/
~
~
~
4L11C CONCRETE PERFO~RMANCE UNDER HEAVY FIGHTER
AIRCRAFT LOADING D.A. TIMIAN, S.M. DASS, W.C. DASS, R.H. SUES, M.B. HARDY, J.G. MURFEE HEADQUARTERS AIR FORCE CIVIL ENGINEERING SUPPORT AGENCY HO AFCESA/RACO 139 BARNES DRIVE TYNDALL AFB FIL 32403-5319
,A,
FEBRUARY 1993TC -;'..
ELECTE JAN28
FINAL REPORTB APRIL 1988 - NOVEMBER 1990 APPROVED FOR PUBLIC RELEASE: DISTRIBUTIQY UNLIMITED
94-02294
J
(
ENGINEERING RESEARCH DIVISION.
Air Force Civil Engineering Support Agency Civil Engineering Laboratory Tyndall Air F~orce Base, Florida 32403
942
5096'
I
I NOTICE
I
PLEASE DO NOT REQUEST COPIES OF THIS REPORT FROM HQ AFCESA/RA (AIR FORCE CIVIL ENGINEERING SUPPORT AGENCY). ADDITIONAL COPIES MAY BE PURCHASED FROM: NATIONAL TECHNICAL INFORMATION SERVICE 5285 PORT ROYAL ROAD SPRINGFIELD, VIRGINIA 22161 FEDERAL GOVERNMENT AGENCIES AND THEIR CONTRACTORS REGISTERED WITH DEFENSE TECHNICAL INFORMATION CENTER SHC'JLD DIRECT REQUESTS FOR COPIES OF THIS REPORT TO:
I I
DEFENSE TECHNICAL INFORMATION CENTER CAMERON STATION ALEXANDRIA, VIRGINIA 22314
U I I I I I I,
I
?'
1
"
I
REPORT DOCUMENTATION PAGE .AGENCY USE 3N.Y
2.
Leave bion.
REPORT DATE
1FebruarT !4
!3.
1993
REPORT TYPE AN)
INAL REPORT,
TITLE AND SUBTITLE
7. PE,
D.A.
TIMIAN (I)
R.H.
S.M. W.C.
DASS (2) DASS (3)
M.B. HARDY (5) J.G. MURFEE (6)
,
Cý,3ANIZATI!3% ;.,,I.
AND ADD.7:SS(ES)
tTQI t;~G ACEN'.CY r;-':
.Z;''.G
APR 88 -
NOV 90
PE63723F PR2104 WU07
SUES (4)
8.
HEADQUARTERS AIR FORCE CIVIL ENGINEERING SUPPORT AGENCY 139 BARNES DRIVE TYNDALL AFB FL 32403-5319 APPLIED RESEARCH ASSOCIATES TYNDALL AFB FL 32403 SXP,:;
DATES COVERED
S. FUNDING NUMBERS
ASPHALTIC CONCRETE PERFORMANCE UNDER HEAVY FIGHTER AIRCRAFT LOADING ,__,_TA10 6 AUHO;Rk
Form A¢pprc~ed
,Si AND ADDRESS(ES)
PERFORMING ORGANIZATION REPORT NUMBER
ESL-TR-91-26
10. SPONSORING MONITORING AGENCY REPORT NUMBER
I
HEADQUARTERS AIR FORCE CIVIL ENGINEERING SUPPORT AGENCY HQ AFCESA/RACO 139 BARNES DRIVE TYNDALL AFB FL 32403-5319
I
•1.~i SU•.M:, '•AR•,
ESL-TR-9 1-26
NZ..LS
IA..
- L. S AT
,12b.
Dir'RISUTOON CC-"E
APPROVED FOR PUBLIC RELEASE DISTRIBUTION UNLIMITED AVAILABILITY SPECIFIED ON REVERSE OF FRONT COVER Rutting of asphaltic concrete pavements is
rapidly becoming a cause for concern among
AF civil engineers. Modern fighter aircraft often have operating tire pressures well above the capacity of the existing pavements. To reduce rutting, a mix design techni. que that explicitly considers the expected loading was investigated. Pavement test sections were constructed and trafficked by high pressure tires. Variations in test sections included mix design (Marshall and gyratory), airfield design (4- and 6-inch Pavement loading was flexible and rigid composite), and wheel loading (F-15C/D). monitored throughout trafficking, including dynamic load magnitude, position, and velocity. Pavement response was measured by taking profilographs before, during, and after trafficking. Damage parameters were defined and calculated to evaluate test section response and performance. Damage varied significantly between test sections, with the most obvious factors being the mix design and the base layer support. The gyratory test sections outperformed their Marshall counterparts with the gyratory composite section performing the best. This study has shown that pavements can be designed using gyratory methods. However, improvements in base layer performance are needed to improve the overall performance of flexible airfield pavements. 14. SUE.LCT TEkRS
15. NUMBER OF PAGES
FLEXIBLE PAVEMENTS RUTTING
GYRATORY TRAFFICKING
344 16. PRICE CODE
MRSHALL
17.
_______
SECUR;TY CLASSIFICATION OF REPORT
UNCLASSIFIED IS%
-
.
-
18.
SECURITY CLASSIFICATION OF THIS PAGE
UNCLASSIFIED
19.
SECURITY CLASSIFICATION OF ABSTRACT
20. LIMITATION OF ABSTRACT
UNCLASSIFIED
,
Sta-cac -c-
(The reverse of this page is blank.)
298 £'e,
2-89)
I I
EXECUTIVE SUNMARY
A.
OBJECTIVE
One approach to reducing rutting is the use of a mix design technique that explicitly considers the loading which will be applied to the pavement. This study was designed to investigate the relative differences in performance of two mix design methods when the mixes are subject to high pressure tires. *
B.
BACKGROUND
Rutting of asphaltic concrete pavements is rapidly becoming a cause for concern among Air Force (AF) civil engineers. Modern fighter aircraft often have operating tire pressures well above the capacity of the existing pavements. Research efforts are under way to identify cost-effective alternatives for the design of asphaltic concrete mixtures under these high pressure tires.
IC.
SCOPE Pavement test sections were constructed and trafficked by high pressure tires. Variations in test sections included mix design, airfield design, and wheel loading. Pavement loading and performance were monitored throughout trafficking. The collected data was reduced for evaluation of rutting performance and resistance.
ID.
METHODOLOGY Two mix design methods, the heavy-duty Marshall and the gyratory, were investigated for this project. The Marshall technique is an impact method and was selected because it is the most frequently used design method for AF pavements. The gyratory uses a kneading action and was selected because it had showed promise in an earlier study.
NE. I
'
I
I
TEST DESCRIPTION s Pavement test sections were designed and constructed at Tyndall AFB to study the rutting damage. Variations between test sections included mix design procedure, airfield design, and wheel loading. The mix design procedures were the heavy-duty Marshall and the gyratory. Airfield designs included 4-inch and 6-inch flexible pavements along with a 6-inch asphaltic overlay on 12 inches of Portland cement concrete. Two lanes were trafficked, one for an empty F-15C/D, the other for a fully loaded F-15C/D. This report documents the analysis of the test data from the fully loaded F-15C/D lane. iii
m A single-wheel loadcart laden with 29500 pounds of lead weight was towed back and forth by a modified front-wheel drive truck. The single wheel was an aircraft tire set to an inflation pressure of 355 psi. The loadcart position, velocity, and applied load was monitored continuously throughout the trafficking. The load was trafficked across a 4-foot wide lane using a normal distribution. Histograms of loadcart position were calculated for comparison to the pavement response. The forward passes of the loadcart averaged 13 mph while those passes conducted in reverse gear averaged 9 mph. The dynamic load applied to the pavement varied from 22000 to 37000 pounds due to the vertical oscillation of the trafficking loadcart. However, the average dynamic load, 29600 pounds, was similar to the static load of 29500 pounds. Environmental measurements were also taken continuously throughout the Although radiation intensity, wind speed and direction were trafficking. Five thermocouples were monitored, they were not used in this analysis. placed to monitor the temperature at the pavement surface and at several Ambient temperature was also recorded at six feet depths below the surface. above surface. Because most of the trafficking was conducted during the hot summer months, the average ambient temperature was 87.4 0 F, while the average pavement surface temperature was found to be 112.8 0 F. The permanent pavement response to the trafficking was monitored with a profilograph, which measures the contour of the pavement surface across the traffic lane. Profilograph measurements were taken prior to trafficking for and at the end of at intermittent pass levels, original conditions, trafficking for final conditions. Cores from the asphaltic concrete were also taken at different pass levels to determine the change in air voids with traffic. Damage parameters were defined to provide a consistent set of variables that quantify the pavement surface response. They included maximum rut depth, maximum upheaval, rut width, rut area, and upheaval area. These damage parameters were then used to correlate damage to the applied loading. Statistical methods were employed to rank order the test sections by degree of damage. A multivariate analysis was conducted to determine the significance of the test variables on the pavement damage in light of the damage variability within each test section and other sources of variability such as environment, material properties, and loading. A Pearson pairwise correlation analysis of the the test variables was also performed to help interpret the results from the multivariate analysis and to further quantify the relative importance of the test variables and their relationship to each other.
m l
3 3 m
m
i iv
!
IF.
RESULTS
3
A nonlinear relationship between damage and passes was found for all
test sections.
The damage measures revealed an increase in the accumulated
damage at the ends of the traffic lane, reflecting the acceleration/deceleration of the loadcart. The lateral position histograms showed a slightly skewed distribution. The greater the asymmetry in loading, the greater was the observed damage. The temperature did not vary much during trafficking hours, therefore conclusions regarding the rate of rutting versus temperature could not be made. The damage varied significantly between test sections, with the most obvious factors being the mix design and the granular layer support. For both mix designs, the composite test sections performed better than their flexible counterparts. The difference can be attributed to the rutting in the granular layers of the flexible sections. Sixty percent of the flexible test section rutting was in the granular layers.
I
The significance of the mix design procedure on the pavement response can be seen by comparing the damage profiles from the composite test sections. The Marshall mix section experienced three times the rutting of the gyratory mix despite having less traffic. The gyratory sections formed broad ruts, characteristic of densification. The Marshall sections formed multiple ruts, characteristic of plastic flow. The cores removed from the test sections showed the severely rutted Marshall mix to have densities greater than 98 percent theoretical maximum while the stable gyratory mix had densities below this critical value.
I
G.
CONCLUSIONS
The laboratory compactive effort used to select the amount of binder (asphalt) must equal that from the expected traffic. The problem with the Marshall compaction procedure is that the impact method cannot create densities similar to those found under high pressure tires. Further increasing the Marshall compactive effort will only degrade the aggregate. The gyratory testing machine applies higher compactive effort through kneading without degrading the aggregate. The mix designer is able to select the amount of binder appropriate for the traffic.
Slaboratory
3
|
I
This study has shown that pavements can be designed using gyratory methods and that these surfaces have superior performance to Marshall mix designs. Even though the gyratory sections were compacted to only 92 percent of the theoretical maximum density, the gyratory sections outperformed the Marshall sections. Heavier field compaction equipment for gyratory mixes may further minimize surface rutting. Lower asphalt costs in the leaner gyratory mixes may offset the additional cost of purchasing gyratory testing equipment.
v
I Independent of the mix design method, the granular layers were found to contribute about sixty percent of the rutting in the 6-inch flexible sections. Improvements in granular layer performance are needed to improve the overall performance of flexible airfield pavements under high pressure tires.
3
v
I I I I I I I I I I I I I vi
I
PREFACE
This report, entitled "Asphalt Concrete Performance under Heavy Fighter Aircraft Loading" vas prepared by Applied Research Associates (ARA) under a Scientific and Technical Assistance (SETA) contract and funded under Contract Number F08635-88-C-0067 by the Air Force Civil Engineering Support Agency, Civil Engineering Laboratory, Tyndall Air Force Base, Florida 32403-5319. The work was a joint effort between the Air Force and its SETA contractor. Co-Principal investigators were David Timian of ARA and Jim Murfee of the Air Force. Dr Robb Sues and ARA's Southeast Division handled the statisical analyses. Groups contributing to the design included Texas Transportation Institute, Army COB Waterways Experiment Station, Resource International, Inc, and the University of North Carolina at Charlotte. Most of the construction and loadcart operation was done by the Civil Engineering Laboratory Operations Branch. Data acquisition expertise was supplied by ARA's Rocky Mountain, New England, and Southwest Divisions. This report covers work performed between April 1988 and November The AFCESA/RD project officer was Jim Murfee.
1990.
This report has been reviewed by the Public Affairs Office and releasable to the National Technical Information Service (NTIS). At NTIS, will be available to the general public, including foreign nationals.
INIS
5
This technical report has been reviewed and is
SJIM MURFEE,
SProject
is it
approved for publication.
UHLIK III, Lt Col, USAF Chief, Engineering Research Division
Officer
DAR SFRANK Chief, Air Base Repair Branch
P. GALLAGHER III, Colonel, USAF r
dhDirector,
Air Force Civil Engineering Laboratory
U
Aoeession For
MTIO QUALITY INSPECTED B
NTISS
RAAI
DTIC TAB
5
Unannounced
5
Justifteaties vii (The reverse of this page is
blank)
i Avaslability Qodes
T val and/or lo
_=peo 18
TABLE OF CONTENTS Section I
II
III
IV
Title INTRODUCTION ...........................................
1
A. B. C.
I 1 1
OBJECTIVE ........................................ BACKGROUND ....................................... SCOPE/APPROACH ...................................
TEST DESCRIPTION .......................................
3
A. B. C. D. E.
3 5 5 6 6
DESIGN ......................... .......... CONSTRUCTION....................................... .................. ASPHALTIC CONCRETE ............. BASE COURSE ...................................... SUBGRADE .........................................
MEASUREMENT AND DATA ACQUISITION TECHNIQUES ............
14
A. B. C. D.
14 15 16 16
LOADCART INSTRUMENTATION ......................... ENVIRONMENTAL INSTRUMENTATION .................... PROFILOGRAPH ..................................... MATERIAL PROPERTIES ..............................
DATA REDUCTION AND ANALYSIS ............................ A. B. C. D. E. F. G. H.
V
Page
VELOCITY ......................................... LATERAL POSITION ................................. TEMPERATURE ...................................... LOAD ................................. ............ PAVEMENT PROFILES................................ DAMAGE PARAMETERS................................ EFFECTS OF COMPACTIVE EFFORT .......... .......... MULTIVARIATE ANALYSIS .............................
28 28 29 33 33 34 35 40 42
CONCLUSIONS AND RECOMMENDATIONS ........................
155
A. B. C. D. E.
155 155 156 157 157
DATA ACQUISITION EFFORTS ......................... RELATIONSHIP BETWEEN LOADING AND DAMAGE .......... EFFECT OF MIX DESIGN ON PERFORMANCE .............. SIGNIFICANCE OF RUTTING IN BASE LAYERS ........... CONCLUSIONS AND RECOMMENDATIONS ..................
REFERENCES .............................................
ix
158
II A B C D E F G H
FALLING WEIGHT DEFLECTOMETER DATA ...................... CORE DENSITY DATA ............................. .. ....... TEMPERATURE DATA AND HISTOGRAMS .......................... THREE DIMENSIONAL PROFILOGRAPH PLOTS ................... REGRESSION PLOTS OF DAMAGE PARAMETERS VERSUS TRAFFIC BOX PLOTS OF DAMAGE PARAMETERS ......................... RESULTS FROM PEARSON PAIRWISE CORRELATION ANALYSIS ..... RESULTS OF STEPWISE REGRESSION MODELS ..................
X
159 169 177 241 255 291 311 341
1
I I I I I I I I I I I I I I
I I
LIST OF FIGURES
Figure
3I
Title
2
Layout of Test Sections ................................ Loadcart With Optical Sensor Attached ..................
13 22
3
Pavement Marking Pattern ...............................
23
4 5 S6 7
Load Wheel Support Structure with Loadcells ............. Profilograph System .................................... Posttest Marshall Mix Design Results ................... Trench Side View in the Marshall 4-Inch
24 25 26
Flexible Test Section ...............................
8 9 10 11 12 13
3
14 15 16 17 18 19 20 21 22
S23 24 25 26 27 28
3
I
Page
Trench Side View in the Gyratory 4-Inch Flexible Test Section ............................... Initial Segment of a Calculated Velocity Profile ....... Typical Velocity Profile, Forward and Reverse .......... Velocity Histograms for Stations 0-48 at Pass Level 2589 Velocity Histograms for Stations 562-612 at Pass Level 2589 ................................ Velocity Histograms for Stations 240-384 at Pass Level 2589 ................................... Velocity Histograms at Pass Levels 2589, 5817, and 10350 Data Acquisition and Algorithm for Calculating Lateral Position .................................... Potential Y Position Error Caused by Optical Sensor Sampling Distance ............................. Mode-Centered Histograms for Station 364 ............... Comparison of Target and Actual Load Distributions ..... Typical Variation of Traffic Mode, Mean, and Standard Deviation ........................................... Mode and Mean of Lateral Position Versus Station ....... Normalized Lateral Position Histograms for 4-Inch Flexible Marshall Section ..................... Normalized Lateral Position Histograms for 6-Inch Flexible Marshall Section ..................... Normalized Lateral Position Histograms for 6-Inch Composite Marshall Section .................... Normalized Lateral Position Histograms for 6-Inch Composite Gyratory Section.................... Normalized Lateral Position Histograms for 6-Inch Flexible Gyratory Section ..................... Normalized Lateral Position Histograms for 4-Inch Flexible Gyratory Section ..................... Ambient and Surface Temperature Histograms ............. Midheight and Interface Temperature Histograms .........
xi
27
27 62 63 64 65 66 67 68 69 70 78 79 80 82 83 84 85 86 87 88 89
I LIST OF FIGURES
Figure
Title
29
Six- and Twelve-Inch Depth Into Base Course Temperature Histograms ............................... Average Daily Temperature for All Trafficking Days ...... Variation of Load During Pass 2500 ..................... Rut Progression in 4-Inch Flexible Marshall Section .... Rut Progression in 6-Inch Flexible Marshall Section .... Rut Progression in 6-Inch Composite Marshall Section ... Rut Progression in 4-Inch Flexible Gyratory Section .... Rut Progression in 6-Inch Flexible Gyratory Section ... Rut Progression in 6-Inch Composite Gyratory Section ...
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
Representative Profile for Marshall Mix Design 4-Inch Flexible ..................................... Representative Profile for Marshall Mix Design, 6-Inch Flexible ..................................... Representative Profile for Marshall Mix Design, Composite ............................................ Representative Profile for Gyratory Mix Design Composite ............................................ Representative Profile for Gyratory Mix Design, 6-Inch Flexible ...................................... Representative Profile for Gyratory Mix Design, 4-Inch Flexible ...................................... Estimated Base Course Rut Profile, 6-Inch Marshall Mix Design ........................... Estimated Base Course Rut Profile, 6-Inch Gyratory Mix Design ........................... Damage Parameters ....................................... Rut Depth Regression Analysis for 6-Inch Composite Marshall Section ..................................... Maximum Upheaval Height Regression Analysis for 6-Inch Composite Marshall Section ........................... Upheaval and True Rut Depth from the Marshall 4-Inch Flexible Section at the Final Pass Level ...... Upheaval and True Rut Depth from the Gyratory 4-Inch Flexible Section at the Final Pass Level ...... Upheaval and True Rut Depth from the Marshall 6-Inch Flexible Section at the Final Pass Level ...... Upheaval and True Rut Depth from the Marshall Composite Section at the Final Pass Level ...................... Upheaval and True Rut Depth from the Gyratory Composite Section at the Final Pass Level ...................... xii
Page
90 91 92 93 94 95
I
96 7 98 99 100 101 102 103 104
3 I I I
105 106 107 108 109 110 I11 112
I
113 114
I I
I LIST OF FIGURES
S
Figure
Title
Page
54
Upheaval and True Rut Depth from the Gyratory 6-Inch Flexible Section at the Final Pass Level ...... True Rut Depths from 4-Inch Flexible Sections ........... True Rut Depths from 6-Inch Flexible Sections ........... True Rut Depths from Composite Sections ................. True Rut Depth vs Station at Pass Levels 1554 and 4784.. Boxplot of Rut Depth for Pass Level 2324 ................ Boxplot of Rut Depth for Pass Level 4784 ...............
115 116 117 118 119 120 121
55 56 57 58 59 60 61 62 63 64 65 66 67 68
SPass
69 70
l
71 72 73 74 75
3 SMix
76 77 78 79 80
I
I
Boxplot of Rut Depth for Pass Level 9715 ................ True Rut Depths from Gyratory Test Sections ............. True Rut Depths from Marshall Test Sections ............. Relationship Between True Rut Location and the Mode and Mean for Marshall Sections ..................... Relationship Between True Rut Location and the Mode and Mean for Gyratory Sections ...................... Affected and Rut Widths from the Marshall Test Sections. Affected and Rut Widths from the Gyratory Test Sections. Areas Calculated for the Marshall Sections at Final Pass Level ........................................... Areas Calculated for the Gyratory Sections at Final Level ...................................... Boxplot Comparison of Rut Area From Each Test Section at Traffic Level 4784 ................................ Boxplot Comparison of Upheave Area From Each Test Section at Traffic Level 4784 ........................ Rut and Upheave Areas for the 4-Inch Flexible Test Sections ............................................. Rut and Upheave Areas for the 6-Inch Flexible Test Sections .......................................... Rut and Upheave Areas for the Gyratory Composite Test Section ........................................... Rut and Upheave Areas for the Marshall Composite Test Section ........................................... Rut and Upheave Areas for the Gyratory Composite Test Section Through Traffic Level 10350 ............... Effects of Compactive Effort on Asphalt Content ......... Effectiveness of Four Levels of Compaction on Gyratory from Paver ................................. Density Change With Traffic, Gyratory Test Sections ..... Post-Traffic Densities, Composite Test Sections .........
xiii
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141
I LIST OF FIGURES Figure 81 82 83 84 85 86 87 88 89 90 91 92 93
Title Matrix of Data Points Available for Analysis ............ ANOVA Generated Predictions of Rut Depth Versus Observed Damage ............ ................................ ANOVA Generated Predictions of Rut Area Versus Observed Damage ...................................... Pavement Impulse Stiffness Modulus Versus Station ....... Initial Bulk Density Data and Interpolated Values Versus Station ....................................... Asphalt Content Data and Interpolated Values Versus Station .............................................. Theoretical Maximum Density and Interpolated Values Versus Station ....................................... Asphalt Volume Data and Interpolated Values Versus Station .............................................. Voids Total Mix Data and Interpolated Values Versus Station .............................................. Voids Filled Data and Interpolated Values Versus Station True Rut Depth Versus Standard Deviation of Lateral Position of Loadcart ................................ True Rut Area Versus Standard Deviation of Lateral Position of Loadcart ................................ Pavement Thickness Variation ............................
Page
m
142 143 144 145 146 147 148 149
3 3 I
150 151 152 153 154
3
I I
m m m xiv
I I I
I LIST OF TABLES
I
Table
Title
Page
1 2 3
MATERIAL PROPERTY SUMMARY OF DESIGN ASPHALTIC CONCRETE... BASE COURSE SPECIFICATIONS AND TEST RESULTS .............. AVERAGE ASPHALT CONTENT AND REFERENCE GRADATION VALUES FROM UNCONPACTED NIX .............................. AVERAGE MARSHALL PROPERTIES ..........*.*.............. AVERAGE GYRATORY PROPERTIES ............................ AVERAGE VALUES OF DENSITY, ASPHALT CONTENT, AND VOID PARAMETERS FROM FIELD CORES ..........................
7 8
4 5 6 7 8 9 10 11 12 13
14 15 S16 17 18 19 20
I
21
AVERAGE RESILIENT MODULUS VALUES OF FIELD CORES .......... PROFILOGRAPH INCREMENTS .................................. FALLING WEIGHT DEFLECTOMETER TEST RESULTS ................ AVERAGE DENSITY VALUES OF CORES REMOVED FROM LOADED F-15 LANE ............................................ SAMPLE OPTISENSOR DATA FOR A PASS WITH AN EXTRA STRIPE... EFFECT OF EXTRA POINT ON LATERAL POSITION ................ AVERAGE VALUES AND STANDARD DEVIATIONS OF TEMPERATURE MEASUREMENTS FOR 10350 PASSES Pc TRAFFIC ............................... STATIONS FOR REPRESENTATIVE TEST SECTIONS ................ DAMAGE RANK ORDERS AT VARIOUS PASS LEVELS ................ TEST VARIABLES USED TO DESCRIBE ANTECEDENT CONDITIONS IN MANOVA ............................................. R-SQUARE STATISTIC FOR EACH DAMAGE MEASURE - ALL VARIABLES ............................................. R-SQUARE STATISTIC FOR EACH DAMAGE MEASURE - VARIABLE SUBSET 1 .............................................. R-SQUARE STATISTIC FOR EACH DAMAGE MEASURE - VARIABLE SUBSET 2 .............................................. PEARSON PAIRWISE CORRELATION COEFFICIENTS FOR SIX DAMAGE MEASURES ....................................... STEPWISE REGRESSION VARIABLE RANK ORDERING ...............
I
I
(The reverse of this page is blank.)
9 9 10 11 12 19 20 21 54 55
56 56 57 58 59 59 59 60 61
m
SECTION I
3
INTRODUCTION
IA.
OBJECTIVE One approach to reducing rutting is the use of a mix design technique that explicitly considers the loading which will be applied to the pavement. Previous laboratory work sponsored by the Air Force Engineering and Services Center suggested that rational mix designs could significantly reduce the rutting potential of asphaltic concrete (1). The study reported in this paper was designed to investigate the relative differences in performance of two mix design methods when the mixes are subjected to high-pressure aircraft tire loadings. The Marshall technique was used because it is the most frequently used design technique for AF pavements; the gyratory technique was selected because it showed promise in the earlier study. Conclusions from the laboratory work were verified in this study through a field test using simulated aircraft loading conditions. B.
3
IC.
BACKGROUND
Rutting of asphaltic concrete pavements is rapidly becoming a cause for concern among Air Force (AF) civil engineers. Modern fighter aircraft often have operating tire pressures well above the capacity of the existing pavements. Because about forty percent of Air Force taxiways are asphaltic concrete, any increase in the rate of rutting will require major reconstruction costing millions of dollars. Research efforts are under way to identify cost-effective alternatives for the design of asphaltic concrete mixtures to resist rutting under increased tire pressures. SCOPE/APPROACH Pavement test sections were designed and constructed at Tyndall AFB to study the rutting damage caused by both empty and fully loaded F-15 C/D aircraft with high pressure tires. Variations between test sections included mix design procedure, airfield design, and wheel loading. The mix design procedures used were the Marshall method with heavy-duty criteria and the gyratory method. Airfield designs included 4-inch and 6-inch flexible pavements and a 6-inch asphaltic concrete overlay on 12 inches of Portland cement concrete.
I
I I
The loadcarts and pavement were instrumented to monitor the loading and pavement response during trafficking. Using computerized data acquisition systems, the loading data was recorded for each pass of the loadcart. Response measurements were recorded at periodic intervals during trafficking.
I All of the collected data was reduced and stored using a database management system. This report focuses on the analysis of the test data from the fully loaded F-1S lane. Statistical analysis techniques were used to develop representative loadinig and response functions for all sections. Statistical methods were also used to investigate the correlations between loading functions, material properties, and response measurements. This report documents pavement loading and response measurement techniques, data reduction procedures, analysis methodology, and results. All tables and figures for each section are located at the end of the section for the reader's convenience. Section I1 describes the test section design and construction. Section III presents the types of instrumentation used to make the trafficking measurements and presents the material properties measured during and after trafficking. Section IV describes how the raw data was reduced into a usable form, presents typical values for each of the parameters, and details the relationships between loading and damage for each of the test sections. Section V discusses the conclusions about pavement rutting that were developed from this study. Supporting data is supplied in the appendices. A complete set of mode-centered histograms and profilograph data are maintained in two unpublished appendices.
I 3
I
I I I I I I I I I. I I
SECTION II
3
TEST DESCRIPTION The test sections for this project were designed to compare the rutting damage caused by high pressure tires between various airfield pavement designs. These designs include flexible pavements and asphaltic overlays of rigid pavements. Two mix designs, two aircraft loadings, and three airfield designs are tested in this study resulting in a total of twelve test sections.
I 3 IA.
3
*
m
I I
This section summarizes the procedures used to design and construct the The as-built pavement section and material properties have test sections. been determined and are reported for use in analyzing test section response. The material property information is the result of extensive quality control during, and after in the laboratory and field investigations before, The laboratory testing is comprised of general classification construction. tests combined with Marshall and gyratory design tests. The field testing includes surveying, density, profilograph, Falling Weight Deflectometer (FWD), A detailed presentation of the and Seismic Cone Penetration Tests (SCPT). data can be found in the design and construction report (2). DESIGN The project was designed to test two asphaltic concrete mixtures under several construction and load variations. Two mix design procedures chosen for use in this project are the Marshall design using heavy-duty criteria and the gyratory design procedure. All other material variables are eliminated by The airfield using the same materials and the same construction techniques. flexible and 6-inch 4-inch 1, include Figure in illustrated design types, pavements over 12 inches of base course, and a 6-inch asphaltic concrete The flexible overlay on a 12-inch thick Portland cement concrete slab. pavement sections extend from each end of the PCC slab. All sections with Marshall mix are located on the east end and all sections with gyratory mix Elevations measured at the site are based on a are located on the west end. benchmark located south of the concrete slab. The elevation of this benchmark The sections were was arbitrarily assumed to be 10 feet for convenience. designed to provide two trafficking lanes. 1.
Asphalt Mix Design
The aggregate was selected and supplied under contract with Florida ASphalt and Paving Company (FAPCO) with ARA completing the quality The selected control tests to determine the aggregate acceptability. as #67, #8910 aggregate was supplied in three separate sizes, identified (both Alabama limestones), and DRAVO, a crushed limestone from Perry, Florida. An acceptable blend of the aggregate was determined by an iterative trial and
3
I error procedure and tested to see if it met the grain size limits called for in the test plan. The LA abrasion resistance, ASTM C 131 (3), of blended material was determined to be 28.3 percent loss, well below the limit of 40 percent loss. The specific gravity of each aggregate was determined according to ASTM C 127 and C 128 (3) along with the weighted average for the aggregate mix. Table 1 lists the design aggregate gradation and specific gravities. Grade AC-20 asphalt cement was used in the design and production of the mix. The properties of the asphalt used are: penetration - 0.57 millimeter ASTM D 5 (4), specific gravity - 1.0293 ASTM D 70 (4), kinematic viscosity - 471.8 centistokes ASTM D 2170 (5), and the softening point - 1280 Fahrenheit ASTM 0 36 (4). The asphalt mix design procedures are specified in the test plan as: Military Standard 620A, Method 100 (6), 75 blow compaction, for the Marshall mix, and ASTH D 3387 (5), using 300 psi compaction pressure, for the gyratory mix. The resulting asphalt contents for each mix are 6.4 percent of the total weight for the Marshall and 5.1 percent of the total weight for the gyratory. The resulting properties for each mix at the design asphalt content are also summarized in Table 1. 2.
Base Course Design
U I l m
The material used for the base course was shipped from northern The Alabama due to the lack of acceptable material in the local area. delivered material was tested for conformance to the base course specifications as listed in the test plan. The requirements and the results of the tests are listed in Table 2. Specific gravity tests, ASTM C 127 and C 128 (3), reveal an average value of 2.707 for this material. The optimum moisture content was determined to be 6.0 percent, resulting in an optimum density of 141 pounds per cubic foot (pcf) using Modified Proctor compaction, ASTM D 1557 (4). 3.
I
3 m
Subgrade
Local materials consisting of clean sand were used for fill subgrade. The material was hauled in from a borrow pit approximately 1/2 mile from the test sections and is classified as clean sand (Unified Classification SP). A Modified Proctor test found the optimum moisture content to be 4.5 percent at an optimum density of 97.5 pcf.
4
m
4 I I I
I B.
I I 3 I
CONSTRUCTION
Test section construction began in Nay 1988, with the Air Force's Operations Group (RDCO) removing the asphaltic concrete overlay from the previous test. ARA surveyed the site and placed grade stakes for the fill subgrade. RDCO transported clean sand from the borrow pit and placed it in 6-inch lifts. These lifts were watered and compacted with a vibratory roller until a maximum density was reached. ARA verified the compaction of the fill subgrade with a nuclear density gauge after each lift. The base course material had been stored in stockpiles on the Portland cement concrete (PCC) slab. The transition sections between the flexible sections and the rigid pavement were constructed before placement of the base course over the fill subgrade. The transitions were designed to minimize the effects of changing pavement layers along the traffic lane. The base course thickness was increased to 36 inches for the 10 feet closest to the PCC slab. The base course thickness was gradually decreased to 12 inches over the next 5 feet. The base course material under the flexible sections was placed in lifts by RDCO. ARA performed density checks and surveyed each lift upon completion.
S6-inch
Before the placement of the asphalt concrete, profilograph measurements were made every 2 feet over the flexible section base course by ARA. The PCC slab elevations were measured at three points every 2 feet along each traffic lane. Prime coat was applied to the base course material by FAPCO one week before placement of the asphaltic concrete. The application rate was 0.43 gallons per square yard and 0.35 gallons per square yard on the east and west respectively. Tack coat was applied by FAPCO on each lift immediately before paving. The application rate for the tack coat was set to not exceed
Sends,
0.05 gallons per square yard. C.
I
I 3
I U
ASPHALTIC CONCRETE
The asphaltic concrete was manufactured at the FAPCO batch plant located on East 15th Street, Panama City, Florida. Before production began, the plant was calibrated using standard procedures with assistance of personnel from the US Army Corps of Engineersz Waterways Experiment Station (WES). All cold feed bins were emptied and cleaned before being filled with the specified aggregate. FAPCO placed and compacted the asphaltic concrete in two lifts. ARA monitored the density during compaction with a nuclear density gauge. RDCP and ARA sampled the hot mix at selected points. ARA conducted extraction, grain size distribution, and compaction tests while paving was taking place. ARA also removed 4-inch diameter cores from the pavement after it had cooled for density correlations.
5
I Several types of tests were performed on samples of the delivered mix. These tests included Marshall and gyratory compaction, theoretical maximum density, asphalt content by extraction, grain size distribution, bulk density, stability, and flow. The average asphalt content values and reference gradation data extracted from uncompacted mix is presented for all sections in Table 3. The average bulk density, stability, flow, and theoretical maximum density of the Marshall samples are shown in Table 4. The gyratory stability index, gyratory compaction index, stability, flow, and theoretical maximum density are shown in Table 5 for the gyratory samples. Table 6 presents the average bulk density, asphalt content, theoretical maximum density, asphalt volume, voids total mix, and voids filled, calculated from cores extracted from the test sections. Actual data is provided in Reference (2). Resilient modulus tests were conducted on one core from each test section according to ASTM 0 4123. The tests were conducted at 40, 71, 104 and 120 degrees Fahrenheit using 0.05 and 0.10 second-load durations and 1-, 2-, and 3-second cycle times. The data shows the expected trend of decreasing modulus with increasing temperature. A comparison between the Marshall and gyratory samples reveals a slightly higher modulus for the gyratory mixture as shown in Table 7. A comparison of the grain distribution data from the plant calibration and the data from tests conducted during paving reveals the presence of excess fine material. This occurrence could have been caused by breakdown of the DRAVO material during the hot mixing process. The LA Abrasion test results did not reveal this breakdown because only the larger aggregates are tested. D.
t
I I I I
BASE COURSE
Base course material was placed in 6-inch lifts and sprayed with water from a gravity feed water and P-4 crash fire fighting vehicle prior to rolling with a vibratory roller. A peak density was typically achieved after six to
eight coverages. E.
I
SUBGRADE
The subgrade material was placed in 6- to 9-inch lifts using the scrapers which transported the sand from the borrow pit. A gravity feed water truck was used to saturate the sand just before compaction with a vibratory roller. The watering and compaction was repeated until a maximum density was reached as determined by a nuclear gauge.
3
6
I
m TABLE 1. MATERIAL PROPERTY SUMMARY OF DESIGN ASPHALTIC CONCRETE.
SIEVE DESIGNATION 3/4' 1/2. 3/84 8 0 16 # 30 # 50 # 100 # 200
CRITERIA
PERCENT PASSING DESIGN 99.9 88.9 79.0 62.0 52.6 41.2 30.2 21.1 12.5 5.6
100 82-96 75-89 59-73 46-60 34-48 24-38 15-27 8-18 3-6 SPECIFIC GRAVITY
ASPHALT AGGREGATE BULK (weighted average of blend) APPARENT AGGREGATE
1.0293 2.591 2.764
I SMATERIAL
PROPERTY
MARSHALL
Unit Weight of Aggregate Only (pcf) Density (pcf) Flow (.01 inch) Gyratory Stability Index Stability (lbs) Voids Filled (%) Voids Total Mix (%)
--148.7 9.5 --3000 82.0 3.3
I I I I
7
GYRATORY 144.5 153.9 10.4 1.02 5500 80.6 1.9
I TABLE 2. BASE COURSE SPECIFICATIONS AND TEST RESULTS.
I
GRADATION OF AGGREGATES
SIEVE DESIGNATION
MINIMUM AND MAXIMUM %
ACTUAL %1
1 INCH 1/2 INCH # 4
100 40-70 20-50
100.0 65.9 33.7
# 10
15-40
22.2
# 40
5-25
12.4
# 200
0-10
6.6
LOS ANGELES ABRASION Maximum Wear % .................................. Actual Wear % ............................................
........ 40.0 27.4
INDEX PROPERTIES TEST
SPECIFICATION
Liquid Limit Non-Plastic - 25 Plasticity Index 0-5 Maximum Dry Density Peak Optimum Moisture Content - - - Apparent Specific Gravity - - - -
1 l 3
ACTUAL 13.8 0 141.0 pcf 6.0% 2.707
I 3
I 8
TABLE 3. AVERAGE ASPHALT CONTENT AND REFERENCE GRADATION VALUES FROM UNCONPACTED NIX.
1 I I
SECTION NUMBER
MIX
I
MARSHALL
TOP
2
6.27
47.2
8.4
2
MARSHALL
TOP
4
6.28
48.6
9.1
MARSHALL
BOTTOM
2
6.57
59.3
11.5
3
MARSHALL MARSHALL
TOP BOTTOM
10 2
6.47 6.07
49.7 57.6
10.0 11.2
4
GYRATORY GYRATORY
TOP BOTTOM
8 2
5.17 5.08
51.0 59.1
10.5 11.2
5
GYRATORY
TOP
3
4.84
57.8
10.8
GYRATORY
BOTTOM
1
5.41
61.6
11.6
GYRATORY
TOP
1
5.09
60.2
11.4
6
I
TYPE
STABLE
S
ASPHALT CONTENT
PERCENT PASSING #8
PERCENT PASSING #200
4. AVERAGE MARSHALL PROPERTIES.
SECTION NUMBER
1 2 3 7 8 9
I
LIFT NUMBER LAYER OF SAMPLES
LIFT LAYER
TOP TOP BOTTOM TOP BOTTOM TOP TOP BOTTOM TOP BOTTOM
NUMBER OF SAMPLES
BULK DENSITY (pcf)
STABILITY (lbs)
FLOW (in)
THEORETICAL MAXIMUM DENSITY (pcf)
1 3 1 6 2 2 2 1 6 4
155.4 155.3 149.8 154.7 150.6 155.4 155.1 150.1 154.5 150.5
3128 2914 3226 2746 2945 2639 2827 3080 2863 3094
.17 .15 .14 .14 .13 .14 .14 .10 .13 .11
159.2 158.7 157.6 158.2 158.6 158.6 158.6 158.3 158.9 158.2
I TABLE 5. AVERAGE GYRATORY PROPERTIES.
