Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
2CHAPTER 2 EFFECTS OF SUBSURFACE DRAINAGE ON ASPHALT PAVEMENTS DESCRIPTION OF LTPP EXPERIMENT SPS-1 The SPS-1 experiment (Strategic Study of Structural Fac- tors for Flexible Pavements) was designed to assess the influ- ence of the following factors on the performance of asphalt concrete pavements: ⢠Asphalt concrete thickness, ⢠Base type, ⢠Base thickness, ⢠Subdrainage, ⢠Climate, ⢠Subgrade, and ⢠Truck traffic level. The Strategic Highway Research Programâs original exper- imental design and research plan for SPS-1 are described in Reference 1. The design factorial for the SPS-1 experiment is shown in Table 1. The first two digits (01) of the number shown within each cell signify the SPS-1 experiment; the last two digits signify the test section number of each design. The base types listed in Table 1 are the following: ⢠AGGâdense-graded aggregate ⢠ATBâasphalt-treated base ⢠ATB/AGGâasphalt-treated base over dense-graded aggregate ⢠PATB/AGGâpermeable asphalt-treated base over aggregate ⢠ATB/PATBâasphalt-treated base over permeable asphalt-treated base. Other than some thickness deviations, the pavement struc- tures actually constructed conform to the experiment design shown in Table 1, with one important exception: which sec- tions are drained and which are not. The field inspection of the drains, discussed in more detail later in this chapter, found several instances of the following discrepancies: ⢠Some pavement sections were found to have drains installed even though they should not according to the experiment design; ⢠Some pavement sections should have had drains installed, but the presence of the drains was not confirmed in the field inspections; and ⢠Some pavement sections should and do have drains installed, but the drains do not appear to be functioning effectively. The geographic distribution of the eighteen SPS-1 sites is illustrated in Figure 1. The originally intended site factorial for the experiment is shown in Table 2. Location data for the SPS-1 projects are given in Table 3. In the wet-freeze-fine subgrade cell, two states were each to have built the two sets of twelve designs, to obtain ârepli- catesâ of these designs. These are not, however, replicates in the true sense of the word, as the sites are not identical in terms of the concomitant variable, truck traffic level, nor are their climates identical. The site factorial shown in Table 2 represents what was supposed to have been built for the SPS-1 experiment. How- ever, the Texas site turned out to have a fine, not a coarse, subgrade. So, although the locations remain unchanged, the site factorial for the actual SPS-1 experiment is as shown in Table 4. The pavement designs in the core SPS-1 experiment con- structed at the Alabama, Delaware, Florida, Iowa, Kansas, Nevada, New Mexico, and Ohio sites are shown in Table 5. The design at the Arizona, Arkansas, Louisiana, Michigan, Montana, Nebraska, Oklahoma, Texas, Virginia, and Wiscon- sin sites are shown in Table 6. Monitoring of the Louisiana SPS-1 project has ceased, and almost no performance data are available for it. So, there are effectively seventeen SPS-1 sites in service. Climate Characterization The climatic distribution of the SPS-1 sites was determined by extracting the latitude and longitude for each site from the LTPP database, searching the National Oceanic and Atmos- pheric Administration (NOAA) database for the weather station nearest the SPS-1 site, extracting the 30-year aver- age monthly precipitation levels and average monthly tem- peratures for the weather station, and calculating the aver- age annual precipitation and average annual temperature for each. These data are provided in Table 7. The distribution of the SPS-1 sites with respect to average annual precipita- tion and temperature is illustrated in Figure 2.
