structure change
IV 4.21 5.91 observed structure change observed structure change
IV 5.91 6.11 observed structure change observed structure change
IV 6.11 6.83 observed structure change observed structure change
IV 6.83 8.07 observed structure change JCT. RTE. 70 NORTH
CAL015 113 Yolo NB IV 0.42 1.08 HUTCHINSON DRIVE DAVIS, RUSSELL BOULEVARD
IV 1.08 2.08 DAVIS, RUSSELL BOULEVARD COUNTY ROAD 31
IV 2.08 4.11 COUNTY ROAD 31 COUNTY ROAD 29
IV 4.11 5.80 COUNTY ROAD 29 observed structure change
IV 5.80 6.11 observed structure change COUNTY ROAD 27
IV 6.11 7.66 COUNTY ROAD 27 COUNTY ROAD 25
IV 7.66 9.23 COUNTY ROAD 25 WOODLAND, GIBSON ROAD
IV 9.23 10.15 WOODLAND, GIBSON ROAD WOODLAND, EAST MAIN STREET
IV 10.15 10.72 WOODLAND, EAST MAIN STREET WOODLAND, JCT. RTE. 5
CAL017 5 Yolo SB IV 28.92 25.57 YOLO COUNTY- COLUSA COUNTY ( COUNTY COUNTY ROAD 6
IV 25.57 23.79 COUNTY ROAD 6 COUNTY ROAD 8
IV 23.79 22.61 COUNTY ROAD 8 JCT. RTE. 505 SOUTH
D- 4
PM from Traffic
Data Physical Reference
Section ID Route County Direction
Climate
Region Start End Start End
IV 22.61 21.80 JCT. RTE. 505 SOUTH observed structure change
IV 21.80 17.62 observed structure change ZAMORA INTERCHANGE, COUNTY ROAD 13
IV 17.62 12.34 ZAMORA INTERCHANGE, COUNTY ROAD 13 YOLO INTERCHANGE, COUNTY ROAD 17
IV 12.34 10.81 YOLO INTERCHANGE, COUNTY ROAD 17 JCT. RTE. 16, COUNTY ROAD 18
IV 10.81 9.41 JCT. RTE. 16, COUNTY ROAD 18 COUNTY ROAD 99/ WEST STREET
IV 9.41 8.26 COUNTY ROAD 99/ WEST STREET WOODLAND, JCT. RTE. 113 NORTH
CAL031 80 Solano WB IV 44.72 42.67 SOLANO- YOLO COUNTY LINE JCT. RTE. 113 NORTH
IV 42.67 41.90 JCT. RTE. 113 NORTH observed structure change
IV 41.90 40.30 observed structure change observed structure change
IV 40.30 39.74 observed structure change Pedrick
IV 39.74 38.60 Pedrick observed structure change
IV 38.60 38.21 observed structure change JCT. RTE. 113 SOUTH
IV 38.21 36.90 JCT. RTE. 113 SOUTH Pitt School Road
IV 36.90 35.55 Pitt School Road DIXON AVENUE/ GRANT ROAD
IV 35.55 32.62 DIXON AVENUE/ GRANT ROAD Midway
IV 32.62 31.36 Midway Meridian
IV 31.36 29.86 Meridian Leisure Town
IV 29.86 28.36 Leisure Town VACAVILLE, JCT. RTE. 505 NORTH
IV 28.36 27.24 VACAVILLE, JCT. RTE. 505 NORTH VACAVILLE, MONTE VISTA AVENUE
IV 27.24 26.46 VACAVILLE, MONTE VISTA AVENUE Mason/ Elmira
IV 26.46 26.01 Mason/ Elmira VACAVILLE, DAVIS STREET
IV 26.01 25.31 VACAVILLE, DAVIS STREET VACAVILLE, ALAMO DRIVE
IV 25.31 23.96 VACAVILLE, ALAMO DRIVE PLEASANT VALLEY/ Pena Adobe Road
IV 23.96 20.80 PLEASANT VALLEY/ Pena Adobe Road FAIRFIELD, NORTH TEXAS STREET
IV 20.80 19.18 FAIRFIELD, NORTH TEXAS STREET FAIRFIELD, AIRBASE PARKWAY
IV 19.18 17.92 FAIRFIELD, AIRBASE PARKWAY FAIRFIELD, TRAVIS BOULEVARD
IV 17.92 17.20 FAIRFIELD, TRAVIS BOULEVARD FAIRFIELD, WEST TEXAS STREET
IV 17.20 15.82 FAIRFIELD, WEST TEXAS STREET FAIRFIELD, EAST JCT. RTE. 12
IV 15.82 15.20 FAIRFIELD, EAST JCT. RTE. 12 observed structure change
IV 15.20 13.49 observed structure change FAIRFIELD, SUISUN VALLEY ROAD
IV 13.49 12.84 FAIRFIELD, SUISUN VALLEY ROAD FAIRFIELD, JCT. RTE. 680 SOUTH
IV 12.84 12.70 FAIRFIELD, JCT. RTE. 680 SOUTH observed structure change