SECTION NUMBER
LIFT LAYER
NUMBER OF SAMPLES
GSI'
Gel 2
1.060 1.026 1.032 0.999 0.983 1.056
0.990 0.987 0.988 0.987 0.987 0.986
BULK DENSITY
FLOW (in)
TMO3 (pcf)
5001 5818 5984 5400 5044 5155
.12 .15 .11 .15 .10 .11
161.9 161.5 162.0 161.4 161.7 161.5
STABILITY (lbs)
I
(pcf) 4 5 6 10 11 12
TOP BOTTOM TOP BOTTOM TOP TOP
4 2 3 1 1 5
157.0 154.8 156.0 154.7 151.9 156.9
BOTTOM
2
153.8
1.016
0.986
6116
.11
162.0
TOP BOTTOM TOP
4 1 2
158.1 153.5 158.2
0.783 1.010 1.054
0.987 0.987 0.986
5776 6690 6385
.11 .12 .11
161.9 163.4 162.2
I I I I I I I 1
Gyratory Stability Index
2
Gyratory Compaction Index
I
3 Theoretical Maximum Density
10
I
6. AVERAGESTABLE VALUES OF DENSITY, ASPHALT CONTENT, AND VOID PARAMETERS FROM FIELD CORES.
SECTION NUMBER
MIX & LIFT'
BULK DENSITY (pcf)
ASPHALT CONTENT (%)
THEORETICAL MAXIMUM DENSITY
ASPHALT VOIDS VOIDS VOLUME' TOTAL FILLED (%) MIX (%)
(pcf)
3 2 3 4 5 6 7 8 9 10 11 12
M T M T M B M T M B GT GB GT GB GT M T NT M B M T M B GT GB G T GB GT
150.6 153.0 148.5 151.6 148.8 148.9 146.3 148.9 146.8 144.9 153.4 152.4 148.2 152.6 148.9 151.6 147.1 149.3 146.0 148.4
6.27 6.29 6.57 6.45 6.09 5.18 5.09 4.85 5.41 5.09 6.51 6.46 6.83 6.24 6.46 5.13 5.00 5.26 4.97 5.12
I I m
I I
iz
Mix Type M - Marshall, G = Gyratory Lift T - Top, B = Bottom 2 Asphalt
Specific Gravity - 1.0293
11
158.8 158.9 158.4 158.5 158.7 162.1 161.5 161.6 161.4 161.7 158.6 158.6 158.3 158.6 158.5 161.5 162.3 162.4 163.4 162.3
M%1 14.70 14.98 15.17 15.22 14.11 12.02 11.59 11.23 12.36 11.49 15.55 15.32 15.76 14.83 14.97 12.11 11.44 12.23 11.30 11.83
5.17 3.71 6.28 4.40 6.24 8.20 9.40 7.85 9.07 10.38 3.25 3.91 6.38 3.81 6.03 6.11 9.41 8.06 10.63 8.57
74.01 80.46 70.89 77.97 69.51 59.80 55.26 59.16 57.74 52.58 82.72 79.71 71.27 79.66 71.49 66.57 54.96 60.29 51.54 58.02
I TABLE 7. AVERAGE RESILIENT MODULUS VALUES (PSI) OF FIELD CORES.
I
TEST TEMPERATURE (°F)
miX DESIGN
SAMPLING TIME (SEC)
3-SEC
41
MARSHALL
0.05 0.10 0.05 0.10
1.85E+06 1.21E+06 1.70E+06 1.19E+06
1.64E+06 1.09E+06 1.78E+06 1.36E+06
1.72E+06 1.13E+06 1.96E+06 1.53E+06
0.05 0.10 0.05 0.10
3.41E+05 2.37E+05 4.27E+05 2.98E+05
3.35E+05 2.45E+05 4.26E+05 3.09E+05
3.80E+05 2.66E+05 4.51E+05 3.33E+05
0.05 0.10 0.05 0.10
7.47E+04 5.91E+04 1.24E+05 9.87E+04
8.24E+04 6.67E+04 1.21E+05 1.06E+05
6.46E+04 5.36E+04 1.26E+05 1.08E+05
GYRATORY
74
MARSHALL GYRATORY
104
MARSHALL GYRATORY
12
CYCLE TIME 2-SEC
1-SEC
I
1 I I I I I I I I I
...
I
.......
+
CI
0 1-0
L
*~ toI
LL
13
I SECTION III
I
MEASUREMENT AND DATA ACQUISITION TECHNIQUES
U
One of the primary purposes of this project was to investigate the rutting processes in asphaltic concrete pavements. Data from both the loading applied to the pavement and the pavement response was observed to evaluate these processes. To better quantify rutting, several types of measurements were made, including loadcart position, velocity, applied load, temperature, pavement deformation, and asphaltic concrete material properties. Descriptions of how these measurements were made are presented in the following subsections. A.
3
LOADCART INSTRUMENTATION
Loadcarts were used at Tyndall AFB to apply the load. They consisted of a front-wheel drive truck with a modified rear frame which held the selected amount of weight. The weight was almost completely supported by an aircraft tire which was mounted through an axle to the rear frame. Computerized data acquisition systems were designed and installed in F-15 loadcarts to monitor the loadings applied to the pavement. These computers were used to record data from each pass and to provide the operator feedback on how the systems were working. The information was gathered from two channels of instrumentation, an optical sensor and loadcells. fll the data was saved and reduced as described in Section IV. 1.
I
3
Optical Sensor
Loadcart position and velocity data was calculated from the output of an optical sensor mounted to the loadcarts, as shown in Figure 2. The sensor responded to white patterns marked on the black pavement by putting out a voltage which was monitored by the computer. The pattern marked on the pavement is illustrated in Figure 3. The optical sensor provides a high voltage whenever a stripe in the pattern is encountered and a low voltage otherwise. The relative times at which the high voltages were first encountered was recorded by the computer. At the end of each loadcart pass the data was stored on a floppy disk and checked for completeness. The automated completeness check involved counting the number of high voltage readings stored and writing the number to the screen. The loadcart operator could then verify that the correct number of pavement markings were encountered. Any deviations from the expected number signaled problems in the system which were then corrected.
14
m
I
l
I
I 2.
rn
I
mB.
m
m
l
I
Loadcells
Static measurements of the wheel loads and tire pressure were made before trafficking began. The total wheel loads used for each of the loadcarts were 19000 and 29500 pounds for the empty F-15 and fully loaded F-15 loadcarts, respectively. These high wheel loads were produced by stacking lead ingots, weighing either 1950 or 3900 pounds each, on the loadcart frame just above the load tire. The total static axle load was measured at a truck scale located on Tyndall AFB. The positions of the weights were marked and recorded so that the same configuration of weight could be produced at the field test site. Each day before trafficking began the configuration of the weights and the tire pressure were checked by the loadcart operators. Any variations in the position of the weights and the tire pressure were corrected. The tire pressures used were 190 and 355 psi for the empty F-15 and fully loaded F-15 loadcarts, respectively. Dynamic variations of the wheel load were anticipated before trafficking. Loadcells were mounted in the structure of the wheel support, as shown in Figure 4, in order to measure these variations while the loadcart was moving. Two loadcells, one in front of and one behind the load wheel, were wired to electrically average the response to the load variations. This electrical average was monitored by the second channel of the computer data acquisition system. Installation of the loadcells was completed before weights were added to the loadcart. The loadcells were placed in the proper position and the connection bolt was tightened enough to just register load on the loadcells. The output of the loadcells was monitored as the weights were applied to provide a calibration relationship. This calibration was used to calculate the variations in the wheel load. ENVIRONMENTAL INSTRUMENTATION Because asphaltic concrete material properties are temperature sensitive, the amount of rutting depends on pavement temperature. An environmental monitoring station was set up next to the 6-inch flexible Marshall design portion of the trafficking test sections. A data recorder monitored the output of the environmental sensors and saved the values at continuous 30 minute intervals throughout trafficking. Five thermocouples were placed to monitor the temperature at different depths below the surface of the pavement: the surface, the midheight, the asphaltic concrete-base course interface, 6 inches into the base course, and 12 inches into the base course. An additional thermocouple was used to monitor the ambient conditions (approximately 6 feet above the pavement surface).
15
m Three other environmental sensors were monitored throughout the test. The measurements included solar radiation intensity, wind speed, and wind direction. These latter measurements were made as a part of the environmental monitoring program, however they have not been used in connection with this
study. C.
i
PROFILOGRAPH
The long-term pavement response to the trafficking was monitored with a profilograph, which measures the contour of the pavement across the traffic lane. The elevations of the end points of the profiles were measured with a rod and level. Variations in elevation across the profile were recorded using One method used a calibrated two different methods throughout the study. roller which rotated I inch for every foot of travel across the lane. A wheel and guide system followed the contour of the pavement, tracing the exact These elevation variations on a sheet of paper attached to the roller. second The disks. on floppy hardcopy profilographs were digitized and stored method utilized a computer data acquisition system in conjunction with the profilograph to directly store the profile data in digital format, as shown in The roller was replaced with a wire wound linear variable Figure 5. The differential transducer (LVOT) with a 150-inch maximum displacement. wheel and guide system had a direct current differential transducer (DCDT) The output from these transducers was recorded by computer and attached.
I
I I
stored on floppy disk. Profilograph measurements were made prior to and periodically throughout Table 8 lists the traffic levels where profile data was trafficking. Also included in Table 8 are the starting and ending stations for acquired. The starting and ending stations each set of profilograph measurements. change with increasing traffic level because portions of the test sections Profilograph Set 21 is were removed from trafficking as they became unsafe. defined as the last profilograph at each station. D.
MATERIAL PROPERTIES
An extensive field and laboratory testing program provided a considerable amount of information on the material properties of the test sections. These tests .jere performed before, during, and after construction, and during and after trafficking. The pretrafficking material properties have been summarized in Section II with complete details documented in the Posttest mix designs, trafficked core tests, falling construction report (2). weight deflectometer (FWD) tests, and trench cuts were performed during and The results of these after the trafficking portion of the test. investigations are presented in the following paragraphs.
16
3
I I
I 1.
I
Posttest Mix Designs
The asphaltic concrete mixture delivered to the site contained excess dust, as defined by material passing through the 0200 sieve. A posttest mix design using the Marshall procedure was performed to determine the effect of the excess fine material on the optimum asphalt content. Figure 6 presents the results of these tests. Optimum asphalt content for the Marshall procedure posttest was 5.8 percent, which is 0.6 percent lower than the optimum asphalt content determined in the original mix design series and in the constructed mat. This value is lower because excess dust tends to reduce the required asphalt. Insufficient material remained to complete the posttest mix design using the gyratory procedure. Based on the gyratory stability index of field samples compacted in the gyratory machine, optimum asphalt content for the gyratory mix would also have been lower. 2.
the
FWD Data
Nondestructive testing with the FWD was performed periodically throughout the construction and trafficking portions of the test at specified stations in each lane. FWD results recorded before trafficking have been summarized in an earlier report (7). FWD testing was not possible on sections where the surface profile had become too irregular, because such irregularity causes a nonuniform loading on the FWD plate. The peak deflection data is summarized in Appendix A in the form presented in Table 9, which is a sample set of FWD data. Other important information documented with the FWD data includes date, time, temperature, pass level, and impulsive force for each test. The impulse stiffness modulus (ISM), a measure of the applied load divided by the maximum displacement, for each load, has been calculated and is also included in Appendix A. m
m
3
U
3.
Core Densities
Cores were removed from the test sections periodically throughout trafficking. The samples taken before trafficking were used by Resource International, Inc., and the University of North Carolina at Charlotte for rut prediction and modeling. ARA performed bulk density tests on each of the cores prior to shipping. Table 10 lists the densities of a sampling of cores removed from the fully loaded F-15 lane at various traffic levels (top 4-inch lifts). The values presented in Table 10 are averages for all cores removed from that station at that traffic level. The individual density values are included as Appendix B.
1 m
I
17
I 4.
3
Trenches
Upon completion of the trafficking portion of the test, several locations were selected for destructive testing. The asphaltic concrete was cut with a concrete saw and pulled off with a backhoe. The backhoe was then used to excavate the base course and a small layer of subgrade. The sides of the resulting trenches were cleaned and examined. Figures 7 and 8 show the sides of trenches in the 4-inch flexible Marshall and gyratory sections, respectively. The rutting in the asphaltic concrete and in the base course can be easily detected. Measurements made in the trench reveal that approximately sixty percent of the total rutting could be traced to the granular layers. In the trench at Station 474, base course rutting was not perceptible to the eye. However, there was degradation of base course in the upper 6-inch lift that was trafficked and some increase in dry bulk density under that station.
18
I m
I I I I I I I I I I I I
I TABLE 8. PROFILOGRAPH INCREMENTS.
I
3
PROFILOGRAPH SET #
01 03 04 05 06 07 08 09 11 21A 218 12 13 15 21C 16 21D 17 18 21E 21E 19 21F
TRAFFIC LEVEL (passes)
STARTING STATION (ft)
0 70 112 224 420 448 882 1554 2324 2589 3049 3286 3942 4784 5137 5370 5817 6808 8080 9715 9715 9716 10350
610 610
0
610 610 610 610 610 610 610 36 84 610 610 610 190 610 334 610 610 348 610 610 526
0
38 86 86 86 86 192 192 336 336 336 528 336 350
I I I I I I I
0 0 0 0 0 0 0 0
U
19
ENDING STATION (ft)
I TABLE 9.
FALLING WEIGHT DEFLECTONETER TEST RESULTS. SECTION 1
Thickness Base - 12 Inches Date - 10/04/88
Thickness Asphalt Concrete - 4 Inches Surface Temperature - 65.3 0 F
STATION feet
TIME hr
LOAD lbs
DI
D2
O O 0 0
922 922 922 922 922
9711 9536 14558 18945 26605
13.03 12.36 18.82 23.98 33.58
6.65 6.54 10.04 12.80 17.91
3.39 3.31 5.20 6.69 9.41
2.20 2.20 3.43 4.37 6.22
1.69 1.77 2.80 3.50 4.65
12 12 12 12 12
923 923 923 923 923
9488 9409 14399 18770 26431
12.83 12.32 18.90 24.69 34.76
6.81 6.61 10.20 13.11 18.15
3.31 3.39 5.16 6.65 9.33
23 23 23 23
925 925 925 925
9520 9377 14320 18722
13.58 12.76 19.53 24.84
23 35 35 35 35 35
925 926 926 926 926 926
26129 9393 9313 14240 18675 25668
34.76 12.87 12.24 18.62 23.66 32.95
47 47 47 47 47
927 927 927 927 927
9377 9313 14240 18611 25827
13.50 7.83 3.82 12.91 7.64 3.78 19.80 11.61 5.75 25.47 14.88 7.48 35.98 20.67 10.35
2.44 1.85 1.42 1.26 2.40 1.85 1.42 1.30 3.62 2.76 2.32 1.89 4.76 3.70 3.07 2.60 6.57 4.96 4.09 3.43
694.6 721.4 719.2I 730.7 717.8
58 58 58 58 58
929 929 929 929 929
9091 9202 14113 18532 25843
12.36 7.28 3.70 11.97 7.20 3.66 18.90 11.14 5.87 24.49 14.41 7.60 34.45 20.28 10.55
2.36 1.73 1.34 1.06 2.40 1.73 1.38 1.10 3.70 2.68 2.13 1.77 4.92 3.66 2.91 2.48 6.73 4.92 3.94 3.31
735.5I 768.8 746.7 756.7I 750.2
o
DEFLECTIONS (oils) D3 04 D5 D6
D7
ISkips/in
1.34 1.38 2.13 2.76 3.82
1.14 1.14 1.73 2.32 3.19
745.3 771.5 773.5 790.0 792.3
2.20 1.61 2.28 1.69 3.39 2.56 4.33 5.94 3.35 4.49
1.30 1.42 2.05 2.72 3.70
1.10 1.18 1.73 2.28 3.15
739.5 763.7 761.9 760.2 760.4
7.40 3.27 7.13 3.23 10.91 5.12 13.78 6.57
2.09 2.01 3.27 4.21
1.61 1.57 2.48 3.27
1.18 1.10 2.13 2.76
1.18 1.10 1.69 2.24
701.0 734.9 733.2 753.7
19.09 7.13 6.93 10.47 13.39 18.70
5.79 2.40 2.48 3.58 4.72 6.50
4.80 3.70 1.81 1.46 1.93 1.50 2.76 2.17 3.54 2.91 4.84 3.90
3.11 1.18 1.22 1.77 2.48 3.31
751.7 729.8 760.9 764.8 789.3 779.0
9.21 3.50 3.54 5.31 6.97 9.80
I
I I. 20
I
II I
-In
.0P%
1.0
I
LL
ii
i
i
LiA 4w
-
*
-
-
i
in .n
in
II L21
-
0
-D
-
i
4LP N~I
n
n
n
en
n
n P. NW NW
n
v
I,
in
.
. N
0
O t 4'".e. tP
v'
w
P% int
v'
in -0 4
&' n
v'
&n
v*
%n 4n
0% -W
%l 0
1
v
v
I in
U
I I
I I j
I
I
Si
lllll
llm•
Jl111
mmJI
••
|
I
I! 2ii
:1!
CIL
I
WI 00 'CCIS
a-L
23
LLI -Ie -ICL
24
Ir
Im I3 Ii *I
ui
cm
25.
VOIS TOTAL. MO
(O FILLEU
70
a
5.5
1
7
.
ASPHAL&T COwnTE
7.5
0
5
MAFIHALL FLOW
&"
0.3
155.I
a,-5
7.5
ASPHALT CONTIENT
COMPATE 08N4f
1545
-
_
1523
5
5.5
7
as5.
7.5
55
5
a
ASPHALT CONTENT
6
6.5
7
7.5
ASPHALT CONTENT
DESIGN ASPHALT
CONTENT DETERMINA11ONI MRHL
TARGET
TB"PROPERTY
PEAK
5.75
VF
4% 75%
DENSITY
PEAK
6.5I
STABILTY VTm
.... .....
...........-
5.55 5.85 AVG=
---.
__
%AC
5.91
REQUIREMENTS TO MEETI ACCEPTABILITY CRITERIA
___
41 ASPHtLT
PROPERTY
TAGTC
VF
70- 80%
FLOW
0.08 -0.16
NO CHANGEI NO CHANGE
DESIGN ASPHALT CONTENT - 5.8% Figure 6. Posfles Marshall Mix Design Results.
263
4
I I .. ,.lm
I
I
U I
I
I Figure 7. Trench Side View in the Marshall 4-inch Flexible Test Section.
I
B/
I I 1 I
I I
I!
I SECTION IV DATA REDUCTION AND ANALYSIS
Some of the data recorded from the measurements described in Section III required conversion into a form that would be useful to engineers and scientists. This usable form of the data was maintained in a database for ease in querying and overall management of the various data types. Loadcart velocity and lateral position were obtained from the optical sensor data. Loadcells provided a measure of the dynamic fluctuations in the wheel load. Environmental instrumentation recorded the temperatures needed to estimate the in situ pavement material properties. The rutting of the test sections was monitored by profilograph. Algorithms used to convert this data into usable form are explained in this section. In addition, overall results from each type of measurement are presented and discussed for the fully loaded F-15 lane. Results from the empty F-15 lane will be presented in a later report. A.
I I
3
VELOrITY
The velocity of the loadcart was calculated from the timing data produced by the optical sensor. The output from the optical sensor was sampled at a specified data acquisition rate. This rate served as a clock for determining the relative time at which each white stripe was detected. The velocity over the M-shapes was then calculated as the time it took the loadcart to traverse the pattern divided by the 12-foot distance between each vertical stripe of the pattern. This calculation procedure assumes: (1) the velocity was constant over each M-shaped pattern and (2) the distance traveled by the loadcart was 12 feet for all M-shaped patterns. Any deviations from these assumptions will be a source of error in the reduced loadcart position data. The assumption of a constant velocity over each N-shape was valid over most of the trafficked length. However, this assumption does not hold true during the acceleration and deceleration portion of each pass. It was not uncommon to find a 1 to 2 mph difference in the calculated velocity from one3 M-shape to the next during the acceleration and deceleration stages. To minimize the error in the position calculations caused by this variable velocity, a modified velocity value was used in the calculation of the loadcart position as it passed the first angled leg of the M-shape. This modification was not necessary for determining the loadcart position while passing over the second angled leg, because the time data used comes from the central portion of the N-shape where the actual velocity is most likely to agree with the calculated velocity. Figure 9 illustrates the initial portion of a velocity profile as calculated by the data reduction code. This variation in velocity is assumed to actually occur in the field. A modified
28
I I
3 I I
I I
m
velocity value was calculated only when the difference in velocity between two adjacent M-shapes was 1 mph or greater. The rate of change (slope) in the velocity between the two H-shapes was used to adjust the velocity to more accurately represent the true velocity of the loadcart as it passed over the first angled leg of the N-shape.
3
Because the loadcart naturally deviated from the intended straight path as it traversed the test sections, an increase in the true distance traveled resulted. The error caused by this deviation is very small. A 3-foot deviation in Y position over a 12-foot length causes only a three percent error in the velocity calculation. Visual observations by test personnel and Y position calculations verified that maximum lateral deviations were on the order of inches across any 12-foot section and therefore would not significantly affect calculated results.
3 l
3
l
The averaJ velocity of the loadcart while traveling forward was generally 13 mph and consistently about 9 mph while moving backwards, as shown in Figure 10. The maximum backwards velocity attainable was limited by the reverse gear. The acceleration and deceleration zones typically extended about 36 feet (three N-shaped patterns) into each end of the test sections. Forward passes and reverse passes were labeled with odd and even pass numbers, respectively. Velocity histograms were developed for the acceleration/deceleration zones of the trafficked length. Figure 11 shows the velocity histograms for Stations 0-48 at Pass Level 2589. Stations 0-12 have a wider range in velocity values while Stations 36-48 are outside the zone of any acceleration/deceleration effects. The velocity histograms for Stations 562-612 at Pass Level 2589 are illustrated in Figure 12. These histograms also indicate that the first 36 feet are within the acceleration/deceleration zone. Further from the ends of the lane the concentrations in velocities around 9 and 13 mph correspond to the nominal reverse and forward velocities. For comparison, Figure 13 shows the velocity histograms from two 12-foot sections located outside the range of acceleration/deceleration. The histograms for these same two sections are repeated in Figure 14, complete to their final pass level. Stations 372-384 show an increase in the number of velocity occurrences at 10-12 mph after Pass Level 5817. Although this section is centrally located to the entire test section length, Stations 0-336 were no longer trafficked after Pass 5817 because of excessive rutting. Therefore, after Pass 5817, Stations 372-384 fell within a new acceleration/deceleration zone. B.
LATERAL POSITION
The lateral position of the loadcart was calculated each time the loadcart passed either of the two angled stripes within each M-shaped pattern
l
I
29
I (Figure 15). Using the optical sensor data, the velocity was first calculated for each H-shaped pattern as previously described. By using the difference in time at which each angled stripe was detected, coupled with the calculated velocity, two separate longitudinal distances were determined for each M-shape. These distances are located between the first vertical member of the M-shape (e.g., Point 4) and the adjacent internal angled stripe (Point 5), and between the two internal, angled stripes (Points 5 and 6). The lateral Y position of the loadcart was then determined algebraically based on these intermediate longitudinal distances, as described in Figure 15. The intermediate longitudinal distances from the H-shape pattern were then recalculated to correspond to the station designations along the length of the test path. Errors in the calculated lateral positions could be caused by several factors including the assumed constant velocity across any one N-shape, missing the true edge of a white stripe due to the sampling rate, or extraneous points detected by the optical sensor where no white stripe actually exists. The lateral position calculations were corrected for Figure 16a shows a acceleration/deceleration and are considered accurate. hypothetical path of the optical sensor as it crossed over a white stripe of the N-shape, with sampling locations highlighted. Note that for a constant velocity, the sampling rate can be expressed in terms of samples per inch instead of samples per second. The distance between samples then becomes the maximum potential error by which the optical sensor may miss the start of a white line. This error will be reflected in the lateral position calculations because these calculations are based on the time interval between the start of adjacent white lines as detected by the optical sensor. Figure 16b shows the magnitude of this error for various loadcart velocities, using the sampling rate used in this study (150 Hz). For a velocity of approximately 13 mph, the maximum potential error in lateral position is about 1.5 inches.
I
I
3
Another source of error in the lateral position data was spurious detection of white stripes by the optical sensor. These extra data points were probably due to either light reflecting off the flat faces of the gravel in the pavement mix, or water puddles. For those passes containing a minimal number of extra data points, the erroneous points were easily recognized and were deleted. Table 11 is an example of data for a pass having an extra white stripe found by the optical sensor. The "icount value" in this table is the time (in counts) at which the optical sensor detected a white stripe. The "delta value" is the difference in time (in counts) between finding each white stripe. Note that the delta values maintain a pattern throughout the data on this pass which reflect the characteristic nature of an H-shaped pattern. The extra white stripe found is Data Point number 72. The effect of this extra white stripe is shown in Table 12. The lateral position data for this pass averages about 35 inches, however, beginning at Station 332 and beyond, the data is obviously wrong. By deleting the erroneous data point the proper
I
30
I
I 1
3
I
3 1
timing pattern in the delta value (and hence the Y position values) will be restored. However, for some passes with several extra white stripes, this Judgment was not easily applied and the data was left as is. For these cases, the lateral position data was obviously incorrect from the point of the extraneous data and throughout the remainder of the pass. The erroneous portion of these passes was removed from the database so as not to bias results. Less than ten percent of all lateral position data was omitted from the database. 1.
Histograms
An important step in reducing the optisensor data to a manageable volume was the calculation of load position histograms. These load histograms show the number of passes versus the Y position of the load tire for a given station. Load histograms were determined for stations and pass levels that corresponded to the stations and pass levels at which profilographs were taken. The applied loads (in histogram format) could then be directly compared with the measured rut profiles and corresponding damage parameters. *
The bar width for the load histograms was selected based on two considerations. First, from basic statistics considerations (8), the following equation can be used to estimate the appropriate number of intervals (k) for plotting the histogram of a given data set,
Ik
-
1 +
3.3 log(n)
where n is the number of data points in the set. At the first pass level of 70 passes, just seven intervals would satisfy this criterion. Given the measurement extent of 72 inches across the lane, seven intervals corresponds to a histogram bar width of 10.3 inches. At the highest pass level, 10350 passes, using this equation results in 14 intervals that are each 5.1 inches wide. The second consideration in determining a bar width was a physical one, tire width. The footprint of the fully loaded F-15 tire was approximately 8 inches wide. Because this width was also in the range of statistically-reasonable histogram bar widths, one tire width (8 inches) was used to calculate all load histograms.
3interest,
3
A computer code was written to automatically compute and plot the load histograms from the optisensor data. For each pass at a given station, the available X and Y position data (these data had been stored only at locations where the optisensor crossed an angled white stripe) was interpolated to find Y position at the station of interest. Seventy-two 1-inch bins across the traffic lane were used to count passes at each station. From the 1-inch bins, histograms could then be constructed with an arbitrary positioning in the Y-direction. For this study, the load histograms were calculated and positioned either about the mode of the load distribution or
3
I
31
I about the position of the peak true rut depth (at that station, up to that pass level). The first type of histogram insures that the central bar of the histogram will usually be the largest (resulting in histograms with a consistent shape). The second type allows the number of passes in one tire width directly above the peak rut to be determined (important in evaluating some of the rut prediction methods). Mode-centered load histograms have been computed and plotted for all 1030 combinations of pass levels and stations for which profilographs are available.' A sample of this histogram data is shown in Figure 17 for Station 364. Note that the beginning and ending positions of the load histograms shown in Figure 17 shift with the mode of the traffic distribution. With these histograms, the build-up of loadcart traffic can be traced from start to finish. The last histogram shows the distribution of completed traffic. The final histogram from Station 364 is compared with the intended traffic distribution in Figure 18. Variations in the mode, mean, and standard deviation of the traffic distribution are shown in Figure 19 for Station 364, which is typical of other stations. The mode and mean oscillate early in trafficking due to the small number of data points. After about 1000 passes, however, the traffic distribution stabilizes and the mode, mean, and standard deviation do not change much. The observation that the mode and mean are several inches apart is also typical of most stations, a result of the typically asymmetric (skewed) load distributions. When taken in series, the load histograms also provide summary information about the wander of the loadcart (i.e., variation in Y position down the traffic lane) during trafficking. The mode and mean Y position of the traffic distributions are plotted versus station in Figure 20 for several intermediate pass levels and for the final pass level achieved at all stations. This load position is compared in subsequent analysis to the lateral position of peak information rut depth. To consider the overall loading applied to each type of pavement, representative load distributions were developed on a section by section basis. These distributions were developed by counting any part of the tire which passes over the 1-inch wide bins set up to track the lateral location. Each loadcart pass would then contribute to the total passes in eight of the 1-Inch bins. By simultaneously considering all stations within a selected longitudinal distance, the number of total observations greatly increases and the distributions can be presented in 1-inch increments. Representative distributions for each length of pavement with an equal final pass count are 1. All the mode-centered histograms are documented in an unpublished supplement to this report, Appendix I.
323
I m
m
I
m 3 I
3 3
I -U
shown in Figures 21 through 26 for Sections 1 through 6, respectively. The dotted lines on each distribution represent plus or minus two standard deviations from the average distribution and indicate the relative variability in load distribution from station to station within a given section. With the exception of Stations 336-348 (Figure 24a), the variation within each length of pavement is small. The variation in load position in the length between Stations 336-348 is probably due to its shorter length and subsequently less reliable averaging within the 1-inch bins. C.
TEMPERATURE
measurements made throughout the trafficking portion of the 3study Temperature were loaded into the project database. The data was linked with the trafficking history data and sorted to provide the number of passes in each temperature increment. Figures 27 through 29 present the number of passes (occurrences') versus the temperature data in histogram format for the final pass level. The complete set of histograms for each measurement type is found Appendix C, along with the actual data at each traffic level when profilographs were taken. The average value of temperature for each measurement location and its standard deviation is presented in Table 13. These values were calculated over the complete trafficking portion of the test, so the actual average for sections which were removed from the test early may be slightly different. The average temperature presented for the pavement midheight is 103.5°F. Figure 30 is a plot of average daily temperature values for the pavement surface, pavement midheight, and ambient conditions throughout trafficking. As can be seen in the figure, the temperatures fluctuated around the average pavement temperatures throughout the course of the test.
I Din I
D.
I II
--
Im
LOAD
Visual observation of the loadcart traveling down the test sections revealed an almost continuous vertical oscillation of the entire loadcart. The loadcells mounted on the loadwheel support structure recorded the variations in the load applied to the tire. The tire transferred the wheel load to the pavement, although it is not certain how the tire pressure varied during the dynamic loading. The maximum magnitude of the load variation was nearly eighty percent of the static wheel load. The calibrated output of the loadcells from the fully loaded F-15 loadcart for a given pass is shown in Figure 31. The load varied a total of 15 kips, from 22 to 37 kips. However, the average and standard deviation for this pass were 29.6 and 1.8 kips, respectively. Based on this measurement, the static weight of 29.5 kips is actually a very good approximation of the load for the complete pass. It should also be noted that although a sampling rate of 150 Hz was used, only every other data point was retained in order to maintain a reasonable and manageable quantity of data. 33
I E.
PAVEMENT PROFILES
The profilograph provided a means of recording the permanent pavement response to the load. Profiles were taken prior to trafficking to represent original conditions, at intermittent pass levels throughout the trafficking, and at the end of trafficking to represent final conditions. The original and final profilographs were taken at 2-foot intervals while those taken at the intermittent pass levels were at 4-, 8-, or 10-foot intervals depending upon the variability of the response in each section. Profiles were measured across a 10-foot wide area, which was centered over the approximately 4-foot wide trafficking zone. Between each test section exists a transitional zone created during construction. These transitional zones did not necessarily respond in the same manner as the adjacent pavement. Therefore, a central group of stations were selected to represent each test section to ensure that the results would not be misrepresented by transitional zone stations. 2 Table 14 lists the test sections, their associated station numbers (eliminating transitional zones) and the final number of passes. The final level of passes changes with station number because the test lanes were shortened as trafficking progressed. Also, the first 20 feet of the Marshall 4-inch test section performed poorly in terms of rutting in comparison to the remainder of that test section and was not considered representative of that section. Figures 32 through 34 illustrate typical rutting progression for several stations from the Marshall mix design test sections. Figure 32 shows profilograph data from the 4-inch flexible, Figure 33 shows the 6-inch flexible, and Figure 34 is from the composite test section. Again, note that the final pass levels vary for each location. The Marshall mix design tended to develop multiple ruts (more than one prominent rut) in both the composite and flexible test sections. In the composite sections, there was a smaller overall magnitude in rut depth than the flexible sections. This difference was substantial because, in the flexible sections, a large portion of the rutting occurred in the base support layers. Figures 35 through 37 illustrate typical rutting progression for several stations from the gyratory mix design test sections. Figure 35 shows profilograph data from the 4-inch flexible section, Figure 36 from the 6-inch flexible section, and Figure 37 from the composite test section. The gyratory mix appears to allow a smooth transition from the original surface elevation to its final state, resulting in a single, shallow broad depression with a small amount of upheaval (if any). The overall magnitude of the rutting was
m
3 m
3
m m
3 3
2. The original and final profilographs taken at all stations are available in an unpublished supplement to this report, Appendix J. Selected figures in Appendix J also contain the profilographs taken at intermittent pass levels.
34
I
I
least in the gyratory composite test section. It appears that most of the rutting in the flexible sections occurred in the base course. Representative profiles were computed for the six sections at or near the end of trafficking and are shown in Figures 38 through 43. Where a portion of the section was removed from trafficking, the representative profile was calculated at the pass level where the most stations could be included. In order to develop these representative profiles, all of the profiles were first normalized to the original pavement surface. That is, the rutting displacements for each profile were replaced by the relative displacement from the original pavement surface. Next, the normalized profiles were used to compute the longitudinal mean and standard deviation of the rut depth for each inch of lateral position along the rut, resulting in
E
m
m
IF.
the "representative* profiles. The three curves shown in each figure are the mean rut depth for the subsection and the mean plus and minus two standard deviations. The representative profiles clearly depict the shape of the ruts at the end of trafficking and also the variability of rutting within a subsection. A comparison of the rutting in the 6-inch composite and 6-inch flexible sections provides an estimate of the amount of the base course rutting. Assuming that the rutting in the composite sections is representative of the behavior of the asphaltic concrete, a subtraction of the representative composite rutting from the representative flexible rutting reveals the base course rutting. Figures 44 and 45 show the resulting estimated base course rutting for the Marshall and gyratory sections, respectively. The Marshall comparison was made at a slightly different traffic level (5817 passes for the composite and 5137 passes for the flexible section) and the gyratory comparison was made at the same traffic level (10350 passes). The maximum base course rutting in the 6-inch flexible Marshall section shown in Figure 44 is approximately 1.5 inches. This amount corresponds to sixty-five percent of the total rutting in the Marshall section. The maximum base course rutting in the 6-inch flexible gyratory section is approximately 0.5 inch, corresponding to fifty-nine percent of the total rutting in the 6-inch flexible gyratory section. In addition to the representative profiles, three-dimensional plots were produced for each subsection to allow visualization of the rutting in an exaggerated scale. These plots are included as Appendix 0. The plots show the multiple-rut channels that developed in some of the subsections and also the lateral wander of peak rut depth that developed in some subsections. DAMAGE PARAMETERS Damage parameters were created to provide a consistent set of variables that quantify the rut profiles. These damage parameters were then used to 35
correlate damage to the applied loading. The damage parameters are based on the change in profile from original conditions. Figure 46 illustrates the damage parameters that were calculated from the profilograph data. Three vertical displacements were measured on each rut profile. Two of these measurements focus on the rutting depths. The true rut depth refers to the maximum difference between the original profile and the measured profile at any point along the profile. The apparent rut depth defines the magnitude of the rut as it would be measured in the field using a straight edge and ruler. This measurement would therefore include upheaval. The third vertical measurement taken was the maximum amount of upheaval above the original profile on either side of the the true rut depth. For each profile, the lateral position of all these measurements was also determined. A least-squares regression analysis was completed relating each damage parameter to total number of passes. The regressions were found to have a better correlation coefficient if the natural log of total passes were related to the damage parameter. Figures 47 and 48 present these regressions for two damage parameters, rut depth and maximum upheaval, from the 6-inch composite The remainder of the regressions are presented in Marshall section. Appendix E. The plots show a nonlinear relationship between load and rut depth as expected. The calculated upheaval and true rut depth for the Marshall 4-inch flexible test section at the final pass level is shown in Figure 49. Note that the final pass level was 2589 passes for Stations 0-36 and 3049 passes for the remainder of the section. The initial 20 feet experienced more upheaval and true rut depth than Stations 38-64, even though it had 500 fewer passes. This increased magnitude of rut depth and upheaval is believed to be due to its location within the acceleration/deceleration zone. It appears that a lower velocity accelerated the degradation of the pavement, although an additional factor may be the loading frequency. The loading frequency is different for the end zones than the central portions of the trafficked length. The end zones were loaded twice within a relatively short span of time, whereas those portions in the center had evenly spaced and longer time periods between loadings. This increased magnitude of pavement degradation at the acceleration/deceleration zone was not significant in the gyratory 4-inch flexible test section. Figure 50 shows the vertical displacements for the gyratory 4-inch section at its final pass level of 9715 passes. All displacements are within a reasonably close range although there is a general increase in rut depth towards the end of the test section. Figures 51 and 52 show the displacements from the Marshall 6-inch flexible and composite test sections, respectively, at the final pass level. Although the composite test section had a smaller true rut magnitude, even with more passes, its magnitude of upheaval was far greater than the flexible test section. This probably reflects the inability of the PCC base to deflect. The gyratory composite test
I
I
I
3 I
36
U
section, Figure 53, did not exhibit a noticeable increase in upheaval magnitude in comparison to its flexible counterpart, Figure 54. Additionally, the rut magnitude was much less for the composite section. The true rut depths from the two mix designs for each type of test section are compared in Figures 55 through 57. The data are only presented up to the maximum number of passes incurred by the Marshall mix design. Regardless of pass level, the gyratory mix design consistently exhibited a smaller amount of true rut depth than the Marshall mix design. Also, the rate of rutting in the gyratory test sections appears to flatten with pass level whereas the Marshall seems to increase as the rutting progresses.
i
i
For both mix designs, the composite test sections performed better than their flexible counterparts. Figure 58 compares the rut depth from both mix designs at two pass levels, 1554 and 4784. The Marshall 4-inch flexible section has no data for the 4784 pass level since all trafficking was halted by Pass Level 3049. The figure indicates the superior performance of the composite test sections and, in particular, the superior performance of the gyratory composite at Pass Level 4784. The differences in performance between the flexible and composite sections can be attributed to the rutting in the granular base layers of the flexible sections. These results are also summarized in Figures 59 through 61 which show boxplots of rut depth across the six sections at three different traffic levels (2324, 4784, and 9715 passes). Box plots are not shown for some Marshall test sections at Traffic Levels 4784 and 9715 because trafficking had stopped before these traffic levels. The box plots show the 25th, 50th and 75th percentile values of damage as horizontal lines and the mean damage as a blackened square. The central vertical lines extend from the box as far as the data extends, to a distance of at most 1.5 interquartile ranges (an interquarttle range is the distance between the 25 and 75th sample percentiles). The plots clearly illustrate the relative damage for the six sections and the variability of the damage within a section. All three figures indicate that the least amount of variability was in the composite sections. Damage variability was particularly low in the gyratory composite section. The true rut depths for all the test sections from the gyratory mix design are summarized in Figure 62. The composite test section exhibited approximately half the rutting of the 6-inch flexible and about a third of the 4-inch flexible rutting. The 4-inch test section also showed a wider variation in its response than the composite test section. Wide variations in response were also found for the Marshall mix design, as shown in Figure 63. The Marshall mix seems to have changed behavior once a true rut of approximately one inch was induced. After that point the true rut depth would either: (1) increase at an extraordinary rate due to the channelization of 37
I
I traffic in the rut; or (2) diminish, reflecting a broadening of the rut width due to an omission of passes directly in the rut. Note that after rutting became extreme in the Marshall sections, the loadcart was sometimes intentionally steered to just either side of the rut so as not to cause extreme wear on the sides of the loadcart tires. The mean and mode of the lateral load position histograms were compared to the peak rut depth location and are shown in Figures 64 and 65. The mode tracked the rut depth much better than the mean, as would be expected, since the histograms were slightly skewed. The apparent rut depth is the vertical displacement measured below two points of maximum elevation across the rut (Figure 46). The apparent rut depth was calculated in anticipation of the need to determine the remaining life, in terms of rutting, of a particular taxiway based on field measurements of apparent rut depth. However, it was found that the apparent rut depth is sensitive to the original pavement crown location and the amount of upheaval. Therefore it was not considered a good parameter in determining taxiway lifespan, although it still is important in determining the current safety and serviceability of the taxiway. Two damage parameters related to rut width were also calculated from the profiles, affected width and rut width. The affected width was selected to describe the distance across the lane which showed significant deformation (rut or upheave). The determination of this width was based on a change in elevation of less than 2.5 percent of the true rut depth or 0.05 inches, whichever was greatest. The 0.05-inch cutoff value was chosen based on the accuracy of the profilograph instrumentation, the accuracy of the digitizing of the profllographs, and the naturally occurring roughness of the pavement surface. When a subsequent profilograph did not match smoothly to the original profilograph, the affected width calculations were too large. For these cases, the affected widths were scaled by hand, employing engineering judgment.