3The need for subsurface drainage can be quantified with respect to a design rainfall, that is, one of a given magnitude, duration, and frequency. For example, a 1-year, 1-hour rainfall is an amount of rainfall that at a particular location lasts 1 hour and occurs, on average, once a year. Rainfall frequency infor- mation is not easily accessed in tabular form but rather must be obtained from contour maps. The 1-year, 1-hour rainfalls for the SPS-1 and SPS-2 sites were determined from contour maps in Reference 2. In the absence of rainfall frequency information, it is pos- sible to estimate the 1-year, 1-hour rainfall as a function of the average annual precipitation, which is more readily known. As Figure 3 shows, there is an evident (although nonlinear) correlation between average annual precipitation and the 1-year, 1-hour rainfall for nearly all of the sites. The excep- tions are the three SPS-1 sites closest to the Gulf coast (Texas, Louisiana, and Florida), for which the correlation curve seems to be shifted upwardâthat is, higher 1-year, 1-hour rainfall at those sites than for noncoastal sites with similar levels of average annual precipitation. Test Section Layouts and Pavement Structures The station limits, layer thicknesses, and material types for each of the SPS-1 test sections were extracted from the SPS_PROJECT_STATIONS and TST_L05B data tables in the LTPP database. The thicknesses in the TST_L05B table represent the LTPP regional support centersâ best estimates of the as-built layer thicknesses and materials. The test section layouts and pavement layer data are given in Appendix A, along with information on which sections are supposed to be drained and which are not, and the locations of edgedrain outlets inspected in the field inspections conducted in late 2001 and early 2002. The presence of filter fabric below the permeable asphalt- treated base in the SPS-1 test sections designed to be drained is summarized in Table 8. This summary is based on infor- mation in the TST_L05B and SPS1_LAYER data tables and the LTPP regional support centersâ responses to an LTPP Data Analysis/Operations Feedback Report (KTH-1, 11 June 2002) submitted by the NCHRP 1-34C research team. In nearly all cases, no filter fabric was used below the permeable asphalt-treated base in test sections with the PATB/AGG base type. (The exceptions are the three Iowa SPS-1 sec- tions, one of the three Kansas SPS-1 sections, and the three Texas SPS-1 sections with this base type.) On the other hand, in nearly all cases, filter fabric was used below the permeable asphalt-treated base in test sections with the ATB/PATB base type. (The exceptions are the three Kansas SPS-1 test sec- tions with this base type.) There are also several cases of sec- tions that do have a geotextile below the PATB, according to the regional support centers, although this is not reflected in the national LTPP database as of release 12.0. Traffic Characterization The 18-kip-equivalent single-axle (ESAL) levels at the SPS-1 sites were determined by extracting the following data from the LTPP database: ⢠ESAL estimates obtained from traffic monitoring dur- ing the experiment from the TRF_MON_EST_ESAL data table; and ⢠Axle load distributions obtained from traffic monitor- ing during the experiment from the TRF_MONITOR_ AXLE_DISTRIBUTION data table. Axle load distribution data were available for twelve of the eighteen SPS-1 sites. Data were not available for the sites in Alabama, Louisiana, Montana, Oklahoma, Texas, and Wisconsin. Undrained Drained Total Base Thickness, inches Surface Thickness, inches Dense- graded aggregate base Asphalt -treated base Asphalt- treated base over dense- graded aggregate Permeable asphalt- treated base over aggregate Asphalt- treated base over permeable asphalt- treated base 4 0113 0103 0105 0107 0122 8 7 0101 0115 0117 0119 0110 4 0102 0116 0118 0120 0111 12 7 0114 0104 0106 0108 0123 4 0121 0112 16 7 0109 0124 TABLE 1 SPS-1 design factorial
4Figure 1. SPS-1 sites.
ESALs were calculated for the years in which axle load dis- tribution data were available, using: (1) the number of axles reported in each load range in the distribution and (2) load equivalency factors calculated as a function of Structural Number (in turn calculated from as-built layer thicknesses and typical structural coefficients for asphalt concrete, treated and untreated base, and treated and untreated subbase). For years during the experiment in which axle load distribution data were not available, ESAL estimates from TRF_MON_ EST_ESAL table were used if available. In a few cases, linear interpolations of annual ESALs were necessary for years in which no axle load distribution or ESAL 5 data were available. It was also necessary in a few cases to extrapolate a year or two before or after the years for which data were available. A growth rate of 5 percent was used for these extrapolations. From the annual ESAL estimates for the different test sec- tions at each site, the average annual ESALs for the site were calculated. The average annual ESAL estimates for each site were used to calculate accumulated ESAL estimates from the date of opening to traffic to several dates of profile, rutting, and cracking measurements. The measurement dates were those on which measurements were obtained for most or all of the test sections at the site. The estimated accumulated Wet Dry Freeze Nonfreeze Freeze Nonfreeze IA, OH