IV 12.70 11.98 observed structure change FAIRFIELD, JCT. RTE. 12 WEST
IV 12.22 11.98 MILEPOST EQUATION = 12.20 FAIRFIELD, JCT. RTE. 12 WEST
IV 11.98 11.39 FAIRFIELD, JCT. RTE. 12 WEST FAIRFIELD, RED TOP ROAD
IV 11.39 9.65 FAIRFIELD, RED TOP ROAD observed structure change/ climate region change
CC 9.65 ~ 8.2 observed structure change observed structure change
CC ~ 8.2 8.10 observed structure change AMERICAN CANYON ROAD
CC 8.10 8.00 AMERICAN CANYON ROAD NAPA- SOLANO COUNTY LINE
CC 8.00 6.81 NAPA- SOLANO COUNTY LINE SOLANO- NAPA COUNTY LINE
CC 6.81 5.63 SOLANO- NAPA COUNTY LINE VALLEJO, JCT. RTE. 37 WEST
CC 5.63 ~ 5.2 VALLEJO, JCT. RTE. 37 WEST observed structure change
CC ~ 5.2 4.43 observed structure change VALLEJO, REDWOOD STREET
D- 5
PM from Traffic
Data Physical Reference
Section ID Route County Direction
Climate
Region Start End Start End
CC 4.43 3.49 VALLEJO, REDWOOD STREET VALLEJO, TENNESSEE STREET
CC 3.49 3.23 VALLEJO, TENNESSEE STREET VALLEJO, SPRINGS ROAD
CC 3.23 2.88 VALLEJO, SPRINGS ROAD VALLEJO, GEORGIA STREET
CC 2.88 2.22 VALLEJO, GEORGIA STREET VALLEJO, JCT. RTE. 780 SOUTHEAST
CC 2.22 1.78 VALLEJO, JCT. RTE. 780 SOUTHEAST VALLEJO, MAGAZINE STREET
CC 1.78 1.14 VALLEJO, MAGAZINE STREET VALLEJO, JCT RTE 29 NORTHWEST
CC 1.14 0.00 VALLEJO, JCT RTE 29 NORTHWEST SOLANO COUNTY ( CARQUINEZ BRIDGE)
CAL033 5 Sacramento/
Yolo
NB IV 29.91 32.73 SACRAMENTO, JCT. RTE. 99 NORTH AIRPORT BOULEVARD
IV 32.73 34.35 AIRPORT BOULEVARD observed structure change
IV 34.35 34.65 observed structure change Sacramento- Yolo County Line
IV 0.00 0.50 Sacramento- Yolo County Line observed structure change
IV 0.50 0.80 observed structure change observed structure change
IV 0.80 2.60 observed structure change observed structure change
IV 2.60 5.53 observed structure change COUNTY ROAD 102
IV 5.53 6.51 COUNTY ROAD 102 WOODLAND, EAST MAIN STREET
IV 6.51 7.09 WOODLAND, EAST MAIN STREET WOODLAND, JCT. RTE. 113 SOUTH
IV 7.09 8.26 WOODLAND, JCT. RTE. 113 SOUTH WOODLAND, JCT. RTE. 113 NORTH
CAL035 16 Colusa/ Yolo WB IV 40.57 39.56 WEST MAIN STREET/ COUNTY ROAD 98 COUNTY ROAD 97
IV 39.56 36.71 COUNTY ROAD 97 COUNTY ROAD 94B
IV 36.71 35.44 COUNTY ROAD 94B observed structure change
IV 35.44 32.34 observed structure change observed structure change
IV 32.34 31.87 observed structure change JCT. RTE. 505; MADISON, EAST
IV 31.87 31.03 JCT. RTE. 505; MADISON, EAST MADISON, COUNTY ROAD 89
IV 31.03 28.27 MADISON, COUNTY ROAD 89 COUNTY ROAD 21A
IV 28.27 27.96 COUNTY ROAD 21A GRAFTON STREET
IV 27.96 27.55 GRAFTON STREET ESPARTO, ORLEANS STREET
IV 27.55 26.37 ESPARTO, ORLEANS STREET COUNTY ROAD 85B
IV 26.37 25.15 COUNTY ROAD 85B CAPAY, CAPAY CANAL BRIDGE
IV 25.15 20.17 CAPAY, CAPAY CANAL BRIDGE COUNTY ROAD 78A
IV 20.17 19.43 COUNTY ROAD 78A INDIAN BINGO ROAD
IV 19.43 19.20 INDIAN BINGO ROAD WINNERS WAY
IV 19.20 18.78 WINNERS WAY COUNTY ROAD 78
IV 18.78 18.13 COUNTY ROAD 78 MOSSY CREEK BRIDGE
IV 18.13 ~ 14.4 MOSSY CREEK BRIDGE observed structure change
IV ~ 14.4 12.21 observed structure change GUINDA, COUNTY ROAD 57
IV 12.21 10.80 GUINDA, COUNTY ROAD 57 COUNTY ROAD 45