The rut width was calculated as the width between the two points of maximum elevation to either side of the point where the true rut depth was greatest. This strict definition did not adequately describe the rut widths for the Marshall mix design sections where there were multiple ruts. For these multiple rut cases, the rut widths for final profilographs were scaled directly from the profilographs. Figures 66 and 67 illustrate the rut and affected widths for the final pass level of the Marshall and gyratory mix designs, respectively. The box plots comparing rut width across the sections at three different traffic levels can be found in Appendix F. In the Marshall composite sections, the rut width is consistently about 45 inches, whereas in the flexible sections the rut width is larger (about 50 inches). For the gyratory mix design the rut width is about 60 inches, seemingly independent of
38
3 I I
3 I
3
3
I I
the base material. For both mix designs, the affected widths are highly variable, ranging from 50 to 90 inches for the Marshall and from 60 to 100 inches in the gyratory test sections. This variation is well illustrated in the box plots for rut width. High variability in rut width is mostly due to the resolution o: the profilographs and the ability to perfectly align subsequent profiles. Two areas were calculated for each profile, anticipating that they could be related to the amount of densification within the asphaltic concrete. The first is the area of upheaval and the second is the true rut area. These areas were calculated as either a positive or negative change in elevation across the affected width of each profile. Figures 68 and 69 present the rut area and upheaval area at the final pass level for the Marshall and gyratory sections, respectively. Figures 70 and 71 show box plots comparing rut area and upheaval area across the test sections at Traffic Level 4784 (see Appendix F for box plots ,,t additional traffic levels). The figures reveal that the Marshall sections upheaved significantly more than the gyratory (exceeding a factor of three for the average upheaval area), for both the flexible and the composite sections. For rut area, the magnitudes are similar at the same traffic level. An important observation, however, is that the gyratory areas do not significantly increase at the higher traffic levels. In fact, the rut areas at the final traffic levels for the gyratory and Marshall sections are similar even though the final pass level for the gyratory sections is two to four times that of the Marshall sections. This is investigated further in the paragraphs below. To quantitatively compare the damage for the six test sections, a statistical means separation analysis was performed. This analysis determines whether or not the average damage in a particular section is significantly different from the average damage observed in all the other sections (considering the damage variability) and then rank orders the damage in the six sections. The analysis included the Multiple T Test, The Waller-Duncan K-ratio T Test, and Tukey's Studentized Range Test. The tests were performed on all six damage measures, for the three traffic levels shown in the box plots. The results are summarized in Table 15. The table shows the rank order of the test sections, from least to greatest damage. Brackets are used to group sections for which the difference in damage was not statistically significant. The results in the tables confirm the observations made earlier: (1) rut depth is significantly greater for the Marshall than the gyratory sections when other design parameters are held constant (i.e., asphalt thickness and base layer design) (2)
rut areas are generally not significantly different for the two 39
I mix designs when other design parameters are held constant (i.e., the Marshall test sections exhibited narrower and deeper rutting than gyratory but the breadth of the gyratory influence compensates for the deeper Marshall ruts (3)
upheaval areas are greater for the Marshall mix than the gyratory regardless of other parameters.
Table 15 shows that, for rut depth, the 4-inch flexible gyratory design is roughly equivalent to the 6-inch flexible Marshall mix design. Thus, the gyratory design allows a savings of 2 inches of asphalt for the same performance as measured by rut depth. Figure 72 shows the rut and upheaval areas for the 4-inch flexible test sections to Pass Level 6000. Both mixes have approximately the same amount of rut and upheaval area through Pass Level 3049 when the Marshall trafficking was halted, although the gyratory rut depth was less than the Marshall. Despite the variability in the affected width calculation, the gyratory design must have a broader basin of rutting than the Marshall. For the 6-inch flexible test sections shown in Figure 73, both mixes appear to have the same magnitude of rut and upheaval areas for only the first 1000 passes. Thereafter, the Marshall areas continued to increase at a faster rate than did the gyratory mix. The gyratory composite and Marshall composite test sections are shown in Figure 74 and Figure 75, respectively, up to Pass Level 6000 when the trafficking in the Marshall test section was halted. For the first 5000 passes both test sections have about the same amount of rut area. After that point, the Marshall clearly shows an increase in rutting. The difference in the response of the gyratory and Marshall composite sections is more apparent in the upheaval area. The gyratory design appears to maintain a relatively small amount of upheaval while the Marshall upheaval area is twice the The gyratory maintains this relatively small amount of upheaval gyratory. throughout the trafficking as shown in Figure 76. In summary, the flexible test sections exhibit an immediate response to trafficking as evidenced by its rut area as a function of pass level. This immediate response in the flexible sections is present regardless of mix type The composite test sections did not show this or pavement thickness. This difference between the flexible and immediate increase in rut area. composite response time is attributed to granular layer rutting. In theory, granular base course materials (not modified) should have an initial, immediate response to trafficking due to their unbound nature. G.
I 3 I
3 I
I
EFFECTS OF COMPACTIVE EFFORT
In design of asphalt mixtures, once an aggregate has been selected, different increments of binder are added and specimens of the mixture are
40
3 m
I
I compacted to determine the effects of overfilling and underfilling the voids in the aggregate. Density and voids filled with binder are two of the parameters used in this analysis. Figure 77 presents data from the test sections regarding the effect of compactive effort on the performance of the mix. The curve in Figure 77 represents the Marshall compactive effort used to design the mixture for the test sections. The top cluster of data points represents the density of the pavement as determined from cores taken from the gyratory test section ruts after traffic. Note that the voids filled are similar for the post traffic gyratory and the optimum Marshall binder content. However, the means by which the voids were filled was attained differently. The voids filled for the gyratory post-trafficking was achieved by compaction whereas the Marshall design curve achieved its voids filled by additional binder. This difference is seen by their performance under trafficking. When F-15 traffic was applied, the gyratory mix remained stable, but the Marshall mix did not. Hence, the laboratory compactive effort used to select the amount of binder must equal that of the traffic. Designing an asphalt mixture using a level of compactive effort which does not simulate the applied traffic may not provide the true optimum design asphalt content. The problem associated with the Marshall compaction procedure is that the impact method cannot reach the density levels achieved under high pressure tires. Further increasing the Marshall compactive effort degrades the aggregate. The gyratory testing machine applies higher compactive effort without degrading the aggregate and can simulate any traffic. The mix designer is then able to select the amount of binder appropriate for the traffic.
U
Specimens of the gyratory mix were taken from the paver and compacted in the laboratory with both Marshall and gyratory methods. The specimen air Svoids were compared to core voids of the same mix taken after construction of the mat and after traffic from the F-IS. Figure 78 dramatically shows the effectiveness of gyratory compaction and inadequacy of Marshall compaction to match the density of the traffic. Figure 78 also shows that the 20-ton rubber tired (RT) roller used to compact the mat achieved an average of over seven percent air voids. Current Department of Defense (DOD) specifications (6) allow for payment deductions for greater than seven percent air voids. These penalties should be relaxed or greater compaction effort used to compact the mats, or both, if gyratory methods are to be implemented by DOD.
I I I 3
*
I
Cores removed from the test sections reveal the difference in behavior of the Marshall and gyratory mixes. Figure 79 illustrates the change in density as a function of pass level for the gyratory test section. The density information is provided in terms of percent theoretical maximum density because it takes into account the slight variation in the as-built asphalt content. As indicated, the gyratory sections densify under traffic. However, at a density level of ninety-eight percent of the theoretical maximum the densification rate decreased to almost zero. The mixture designed for the gyratory test sections was able to support the applied traffic without
41
I failure. The data also reveals that a higher initial compactlve effort in the gyratory test sections could eliminate most of the initial asphaltic concrete densification. It is not known if the 4-inch data is valid due to lack of additional data. Figure 80 compares the final densities attained by both mix designs. The Marshall mix, which had rutted so severely that trafficking was stopped, had densities greater than ninety-eight percent theoretical maximum while the stable gyratory had densities below this point. H.
MULTIVARIATE ANALYSIS
A multivariate statistical analysis was conducted to determine the significance of the test variables on rut damage in light of the damage variability within the sections and other sources of variability such as These results also form a environment, material properties, and loading. basis for developing improved rut damage prediction models. The statistical analysis techniques used were MANOVA (Multivariate ANalysis Of VAriance), Pearson pairwise correlation analysis, and stepwise regression with multivariate incremental R-square measures. The MANOVA was used for initial screening and identification of the test variables, while the correlation and stepwise regressions were used to further quantify the relative importance of the test variables. The database for this analysis consisted of damage measures (response variables) and test variables (antecedent conditions). The antecedent conditions are the measures of load, environment, material properties, and pavement geometry that are used to explain the observed damage. Figure 81 illustrates the 1030 observations of rut damage data available for the analysis, corresponding to the profilograph measurements made at the various stations at various pass levels. For each point in the figure the six damage measures as previously described (as shown in Figure 46) were evaluated. These included true rut depth (DT), maximum upheave area (HUI), second maximum upheave (HU2), rut width (WR), upheave area (A2), and true rut area (Al). To describe the antecedent conditions (loading, environmental, and pavement) for the multivariate analysis, 51 test variables were compiled. In total, input for the analysis therefore consisted of a matrix of 57 variables Due to (51 test variables plus six damage measures) by 1030 observations. some missing upheave area (A2) damage measures, only 1009 observations were available for that damage measure. The test variables can be categorized into three groups: load variables, environment variables, and pavement material and geometry variables. Each of the test variables are listed in Table 16 and briefly described below. Seven test variables were used for characterizing the pavement loading. These seven variables included the number of passes, the natural log of the
42
m I
1
I
3 3 I
I
m
number of passes, and five variables for describing the lateral distribution or position of the load. The primary description of load was total number of passes. The natural logarithm of the total passes was included because earlier work showed a log-linear relation between damage and pass level. The purpose of including the position variable was to study the effect of the load lateral spread on damage. The lateral spread is characterized by both the lateral position standard deviation and the lateral position variance. An additional measure, quasiskew, was also evaluated. Quasiskew is asameasure designed to identify asymmetry in the load histogram; it is defined
i i
*
as: Q-
I(Y
-
U
where
I
The environmental parameters used in the multivariate analysis consisted of the number of passes at different temperature intervals. Earlier work has shown that the interaction of load and temperature is important in modeling pavement damage. This study looked at the effects of this interaction. Interaction cells were defined at each pass level and 10-degree temperature intervals. For example, as shown in Table 16, the number of passes for which pavement surface temperature was between 80 and 90 degrees was used as an interaction cell. Cumulative pass numbers were produced for each temperature type and 10-degree temperature interval, for each of the twenty pass levels at which profiles were taken. Only the eight intervals from 70 degrees to 150 degrees contained passes and for any given temperature type only those intervals that had a nonzero total number of passes were used as variables in the analysis. For example, as shown in Table 16, ambient temperature is characterized by four variables plus average ambient temperature. The average for each temperature type and pass level was estimated as:
Q - quasiskew, - mean load position, I - modal (most frequent) load position, and oY - load position standard deviation.
I
tave.where
(n, X ti) / Z ni
tave - weighted average temperature,
mt n,
-
the midpoint temperature of temperature interval i, and the cumulative count of passes for interval I.
Thus, a weighted average temperature was obtained at each of the twenty pass levels. Eleven variables were used in an initial (Phase I) analysis to describe the pavement configuration and material properties. As shown in Table 16
I
I
43
I these included: mix type, design type, design thickness, asphalt content, asphalt volume, total voids, voids filled, theoretical maximum density, bulk density, base course FWD ISM, and pavement FWD ISM. This set of variables is augmented in a later (Phase II) analyses to include aggregate gradation and constructed asphalt thickness. The first two variables, mix type and design type, are categorical variables, that is, they do not take on continuous numeric values as for the other variables. For multivariate analysis, the standard approach to handling such variables is to represent each value of a categorical variable by an indicator (0-1) variable. This approach was used for the MANOVA. Falling weight deflectometer (FWD) and pavement material property measurements were not taken at all stations and so were estimated for intervening stations by piecewise linear interpolation. The Impulse Stiffness Modulus (ISM) was measured at the top of the base course to represent the underlying granular layers and also again at the top of the pavement to represent all pavement layers. When repeated measurements were taken at a station, mean values were used. Sections 1 and 6 do not have lower lift thickness measurements because a 4-inch thick pavement was laid in one lift at the same time as the top 4-inch layer of the other sections. Thus, the material properties are based on the top asphalt lift only for consistency. 1.
MANOVA of Six Damage Measures
The first step in the statistical analysis was a MANOVA to screen the significance of all the test variables on the six damage measures, both individually and collectively. Due to the large number of test variables, a two phased approach was used in the MANOVA. This was necessary because many of the variables are expressible as linear combinations of other variables and are highly correlated with each other. This has the tendency to mask the importance of some variables and can lead to incorrect conclusions. In the first phase, an analysis where I1 the test variables were included was performed. In the second phase, the results of Phase I were used to select subsets of those test variables that were relatively independent of one another. The dataset used in the Phase I analysis (Figure 81) was also used for the Phase II analysis.
I I
3 I I I
3 I 3
Because the initial analysis used all of the test variables it was expected that many would not be significant. Individual ANalysis Of VAriance (ANOVA) results were produced for each of the six damage measures. In each case a damage prediction model was developed based on the data and the test variables. Although the resulting models are not necessarily useful for damage prediction, they can be used to investigate the importance of the test variables. For each there is an R-Square statistic which measures the proportion of the damage variation explained by the model. The R-Square statistic was significant in all cases at any standard confidence level.
44
I
m
I m
Table 17 shows the R-Square statistic for each of the damage measures from the Phase I analysis. The ANOVA results indicate that the test variables explain the variation in damage across the test sections, particularly rut depth (OT) and rut area (AI). Figures 82 and 83 confirm this, showing the damage predicted by the ANOVA generated regression relationship versus observed damage. It should be noted that the significance of upheave area was reduced by the inclusion of the gyratory composite (Section 4), which did not exhibit plastic flow but did rut some. The results of the ANOVA also showed that the main effects of only a small subset of the variables were estimable, because of the lack of independence among many of the test variables. For purposes of understanding the test results and developing predictive models, the significant variables identified in this initial analysis are not necessarily the best variables to rely on. This is because the analysis may arbitrarily identify as significant only one of a pair of highly correlated variables. In some instances this may be due to confounding in the experimental variables. Engineering Judgment must be used to sort out the important variables and select subsets for further analysis. Based on the Phase I analysis, two different variable sets were selected and used in Phase II. Both sets included a core of fifteen basic test variables that were found to be significant in the initial screening analyses, and a group of seven additional variables, different for each set. Note that several of the variables in the core group are functionally dependent. However, these were retained at this stage based on the initial screening analyses. The following analyses were used to further reduce and rank order the significant variables.
l
The core variables included the two categorical variables: (1) (2)
m
and thirteen other numeric variables: (1) (2) (3) (4) (5) (6) (7) (8)
I I
mix type (Marshall vs. gyratory) and design type (flexible vs. composite)
design thickness natural log of number load lateral position load lateral position load lateral position base course FWD ISM pavement FWD ISM top lift bulk density
of passes standard deviation variance quasiskew
45
I (9) (10) (11) (12) (13)
top top top top top
lift lift lift lift lift
asphalt content (%) theoretical maximum density asphalt volume (%) total voids (I), and voids filled (%).
1
Two different groups of seven additional variables were used in conjunction with the core variables. These seven variables represent different combinations of pass number and trafficking temperature information. The first group contained: (1) (2) (3) (4) (5) (6) (7)
number of passes average temperature 12 inches into the base course average temperature 6 inches into the base course average temperature at base course-pavement interface average temperature midheight in the pavement average temperature at the pavement surface, and average ambient temperature.
As described earlier, these average temperatures are weighted according to the number of passes at 10-degree temperature intervals. The second group of variables contained the cumulative number of passes when the surface temperature was: (1) (2) (3) (4) (5) (6) (7)
I I I 3 I
80-90 degrees 90-100 degrees 100-110 degrees 110-120 degrees 120-130 degrees 130-140 degrees, and 140-150 degrees.
Note that in this second group, total number of passes need not be included since it is simply the sum of the passes at the seven different temperature intervals. Additionally, only surface temperature was used for this initial analysis since the temperatures at the other locations are highly correlated with it. With these two sets of variables it was now possible to compute multivariate statistics of the overall effect of each test variable on the damage measures collectively. Several test statistics were used to assess the significance of the test variables. These test statistics included, Wilk's Lambda, Pillai's Trace, Hotelling-Lawley Trace, and Roy's Greatest Root. All of the test statistics tended to indicate the same relative significance for the test variables.
46I
I
m
The MANOVA produced individual ANOVAs from the first variable subset for each of the six damage measures and the multivariate significance tests for the effects of the test variables on the six damage measures collectively. All test variable main effects were estimable, but several were nonsignificant in the individual ANOVAs, that is, some variables were not significant for some damage measures. The ANOVA for each damage measure was significant at any standard confidence level. Table 18 shows the R-Square statistic for each of the damage measures from the ANOVA. The reductions in R-Square from the Phase I analysis results from using far fewer variables. However, the individual estimations of the main effects are far more reliable
i
in this analysis. The results of the MANOVA indicated that all twenty-two test variables were significant, that is, in terms of contribution to overall damage as measured by all six damage measures.
m
As with the first variable subset, individual ANOVAs were conducted for each of the six damage measures from the second variable subset and a MANOVA was performed for the effect of the test variables on the six damage measures, collectively. Again, all test variable main effects were estimable, but several were nonsignificant in the individual ANOVAs. The ANOVA for each damage measure was significant at any standard confidence level. Table 19 shows the R-Square statistic for each of the damage measures from the ANOVAs. As in the previous analysis, the R-Square values are smaller than those from the Phase I analysis. Again, this results from using fewer variables. The results of the MANOVA, indicated that twenty-one of the twenty-two test variables were significant, that is, in terms of contribution to overall damage as measured by all six damage measures. The number of passes when pavement surface temperatures were between 140 and 150 degrees was not found to be significant. This is most likely because there was only a small number of passes at this temperature interval.
I
2.
Pearson Pairwise Correlation Analysis
A Pearson pairwise correlation analysis of all the variables was also performed to help in explaining the results of the MANOVA and to further quantify the relative importance of the test variables and their relationship to each other. This analysis is a basic multivariate technique to compute correlation coefficients for all possible pairs of numeric variables in the data. Again, to include information about the two categorical variables, mix type and design type, two indicator variables were computed. The mix type indicator variable was set to zero for the Marshall mix and one for the gyratory mix. Similarly the design type indicator variable was set to zero for
I
I
47
I the flexible design and one for the composite design. variables were added at this stage:
Also, three additional
(1)
actual asphaltic concrete layer thickness, taken as the difference in pavement surface elevation to the base course surface elevation
(2)
dust (percent aggregate in the asphalt mix passing a #200 sieve)
(3)
sand (percent aggregate in the asphalt mix passing a #8 sieve and retained on the #200 sieve)
Appendix G contains the results of the correlation analysis in the form of a matrix entitled "Pairwise Correlations of All Possible Pairs of Non-Zero Variables.* For each pair of variables, the matrix contains three numeric entries. The first entry is the Pearson correlation coefficient. The second entry is the "p-value" for the test of significance of the correlation coefficient. For small p-values, usually < 0.05, the hypothesis that the correlation coefficient is zero is rejected. The third entry is the number of observations available for computation of the correlation coefficient. Guided by the initial ANOVA and NANOVA results, several interesting observations can be drawn from the correlation analysis. The objective of the correlation analysis is to identify variables that could be used in a rut damage prediction model. First, there is a very strong and significant correlation between design type (i.e., flexible vs. composite) and the pavement falling weight deflectometer measurements. The correlation coefficients are approximately 0.99. This indicates that the FWD measurement is an excellent indicator of the base support type, independent of the pavement surface mix type. This is further borne out by the lack of correlation between the FWD measurement and the mix type (nearly zero). Figure 84 illustrates this graphically, showing the pavement FWD readings taken across the test sections. Note that the plot shows the raw data as circles and the interpolated values used in the ANOVA as crosses.
m
I I I
I
I
Next the correlation between the damage measures and the FWD measurements can be examined. The FWD measurement correlates well with rut area (correlation coefficient of approximately seventy percent) and is also somewhat correlatable to rut depth (approximately fifty percent). This is true only because of the overwhelming Influence of design type, i.e. difference between the underlying PCC and aggregate support shown in Figure 84. There was no correlation between FWD ISM and rutting within a given test section.
48
I I
Finally, it is seen that the mix type is highly correlated with all of the asphalt properties, including bulk density, asphalt content, theoretical maximum density, asphalt volume, total voids, and filled voids. This correlation indicates that these measured properties were significantly different for the two different mixes and could, therefore, form the basis for a quantitative model. This is portrayed graphically in Figures 85 through 90. Based on the correlation analysis, the pairwise relationship among the six damage measures can also be examined. Table 20 shows the correlation among the six damage measures, extracted from the matrix in Appendix G. The analysis verifies the expected correlation between rut depth and rut area and also the expected correlation between upheaval height and upheaval area. It also shows that there is a strong correlation between the two upheaval height measures. However, for several of the measures, the correlation is not particularly high, for example, the upheaval measures versus rut depth measures versus rut width. Thus, depending on the relative importance of each of the damage measure in describing the functionality of the pavement, it may be desirable to retain all six measures in further analysis and development of prediction models. 3.
Stepwise Regression Analysis for Variable Rank Ordering
A stepwise regression analysis was performed to build the largest, most explanatory model for each of the six damage measures. Starting with the most important test variable, new test variables were added until no more significant ones could be found. During this process if any test variable lost its significance by virtue of being in combination with others in the model, it was removed, and the process continued. This analysis was used to rank the test variables in terms of their importance in explaining observed damage. It can also be used to build the best predictive model with the least number of variables. As for the MANOVA, the stepwise analysis was performed in two phases. The stepwise analysis was begun (Phase I) using both variable subsets used in the MANOVA Phase II analysis. The results of this Phase I analysis were used in conjunction with the correlation and MANOVA results to identify a reduced set of variables for the Phase II stepwise analysis. The results of the Phase I analysis showed that the temperature measurements contributed relatively little to explaining the damage variation in the test sections. On closer examination this should be expected because of the manner in which the test conducted. There was not a sufficient variation in temperature at different stages of trafficking to quantify the relative contribution of the number of passes at the different temperature intervals. Most of the rutting occurred early on and the temperature did not
49
I
I vary widely. Hence, the temperature variables were deleted in the Phase II stepwise regression analysis. Several additional variables were also deleted for the Phase II stepwise regression analysis based on the results of the correlation analysis and the functional relationship among several of the asphalt property descriptors. The variables that were deleted included: total voids, voids filled, asphalt volume, base course FWD ISM, mix type indicator (Marshall versus gyratory), and design type indicator (flexible versus composite). The latter three variables were deleted because the correlation analysis showed that pavement FWD ISM is a good indicator of design type regardless of mix design and that mix type is adequately characterized by the remaining asphalt variables of asphalt content, bulk density, and theoretical maximum density. Hence, the final retained variable set for the Phase I1 stepwise analysis represents a set of relatively independent variables that can be measured in a field evaluation. The best models found for each damage measure are listed in Appendix H. Also shown in Appendix H are the R-Square statistic for each model and the mean square error (mse) of the estimated unexplained variation. The F-test statistic shown for each variable measures the significance of its parameter estimate and is a good indicator of the test variable's relative importance and proportion of explained variation attributable to that test variable. Used with a normally distributed error term having a mean of zero and variance equal to the mean square error, these models would provide reliable predictions or simulations of damage for pavements and conditions within the ranges of the test measurements. The overall importance of each final test variable on all of the damage measures collectively was calculated using multivariate incremental R-square measures or average "percentage of damage variation explained" by the test variable. These measures were calculated as:
I
I I I
OVERALL IMPORTANCE - 100/6 * I [ (R-square * F)/ E F] where the outer sum is over the six damage measures, the inner sum is over all significant test variables within each the damage measure model, and F is the order-independent F-test statistic for the significance of the parameter estimate of each test variable. The results are shown in Table 21. Also shown are the number of damage measure models for which each variable was significant, as well as statistics for that variable across those models. 4.
Conclusions from Multivariate Analysis
I
Immediately apparent is the importance of the base support as quantified by the pavement FWD measurements. This importance is expected based on the results of the MANOVA and correlation analysis, and the fact that
50
I
I
I
*
a large percentage of the total permanent pavement displacement of the flexible sections occurred in the base layers. The results show that the base support is the single most important variable when the set of six damage parameters is considered collectively. However, these results do not provide a basis for studying the relative importance of base course properties (as measured by the FWD) to rutting in flexible pavements because the correlation does not hold within pavement types (i.e. for the flexible and composite pavements separately). Surprisingly, in regards to using the FWD as a predictor of rutting potential, it should be noted that the pavement FWD measurements could not discern mix design type. The measures for the loading (passes and log of passes) collectively explain roughly twelve percent of the damage variation. This number is not as large as might be expected because this analysis identifies those variables that explain the variation of damage in the dataset. Although the dataset captures the variation in damage with number of passes, number of passes does not account for differences in damage for the different test sections. Also, the rate of damage tends to level off as trafficking progresses, so that pass number can explain less damage variation as trafficking progresses. The regression analysis results for damage versus passes for the individual sections more clearly shows the relationship between loading and damage. Taken together, the five pavement material measures (bulk density, theoretical maximum density, asphalt content, and the two aggregate gradation variables, sand and dust) explain about fifteen percent of the damage variation. This contribution is expected since the density and asphalt content measures were shown earlier to discriminate between the Marshall and gyratory mix design. The gradation measures contribute to explaining rut damage variation within a section.
I
The significance of the load distribution spread and shape (as measured by the variance, standard deviation, and quasiskew) on damage is not clear at this point. While there is a statistically significant contribution, it is relatively small. These results can be explained by examining the correlations between the load spread and damage measures shown in Figures 91 and 92. Although the correlation analysis does show a greater correlation between load spread and rut depth than for load spread and rut area, as would be expected, the correlations are very small for both. This is most likely a result of the small variation in load spread during the course of trafficking as can be seen in the figures. An important observation, however, is that quasi skew, which is a measure of load distribution asymmetry is in all cases Hence, the positively correlated in the damage as might be expected.
indication is that an asymmetric load distribution will lead to greater damage.
51
I Asphalt thickness was found to explain only about two percent of the damage variation when considering all six test sections. The contribution of asphalt thickness is overshadowed by the much greater effects of base support and mix design types. However, within a given section, for a particular pavement design, thickness may be a very significant variable. For this reason an additional study was performed to investigate the importance of asphalt thickness. This study was conducted by performing a MANOVA and then individual ANOVAs for each section. The analysis examined the effect of thickness by evaluating how the damage variation is explained by four variables: section identifier (ID), actual asphalt thickness, number of passes, and log of number of passes. Figure 93 shows a plot of the actual asphalt thickness for the test sections. The section ID is included in the variable set for the MANOVA since this effectively 'factors out" the effect of design difference, leaving only the variation within a section to be explained. Considering all six sections and all six damage measure, the 4ANOVA showed that asphalt thickness was highly significant in explaining damage variation within a section. Next, the individual ANOVAs were performed. The results of this analysis showed that asphalt thickness was again significant for all sections, except that it is only marginally so for the 4-inch flexible Marshall section and the 6-inch gyratory composite section. As mentioned earlier, the test conditions did not provide sufficient data to quantify the effects of temperature. There was not a sufficient variation in temperature at different stages of trafficking to quantify the relative contributions ýJ the number of passes at the different temperatures intervals. The results and conclusions analysis can be summarized as follows:
of the multivariate
statistical
(1) Pavement FWD measurements can differentiate between different base layer designs, regardless of the surface course design.
I I
3 i I
m
(2) Mat asphalt content, bulk density, and theoretical maximum density may be key indicators for predicting surface rutting. However, this conclusion requires further evaluation. (3)
Load asymmetry was found to be positively correlated with all damage measures for all sections. Hence damage prediction models should incorporate an allowance for expected trafficking asymmetry.
(4)
Multiple damage measures should be retained for further analysis and development of pavement performance evaluation models.
52
I
m
I I
I
(5)
Actual thickness of the asphalt layer is significant in explaining rut damage variation within a section.
(6)
Pavement aggregate gradation is significant in explaining rut
damage variation.
I i
(7)
The quantitative effect of temperature on rut damage could not be evaluated.
(8)
The effect of load magnitude on rut damage has not yet been quantified.
I
i I I I I I I
53
I TABLE 11.
SAMPLE OPTISENSOR DATA FOR A PASS WITH AN EXTRA STRIPE.
PASS NUMBER - 4929, LANE - Heavyweight NUMBER OF LINES FOUND - 134 NUMBER OF LINES EXPECTED - 133 DATA ICOUNT DELTA DATA ICOUNT DELTA DATA ICOUNT DELTA PT.# VALUE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
1137 1233 1332 1379 1428 1523 1565 1610 1694 1735 1778 1860 1899 1940 2020 2059 2100 2180 2220 2260 2340 2380 2421 2500 2539 2581 2659 2698 2740 2817 2858 2900 2979 3019 3060 3139 3178 3218 3301 3338 3378 3463 3502 3542 3634
VALUE 1137 96 99 47 49 95 42 45 84 41 43 82 39 41 80 39 41 80 40 40 80 40 41 79 39 42 78 39 42 77 41 42 79 40 41 79 39 40 83 37 40 85 39 40 92
PT.# VALUE 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
3673 3714 3792 3825 3858 3920 3948 3976 4033 4060 4088 4139 4163 4188 4237 4258 4281 4330 4351 4373 4421 4442 4465 4513 4535 4557 4584 4605 4626 4649 4696 4718 4742 4788 4811 4834 4884 4905 4927 4977 4998 5021 5068 5091 5115
54
VALUE 39 41 78 33 33 62 28 28 57 27 28 51 24 25 49 21 23 49 21 22 48 21 23 48 22 22 27 21 21 23 47 22 24 46 23 23 50 21 22 0 21 23 47 23 24
PT.# VALUE 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
5159 5182 5207 5247 5273 5300 5340 5366 5390 5434 5458 5482 5528 5551 5573 5623 5644 5666 5716 5737 5758 5809 5830 5850 5901 5921 5943 5993 6014 6036 6085 6107 6129 6177 6199 6222 6273 6296 6322 6377 6403 6432 6500 6532 0
VALUE 44 23 25 40 26 27 40 26 24 44 24 24 46 23 22 50 21 22 50 21 21 51 21 20 51 20 22 50 21 22 49 22 22 48 22 23 51 23 26 55 26 29 68 32 0
1
I
I
I I
I
I
I
TABLE 12.
I
EFFECT OF EXTRA POINT ON LATERAL POSITION.
PASS NUMBER - 4929,
U
X POS. FEET
Y POS. INCHES
TIME SECONDS
X POS. FEET
608.972 603.143 597.214 591.085 585.199 579.269 573.229 567.229 561.300 555.300
39.165 34.881 36.265 34.186 36.448 36.399 36.086 35.914 35.230 36.770 35.001 37.446 34.330 37.670 35.462 36.991 36.368 36.991 36.130 38.570 35.895 37.446 35.462 36.991 34.330 34.970 33.452 35.914 32.014 34.512 37.172 32.933 35.121 36.293 34.330 36.384 37.476 35.223 35.489 34.237 33.943 34.186 33.143 34.901
0.343 1.003 1.643 2.276 2.856 3.416 3.976 4.523 5.056 5.589 6.123 6.656 7.189 7.723 8.263 8.789 9.329 9.849 10.389 10.903 11.456 11.983 12.523 13.049 13.576 14.129 14.643 15.209 15.736 16.349 16.883 17.403 17.843 18.256 18.629 19.009 19.376 19.716 20.043 20.369 20.663 20.989 21.276 21.596
345.407 339.214 332.604 327.975 321.606 318.573 309.505 306.375 297.500 294.625 285.665 282.827 273.606 270.573 261.342 258.177 249.239 245.830 237.020 233.536 225.056 221.992 213.311 210.247 201.470 198.691 189.665 186.827 177.665 174.956 165.636 163.027 153.766 150.897 141.636 138.766 129.505 126.636 117.625 114.750 105.721 102.721 93.723 90.893
I549.319 543.356 537.375 531.375 525.281 519.318 513.205
I507.318 501.225 495.450 I489.245 483.356 477.281 471.318 465.375 459.150 453.448
I447.229 441.568
435.112 429.138 422.980 417.309 411.260
S405.375
399.268 393.113
S387.171
381.278 375.089 369.407 363.085 357.474 351.144
I
I
LANE - Heavyweight Y POS. INCHES 33.943 35.735 43.587 44.870 31.561 76.044 32.765 73.670 32.830 76.670 30.846 79.089 31.561 76.044 34.726 71.296 35.966 67.125 38.588 63.605 38.160 69.074 35.096 72.138 33.193 77.459 30.846 79.089 30.846 80.638 31.200 81.496 29.634 79.931 31.200 78.366 32.765 76.800 31.330 78.170 30.176 77.824 30.159 79.890
TIME SECONDS 21.889 22.209 22.503 22.683 22.963 23.116 23.576 23.736 24.196 24.349 24.823 24.969 25.443 25.596 26.063 26.223 26.669 26.836 27.276 27.456 27.896 28.056 28.509 28.669 29.129 29.276 29.749 29.896 30.369 30.509 30.989 31.123 31.596 31.743 32.216 32.363 32.836 32.983 33.449 33.603 34.096 34.269 34.809 35.003
I TABLE 13.
AVERAGE VALUES AND STANDARD DEVIATIONS OF TEMPERATURE MEASUREMENTS FOR 10350 PASSES OF TRAFFIC.
MEASUREMENT TYPE
DEPTH BELOW SURFACE (in)
Ambient Surface Midheight Interface Six Inch Twelve Inch
72 (above surface) 0 3 6 12 24
AVERAGE VALUE (OF)
STANDARD DEVIATION (OF)
87.4 112.8 103.5 97.6 92.9 92.0
4.4 13.5 10.8 7.8 3.8 4.4
I I TABLE 14.
STATIONS FOR REPRESENTATIVE TEST SECTIONS. REPRESENTATIVE PROFILES
1
SECTION NUMBER
mIX DESIGN
SECTION TYPE
STATION
FINAL PASS LEVEL
1
MARSHALL
4-INCH FLEXIBLE ASPHALTIC CONCRETE
20-36 38-64
2589 3049
l
2
MARSHALL
6-INCH FLEXIBLE ASPHALTIC CONCRETE
78-84 86-140
3049 5137
3
3
MARSHALL
6-INCH COMPOSITE ASPHALTIC CONCRETE OVER 12-INCH PORTLAND CEMENT
156-190 192-304
5137 5817
6-INCH COMPOSITE ASPHALTIC CONCRETE OVER 12-INCH PORTLAND
326-334 336-348 350-454
5817 9715 10350
6-INCH FLEXIBLE ASPHALTIC CONCRETE
470-526 528-532
10350 9715
4-INCH FLEXIBLE
548-610
9715
4
GYRATORY
I I
CEMENT 5
GYRATORY
6
GYRATORY
ASPHALTIC CONCRETE
I 56
U I
I 3
TABLE 15.
DAMAGE RANK ORDERS AT VARIOUS PASS LEVELS. a) 2324 PASSES
DAMAGE
RUT DEPTH
MAXIMUM UPHEAVAL
RUT WIDTH
RUT AREA
UPHEAVAL AREA
least
6CG
4FG
6CG
6CG
4FG
7
6CM
6FG
6CM
6CM
1 6FG
II4FG
6FM
6FM
6FM
F
6FM
4FM
6FG
4FG
6FM
4FM
6CM
4FG
4FM
6CM
UPHEAVAL AREA
3
most
b) 4784 PASSES
DAMAGE
RUT DEPTH
MAXIMUM UPHEAVAL
RUT WIDTH
RUT AREA
least
6CG
6CG
6CM
f6CM
6FG
6FG
6FM
6CG
6FG
6CM 4FG
4FG 6FM
6CG ;6FG
6FG 6FM
4FG 6FM
6FM
6CM
1 4FG
4FG
6CM
RUT WIDTH
RUT AREA
UPHEAVAL AREA
l most
(6CG
c) 9715 PASSES
SRUT I
DAMAGE
DEPTH
MAXIMUM UPHEAVAL
least I
6CG 6FG(
6CG 6FG
j6CG
1'6FG
6CG 6FG
(6CG 16FG
4FG
I 4FG
i 4FG
4FG
14FG
Smost
Brackets indicate groups of data which are not significantly different.
S4 6
= 4" overlay = 6" overlay
F = flexible section C = composite section
3
I
I
57
M - Marshall mix design G - Gyratory mix design
I TABLE 16.
Load
TEST VARIABLES USED TO DESCRIBE ANTECEDENT CONDITIONS IN MANOVA.
Node Lateral Position, Mean Lateral Position, Lateral Position Standard Deviation, Lateral Position Variance, Quasiskew, Number of Passes, Log of Number of Passes Number of Passes in 10* F Temperature Intervals
Environment
I 3
Intervals
Location
Surface
70-80, 80-90, 90-100, 100-110, average 80-90, 90-100, 100-110, 110-120, 120-130, 130140, 140-150, average
Pavement Midheight
80-90, 90-100, 100-110,
Ambient
3
110-120, 120-130, average
Pavement Material and Geometry
Base Course/Pavement Interface
70-80, 80-90, 90-100, 100-110, 110-120, average
6" into Base Course
80-90, 90-100, average
12" into Base Course
70-80, 80-90, 90-100, 100-110, average
1
Mix Type, Design Type, Design Thickness, Asphalt Content, Asphalt Volume, Total Voids, Voids Filled, Theoretical Maximum Density, Bulk Density, Base Support FWD ISM, Pavement FWD ISM
I
I I I
S~I
I I l
TABLE 17.
R-SQUARE STATISTIC FOR EACH DAMAGE MEASURE - ALL VARIABLES.
R-Square
Damage Measure True Rut Depth (DT) Maximum Upheave (HUI) Second Maximum Upheave (HU2) Width (WR) Upheave Area (A2) True Rut Area (Al)
0.8349 0.6171 0.6236 0.4102 0.5979 0.8463
SRut
I TABLE 18.
R-SQUARE STATISTIC FOR EACH DAMAGE MEASURE - VARIABLE SUBSET 1.1
IDamage
I I
TABLE 19.
R-Square
Measure True Rut Depth (DT) Maximum Upheave Upheave (HU1) (HU2) Second Maximum
0.7460 .5513 0.5583
Rut Width (WR) Upheave Area (A2) True Rut Area (Al)
0.4001 0.5391 0.8295
R-SQUARE STATISTIC FOR EACH DAMAGE MEASURE - VARIABLE SUBSET 2.'
SDamage
R-Square
Measure True Rut Depth (DT)
0.7622
Maximum Upheave (HUI)
0.5517
Second Maximum Upheave (HU2) Rut Width (WR) Upheave Area (A2) True Rut Area (Al)
0.5576 0.3771 0.5350 0.8350
where the "p" values are probability significance levels associated with each damage measure model.
1All "p" values < 0.0001,
I I I
59
TABLE 20.
PEARSON PAIRWISE CORRELATION COEFFICIENTS FOR SIX DAMAGE MEASURES.
True Rut Depth (DT) Maximum Upheave 1* (HUI) Maximum Upheave 2* (HU2) Rut Width (WR) Upheave Area (AI) True Rut Area (A2)
DT
HUI
HU2
oR
A2
1.00
0.36 1.00
0.31 0.83 1.00
0.19 -0.16 -0.17 1.00
0.27 0.90 0.89 -0.18 1.00
SYMMETRIC
I I I
Al 0.78 -0.04 -0.11 0.45 -0.10 1.00
Two maximum upheave measurements are calculated: one to either side of the maximum rut depth.
60
I I I I I I I I I I I I
IL C"
*n
-44-
6-4 LUi
-
0*
iN
0
-
c4 LL, K
j
0
%C.
Ln Nl~
r. .
-n
0%
C0 m
CA
V)
Lf
0
N0
&4Dr, EJ
0 -4J
(no
M
-
W
m
n
%
n
-
a-
2~
-J
1-!
I
=-
-J
-
3ifW-Y3 -IV
-
-
-
-
~
-e
-
-~
Z
w1
I I I I
___________________________________
a
-
I -
I
-
Oil
0
I I
0
I
0 9Co
irE +
o q.
-o
_______
'a
0
in
'a CJ
o
in
o
I I I
SdD4
92
I
IS.
acJ
Ic2
M*? fel 912 U.
VZ
0
N011003
ad
ad
B *r
i ED
3cB
Ij
9%(1 1
II 9*'
1
02tS7
CZI f~
BIIMU B9
11 VN 91
QI~
Ic
I ;;Pil
Pfil
3
.
.
.
-...-.
0
..
:
S:
g
~g
/
-
i
I•'+
*
ZII Z*
I D2
Z
I
!IZ*Z
-
I
w b 9 i
I
9Z
II
I.. M
ZI
tX*I
~
D3
Mila•,
I
_
Li
/i
8
I94
I
I
-
g
* K3B I
le l
971 11
agS
- d
271 00 z 1.1 aj DIM13
5
111
17 IZ (A 010
071
161
UAI
MEDVA
UJ DIOA
mlb
I
-
lull
I..
g
g
B
B KTa BI
RIZ 1'
02
01W7
11
96
Z7
01
M
IIWO
LI~Lb
*
gad
CMAI
'a
1los
3cj OLD
-zz
917 Z11 91
0S87
iz 97
01 WZ
8Ama
I
NOVO
m
-U,1
UI
j %
'a
hIIet
V
ziZIeI
901 WTt
1
R7 T~II 9 z 1 I
90,21
OOIZU
UA~~~01n(L NU a
98
0)v
*~
.2
Go: 0
o4
~
4~3.00
0
0l-,o
o*
00
0
0 0
00-4
0 0
0
0
.
152
I I
I°
I
0
3
I
0
0
0 0
I
.o
I
o
0
0 00
4
z
0
000
I
0
*
I0o 84 0 d>0.g0 o o• q
.
.,....
4
,i,1.
0
4
0
0
.'4,
0 06)0
00
~
0
.~0
00~
-
+ *++ I
00O
+
4+ +
+
2
t > X.
I
+
I.-
0 0
(gLOM
I
I
6VUI)1ý8 Ln 3nW.