AL KS NM Fine subgrade VA, MI LA NE OK DE FL NV TX Coarse subgrade WI AR MT AZ TABLE 2 Intended SPS-1 site factorial SHRP ID State County Nearby city or town Route Latitude Longitude 010100 AL Lee Opelika US 280 32.61 85.25 040100 AZ Mohave Kingman US 93 35.39 114.26 050100 AR Craighead Jonesboro US 63 35.72 90.58 100100 DE Sussex Ellendale US 113 38.79 75.44 120100 FL Palm Beach Coral Springs US 27 26.33 80.69 190100 IA Lee Burlington US 61 40.42 91.25 200100 KS Kiowa Greensburg US 54 37.60 99.25 220100 LA Calcasieu Lake Charles US 171 30.33 93.20 260100 MI Clinton Lansing US 27 42.99 84.52 300100 MT Cascade Great Falls I-15 47.41 111.53 310100 NE Thayer Hebron US 81 40.07 97.62 320100 NV Lander Battle Mountain I-80 40.69 117.01 350100 NM Doña Ana Las Cruces I-25 32.68 107.07 390100 OH Delaware Delaware US 23 40.38 83.06 400100 OK Comanche Lawton US 62 34.64 98.66 480100 TX Hidalgo McAllen US 281 26.74 98.11 510100 VA Pittsylvania Danville US 29 36.66 79.37 550100 WI Marathon Wausau SR 29 44.87 89.29 TABLE 3 SPS-1 location data Wet Dry Freeze Nonfreeze Freeze Nonfreeze IA, OH AL KS NM Fine subgrade VA, MI LA NE OK, TX DE FL NV Coarse subgrade WI AR MT AZ TABLE 4 Actual SPS-1 site factorial
ESALs to each of the selected measurement dates, for each site with data available, are provided in Appendix A. The age of each SPS-1 site and the accumulated flexible pavement ESALs for each site with traffic data available, as of the latest date of condition measurements available for this study, are shown in Table 9. Also shown is the accumulated ESALs divided by the age, which gives a rough estimate of the average annual ESAL level during the time that each site has been in service. The annual and accumulated ESAL lev- els are fairly low at all of the SPS-1 sites. The ages and accumulated ESALs at the SPS-1 sites are also illustrated in Figure 4. The vertical scale of this graph was cho- sen to be compatible with the range of rigid pavement ESALs for the SPS-2 sites (shown in Chapter 3). Most of the SPS-1 pavements had carried considerably less truck traffic at the time of the analysis than most of the SPS-2 pavements. 6 SPS-1 Construction Some information on construction of the SPS-1 sites is available in Reference 3 and in the construction reports pre- pared by the LTPP regional support centers. At most of the SPS-1 sites, one or more problems or deviations occurred dur- ing construction. However, most of the construction reports contain little information on the installation of the drainage systems. The information on construction deviations avail- able in Reference 3 is summarized in Table 10. FIELD INSPECTIONS OF DRAINS AT SPS-1 SITES Video inspections of the drains at the SPS-1 and SPS-2 sites were conducted in late 2001 and early 2002 under an FHWA Undrained Drained Total Base Thickness, inches Surface Thickness, inches Dense- graded aggregate base Asphalt -treated base Asphalt- treated base over dense- graded aggregate Permeable asphalt- treated base over aggregate Asphalt- treated base over permeable asphalt- treated base 4 0113 0122 8 7 0115 0117 0119 4 0116 0118 0120 12 7 0114 0123 4 0121 16 7 0124 TABLE 6 Core SPS-1 test sections built at the Arizona, Arkansas, Louisiana, Michigan, Montana, Nebraska, Oklahoma, Texas, Virginia, and Wisconsin sites Undrained Drained Total Base Thickness, inches Surface Thickness, inches Dense- graded aggregate base Asphalt -treated base Asphalt- treated base over dense- graded aggregate Permeable asphalt- treated base over aggregate Asphalt- treated base over permeable asphalt- treated base 4 0103 0105 0107 8 7 0101 0110 4 0102 0111 12 7 0104 0106 0108 4 0112 16 7 0109 TABLE 5 Core SPS-1 test sections built at the Alabama, Delaware, Florida, Iowa, Kansas, Nevada, New Mexico, and Ohio sites
7SPS-1 Site Nearest Weather Station State State Code Latitude (degrees) Longitude (degrees) ID Name Latitude (degrees) Longitude (degrees) Average Annual Precipitation (inches) Average Annual Temperature (degrees F) AL 01 32.61 85.25 014502 Lafayette 32.54 85.24 57.56 62.60 AZ 04 35.39 114.26 267369 Searchlight 35.28 114.55 7.42 63.30 AR 05 35.72 90.58 033734 Jonesboro 4 N 35.53 90.42 47.19 60.60 DE 10 38.79 75.44 182523 Denton 2 E 38.53 75.48 42.55 55.90 FL 12 26.33 80.69 081654 Clewiston US ENG 26.45 80.55 45.01 74.00 IA 19 40.42 91.25 135796 Mount Pleasant 1SSW 40.57 91.33 36.98 49.80 KS 20 37.60 99.25 144333 Kinsley 37.55 99.25 24.51 54.90 LA 22 30.33 93.20 162361 De Quincy 4 N 30.31 93.26 58.20 66.20 MI 26 42.99 84.52 207280 Saint Johns 43.01 84.33 31.39 46.20 MT 30 47.41 111.53 242857 Fairfield 47.37 111.59 12.46 42.90 NE 31 40.07 97.62 251680 Clay Center 6 ESE 40.30 97.56 26.98 49.90 NV 32 40.69 117.01 263245 Golconda 40.57 117.29 7.58 50.00 NM 35 32.68 107.07 291286 Caballo Dam 32.54 107.18 10.17 60.20 OH 39 40.38 83.06 334942 Marion 2 N 40.37 83.08 36.91 49.20 OK 40 34.64 98.66 349629 Wichita MT WL REF 34.44 98.43 31.12 59.60 TX 48 26.74 98.11 413063 Fulfurrias 27.14 98.08 25.88 71.60 VA 51 36.66 79.37 441614 Chatham 36.49 79.24 44.39 54.80 WI 55 44.87 89.29 475364 Merrill 45.11 89.41 32.21 41.00 TABLE 7 Average annual precipitation and temperature levels for weather stations nearest SPS-1 sites 32 48 64 80 0 21 42 63 Mean annual precipitation (inches) M ea n an n u al te m pe ra tu re (o F ) NM AZ TX FL NV KS OK NE AR VA DE IA OH MI LA AL WI MT Figure 2. Distribution of SPS-1 sites with respect to precipitation and temperature.