IV 10.80 7.15 COUNTY ROAD 45 RUMSEY, MANZANITA AVENUE ( TO
ARBUCKLE)
IV 7.15 0.00 RUMSEY, MANZANITA AVENUE ( TO
ARBUCKLE)
Yolo- Colusa County Line
IV 7.26 0.00 Yolo- Colusa County Line BEAR CREEK, JCT. RTE. 20
CAL041 80 Solano WB IV 44.72 42.67 SOLANO- YOLO COUNTY LINE JCT. RTE. 113 NORTH
D- 6
PM from Traffic
Data Physical Reference
Section ID Route County Direction
Climate
Region Start End Start End
IV 42.67 41.90 JCT. RTE. 113 NORTH observed structure change
IV 41.90 40.20 observed structure change observed structure change
IV 40.20 39.74 observed structure change Pedrick
IV 39.74 38.60 Pedrick observed structure change
IV 38.60 38.21 observed structure change JCT. RTE. 113 SOUTH
IV 38.21 36.90 JCT. RTE. 113 SOUTH Pitt School Road
IV 36.90 35.55 Pitt School Road DIXON AVENUE/ GRANT ROAD
IV 35.55 32.62 DIXON AVENUE/ GRANT ROAD Midway
IV 32.62 31.36 Midway Meridian
IV 31.36 29.86 Meridian Leisure Town
IV 29.86 28.36 Leisure Town VACAVILLE, JCT. RTE. 505 NORTH
IV 28.36 27.24 VACAVILLE, JCT. RTE. 505 NORTH VACAVILLE, MONTE VISTA AVENUE
IV 27.24 26.46 VACAVILLE, MONTE VISTA AVENUE Mason/ Elmira
IV 26.46 26.01 Mason/ Elmira VACAVILLE, DAVIS STREET
IV 26.01 25.31 VACAVILLE, DAVIS STREET VACAVILLE, ALAMO DRIVE
IV 25.31 23.96 VACAVILLE, ALAMO DRIVE PLEASANT VALLEY
IV 23.96 20.80 PLEASANT VALLEY FAIRFIELD, NORTH TEXAS STREET
IV 20.80 19.18 FAIRFIELD, NORTH TEXAS STREET FAIRFIELD, AIRBASE PARKWAY
IV 19.18 17.92 FAIRFIELD, AIRBASE PARKWAY FAIRFIELD, TRAVIS BOULEVARD
IV 17.92 17.20 FAIRFIELD, TRAVIS BOULEVARD FAIRFIELD, WEST TEXAS STREET
IV 17.20 15.82 FAIRFIELD, WEST TEXAS STREET FAIRFIELD, EAST JCT. RTE. 12
IV 15.82 15.20 FAIRFIELD, EAST JCT. RTE. 12 observed structure change
IV 15.20 13.49 observed structure change FAIRFIELD, SUISUN VALLEY ROAD
IV 13.49 12.84 FAIRFIELD, SUISUN VALLEY ROAD FAIRFIELD, JCT. RTE. 680 SOUTH
IV 12.84 12.70 FAIRFIELD, JCT. RTE. 680 SOUTH observed structure change
IV 12.70 11.98 observed structure change FAIRFIELD, JCT. RTE. 12 WEST
IV 12.22 11.98 MILEPOST EQUATION = 12.20 FAIRFIELD, JCT. RTE. 12 WEST
IV 11.98 11.39 FAIRFIELD, JCT. RTE. 12 WEST FAIRFIELD, RED TOP ROAD
IV 11.39 ~ 10.5 FAIRFIELD, RED TOP ROAD observed structure change
IV ~ 10.5 9.65 observed structure change climate region change
CC 9.65 ~ 8.2 climate region change observed structure change
CC ~ 8.2 8.10 observed structure change AMERICAN CANYON ROAD
CC 8.10 8.00 AMERICAN CANYON ROAD NAPA- SOLANO COUNTY LINE
CC 8.00 6.81 NAPA- SOLANO COUNTY LINE SOLANO- NAPA COUNTY LINE
CC 6.81 5.63 SOLANO- NAPA COUNTY LINE VALLEJO, JCT. RTE. 37 WEST
CC 5.63 ~ 4.6 VALLEJO, JCT. RTE. 37 WEST observed structure change
CC ~ 4.6 4.43 observed structure change VALLEJO, REDWOOD STREET
CC 4.43 ~ 3.7 VALLEJO, REDWOOD STREET observed structure change
CC ~ 3.7 3.49 observed structure change VALLEJO, TENNESSEE STREET
CC 3.49 3.23 VALLEJO, TENNESSEE STREET VALLEJO, SPRINGS ROAD ( Solano)
CC 3.23 2.88 VALLEJO, SPRINGS ROAD ( Solano) VALLEJO, GEORGIA STREET
CC 2.88 2.22 VALLEJO, GEORGIA STREET VALLEJO, JCT. RTE. 780 SOUTHEAST
CC 2.22 1.78 VALLEJO, JCT. RTE. 780 SOUTHEAST VALLEJO, MAGAZINE STREET