153
0
00 0
0
W
§ UI 'I
cnI
I I I I I I I I I II
U'I 0.
2
U
Ij 1
I 0
(ui) SSN>IOIHJ.
1NVtAYd
154
I I I I
I SECTION V m
CONCLUSIONS AND RECONNENDATIONS
This field study of rutting in asphaltic pavements provided a significant amount of data from which several important conclusions can be drawn. The quality control aspects during construction provided test sections with "consistent material properties throughout their length. Instrumentation mefforts
weesuccessful
in poingdata
onmny
important variables associated with pavement performance.
m
3
DATA ACQUISITION EFFORTS
RELATIONSHIP BETWEEN LOADING AND DAMAGE
All of the damage measures for each section were correlated to the applied loading (traffic level and loadcart position). A nonlinear relationship between damage and passes was found for all cases, with the rate of damage typically decreasing with increasing traffic. The loadcart velocity was bimodal, reflecting forward and backward directions. Profile damage measures revealed an increase in the accumulated damage at the ends of the traffic lane, reflecting the acceleration/deceleration portion of the passes. Lateral position data was reduced into histograms, which show a slightly skewed distribution. The greater the asymmetry in loading, the greater was the observed damage. The mode of the position histograms was the best indicator of the peak rut lateral location. The average of the loadcell data over the length of the pass averaged around the static weight, however oscillations as great as eighty percent above and below the static wheel load were detected. Further analyses to detect possible patterns of load fluctuations and the relationship to rut variation are planned.
I I
I
the
A low-cost, high-output data acquisition system was developed for loadcart application. This system provided longitudinal and lateral position, velocity, and dynamic load variation for each pass of the loadcart. The data was loaded in a project database providing a means of comparing loading and damage data. In addition, an extensive set of FWD data was acquired, monitoring the pavement stiffness throughout the study. B.
I I
lof
The data reduction
processes used current technology to provide a complete database from which information can be drawn by future researchers. The analysis of the data incorporated both traditional and statistical techniques, providing a multifaceted review of results. A.
l
f no
155
I The temperature did not vary as much as expected during trafficking, therefore conclusions regarding the rate of rutting versus temperature could not be made. However, the pavement midheight temperature averaged 103.5 0F and rutting occurred rapidly. Today's high tire pressure aircraft will continue to cause significant rutting problems in similar environmental conditions. Further study is required to fully quantify the relative performance of gyratory designed pavements under various temperature conditions. C.
I
EFFECT OF MIX DESIGN ON PERFORMANCE
The analysis presented in Section IV showed that the applied loading across all of the sections did not vary significantly. However, the damage did vary significantly between sections, with the most obvious factors being the asphalt concrete mix design and base support. The gyratory sections formed broad ruts, characteristic of densification. Marshall sections formed multiple ruts, characteristic of plastic flow, with one rut eventually becoming primary. Comparisons between the composite sections, which reveal the behavior directly attributed to the mix design procedure, show that the Marshall mix sections experienced three times the rutting of the gyratory mix. The Marshall sections failed at or below 5800 passes of the loadcart. The gyratory sections had withstood between two and four times the passes of the Marshall sections when the trafficking was stopped because of dropping ambient temperatures. Comparisons between the mix types for each of the similar sections reveal several important findings. The gyratory sections consistently had less rut depth than the Marshall counterpart although their rut areas were similar in magnitude. This indicates the broad width of the gyratory damage zone relative to the Marshall sections. The rate of damage increased with traffic in the Marshall sections and decreased with traffic in the gyratory sections. The upheaval area in the Marshall sections was three times higher than the upheaval area in the gyratory sections, despite less traffic in the Marshall sections. The gyratory sections had experienced at least twice the traffic. This study has shown that pavements can be designed using gyratory methods and constructed with conventional equipment. It has also demonstrated that these surfaces have superior performance to Marshall designs. Common arguments preventing the widespread use of the gyratory mix design procedure include the expense of equipping a laboratory with a gyratory compaction device and the problem of inadequate compaction. The gyratory mixture used in this study called for twenty percent less asphalt cement than the Marshall mix. This decrease in binder represented a three percent savings on the in-place cost of the asphaltic concrete. Similar savings on a paving project would offset the initial cost of the gyratory compactor.
1
I
I I
I I
I 156
I
I Although the behavior of the gyratory mixtures under the applied traffic shown in this study should alleviate concerns of inadequate compaction, heavier equipment will be required to meet current DOD criteria for The gyratory mixture was not compacted to its design density compaction. during construction, probably because of the equipment used. For the Air Force, the compaction problem may be academic. The substantial increase in life span of pavements which are currently rutting after only eight months of operation will compensate for relatively small reductions in life span due to excess permeability. I
D.
SIGNIFICANCE OF RUTTING IN BASE LAYERS
A comparison of representative profiles between the 6-inch flexible and 6-inch composite sections shows that sixty percent of the rutting experienced in the flexible sections can be traced to the granular layers. Visual observations of the base course conditions were made in trenches at the end of trafficking, and base course rutting was identified at all locations. Statistical analysis also showed the importance of the base support type for its contribution in resisting rutting. An additional investigation is planned to determine the cause of, and methods to reduce granular layer rutting.
HE.
3
l
I
CONCLUSIONS AND RECOMMENDATIONS This experiment has shown that rutting of asphalt mixtures can be prevented if the mix is designed with laboratory compaction equal to that of the traffic. Failure of the Marshall sections was attributed to inadequate laboratory compaction during mix design. This was exacerbated somewhat by manufactured fines which prevented construction of mixtures at optimum binder content. In this field study, gyratory designed airfield pavements performed about three times better than Marshall designed pavements in terms of surface rutting. Even at compaction to ninety-two percent of the theoretical maximum density (one percent below DOD criterion), the gyratory sections outperformed the Marshall sections. Heavier field compaction equipment for gyratory mixes Lower asphalt costs in the leaner may further minimize surface rutting. gyratory mixes may offset the additional cost of purchasing gyratory design equipment. Independent of the mix design method, granular layers were found to contribute about sixty percent of the rutting in the 6-inch flexible sections. Improvements in base layer performance are needed to improve the overall performance of flexible airfield pavements under fighter aircraft loading. To account for failure mechanisms other than rutting, cooler weather and long term performance of gyratory designed pavements should also be evaluated. Additional performance data (in-service test sections) will be needed to support Air Force-wide application of gyratory design methods.
157
I REFERENCES 1. Regan, G. L., A Laboratory Study of Asphalt Concrete Mix Desian for High-Contact Pressure Aircraft Traffic, ESL-TR-8S-66, U.S. Army Waterways Experiment Station, for the Air Force Engineering and Services Center, Tyndall AFB, Florida, June 1987.
3
2. Timian, D. A., Desion and Construction of Pavement Test Sections for the Study of Hiah Pressure Tire Effects, Applied Research Associates, Inc., for the Air Force Engineering and Services Center, Tyndall AFB, Florida, May 1989.
m
3. ASTM, Annual Book of ASTM Standards. Concrete and Agareaates. Volume 04.0, American Society for Testing and Materials, Philadelphia, Pennsylvania, 1988.
I
4. ASTM, Annual Book of ASTM Standards. Roofing. Wateroroofina. Bituminous Materials, Volume 04,04, American Society for Testing Materials, Philadelphia, Pennsylvania, 1988.
and and
5. ASTM, Annual Book of ASTM Standards. Road and Paving Materials. Traveled Surface Characteristics. Volume 04.03, American Society for Testing and Materials, Philadelphia, Pennsylvania, 1988. 6. Military Standard 620-A, Military Standard Test Methods for Bituminous Pavina Materials, U.S. Army Engineer Research Laboratories, Mobility Command, Fort Belvoir, Virginia, 1966. 7. Anderson, M. and Timian, D. A., "Backcalculated Moduli for High Pressure Tire Effects Sections," Applied Research Associates, Inc., for the Air Force Engineering and Services Center, Tyndall AFB, Florida, September 1989.
I
8. Benjamin, J. R. and Cornell, C. A., Probababilitv. Statistics. and Decision for Civil Engineers, McGraw-Hill Book Company, New York, 1970, p. 8.
158
I I I I. I I
I I mIPENI
I m
FALLING W~APEIGHT DELCOEE A
T
I
I
I I I I
I I15
I
FLIGWIH ELCO4TRDT
I I APPENDIX A. FALLING WEIGHT DEFLECTONETER TEST RESULTS SECTION I Thickness Base - 12 inches Date - 26 June 1989 STATION feet
Thickness Asphalt Concrete - 4 inches Surface Temperature - 89.6 OF
TIME hr
LOAD lbs
Di
D2
1412 1412
9075 9027
13.15 12.56
5.51 5.39
0
1412
11602
16.22
7.09
0 0
1412 1412
18563 26017
26.34 35.51
11.34 15.87
12 12 12 12 12
1413 1413 1413 1413 1413
8948 8932 11523 18484 25970
13.94 13.35 16.93 27.17 37.32
5.75 5.75 7.32 11.73 16.38
23
1414
8948
13.11
5.75
23
1414
8916
12.52
23 23 23
1414 1414 1414
11443 18357 25668
15.87 25.39 34.80
7.24 11.54 15.91
35 35 35 35 35
1415 1415 1415 1415 1415
8821 8821 11332 18293 25620
13.27 12.68 15.79 25.31 34.33
47 47 47 47
1416 1416 1416 1416
8900 8900 11396 18389
58 58 58 58
1417 1417 1417 1417
8853 8821 11443 18404
0 0
47
58
1416
1417
25684
25795
DEFLECTIONS (mils) D3 D4 D5 D6 2.28 2.28
1.77 1.73
1.46 1.46
4.17 2.87
2.24
6.61 9.29
4.49 6.26
3.46 4.80
3.23 3.23 4.13 6.57 9.29
2.28 2.28 2.83 4.41 6.14
3.07
5.59 3.07
3.27 3.23
D7
ISm kips/in
1.22 1.22
690.1 718.7
1.81
1.50
715.3
2.76 3.90
2.36 3.27
704.7 732.7
1.73 1.81 2.24 3.46 4.72
1.30 1.38 1.69 2.68 3.66
1.06 1.14 1.38 2.24 3.11
641.9 669.1 680.6 680.3 695.9
2.24
1.77
1.34
1.14
682.5
2.09
1.69 1.38
1.18
712.1
3.90 6.26 8.74
2.64 4.17 5.83
2.13 3.31 4.57
1.73 2.68 3.78
1.42 2.32 3.19
721.0 723.0 737.6
5.63 5.55 7.17 11.54 16.06
3.31 3.39 4.17 6.77 9.53
2.32 2.32 2.91 4.69 6.50
1.73 1.42 1.73 1.46 2.20 1.77 3.50 2.83 4.88 3.94
1.22 1.22 1.54 2.36 3.31
664.7 695.7 717.7 722.8 746.3
13.54 13.11 16.57 26.38
6.30 6.22 7.91 12.80
3.39 3.43 4.33 6.93
2.32 2.36 3.03 4.69
1.93 1.93 2.44 3.82
1.54 1.54 1.85 3.07
1.26 1.26 1.38 2.48
657.3 678.9 687.7 697.1
11.97 11.61 15.04 24.25
5.91 5.83 7.56 12.17
3.54 3.50 4.49 7.13
2.52 2.44 3.11 4.88
1.85 1.81 2.32 3.66
1.50 1.50 1.85 2.83
1.30 1.22 1.50 2.40
739.6I 759.8 760.8 758.9I
36.10
33.54
17.72
16.97
9.61
9.96
6.50
6.77
5.16 4.17
5.08
4.02
3.46
3.35
m
I
II
I
711.5
769.1
I 160
I
I
APPENDIX A. FALLING WEIGHT DEFLECTOHETER TEST RESULTS
I
SECTION 2 Thickness Base - 12 inches
Thickness Asphalt Concrete - 6 inches
Date - 26 June 1989
Surface Temperature - 89.6 °F
STATION feet
TIME hr
LOAD lbs
D
82 82 82 82 82
1418 1418 1418 1418 1418
8789 8868 11443 18404 25858
10.91 10.67 14.17 23.90 32.99
5.91 3.50 5.91 3.54 7.64 4.57 12.28 7.28 17.09 10.16
2.52 2.60 3.19 5.00 6.97
1.77 1.93 2.36 3.74 5.16
93
1420
8868
11.22
6.57 3.31
2.17
93
1420
8837
10.79
6.38
2.05
93
1420
11443
14.21
8.23 4.33
93 93
1420 1420
18246 26192
22.83 31.89
13.07 18.11
104 104 104 104 104
1421 1421 1421 1421 1421
8868 8821 11459 18420 26176
12.13 11.61 15.20 24.29 34.21
116 116 116 116 116
1422 1422 1422 1422 1422
8932 8868 11507 18373 26002
128 128 128 128 128
1423 1423 1423 1423 1423
139 139 139 139 139
1424 1424 1424 1424 1424
DEFLECTIONS (mils) D3 D4 05 D6
D7
iSm kips/in
1.42 1.46 1.73 2.83 4.02
1.14 1.14 1.42 2.36 3.35
805.6 831.1 807.6 770.0 783.8
1.61
1.38
1.14
790.4
1.57
1.30
1.10
819.0
2.72 2.13
1.69
1.38
805.3
6.85 9.61
4.29 6.02
3.31 4.61
2.68 2.20 3.74 3.11
799.2 821.3
5.79 5.67 7.52 12.01 16.69
3.23 3.23 4.21 6.69 9.37
2.20 2.24 2.87 4.53 6.26
1.65 1.69 2.24 3.39 4.65
1.38 1.42 1.73 2.72 3.70
1.06 1.10 1.42 2.28 3.11
731.1 759.8 753.9 758.3 765.2
13.82 12.95 16.50 26.73 36.61
5.08 4.96 6.65 10.71 15.08
3.03 2.99 3.98 6.34 8.86
2.09 2.05 2.72 4.29 5.94
1.54 1.57 2.05 3.31 4.53
1.26 1.30 1.65 2.56 3.54
1.06 1.06 1.30 2.13 2.87
646.3 684.8 697.4 687.4 710.2
8821 8837 11443 18373 25970
16.14 15.39 19.92 32.64 44.37
5.55 5.43 7.28 11.97 16.69
3.07 3.07 4.13 6.69 9.37
2.01 1.97 2.68 4.25 5.94
1.50 1.77 2.05 3.11 4.33
1.22 1.18 1.54 2.52 3.35
0.94 0.91 1.26 2.05 2.72
546.5 574.2 574.4 562.9 585.3
8932 8884 11475 18436 25827
14.06 13.15 16.77 26.77 36.26
5.24 5.04 6.69 10.83 14.92
2.60 2.48 3.27 5.31 7.40
1.73 1.65 2.09 3.39 4.72
1.34 1.30 1.69 2.64 3.66
1.18 1.14 1.38 2.13 2.87
0.98 0.91 1.14 1.93 2.48
635.3 675.6 684.3 688.7 712.3
D02
3.27
I I
I
161
APPENDIX A. FALLING WEIGHT DEFLECTONETER TEST RESULTS
U
SECTION 3 Thickness Concrete - 12 inches Date - 28 June 1989 STATION feet
TINE hr
LOAD
lbs
Di
D2
161 161
1320 1320
9107 9091
6.89 6.69
161
1320
11129
161 161
1320 1320
18802 26208
173 173 173
1322 1322 1322
9027 8980 11602
173
1322
173
1322
185 185 185 185 185
1323 1323 1323 1323 1323
191 197 197 197 197
Thickness Asphalt Concrete - 6 inches Surface Temperature - 89.6 OF DEFLECTIONS (muls) D3 D4 D5 D6
D7
iSm kips/in
1.57 1.65
1.61 1.69
1.02 1.10
1321.8 1358.9
8.21
2.01
12.76 17.01
3.23 4.57
5.75 5.39 6.85
18675 25938
1.54 1.61
1.42 1.46
2.05
1.89
1.69 1.46
1.22
1418.3
3.31 4.72
3.07 2.76 4.29 3.82
2.36 3.31
2.05 2.80
1473.5 1540.7
1.65 1.69 2.01
1.85 1.81 2.20
1.65 1.65 1.93
1.30 1.30 1.50
1.14 1.14 1.22
1569.9 1666.0 1693.7
10.94
3.43
3.62 3.23
2.83 2.44 2.13
1701.0
14.80
4.84
5.04
3.98
3.43
2.87
1752.6
8996 9043 11682 18818 26272
4.88 4.61 5.91 9.29 12.56
1.97 1.93 1.93 1.89 2.44 2.36 3.86 3.78 5.39 5.31
1.73 1.54 1.69 1.46 2.17 1.93 3.39 2.95 4.76 4.13
1.38 1.30 1.65 2.52 3.58
1.18 1.10 1.38 2.20 3.03
1843.4 1961.6 1976.6 2025.6 2091.7
1325 1325 1325 1325 1325
9059 9021 11697 18802 26733
4.76 4.57 5.87 9.21 12.52
1.89 1.89 2.52 3.98 5.55
1.89 1.89 2.44 3.90 5.39
1.69 1.69 2.17 3.46 4.84
1.30 1.26 1.61 2.52 3.54
1.06 1.06 1.34 2.17 2.95
1903.2 1975.3 1992.7 2041.5 2135.2
209 209 209 209 209
1326 1326 1326 1326 1326
9012 8964 11650 18706 26367
5.08 4.80 5.98 9.45 12.91
1.81 1.81 2.36 3.86 5.35
1.81 1.61 1.81 1.61 2.28 2.05 3.82 3.39 5.28 4.69
1.42 1.26 1.38 1.26 1.81 1.54 2.95 2.56 4.06 3.46
1.06 1.06 1.26 2.20 2.91
1774.0 1867.5 1948.2 1979.5 2042.4
221 221 221 221 221
1328 1328 1328 1328 1328
9091 9027 11729 18865 26892
4.69 4.45 5.75 9.21 12.56
1.85 1.77 2.28 3.78 5.16
1.81 1.65 1.73 1.54 2.24 2.05 3.74 3.31 5.08 4.53
1.42 1.34 1.77 2.91 3.98
1.26 1.18 1.57 2.48 3.39
1.06 0.98 1.30 2.13 2.83
1938.4 2028.5 2039.8 2048.3 2141.1
233 233
1329 1329
9043 9012
4.17 4.06
2.05 1.97
1.93 1.89
1.65 1.54
1.38 1.14 1.30 1.10
233 233 233
1329 1329 1329
11697 18818 26923
5.31 8.66 11.89
2.52 3.98 5.47
2.44 3.86 5.28 162
4.49
1.73 1.69
2.17 3.46 4.72
1.46 1.46 1.69
1.46 1.46 1.93 2.95 4.17
1.97 3.03 4.13
1.18 1.26
1.65 2.60 3.54
1.42 2.24 2.95
I
i
I i
2168.6 2219.7I 2202.8 2173.0 2264.3
I
APPENDIX A. FALLING WEIGHT DEFLECTOMETER TEST RESULTS SECTION 3
m
Thickness Concrete - 12 inches Date - 28 June 1989
1 I
Thickness Asphalt Concrete - 6 inches Surface Temperature - 89.6 *F
STATION feet
TIME hr
LOAD lbs
Di
D
245 245 245
1330 1330 1330
9059 9059 11666
4.84 4.53 5.79
2.09 1.97 2.56
245
1330
18881
9.25
4.09 4.02
245
1330
26955
12.64
5.71
5.55
257 257 257 257 257
1333 1333 1333 1333 1333
9457 9361 11904 18865 26812
4.96 4.76 6.18 9.53 12.95
1.69 1.77 2.28 3.62 5.08
269 269 269 269 269
1335 1335 1335 1335 1335
8948 8980 11634 18865 26955
4.80 4.61 5.75 8.98 12.20
281 281 281
1336 1336 1336
9091 9027 11570
293 293 293 293 293
1337 1337 1337 1337 1337
281 281
1336 1336
DEFLECTIONS (mils) D3 D4 DS 06
D7
ism kips/in
1.14 1.02 1.38
1871.7 1999.8 2014.9
3.54 3.07
2.64 2.24
2041.2
4.92
4.25
3.62
3.03
2132.5
1.73 1.77 2.28 3.58 5.04
1.57 1.57 2.05 3.19 4.53
1.34 1.34 1.81 2.80 3.98
1.26 1.26 1.61 2.36 3.39
1.10 1.10 1.34 2.05 2.87
1906.7 1966.6 1926.2 1979.5 2070.4
1.69 1.69 2.20 3.58 4.96
1.81 1.81 2.32 3.74 5.08
1.61 1.61 2.09 3.35 4.53
1.42 1.42 1.81 2.91 3.94
1.22 1.26 1.57 2.48 3.39
1.06 1.02 1.34 2.13 2.83
1864.2 1947.9 2023.3 2100.8 2209.4
18611 25858
4.02 4.06 5.20
8.58 11.85
1.89 1.97 2.52
1.85 1.97 2.32
1.61 1.73 2.05
1.38 1.46 1.81
1.22 1.34 1.50
0.98 1.14 1.22
2261.4 2223.4 2225.0
8964 8932 11618 18722 26685
5.20 5.04 6.26 9.72 13.31
1.97 2.01 2.52 3.98 5.47
1.93 1.93 2.36 3.82 5.39
1.69 1.69 2.17 3.35 4.76
1.54 1.54 1.93 2.99 4.09
1.30 1.30 1.61 2.48 3.50
1.10 1.06 1.34 2.05 2.91
1723.8 1772.2 1855.9 1926.1 2004.9
2.05 1.93 2.52
4.06 5.71 3.90 5.47 3.43 4.84
I I I I
I
1.85 1.73 2.24
163
1.65 1.50 1.97
1.42 1.26 1.69
2.95 2.48 2.13 4.13 3.50 2.87
2169.1 2182.1
I I APPENDIX A. FALLING WEIGHT DEFLECTOMETER TEST RESULTS SECTION 4 Thickness Concrete - 12 inches Date - 28 June 1989
Thickness Asphalt Concrete - 6 inches Surface Temperature - 89.6 °F
STATION feet
TIME hr
LOAD
lbs
Di
D2
DEFLECTIONS (mils) D3 D4 D5 D6
305 305 305 305 305
1341 1341 1341 1341 1341
8964 8948 11586 18675 26478
9.76 9.49 11.22 16.38 21.73
1.65 1.69 2.17 3.50 5.00
1.81 1.85 2.28 3.62 5.16
1.65 1.65 2.05 3.27 4.61
1.42 1.42 1.81 2.76 4.02
317 317 317 317 317
1342 1342 1342 1342 1342
8964 8900 11539 18659 26303
8.54 8.35 10.12 15.83 21.50
1.34 1.38 1.77 2.99 4.25
1.71 1.77 2.09 3.54 4.92
1.57 1.57 1.89 3.19 4.41
329 329 329 329 329
1343 1343 1343 1343 1343
9027 8980 11666 18865 27050
4.17 4.02 5.20 8.27 11.34
1.85 1.85 2.44 3.86 5.43
1.77 1.77 2.28 3.70 5.16
341 341 341 341 341
1344 1344 1344 1344 1344
8948 8884 11602 18675 26399
4.61 4.25 5.51 8.70 11.89
1.81 1.73 2.40 3.66 5.08
353 353 353 353 353
1345 1345 1345 1345 1345
8996 8916 11666 18802 26621
4.29 4.21 5.31 8.39 11.50
365 365 365 365 365
1346 1346 1346 1346 1346
9027 8884 11682 18802 26764
4.09 3.94 4.96 7.87 10.94
D7
iSM kips/in
1.26 1.30 1.54 2.36 3.39
1.10 1.10 1.30 1.97 2.76
918.4 942.9 1032.6 1140.1 1218.5
1.38 1.38 1.65 2.76 3.82
1.22 1.22 1.42 2.32 3.27
1.06 1.06 1.26 2.01 2.72
1049.6 1065.9 1140.2 1178.7 1223.4
1.57 1.57 2.05 3.27 4.57
1.34 1.34 1.81 2.80 3.94
1.22 1.22 1.54 2.40 3.31
1.02 1.02 1.30 2.05 2.76
2164.7 2233.8 2243.5 2281.1 2385.4
1.77 1.57 2.13 3.54 4.80
1.57 1.46 1.97 3.15 4.29
1.38 1.30 1.77 2.76 3.82
1.26 1.10 1.50 2.40 3.27
1.10 0.94 1.30 2.09 2.76
1941.0 2090.4 2105.6 2146.6 2220.3
1.81 1.89 2.28 3.78 5.24
1.73 1.81 2.20 3.58 4.96
1.57 1.65 1.97 3.23 4.45
1.38 1.46 1.73 2.83 3.94
1.22 1.30 1.50 2.44 3.39
1.06 1.10 1.26 2.13 2.87
2097.0 2117.8 2197.0 2241.0 2314.9
1.69 1.69 2.01 3.23 4.61
1.65 1.65 1.97 3.15 4.49
1.50 1.50 1.77 2.83 4.06
1.34 1.18 1.34 1.18 1.57 1.38 2.48 2.17 3.58 3.11
1.02 1.02 1.18 1.89 2.68
2207.1 2254.8 2355.2 2389.1 2446.4
164
m
I
I
I
I I APPENDIX A. FALLING WEIGHT DEFLECTOtETER TEST RESULTS
I
SECTION 4 Thickness Concrete
-
12 inches
Date - 28 June 1989
i
Thickness Asphalt Concrete - 6 inches Surface Temperature - 89.6 OF
STATION feet 377 377 377 377 377
TIME hr 1347 1347 1347 1347 1347
LOAD lbs 8980 8900 11650 18865 26844
D 4.49 4.37 5.43 8.58 11.85
D 1.27 1.85 2.20 3.54 4.92
DEFLECTIONS D D 11 112 1.69 1.50 2.05 1.81 3.27 2.91 4.69 4.21
(mils) D D 130 1.14 1.57 1.18 1.81 1.42 2.76 2.24 3.74 3.27
ISm D. kips/in 1.02 2000.0 1.10 2036.6 1.26 2145.5 1.97 2198.7 2.80 2265.3
389 389 389 389 389
1349 1349 1349 1349 1349
9186 9059 11507 18659 25922
4.17 3.98 4.96 8.11 11.22
1.81 1.77 2.17 3.58 4.96
1.77 1.69 2.17 3.50 4.84
1.61 1.54 1.89 3.11 4.29
1.42 1.34 1.69 2.76 3.82
1.26 1.22 1.54 2.36 3.27
1.10 1.02 1.22 2.01 2.72
2202.9 2276.1 2320.0 2300.7 2310.3
401 401 401 401 401 413 413 413 413 413
1350 1350 1350 1350 1350 1351 1351 1351 1351 1351
8948 8932 11634 18722 26415 8948 8948 11618 18786 26590
3.94 3.78 4.88 7.68 10.59 4.21 4.06 5.31 8.43 11.57
1.93 1.81 2.36 3.70 5.24 2.13 2.05 2.72 4.41 6.02
1.81 1.69 2.17 3.50 4.92 2.01 1.93 2.60 4.13 5.67
1.65 1.54 1.93 3.15 4.37 1.77 1.69 2.32 3.70 5.08
1.54 1.34 1.73 2.76 3.86 1.54 1.42 2.05 3.23 4.37
1.30 1.18 1.50 2.36 3.35 1.34 1.26 1.73 2.76 3.70
1.06 1.02 1.22 2.05 2.83 1.14 1.06 1.42 2.32 3.15
2271.1 2363.0 2423.8 2437.8 2494.3 2125.4 2203.9 2187.9 2228.5 2298.2
425 425 425 425 425
1353 1353 1353 1353 1353
8964 8900 11650 18786 26812
3.98 3.94 4.92 7.88 10.67
1.85 1.93 2.44 3.82 5.39
1.77 1.85 2.28 3.66 5.08
1.57 1.65 2.05 3.27 4.53
1.38 1.42 1.81 2.83 3.98
1.22 1.30 1.57 2.44 3.43
1.02 1.10 1.30 2.05 2.87
2252.3 2258.9 2367.9 2408.5 2512.8
443 443 443 443 443
1354 1354 1354 1354 1354
8980 8996 11650 18881 26892
4.33 4.21 5.43 8.78 12.05
2.05 2.05 2.60 4.25 5.98
1.89 1.89 2.44 3.90 5.51
1.69 1.69 2.09 3.39 4.76
1.46 1.46 1.81 2.91 4.09
1.30 1.30 1.57 2.48 3.46
1.10 1.10 1.34 2.13 2.91
2073.9 2136.8 2145.5 2150.5 2231.7
I
I I
I
165
I I APPENDIX A. FALLING WEIGHT DEFLECTOHETER TEST RESULTS
I
SECTION5 Thickness Base - 12 inches Date - 26 June 1989
Thickness Asphalt Concrete - 6 inches Surface Temperature - 89.6 OF
STATION feet
TIME hr
LOAD
lbs
Di
D2
D3
D4
D5
D6
D7
ISm kips/in
483 483 483 483 483
1426 1426 1426 1426 1426
8996 8900 11539 18484 26447
10.31 9.80 12.76 20.63 28.27
5.20 5.04 6.65 10.87 15.08
3.07 2.99 3.94 6.38 8.90
2.28 220 2.76 4.41 6.14
1.69 1.69 2.13 3.39 4.57
1.30 1.30 1.69 2.68 3.62
1.10 1.02 1.38 2.24 2.95
872.6 908.2 904.3 896.0 935.5
495 495 495 495 495
1427 1427 1427 1427 1427
8868 8853 11475 18468 25874
10.55 10.24 13.31 21.61 29.96
5.67 2.80 5.63 2.87 7.36 3.78 11.77 6.02 16.42 8.58
1.81 1.89 2.48 3.94 5.59
1.42 1.50 1.93 2.95 4.25
1.18 1.26 1.57 2.40 3.46
0.98 1.02 1.30 2.01 2.83
840.6 864.6 862.1 854.6 863.6
506 506 506
1428 1428 1428
8900 8853 11523
10.79 10.39 13.50
5.31 2.91 5.31 2.91 6.77 3.82
2.01 1.57 2.01 1.65 2.64 2.05
1.22 1.26 1.61
1.02 1.06 1.34
824.8 852.1 853.6I
506 506
DEFLECTIONS (mils)
1428 1428
18452 26367
22.17 30.75
11.18 15.91
6.38 9.13
4.21 6.10
3.23 4.53
2.56 3.66
2.17 3.11
518 518 518 518 518
1429 1429 1429 1429 1429
8884 8884 11443 18373 26415
10.87 10.47 13.39 21.54 29.69
5.16 5.16 6.69 10.91 15.24
2.91 2.99 3.90 6.34 8.94
2.05 2.17 2.80 4.41 6.10
1.61 1.69 2.17 3.39 4.61
1.30 1.34 1.65 2.64 3.58
1.10 1.10 1.34 2.20 2.95
817.3 848.5 854.6 853.0 889.7
530 530 530 530 530
1430 1430 1430 1430 1430
8916 8884 11491 18484 26303
10.55 10.08 13.07 21.54 30.31
5.08 4.88 6.50 10.71 15.20
2.99 2.87 3.78 6.18 8.78
2.09 2.01 2.64 4.25 5.98
1.57 1.34 1.30 1.30 2.05 1.57 3.15 2.60 4.53 3.62
1.18 1.06 1.30 2.13 2.95
845.1 881.3 879.2 858.1 867.8
I I
i
832.3 857.5
1 I 166
I
I
I I APPENDIX A. FALLING WEIGHT DEFLECTONETER TEST RESULTS SECTION 6 Thickness Base - 12 inches
Thickness Asphalt Concrete - 4 inches
Date - 26 June 1989 STATION feet
TIME hr
LOAD lbs
Di
D2
553 553 553
1431 1431 1431 1431 1431
8996 8853 11491 18452 26415
10.91 10.55 13.94 23.23 32.17
5.55 5.67 7.32 11.77 16.38
3.23 3.15 4.21 6.69 9.41
2.32 2.36 3.03 4.69 0.42
1.77 1.81 2.36 3.62 5.08
1.38 1.38 1.81 2.80 3.78
1.14 1.10 1.46 2.24 2.95
824.6 839.1 824.3 794.3 821.1
564
1433
8932
12.40
5.98
3.11
2.09
1.54
1.42
1.18
720.3
1433
8868
11.93
3.11
2.09 1.54 1.42 1.18
743.3
564 564 564
1433 1433 1433
11507 18500 26351
15.63 24.84 33.62
5.8 7.68 12.40 16.97
4.09 6.46 9.09
2.72 4.25 5.91
2.09 3.31 4.57
1.77 2.64 3.66
1.38 2.24 3.07
736.2 744.8 783.8
576
1433
8932
13.27
6.26
3.62
2.44
1.73
1.42
1.10
673.1
576
1433
8868
12.12
6.14 3.58
2.44
1.73
1.42
1.06
697.2
576 576 576
1433 1433 1433
11507 18420 26256
16.54 26.93 37.09
8.11 4.65 13.31 7.44 18.70 10.47
3.11 4.92 6.81
2.28 3.54 4.88
1.73 2.83 3.86
1.34 2.24 3.11
695.7 684.0 707.9
588 588 588 588 588
1435 1435 1435 1435 1435
8932 8932 11507 18389 26272
12.83 12.36 16.22 26.22 36.34
6.26 6.02 7.99 12.76 17.80
3.39 3.31 4.41 6.89 9.61
2.24 2.28 3.03 4.61 6.30
1.73 1.69 2.28 3.46 4.69
1.34 1.34 1.81 2.76 3.78
1.10 1.10 1.46 2.32 3.11
696.2 722.7 709.4 701.3 722.9
600 600 600 600
1435 1435 1435 1435 1435
8868 8884 11507 18389 26303
11.97 11.46 15.16 25.08 35.00
5.91 3.46 5.59 3.46 7.52 4.53 12.32 7.28 17.24 10.00
2.48 2.40 3.15 5.04 6.81
1.93 1.77 2.28 3.74 5.08
1.38 1.34 1.77 2.87 3.94
1.10 1.02 1.38 2.32 3.15
740.9 775.2 759.0 733.2 751.5
611 611 611 611
1436 1436 1436 1436
8916 8916 11491 18404
11.65 11.30 14.88 24.72
5.83 5.75 7.56 12.28
2.48 2.56 3.27 5.00
1.85 1.93 2.44 3.70
1.38 1.46 1.85 2.87
1.06 1.14 1.46 2.32
765.3 789.0 772.2 744.5
611
1436
26319
34.49
17.24 10.08 6.81
5.00 3.86 3.15
763.1
3553 553 l564
S600
I
Surface Temperature - 89.6 °F DEFLECTIONS (mils) D 0 D4 D5 D6
3.54 3.58 4.57 7.28
I U
I
167
D7
ISM kips/in
I I APPENDIX A. FALLING WEIGHT DEFLECTOMETER TEST RESULTS
I
SECTION 5 Thickness Base - 12 inches
Thickness Asphalt Concrete - 6 inches
Date - 5 September 1989
I
Surface Temperature - 94.1 OF
STATION feet
TIME hr
LOAD lbs
Di
518 518 518 518 518
1226 1226 1226 1226 1226
9488 9425 12079 19088 26399
10.87 10.35 13.27 21.06 27.91
6.97 3.90 6.69 3.82 8.62 4.88 13.90 1.87 18.66 10.83
2.52 2.48 3.31 5.16 7.09
1.85 1.77 2.40 3.74 5.16
530 530 530 530 530
1225 1225 1225 1225 1225
9345 9282 11872 18913 25858
11.61 10.91 14.02 22.64 30.08
7.05 6.65 8.58 13.78 18.31
2.48 1.77 2.48 1.81 3.27 2.40 5.16 3.78 6.81 5.04
D2
DEFLECTIONS (mils) D3 D4 D 0 D6
3.86 3.78 4.88 7.80 10.59
D7
Ism kips/in
1.46 1.42 1.77 3.07 3.94
1.18 1.14 1.50 2.48 3.27
872.9 910.6 910.2 906.4 945.9
1.46 1.54 1.81 2.83 3.90
1.18 1.18 1.46 2.32 3.19
804.9 850.8 846.8 835.4 859.6
I
SECTION66 Thickness Base - 12 inches Thickness Asphalt Concrete - 4 inches Date - 5 September 1989 Surface Temperature - 94.1 OF STATION feet
TIME hr
LOAD lbs
Di
553 553 553 553 553
1219 1219 1219 1219 1219
9361 9313 12031 19120 26367
14.92 13.54 17.13 26.30 33.62
564 564
1214 1214
9647 9600
13.82 13.35
564
1214
12270
17.01
564 564
1214 1214
19358 26399
26.22 33.66
DEFLECTIONS (mils) D2
D3
7.52 3.70 7.17 3.70 9.17 4.76 14.41 7.48 19.17 10.16 7.56 7.40
3.86 3.82
9.45 4.92
04
D5
D6
D7
ISm kips/in
2.40 2.40 3.07 4.80 6.50
1.85 1.77 2.28 3.58 4.80
1.42 1.46 1.85 2.83 3.86
1.14 1.18 1.50 2.40 3.27
627.4 687.8 702.3 727.0 784.3
2.48 2.44
1.69 1.69
1.61 1.57
1.26 1.22
698.0 719.1
3.19 2.36
1.81
1.46
721.3
2.99 3.86
2.48 3.27
738.3 784.3
14.72 7.64 4.84 19.33 10.16 6.54
168
3.58 5.00
I
I I I I m
I
1 1
a
APPlEIOIX B
I S~CORE
DENSITY DATA
I I I I
I I I I I
I
169
TABLE 81.
PRETRAFFIC BULK DENSITIES - HEAVYWEIGHT LANE
CORE IDENTIFICATION
NIX
LIFT
SECTION
BULK DENSITY (pcf)
AO+25B AO+25D AO+25F AO+49B AO+49D AO+49F
N N N N N
T T T T T T
1 1 1 1 1 1
150.50 150.60 150.30 150.60 151.20 150.10
AO+95B AO+95D AO+95F AI+41B AI+41D A1+41F AO+95B AO+95D AO+95F A1+41B A1+41D AI+41F
M N M N M N N N N N N N
T T T T T T B B B B B B
2 2 2 2 2 2 2 2 2 2 2 2
150.30 151.60 151.50 155.20 154.70 154.40 146.10 146.10 150.80 148.40 148.80 150.50
A1+73A A1+73C A1+73E A1+73G A2+13B A2+13D A2+13F A2+15B A2+15D A2+1SF A2+57B A2+57D A2+57F A2+61B A2+61D A2+61F A2+79A A2+79C
N N N N N N N N N N N m N N N N N N
T T T T T T T T T T T T T T T T T T
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
153.50 153.30 152.90 152.50 151.40 152.10 152.10 151.00 152.90 152.50 151.80 152.00 151.80 151.50 151.80 151.00 144.40 149.00
N
170
I I I
3
I I
I I I
3
I
I STABLE
B1.
PRETRAFFIC BULK DENSITIES-
HEAVYWEIGHT LANE (Continued.) BULK
CORE IDENTIFICATION
LIFT
SECTION
T T T T T B B B B B B B B B B B B B
A2+61D A2+61F A2+79A A2+79C A2+79E A2+79G A3+05B A3+05D A3+05F
N N N N N N N m N N m m N N N m N m N m m m N m m m m M
B B B B B B B B B
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
151.50 151.60 150.00 151.60 151.60 147.80 148.90 150.80 149.50 149.00 150.40 150.90
A3+31C A3+31E A3+31G A3+69B A3+69D A3+6VF
G G G G G G G
T T T T T T T
4 4 4 4 4 4 4
150.80 151.10 151.70 150.90 148.50 149.50 149.60
A2+79E A2+79G A3+05B A3+OSD A3+05F AI+73A A1+73C A1+73E Al+73G A2+13B A2+13D A2+13F A2+15B A2+15D A2+15F A2+57B
A2+57D A2+57F A2+61B
SA3+31A
B
171
I
DENSITY (pcf)
miX
148.60 147.90 154.20 154.40 154.20 147.80 147.30 150.10 149.10 144.70 146.30 148.00 144.00 146.10 146.30 148.40
I I TABLE BI.
PRETRAFFIC BULK DENSITIES
CORE IDENTIFICATION
-
HEAVYWEIGHT LANE (Continued.)
BULK DENSITY (pcf)
mIX
LIFT
SECTION
A3+77A A3+77C A3+77E A3+77G A4+OlC A4+15B A4+lSD A4+15F A4+27A A4+27C A4+27E A4+27G A4+45B A4+45D A4+45F A4+49C A4+49G A3+31A A3+31C
G G G G G G G G G G G G G G G G G G G
T T T T T T T T T T T T T T T T T B B
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
144.70 147.10 147.50 148.20 148.90 147.10 150.20 149.50 144.90 146.40 147.80 146.90 149.70 150.80 151.80 148.00 151.50 144.80 144.60
A3+31E A3+31G A3+69B A3+69D A3+69F A3+77A A3+77C A3+77E A3+77G A4+01C A4+15B A4+15D A4+1SF A4+27C A4+27E A4+27G A4+45B A4+450 A4+45F
G G G G G G G G G G G G G G G G G G G
B B B B B B B B B B B B B B B B B B B
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
144.70 145.30 144.50 146.00 147.20 142.60 145.20 146.60 145.90 146.40 147.30 148.10 147.80 146.00 147.60 147.30 148.40 147.50 148.60
172
i
3
3 l
I 1
I
I I I
TABLE BI.