8(c) Subdrains with outlets were placed and could have been located but were not marked for inspection, or (d) Subdrains with outlets were placed and were marked for inspection but were not inspected. ⢠Second, there were several instances found of test sec- tions that should be drained (test sections numbered 0107 through 0112 or 0119 through 0124, depending on the site), but at which no lateral outlets within or near the sections were found or marked for inspection. These instances are indicated by an ânâ rather than a âyâ in some cells in the last six test section columns in Table 11. These discrepancies were reported to LTPP in an LTPP Data Analysis/Operations Feedback Report (KTH-3, 11 June 2002) submitted by the NCHRP 1-34C research team. The responses received from the LTPP regional support centers are summarized below. SPS-1 test sections designed to be undrained, but drains were found and inspected: ⢠Iowa 0102, 0103, 0104, 0105, and 0106âDesign plans show that outlets were planned in each of these sec- tions. In section 0103, the outlet was to be placed on the right side of the road; in the other four sections, the out- let was to be on the left side of the road. No as-built records are available to confirm whether or not these outlets are present. contract with NCHRP support. The completed data forms for each site were furnished to the NCHRP Project 1-34C research team for use in this study. Highlights of the inspections are summarized in Appendix A. The test section layout diagrams in Appendix A show the locations of the edgedrain outlets inspected. Observations on Results of Field Inspections A summary of the drains inspected and not inspected is given in Table 11. In cases where an outlet was located out- side the limits of any test section, it was assumed to be asso- ciated with the nearest test section. The field inspections reveal some surprising things about the drainage installations at several of the SPS-1 sites. ⢠First, there were some instances found of test sections that should not be drained but at which lateral outlets were found and marked for inspection. These are indi- cated by a âyâ in some cells in the first six test section columns in Table 11. There are several possible expla- nations for these discrepancies: (a) Longitudinal subdrains with lateral outlets were not placed in the test section during construction, or (b) Subdrains with outlets were placed during construc- tion but could not be located and marked for inspec- tion, or 0.00 0.50 1.00 1.50 2.00 2.50 0 10 20 30 40 50 60 70 Average annual precipitation (inches) 1- ye ar , 1 -h ou r p re ci pi ta tio n (in ch es ) Gulf SPS-1 sites (FL, LA,TX) All other SPS-1 sites All SPS-2 sites Figure 3. 1-year, 1-hour frequency precipitation rate versus average annual precipitation for SPS-1 and SPS-2 sites.
⢠Michigan 0115, 0117, and 0118âDesign plans show no outlets planned for the first two of these sections; they show one outlet planned for the third section, but the station is not indicated. No as-built records are available to confirm whether or not these outlets are present. ⢠Ohio 0101, 0102, 0103, 0104, 0105, and 0106âDesign plans show that outlets were planned, for both sides of the road, in sections 0101 and 0102. No as-built records are available to confirm whether or not these outlets are present. ⢠Virginia 0114, 0115, and 0116âNo edge drains were installed and are not present. Edge drains may be present outside of the section limits and may be the reason the drainage inspector found outlets. The stationing should be checked to verify the correct location of the outlets. SPS-1 test sections designed to be drained, but drains were not found: ⢠AZ 0124âAccording to construction information, this section has three drainage outlets; but, due to the over- 9 grown bush and the presence of rattlesnakes, these out- lets were not inspected. ⢠FL 0108 and 0111âThe Florida SPS-1 laterals have been damaged by farming traffic. The missing laterals are most likely buried and not operational. ⢠Iowa 0107, 0110, and 0112âDesign plans show that outlets were planned in each of these sections. In section 0107, the outlet was to be on the left side of the road; in sections 0110 and 0112, outlets were to be placed on both the left and right sides of the road. No as-built records are available to confirm whether or not these outlets are present. ⢠Kansas 0107, 0109, 0110, and 0112âDesign plans show that outlets were planned on both the left and right sides of the road in each of the sections. No as-built records are available to confirm whether or not these outlets are present. ⢠Louisiana 0119 and 0122âLouisiana SPS-1 was already at close-out status at the time of inspection. Laterals were retrofitted in all sections with PATB layer. The missing laterals are most likely buried and not functional. Base Type Permeable asphalt-treated base over aggregate Asphalt-treated base over permeable asphalt-treated base Section number State 0107 0108 0109 0110 0111 0112 AL 01 no no no yes yes yes DE 10 no no no yes yes yes FL 12 no no no yes yes yes IA 19 yes* yes* yes* yes* yes* yes* KS 20 yes* no no no no no LA 22 no no no yes yes yes NV 32 no no no yes* yes* yes* NM 35 no no no yes yes yes OH 39 no no no yes* yes* yes* Section number 0119 0120 0121 0122 0123 0124 AZ 04 no no no yes* yes* yes* AR 05 no no no yes yes yes MI 26 no no no yes yes yes MT 30 no no no yes* yes* yes* NE 31 no no no yes* yes* yes* OK 40 no no no yes yes yes TX 48 yes yes yes yes yes yes VA 51 no no no yes yes yes WI 55 no no no yes yes yes * The regional support center reports that geotextile was placed below the PATB, although this is not reflected in LTPP database release 12.0. TABLE 8 Presence of filter fabric below permeable asphalt-treated base in SPS-1 sections designed to be drained