D- 7
PM from Traffic
Data Physical Reference
Section ID Route County Direction
Climate
Region Start End Start End
CC 1.78 1.14 VALLEJO, MAGAZINE STREET VALLEJO, JCT RTE 29 NORTHWEST
CC 1.14 0.00 VALLEJO, JCT RTE 29 NORTHWEST SOLANO COUNTY ( CARQUINEZ BRIDGE)
CAL047 50 Sacramento EB IV 12.50 14.30 SUNRISE BOULEVARD observed structure change
IV 14.30 15.76 observed structure change NIMBUS ROAD/ HAZEL AVENUE
IV 15.76 16.10 NIMBUS ROAD/ HAZEL AVENUE AEROJET ROAD
IV 16.10 17.01 AEROJET ROAD FOLSOM BOULEVARD/ NATOMA
IV 17.01 17.20 FOLSOM BOULEVARD/ NATOMA observed structure change
IV 17.20 18.70 observed structure change observed structure change
IV 18.70 19.23 observed structure change PRAIRIE CITY ROAD
IV 19.23 21.50 PRAIRIE CITY ROAD SCOTT ROAD/ E Bidwell
IV 21.50 22.70 SCOTT ROAD/ E Bidwell observed structure change
IV 22.7 23.14 observed structure change Sacramento - El Dorado County Line
CAL049 45 Yolo SB IV 12.92 8.02 Yolo - Colusa County Line observed structure change
IV 8.02 5.80 observed structure change COUNTY ROAD P98A
IV 5.80 0.27 COUNTY ROAD P98A COUNTY ROAD 108
IV 0.27 0.00 COUNTY ROAD 108 Yolo - Colusa County Line
CAL050 505 Yolo/ Solano SB IV 22.36 20.11 DUNNIGAN, JCT. RTE. 4 COUNTY ROAD 12A
IV 20.11 17.45 COUNTY ROAD 12A COUNTY ROAD 14
IV 17.45 13.43 COUNTY ROAD 14 COUNTY ROAD 19
IV 13.43 10.93 COUNTY ROAD 19 observed structure change
IV 10.93 10.62 observed structure change JCT. RTE. 16; MADISON, EAST
IV 10.62 6.53 JCT. RTE. 16; MADISON, EAST COUNTY ROAD 27
IV 6.53 4.03 COUNTY ROAD 27 COUNTY ROAD 29A
IV 4.03 0.40 COUNTY ROAD 29A JCT. RTE. 128 WEST; RUSSELL BOULEVARD
IV 0.40 0.00 JCT. RTE. 128 WEST; RUSSELL BOULEVARD Solano - Yolo County Line
IV 10.63 9.36 Solano - Yolo County Line observed structure change
IV 9.36 8.96 observed structure change observed structure change
IV 8.96 8.76 observed structure change observed structure change
IV 8.76 8.16 observed structure change observed structure change
IV 8.16 5.76 observed structure change observed structure change
IV 5.76 5.57 observed structure change ALLENDALE ROAD
IV 5.57 5.06 ALLENDALE ROAD observed structure change
IV 5.06 3.36 observed structure change observed structure change
IV 3.36 3.06 observed structure change VACAVILLE, MIDWAY ROAD
IV 3.06 1.45 VACAVILLE, MIDWAY ROAD VACAVILLE, VACA VALLEY PARKWAY
IV 1.45 0.00 VACAVILLE, VACA VALLEY PARKWAY VACAVILLE, JCT. RTE. 80; BEGIN FREEWAY
E- 1
APPENDIX E: GPR DATA AND UCPRC CORE COMPARISON: PLOTS
E- 2
AC AC AC
AC AC AC AC
AC
PCC
PCC
PCC PCC
PCC PCC PCC
Base Base
cmnt
Base cmnt Base cmnt
0
5
10
15
20
25
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 8.1, # 1 PM 8.2, # 2 PM 8.3, # 3 PM 8.4, # 4
PCC
Figure E1. GPR/ core thickness - Solano 505 SB, CAL050- 1a.
E- 3
PCC
PCC
PCC
PCC
PCC
Base
Sub
Sub
Sub
Sub
Sub
Sub
PCC
PCC
PCC
PCC
PCC
PCC
PCC
AC
AC
cmnt
Base
Base
Base
cmnt
cmnt
BB
BB
0
5
10
15
20
25
30
35
GPR
core
GPR
core
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 2.9, # 1 PM 3.0, # 2 PM 3.1-, # 3 PM 3.1+, # 4 PM 3.2-, # 5 PM 3.2+, # 6
hard material
Figure E2. GPR/ core thickness - Yolo 113 NB, CAL015- 5.