PRETRAFFIC BULK DENSITIES
HEAVYWEIGHT LANE (Continued)
NIX
LIFT
SECTION
BULK DENSITY (pcf)
A4+49C A4+49G
G G
B B
4 4
146.70 146.40
A4+97B A4+97D A4+97F
G G G G G G G G G G G G
T T T T T T B B B B B B
5 5 5 5 5 5 5 5 5 5 5 5
150.40 150.90 152.00
G G G G G G
T T T T T T
6 6 6 6 6 6
144.10 143.70 146.70 144.10 144.80 146.10
CORE IDENTIFICATION
A5+1OB
A5+1OD A5+1OF A4+97B A4+97D A4+97F A5+1OB A5+1OD A5+1OF A5+67B A5+67D
A5+67F A5+91B A5+91D A5+91F
I I I I I I I
-
'73
146.10
146.30 147.50 146.80 147.20 148.00 144.70 146.20 147.70
I I TABLE B2.
BULK DENSITIES AFTER APPROXIMATELY 4600 PASSES. HEAVYWEIGHT LANE
mIX
LIFT
SECTION
BULK DENSITY (pcf)
A3+05H A3+051 A3+05J A3+05H A3+05I A3+05J
N N N N N N
T T T B B B
3 3 3 3 3 3
155.6 155.2 154.1 153.1 152.3 153.0
A3+78H A3+78] A3+78J A3+78H A3+781 A3+78J
G G G G G G
T T B B B
4 4 4 4 4 4
155.7 155.03 154.5 150.1 148.5 148.8
A4+26H A4+261 A4+26J A4+26J A4+26H
G G G G G
T T T B B
4 4 4 4 4
155.6 156.2 155.7 150.8 150.8
A4+261
G
B
4
151.0
A4+96H A4+961 A4+96J A4+96J A4+96H A4+961
G G G G G G
T T T B B B
5 5 5 5 5 5
157.6 157.6 156.9 150.6 151.9 151.0
CORE IDENTIFICATION
1
174
I I
I
TABLE B3.
CORE IDENTIFICATION
BULK DENSITIES AFTER APPROX114ATELY 7500 PASSES. HEAVYWEIGHT LANE BULK DENSITY (Pcf)
mix
LIFT
A3+79K A3+79K A3+79L A3+79L A3+79M A3+79N
G G G G G G
T B T B T B
4 4 4 4 4 4
156.7 150.7 156.7 150.5 155.9 149.3
A4+27K A4+27K A4+27L A4+27L A4+27M A4+27N
G G G G G G
T B T B T 8
4 4 4 4 4 4
156.9 151.6 157.1 151.8 156.5 151.7
A4+95K A4+95L A4+95M A4+95K A4+95L A4+95N
G G G G G G
T T T 8 B B
5 5 5 5 5 5
158.1 158.0 157.6 151.6 152.2 152.3
SECTION
175
(The reverse of this page is blank.)
APPENDIX C
TEMPERATURE DATA AND HISTOGRAMS
177
APPENDIX C.
NUMBER OF PASSES RECORDED IN 10-DEGREE TEMPERATURE RANGES.
I
TEMPERATURE RANGES (OF) MEASUREMENT TYPE
PASS LEVEL
LOW - 70 HIGH - 80
80 90
90 100
100 110
110 120
120 130
130 140
140 150
44 86 194 394 422 678 1033 1669 1864 2186 2190 2574 2817 3082 3185 3353 3773 4355 5894 6455
26 26 26 26 26 202 428 437 501 630 863 1135 1730 1794 1924 2203 2114 3308 3308 3308
0 0 0 0 0 0 12 12 12 12 12 12 12 12 12 12 12 12 12 12
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 4 0 12 0 21 0 21 0 21 0 39 71 124 165 211 173 258 233 349 233 349 233 503 244 613 244 735 244 735 259 780 259 880 351 984 351 1374 351 1820
18 34 38 89 103 178 293 509 535 651 655 740 880 937 937 1045 1187 1313 1815 1869
22 22 66 138 152 244 422 668 740 789 805 977 1111 1211 1211 1251 1431 1689 2209 2291
10 28 69 120 120 345 498 622 734 799 900 975 1130 1201 1341 1580 2051 2741 2955 2955
16 16 26 52 52 74 142 142 142 206 322 467 755 755 848 848 946 946 946 946
0 0 0 0 0 0 0 0 0 0 0 25 25 25 25 25 25 25 25 25
AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT AMBIENT
70 112 220 420 448 882 1554 2324 2589 3049 3286 3942 4784 5137 5370 5817 6808 8080 9715 10350
0 0 0 0 0 0 77 199 199 199 199 199 199 220 220 220 220 374 461 535
SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE SURFACE JRFACE SURFACE SURFACE SURFACE SURFACE
70 112 220 420 448 882 1554 2324 2589 3049 3286 3942 4784 5137 5370 5817 6808 8080 9715 10350
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
I
I I
I 178
I I
APPENDIX C. NUMBER OF PASSES RECORDED IN 10-DEGREE TEMPERATURE RANGES.
I
TEMPERATURE RANGES (OF) MEASUREMENT TYPE
PASS LEVEL
OIDHEIGHT MIDHEIGHT
70 112
MIDHEIGHT
220
LOW - 70 HIGH - 80
80 90
90 100
100 110
110 120
0
0
24
0
48
20 22
19
0
35
0
420
0
90
64
0
59
0
134
160
MDHEIGHT
448
0
0
148
174
MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT I4IDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT MIDHEIGHT
1554 2324 2589 3049 3286 3942 4784 5137 5370 5817 6808 8080 9715 10350
882
0
8
270
INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE
70 112 220 420 448 882
NIDHEIGHT
UIHIH
0 100 435 0 260 658 0 285 743 0 309 991 0 309 991 0 340 1109 0 376 1342 0 397 1511 0 397 1511 0 446 1619 0 546 1900 0 786 1982 0 1204 2254 0 1443 2554 0 0 0 0 0 0
0 16 16 16 16 55
241
120 130
130 140
140 150
7
0
0
7
0
0
7
0
68
0
58
0
0
68
58
0
0
87
0
0
274
506 758 794 842 873 .1073 1191 1231 1268 1361 1404 1818 2685 2781
360 492 611 694 761 904 1286 1406 1602 1799 2366 2900 2969 2969
149 149 149 191 330 494 563 563 563 563 563 563 563 563
0 0 0
0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
42 60 144 224 238 386
21 29 53 128 142 358
7 7 7 52 52 81
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
INTERFACE
1554
0
116
803
IINTERFACE
505
126
0
0
0
2324
0
386
2589
1083
722
0
126
401 1231
0
824
0
126
0
0
0
154 293 447 597 597 597 597 694 773 773 773
0
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0
INTERFACE
INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE INTERFACE
I
I I S]179
I
3049 3286 3942 4784 5137 5370 5817 6808 8080 9715 10350
0 430 0 430 0 434 0 469 0 547 0 547 0 547 0 587 0 909 9 1479 9 1963
1522 921 1553 988 1793 1246 2130 1562 2282 1682 2319 1878 2526 2118 2909 2589 3380 2987 4364 3050 4515 3050
I APPENDIX C. NUMBER OF PASSES RECORDED IN 10-DEGREE TEMPERATURE RANGES.
I
TEMPERATURE RANGES (OF) MEASUREMENT TYPE
PASS LEVEL
6 INCH 6INCH 6 INCH 6 INCH 6 INCH
6 INCH
LOW - 70 HIGH - 80
80 90
90 100
100 110
110 120
120 130
130 140
140 150
70 112 220 420 448
0 0 0 0 0
0 34 48 48 48
70 78 172 372 400
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
882
0
108
772
0
0
0
0
0
6INCH
1554
6 INCH
2324
0
236 2081
0
0
0
0
0
6 INCH 6 INCH 6 INCH
2589 3049 3286
0 0 0
2346 2791 3028
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
6INCH
236 236 236
3942
0
236 3684
0
0
0
0
0
6 INCH 6 INCH
4784 5137
0 0
250 250
4508 4858
0 0
0 0
0 0
0 0
0 0
6 INCH 6INCH 6 INCH 6 INCH 6 INCH 6INCH
5370 5817 6808 8080 9715 10350
5091 5538 6529 7799 8099 80990
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0
0 0 0 0 0I
0
108
0 250 0 250 0 250 0 250 0 0 1576 2211
1442
0
0
0
0
0
12INCH
70
0
42
28
0
0
0
0
0
12 INCH 12 INCH
112 220
0 0
28 28
0 0
0 0
0 0
0 0
0 0
12 INCH
84 192
420
0
392
28
0
0
0
0
0
INCH INCH INCH INCH INCH INCH INCH INCH
448 882 1554 2324 2589 3049 3286 3942
0 0 1 1 1 1 1 1
420 651 785 1083 1083 1083 1083 1083
28 229 764 1233 1498 1943 2137 2715
0 0 0 0 0 0 43 121
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
12 INCH
4784
1 1112
3524
121
0
0
0
0
12 INCH 12 INCH
5137 5370
1 1133 1 1133
3853 4086
11 121
0 0
0 0
0 0
0 0
121
0
0
0
0
1133 1133 2459 3094
5524 6794 7094 7094
121 121 121 121
0 0
0 0
0 0
0 0
0
0
0
0
12 12 12 12 12 12 12 12
12 INCH 12 12 12 12
INCH INCH INCH INCH
5817 6808 8080 9715 10350
1 1133 4533 1 1 1 1
180I
,8o
I
I
I AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 70
I I
0
-i
Go _
50
60
70
80
100
110
120
130
140
150
140
150
TEMPERATURE (F)
I
I°,
i
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 112
i.oC
I 50
I181
60
70
80
90
100
TEMPERATURE (F)
110
120
130
I 3
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 220
,
0 co
50
60
70
so
90
100
110
120
130
140
150
TEMPERATURE (F)
I AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 420
I
L.w
zI
I
°z
I
0
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
182
I I
I AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 448
I
I
z
50
60
70
80
I
100
110
120
130
140
150
140
150
TEMPERATURE (F)
I
I
90
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 882
I
z 0
z
50
60
70
80
90
100
TEMPERATURE (F)
I
I
183
110
120
130
1 I
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 1554
8
_
°
I
z
-
o
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 2324
I.I
,.,
I
Co
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
184
I I
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 2589
Iz ,D
AE1S
I18
~,
I!0 --
L
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
6
0
8
0
¶0
1
2
3
4
5
I i
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 3286
I
C -
50
zi G0
70
80
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 3942
0-
c.! 50
60
70
80
100
90
TEMPERATURE (F) 186
110
120
130
U AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 4784
i
,x
S
C>C-)
50
60
70
so
90
100
110
120
130
TEMPERATURE (F)
IW
uL.J
Cr--
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 5137
I mJ
l I
I
TEMPERATURE (F) 187
140
150
I AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 5370
I
§
z
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I I
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 5817
zI -o
I
LJ•
0L
50
70
o
0!
100
90
TEMPERATURE (F) 18
18I
110
120
130
140
150
I0
I AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 6808
I
(
50
60
70
80
I
90
100
'
'
'
110
120
130
140
150
140
150
TEMPERATURE (F)
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 8080
I°
0(50
I
I
60
70
80
90
100
TEMPERATURE (F) 189
110
120
130
AMBIENT TEMPERATURE HISTOGRAM FOR PASS LEVEL 9715I
VI)
z
coI u0
m
0
6
0
8
0
10
CL,
LxJ
19
11
2
3
4
5
I
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 70
TEPRTR
50
60
70
80
90
(F)U
100
TEMPERATURE (F)
I9
110
120
130
140
150
I I
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 220
I
I
LU 50
60
70
80
90
100
110
120
1:30
140
1,50
TEMPERATURE (F)I l~cn
S~I
Co)
I
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 420
0
o
o 50
60
70
o'° 80
•I 90
100
110
120
130
140
150
I
TEMPERATURE (F)
192i
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 448
ROI 506I0 8 0 10 11 2 3 4 5 TEPRTR
I50
60
70
50
F
90 100 110 TEMPERATURE (F)
19
120
130
140
150
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 1554
IX
050
60
70
80
p
90
100
110
120
130
140
1501
140
150I
TEMPERATURE (F)
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 2324
L.J
50
60
70
80
90
100
110
TEMPERATURE (F)I
1941
120
130
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 2589
9--
UU
co
50
6
0
80
TEPRTRIF 19
9
0
10
10
13
4
5
I U
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 3286
I 50
60
70
so
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)I
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 3942
L.!
!W j0 mI
50
GO
70
80
90
100
TEMPERATURE (F) 196I
110
120
130
I I
I
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 4784
I
I 50
60
70
80
i
, i
SURFACE
~
10
90
110
120
130
140
150
140
150
TEMPERATURE (F')
TEMPERATURE HISTOGRAM FOR PASS LEVEL 5137
z
I
I
I
I
II
TEMPERATURE (F)
! 50
I
60
70
so
90
T100
197
110
120
130
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 5370I (N
WI
cr2
19
I SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 6808
II 0
I I *
0
50
60
70
80
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
I SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 8080
I I I I I I I I I
cdS L.J (-)
8
0
60
70
80
90100
199
(F)
110
120
130
SURFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 9715
50
60
7I0
9
0
1
2
TEMPERATURE (F)
200
3
4
5
i ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 70
i (-)
C.i
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 112
L.J(
j
•j
°cc
I
60
i50
70
so
90
100
TEMPERATURE (F) 0201
110
120
130
140
150
I I
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 220
0l
6 50
7
o
9
100
110
120
130
140
150
TEMPERATURE (F)I
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 420
I
0
20 5
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)I
20
I I
I I
ASP'-k'T MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 448
I If STEMPERATURE
I
,
I!(
(F)
0
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 882
0
/
S-) 00
50
60
70
80
90
100
TEMPERATURE (F)
I
110
120
130
140
150
I I
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 1554
Z -5
I
"L
oI
co
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 2324
I
oI zI
I
-o 0
(-L.J
I
zo
C..'!
I I
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 2589 8
C.J
I so
GO
70
80
t
90
100
1 10
120
130
140
150
TEMPERATURE (F)
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 3049
I I
I 50
I205
60
70
80
100 TEMPERATURE (F) 110
90
120
130
140
150
I 1
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 3286
|
L.
',
: I
o
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 3942
I
S
0!
Cd,
I
II
50
60
70
80
90
100
TEMPERATURE (F) 206
110
120
130
140
150
I
I
~ASPHALT MIOHEIGHT TEMPERATUJRE HISTOGRAM FOR PASS LEVEL 4784
IL CL)
I
zj L.J
3
TEPRTRIF
50
0
7
80
9
10
11
12
13
14
15
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
20
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 5370I
ClI
IO
w
mI 50
60
70
so
90
100
110
120
TEMPERATURE (F)
208
130
140
150
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 6808
20
I I
ASPHALT MIDHEIGHT TEMPERATURE HISTOGRAM FOR PASS LEVEL 9715
-
I
I
I
z-o L.
S~I z•
I C
o
I,
,
050 690 70 100
110
20
10
14
15
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 70
Ln
z
o
50
60
70
80
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 112
2 -
i
I
zII
50
60
70
80
90
100
TEMPERATURE (F)
i
I
211
110
120
130
m i
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 220
z
I 50
60
70
80
90
100
110
120
130
140
150
140
150
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 420
,50
50
60
70
80
90
100
110
120
130
TEMdPERATURE (F)
INTRFAE EMPRAUREHITOGAM ORPAS LVEL42
212I
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 448
U£
I I
21
I50
60 [60
50
I
70 70
50 so
0
90
100 100
110 110
TEMPERATURE TEMPERATURE (F) (F)
I2i
I II
120 120
130 130
140 140
150 150
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 1554
RI
z
050
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 2324. 0
(.)
LAzI
7
5=0
o
9
0
1
2
3
4
5
I0
z
I
o
,50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
214
I
I
I LEVEL 2589 INTERFACE TEMPERATURE HISTOGRAM FOR PASS
I
Io Iz
I'
' 50
60
70
80
90
100
110
120
130
140
150
TEMPERTURE((r)
I I
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 3049
I
I I I215
50 L.Ji
60
70
80
90
100 110 TEMPERATURE (F)
120
130
140
i
150
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 3286
l
•_
I I
-
I 50
60
70
80
90
100
110
120
130
140
150I
TEMPERATURE (F)
S~I
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 3942
S~I -oI
S~I (-
|
I I
z
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
216
I
I
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 4784
I
hj
0
I
ILI.
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 5137
0 CD,
Io
*o
I
CD.
z
I
I 50
60
70
80
90
100
TEMPERATURE (F)
217
110
120
130
140
150
I U
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 5370
(-
I 50
I 60
70
I
. 80
90
100
110
120
130
140
150
TEMPERATURE (F)
3
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 5817
(z Lo
,I
z 0I-
020 50
G0
70
L~! so
90
100
TEMPERATURE (F) 21
110
120
130
140
15n
I INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 6808
i
uC-)
C,-
50
60
70
50
90
100
110
120
130
140
150
TEMPERATURE (F)
I U
INTERFACE TEMPERATURE HISTOGRAM FOR PASS LEVEL 8080
II.
CD)
!0 50
I
I219
60
70
so
90
100
TEMPERATURE (F)
110
120
130
140
1t50
INTERFACE
IEMPERATURE
HISTOGRAM FOR PASS LEVEL 9715I
C4)
50
60
70
80
90
100
TEMPERATURE (F)
022
110
120
130
140
150
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 70
IB
L.J
Iz
C-
|
50
60
70
80
I I
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 112
m
o•
ui
I 50
I I221
60
70
80
90
100
TEMPERATURE (F)
110
120
130
I TEMPERATURE HISTOGRAM FOR PASS LEVEL 220
6-INCH
oU ,
-
oa
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I TEMPERATURE HISTOGRAM FOR PASS LEVEL 420
6-INCH
_
I
_
_
60
_
_
_
_
_
_
I
_
I:!
(AL.222
50
_
70
8O
90
100
TEMPERATURE (F)
110
120
130
140
150
I
I 1
6-INCH
I
-
50
60
70
TEMPERATURE HISTOGRAM FOR PASS LEVEL 448
s0
90
100
110
120
130
140
150
TEMPERATURE (F)
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 882
IO
N -
I 50
60
70
S0
90
100
TEMPERATURE (F)
I
223
110
120
130
140
150
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 1554
8I
z
o4I 50
60
70
80
90
100
110
120
130
140
150
I
TEMPERATURE (F)
6-INCH
I
TEMPERATURE HISTOGRAM FOR PASS LEVEL 2324
zI
°
50
60
70
I
80
90
100
TEMPERATURE (F) 224
110
120D
130
140
150
I I
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 2589
I *
-
50
60
70
80
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
I I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL ,3049
40
Iz
I ,,-)
z
I50 I
I
50
70
50
90
225
100
()110
120
130
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 3286
i
'"
I
I
S
a
50
60
70
50
90
100
110
120
130
140
150
TEMPERATURE (F)
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 3942
I
S
I
C>
-o
I
C)4
(-L.
50
60
70
80
B!
90
100
110
120
130
140
150
TEMPERATURE (F) 226
I
I 6-INCH
II
TEMPERATURE HISTOGRAM FOR PASS LEVEL 4784
I:
I
' I50
60
I I
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
6-INCH
I
TEMPERATURE HISTOGRAM FOR PASS LEVEL 5137
-
L.J
I
z0
50
60
70
80
90
100
TEMPERATURE (F)
I
I
227
110
120
130
140
150
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 5370
I
'"
I
0
"5
-
0
I
o
I
0
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
6-INCH
I
TEMPERATURE HISTOGRAM FOR PASS LEVEL 5817
'.'
L.J
50
60
70
80
I 100
90
110
120
130
140
150I
TEMPERATURE (F)I
228
I
I
I 6-INCH
I
Io
TEMPERATURE HISTOGRAM FOR PASS LEVEL 6808
"
I
o 50
60
70
so
I I
I
1 1
I
100
I
110
120
130
140
150
140
150
TEMPERATURE (F)
6-INCH
I
90
I
TEMPERATURE HISTOGRAM FOR PASS LEVEL 8080
C-,z
0
50
60
70
50
90
229
100
()110
120
130
I 6-INCH
-4
TEMPERATURE HISTOGRAM FOR PASS LEVEL 9715
I
° ,.
I iI
I
L.
50
I
II
c
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I
I 6-INCH
TEMPERATURE HISTOGRAM FOR PASS LEVEL 10350
•"
I
°
I
C-,
I
0
50
60
70
80
90
100
TEMPERATURE (F)
110
120
130
140
150
I
230
I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 70
S-)
z
IL
50
60
70
80
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 112
I
'
z
CD
I I
I
23
I
50
60
70
so
90
100
110
TEMPERATURE (F)
231
120
130
I I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 220
I
I
U, uJ L:z
i 50
60
80
70
90
100
TEMPERATURE (F)
110
120
130
140
150)
,.
I
12"INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 420
C9-)
I
I 50
60
70
s0
90
100
TEMPERATURE (F)
232I
110
120
130
140
150
I 12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 448
I
•50
Ii
I
I
I
I
I
I
I
II
60
70
80
90
1003
110
120
130
140
150
140
150
TEMPERATURE (F) C-)
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 882
C,
I
50
i
60
70
80
90
100 TEMPERATURE (F) 110
233
120
130
I I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 1554
-
I I
,.
I
CO
I
z
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
I 12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 2324
°
I
I
-o 0
I
I~C
-
i
234
I
I 3
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 2589
I I
I° I
Nm
*
,m
:Z 50
60
70
80
I I
90
100
110
120
130
140
150
TEMPERATURE (F)
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 3049 W
I
• 00
I
-o
I
50
I
I
60
70
80
90
100
TEMPERATURE (F)
235
110
120
130
140
150
I I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 3286
-
I
I
,0
50
60
70
80
90
100
110
120
130
140
150
I
TEMPERATURE (F)
I 12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 3942
I
-o
I
0 zI
50
60
70
80
90
100
110
120
130
140
150
TEMPERATURE (F)
236
I
I
I I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 4784
0
0
I
I
zI
50
60
70
80
90
100
110
120
130
140
150
140
150
TEMPERATURE (F)
I 12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 5137
I0
0
0D
""z
3 1
I
50
60
70
80
90
100
110
TEMPERATURE (F)
237
120
130
I 1
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 5370
CDI
Ouz
I
0
-o
50
50
60
60
70
70
80
s0
90
90
100
100
110
110
TEMPERATURE TEMPERATURE (F) (F)
120
120
1,30
130
14,0
140
150
150
I
I 121NH EPEATREHSTGRMFO
,
PS2LVE381
I
I 12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 6808
I
z
_
50
60
70
80
I I
90
110
100
120
130
140
150
140
150
TEMPERATURE (F)
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 8080
I
38
I I
z=
I50 I
I
I
I
60
70
80
90
239
100
(F)0
120
130
I I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 9715
-
1
QI tI
C50
60
70
80
gO
100
110
120
130
140
150
TEMPERATURE (F)
I
12-INCH TEMPERATURE HISTOGRAM FOR PASS LEVEL 10,350
,.-,L
I
S-,
zI
50
60
70
80
go
100
TEMPERATURE (F)
240
110
120
1;30
140
15,0
I" I
I
I I I I
U
A•PPOWX D
iTHREE
DIMENSIONAL PROFILOGAP
i
241
I I i I I i
I
PLOTS
Representative Profiles Using 2/3-Foot Intervals SECTXON-1
START-O
NIXTURZ-MAAkSHALL
DEWTH-4
DES ZON-VLUXXI3L
VAAMZS-2S09I
36
-0.2.
-0.2
242
*
Representative Profiles Usin &XCTXON-1 STAAT-36 DXXTURZ-14AASHMALL
EpT
2/3-Foot Intervals -4
oDZSGkd-rFKXb]LE
PAS3SES-3049
64
5
0.0
I
-0.1
I
243
U3CTXOI4-2
Representative Profiles Using 2/3-Foot IntervalsI
BTJA~T-76
MIXTURS-MARBHALL
DUPTH-6 DZBXGN-FLEXXBLE
VABSZS-3049
84
I
NORIMSLEV
0.1 0.0
-8
-0.1I
244
Representative Profiles Using 2/ 3-Foot Intervals S3CTXOM-2
STAAT-66
NX XTUR3-btAaS8HALL
OMP"
-6
BESGW-FLEXXSZ3
PASES-S.5137
140
NOPJ497-V 0.1
0.0
-1.
-0.2.
-0.2
245
BEC'rzOb-3
Representative Profiles Using 2/3-Foot IntervalsI
BTAWT-154
D4XXTURZ-I4AAt8IALL D5PTH-6
ZCSIGN-COMPC8XT
PABBXS-5137
190
NOM-MLEVI 0.0
-0.1I -0.2
246
I
Representative Profiles Using 2/3-Foot Intervals SZCTION-3
BTART-192
MXXTURZ-MARSHALL
DEP
14-6
DESXQN-COMPOBIT
PAaS&Z-5017
304
I I
0.1 0.0
24
-0.1
I
247
I Representative Profiles Using 2/3-Foot Intervals SECTION-4
BTA•T-326
MXXTURZ-GYRP.ATORY
DZVTH-6
DEZ"GN-CONPOSIT
WAJStE-5117
I I I I I 334
I STATION
14ORM4ELEV
330
0.0.0
--0.1
I
-0.2
YVOSFT-I
I 248
U
I
I I
Representative Profiles Using 2/3-Foot Intervals
SECTZON-4
START-334
NXXTUlJz-GYATORY
DEPTH-6 OMSOZN,-COMPOB1T
PASOZO-97135
I I I I I 348
I I
STATION
•
m~NOI•MELIEV
342
0.0I
l
I
-0.1I -0.2
I--
I336
I
I
249
I 6ZCTXON-4
Representative Profiles Using 2/3-Foot IntervalsI HXXTUR3-GY3ATORY
STAAT-350
DEPTH-6
OEZBGN-COEWOBXT
PAZBZB-.10350
I I I I 454
I
I, STATIONI
I NOPI-MLEV
i
0.1 0.0
-
402
I
-0.1 -0.2i
XPOS
T
--71 -73 350
250I
I i6
I
I
ZegrsenttiveProfiles Using 2/3-Foot Intervals
526
I
NObMELEV
0.1 0.0
I-0.3.-19
251
Reresentative Profiles Using 2/3-Foot Intervals
SECTXON-5S TA1PL-528
NZXTURg-GYXRATORY
03VTH-.
CSZBXtd-VLXIULEX.
VAISSS-9713
532
-0.22
-6I -6I YVOSFI NO-kEL3
0.12
52
I
USCTXON.mg
4&?ý.sefltative
Proffes
2/3Fo
- 0 .~
0.
I7
P A at
II3I4 (TervesI5 Of tiIa ei ln.
I I
I B
~APPENDIX Er
I I
REGRESSION PLOTS OF DAMAGE PARAMETERS VERSUS TRAFFIC
I I I I I I I I I I I
I
255
I
I
O
"0
0'
+
ot
0
00
•
+
+
•0
+
++
020 0
0
o+
I
t 44m
4
44
256O "4+
S..
S
..
.
.
...
.
.
.
. .
.ii
.
.
.
4
..
I
.
4
4.
.4 .
.
.
.0r
.
.
.
.
I
O
1
256
aI
III
't
I
i
4.
I
I
p I|
4
4.
4.
4.4
0
4.4.4"4
*4,
.4.4 4
4
.
,
4.*4
.
4. 40'41
m
o
00
0
U)
x
0
lC
"0 to
0a 0
0
45
0 0
*
OK
--
S
00
tioo-
4.4.
1.60
0o0
".
4.0
4.
o>H
(9 a0 . . .
0257
N
V44.4
.
4.0
o0
0U
+
0A
00 1-
I I
..
0U
• --
0
257!!
""4
tI
i n, I . . .
.. ..
,. .
. a .
-
.
. .
i .
. .
. ..
. .a. . . . . .
1 ..
. . .
. .
.
- . . . . . . .
.
.
.
.
I
00 .0
tii 0
+ X
0
0
44
;I
0
4I.
4,•
Win tko
0
II
0
0
20
o0
z
b10
0
4
N
P-
qjI4
4.
4
4
4.4.4
-z
M
+
to
0
0
00
0
0
0
258
0
0
0
0
3
I
I S. .
.
..
*.
.
.
.
a. .
.
.
,
.
.
.
.
|
.
.
+a 4.
4,
40
0 0 .,4
.
3I
ri
o
"
V.•
Cal 100 a
>0÷ gP34
aa
vi
02
om
°aS a.0 baa
"'o too
ONS
IDX
.0
0i1 *0
,,0
a
•
o o
%40
al
4
545
•)r.-
af. .'4
a• 0
0
0
.
S I
I
0
0
259
0
ri1
0
N
0
.4
0
0
0
1
I ..... ..... ......... . 0
..................... . .. ........... .......
0
0
0
""
I
05 0
V3
0
x
0 0
0iz 0
13N
0
0 a
$4
.o 00
+
÷
÷ **.,1
..
|
4.4 0 0
4
,44 00-
4
4
4
0
0
+
0+
0
44I
4
4
I
I4
I
I 260
.4 44.44' @
tko
0 a0 0
@4.4. 4.
0
o0
0
to
6o
oil
.0 NO
•X 0C
0 0 0
S,
>0
0)60
o
0
OiN
020 00
0
0 0
b)x
0
Q
+Z
%,
l mm
•+
mi
m
n m m
m
mm ni
+ +
•
0 m0
N0
o+
+ '40
O0
O
261
0
1 o,1 0
V
0
4
4
b
4.
*
'
.40
0
W
I
0m+
0
>
10
10,.
0
++
0
00
0
ON,
°w
in
v
'0
0h +
0U
.,.
4.
+
+.
I
0,,_
. .l 4.
2
6.2
M
0f
262I,
I
I I+
+ 4
*
-
I.4 .4
''4a
4"
0 0 0 0
-00 0
-
>
0 0
a
oz N0
I,
+
0 + 0
|.~
+
(C *4.4
.)
.,,..
+
+"
0
o 0
4
.
0 ©D. q)14
.4.)
÷4÷44÷40 ++
+
+
+.
0
,,,
O 0 0
0 0F
+ 0
oo 00
0o
..
.
V
AN
I263
I
4
÷\2
.
4,4.4
4
4
44
.4
4,
1"
\4
0
0) 434
0
0
0
14.
tax
a
a
%-40
PI
0H u0
NN
VU
Wem
I-264
I 0 .4
ozo .0
>
0
3
0 0
a
0
.0
0
ozo
0 0
.00
3
I!
o
U)4140
o
&0
0~l
IN
0 0 4.
4
4.
4.. 4.
If
I0 0
04,
1" |
Io
oo
+ 0
0
+
0000000000 265 265
.4.
I
aI
04 +,
*i
4.44.
*
4. 4.4-4,4,4
40.
*el
/a
a,
4.
I
0
"w)
0
00z
0
.0
0
...
0 0
z÷0
W
z)1
0
i
-
i
4,003
0
.4j
9)0 0
0
I
26 '4. 0
o
00
+!
i 266
4
r
'I og
*
.4
S"
.4 Va
doo
0 13.
u1
0
a
> 00
41.
04
4*
4.
0
+
3oo co"
q~)54
oh
5
444646.40-0+
V0
/0
0a
I
+**
•.0
+*4
4~)00-4,
3
.
4
* *+4
-267
Aoo
I
2aa
Ia II
I
4.
'4 4
+
4.
.4"
I F4 .. .
.
.. . .
... . . .4O'
6-44
a
..
oCo
00
I *4
00
•U)0
.
020
4
.4- 4
+
,
0.N
old
._0 0
48
44
+
40-
U
*~
00 ul
0
5.4)4
00
eq
In
268
0
4W
+
V4
0
I
II
I • . .
.
.
.
. .
.
.
.
. .
.
.
.
. .
rn
.
.
_ .
. .
.
.
. i . .
. .
.
- . . .
a,
00
0 z. 4h.
4.
,4. 4
,, .4.....
4
.
,44
..
m0
0
0
•
0
+o 0
I:Z,,o
co
o
02~
4.4 @ 4.@@@4.p 4 .4.4.4.4..4. 4.
71.
,
4
+
0
~4M
0
i\
I
0
0
.J
4
0 I44,
4444.44.
ISI
4.
""•4."'
41.
'S
0
0
4. f4
@g
4
.
4.
0si
mmF -0 260
I
1.
-.
.
..
l.
..
.
.
..
.
..
.
..
"""l......... .........
+
+. .
.
.
.
.
.
.
0
9.4
4
0 0
4
0 .0 0
4.)
03
U3
.+440
1
00
0J 0
0480 0
14
4)
45
4w
0
&
4.)0
.0
ON 4
m
'.
0 0
a
0
4.4
4+44
*~0I
4
Vg 4
v4
X.. 0
0
a
270
0
00
0
0
0
0
I
I I
i4.
+
•
I
-
e .
t
.
,
*
n .
.
4.
*
** n
.
*
J
-4-
414
ozo 'I-.I W) i4.
0 0 0
0 +
U) do
4,4
4.44 4-
4
4'4J
4
4.,.
00 u0 +o
"U
.
,.
0
*4,• B 0
4*,ib
00
U
024
>
0
-0
.144
oi
.0 °
-
Ig
E z
0
0
.0
"0
00
.+
4
4.
4. ++
+ 4..4.
4.
I0X
ou
0
0
.0 0
M
4+
4.
'9'
4.,,I'41.
4.
@
*4
0 0
..
.4~ U
V
..
0
0. 0~~~
0
0 00 000
a
r. 00
w
Ii
v 000
271
; 0
C,
000
9
44
.
I . I .+ .. . ..
;40 6
r
0 0
V 0
000
0
N
. 0
0
0I I 0
*•
4.4 *e,41*4,'Oa4F
4
,4,.
4,
9).4
144
0
2
V, 00 U)
0
+a
a U)4
0 0
+
" +/'+' "a•""" "•I
w•:ooo co -
+
+
. *,.4.40-
+4.4.4.4. +444
a
4144+.+.
0oo
0
mU 4.I.4I
0 co+
+
"+
*44.4...
4.
*
.I4
I.
I
I.
.!
44
o
~0I
-272
......
...................................................
272I
lw4
00
. 4.J
U))
14
(10 5-44
0
*273
0
I
.0
o.
I
. .. .... ..
. .... ... ..
"*O
II
3 0
a..,
4.
++
+.4
4.4
.
+..44
0
en
.4
,-.4.
*4
./4
4
0
°1 to
4
.
4
4
.
0*4
. 44
C5OiU %.Os
Z\
574
00
coH
+
21 Mi N
4
scro
o
r4
,
1
0
1
4.
.
4
4.
1
1
1 4.4
4
r4.44.
0
V4
274
I
0 0
o .. 4.
4. .
.. 4.
.
.4
....
4..
Q/)
0
U)o
•0
*
woO oz
E941
tl
m 4.4,.4..4.4.-.
4-.
4.
4
. 4 4 . 4. 4. 4".-
÷
4' I
0"1I
0W4
wI
tH
0 0
00
I J42 0 Ho
0
99~9
H
II0
00
30 0
I 275
0a
I +
+
+
..
+
+
.
0
+ +
0
I
o 00
+
+-
l
4•.•
+
41+
4.*E÷
44
0om
0
Q to
0
> U)D
l0x
40
I
0 00
,
0
H
0 coo 00
.0
Col.
+..4 ++.4404.
+toVo
00
276
r4
0I
I
'+
Ii ..
.
.
.I.
.
-..
.*.
.
..
.I..........
I.
,,|,
•I........
.........
+
•(n
I
0
0
4*
4++
ozo
4
4
4.
4'
+4
4
o0
4+++
4
*4
*
4444
-"4.41
co 0 z0
4**
•m
f,
°!4
4444++
44 +4
%a0
I
0
H
'0°
poo
0 mo
0
277
4
I I
277
4
I i
.
.
.
.~
*
I
-
- I
.
i
*
* *
~
i.
. '.... '
' . . '..
*
"
i
'
l
"
*
•" 1
•
ul- ,4++
+
+
I
0+I
0o° 0
b 0 0
Io
0
9.-F
m
÷
t.• +÷
֥+
+
+,m,
!O X000
m
0
281
0
0
I
I
o (0
w! too 30
.
t x
CO4
°II
0 0
-
I
I
00
oz
)
++
+
4
P+
A .
.
. . ..
0C
(4
(4
(4
04
.4
. . .
04
.
.
.
04
.
. .
04
.
+
I
r4 .
.
.
.I
+
+
.
0
44.
1.
. . .I
..
4
.
0
0
0
0
0
0
0
0
0
0
282
I
I I
o
I 030 4
+
+o
+ 4
0,
z
+
4
*
+
*
++
0 0
+,
!.Z x V) u
o>
o
*
00 Soo +•4
4. +
+÷+.
+.
H
0
*0
Vs 0(
020
*
*0U
aox
0
..
+
0
+
. ++
!I MW
++,+
;"W
+I÷,•r
0
!H
i
283
~
0.. 0
0
444+44.+4
464
4
c
FA
c440
"o 44p
0
(AC
0.
4
4
4H+44+*
4+*
4
blI H
"CA
CO)
o
crc 0Ih
00
284
I L-4
I0 iI.
. , , . .
.
.
..
.
..
.
.
.
.
.
.
.
...
.
...
( a
0
I I
.
IH
W)
.0 +
4+
4*4
4+
.4
+
00+
o
co,
0 14
+'
+÷
+
+
05
00
I
I
0
OO.4
I
*,
4
+.
0)285
++ 4
4
II0
.044
4
0
I Eii
*0
I
°"
a
N" z +.
+
,,
+
+
4
+0
S.
+
a
00
oz0 (q•
q
~~286
0-!
I
,4
(1
4
44
4
*i 0i
L,
. , . -
.-
i
.
.
.
.
.
.
.
.
.
.
,
.
.
-
.
0
.
a 0
]v
CO m
fnx (D2 N
0
E4-41
.
N 00
U/)
0ON
o,
'
..02
4
+
+
%,.4
0,. ) oW
)14
0
00
4+
+
o
H
o
4
0oo faIn
v 287
0
0
II o .
........
,.
.... .
.
.. . ...
..
..
.
-.
., .
..-
- .-
.
.
0
.
.
0
3 I
r..
I
4.)0
U)
'00
0
-'
3
+
pi
3
0
+ +.
oz
I
z
÷
'
,1.1
4
-. '1
0oa
NiI i)
co
.
....
""-,4
%,00
+...
4
•
,4
F
E4
O
r
0
0
0
0
0
I
0
88
288HI
0
I
I
g"
l
U)
+
H0
. -.
+
:z÷
+
+.
+14
+
.
0
.ozo
(0
,.4 +4
+
+ +
+*
+
+
IZ:
I
As@
*-
M4
44
+
44
44+
+
00
*
I4 I
;p4(Ui•
N
H4
V
00
0
0
I0
00,
0
0
a
0
•
0
4289
00
0
0
0
0
a0
00
A + +p
+
+
+
+
0
VV
cl
I .*.
FN
00
029
4
4
4
4
0
I I I IAPNI APEDII
I I
BOXPLOTS OF DAMAGE PARAMETERS
I I I I I I I I I I I
I
291
I I II
]-:-
"
I
I
I
iiOi *__.
__14"
a
-I--
-13II-
+
1.S
292o
I
I
I I * I
i1
II
I",,I I
I
i
i-
,
;
I
I
j
-
I
--------------------------
I
I
--
I
+" I I ----- - '-- -- -
-10
293
i
-C
I
I I --- -,
I
4-
---
I NN
N
---
-
-
-
-
---- --------------- --..."----------- II
----.4
I
-----
----
I
I
....
*
4
a
4
*
.4
294
'
i .0
0
0
I
I C! 0
U
I I
1
t
-
I
---- 1_1
III
---------
1
----------
--J__.tIs
*
S4.--.------I
A---------------------------------------------+
in- v4
I..
II
2 I
I
295
7+li
I I
!- I!
+,+I in i & V
"296
3q f D - -I---1~
! •I01 i1wI~ a N.
j
-N
toL
I
i.,
I
-I
.2
I I
' i 296
I
I
I I
I
.
4-,
-----
- - --
- --
I
-'iI
--1 +1 ÷
I_ .,4
II
L I"
*
I
.... %a i
0 0 0tomon
0
0
4
I
I297
0
IInin
4
.
,,.t
I I I
I i
SLEL---
F
-
!!!
I
~.w
÷1II--
a
I
I I
. ..Il-- ' ~II o --. I"
--
4.
I
.*
-,I
4÷.SI
*
I
--
I
U
4.
*
I
-
,I
+- - -
* 299
I,
I I
,, +
:
I ~~
+
---
--------
-- I S-
---
÷----÷-----4.------
o; -l0
C;
; C
o 0
0
+
~
-
I-4-
÷------÷------÷------
÷----÷
÷------÷------
49
; 0
10I
30' 300I
I
I I II
I
I~
II' !!-
I
'I"
4
t-
T
.t
I I
,-t
+
S'H
i
-'
I
-- ---- ----
--
--
- -
--
--
0------
- -
I-
--
j.