⢠Michigan 0119, 0120, 0122, and 0124âDesign plans show that three outlets were planned in each of these sections. No as-built records are available to confirm whether or not these outlets are present. ⢠New Mexico 0109âThe project is supposed to have concrete headwalls. However, the State Coordinator could not locate the headwalls in the locations where he did not find the lateral [sic: presumably, where the lat- eral should have been]. This is an indication that the headwalls and lateral are buried. ⢠Ohio 0107 and 0109âDesign plans show that section 0107 had two outlets planned, on both sides of the road. Design plans do not show any outlets for section 0109. No as-built records are available to confirm whether or not these outlets are present. ⢠Virginia 0119âEdgedrains were installed and are pres- ent. The drainage inspector must have missed the outlet. ⢠Wisconsin 0119, 0120, 0121, 0122, 0123, and 0124â Designs plans indicate that two or three outlets were planned in each of these sections. Outlets in sections 0120 and 0122 were planned on the left side of the road; outlets in the other sections were planned on both sides of the road. No as-built records are available to confirm whether or not these outlets are present. [Note: the video inspection contractor reported that LTPP regional cen- ter support personnel visited the Wisconsin SPS-1 site and could not locate any lateral outlets to mark on either occasion. However, Wisconsin DOT personnel and a 10 research team member who visited the site confirm that the outlets are present.] Test sections designed to be drained but for which drainage outlets could not be located were not used in this analysis of the effects of drainage on pavement performance. Similarly, test sections designed to be undrained but at which lateral outlets were found and inspected also were not used in the analysis presented in this report. After reviewing the forms summarizing the video inspec- tions of the drainage installations, the 1-34C research team made a subjective assessment of whether the quality of drain- age functioning in each test section was âgoodâ or âpoor.â The ratings assigned are summarized in Table 12. Conditions that garnered a âpoorâ rating included lateral outlets being buried or fully blocked with silt, gravel, or other debris; lon- gitudinal drains being fully blocked; or a considerable amount of water standing in longitudinal drains and not flowing out. Longitudinal drains and lateral outlets in a pavement struc- ture that has been in service some years are never pristine; there is nearly always some silt accumulation and some rodents and their nests. Whether or not there is enough mate- rial present to block the flow of water in the event of a stormâ or whether the flow of water caused by a storm would clear out some or all of this materialâremains a matter of judgment, until someone investigates this by conducting video inspec- tions before and after storms. In general, if the amount of mate- State Age, years Accumulated flexible pavement ESALs, millions Accumulated ESALs divided by age Alabama 8.04 Data unavailable Arizona 7.75 1.93 0.25 Arkansas 6.39 2.66 0.42 Delaware 4.60 2.01 0.44 Florida 4.86 2.25 0.46 Iowa 7.99 1.22 0.15 Kansas 7.53 1.88 0.25 Louisiana 2.36 Data unavailable Michigan 4.87 0.32 0.07 Montana 2.62 Data unavailable Nebraska 5.79 0.56 0.10 Nevada 5.62 3.01 0.54 New Mexico 5.51 0.84 0.15 Ohio 5.44 0.39 0.07 Oklahoma 3.51 Data unavailable Texas 3.89 Data unavailable Virginia 4.95 1.66 0.33 Wisconsin 2.60 Data unavailable Average 5.94 1.56 0.27 TABLE 9 Age and accumulated flexible pavement equivalent standard axle loads (ESALs) of SPS-1 sites at time of analysis
rial described as being present did not seem sufficient to block the flow of water, a âgoodâ rating was assigned. In Table 12, a question mark alone in a cell indicates that no rating can be assigned because no laterals were inspected within the test section. A question mark with an asterisk indi- cates that a lateral was found and inspected visually, but the video camera could not be inserted in the lateral to inspect the interior of the longitudinal drain. In some cases this was because the inner diameter of the lateral was too small. In other cases this was because rodent screens were placed too far up in the lateral, out of armâs reach, and could not be removed so that the camera could pass. EFFECTS OF DRAINAGE ON SPS-1 ASPHALT PAVEMENT PERFORMANCE Tables 5 and 6 show that the following test section com- parisons may be conducted to assess the effects of drainage on pavement performance in the SPS-1 experiment, holding asphalt concrete thickness and base thickness constant. (A) Undrained dense-graded aggregate (AGG) versus dense asphalt-treated base over drained permeable asphalt-treated base (ATB/PATB) ⢠0101 versus 0110: 7-inch asphalt