Sub Sub Sub
PCC PCC PCC PCC PCC PCC
Base Base Base
ATPB
0
5
10
15
20
25
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 8.5, # 2 PM 8.6, # 3 PM 8.7, # 4
stripped
ATPB
stripped
ATPB
stripped
ATPB
E- 4
Figure E3. GPR/ core thickness - Yolo 113 NB, CAL015- 5a.
AC
AC
AC OG/ AC AC OG/ AC AC OG/ AC OG/ AC
AC
AC
AC
AC
AC
BB
Base
BB
BB
AC
BB
Base
Base
Base
0
5
10
15
20
25
30
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 17.2, # 1 PM 17.3, # 2 PM 17.4, # 3 PM 17.5, # 4
loose aggregates
hard material
hard material
hard material
Figure E4. GPR/ core thickness - Sacramento 50 EB, CAL047- 9.
AC
AC
AC OG/ AC OG/ AC
AC OG/ AC AC OG/ AC
AC
AC AC
AC AC
AC
AC
AC
AC
AC
BB
BB Base
BB AC
Base
Base
Base
BB
AC
0
5
10
15
20
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 20.0, # 1 PM 20.1, # 2 PM 20.2, # 3 PM 20.3, # 4
E- 5
Figure E5. GPR/ core thickness – Sacramento 50 EB, CAL047- 10.
AC AC
AC AC AC AC
AC AC
cmnt
cmnt
Base
Base
AC AC
0
5
10
15
20
25
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 10.8, # 1 PM 10.9, # 2 PM 11.0, # 3 PM 11.1, # 4
hard material
hard material
hard material
hard material
Figure E6. GPR/ core thickness - Yolo 45 SB, CAL049- 11.
AC AC
BB
BB
AC AC AC AC AC AC
AC
Base
Base
0
5
10
15
20
25
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 7.8, # 1 PM 7.9, # 2 PM 8.0, # 3 PM 8.1, # 4
hard material
hard material
hard material
hard material
E- 6
Figure E7. GPR/ core thickness - Yolo 45 SB, CAL049- 12.
AC AC AC AC AC AC AC AC
cmnt
Base
Base Base Base
cmnt
cmnt cmnt
0
5
10
15
20
25
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 9.1, # 1 PM 9.2, # 2 PM 9.3, # 3 PM 9.4, # 4
Figure E8. GPR/ core thickness - Yolo 45 SB, CAL049- 12a.
OG/ AC
AC
AC - core broke
approx height
AC AC OG/ AC AC OG/ AC
AC
AC
AC
AC AC
AC
AC
PCC
PCC
PCC
PCC
0
5
10
15
20
25
30
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 8.9, # 1 PM 9.0, # 2 PM 6.3, # 1 PM 6.4, # 2
PCC
PCC
PCC
PCC
CAL009- 14 CAL009- 15
E- 7
Figure E9. GPR/ core thickness - Sacramento 99 NB, CAL009- 14,15.
AC AC
AC AC AC
AC
AC
AC
Base
Base
AC AC
AC
AC
AC AC
Base
Base
AC AC
AC
AC
AC AC AC
0
5
10
15
GPR
core
GPR
core
GPR
core
GPR
core
GPR
core
GPR
core
Thickness ( in)
PM 3.0, # 1 PM 3.1, # 2 PM 1.8, # 1 PM 1.9, # 2 PM 0.6, # 1 PM 0.7, # 2
loose aggregates
CAL035- 16 CAL035- 17 CAL035- 18
Figure E10. GPR/ core thickness - Colusa 16 WB, CAL035- 16,17,18.