+
t
aK
I I
@1
IJ
fnI
(
4
0
0
'
-4
A
301
I30
%% 0
in
-W
!
fn
IR
I I II ,-+ '•I
ii
~
"~4
I
A.
U. *
I
I I !
A302
Si
r---!
l
I.
I
I
t --.-
I
tin.
I I
4--
..
-
IN
Z
I I
I
Iz
SJI
*
I
.1
-C 4
303
I I I I I I
I 1
-I
I +
I I--+
I
U
I
-- 1
cm MU
I
-
-4. 6
I
4.1 1 _j-j.
ii
a
-4.---
+ --q
---
4. (4
4.
--(4
4.
* -.
---
---
4.
4.---4. U 4
-E
--
-4.
4.--.
a
-
--N .4
4.
---
4.---
4.---
_
0
0
.4 0
4efl
U
1" 1
3
K 1 0
a 0
-.4
a
'aA
.1 I I I I I
6 'a-4-4 0U Ad
a14
U
304
I
I I
I.
I I
--- ----- -----+
I
--------------- - -
I
,I N
N
*
.4
.4
.-4
.4
*-
0
0
I4
I
!0 C
4'
u4
I
0
0:=
0a
.1
I
d
----------
+--4.--4.--4.--4.---4---,----.---,---4---,
-
I
1
305
l I l
I I iI ii,
4
4"''I
iI
ii
4.
in1.
mU
to1 -
+------
------------
o
N
0
4
,----
a,
10
r
%D
0
0
0C
0
--to 0
,------
---
fl
-----N
!
4--I .
C! 0
-4!
306I
I I
I
i1i
I
Ii
,
't
i
-- ------ -------1 ------
a-
------
U...
I
fn .4
4
.4
in mtf%
.4
... 307
I II
i
I
307
C, fl
4
I I I
i II
------------a~
--
--I - -t W
-
I
a
Uq
-
I 3080
I
...
I0
w
I
IL
,--
00
309 (The reverse of this page is blank.)
I
""
I
~APPENDUX Bi
RESULTS FROM PEARSON PAIRWISE CORRELATION ANALYSIS
I I I I I
I
311
in r.
momwS
I'D
SO0 3a f
14-4
31
in
~ ~ ~tO ~0* O~00 OO ~ ~ O400C0000'00.O 44t ~~ ~~0 ~r .00
O
* 0
*
~
Ui:-wU). C occ
0
00
--
inOOOOOWU a
0 10
00
4
00003
.ýq.tooWw
~
(I
lVw on A'40 .4
on. .4
"d~og@0)wO 000
W)
oC
00U)
r.C at
W"'
00000
N
0 .4
.4
4
(44 " N
.4.4
U)o
.
~
.4 04(4
~
to
Wt4
a
wI"
.4.4
O
-0.4
.4
.
1r
)W44W .4~-W4 UtU0~t(4)' Ono)*U). ON
aw
1=014rs.4 ...4 . ....
.
4440. I4W44 to. . .IWO' .4)~oU .4.))0' .. 4
,('14
e .42.v44..:
t4ton
(01.(
In .
w-.
v
WCI 10 .
.4 "4.4
.4i.A4
('4
4.4.
~
wo M~OU)CC .0-o U)f
@t .0 6,)0
~toww.44O4o
=aC.4U).4qU~wwU.4wo a..wtto N.4)W NUW.44l
3-1 00U 4U
.w
M.4t4
)0U .4
*WOO ot el)
"
5oIw
V-I-Wtc4N.Uto(409w..44to.Mm )U 14 .. 4U
.414ý
V61iM ))))VMVVVVV
313
O O~.C .4)qW4)0U.4~o))4--
ft-)fe
4U
V4" .ý4.4M4, 4. V l
I
00
Is
o"W)C4.O
80 3 ~
.4
m
.4i U4 k4
o~
xI
wt..
W.4'iwo OO
tg4.
.
C~
00000
00
NOwMor-4oR-
4pW w..
.4
.4 .4
(4.44 A
inH 00IUU0-w)
f
"'4(,400
M.4UrtU)tol)UU)
SH
M
B~ ~0 .4Nl
000 000
@00)
a
A
coo
400040000 so
a
t
@
~.4('40
0
I0.4woNo
ew~or ea"N 4 U)CoU4-Vn. t.0U.(4.4 U)W) W t. 0 )Cw
U)O.4U)44Uf.44
141.4.
4
meU)
CIOw(
I.4
))VUUUU
.4
.4 r4
I 14.4
))))NUUU
0 .4 N
I h'4
.4i Dq
000Is
0* C4W
O C4.4 '
:woww
w4
f
0
0 00ý
0 Ia 04 14
0 ca
4
a
0
NWrý C fn NqU N 0000in &
0
00
0
a
I
*
0.
-40
0
C
*
~
j..
M
,
"
I0
I-4 f%
0 0.e
0. in-
00.
0 00
.4
~
0
0
WrNflml.~ 4
*14
w-CmnO~n314
0
0
"3 64
;0
goo 'm i0m oo4 0.0 .0 4
040 0*.40 .40M~ .40P 000 @00
tfl..o P on 1%30. C.4." .0
"OF ean 00.4 w. .0
C44 moo e .4 Ssn v.. i-.4 C.. 0 i . 90 0
47
4
.44o non 93 .4 . .;0
M no Cto O.4oa" 0. .0
.
o
n 0. r400 0.
0
e
O
O
4 i-M 0 0. C
P1.r0. C 0O 0. 1
7!
a
0
Ca
C
.
M4
a
M 4
""4
N.02 wo 0 40 e O
~ n.
0. n
@0.9
004 o.
4n.
n o ZO0
N.0
NO
0
=
(4
1 10 00
1 w
0.
04ý n.
0.
~.
,4~
Cf 90.4 .4 0 C i COO 0.
.4
CO 0 :a,4 4 ý
@. 0 C.z 0
000.4 -o ie 1z1 i .0 .0 N
*
4 00
00Cn.4.0 0 0. C4 9.4.44 @-0I ý 4 C. 0. .4.
o .0 .4ýo
C:00 .40 4i-..0 0
r..4 oC 0 1 ,C; *0 0 .
0 00s. 4i-04 .0 04
M%
W d0
0
.0.0
04
0.4 R
C"P 0*00 .4
90.
0
.0
0 M4x
a.4 . 00'.0 .0i.4 -. 0 .0 .0" 90 0.
47 0 go
UC
N.OC4
:go *4 CM 1 0OM
.
n 0 0;a
'90 C0
"go0 ' V40" 4
0
.0
0* C
M
4M
x f
.4 i-o M0 ne0 4 00
.0
0.40 v i-o SCO I 0 4
.0
n. .0 0ýa
o
en 10 . 0
0
e
a
0
44
0.4 0
O
i00 .0 i-0.4 *;a 0
0ý
0.4 i-Oe 0. .;0 .
4 7
M00 .4No @10 .e0 0NC.4 .w . 0 !0. 0 !
*. NO~ 0C.4 0 .4
47: 47o 470 47
00.403 40 '4C no. W.0! 0
i
w00 44 f%49C.4
0
CO N0 0.4! .0 0.
N10 MOMC 0 .C00 @Ct 4 r.0 .0.
rr47
A~M.. .- 0 C0
0Io 00 i.4 mC
a
C.00 COO 00.4 .90 .
0 04 C.4 a!0 0.
0. o COO
4.
4t
ca 0 0 '*' o se n30 *40 ZOO 00 4 0. 4 '*0.4 n.4 0. 94. 0 0 . 0 e 0. .0. 0 40.
4747 x w * CO
.4
i
0 C.4O
"0 10. 4 cM .40
ý01
O m0
MC.40 00 an 44 00 C0 O CO C.4 N .4 .4 . 0o .0 C 0 . 0
e
C.0 .00. 00 . 0 .4 0 40 40
0 ý 0 .. 40 o4 n0 Me 100Z M 0! 4*4 1 0 0 0 0
44 0
00
!*0S.40 N.
0
,
n00
HC 0. .0
e
00.4. 0 -OP C!0 -O 40
W
-o
.0 40
Ag en H4 Is. .0
4
CO
.4*
0
:*4 440 .
000 0.91 0
0000 C0
@4C 10:.40Oo C" C 000-c or .. (. .0 0 0
OO i-on 014? C 0 .0 ; 0
0
'.0 -10.0 C .04 .14n. 00; 0
"No4
no '*00 C 10.
=440 on i-00 .4 0
.0C 0
.0 0 C tz i1 a . 0 C oz :-0 LO04 ý0 04g MOO4 " .0r4 44 C,0 .
0
I440
.
ý :
';a
0. i--Cu n 00.4 i-0 oq 0. ,. .0 .0
C .4.C
004 '0.4 P1
nc 0 .. 0 4 00mý.4 P0 .4
0
0.4'* C 00. .* M0.0 .40
0 :.
0
1
-.
a
0 40
'
0ýf a 0 i,40
moo
n4
W
0
.. Nf 000 Wo $.
n. 4 c00
C07,4
" ý 0
.0 0O
:
0
0.40 :., 000 WOO 0 0 i .0.4 00 1i-Oq. . 0s~ o 0a C. C .4.
Id14
I
0
.. 40
..40.4P0.4:0 a.00 , P4-. 0 0
" .0
*
COwI .4 0* .0
C. .0
0
94'*.4
470
6
M"O4
0~Z
nec0 ne N.40 "so0
moo 0moo 0 OO '0 0 i04C! g . ..0 10 0. A.
W
Ul
eno0
*.4. '*.. 40
*0 0
4J
* 0
PN'*
.40.4o COO0PP P0 !~
@00ý 0.4
%
4
C.-40
40.
9 0
to
a
moo4 ":0.4 0o i-o
n
A* 00. so COO 0 0 .4i 4. 0. 0
.4*
00 0
* o eo-
i .00
1.0
0
.
-0
47
0
17
.4..0
00. Coo
.. I..
:22
.03
=4.40 : *0 0P1 i-C Q014 .40.4 I 0 .
ne0oo x Qn
' . 0
P . 0
. . 0
f%.4 0 to" .400 .0. . 0 0
.0
t
"
40
z
6 5~ .40 c - ai n" "~e on" 1400 moo No~4 @
[email protected]. C 4 in 0. C! A0 oi.0!
a4
.4
4ýl
3ý
ý
C
4
4c
*0
3 ý t.41-0
0
Ct
4
94 M1ý
03
05
x
0
Ma
e a"e ia Pcn '4 NNn.',r!4 0a
CC0
M0A
a
woo
:n4A
.4
0
r..49 m no COO0 i-.
::0 in W.0c 1*04
04
W..4 no 0 "moo 00o 1
0
rIiS
4*
0140 .0
.01.0
00XO0400
034 04
i
a
-1 0
.9"s0
* 14iiU
i
N4Q 0 -- a
.4 -c
o
-
*.
0
~a
o
C.4
amo .
~
4
i.40 oo c go weICw 40
~~ ~~~
*~ . 0
Pc 03R.
Nn-I
~
~
mU 1 noI 7 .0
o 00 10. .e
0
~
N.4 00~~ -Ic N4.4 *e . *C 0
0
CIC o 4.0C-C n.0 ne I-cewOO WOO 00C.4 NO.4 4804 r. in. in. X. !.0a0 . 0 0 0
u0 14
i
in
i-4 00 A~ inO
~
.4 mon C aC O4 ' C! .0
goo n.4 N
e1*00 i
0.
1.;0
4;0. .
* k.
t4 00 1
S.4 N
~ Poin ww0 o
ceo cNO cc
MOO .4 0. 0M9 0. .0
In C I
0~~~
0
weI Min0 4
0I M Noo
ao0 .0
*4
0
I
.oC0 o
.
*c
40.,
0
.4 M Mnao * .0 e
.;0
;0.
O00
0wo
I
0 c.-4 0.4 000o CO4
inno nanm ec 4.4 .0
7
.
Maon 0 00.4 .0 0C
oom iO 10 00 1*0Is .0 0
.4#.
-I* C
.4 .4.
.4
,I:
I-so
oo co woq004r.4 0.
woo~
inca
inO.4
100.4
.;0 0
i-40 caeon Ce-I 4.N .0 0
ONO 0
n.4 .4n o n . .ý0
1-1C0
in-I0 nON ic *4 *0
i-on inca 1. C;0 0
.44 .00: NO.
0i-. ceo 0. .;0 0
Won Ni-O oo .1.
.0
cooo -ci0 CO.
. 4
.0
.ri0
0
eec 0inN ee .er nMn .4c 0 in0 nw0 . *.4 3 0* O N4 0 .0 .0
0
-Ic.'.a wwo *0 0
.
.
C0 a0 0
COO ro4 C ac. r-
.
N.
*c
C; 0
cc
If
In
.0
0
-I3 1 ewn00n ow IWO00 .0
O.0.4 'n 0 0
.0
nN.4 nia 0. .0 0
eo awoOe;c CO-I NO 0.C 0.a0 0 0
0
0
nN-O ew0e.M 0c. 9.0 0
W
0
316I
fA
I
M
.:0
0 1.4
No Mg
as.0
0 a
we
-0. 0
C-Ic OO .40. x .. C0 .0
.4 nOn 43 ee rc O4 .0 0
c
.40
con oro0 w . .0 0
03
9
C.
n
.!0
0
.0*I
0
0
vN ci wn .4004 0. n .0
N-I i-no: on m n 04n 010.4 Oi-.4I 0. 0 o *a
CO.0 0 .c 0
.0.
7; 0
I.
in.0N.40I woa so"i Inn on 0004 N .0 .0 0 0
c 0 0
100.4 N4. -N.
.0
0
In
*0
7ý04
i.4I ~agc 4
M.oo
.0 0
.0 0
nac0 COI N .;0
rco n4. n.0
I
"N N.;a*c
0
in
M" w"
W~ln ina .e 0
.0
.0
0
C;0
ace0
00
V.~r- 0
AO4 CI
I .0
M.-o 4lO n a., v0. . 0
a..
w.
0
0i
w .40M N.CO-INO-I4
a.
.
0;O
0o =04
N000 O
.40.
.0
.04
0
N.494 -I90 wn .0 0
.0
0
c
.40 :nN9 c nIc .4 0. 1 .4 N 0 10.0 *0 .0 9 9 0
4. "! 0
M. n.40 C 0 P. 4 0 -4cn4" corn a*0 0O woO C 10 . 0 .404 0.4C" O.4
.0 0
woo4 cNM no .
.-
.
mg0.
-0 0
00. 00 . .0
C;0
0ý
N. o ft0. MIC
-ý C.I
Cin rO oom moo inO. r .0
.4.
.;0
0
0
ea. eec 4
.0
eec wain we MN4 4 0.. 00.0
c
0
0
0
OO en .4 0 .0 0
.
ni-c wan inco 00.
.0 0
0.0
a .
a0
7
0. -Icc
0.40 33M .
C0
;0
.4
0
*;0
0woo moo4 Nec 0 009 . an0
wO 0
.40
:10 N-IC
9*00C .4 00-
I wo ec 0. .I4 0
~"
W-Oc C-IC C.40 NO on oc W-. ann mot.. n-Iceno. . w 0. .0 W! . W!0 a 0;
A
%4 04
0
;0
0. .00 0
M 104
in.0
4
.
i.40
0
'0.4
C.0
C-4C!.4 0 00
.
:0o
a~da.40.
ia 0o
C 0 .;0
*;0
0
00
.
.4
0
#-..I0 .4a r.40 0 i-0. C.4
0.4
0
no 0.
AO ORNecxA in-
.0 0;
rnN0
.
00C
to 0
@ 0;
:.40 On..4 wo i4ncnennn " 0 eec a C A 4 40. .0
40
.0.
*
0.
a. 00 C;0
wngMge ýrný
0
;0
40
90
.4.40 -IOn moo MnC-I
~on
n0
Q
.0
0. 0 0
3
0. 0;0
:
.0 0
*.*i woo wO.!4
.0
;a
0
.0 C!
i o
*O
00 e
:o
0
g
.400Oz 0.
0
1 a
-o oa 0ý
0
S,
R
mo
o
in
'
I
M .40
.4o
e.a .0
0
W'O.40 m a. C0
;*e
a. C;0
0o 0A o." 00 0
A8
Mo
r. *o
0
405
:0
0
.0
0
""
Io 0f
C
.4
A A.4
A.
0
*0 0
0
0,40
"Oo
%00
.0C ;a
D.I
n
4U .0
0
.0
9.
%4
C.-1 01i 48M C0O 0.4 . S0 0*
0
-a .0
0;
"0 0
0
aC.0.
0n .e
W ,40
0"9 0
Coi M-~i
r.i (0P
4. r0 00
in
Cý
0
C-0
Nn
0 0 .I,0
O P1CM 00o 4
.-
.
*e 0j
,4
C0
00
o
A1 0 @0.- I
0 ¶00.c0 aP1. e;a
C0
4 4.4
s.
C
00; 0;0
C9
0
o
@-0 ico
M
."
-c
iwnI 0 i-co do 0,4 0;a
C0
M
9C
A
-
0
.0
,i 0.0 *o
0
M-I
P1,40aft"
Ci
-i
woo 1.
tooi
9.0
4o -0
e91 in
0
0e
01
0
001 P1CM moo002 0 A ,40.-
:-Cýe i-in P-c2 O4ý
0100 O4%
0
CW MOOa "104,4
,0 P.0 ,400 Mo0 0M.4 1 0%, .-
40
4
0
c8
.0 e
014 i0 ¶0, e;
91 Cini
.r!0
0 1% 0 010 i-1 , e4
47
.0
C MC0 M01 00 A1@1 CO.MOO 0.4 a. *O .0
91
-4O CWm
i-c0
004
-0
@4 10 P10,4 0
c; 4;3
1
o
COO
0
0
.0
Z.40 O 01n Mo00 CC4 ý
0I
-0W 9 C;
.0 0
90
90
0
0o
,O
1
*e
12 c
maCn
4. 000
.e
-
P1-I
-I
a 0
@040 P1CM r 049
0
woin
oý
0
.
O*a .0 0
r. w0 M. -0 :00C ..0
C;
0
* .0
0
i-40 to
go0A
.
0
.0
.0100 o -i .-I10 .0
-
0o
00
o 4iO" MAC .- c eno 0 0-04 C, 0 4-arI .0 .0
.1
.40
C0
0* 1
0
01
-
0:o
0
Oi-C Ctdi 80 CO4
4
.0 0
*
r
[email protected] C. .
0 a
Co;0
-e
,4O S.0
CCC0
t-0
0.
r.
0
inC0 0091 i o 00%,,4
041nCM ¶84 i'O, N. .0
!V
Co -a
a0 ino "so.a . .0
¶0,0 i-CM OO
44
0
W,9.
N.0 CC CCa. !4 -C;
i coo
f
wo 0-9 .0
.10.
*o
0
0e 0
04
~
OM 0
.0
-
0
-0:000
.
0w 4 0
M.4
0
0
¶8.woo 0" P1CCm0MNE .40 wic :-vcc -..-. @.-I W1n, 0.4ia-! a. "40. .;0 0 C;0C; 91 0
maaeMoo o .1,4 V,9,4"
II0
00
-e
OO1
C
I 0
o N
o
0
@000.
Io0
90.
OwoP.
M-C a~
act 40
co00a0
WO
n
to01-
c0
01Cw
0c
000
o0-
4
-
00 0o CO.- 0 I0 . 0 C; .0 0
8 .0
S4,4
*o
0
II
i
0
a~i m-0i M
In.
0 *
on) .0 1. W. C! -o
momi
C;0
,4
-. 02
Tn0 0. M ,40
0
100
r, a 0~~ t--.9~ ai~ "g
0.
0. &i2o
IIII0 (I"
14I
.0 a
n~
Ný4
A
a 0 .
0 a
00
0
0. 440
0 0 C
44
0
.4 002
UC
0 . 40
0 00
00
in
ri 0000C 0
NO..a.C;0 *e
0
0 p-ig 0 c0 ¶80 *e 0
P.0 CoO 004 4ý0
P1
C
0o4
ARC Cro
.4.
P1@ 2aw
0.
0!
.0 0. *0
04e 0. .0
. 0
a1
0 V.*4
C04
.3
c
-a
*C 1
e
q-0
.0~
P1.0
~~ 0.
44 -*
*O
*c
C
1
p-am 0 000
U
.4' .4 14 4. -i.
1'9"
0.
0
P. 10 r- 0
00 0f 4
.0C;
-!C .
0
f0-a"
kI
0f-aý
.0
0
m
O-ac
oo-a ""00110 190c
... ec nOO W040
.0
0
,4
P.40
4.
r4. 0
0 " Mo-
000 1
;.0
~.
.4
40
CP00
-0P0.1
.0
0
01
0
ZO0
04
P.40
@0.4
.
.0
n.4o
0.4 Ca" eo-a 44. .0
0v.4
0
.0
,-.40 "S" co .4.4
0
P100 g0oo
COO p4 c-a .0 0A ýN"0 I0
I:
1
4
azo
rw 060 11 a.
0
.0
Nve e l~.e .0
% p.c
4 .
0.
.0
0
Q
3:0~
c2oo
3:.ý
4i
.0
400
O CO-a
.0
P1a
P0 l..
00 1- 00.
.0 0a
1S.44
P0 M
:0
- .4
P1-a0 in.. so" "on .0 0200 00. C
.40.. 4
.0 0
a
.0
P: 1.41
W ":0M 00"
rn-c 0.04
:o
3000 10 @0.4 .0
acMoo S0oo .0
0
.
1.4 X
04 ".0X :14 rt:0 W1000 0.4 Z0
44 g P1 600.4 .4 4;0 0
*
00
CS .4
P1.4
.0
*
-so C4 p.00
11 c;i
f%1
2:
~a menp.W 0
032 0 @0.6 . 0
81.04. 0 .0
0
20
P0 .0 @.
PC 4.
go1.400 a-a P1 0 mg.4
0 a04
.0
.0
0
0
0
egg
.-Ce-a 4a
&¶4
0 .0
.0
p. c0
P1
¶ -a
¶t
r~o 00
MAC4
p- a0
:.40
P101
or0"0
P101
p.0" MI
.40.0
p9c-a
o @ P0-a00.
WO
-0
0
.0
n
0
a-ac P100M 104: .0
P1-a
.4
1
0n00 9e pc-
.0
.0
.0
0 f.OR
00 in I
0,9
NO
I
0.
M
1P0 01-a0 C-a' 001P01P001 MP0 000CO00P1010 SIP1.4 00. CO-a p.0.0 .0 .0 .0
.0
c0
.0
.0
0060 a
-10
:o
;4 0
4
.0
.0
.0
0 w
.r X
1"4 :194
XI-
0
8.c;14 0 C-m WOO
0
0 00c
10
r.
0
m@.
.
90 0S
W
001 0e 0l"10.4
woo
-cP:e0 40
00
.0
0
@440
X.4
0
1M.4A
X
P1-a0 P0 "P0 s~0 CM0 @0 a
0
0
.
t! 0 aie co
.0
-0
0
.0
o 000 0.4 P0a .0 .0!
4M
X1 C O P10-
4
V!0
C001 ý 4
0
0 00
0.4
¶
¶
.
..
Ie M1 *0 0
O1%0
0 iic
P1-a
4-0 0
.40.40
P1@ a
.0
04
.0
0
CO
0 1 :
0P.4
.C
0.
¶
0
. O 0"..%0 an ~~~~ r0.4 M10 ~ '4-
2Mo
.40
0
.0
0 e .40 w MOM 0.0" nec P1-a M @0. .4 0 .0 ¶4 0
a0M00 0 0'4
P0 .4 V1000 p.c-a P.0 0
0
;0
0
1
0
*.0
.P.
SM
a4
¶
0e 4 M1 0" .0 0
W4 .4 *
0
.- 0
*0-
.02
0
0 cMO
O
0 *00.4
0
O
00
0
C 10
0 00.4
0
A
I.
p.
0.40 10 @..
0 1.
i.4
ZO 0.0
0 0.4
0
QP 00a
i.0
0
.409
4.c0
.d
1o.
0
0 0 1
0
9
-a
0"
-c.0"4.40" p.00 P001 "W00 .. .. 0P1. 01 0
.0C0
0. *2
i--a M C100
In0.4.-1002
P100 P1-
0
M.
0 -4
P1C 4; t o
g:0.4
.4 01 00-
0*0
.0 .
.0
'
.0
0
OCO
~~~co
.@ 0.4 p :0a"
0
0 0 0;0
.0
-a-c
9...
Ar-
0
.0
0
0.
9.0
A1 .0 . 0
r40
*i@
a
0
0
w
14482 113
.0
"'40
0
:090
.
99
0
¶
0
qeXo
4.
.0x
"-a 0~ .
0,
0
.0
"
0 b,
.
.0d
a
0.40
j
0
0
*
i.e .40
*@-a
MO
ii-H"C
:92 99 c;
.""0
.4. 0.
0
*0 0;
.0
0
.,
5.
o04
@41
*o-"a
0"
a1
.10
9 .4.06.-a
*
..
'410a
0. .
E'
P0-a0 **0.
o
0
~~~~~~~~~ e O *.0 W0 -P1001~~~~~~ 0, 161 00.4
'430.
...
1
0
0
.
.318
W, S400ý4-40 0,0 .40 ý4:#O Oý0 v~ rý4040 0.I ~f
:0 0
0o
. In..40
-1 'Cot1 0
60
M.4 'a 0
to 0 NO. 60c
.~
.4.4
0.40
t-.0 4
000 04
7 (00 no4 .0
to. 0
60. 4 "400
0
on M0 I
0
0.4 C4. 0
In.4
6.0
6
o
P4N on
eon0
n
"On
con 001 60 .
.0 0
40
mor Monor Con0 n0 mo 0. Sa 00.4 04
0.
.
.4 0 m-om 4' t400. .0
9.0
0
I.4
a .
*:
6.40
NA 600 4100 .0.
0
0.40 0 "0 60 " 010-d,.'4 60.4-
a
0
g160 01.40 :nO con o(0.4 60.4 .10
01n
.
31
.:C
¶4
C0
0
6.40
660 N0
w0
N00 60.
141.4 10 o
C4Ie
-0. use4
0
w 0ý9
0
34'
60.
60.4
0609
.
I
M
N M.401 AsoC
:M0 .0 woo ,4.
600 604
.0
N.4 0. .
N.
01N.4
.
0.4 C4IC .0
M.4.40
in0.4
¶
.0
.4
01(4.
6.40
'"O. 40
6l0 On0 0 *.60" 6m.400 momi m00 01O0.40
¶4 c 2 60.
.40.4
60 M on0 goo 40
.400. 9200 60.4
6A0n4 (-0A
.
*.0
"OR00 @..40 16D0 "
**
60
0
.44 o
.4
n; C: ::0 10. 6.4 0 9000 000C.
in o ýa. :04
me .0
;
'4 0z
0.40 (0.
0.
6.;4 0
9
a4
60.4
.0 .0.0 0 0c
.0
0
S@0 P4 O (0 .4
0
';0
.4g 0 6. ao won .o .8
.0
0MI4
0
0
0 Moo aion
.
.0 40
6.0
.40
0 a
0
0.40
o4
0.40 00
.
0
W0110 "SO" :mo :so0 .0 " N1. 1 6 N! 1 0 10 0 .
146.14
01 44 4m 0 60.
Ono4 "00
.0 ia a
a.~:
0.
0.46 000
.0 0
0 C! aC4 C; C;
4
A4! 0.0
1%.40 1
14 zoo
.4
440.-
MO
.0
.0
.0o "ao 4
O
:~ a
.0 0
no
w " v00909N92
0I.
* U A4
a0
tea
n
oft
0
I0
7
.0
.40
N
c
a
" 0000So I1. C;0
Ho
C04On z. * 0
.0
0
*
.0
14
0I
.0
0
7In.
.4*
""
0a.4
0a
.N0
z.
C
in~
0.4
0
0,
¶
¶ ' 0
w~ ..
0
.4w0 4s 4nn 0
o
3S
62
..
0 12
4n
0
¶4 ti
34 to
0
1-:0 eon
ip--
0.
.0
4
me a .0 - .40 0 .403 li r! 4; 0.'c 0
0
g
0i-. 0. *O 0 0 I4
0
*0
.
.0
~
ri
0
0
0' .4
.- 44.0
o 000 .0
.0
i4 *
04 :
4
7
wo0o ', I ýl 4;0
0.40 00024 a.
0.40
.0
R i-n oo go*. M .0
0
0
0.40
9%"
.0 ;
43 40.0ý a 0
00.4 *0
0
0.40
.0
4."
0
*0
0
.40a
0m
*
*0
nec A 000 I3
40
4
i
0.ý40
04.40
00
0 :00
00(.4 0 1; 0 000
.ý0
C;0
40 i-i 404 0 .0 0
04 -"so3 4 0. . . 04
0.40 i2gr4 "V
noo4000 . 44 ., 0.1
0. .0
0. .0 4ý0
.0C 0
04 02"40 n4 0.4 .0 0
0
.0
0.40 P3.40 0.40 41t-3M 000 P3.40 0~.40 w.n .f .0 .0 .0;
0.40 P0
0
0
¶0
0
.4 0 000 (40. 4d 0
: .0 00M0 fn i-CO eO4 4 0
04 40 .. 400 00 "s00 i-CO 00. 4.
00404 1c400 14. C0"4 0.
0.40 00 IS 0.4,
.0g 0o
.0 0no¶
A.40
.0: 0
0
1.40 0 000
~
.
.0
0a
4 03
0
0004
~
0
.0
.!0
0
:494 (O 04 woo2 00 .0
2.
7
0
.0 ;
.0 0
.0
009 0.40 '.0
i-2
U.0
-g 8
0
04 a
00".4 414 .;0
r0
.0.40 0 woo 0 i03 on 4W3
.
.0
*0
0.40
.0
¶4
.00
0
0
0.040 i-0 1. 0.4
00a .14
000;
00
0..: .43 0,
f.o
0000"aa
0.4 a. 0. .0
.0
.40
V1 0
P.
*0
0
9~~".
t!.44
07I
0.0
fot
0
0
0
0
0
¶4
0
WOO.4.
0
0.40
0
.
000
*000 "00000
0.40 no
0.40 *S-4o ag~o n Iwo rio 0,4 .0W n m. -. . on -. 1 3
4140410. . '.;0
M 040. .0 0 12
0..0 -o
0a
0
.400
1000
n00
." " .
0
~
.40
""A~4
0 4
0
(0
.0
%so
t Vý0 A 0
0 I" 01
.0
0
0..om
P3 0
00
0
¶'0 0
.4on
0
.4. .
M 00
0
a
. N!.4o o
0
0 I
'400
a
0.
..
14
0
0
040
0.40 m
0
1 ~ 1
m443 o.o 4 4.
0
"400
*O '4
k
0
000 gr 0.
*.0
(40
0
3
01:1 0
*0
4 0'40
¶ 4
"
s
*0
i-00 F41 00.oo o. .0
A-.40
0.0 000n 0:0 0.4
04 000 M40 0043
0
P
.0
.0U
.4 m3
~~O
0 o
goo 00 O
1
0.4
..
.*
.0 M¶ .40
.4
%4
On
00 ".40 i-.4 0 4400
%604 ýq
4.
%*30..0
.0
V000
goo
4"
00 a
140
0.0 w noo .0 oo .44 i-co
44 00 014 00 o0. 3o 0040.4 '9 0.9 . .0 0 I 1.40 g 0
4
i-N
v
0i00
m(4. ..0 0.
"4.4 -4 .0 440 (40 00 03 0.4 004 IPC! 0 0 .0
oN0.40 1%.40 140 gi0 40.4 00.4 09.9 4 .0 0 0 1
0o "i0.4S0.4 .4 .4 .0
A
O.0
00 r. -.0
'a-a M
W4 0 00 i-so 00. z 3
000S
woo .0 ¶
00 'c. 4
(4.40ww Va0n i-0 00 0043 0003 .04 000 00. 44. .0 .0
Ago0 M.44a ag 0.4 0 000 W 00 MO.4S 0. 00.4.~ . .0 0 .0 0
ýq 9f 10: lo N: .4O .4cm ~ ~ ~ r-oo~~ wo
:.0
.40 000 10
0
0 . .0
.44 400 (4.4 0. .0
0
0. *
i-.0 00 140.4 n* 0. 0;.0
.*lo
, %0 :.0
0
440
g
. 0 0
0
.4 f0 0
3 0 20 0a. Mo. 00. 0. .0 .0 * .0 0 0 a4
0.4
0
i-.4 i-094 .. 0;!.0 0
i-40 r0.4 : 0 .40 ;00 "s00 400 "a40.4 m4 ~ 9405.49 1 0 0 10
ý4ý49 W 49 1-49 0ý4 00 ~ 00 M~ am O a"~ 0 moI.wo0 a:
~
zo 44.
4 0 00 00.4
8 19 C00 ~
0.0 r0.0 .4.40 00140004 M0O"0 to"" .40 00-0.-00.4 C!0. Z0 a* . 0* .0 .c0 0.0 0 .4.40 000 00. "S0. .4* 0
i-.0 ago0 .04 1,4.. .0
01. 0 00 f"00.4 0 . .0 .4
14 l 3a I"
1 .4 13
a, A UN
Oý
0 .;0ý
40
ind
PS.i " a
.. 4$0 O".4 0* SP002~ "nc -0e00.4 wo"
4,8002
00.4
MM.4 2: *@ .
M1.4 0
.0
0
.
4, dl
I
4
0
9,
9
0
.So0
4Ia
.40
0. 1 0
0.40x
.004
P100
SO
('4
to
1.0fa
p.0. P P1. P1.
*00
0
0
0 w p.0 . ý l 0
:02. .0
0
?,0 9
0
0.40
#-4
0
a *0.4
.P NS00
dla .0P.40 N. pOP 01 aIn0P*. P"100 WS 00 4 00
9,
0
.
(
.
0
p. .4 .1.*p. I, .0'c; .0 47 0 0:
0 0 woo. *0.4
39.41 .0
0
0
0.40 0
0
p
0
0
4. .0
.0
N4.40
P"
.4
.0
0
0
""X P40
9-" 4
.0
p.40a
.
F d.
E W It z
O W.4p.A dlP in 00( 40W
r-1dO1 .P 0
0
0
00
In
ft.40 p.0(4
in
.4.40
.
W-.0 P1
Von(
is"
VO
C 0O
N1440
2
Oý 0 4
01.40
van C P.
0
0 P1
0
R ;.40
moo1 AP.000 ae 0
o.
0.
p.
.40
$30
00. aM
P0. a.
v,
9,O
~ 0h 4 '4U
.0
.
.
a
0 a 04
0 0
0 24
.O
(40.
9,
0
"1
.0.040 0
P1.40 00on0( 0o0 .
0.4f0
dO.
P 04;
00 . 4
P10.
P 0 .4
0.40 0:0 CO
P1 0 p . P1
.4.40 ('0 0
0.40 ('0P
0 a.40 on0P P1 m.P
M' 0 "OPmP
('4.40 0 Wv40 mo0m mo0
0 00.4 04 0;
OO 00.4 P1 0!.0
#400 (.40 1
"g P.0 P10P0
04 0 0010
0
m 02
0 P1
P1
0
lC
0
.-
0
.4 P1"a 0 a-.-
0.0
00
p.0. "1
P10.4 "a
0
P1.40
0140
p0P -400 0
0.40 *0P
2PaP
%00
200.
CO
0.
'..
10
P10.40 P1. 01 M1
@0.40 04m
4.40a1
9, 00
.
('440 "Of O 1
0 0 a
.4 0 0 .
00 CO
:.400. 9
dl 0 " M p. a 2 a
0
20.0 0.40 woo1oas on on( mom P0 CO000 000 o. 0. 0. a
a.
Uo
00. o. m .4
0
@cc 0 .
0
0.a.
0
00
1.(. mg-1
N,
0C
0
0
C;.*
- .
0
.0
0
0
2g 04112 0"1
.40.4
.0
.0
.4
x.4 0
0
4
a
~~~~~~~~~~~w 0 MOP 000
~
o0
0
3020P0 .
.
0.
0
000N*a0
-
r. .4 m1. *0.0
0i
0
0
0
-0
dl0.04 P. 0
(
'.0
.0
0
r-
.0
.0
0.40
""a4
do 00
0
0.40
M.40
m1
.0
.0
0
.04
.0
C;
.;0
.0
ic
v
.0
0.40X aP 40 .400
('4g4 HE P a10
",40 mom( p.0 P.
.0
S.40
P1.40 1 .00
.000
;0
ý 0
dl.40 C.40 0.40 0.40 P..4O0 X40 COO 04 s1 o. *0P1 00 410
0 1.
0
0 a 0 0 a "OP04 P1 m1 0 004
IX
0 0.. 0. Ct0 0
4 0 4 .4 .4000000 .0
.ýC
6
CO c 00.4 0a 0-
aO. 4 a. G0 I 0
0
0 1 ('4 00. a."0 a0. .0A .0 .0
440 0. 0.
0a
04
!4 0
a
104 #40a
.0
onImmmMM""x3 ~
0~~~~ 0
0 wo '05040
~
0.4
('4R.4 0 X 00 C40 P10.%54 O.4 '. (4. .0 .0 0.40 0P m
P-1.4 U00 00.4
A1 .0
0
0
09
P.40 P1.4
V-.0 10
24.4 0 004 e0.4
('4.4 O P0.
m. #4
09
#0.
(4.
0
.0
.0 1c
a
0
0
.0
.44
W.4 0
04.40
.40 " CO P10. 'tmo.4 P1 P1. .0 .0
,
,
P.4 0 pc C. l0,4*
0.4
.0
p.a a
9a,
40
0.X v
F
10 1*0 4 #4. 0
v4.0
c moo4 a .4~
.0
c
.0
.
o 0o 0IS 0 0000.3 0 o
. ýl
940 32Sa
ý0
~q
n4
m
ý0
as
2.4
va
I4 0
,0
44
00
0
.0
I
o
@040 r-Wor
0
9
40
ga4n1w
1-004 .0 a.
@0.4 9 .0
0 4?
0
-040
0
*
a
"00*00t4
2040 e .0 a 0
'@00 a a .0.
0040 *040 . o a
a. C.
o
102
!arn
w0r 0a 0
3024 .0
a0 24
4a:
0
@0.4
Wm04 @ a 0.
2@ 000
0
a
0
$1
0004 ."
r-0am 4. .0 0
90.0
0
H
0*.@
j
"1
.
?
4004 0
z a.
0
F-
0
'0
.0
.0.0
-
Ho2
aa
0
0
1-004 a. 00
aMO
""a4
0.40 @a
M a
"02
ar
,
.
W0 4:0.
400s .0 400 0
'0.40 .aiM
Mm a
1-m00
9m
6:0 0a0
2
220 .0
0
UR 0
0..4 33-U
0
4~00
0
0
F.04 a
0
C0
0
N4
-20 .
*@
.0
0
0
0
1-.4
0
0
0
0
o.4
X,
0004 . .0
ft
?
I
*4
(
O
2140 zq 00
0 00
.0
4'A -'
m
~~~ 0 AmE '00
0,A
0400
0
.0
0
*
earO n ZO ace
00
*o
.0
go*
A9004
.49
.0
oo
0-
0
morn
wo
P@0
00
0400
.0
0
0
0
0
.4.2
a0.
.0
.0
.0
0 W
.0
0
n
c
!
.0
.0
0naS
080
.
.0
0%O
@0'm .. 004
?
A-0
0
c
a
c
.0
.0
MO -S
am0 0400
aor onwr 0
.0
;8
In04 @04
oo
(400
.0
.0
0O
-sr
ioc
00 en
ace Voo m@0 M@0 000 1@A10A9
:004
ar
000.
C;0
i.0
.0
(00 (40
rn00 '@0
.04 00
"040
en
040 04 .0
@0400 wr aco
'@00 C;0
wo .0 0
.0
cr
9-
?
WWI *0
4? .
ac
? 0
4? .e
ina
-4
or ca
or
?
0:
ý
00 .0 0
.0
-O O
-0 O
r030 @0 '0
(042 0 @0 2 wor 0.r 00.0 '@00 on00 aM 04
.
03 l
0 *0
040 0
.0
C;.0 -40
.;0
.0
.
0 @0
-S
OM
0'
cor 0
40'@0
:So4
wor
cr.
'@0
.
.0
I
0
I 0400 '0 0
1-0 '00
4?
(
0
0? 0
A0
m4 4
m0.
'@04 '@4
-04 (40
Z
.0
0 10 0 rn a 00 -0
(0
C
00
.0
.;0
094 .;03
4;0
A0
.0
3
'04 a.000
0
c00minorM .0 9:ee ao (4
ace
S
4 04 MO. 1.4 a.3 .0
00
W
.0
a O rlo ar4 ""
aer mooe ace
00
90
a
000
4 mco
0'@4 0 o
494!C;
a M. ee;
0
. 0
0
0
@0 10.
..
u.
0 n
.0
4?
2.40m -0 100.
C;
N-o "mA-cWa02OZ wor .0
0
@0 9. a.
x3o
.
.0
P4.;0
0
*0 0
S a00 a
t. 00
a*
A.
&b
.030 004 2@4 M@4 m w o
0I
00.4'@04
0
0.4
0V@4 0a
000
0
0
0.