concrete over 8-inch base ⢠0102 versus 0111: 4-inch asphalt concrete over 12-inch base 11 ⢠0113 versus 0122: 4-inch asphalt concrete over 8-inch base ⢠0114 versus 0123: 7-inch asphalt concrete over 12-inch base (B) Undrained dense asphalt-treated base (ATB) versus permeable asphalt-treated base over dense-graded aggregate (PATB/AGG): ⢠0103 versus 0107: 4-inch asphalt concrete over 8-inch base ⢠0104 versus 0108: 7-inch asphalt concrete over 12-inch base ⢠0115 versus 0119: 7-inch asphalt concrete over 8-inch base ⢠0116 versus 0120: 4-inch asphalt concrete over 12-inch base (C) Undrained dense asphalt-treated base over dense- graded aggregate (ATB/AGG) versus permeable asphalt-treated base over dense-graded aggregate (PATB/AGG): ⢠0105 versus 0107: 4-inch asphalt concrete over 8-inch base ⢠0106 versus 0108: 7-inch asphalt concrete over 12-inch base ⢠0117 versus 0119: 7-inch asphalt concrete over 8-inch base 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 0 1 2 3 4 5 6 7 8 9 10 Age (years) A cc um ul at ed fl ex ib le p av em en t E SA Ls (m illi on s) Figure 4. Age versus accumulated flexible ESALs for SPS-1 sites with traffic data available.
12 SHRP ID State Problem or Deviation 010100 AL Mechanical problem with paver; construction joint placed in section 0111. Deformations occurred on top of the PATB1. DGAB2 contained excess minus 200 material. 040100 AZ Rain delays during subgrade preparation. Fill material pumped, but was replaced prior to paving. Section 0122 included a layer of DGAB below the PATB. DGAB for sections 0119 and 0122 did not meet the gradation requirements. 050100 AR Rain caused construction delays, but surfaces were allowed to dry prior to resuming construction. DGAB thickness on section 0114 was less than half of the required value. Many other sections were also less than the design value. The stability of the HMA3 mixture was less than the specified value. 100100 DE High water table along the project. Ditches were shallow, so outlets of drains were not placed at the 76-m spacing. The number 4 sieve from the gradation tests for the HMA surface did not meet the project specifications. 120100 FL Rain delays caused the DGAB to be reworked multiple times. The number 4 sieve from the gradation tests for the HMA surface did not meet the project specification. 190100 IA Multiple rain delays, but surfaces were allowed to dry and were reworked. PATB ârolled outâ on the sides, which resulted in the placement of an extra lift to meet the thickness requirement. The number 4 sieve from the gradation tests for the HMA binder layer exceeded the project requirements. 200100 KS Excessive moisture in the subbase, which caused difficulty in compacting the material. Fly ash was added to the subbase layer for stabilization purposes. 220100 LA Test sections for thickness cells 1 to 12 rather than 13 to 24 were built. Rain delays. Subgrade was stabilized with cement. Fabric did not meet overlay requirements. Aggregate in drainage trenches contained fines. DGAB was compacted in one lift. Select material was used at site to achieve the final elevation. 260100 MI None noted. 300100 MT None noted. 310100 NE Three test sections were constructed over culverts. Rain delays. The minus 200 material for the PATB exceeded the project requirements. 320100 NV HMA facility breakdown. High air voids reported in the ATB4 prior to plant breakdown. Localized tenderness problem noted. 350100 NM HMA facility breakdown. High air voids reported in the ATB prior to plant breakdown. Localized tenderness problem noted. 390100 OH Fill material placed on all sections. DGAB thickness was much larger than the planned thickness. The number 4 sieve for the HMA surface did not meet the project requirements. 400100 OK The number 4 sieve for the HMA surface did not meet the project requirements. One of the two ATB lifts exceeded the project thickness requirements. 480100 TX Transverse interceptor drains not installed along the project. 510100 VA Subgrade treated with cement. The number 4 sieve for the HMA surface did not meet the project requirements. 550100 WI None noted. 1 Permeable asphalt-treated base 2 Dense-graded aggregate base. 3 Hot mix asphalt. 4 Asphalt-treated base. TABLE 10 SPS-1 construction problems and deviations (from Reference 3)
13 ⢠0118 versus 0120: 4-inch asphalt concrete over 12-inch base The test sections on 16-inch-thick base (numbered 0109, 0112, 0121, and 0124) are not directly useful in this analysis of the effects of drainage on pavement performance, since all of these sections were designed to have drains, and there are no corresponding undrained designs in the experiment. How- ever, the test sections on 16-inch-thick-base are anticipated to be useful in future analyses differentiating the effects of structural capacity and subdrainage on performance. The sites with test sections useful to each of the above comparisons were identified by combining the information about which test sections had drainage inspected (Table 11) with the information about the subjective ratings of drainage functioning (Table 12). For each of the three sets of comparisons listed, and for each pavement performance indicator considered (Interna- tional Roughness Index [IRI], rutting, and cracking), paired t-tests were conducted to determine whether or not the mean difference