E- 3
APPENDIX F: GPR DATA AND UCPRC CORE COMPARISON: TABLES
F- 1
Table F1. GPR Data and UCPRC Core Thicknesses
GPR - Layer Thicknesses ( in) Cores - Layer Thicknesses ( in)
Site, Closure,
Core #
Co., Route, Approx
PM, Dir
Layer
1
Layer
2
Layer
3
Layer
4
Layer
1
Layer
2
Layer
3
Layer
4
CAL009- 14 # 1 Sac 99, PM 8.9 SB 8.22 4.05 9.56 n/ a
CAL009- 14 # 2 Sac 99, PM 9 SB 7.55 8.44 9.44 7.67 7.68
CAL009- 15 # 1 Sac 99, PM 6.3 SB 7.25 3.34 8.59 6.21 3.11
CAL009- 15 # 2 Sac 99, PM 6.4 SB 7.41 2.93 9.00 5.59 3.05
CAL015- 5 # 1 Yolo 113, PM 2.9 NB 8.50 7.22 15.68 8.41 5.05
CAL015- 5 # 2 Yolo 113, PM 3 NB 8.22 7.22 15.77 8.35 0.00
CAL015- 5 # 3 Yolo 113, PM 3.1 NB 8.57 7.48 4.00 8.46 5.83
CAL015- 5 # 4 Yolo 113, PM 3.1 NB 8.50 7.51 3.95 8.56 5.10
CAL015- 5 # 5 Yolo 113, PM 3.2 NB 9.38 5.93 16.02 9.26 1.37
CAL015- 5 # 6 Yolo 113, PM 3.2 NB 9.45 5.82 15.95 9.17 6.70
CAL015- 5a # 1 Yolo 113, PM 8.4 NB 9.46 1.66 12.01 n/ a
CAL015- 5a # 2 Yolo 113, PM 8.5 NB 8.97 1.65 12.83 9.26
CAL015- 5a # 3 Yolo 113, PM 8.6 NB 8.90 1.86 12.00 9.68
CAL015- 5a # 4 Yolo 113, PM 8.7 NB 9.29 1.53 12.69 9.87 2.14
CAL035- 16 # 1 Col 16, PM 3 WB 4.09 6.20 4.06
CAL035- 16 # 2 Col 16, PM 3.1 WB 5.87 4.25 4.77 2.55
CAL035- 17 # 1 Col 16, PM 1.8 WB 3.22 2.49 3.04 2.76 2.79 3.02
CAL035- 17 # 2 Col 16, PM 1.9 WB 1.29 4.25 2.73 6.68 1.40 5.42
CAL035- 18 # 1 Col 16, PM 0.6 WB 5.58 5.33 4.65
CAL035- 18 # 2 Col 16, PM 0.7 WB 5.01 5.16 3.25
CAL047- 9 # 1 Sac 50, PM 17.2 EB 9.55 4.63 5.07 5.54 8.71 3.70 3.78
CAL047- 9 # 2 Sac 50, PM 17.3 EB 8.05 4.77 4.45 4.97 6.89 4.29
CAL047- 9 # 3 Sac 50, PM 17.4 EB 6.98 11.67 3.51 3.92 7.13
CAL047- 9 # 4 Sac 50, PM 17.5 EB 7.45 7.23 3.78 5.60 8.62
CAL047- 10 # 1 Sac 50, PM 20 EB 5.79 5.29 4.72 3.28 6.01 5.75 3.74
CAL047- 10 # 2 Sac 50, PM 20.1 EB 3.13 5.32 4.98 3.88 2.64 5.28 4.68
CAL047- 10 # 3 Sac 50, PM 20.2 EB 3.39 5.56 4.44 4.44 4.13 5.15 3.42
CAL047- 10 # 4 Sac 50, PM 20.3 EB 4.64 5.17 4.72 4.08 4.99 5.19 4.09 3.55
CAL049- 11 # 1 Yolo 45, PM 10.8 SB 3.25 13.52 3.55 5.51
CAL049- 11 # 2 Yolo 45, PM 10.9 SB 3.26 10.50 3.58 3.31
CAL049- 11 # 3 Yolo 45, PM 11 SB 5.31 16.75 5.72
CAL049- 11 # 4 Yolo 45, PM 11.1 SB 5.50 16.44 5.47
CAL049- 12 # 1 Yolo 45, PM 7.8 SB 2.78 7.02 3.01
CAL049- 12 # 2 Yolo 45, PM 7.9 SB 2.35 13.36 2.60
CAL049- 12 # 3 Yolo 45, PM 8 SB 2.41 5.33 15.09 2.57
CAL049- 12 # 4 Yolo 45, PM 8.1 SB 3.14 13.78 4.12
CAL049- 12a # 1 Yolo 45, PM 9 SB 6.70 11.66 7.48 8.66
CAL049- 12a # 2 Yolo 45, PM 9.1 SB 6.57 12.38 6.57 10.00
CAL049- 12a # 3 Yolo 45, PM 9.2 SB 6.27 13.98 6.41 8.73
CAL049- 12a # 4 Yolo 45, PM 9.3 SB 5.88 16.48 5.56 8.97
CAL050- 1a # 1 Sol 505, PM 8.1 SB 2.12 4.99 8.17 5.14 2.54 3.92
CAL050- 1a # 2 Sol 505, PM 8.2 SB 1.87 4.86 8.69 4.80 2.40 4.92 8.46 1.65