35 IM
0 Wn0 I0. no
4:
4
4
4
004
44
(er
4on
00
a4
ro
or
00n
n.r
0Q0 0 040 '00
1? .0 4
00
4a
? 14* .
0
4
4
'00
04 in4
e322On
? .
44
e
I
a 4aa ~a :a :ag
:a
so .*
47
4!0 @ .. o
0
.0
0
0
40
~
In
00
0
0.0 .
3
:.2*0
2
0
*.i
0140
a
0ý
0f6
7
a0.. 0 a0
0
*0d
M
3Ic 14 0 4
0..
0 0"
.0
0
0 w
W.4
.40
1
W."*
4i
1.
0.0
*@
mo
0.
2i00 g
0.
act
.0 0;
4i a 0
0
a0
*O
.04
zo
40
Vt
0
.0
0
0; V40
04* V. !.0
. .0
z
00. m
C0
0;a
01
.ý0 0
4
.4 0atoga
0
0
0 .4
.4 4n . 0
3
fo*
a s.40
a1 0
44.
0. .0
?
""X
Ir ZO
V.0
.01 4
t-00
0@
4
0t40
01 a
* a0
@00.4X0100
A10
.0
.0.4" o4
0.4 0
r
0. !.
0
0.0
.0
14
14
0.4
.4.4
00. !.0
00
.0.94"a @0.n 01;i. A. .0 .0 0 00
: .0
0
0
040C01-G
V4 1 mo0
0000
0.4 1. .;0 0
0 :00. .. z0 0
@0.4 V.4 .ý0
zo-n 01., .0
n
0.0
0
0.4
n
.4 0
0.1 . ~~4i 0
U9
X
0
1.000 O
a0
0
7
.. 0
0.0
014
IC! 0
-
:00101 01001
0.0 1 0
0
.0 a 4o 0
0.@moo
00
0
037
47
47
V.0
..
S0
.0
.0
0
0Is"
go"
Inc
. 0n
0.40 0
0
a0
0 .0
001
0. 040
040
S .
.0 M
V.0 .0
0
04
moo 0 0
47
.0 0
0o
.. 0 0
It40"01.0.
0.400 l 0
0. 01.40 .0
0
~
0~~~ :"
.4 ,1
.0 0k.
.47 V.40 .0
0.40 0
.00 0
0 0
I
m
:an~g
.0
40
0.40 .0
00 291
04002
*
u
.
0.
o
*0
X04001 OW-
A
A0
4!~ 09
a
.
0
n W1 .0.404. 0.4 0 00.i4.0
n0 .0
04001 40 ;0 0
z4~x
M
n4V
0. * i
r t @010i
0 -o:
0
Ven
0 4. 04 n1ý .0 47
.0 04 C!00 .0
47,
a
0
woo 044 049. 01001 .0
47
040
:1V0
04V0
00. .4 .0
0o. 0. 0 t .41 .0 .0
W40
"4
4
on. 0
.0 so4
04. el0 0101 .
7
4 "
u,.9 04
0.40
V0
01.0
.44
.0". .00 0
0 401 .
no'. 000 0
:_O 00 .
7
.r.O
c7 4 0
W4
mo .4m1-0 ~l 04 Ol9 0 4.-10 No"n323m 00a
on"On:0 wo
0.
"
47 O
ws m~
a :3a.
U
-dI
o an
a
0 lV%0.4d
*-.e
,11
moo
".so !00
0
o * 0
5
0
40 *0 o
(1.4
~S
0 .0a
a
2 0 infOP0 0 * ;0
~
0.
.w~*. .
a
a
:goe
" a0
0 4.4 f!a0
0a
..W
0
¶
0e
0ea
.0
ew "rn .e
le
cw.ae WS 00
0.40 E-00
0
0
in440 0 inw .04 S.
;A!!
Za ~AR
a4PUs x VWe-aS a : C a *4
440 ""
. S1
0
0I
0
0
0 S
0e
00
en..
9.~w4 in0.
OC
1
a. n aI
4w
0 in... erM 0 ". 0 Mnw do... ( m.. s-am .0 40.0 w f% . a :9ý4 9*0 0 0 . 0
W
in 3 ¶0
I.
s-ce A won 0 G. M~ eO0o *)
29 .4 .0
*
A4
9
0
wsrn.i n 0(4.4 NCO.0 0. 10
C0.
O-4
P1.4
e
0 a .444 vN '00 A-4' P1000 E0.40 a .' P .4 .0 .0
Nog :1.aQ 0 amo "11 0, 00-.4 coo w0
ze we
.40
3-140
P0 M W49-4 M 0' 10
a..
0
n wo NC.4
.0
0
0
0
0.d .0
.4
.0
04 14 so . 0
0
000 400P 9..4 004 .0
9O
.40, P00 wO4 %
10 wooa 0 to. .
a
0
0
V
r.
0
..
I
(40. in. .0 0
@
.4
a 0w cooe 1- 00-4 .
Imo
0
*~~
4
14 %4
o u
.4
W
1
*4'
-0
ee
~
-
4
I ~
0 a
.40 ear a. co X9 E 0
"
E-
C0
0 00
00-4 0 0
r M... C.4 .. a wa wor woo 00o2 004 9"O4ýNn ;0 0
M
q410
90
00 4 m0.4 1 0
'c0 0
-. 1Pot 440 woo .0 00
100 0. 0 .-. 4
w4 9."
.0 90P 04 :10 1P
s
40 94.
0
0
0
P0 w40 w'.
P10
,1. "
0.
0
0
91
4 000 (0. 0
0
0P10
014
*ý
94a4
0
a
9
0.
9 0
0~
M
.. A c 00. 01 ;0
4 I
MC (00.afla . (4. C;0
:0
;.
A
0
warn -4 oo c
3-g 0 C.40 rM con-sr 04 s.40 " C!0 o~
4ý0
C 0c
w
IA
.4.40wrna .40
04 0
.
a. 400of0 cc0-.0 14 0
.4'9
C
*
0
w.
E00.$ ';0
rn
0m
Pf f.0: 0
0
0
0
9.491
.. 40 war 000 (1.4 in .0
s-..o ern"mo .O .40.4C . M
.4 ago. 0 0'9 ;0
0
"a
x 324O.G
.0w4
14
"*4
PAX
22
"
0.4
r.0
moOO
.0
0
0
0 00 " 0
10 21
'a
A
ingo
¶
0
o
:34ýO0
=2
%d
.0
0 0*
rN
oC! A
N.
0C.4 06 0
0
0
0
N.
0
0
WNC
000
0
-4
in
00
000v
0 W-0
.0
00.0 .0
0 N00
04
000
NON ec ago
0o
N
~~~C 0.0
"No 0c 0 4
0040 00 .0 o
v 0-N .0 00-
¶6 ¶ 6
"N.40* Sao0 N
.4NOO
I4
,n
"a .0
14 M
.
~
4.
0
60
:g.4 40. @.4 N0no .a0
.
0
4 . , 00
00
.O 0 .404 0.4 *.
0
0
0
.400 00 coo 0
"0.40
0. .0
0.
0
0. C;0
0. 4;0
4
¶.0
NO4.4
00
CO4!
:00. 0.4
0
.0-
MO
toN
0a
0 0
0. C!42
0.-
0 0 SAO
.
0.-
.
0.4
0-r.
"000
NO0
93
.4*
0.
0
00.
0.
00 0
0
00040 N
00
NN 00
.0 NO
@00
.40
"
M
0
0
-40
"90-N Nc M0
0
.0
.4.
0
t
0
.400 f 0N0 00 0.4 q0
0.40 .0 0 NO.4
0.40 0.4R
00 "cNO
040 .
NO
.
140 0.4 0.
325
0.
.0
0
.0
0. 0.0 .0
0.
.0o
0
"* .0o
a
.0
. .0
N .0
0.
N
0
96 .
0
.
:00
3
Nt2
0.
00 . 0 004
0.40 0N
0.4 0
N0"
.0"0
0 M0 0 0N
v0 30
¶
04
W0410 0 00 ag N.0 O 90020.0 9.0 A0 0 a :6 0 0
0.40
N0 .0 .0 .4M
00. 4
so
0 0.40
0
000 00.
C00 00.
0 .0 0sg
t0
00
m14c 940c 4 94s 0
N. N.4
900 .00
0
so
0.40 0. X.40.
.49
.
.0
4.
N 0
.0 0N
.60 M.00
.0 00
¶
.0
. a!*00
000
.0
.4. *o
20- .40
-00
0
¶
pf-
.4 0.40
.
2.v4 .
;.40
Ct
Nag4
I
N. .0
*M!
@0
"~
32
".
00
0 NO4
N. fa Mo" .0
0
a~
0:
0 00 0.40 t020
00 o 0.
0
.0
.400
.0 0 00.4
0 -4
0
W,
0
a0
.4.4 0
o 0." .0
.
V*.* .0
0
moo 00 .0 0q
4i
0 0
0
4
*00
r-No
0
00
:0.49
0 0o 0
o
m
0
Sa0
00.
N.
.41-
0
0
¶6
0.0
0
iv4O
0!
.0
0
0
O
M¶4 0.0 0 Oi:~ 0r.49 .40.4 0
32t .0
0
N.4a
0.
.0
¶ .40
0.0
N
0
0. E~
M 0
00
0
.
¶6
0.
N. 0
A00 0 0
M.4 N
o0
0 .4
0.
.0430 *Ho
0
0
1
.4000 N %go0 NN
M.
NO 0.. 4i0
x
:0o
**
0
NO.4
M . .m0
:.40 0* .0
0.
~vO
0
.0 ¶6
0
040 N.c! .0
¶
u
ss.0o 0
09 .s4.
.0.4@
N. .0
:oo 0
i¶6
0-1O.20. 0. 0.-w a0 4i0
.0
0
o0. .0 0
40.4 .:
140
0
0.4
:
"0
¶
-.4.2 0. 4i0
N "..0
0
0 .a
.
z .0
*o
C4
0.40
.0
.
00
r *.9 a
0
.4
0
0
0
0
coo
..
0
-192.
0 0
ý 0.40"
a0 'or
2:
02
00. i.4
-0.4ý
F 0
0
0
i
g¶6 ¶6
.
2ý2
-4f
.
'
0
*
r.ON 0-O a
On .0
N.O~40
i6
.
f!
.4
Mr 0ig
"4
=0-
00 0
00
0.40
M .0
g
o
AO " f.4O
mN.
0
f- NNv4
00.
A
1002 .0
"
a6
.4 .0 0
30
14.4
0 *.! 0.p .0
"3 N02
000
A
9x39
05O4003 .4 M. 0.C .0I;
0
¶
00.4a
M
0n 0.4 0.0
V%0
004 0I .0
0
0-00 NS 002 0. 0 *0
* 0
1wo
0;
*.oO
N 0"
.0
*0
0
ýo 0
aa
0
000 .0 00.9
4
.
.4.
080 N
.
100 00. 0."o 20M
00so
0.4 00 0. N.9vM .0
v)VI U U
0
N
.0
,a :" 01A
4a
N.90
~~0*. c ". .0
0
~4i*0
.-
i... .0
¶4 0
.0
,am0
MOb
US:4
*00.
0.
0.
40.0
4
:1
n~~ C; a
.0
"
C1O
SM'
0 0
c
2 0
"
. @1%,4 3
1
4ý0
0.4
0.
i0.0.0
4Uii
I
.00 0
.4or 0
S
¶4
oo)
1) *
0
.0
N'
.40o
8 r4U) 9.4 M.40 rnor 0orM
C..4 440
@. 40 C 4
oa
in
*
.-
4
A
~ ~ *qý =. 0
4
NO4 rN4. .0
¶4a 0.
:a Coo 04%@.
$4
.4.c
.4
.0
¶
0
¶
W.4
0.40
C.10M
.4.40
MWOO 0w~ wO4 .0
U)0 04 .0
1 .0m
mn !n
*0
.0
¶
1
0.40
C.40
NI InI
l0.
w.
0
.0
0 x )C
*.
.0
.4
0!0
0.6 6!
0
0. 4!
0-
0
0I
No. 0
0
S.49 0 Nm moo C-. .0 0ý 0
4
0
C;0
0 0O.4 'a .0 0a
N
0
0
.0
0
I
0
¶
I .0
a)
0
@.4
0.4
0.
.4"0
0
¶ w
*0
00.4
C0 ýO ¶44ýa
w-.
C-0
40
z
326
'
¶
¶
.4ý
O.40
0 NO600 O.
U)oO O.-
.
'8.2 C-. ). .0
.0
4
0% m .0. C* . 0
io
0
0
we
a0.4
!t 0
.0.
4
0.
.
N-.4
0.
.0
n0 .-
@0 . 0. .g .
0
.0.
.0a
C*0
00CO *R0 04 C 30 0 * 0 ý0
N
0
N.40
.0
0
.ý0
". 0
.0 0"
gorn 00 M). r. .0
nor .400 00.4O N .0
worn 'WOO r 4 w .0
Norn
@0.4
0I
w
WOo)Cn r,30 PI0.4 U .
.0
za
0
0-4
*
000 I.1 NO4
.0
0.0
'.
w~o.
@0.4
.0
".4
sr 0
C.0.
0
0
r.a
or
00 0
*
no.00 O.4
¶4
.-.i-
C)0
corn r-02 NO4 M
a .40
0 0.. U)0. t C
C;0
0
1-0
orn .00
30
I
0
U).40
.4! -)
4
"
¶4
,rn mnoo CO.
orn 0
.
!0
*0.4 ' .0 0
.4
¶
x
C
.
go"r Nrns @00 @0.
40
Nf .
0
a Coo
0 io
0
MO
a0 "U.0 0 04 COm 0)0.4 . 0
torn 0o U0. in. rn. 40
) o f-0.4
"Old. 40.40 . 1C0 as .1 :*ý0 a .0
.194.
U). 0 i
.40 Ca) .4
q i
4
COO aO ct .0
.4
1m v0. 0
.
¶
W.1g
0
0
7
9.40
M0 ek. *!
ft
.0
4
0 x
N
W. .0
*. .0)
"0
¶40
0.4x
.0
0
o
I
4
040 4 0 ý0
W"O.
.400 wo O41 .
C!.4
!IC*
.
¶~c
.0
.49
11
0
C;0 N. . 0
0
0.40
P-04 moo N32 .0 ec 0ý
mom 0%
0. o
4
N.qx0 U)4X
¶4 .0
304
03.6
-
*0 0d
0
0
C
0
r. rC0 4 0 0
% .
N-40
"4 As*mor g~o. 440
W.40
~o:oa0 COO .40. 0 M0.4 )
Mn.-0 .40mwo a0 O Nh 4 rn. N.0 0!
c
i
4
0
.0 0i
.00
0
COOO MC onO.2 C0. r.wO.4 m) I C. *4 .0 *0
-of @00
-
I0.
4
N.40
.14
0
W"X .4 v4@o
600 00 SC 4
06
.4 N
W.4 x .40
W160
x64
i
4
Ne f-04.4 0
ýf
0 M.4 0
1
¶
.4.40
00.0n .0
P5 0
0
0.40
0
Nl .40
w o
a4
9000 -0. .0
.0
.40 in 00.4 If * 0
¶4
4ý) M
0
0.
0
r.
.
U x
.0
@0.
00. 00.
0I
U) M140
1;
")
C4
4
*0
S~
.00 00
0 0
0a C.r-0.4
.0
.4x
-) .0;
0
x '. a
00
*O 00
'.
.4
Ux
0.40
-0!
0
04
0a .4%aM"l0*.40 Ef
.;0
1-90
¶
0
30.
0
C. .0
.0
4
4
0. rn rn
1.
004 C!.
)0U @0.
COC0.4
';0
C;0
';0
0
aO 19c
. .4
wO4 .00
aR
0 rn:
04
0
0
@. .4
C0
I
~
004004
0
0
400
4i00
0.
0
0
0
4
0.0.~~W. e *
A
i.0, 01 en.c!i~ I ,aa
.0
a w ie *a
C04-a a wa 4:0
@
00
014
Sb 30
=
3
I
0100.040. o ie
o
0
0a. 1
-4m 6.
-M.
w4
Mo.
goo
Tog-a-
0
4
0
a..i0.4 en.
0.
en 0
u .4
en 04 .4
.0
a0 W a0.4I 0
a
P.40
.i40
0.-
9004
004
01 .0
.
.4
0"4 .0
0.
W.4
40 no W40 ie0ý 01 - 0 C; 0
:a4001 0 a 0010
400.4 . M 0 010.
0.
0-. i
_0
Oi
0
0
0
M4 in
0
.0
M VIN ! a,1
.4
0
00.4# 0 go". _9, .
0C M
0
;"
01i
44 moo en 0
_9 0
0
I4Io
.0ýo
0
Z00
00
@ý0. " .on
0
n
004 4c!
9312
-
v~ .0
*0
Mao
A0
a0..0
e-a 0. o0 0
0.
0"A
wW
04
a1.O .o
0
a-i5
Sn-4 HE
a
0
0
S.0 .
w.
..I
0
a
g
M
a
.
0
.0
g f-0-4 , . .0 0
04.0f
0
i no.
en00
0M. 00
a4~
.i
cc1. .0
a
a 0 0
.
04
0
0
0
01 0
r09 4c; 0
Po0. .0"
40
0:
0
0
r 0 0r .
0
0
0 04 woo 01 -40
0ý 4;0
C04 0.40 0M .0
1
.0
0
.
0.0
40
-,94 .0
g-o0
@0.-i .
0
4
0
OWM9
0
0
I4
a
.0
ý0-
M11. 1
0
M40.mo. .0
a
41 !-1094 3271o- 49.
.0
00m 00 0 1
en..
w.4 0.i
.
W
40 e
.0o~
e 0 W 0a 0 W005 0 .0" r 0
[email protected]*a *o40 *;0
P050
00
0
a
-4.4 * 4.0
v
a
:00.4 00
00.04 a . .0
0 en0a
01.40 W
I-
*.4 a. *a
0
2039-
~00
0:
W 0 0. G!o..e
ag o
W0
0 en
"MA0. goo .
aa.
*
b.0 14S .0
1000 !4 . 0
4
0.0 go 4.
.
n
O~
0100 . z1. 0 .0
a in a a~ O a0-0 ~o 0 0 . 01 04 0
M02 -
1
q go*e
W
S.0
14
0 (4.-i n
0 010-i .0
00 1S
010
0
0
0
0A
.4
a
.0
a.
0
4
* *i
0t 0 a 00.i0 0 a. 0a 0 0.0 0 00.4 .91'".04 .0.0 0 . 0 n0 a 0. 00
M
0
*
.i0
-r0-
t 4
-i
0
.4a.. 0.
0O
0
0
0 e 0 00 0Co 009-i r 10..0 n.0
0
k
M
0
0
n-
0
aa 20" a
0
0
no0.0 0a
n
0
0
n 44 a4 000 0-4 a
U
0
.4
0
04
w.0 -i .0
0 sn"
0
s0 . 0. b.0
Xw 1
.0
.0
WO.4 000.4.1 a. aa. .0 0
f% a
a enSen
IIs 0.4
4
1
0
0 act .0
a
.4 0
0
a0.4 0N a,4 a 1-00 0.4
10 0
~
no00. 0a.04 a. 01 a. .0 .0 a04 a
9 0
0
OIm ma
4
0
0 oo a
M
.0
10
i
0 oq0. we r.
ws00 0 b. .0 *
:0-i
00
04
*
0
.0
Sb
o
c
*0
a
a.4 en"i 0 01 15I-H
Soon
j
.0
Sb* Sb
.0
cn@ . 00 Ago .e :~ eýa. o..i wag. 04e ieo . a 0 . 0
:0 n 40* 0
0
0.-i
§2-. 3
.
C;
-0
01
0.40 a 0
0
.0
w1 0 W 1 C; 0~
aso
W.0.-
H 0.
00
.0
C;a
o 0
40
.
-~
2 -n :04:C l:C' a- a- I
@
W .4
.0
0
.0
.4.40
28
:
3o @
0
0
2'
.0
no4
Co
04
".
0.00 0 . P'. 0' 00..
0on
0
AS 000(00 *0
7
0
.0
0: o .40 00.4 0 1 0
:so 'A 4!
on *C
0
~
4
3
-
I ,
W
a,
0 9
CS oo @04 0 00~ 00. C Ai"4 .0 .;0
v4 ag"0.
r4 v 40 (03"
!.;
in .4
a2 0 a. 0 0
1-.4 -a.
.0
.4 0
C% . 10 N
@A a~.
0.
*.40
M1
01.
0
0
9
@m00 " . 4000 a4 0 C 0 1.4 C0 010ý.4 in00.4 00,1 -!. 0 . 0 0 C 0
¶4
0
010x9.9 3 0w
0
0
0" .in
O
0
9.
Z. 4
9-10@00 0
v
0
00
40 0. 0
0
a
.1
0
0
.-
C
.0
9 29 .0
0 0.! C!0
a 0
1j
1X
-
0. a0
C
.
0.
40
07
." 40n.4a 401 0 001.01 014 a . 4400 00
a.
1.
1
"1
0
0
0
1
-. .0
0a
0
.00 00 0woo 0 .4 0 @~4 0, 00.4 a0, @0. .C 0 C .0 .0
0
0
0 (4 10G
0 0
404. .0
4
0
0
C.40v 0
vvd0 0.4
(40 00, v0 .0
0
@0
0
0
U0 .
@ 15N 0. 0
0
7
04"
40
1-C
0
on 1"0. 04 0. 0C;
t @0d .o% v0 a. !4 o 0O ZO 0
..
0119
C.
0
-10 CO 01 10 C
.:n
0
0
0
z0
0.40a9 0
0 0"
.
0
0
10
0
*00
*@
0.0 t
010 cooto -C!.0 .0
0
onHOWS0 00 .00
.0
0
0
7
a00.
'm .O
040
00.4 00 n1 . .0
7
00
.
0
0
HE0. 0 ,C! .0
0
0
0
0
0
0
0%19 g~wo
9
40
0
0
C0 a
.
0 F0. -042 '0 - .4 .0 10
O 0. 0
n.4
0
a0:-
!v
*
0.0 N4
.0
0
0.0
0:2"0:~
0 244 04 0 (14 0.4 IF901OR10901I on00 @0n1 a 0 00 (0 90
r.0MGp:'400 0 @0, 00.4 0141, *.uS . '. .~ -9 .0 .0a .;0 r:0
0:
0
a.4 01 040
1.
0
'001 @.iS.4
0
n-
0
a . j.0
0
-.
0.4 1 000
0
40
.0
0
.0
0
0
X
900o 010, :*4 .;0
000..4A 4A 040
0
0A
o
.
0
40.0
.0 00ý
MO 00 0w0. '1 w! 0
0
l
.4 0920oo1
a
jI
4*0a0
*0
0
*
ý
0 W49 0
ic
o"
moo
~04:.4
-' 0
XS0
.;a 0
0
.
.0 .0
2 0" V§24
0
0,1a3
0
'i0
.0
0 a
w1.
o.
C
*0
0
o
40
41.0
.0 0
0 :00 .0
0 .04
zm1 *'00 1ý401 .400 0.190
7
*.40
.4A0
-00 .44914 .0
0010 W'IC. 90
0
0
C
0.40 2 @.409 00. C01
n .40 00.4 a 0.0
01.0-01.40 0.40 0O COO" .4 M 1 . 00 ,1WCOg400,1 1 0. 1.
0100
1M.
0
: r40. 0 X01
0
0
.4 .4
01.l
. IN
S. .0 ;0. M M"" 0 .
0 0 0ý4 0.0 A-0 ('400 Mo0 PR00 ", 0 .4 1014 0C. 0 *0~ #40 1.4 0ý1 0 001
0,
v1 0
A
0*.
.0.
0
01
",10 9 0 .4 000
"S"4 401 '40 V,1200
.0
0
in
C
.t0
0
w
.0 0
0
.4 9 N -4 0'0 .4 040 C"0" @.400 .4 Cý 40 4*01. .0. @0 901 01
0 0
10.
o . 0
01 010 2 0M 01
.0
0
0
002
0
01
.0! 0
o~~
a ý001
0001
0001
10 , .410
1-0, on 000
1-0, N" CO
03" 10
'0
0.
. '90 0-. .0
-9.0 0
0
C 010, 0.
n 0 01C
a1 0 01 0
4.90 001 o
!0
0
-
.0
0 10
S
0.40 M'0 0 "1 .
.0
0
C.
('4.40 A 01 ,04 "40
Co,1 CO.. 0 0- ý
0 0
.40 WS" 00 wo00
!0,0
0;
0
0 3'4 " 0 3' " 043' w
0::
10, 10 , 004 0 0 .4 -1-001 0 a. 0*.4 0 .0 .0 .0 0
.4
X
g 0 0R "02
.0*
0
001
.
C1 01 0.4
0
0
.0
0
0
On 1@00 COO c! 0.0
0
CMo
01.4 -0ý,1 0 'wo01 00.4 0.r C0 0
, Us1 (41-00 !. .0
0I
.9 2 4 v 0 04 x 14:I ts -I Mon oo x " 0 O w" 3 "n 2v~"
In
IIn
~
i
A
f
I
I
a-
I1
kn
wao a .40i or c-a0c N40.
a
0
min
a~ 0
Wle 0
.t0
0
NO*e
m..o
c-MWomar c-a i wo ý00. :.0400. in. * .a. ic aa
.n 0 "N.O 0arn *ain 3 a w....0@-
.0
a
2"g i-I ra i-=.'
.4cc
a
'O
ana
.0
*04
c-a
oa
in
,
0
.0
0 0
4
woo warnf
Ni'.. oe0 4r
10.0
Ca0I
0
0i..4 ",d
'.0 N
m800 0 :00 "a40.4'04iO
04 a. a
.-. ao z
"ina m
:N0
go.'
ac..
000 0'8zfcc
0
a
0
0 a.
a30
C!.
C
ao o1
240 maien
aýO ona
ca
a
00-
Ca.NO
4a 00
00 A
0.4ý .44
0 c-ma
fe
'04
inC.
004
'0.4
amG4U
W.40 Nn 11c
NNO
N4-c c0c:
.0.0 00 .0
M00 OiO
O il
afg NW..l
g
00..4
0
*0
0 a0n
fd
MSO Na MOMc
*0 00"
i
00
0
*
0an f* * 0 0
.4
1
0d *4=
:
A It
-
in
0
0
0
01.40 a 0 " NO4
Wwo hlNin
0
W
Sn 0 n
in
0 0. a aI
Mn.4e 0
arn 0c
'8 "a a
ot'.
z I0
NO. .0
a
moao P. 09 *0
%A A
%4
M-n 0 Mn.0 .c Mnr 0M OeM Noe S0124
0
.
.w
c-a
44
40M
MOM .0
Ca .0c 0
zo
.0
ON win
i-
a
a
A. aa4 e.
in M a 0-
I
a;
In
a
o
.4
'800 'f 0
a
.-.
in
M
.
0
W
4
0c .0
Nin00e f-a a..40 oner o mrn Wnc woo goo 0 c~'4a0.4 In. 9 4 9a 00 C0 .0 . a a a
a
N.-40
O1W0
a'0
0a
mane
ear: m8 An
z..00
.4
N-IC
0 C4r 00.4 0.0
0
awi a
.0
W-c
nzo
morn :Mr X a 0 -00 4w'.. wad t-..0 0 0
:No -WOM.40000ONO A n ina A in.' 0. *:
W nI .4 x
ccco
.40. *:
0.0 0
*1 .4
a A a
00. .0
Q C)
o
e
0.40
mnom woo in 0
.04
0 0
". NAM 0 0.0 0o 0 0
0
9nc
OfNO
amcca .0
W.0
*40
0.40
0
x0
c0 ..
I
.m0
0
.09
.0
.0
4!.0rn. 0*
-
a M .0
mom 0
0 i.n211 w rn 0 a '8.00 ý o
IW
0
4
.4
M
W.40
In 0
go Ne
.0
carn 400 N-.c.0 0
W-io
08.40 t-a0Mr. c-cm 300. '0.4 N. .4. N. aý
W~0
momn .400 11in104 .!0 0
'8-f0 orn w00 4a.4 .0 0 0.-Ic -IOi m 0 0oe 1104 .
MinN.4 0~
00.14 0.
0.40 NOMn 40 t- 4 4 CI.
ONO4
M~oo
aMe
.0 0
.0 0
N.W 0 N0 _0Mr 0a goo. a' . a
W.40 waon c wa..
Orng
'8c-0
in
"S
owe
ar
m411000.
0
W
329
.0
0 a
:0
4
a
earn 0
*
0c:
.8
4.40I amom
mo0.o
N.
.0
wio
a
.0
:ae '1 a woo iO . a;
M .4 0 ýcrn ace 000 9 4 '8. .0 0 0 N'O
worn
wAM2:N wee~
w
~~I04 *4
.; 0 0
InP140 ar-.4 0 0,c 2e '-men2 . 1: 410
Co
0
C0 a
In .40 "a" e P04
"4.40 " a-en P1 "1 04 14
41 mwe mme nwn
00N0 0n @0
i
1.0.4
.0
41 0 0eP.40o en
mo
a-n
~
N
nW.4 0 "40"
0
0.:0o so
I z0
I.!0 0
Ce a-G 0fO mo.4" -
9"R
Mwo9 wee 0.
c
0w ý
0
0
41
ý 0.s
ý
.
a41
zo
:"a rm
U.40 e *0.
w. 0 "14 0 Oaa wa6 0 M n a-e" 0.4 N4. 09.4
0.4o .wn a~
4;0 0
;0
4 0
"44 son
0.40 Ma-n
0
0
o
0
.4F
lo .4. ý
m
!O'4
.0
.0
*0
0
v mam a SO
44
as
" awn
Is
PIP4.
0
0
41
0
*-4a r.30
m.40 a-o
.0 5-Co 00.
:.4 wee Of
i
:a
00
p.MO
0
0
0
P1.4 me
0
05
o
0 0'0
0
a
1.0
0
.0 pa"a 4 p0.
*o
5-0.4 4m . "4
I
C! -
0
.4X on
a0 a
.490 Wi
1.n a~ ae N P40 e~osena
C!
0
0."0 Sao
@eewew
.: "n WP4 0 a 0
g4 we 4PIn .0
;0
.0 moo w4e4.
e
c4
Me. 30
M44.
c0
4
10,
Y O.. 0
V
to W4
I40
NOP1
ý4on
H~
t.. 0.
VIf 0
a4
4)40
41
mW*g 100
2 00on
1004 C4 .
U004
o 410
=
914M
ro c
~
0~ .4
P1.4 oC 4
W3
.0
M
we"1 P00 00.4-
o *
.4 Id
:c .0
.0
4%: won -19 4 P NC 00C
w4
O
P4
W
0 :4 0W
0
0,00 0
ýý e
0
aM
.0. 10 4.0
0. 4 .0 w
1
M1
.0
w.0
no Mo P0 P0 MOO . .0
40
!.0
10
40
W4P.0a-4 00 N A Mw M a wo W.4 wom-a00 0a m. P .4 a. 0 .0
C!01
41 0
:a wonM P400 p.0. O .0
P10 . 4;0.0 0
a0
r,1
.40.
a-we
w
O
PP0
0); . 0 0
.0
0
I I
0
S
P
0
. 0
omn
aNoo
"W2. 0.
.0
0
0
.0 0;
V.0 4 P.
moo
0.40
0
rie .-
AMP. 0 . .;0
wong
awn
4
W0.
f.4.4 -.
O
0
4fl0
moo
0ý,. .0
0"00..c!0
992 O. P. .0
0q
41W.
P400
w.
4;0
04
0-.0
A -. 4 .0 0 w
WI
woe
n in ACv~ 0g
V=
w..
1 C!04 0al
A0
Mo0
.0 0
Mw2140
-4*
0
0
m
0
C; 0.0
P
0
44i
41
.0
0.4
P.0
W~on
.000 N40 P1. .0
mnom n42 100 P04 0.0
*amP 'o .00 1 .0
4.940 io -e .40m4 4 C a
P4.40
wino a-a-n w .w 100.4
m.omI OP"c .0 O4
won moo P10 C!.4 CO woo. m0. .0 0 a
Na4 1000
P4.40 nn p.00 100. W1 .0c;
40
a-40040 SW w n 4 n 40n "mo0 p.0t.4 P490m4 P1 04 .0 .0
0
0
C.4:4 n 0 a-e W0.4
.8
a0
.0
00
I e
0
a
.4 .4 i 04ý
4
woOc @0.4 N4 4 .0
0
.0
c!o P.0
0
0 p
W9a P
a
.0
01 0
1.e"
"on4 SAO4 0000P1 ,9OO W0N M. 0. am. P
.0
*0
0e
.0
m 01
P.0I
0
0
-40WO
P40. .0 w.
P
NN
VC; 0
.0 0
0"
a-.4 0
0C;wH
.40
0-4 on10 00o 00 O.4 P1.4 .0; .0 0ea4 .404
0 .0
0
:0 Cia
0
.0 .4 wv. .0
P.0
P.40
PPIO
.0 0
.0
v4.4 0 0.40 a-e wo P100 moe Aw0.4 W40.4 n. . - 0c .0
nom ClOG o .4.40 4 N0 I 0 woo 3 a P1.40 00 w.4 CA, .4 .4 0 -4 0.4 O 10. p
P4-4a
.0 0
*.0
O304 C;0.0
C0
a* won0
I1
.a
P.0
P.40
0
Man
100v 4 .0 P.0
woo 04 M4.
*
P00 W@ 40-1~ 0 0. *c!0 .;0
w40M
P1 P.0
ar0
;';~41
.
:on Own 10C 0 0.4 O. M'g.49 .0ý 1 0
90
p.04 N4 4;0
n
4
0 .4;
-G0
.4j
.0
P0.0
0
.O .4 'a . *;0
0-.40 n~O
n0:.f100. N.
w
40a40.4 .4.0
.0
wP4O a-w 4oo100"c000 P1P4.. ý4 100.c4 ';0 .0 .0 mo 1.0
.
.4
040
4
.40"
0
.
14
P10 ;0 -0
-0
.0 0
41
ý moml.4
V
an
.410. 0
-0
r-.4ý 2
It
fi-o a0
M9
[email protected]
4
0
*
4
q. 0
m.-0
P1 .0
.0
ao
09
10
a
4;
4.0
i .0
0~~e M-o0
U
= 0
14
4330
. 0
9I 0IC
14P
In oe
.5e ma
a
r~
W~ moo4 4." 23
flO.. *0
ego4 '0 ?~
MNO4 Mwo4 Con AN
0O.-. .0
Nlo.
NO*O
.
670
In
I-0
C.430
31 .9~~~
m-
.
g
n I
00.4 . f
:0-0
n4 40
an
an-. N.4 a on iee A0C"4-I
in
40 momn V400 me100 1 -O. 0'4
V.-4
go
M-o "a
0.40 ".4
i-.I .0 0
.04 .0 0 0
0.4 ...0 0
zUO0.a. M.
.0
O
0
moo '0
C! AO 4
i'0 n
C
C0
I0A.-4 Pn0 .
Ae. d0
~
r-40.
.0
1.0
-o
..01, .0
a
o 4'
0
A
0
.0
0
.
0
;a
W" o 4. 0
=64.0
.44 *.. .0 0
400
104 t-.4ý
o i-
0
.0
.0
W* .0
0o
0.4
Nm
.=19.
e
0
:go I.0
4
C;
0
=IX-a
.0
(-0
0c
0
('00s00 00
I
.0 0
a oanna 0'00 . '1
91 F' a ae.
0. 0
0
00.
0
M.40 oo
O0
0.490C.40
i-c
0~
0N
0
.
0
a1. N .4 0 "SMma
0
.40
@40. .4
r-.a 0.-lo 0.40 eo A a.a no non aAn oo 0 COO e0g4 '0.-I4 n00.4
0
(
.0
:,.0
1
n
0
'
f
Z4-OO
0ý 0
0-
O 4l.
mooI
.;0 0
e.4.o i4n a woo 00.4
*@
rn-I
"mworn0.0
C-
(4. .4 lo
me
-40
P.~ a
4c 0a
a:I04
iOO
0
a
.0
o
r0.0
C .;0
.;0
; 0 (4Ii0
*
.4
9-0 M1
0--..C
.0
7
$
As00
0-mo
.0
a0
*4 0c
n=o4.4 -40-
$4 9
o
0
U
*.0
w
7
Ifl.40
ý
m
0
ano
0
.4 .0
.0
.40 0
a
0
C-O mea A w 000a NO.4 9 .0 0.
I"-I won mAr a0 a .4 .0 0
l.0 (100 .10..0 0.
40 .4
0.-I0
*04
0
0
ea
iCrw
0
C;
i-0. w.m 0
.0
M.40 m0 A 000 ie..4 .0 0 0.
40
a
*0 0
4fc 0
a
9,0- ('0. an.
X4
nwM0. 00A '0:
0
F'0 -m rq 0 M0.4q !0
.
w
41
t-a
g
r
0 0
won .4On i-C-I
:
o
eon 0040 111.-ýf 01 ('4
;
O
worn WOO 04
C,4
.0
r.44 g1 1n (40 C,4 .4
.0
.;0
'c; 0
In
me.a e
a-
-o Con
l.a ao Ie
.4
000 41 0.
.400 nO 4
2000 1 ).
(4 0 0 .
.0
u
0
0
.0
0
.0
a
l.a on
.0
mooo O 4 0
a0
ae4 An-4 0 a44 in.4 OCO o' on 40 a leO.4
V'4
in
.0
.0,
e
.I
n
.4000 nO I 0
-0
0.4 i :00 0 .I 0
.0
0 i-0 .40 'an mnnIom anon mo 10000 0 F' 0 no.q 4 ('0490-
:MM W'i0 -4. V! al 0*
M
.0
.0
-00 '
'
n mo (4 . 0
.0
.0
.0
0 0
I an0e - . 0
.0
I 0
0 oo C-
.0
331
f a'a
Cam 0nO 004
.0
In
.4
wooG
onm r-0M e. C 00. to0.
.
ý0
N4
40
N4
ý4
M
-4
.A'
Cý 0 "Onn -00C e-.4ý
1me40
"amr (ý00 CO.W-E
.0
0C
.4
0 C14('
n 01F'n
moo O-
wo 0 0 0.4
moo0 W 0.4ý
.0 C
0! 0
0
.0
400
900
.4~a~
:.-
a
94 0 uf r.
.-ac 00m'
H-ac W04o
*
.. 40 M
H.40
0-a I-on
C..4 Rom
0-a .40M
W.40 HOCon0 0 00.
an0
%n-cc1
ano
an "
In 00 a 0.40
0
COO Fac -0 .4 *0 C;0
w .0, C
03" 2.4 Ho.
0.40 so 0
v0
0
4 C
C.
H -
Uý0
.
0
0
C..4 0
0.40
a.-40
0.4
M.40
am 0 02
.0m
.0
.0 .4
a~ N-0
.400
.
.
4(
.0
0.40 ua Co09 ac
.0 0
0.49
.4.40
00.44 co .ge .0a
04 moo .0
0
0
C. C0
a%. ;a
0
.0
0OM
*0
0. 0 0 wo nO. -I *0 *0 10 0 V.4 0
C. C;0
0.40
.40 rv
4i0
S
f.)
*0
00.0
" .
0 .4
0
.0
0
.0
0
qW
10.40
;0
H
H
4
a-0
cm :am0
0.mo
.0
ao .i0
*0
*0
0
0
Z.0
0
0
4
0
0
0
T.o N.0
v.0 0
C;0
C
o
0
0
0 4 0-ao .4W40 so" -a0"
a0.0
0
C.
aC. C
c
HE.a:
*0 .0
"So .0
i 0.04
0
09
0
W'4 00.4
00 m
a-
I.40 .0W40 eon
0.40
4
0i0n
0
a
0 m
a
a.
0 .. Sam
.0
:4 c
00o,
Ho.0 0
0
.0
04040 0 -aen
10 0
0
.00
0
40
ý 0
40 4
io
4-0
9-a0 won
X140 .
.004 .0 004 004
W4
0
.400 0 0. .4a a o ''we . 0, C0
0
a &0
0
*0
000 00.
W0
wO.4 00
.
0
40
:99 a
0040:40 WOO 0-, 0
SO0 0
00000
.0
HOOa9
.
0
0
00 .0
0 0.
04 .0
-4 egMo .
a
0no
0-
0
0
.04 0
H 000
'A
C. 0
C. 0o
at 0
o
a. 0
0)
o * U0 C l
* 0a
0
-
-A.
0 C .4.0 40 O"C' n" m N Aso" con 0 0 a0 40 Wo00.0 a .0ne0 0ý41.e 7 0 .0 .:0
nm 0 m 0 n a H 0 0 0 ta ý0.4 Ocý4a v0 .0 .0 .0
A4
0
0
0
0
wee
1v-c
e-"'0. v- . 0 0
0.0
N0 0 000 -0 040 -aom.4m 400 v-o 00. an 'M n . . 0 0 .0 .0 0 0 0
wOO n-c ano CO-a MO-a w. 0
e w Z40 0 n won n 0 0-o 0 9w 0 in00.4 0.4
m M4 0..40 M nHon a cc e W0.4 W00.4
0.-l "eon -0 a.~~~~a .. oc .
no
0r 0
e.-ao 0 0m W( H0 0 a 4" " 4" a 0 W 0 v4 0 00.04 . afav-.. 'a9U .oC;0 C;e
4
0
nan
.-ýc o HOo .0. 0
0 eorn c zoo 00.4
4
0
a
o
0 a om C 0 0o.4
0
0
.4
24
a.0
N
0
v-c 0-e .40 aae -c o nn HO-a ne -a no. -1. . .0 .0 .0 0 0 0
00" na aen 00.4 0 .0 0
0~f 1in n H0 00.4
.0
94 ana-4
v-c-a.4
~0
0.