between the undrained and drained test section pairs was significant. Three paired t-tests were conducted each time: ⢠For the subset of test section pairs with drainage func- tioning rated as good (denoted by G in the column âDrainage functioningâ in Tables 13 through 21), ⢠For the subset of test section pairs rated as poor (denoted by P in the column âDrainage functioningâ in Tables 13 through 21), and ⢠For all valid test section pairs regardless of drainage func- tioning (denoted by G, P, or ? in Tables 13 through 21). The test section pairs excluded from these comparisons were: ⢠Those that had subdrainage outlets located and inspected in the section that was designed to be undrained (denoted by X in Tables 13 through 21), and Test Section Numbers 0101 and 0113 0102 and 0114 0103 and 0115 0104 and 0116 0105 and 0117 0106 and 0118 0107 and 0119 0108 and 0120 0109 and 0121 0110 and 0122 0111 and 0123 0112 and 0124 Base Type Dense- graded aggregate Asphalt- treated base Asphalt- treated base over dense- graded aggregate Permeable asphalt- treated base over aggregate Asphalt-treated base over permeable asphalt-treated base State Undrained Drained AL y1 y y y y y AZ y y y y y n2 AR y y y y y y DE y y y y y y FL y n y y n y IA y3 y y y y n y y n n y KS n y n n y n LA n y y n y y MI y y y n n y n y n MT y y y y y y NE y y y y y y NV y y y y y y NM y y n y y y OH y y y y y y n y n y y y OK y y y y y y TX y y y y y y VA y y y n y y y y y WI n n n n n n 1 y = in unshaded cells, subdrains found and inspected 2 n = subdrains not found or found but not inspected. 3 y = in shaded cells, drains found and inspected, even though according to the experiment design the section should not have drains. TABLE 11 SPS-1 test-section sections with drainage outlets inspected with video
14 ⢠Those lacking measurement data for one or both of the test sections (denoted by --- in Tables 13 through 21). The difference in the performance measures (IRI, rutting, or cracking) for each test section pair is calculated as the mea- sured value for the undrained section minus the measured value for the drained section. Thus, a positive difference indi- cates that the measured value was greater in the undrained section, and a negative difference indicates that the measured value was greater in the drained section. The mean difference for all the pairs considered is the sum of the pairwise differences divided by the number of pairs. Whether or not the mean difference is significant (at a selected significance level, e.g., 95 percent) is determined by calcu- lating the lower and upper limits of (95-percent) confidence interval around the mean difference. If zero is not contained within the lower and upper limits of the confidence interval, the mean difference can be concluded, with 95 percent con- fidence, to be significantly different from zero. Effect of Drainage on Asphalt Pavement Roughness (IRI) Development For each SPS-1 site, IRI data were extracted from the MON_PROFILE_MASTER table in the LTPP database. An IRI for each run was calculated as the average of the runâs left and right wheelpath IRIs. An average IRI for each test- ing date was then calculated as the average of the run IRIs for that dateâmost often five runs, but sometimes as few as one or as many as fifteen. The expectation is that IRI will tend to increase over time, as the pavement deteriorates. However, IRI does not always increase steadily over time. Sometimes the IRI of a test sec- tion is lower than the IRI measured on the same test section Test Section ID 0101 and 0113 0102 and 0114 0103 and 0115 0104 and 0116 0105 and 0117 0106 and 0118 0107 and 0119 0108 and 0120 0109 and 0121 0110 and 0122 0111 and 0123 0112 and 0124 Base Type Dense- graded aggregate Asphalt- treated base Asphalt- treated base over dense- graded aggregate Permeable asphalt- treated base over aggregate Asphalt-treated base over permeable asphalt-treated base State Undrained Drained AL G1 G G G G G AZ G G G G G ? 2 AR P3 P P P P P DE G G G G G G FL P ? P P ? P IA P G P P G ? G P ? ? P KS ? P ? ? ?* 4 ? LA ? P P ? ? ? MI P ? ? P ? P ? MT G G G G G G NE G G G G G G NV P P P P P P NM P P ? P P P OH G G G G P ? G ? G G G OK ?* ?* ?* ?* ?* ?* TX P P P P P P VA G G G ? G G G G G WI ? ? ? ? ? ? 1 G = Drainage function rated as good. 2 ? = Drainage outlets not found. 3 P = Drainage function rated as poor. 4 ?* = Camera could not be inserted. TABLE 12 Subjective ratings of drainage functioning at SPS-1 test sections based on video inspection results
a year earlier, a month earlier, or even a day earlier. Physical reasons why IRI might decrease from one testing date to the next. include the following (from Reference 4): ⢠Rehabilitation or maintenance between testing dates; ⢠Seasonal variation; ⢠Measurement in different paths; ⢠Different starting locations; ⢠Spikes in the data caused by reflection of light from the white paint stripe at the start of a test section; or ⢠Problems with the profilometer electronics, sensors, or distance measurement. However, it is not necessarily true that an IRI decrease, or an IRI increase for that matter, always has a physical explana- tion. Some portion of the variation seen in IRI data is random variation. That is, some fluctuations in IRI, both upward and downward, are not significantly different from no change