CAL050- 1a # 3 Sol 505, PM 8.3 SB 8.92 6.44 9.05 4.81
CAL050- 1a # 4 Sol 505, PM 8.4 SB 8.53 7.14 8.75 6.27
F- 2
Table F2. GPR Data and UCPRC Core Layer Materials
Site, Closure, GPR - Material Type Cores - Material Type
Core #
Co., Route, Approx PM,
Dir Layer 1 Layer 2 Layer 3 Layer 4 Layer 1 Layer 2 Layer 3 Layer 4
CAL009- 14 # 1 Sac 99, PM 8.9 SB AC AC PCC ( OG/ AC)
CAL009- 14 # 2 Sac 99, PM 9 SB AC AC PCC OG/ AC AC ( PCC)
CAL009- 15 # 1 Sac 99, PM 6.3 SB AC AC PCC OG/ AC AC ( PCC)
CAL009- 15 # 2 Sac 99, PM 6.4 SB AC AC PCC OG/ AC AC ( PCC)
CAL015- 5 # 1 Yolo 113, PM 2.9 NB PCC Base Sub PCC cmnt
CAL015- 5 # 2 Yolo 113, PM 3 NB PCC Base Sub PCC ( cmnt)
CAL015- 5 # 3 Yolo 113, PM 3.1 NB PCC Base Sub PCC cmnt
CAL015- 5 # 4 Yolo 113, PM 3.1 NB PCC Base Sub PCC cmnt
CAL015- 5 # 5 Yolo 113, PM 3.2 NB PCC BB Sub PCC AC
CAL015- 5 # 6 Yolo 113, PM 3.2 NB PCC BB Sub PCC AC
CAL015- 5a # 1 Yolo 113, PM 8.4 NB PCC BB Sub ( PCC)
CAL015- 5a # 2 Yolo 113, PM 8.5 NB PCC BB Sub PCC ( ATPB)
CAL015- 5a # 3 Yolo 113, PM 8.6 NB PCC BB Sub PCC ( ATPB)
CAL015- 5a # 4 Yolo 113, PM 8.7 NB PCC BB Sub PCC ATPB ( ATPB)
CAL035- 16 # 1 Col 16, PM 3 WB AC AC AC
CAL035- 16 # 2 Col 16, PM 3.1 WB AC AC AC AC
CAL035- 17 # 1 Col 16, PM 1.8 WB AC AC AC Base AC AC
CAL035- 17 # 2 Col 16, PM 1.9 WB AC AC AC Base AC AC
CAL035- 18 # 1 Col 16, PM 0.6 WB AC Base AC
CAL035- 18 # 2 Col 16, PM 0.7 WB AC Base AC
CAL047- 9 # 1 Sac 50, PM 17.2 EB AC AC BB Base OG/ AC AC AC
CAL047- 9 # 2 Sac 50, PM 17.3 EB AC AC BB BB OG/ AC AC
CAL047- 9 # 3 Sac 50, PM 17.4 EB AC AC Base Base OG/ AC
CAL047- 9 # 4 Sac 50, PM 17.5 EB AC AC BB Base OG/ AC
CAL047- 10 # 1 Sac 50, PM 20 EB AC AC BB BB OG/ AC AC AC
CAL047- 10 # 2 Sac 50, PM 20.1 EB AC AC BB BB OG/ AC AC AC
CAL047- 10 # 3 Sac 50, PM 20.2 EB AC AC BB BB OG/ AC AC AC
CAL047- 10 # 4 Sac 50, PM 20.3 EB AC AC BB BB OG/ AC AC AC AC
CAL049- 11 # 1 Yolo 45, PM 10.8 SB AC Base AC cmnt
CAL049- 11 # 2 Yolo 45, PM 10.9 SB AC Base AC cmnt
CAL049- 11 # 3 Yolo 45, PM 11 SB AC Base AC
CAL049- 11 # 4 Yolo 45, PM 11.1 SB AC Base AC
CAL049- 12 # 1 Yolo 45, PM 7.8 SB AC BB AC
CAL049- 12 # 2 Yolo 45, PM 7.9 SB AC BB AC
CAL049- 12 # 3 Yolo 45, PM 8 SB AC AC Base AC
CAL049- 12 # 4 Yolo 45, PM 8.1 SB AC Base AC
CAL049- 12a # 1 Yolo 45, PM 9 SB AC Base AC cmnt
CAL049- 12a # 2 Yolo 45, PM 9.1 SB AC Base AC cmnt
CAL049- 12a # 3 Yolo 45, PM 9.2 SB AC Base AC cmnt
CAL049- 12a # 4 Yolo 45, PM 9.3 SB AC Base AC cmnt
CAL050- 1a # 1 Sol 505, PM 8.1 SB AC AC PCC Base AC AC ( PCC)
CAL050- 1a # 2 Sol 505, PM 8.2 SB AC AC PCC Base AC AC PCC cmnt
CAL050- 1a # 3 Sol 505, PM 8.3 SB PCC Base PCC cmnt
CAL050- 1a # 4 Sol 505, PM 8.4 SB PCC Base PCC cmnt
( ) materials in parenthesis were determined from field observations and review of the DCP
results.