W
0
0
mo .o We.0. 0 0
m0000 HOo We00 .0 . 0
o 900 w .0 0
a .4~ Wea a." .0
.0
0
0
C o- 0
0 C 0
.0 0
.0 00
S03 e
.0
0
e3
0
:00 nec
4.0
'oC
a-
.a Ce . .0 0 000"00
-a .0. n .0
0 a
94
W4 0 W .0 0
ao o .0 0
..
0
0 0
a-0 co t a no-a .0 0
0
. w
CW.40 0. 9!0
0 n-a
m
O A0 0 ano,- vca .. co .0 .0 0 0
:"0. . .0
a -.
.
40
00
.
M
.
.
a
0,
0332 .
0
0
aC
. 0
3 0 a
0
.
-
a
900 no.0
m
1.40:g a .0 0
.0
.0
.
00 a
.4
W
0 0
no 404 .0 a
0 1.
100 007 0 0 0
w.4 0q0 04 0 no 0. 0 40 c-a o.4
0
.
n
.404
00-a . .0
0 0. v.4 .0
.4
0
0 0 "e ' 0 00-a .0 4 0
0A M ".
0
H4 o o n .0
0
H.4 0 0 a-Cx . 0 no2Cn wom 00 :g ac e 0o Coa .4In0.4 H.
eo
14.440M.4
3
0 20 m nm m " a 21 a ano.4a .0
c 0
.
0
0 H 00 HO-a 4 0..0 C!0 1 0
0 H 0
.
0
I
.
ý4I
I
g
woo 00 P-C .4 .0 !
0~
40.4 *0 0 @0 ;a
a; 4
In
0.40 f- It e .~ mM :'00"aA *'a00 0* .40 .4
W 0. a r.04 8.40
.0
In0
M
a mo
.a N
om p00 4
~0
.0
'0.40
" 40 10
.0
0
'0..q
a0 4
eo0n. c0 04
a a
0.
.0
to0
C0
4;0
C0
'.0
w14
'0.40
.4.40
Is" 000
N14 14 000'
0
S
a
.,
H 0w
s 0C
4
0
0
0
!0
0-
0144 a
Von4 1 ao
.
'04
C;40
in
a0
0
:0.40
%gom We
ý, o. N
41*04.40 "'0
4
.0
0.40 O~ t o"
i
4 eo.041-
*0
so
.
!.0
0o
. go
.
s
4.0
"s
n
40
'0.
XCi0.
0.
C
;0
C
O
0 oMM 1000040 40 .4
can 4 0 00.4
4.0
* 40
0 0
0ý
0.o 3041 *a0 a 4.
440
S.40
00. .0 .40
a M. 0 -6
c
2
0
0
?04 0
lo0 0 0
0
r4
41
W* 1ý0
0
0
0
4M.40
p04 r. poe.00
0-ii0.0
0.40 0 03" mM .40 M00.
w!
0
'0.-CA
d A04
4
M0. a *0% a
4a
.0
f
40
C;0
"
.0 o .
4*0 i
.4
-0
'0
p.. 0
C;
3W4
0
0 M
.4
*00
..
:90~
Id~' x in
m0
0
0
1
i~A
41.
"-
r: 0
0 1
4
$a 9k @4 @
.
0 4
0 04 0 41 mo 0
0.40 .44 M '000
w 14
0
L
c
to U
0 'W0. M a. C!.0 0
14 M
p04
41 0.4
'.
4X1.
0
'0.40 9om mo 00
a1
00
.0
0
1
p3 4
.4-40
a,0 a1
a.
.0
'A0 1
.000
.041ý
0 m
N41 0 MO 0 Woo 0 410.4
a1 0 0 " 0 0 '0. .4
4.00 mo
0n
.40 0
a4
a.
a
.0
.
C;0 0
a
0
41
g
;0 0.4
C40
.0
0
.
Wpa0.4 .0
.
.M4M0
:a 00
41
0
.440
.0 a
'0.40 wonm woo0
9"R oem aos 'I
'0.490 i-a wo0
p.40 io W 0 o
ý41-40 n 41O00
0.4 0 WS1 0
a 00
'0.
r.'0. .0
.0
0
10 '0. .
'0 al
0 '0.40 '0.40 M1 aem ion 0 0 0 0 0 0 '000 ',4000 '00.,4 00. 4ý10.
.0
040
.0
.0 0
0
.0
'0.4 "ao" 0 00. 0
0 0
0f a m
goo a '00!
.
'0.
.4.4 o
i.0 10
C; 04 mo
;a m
a
'4; 4
'0
'0
'ý0W.4
a
0
a
.0
C;0 0
.ý0 0
.0 0
0
a
.0
0
.0
p..40
=no4 pa .0p '0410 -'0
414..4 0 00
40
a.4
p..
0.
00 m 0.
.0
0
0
.0 0 1
in
rý40
wo' I4M 1'n ,4
ftI04'
:n.o 0.. a.
a.
9
,4M 3, o
9.0 404 a. .0
.0
0
So0o44
0am
290 M00 .04 p. a
.0
0
'0 '0.4M
* .0
'0140 mmno 0 000 '04 0
..
.0
ý0.40 momM a
m4 0 1.4 a.-
" 0 a"o .0 '04 0 a. a.
'0.
.
0"
00 A440.4
.0
0 '.
*ýa.
.
0
0*.
.
.
o
.
'0.ý
*.
p
'00 . 0
i
0
a
.0
.0 4
p.04 W04
.41
4
.0 4
iWOg
'0.4
4104 WO.0
'0.
1!0 .0
0
04 0
'04 n'00
14
M1.4 0
0
0'0 '0
.0
I.
'0.040
r: 0
'.4
04
0
14
.00
. 0
414
00
.
.0 F
04 0
0
d
o
0
'.
400 '00.4 on.
a
0
0
10
R .44
;0
0
a .
0
0400
0"
'0.4
.4
Mo
4!
a
wo
0
0
" .0 a
0@
0.
a. :~ 04
0
1!
!
0
a
0
4
0
4
.0
.0
0
'o
1!
A.0
i.40
'0 0 '0O.4
C; 4
0
4.0
a
'004 '0
.0
1.4 0 aM Wo 0 410 '04
.0 0
'010 40.4 0
C;0
4.0.
'0 0
0
.
ft.-i0
.0
0
4 ..
'N001 '.40
.0 41
m 41 a0
0
4.40
C;0.4
440.4
.0
jd0
1cm
410
0
C!
0
0 i-C0
00
6! 0 0
040 : m 0
.0"
.40
'0 in.
00
3g
00
M
0
'0.
.0
'000 mO o a .0
M.
'.0 4 04 '.4 pWOO '00 w a. . - 0.
4.4
0.40
140
0400 40.4
.4M
.ý0
inM
44
a0
"
0n
0 '0.4
S
0
0
.4
.0
ac
,; 00 4f
~
'014 '0C0
00
..
00. 4ý 0. .0
41
0
0 .4 0
M0
'0.4
S100
4.0
o
04 U4
0
w.0
".400
.4 .4
0
'40"
~
4
a% m. 00 0
V"
MO
014.
.4041
.0
0o
41. O40
.0 '0
,
e
o0ýý 0
W.0
n0
a
0
' 40
4
104.0
ý0
f
I
0
40
0
ftI C!
i
0
'000
.
'03".
.4 0
0
0
2
0 4'0'0 o40
0
Ma
.40M
90
-o
m
ýf
0
o
O4r-04 0"M0ý
0
0
0
.40
wo
%3330
MM
:a NO" 0
ýn I . 4V.
WI
('4.
. ae
.0
o
.0
.0
C
*;0
0 WCoo"a000 0
O0
.4 19
.
03
eom 0
0
410
0
('.0 N 400
0
404 C!
.0
0
0
4M4-4
MM.4
.4.40 0 ;N 4
goo
moo
ac:"cc
.0
*
o0
.0
.;0
0M
0l~
C0 a
.4
.00
r
0
0I
4.40
1- "0 moo moo
"Co
:
1'0 e .0
.0
.0
Iw
0
" .0
( *
0-C " '.0 04 P04 '00. 49 " 4e .40.4 WZ * 0 .0 00
0
.4
M 0...
ano
-.410 0 44'1
* w.4 w0 rNE' S4. al em4r449
r
z0 .0 0
a
t 0
M440
'0.4
";a0
ONO
10 .ý
1%. M"04
0
*0
.0
ago0 00.4 00.4
.00 04. Z io.4
0
4c 0o 0
4M
A
0
44 Ie
we
0
'0.4
0,00 r-5 .
momo P10 q 00.
2,00 wo .4 .0
44
400a
.
so" c cc .90.4
*e 00.4
04
.
0
('4.4
4.4
"90
'49
:amcm 0m coc 00a
oem 000 40.
~1
III &0 a0
MN o en
0
i4 0wo
~
o
t. in. 0. w 0
N
C0
to
:0moe
C;0
.0
C
4
.4
C) )
0
C 0
0
0
.09
49
n
.0
0
0
0
0
44 0.
.ý0
0.4
.0 9 00
WAO mo N.2 00.4M 0.o 0 .0 e .0
0
0.
) .0 0
m
.0
0 0 4m
A-.4
.0
f-o mo in4. '0. ('w
0
o
c0
0
00
00 cm
0.
O0M
n
.0
0
0.40
"
*90
0r.40
00 0.
0M
0.M
C0a
.;0 0
0
40
90 a
04ca
0
:.400c 00.4 '4.10 %0 0; 0
.0
.400 004 Do o0
4I 040 a
m
9.40
0
-
0.
C;0
0
calf0
.
0
ý0
40ý
0.40
0 f"
o a
en0 %.4r
en
C)
4oom
0
.0
0
0z .
000an .40 0
.0
m
00.4 400C0 0
.0
4.40 oeCm moo 0
.0
00.4
r- 40 WOm 00 0oo 00
40.4 00
.0
.0
.0
0.40
0
0
a
00 40 00 .0
0
*.e .4.0 000 400040.0(4000 0 :n . .0 .ýV40 :M0O 004r.0 .0g ý 0 P 0 0ýc o 0 0 0 0.an.0 .0404 0.4 20 N# 004;0 4.'004'0. 40 0..40o 004 004 n.4 00. 4 .0 * . 0 0 0 0 00 0 0 9. !a.9 C, 0 90 . !0 zo 9, . Cc ;0e9. 1' 0g en a
N n-w 3344 in aW
I
*
0
.1 ewe
wwo
.4M. 0. .i0
.4C.4 0." '1;0
14
0woo4 .4 .0o
.4 *igr.0@ mw OO 6,xci .4Cme'mie HMr REc 0~ .1 00"0 2O
~
*
k
-04
X1
0me ear AMO fe2 .400If 0 . me01141i-m *C.'0.4rnC. 0.4 .0 .0 aa dju
"00n
in00. i-e
aM
.
a0 2
0.aI 40.
.;0
.0C
.
NU.0
0
c
On *0
f
.0404
040ý 0
00
i-.4
N40.4
"O. 00.
0.4
14.40 Ma e"SO .0
.0
.40
"149
wago .4m.0 O .4 M 0 e..4 .0 a0
40 . ccmoSo
0
we M.4
0.40
.40
O NO4
Agoe i-e
0.40
o.4 N mo #t .00 a
M.C .4: CC N 3.4 CC 0
:
Ma9 44mCr 20.4 400
s
An0' em00 i-c
#w ear
1-.4s moo CO. .4~ ".0
C NO
C:0
WNS
:N94
omr
oem
gem
0140
XWOi
a.0
0
0
0
0e
0
-".
C. Og
CC4
~C4
C.,o
4
1 .0
l C C.40 P" tw ion worn .m 33 me: N c C 0. .;0
.0
VO4~0. on c!-i
.0
.0
M.4 0 worn woo . zo
'ý0 onq umo 0.0 1 00
24
10.4
Cn
0i-.4 C; 0
C mm
.0
0
0.4g
N.0. a .0
00. .;0
f%.4O corn
W4C i-C CC
"0. zC
0 . 40.0
MO. .;0
on wom0 No. i0C;
W.
i-C
0 4tm CCý
0-
0I 0C
a .me t4e0
04 0n m
010
.0
0e
.C
0
04
0
0
"ee eAso -O.002 C. 0. 0 0
:so cer 00. 0. a ;0
moo . n. 0 .0
Nina
.414
"WC
a~
4. *1-2 o
"0. a
wa
w2.
C.40 oem iC woo win w. ' .0 0 0
t#e
104 '1 S
a
.0
Wm
f-i 0.40 -
0
.4C0
CO
i-f.we
w4: Vn4 0.4
W
US:0"o
-mel
o
NM
"
e-.." 0
0..40
r.
0
Cn40 .44m aee '0. r. A .0 e;
.0 4I
0
mo9.ml r'0 rn i-a."a w1 0. :0 4 aC 0.4 . O. e 0.4 .4 .0 *0 .0 0 q; 0 0 0 4
e
."C ern we ec '04. 0. 40 e;
i'0. *0
0
1-.4
'0.
oo
04 go
C;C C
nia
.
.4
0
W.0 aoCM .4C 0 . N 0 r 1; C C;
. 4!.4 M !a 0
Ho"" 141"o
0ý
wo
-0M
C0
.- e M
:"A
:"RM.40 u-em mom :'00.4 w S0.4 COO COO
00 a0m
0
I04
'0.40
.0
Ag Cin0
.C
.;0
g .4 M.40C 4 mOn "O :oo 3:.C N .4 . .C .ý0
4i;
oe0 foect. w0. .0 0
0
w.
.0
"04S0034
.40
00. 0.e4 w w.
riM
0. fte 0 :mom@0@
wo
.0 C a0 C 0
N00C0
0-C u04 C4 .4 .0 .0 0a
oomM
0"0 earn
I
.0
a.4 0 worn W_0.4 wO
.0 0
C
0a
me
Naem acorsnSmM "r Me
0,.~o
0
0.40 iS~-am Man e
3
0.4
@
OI0
91
.0
.0 *
,
~
0
won
02mo
00 .0
w U)0 9gd 40 .4.49 O .onia wm 4 M.u mo Weo O 4 we9, o 4 . I .0 0 * 0 0 00
C q
40.4"4 N4. .0
m.e
.0#.0C0~ 0o em 0or
'000 .0
j00:.
wooC 004 an 4 . 144. .0 .0 4
A0" M0
.4
A.q0
49
0
ela 9 S 0 W.40 Mam mmCon 4 00.0o iO 4 040. .40 wee i-e
wo .4140 .amem * .4.4ý 4ct C. C; 0m a
.. c
0
W 004
.0
~N.0
0 wi
3
~
a.4
W.4e 0 00e
E*4
p.0. 14. 4i0 a0 a
.4e'4e aam
i-cm e1-m .0 .09
4@.
0
-C
0640
mo ;0
mi-C 4m va o 0* .0
.0
".4.40
e0r
.01
mP-0 M4C .4
me. .
0.dwo
m .4 .0
0 .W40 Xý e0 .o Con 0.0'C so "so "39 4;0 .0
0
999~ ~~
~
o'
oM00M
~~~~~~~4 .4 r
g . C.I o
335
.M
'CIn
0.t 4W 1 a:92.
Mg0M
,
oa
0
0-0OO
UNn9"R9094
111 00to n3 *0
0*
.4
~~91~6 04 91 ~~ ~t0 - 44 0 9
0 9 ~040 ~ .. ~0 C;0 91
0 0
0
go4 -.4 4ý Z .4.0
14.0 U; ..
144
.4.r4 3 -
.
.0.4 r.4 4:0
& 5
Oa 0.ý .a
fff (-0.4 .0
0
.0
*44-
0
040
209 .400 0
91
1
m
lb
o9
: ~
.
aU09 %
000 0. 0 0
Cor 1 .0.ý4 0. .0 0
.4.f 00.
ft.
.
"
40 .i
10
.-
0
W.49
VO
.400 100. .4. .;0 0
0tO40
0
00oo " -
91
91
f0.40
00.
.40.
SO"
9
0
W.40 R0a.10
.0 00
0
4
29.4
%-.9
0
000 ft ftO.4 .
.0 91 0
4
AMR
ftO 1-0.-4
0
0.40
000 0 00.4 * 91
0
Ns00 00.4 .0
0
.4.40
fO
~
f
0.40
0li
.0!0,;
r
0
00.4 .
0. .
o Ag U P4.4 .0.4
0
.0
.40 WO
.4 0 00.
Of
WOwn 00(1 3"m f 0 0 C. f-.s0 0. 0. . 0 0 . 0 00
.0
0
000
W.40
:tt0v .
v0 v.40. 0w
00.4 0 .
0009 00. 0o . .0
.0
a
0.0.
ma t." .0
0
94
.
0
004 0
0.40
ag0 M0O.4
.
0
0 00.fPlf 14
40
.0
;0
0
40
V0.40
0
0
.4 0 (-. 4ý 0
R4
0 0.
2*
40
0q;0
. :004
0
0
.
W
-
: 0 004ýý .0 0
0
ftm 0
40v0 a4 n R11
.0
0ý0.
.0 4;
I0
(-0 0.4 0.
-
X
0
W.40 .4
0 .4 s r* MftO.4
fto t 40 0M - 0% Moo. :s 0
: f0"
V0.4
.0
0 0.
.0
000 r. 0.4
f
0
moo0 Wr 004
0.40
-O@0(0 a a
T
"0.0 000t .4
V00 .40g.4
444.O 0
.1 .00 004
.0
IVO -0
N.40
0
C;
0
91
0
mom.
Woo 00.4
91
WO Man a0
f78
Is40.4
~ .01~
.0 91
0.40
*0 0; 0
o
*0
C;0
.4W. 0: 1
R 0
.0
VI00
0.
0
WO .40.4 f.4 .0
Ut .. 0
.~ 0qý0.4
0
p040 0.oi .ý0
ftt
moo4 0U4 002 ft f-0
WOO
0
0 W.1 ft.. 4i0
1 "2i10
f.40 0.
0..
0
a1
moft
0 0 00 CO. 1 00.
fV=t 00.
i000
.0.40 .0
0
V00 ; 000 0
00
.0
0
9.40 R
10.49
0 ft.
10
0
m.
.0n 0 04
0.40aao 2 f-Of f00. 00 000 0 0
0
0
HE4
02
fO f ag
N:
00 wnt
H"- o
0
a0. .ý0
"
ftWO0.0 R 4.4 ft. 0!C!0n .0 !0 *. 0 90
ft .40 Of 4Ot 2000. 900
a00 0004
ft w0.4
.0 0
"0 0f 000 0
00 .40 2g 3 24. 00.4
I -OO
0
-!
ftX'2. 0. e! *;0
4.
0.4 ft f00 fta.4 A0. . 0
0*
.0 0
0
1440 4
0
`50. 0. 1 .0
Wtft oo.. 1.4 0 .0
C0 404
.4-
91
v .
0. 40
tO HE0. at .ý0
f-sof f
0 v.4 0.4
-4-o
0 0 . C;0
2.W5 0 0
0 91
00
.4. mom
0
0
.4.0 00.
*0
W0.4 0. ý .0
.4
coo0
0.60 cr 00.0
.0 0
0
Ong 0 atO f-SO 9 0. .0
9
.0
IV0
V.4Of 400 M0 9 0
a00..4 040 "0. 0 ..ý0 4U 0 0
V000
a
.0
0
0
.0
0
@
0
ff4 0 .0
0'4
o
*
0:0(.4f940f tO .0 O
[email protected] 0 . 0 . i 0 0 .0 0a 0 0
~
a-f 3sof 004
00.
. .4*
S0
*4
WU0 0. *@95 ý.0
.4.
ft.4 0oa 0.0 f nft :. ý400 voC 0ý t.
W..a 00a
00. "0.4
0
.4.0
4
0
0M V0009 w4 o .0
.40
%-.
0
~
.4
"Mao" f . .0
Xo
0
90
0
4!0
0M -- 0 as. "Z *
Avg0
9
on 00 f.4 Vo ýl m .0. 9 .0
0M or 900 .4. . 0 0
.4~~ -4 00
.4
0.0
M0oo 4 . 14 0 r
0
.0 .0. P0.4 00.
t a44
-1 0
.0
Z.0 0 .4 2 0, -. 4 C4 . .0 0
A 4
04C
9
&~40
Sa 4 14 .a
0.
*0 10
.0 0
Mt t V"0 0 ft.'(-. 'I C!* 0 * . .0
"I
2 ~¶9
. 0 13
.00*@.
'"W8
ft00
0
Soo
:sWOO"s
:o"~.o
*
OR
orn
0 "so
0
.
InWI In bIin in 336
0
0 00.4P4-0.04 Z0 0 0 .
I;
.-
3.
.c . 0
0
.0 0
#400
0-40
W.40 0
Um 0n 105
naF, asmF 440.0 ý r
7
*0
190 .
09. N *0
00
#4 Owe 4eo
o
M.
4440
54
a4
U31
*2
0
44
0
*
o
0400
.
0.4..4 0 0
U~44C
A4iA -
0
0
M
4J .4
0
0
47 .4 00
404
i.
r.0
440
3
e *..0 9 .. 0
o-o Z 4030 404
0
*
.
47 "a
0
0tq .40 o~ 4s O
0
0
'040 040 no"F a4F -040 000 00. #4,. *0 .0
ý@a 0%.
0 40
ý0.4
90 0 @0Z .0 -0 0 .
M00
0 0"
0
3
0
4;
i
-.
0
4a
0 0.4 .0" 0
0
0
t0 f.0
2.4 42w40 .-.40 00 M2 "q0 D0 ..0C .ý0
%4
4400 M 42. 4 0
4!0
F, N0. . 0
0
0
an
42o
no0
0
0
0
WO
g
.4
.0
0
0
0
337 0
#. .0
0P
0.4
ai-
z.
40
.0
. on
0.440 3.424 F lin. !in4 0F #0 t4, 204 C!.4 .4. 0.
4
.
-5
4.4
i-35
so0 .;0 0
so0 .0 '0
$040 .0 C;0
.49 U20
r-49
P400
$300
C;0
#442 ago2 40 44404F,40. . 0 *0 .0
40
40
0O
0t"
0-
0o
0S
044
",4 40
*.40
0.
:
wo
W,
'.4 0i-o
:04 .
0
0
0
0an ".404
0044 0 .
3i
v 4
.
.0-0
0
;no"Mn S40o
moo4 42i-0 F,4.0 0.00 42 #404 " 1 n 004 " 2F 0i-OF,:s i-v i-00 0i-0 040 42040 420.-1. 042.4 -en1.. .44- e .0 .0 .0 .0 '0
0114.
.0
40
0440 -
0
-
0
i00.
0
.-
:04 000
1
0n0.4
0 0
0
0
Z0.4.244 #40 t- 00 0 0 0.4 00. 09. i-.
0
40
000
.400
00.4
0
.0
0
0o
:
0
Z
42
,004
0.! 0
0
0-.0
04
.40F,
0
42
W-.4 420 R .0 4. ia.
00
0
n"4.4 *ý !0 0
0
n 0eoo.4
,ci-
.
0.4 00nft .10 ,.4 0.
0
0
40
m
; 4;
i-A.v0 #400 zoo i-. i
@.0 .o
F
or.4 a0 0
0
00
0
ano
? "M
.
i-am *
0.
1 .9C0 0
.0
yo 0;
0o
0.
40
.0
4200 0.4 0 0
0
.4 4. 0 0 0 .0c .0 .4 w00.4 F, 4.a
02 a.
i-9
.
40
94
4444 *on4F 1400 .0
M. 00
90
4..
in4420 aFF .0
r-
-3 F 2 0.
w7 0047M
0
.49.4
2.4
no44
F
.40
0
ý.4.0
'0F 0
ms. 0
0
0
ftee 9
#.40
.0
r.V5
.0 C
.4ao 6230,0 00.4
C0
in.0 *0
.4
.0
47 90
0
o .0 4
.0
*O
0
*0
*Oa
204
0.
C;0
.O0
.4
0.0
47.
42
I,1
:s-,
Fee
0
04
11O
W.40
~
V
O in*
fl.40
M..4
0 0;
16
F, 4q0 0
ý10
X0.
0420 . 0.4 49
P0.9F
0
.0
%
420
"-*A
00 00
10
A4 0.
4.0
2..
0
C! *0
:90 14o.a
o
@440
w4 .0
@0
0 o-e .
.
#..0
*
MVRUo.
00
40
oo
0440 020 sort MO000040
0-4
40
#4* -f .0
FM
?,~ "-19 :j
.0
.0.
C!
0. .0
.0
0
00C4
0'
.0
.40
0 U
M-4 ewe
.0
rw 0n
I
.
.40.
#40 .0
Nnn
0
hi0
.0
"
09
S4 !0 .
0
0
c.
44
in ý4' 0
ns"
o
W 0.0
ugon . wo
0 Sao
.40 .0
#4 .4 0
*.0
e-om i-c *0
0
4*
0
am ft
.0
.
0
4~.400
m.4, 0.4 044 c2F o..ao490
44
0
0.40
.4=4 *g
444 .40
4
OF ;go.
-
gsn
'OF, *00
f0 W
i.-4
-
a
4i.4 40.00
0
he
4.40
.0
"am 140 *0
0
9"R,#'e~ W149 NO 21 -eec F",o
-4
s;-021
F,1a
a
I-a :0 *
a0
MFm040
4 .0
o
.
a
-.
C;C;
.
4;
12 HE"I
~0 *
.1
0
00 :*.
#4 0
ao
*
0
*.
4;0
04
#10
02
Z#4 ~
0"
"as
00. 4
14~~2..
j
0
.0
0
0
000
0
-;0
0
i**
19 0 z 0
. 0
0
t0
4t
C;0 00
H'S 104
0
*
r0
a, 0
O 19.0 0
0
0
a0 .0 0
00
Ago
"a"
14.0
#46U04
0*0 0
0
*0.0
0.40 04qr ý0 0.4
00
.0
.ýa 0
0
m
0
O
1
0.
9
"14 *!
o
0 14
40
0
"a
4
.
40
.#4.4
---
6
*#40
0
0
"M.4
0.
t0# 00
ý0
#49 0
# 0o.
0
0
0 0; C
0
0
.0
00
C!
0*
00. a4
#4.4 .. f 0# a4. # 0 0. A6 .0 .0 a
ag
90goo
Ut
0
000
".400
#.0 00.
M0
i .4
.4
R
0
#4
.
0F
10
00 0
0
00
#0 60. 0.r C..0
.4 0 104
"g0 00
M600
*oo
000 190 0.4 40 0
in
ný 00
"c0 0
0.4
0 0 0m
0
#.40 004
000
"1. 0
4
.0
0.0!
01
.40
1
V4 @.0
.0 0
.0
C!404 . .;0 0 0
0
00.,4 0 4*0
0o
40 0.
a
0
A1 0
0
V!.0 0
0
O.40
0.40
#mo
0.4a 600.
.4# 00
0 0
rto ý
0 .0
¶
0
0. 44
00 0.-
0'
9
#4.4
10
.0
0
.010 0
0 %.00 0 0 0.co A0 . 0
0
00 no
0c . .0 0
.101 0
90.
.0
#
0
X4
a4. 4 00 1
10
90
0.4
0
r- 0
.444 #-00 0
0'
V #40w .0
0
.40164.4
.0
gel
0*
0
.0
0.4 0. . 0 X .0 Mc0 0
0
.0
.0 .0 0
1ý
00
0
0Ic
0
#00
.00.4 0.
.4.4 - C;4 0
100
0c
000 00.ý4 . .ý0 0
60.4 6.4 *ý0 0
.40
10
0
a0
#00
.400 00.4 1-
1
0
a0
mo 0 3'0
0
..
00 0 0
a
0
0
00 O.400 0.4 6. C0.0
In. . .ý0
.
0
a
"
c
0i2 0
:9~33 W91a.4I 1 !,09.
0"
.0 I 0
I0
0
0
0c00
.40
0 0! 00. 0. .;0
144 a. .0
0
.4
W)'
.0
9 .
00
4 0
0
0
.0N 0
194- 0
1
0 #00
0 0
W.0
1440 04I
.4
09 C
.140
0.4 1.
90
00 444 0
.0
?
o 4
c
000
W.0
#
0; 0 004
.04
.400
0
00
r.000
0ýI
4
.n"00
0
.0 ' i00
0.0 104
004
0 44
:W
20.40
-a0t
ao
0.0 0 r.
a4.40 0.
09
a
'I1
"02
0 .4.4 o 0 .4. 0 -!, .0 0 0
Ma
:.00
a
.0 10.
#14
a44
.0
~;00.4ý04
0
0 0
. ;0
0
0
.40
.0
0n 0"4 40.4 0 4 0
0
t10.4 .0 C;0
.0
11 281
0
0 4. 0
OU
W
0
.0
.;0
0 A#4 #4 0 o 0
t
030. a .4. ".0
.0 .0
0 0.0 :"4 #
.0 'i0
It 00.4"c a~0.4. .0
.
M
g~ .40
#0 aO 00
C 0
mA 040 ,400 4 00
u
Z0.0
C0
a4 .0
.
200
0
6.4O m
400.4 00
.;0
14
oft40.4 .
*.0
#.0#44
.0
0
x
#4
0
#.0
#404
14604
40
400
4 .0
uOWS
i0.4 2 # .
0
,
4.
4 . 0C
0
*
a
0*4
#.0
3
m. oo 0
.40
H
mo
V4
04
#.4
i
a0 .0 .0 4;0 0
C0
#.4
.U
.4*
.0 t0
#40
: .. ., 00. 4
#0
C
*
0o
m40
09
I.0
o0i0
-04-o~4 ON
* U
4FH
4.Q
0
-W
o
~4
K44
2itst2
4I
.44
:
0
0
01
I0 04
n
ago "
0 ..0.0
4U
4
ý
I
S.
og-.
1 0 a.
0
*0.0
i.. N w 0
1N
0
9:0 x1e.. 0402 IAN^ go0 00 0 00. 4. 6=0. g.4. 1.0 4. 0. *0
10
o144090a
0
g*0
?
0
T
Mri
PCe oe. a. 01 0 I@n a0~ 0 0.4 a0. N a0..4
0
j
.00
49 *0 S0
6
v. i 0
0.6.
0 0
.0
.0
0 0 0% 0 00.4c 0 .
.4.0 : am WO6
N.40 @00 NSO.
.4
4 -ý .0
V@ @0
.
0
0.4 ^a
M CIA . 0
.0
.
0.40 MOM NOa.2 . I .0
v;0
0
V 01@ 0~
0
0
*Q
0~
0
.
l04~ 0W V=W on in 0..4 a0 M9N ean 40 NO.4 AA401. i* . 0! 0 .
win 0
*I60.
:.4x .400 690.
900
N30
*01 0
c
0?
0
0
.
.0
0
0.
0A1
.0
.0
0 0 0 0 0 000.4 .
:01 0001 .0 M woac 9 540 in*
006
.0 M
Ncc24040 0
An.
@0 gv4 .0
0
0
.0 If-00
.0
.4
0
4
.440 @00 woo 'w0.4
60 @of, i-c a@00. N.
0
.4 .4 .
occ. 4
0 .0 ;0
.9
. g
14
x
0
0
.
.0
:00 .0
0 0
0 oh
0
1416
~
000 inn hi -:A-
0
M9A9.4 0 0
0
0
-,.M .0
g4 0
00g.0 0
ý0
0
09 CmV@
00.0
4
go .
0
144
4a
401 40~i
0g. .0 0
0 0 0 in 000 00.4 0 g9 0 0
@ 0 69496 000
0.0 NCO
a0.4c
N 4l 0.
.600
NCO4 009
.0
0 a4 0404 . .0
.4 nn6
NO.
1!0
0
0
OO 0 lAe .4 w .4 0 . .0 0
0N04 M
in.40 0OM 400 NOg.4 0n .0
'4
A..1 1.0 0049 44
*
cc
10,
0
..
69
0
.40
w
aw 0
:V4 00
0.
04 9
o
46
.0
.0
0
Ao
U
0
@
=I
0
0
16- 0
*
0
-*0 .
z
006
0
.0
0
0
...4
0400
r..40
N0~
In0.4 .0
0
0
.0
@10 a0
N4
0iie
0
oNo 004
1%N
0
O0O
@0.ý 4 0g0g
0.60 g.40
4.00 @04 0
10
0.4
=0 000
.400
r-0.4 0 0
000
orI
-4
.4
6d .4
-!4
@
4hi
* .4 14 .4 6t U
0
0
006. 4 00 !4 104 00
@
S
M0a 0
mo 000
0 w
40
0
**
0
0.
00
NiC
04
@6=4 69.4 Ng0" 'c;0
009 In*n 1sf.4 .c0
o
ý0
M*90
060 NOC
Rom4 i.4 6=00
-906 01=60. moo100AC 00.4 410. z0 .0 0M~ 0n
4
0
.0
0
0 -
0c 4c0
.4
0
.0
0
(The rev.40 @00
940 P026 0. C0
N x .0
4*
0
0
.0
thi0 69ag0
0
10000 in06 0 0i0 n=. .40. 1 .;0
.40 .00 0 .04 0
4
.0
0619 O00 '.4 Z0
.4 .406 @00 00 .0;
=0
0ý
0.40 ,00
.0M0 6900
.0 ';0
.0
.0 C;0
.
0309 is
blank0.60
APPEIIX N
RESULTS OF STEPWISE REGRESSION MODELS
34 1
I I Stepwise Regression Results for Dependent Variable DT R-square - 0.69537191
Regression Error Total Variable intercept LN PASS ST DEV VARIANCE QSKEW PVFWDISM BULKDENS ASPHCONT TMAXDENS TRUTHICK PCMEDIUM
C(p) - 10.62977138
DF
Sum of Squares
10 998 1008
177.01685943 77.54743560 254.56429503
F
Prob>F
227.81 17.70168594 0.07770284 -mse
0.0001
Mean Square
Parameter
Standard
Type II
Estimate
Error
Sum of Squares
F
-27.46644280 0.23553798 -0.38563620 0.01585799 -0.12211429 -0.00015742 0.05792197 0.48164218 0.10870507 -0.10545457 0.01826960
3.29647395 0.00819384 0.07995716 0.00308606 0.04019436 0.00000663 0.00732821 0.03505400 0.01427616 0.01478513 0.00487709
5.39439121 64.20727273 1.80749794 2.05175020 0.71719913 43.75173900 4.85431126 14.66934495 4.50519510 3.95291344 1.09036693
69.42 826.32 23.26 26.41 9.23 563.06 62.47 188.79 57.98 50.87 14.03
Prob>F 0.000). 0.0001 0.0001 0.0001 0.0024 0.0001 0.0001 0.0001 0.0001 0.0001 0.0002
I
I
I
Stepwise Regression Results for Dependent Variable HUl R-square - 0.44344981
Regression Error Total Variable intercept PASSES LN PASS Q SKEW PVFWDISM BULKDENS ASPHCONT TRUTHICK PCMEDIUM PCTSMALL
I
C(p) - 15.64541534
DF
Sum of Squares
9 999 1008
5.52039498 6.92835310 12.44874808
Mean Square 0.61337722 0.00693529 -mse
F
Prob>F
88.44
0.0001
Parameter Estimate
Standard Error
II Squares Sum of Type
F
Prob>F
-4.83864348 -0.00000602 0.04763292 0.04240207 0.00000750 0.02332974 0.06654293 -0.00686042 0.01405719 0.01593107
0.36895435 0.00000193 0.00511533 0.01139746 0.00000190 0.00225872 0.00585521 0.00441094 0.00136412 0.00320824
1.19279690 0.06737175 0.60135626 0.09598929 0.10857124 0.73987613 0.89574281 0.01677660 0.73647272 0.17100947
171.99 9.71 86.71 13.84 15.65 106.68 129.16 2.42 106.19 24.66
0.0001 0.0019 0.0001 0.0002 0.0001 0.0001 0.0001 0.1202 0.0001 0.0001
I 342
I
I
Stepwise Regression Results for Dependent Variable A2 R-square - 0.41058484
Regression Error Total Variable intercept PASSES LN PASS QSKEW PVFWDISM BULKDENS ASPHCONT TRUTHICK PCMEDIUM PCTSMALL
C(p) - 14.23994920
DF
Sum of Squares
9 999 1008
2599.55512808 3731.79188451 6331.34701259
Mean Square 288.83945868 3.73552741 -mse
F
Prob>F
77.32
0.0001
Parameter Estimate
Standard Error
Type II Sum of Squares
F
Prob>F
-128.19288355 -0.00020221 1.06923425 0.72186403 0.00021697 0.65517030 1.02529012 -0.18597876 0.34876876 0.44414649
8.56281056 0.00004482 0.11871819 0.26451583 0.00004402 0.05242113 0.13588960 0.10237046 0.03165892 0.07445790
837.23444724 76.03712417 303.01427682 27.82014550 90.76029514 583.50910627 212.65375400 12.32904070 453.35139590 132.91768790
224.13 20.36 81.12 7.45 24.30 156.21 56.93 3.30 121.36 35.58
0.0001 0.0001 0.0001 0.0065 0.0001 0.0001 0.0001 0.0696 0.0001 0.0001
F
Prob>F
10284.67626082 423.08 24.30896292 -mse
0.0000
Stepwise Regression Results for Dependent Variable Al R-square - 0.82356787
Regression Error Total Variable intercept PASSES LN PASS ST DEV VARIANCE Q_SKEW PVFWDISM BULKDENS ASPHCONT TMAXDENS TRUTHICK PCMEDIUM
C(p) - 12.28329668
DF
Sum of Squares
11 997 1008
113131.43886897 24236.03602934 137367.47489831
Mean Square
Parameter Estimate
Standard Error
Type II Sum of Squares
F
Prob>F
-377.28036258 0.00050598 4.01861935 -2.45176964 0.11974458 -2.56905291 -0.00395845 0.49918104 4.72153434 1.78509208 -3.90395080 0.39399180
58.54883824 0.00012118 0.33329459 1.49439686 0.05743266 0.71210518 0.00011734 0.12978246 0.62254634 0.25278137 0.26160671 0.08638314
1009.38830544 423.80398839 3533.97638182 65.43255298 105.67209951 316.39108763 27665.82206725 359.62557251 1398.26461326 1212.26490420 5413.48996051 505.68820851
41.52 17.43 145.38 2.69 4.35 13.02 1138.09 14.79 57.52 49.87 222.70 20.80
0.0001 0.0001 0.0001 0.1012 0.0373 0.0003 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
343
I I Stepwise Regression Results for Dependent Variable HU2 R-square - 0.42016959
Regression Error Total Variable intercept PASSES LN PASS Q_SKEW PVFWDISM BULKDENS ASPHCONT TRUTHICK PCMEDIUM PCTSMALL
I
C(p) - 10.62453987
DF
Sum of Squares
Mean Square
F
Prob>F
9
3.33721118
999 1008
0.37080124
80.44
4.60532260 7.94253378
0.0001
0.00460993 -mse
Type II
Parameter
Standard
Error
Sum of Squares
F
Prob>F
-4.47261989 -0.00000508 0.03184627 0.02328770 0.00000809 0.02376895 0.03662013 -0.01187245 0.00983680 0.01584263
0.30080681 0.00000157 0.00417050 0.00929230 0.00000155 0.00184153 0.00477373 0.00359622 0.00111216 0.00261567
1.01916204 0.04800810 0.26880327 0.02895355 0.12617165 0.76799660 0.27128088 0.05024389 0.36063464 0.16911608
221.08 10.41 58.31 6.28 27.37 166.60 58.85 10.90 78.23 36.69
0.0001 0.0013 0.0001 0.0124 0.0001 0.0001 0.0001 0.0010 0.0001 0.0001
Estimate
Stepwise Regression Results for Dependent Variable WR R-square - 0.36151213
Regression Error Total Variable intercept PASSES
C(p) -
DF
Sum of Squares
6 1002 1008
98162.83239196 173371.16166155 271533.99405352
16360.47206533 173.02511144 -mse
F
Prob>F
94.56
0.0001
F
Prob>F
Parameter
Standard
Estimate
Type II
Error
Sum of Squares
231.22301551
36.14714326
7079.83764301
40.92
0.0001
0.00212137
0.00013938
40078.85801817
231.64
0.0001
6.87 96.04
0.0089 0.0001
Q_SKEW PVFWDISM
4.58598165 -0.00252858
BULKDENS
-0.87724708
0.23601574
2390.40275840
13.82
0.0002
ASPHCONT
-4.43946888
0.86553516
4551.99463681
26.31
0.0001
PCTSMALL
-3.22328768
0.47118347
8097.05666926
46.80
0.0001
1.75005817 0.00025802
344
1188.14322475 16616.91815383
I I
I 3
19.61208843 Mean Square
i
3
3 I. I I