F- 3
Table F3. DCP Results from Coring
DCP Results
Layer 1 Layer 2 Layer 3
Site, Closure,
Core #
Co., Route, Approx
PM, Dir
Thickness
( in)
mm per
5 blows
Thickness
( in)
mm per
5 blows
Thickness
( in)
mm per
5 blows
CAL009- 14 # 1 Sac 99, PM 8.9 SB No DCP - PCC underneath
CAL009- 14 # 2 Sac 99, PM 9 SB 2.2 0.7
CAL009- 15 # 1 Sac 99, PM 6.3 SB No DCP - PCC underneath
CAL009- 15 # 2 Sac 99, PM 6.4 SB No DCP - PCC underneath
CAL015- 5 # 1 Yolo 113, PM 2.9 NB 11.0 2.3 24.5 13.4
CAL015- 5 # 2 Yolo 113, PM 3 NB 10.6 0.8
CAL015- 5 # 3 Yolo 113, PM 3.1 NB 3.9 2.5 21.7 8.1 23.5 3.0
CAL015- 5 # 4 Yolo 113, PM 3.1 NB 7.0 2.5 20.7 8.7 22.4 0.7
CAL015- 5 # 5 Yolo 113, PM 3.2 NB No DCP - rest of AC core stuck in hole (~ 2- 6 inches AC)
CAL015- 5 # 6 Yolo 113, PM 3.2 NB 23.9 7.6
CAL015- 5a # 1 Yolo 113, PM 8.4 NB No Core/ DCP - problems with closure
CAL015- 5a # 2 Yolo 113, PM 8.5 NB 8.9 2.3 28.3 6.6
CAL015- 5a # 3 Yolo 113, PM 8.6 NB 11.2 1.8 27.3 5.9
CAL015- 5a # 4 Yolo 113, PM 8.7 NB 10.2 2.2 26.7 7.0
CAL035- 16 # 1 Col 16, PM 3 WB 18.3 5.0 25.8 2.7 32.7 8.8
CAL035- 16 # 2 Col 16, PM 3.1 WB 26.0 3.0 28.2 0.6
CAL035- 17 # 1 Col 16, PM 1.8 WB 17.3 4.9 27.8 3.3 33.3 7.1
CAL035- 17 # 2 Col 16, PM 1.9 WB 4.9 2.5 34.8 6.9
CAL035- 18 # 1 Col 16, PM 0.6 WB 1.9 1.2 5.1 4.2 22.6 19.9
CAL035- 18 # 2 Col 16, PM 0.7 WB 6.9 4.4 10.3 0.6 32.4 7.6
CAL047- 9 # 1 Sac 50, PM 17.2 EB 11.9 0.8
CAL047- 9 # 2 Sac 50, PM 17.3 EB No DCP - stopped coring b/ c of stripped AC layer
CAL047- 9 # 3 Sac 50, PM 17.4 EB No DCP - hard material underneath
CAL047- 9 # 4 Sac 50, PM 17.5 EB 6.2 0.4
CAL047- 10 # 1 Sac 50, PM 20 EB 19.6 2.4
CAL047- 10 # 2 Sac 50, PM 20.1 EB 10.6 1.3
CAL047- 10 # 3 Sac 50, PM 20.2 EB 5.8 0.8
CAL047- 10 # 4 Sac 50, PM 20.3 EB 8.5 1.1
CAL049- 11 # 1 Yolo 45, PM 10.8 SB 6.7 0.7 29.6 9.7
CAL049- 11 # 2 Yolo 45, PM 10.9 SB 5.1 0.9 33.7 10.4
CAL049- 11 # 3 Yolo 45, PM 11 SB 2.6 0.6
CAL049- 11 # 4 Yolo 45, PM 11.1 SB 8.3 0.9 30.6 11.3
CAL049- 12 # 1 Yolo 45, PM 7.8 SB 8.7 0.6
CAL049- 12 # 2 Yolo 45, PM 7.9 SB 10.9 0.8
CAL049- 12 # 3 Yolo 45, PM 8 SB 3.1 0.3
CAL049- 12 # 4 Yolo 45, PM 8.1 SB 4.9 0.4
CAL049- 12a # 1 Yolo 45, PM 9 SB 22.3 14.2
CAL049- 12a # 2 Yolo 45, PM 9.1 SB 19.4 16.4
CAL049- 12a # 3 Yolo 45, PM 9.2 SB 22.8 14.5
CAL049- 12a # 4 Yolo 45, PM 9.3 SB 22.6 19.1
CAL050- 1a # 1 Sol 505, PM 8.1 SB No DCP - bottom part of core stuck in the hole
CAL050- 1a # 2 Sol 505, PM 8.2 SB 8.5 1.2
CAL050- 1a # 3 Sol 505, PM 8.3 SB 7.1 2.0 12.6 7.0 13.4 1.2
CAL050- 1a # 4 Sol 505, PM 8.4 SB 3.3 2.1 21.9 6.9
F- 4
Table F4. GPR Data and UCPRC Core GPS Coordinates and Relative Distance Errors
GPR GPS Coordinates Core GPS Coordinates
Site, Closure, ( d) ddmm. mmmmmm ( d) ddmm. mmm
Core #
Co., Route, Approx
PM, Dir Latititude Longitude Latititude Longitude
Relative
Error
( ft)
CAL009- 14 # 1 Sac 99, PM 8.9 SB 3821.748054 12120.881090 3821.745 12120.878 23.69
CAL009- 14 # 2 Sac 99, PM 9 SB 3821.815252 12120.951460 3821.814 12120.951 7.92
CAL009- 15 # 1 Sac 99, PM 6.3 SB 3819.697283 12119.729590 3819.694 12119.72