at all. The first available IRI is not necessarily the IRI immedi- ately after opening to traffic. The first testing date for which IRI data are available for all or most of the test sections at a site may be a year or more after the opening date. (Note that the date of opening to traffic is shown for each site in the ESAL calculation summary in Appendix A.) Similarly, the latest IRI measurements used in the analysis are not neces- sarily those for the latest IRI measurement date at any one test section at a site, but rather those for the latest IRI mea- surement date for which measurements are available for most or all of the sections at a site. The IRI histories of the test sec- tions at each SPS-1 site are shown in Appendix A. Note that no IRI history is shown for the Louisiana SPS-1 site because only one set of IRI data, from 1997, is available for that site. The comparisons of IRI change between drained and undrained sections of matching designs are shown in Tables 13, 14, and 15. Which comparisons were deemed possible was assessed on the basis of the drainage detection and drainage functioning information summarized in Tables 11 and 12. Table 13 shows the comparisons of undrained dense-graded aggregate base (AGG) versus dense asphalt-treated base over drained permeable asphalt-treated base (ATB/PATB). For drained ATB/PATB sections with drainage functioning sub- jectively rated as good, the change in IRI was significantly less than in the undrained AGG sections of corresponding design. (This is indicated by a positive mean difference in change in IRI, undrainedâdrained, that is significantly dif- ferent from zero). The same was true when all drained ATB/ PATB sections combined (good, poor, and unknown drain- age functioning) were compared with the corresponding undrained AGG sections. When only drained ATB/PATB sections with drainage functioning rated as poor were com- pared with corresponding undrained AGG sections, no sig- nificant difference in change in IRI was detected. Table 14 shows the comparisons of undrained dense asphalt-treated base (ATB) versus drained permeable asphalt- treated base over aggregate (PATB/AGG). For drained PATB/ AGG sections with drainage functioning rated as good, the 15 change in IRI was slightly, but not significantly, greater than in the corresponding undrained ATB sections. (This is indi- cated by a negative mean difference, undrainedâdrained, that is not significantly different from zero). The same was true for drained PATB/AGG sections with drainage func- tioning rated as poor, and for all drained PATB/AGG sec- tions regardless of drainage functioning. Table 15 shows the comparisons of undrained dense asphalt-treated base over dense-graded aggregate (ATB/AGG) versus drained permeable asphalt-treated base over aggre- gate (PATB/AGG). For drained PATB/AGG sections with drainage functioning rated as good, the change in IRI was slightly, but not significantly, less than in the corresponding undrained ATB/AGG sections. The same was true for all drained PATB/AGG sections considered together, regard- less of drainage functioning. For drained PATB/AGG sec- tions with drainage functioning rated as poor, the change in IRI was slightly greater, but again not significantly so, than in the corresponding undrained ATB/AGG sections. Effect of Drainage on Asphalt Pavement Rutting Development The rutting histories of the test sections at each SPS-1 site are shown in Appendix A. The estimated accumulated ESALs corresponding to the most recent rutting measure- ments are reported in Appendix A. The accumulated ESAL values are slightly different from those reported for the IRI histories, since rutting and IRI were measured on different dates. Rutting histories for the Florida and Louisiana SPS-1 sites are not shown because of the limited data available. The comparisons of rutting change between drained and undrained sections of matching designs are shown in Tables 16, 17, and 18. Which comparisons were deemed possible was assessed on the basis of the drainage detection and drainage functioning information summarized in Tables 11 and 12. Table 16 shows the comparisons of undrained dense-graded aggregate base (AGG) versus dense asphalt-treated base over drained permeable asphalt-treated base (ATB/PATB). For drained ATB/PATB sections with drainage functioning sub- jectively rated as good, the change in rutting was slightly, but not significantly, less than in the undrained AGG sections of corresponding design. The same was true for drained ATB/ PATB sections with drainage functioning rated as poor, and for all drained ATB/PATB sections combined (good, poor, and unknown drainage functioning). Table 17 shows the comparisons of undrained dense asphalt- treated base (ATB) versus drained permeable asphalt-treated base over aggregate (PATB/AGG). For drained PATB/AGG sections with drainage functioning rated as good, the change in rutting was slightly, but not significantly, less than in the corresponding undrained ATB sections. The same was true for drained PATB/AGG sections with drainage functioning rated as poor and for all drained PATB/AGG sections com- bined regardless of drainage functioning.
16 IRI Difference in IRI AGG ATB/PATB Age, years Drainage functioning All Good Poor Site 0101 0110 AL 01 0.086 0.002 5.37 G1 0.084 0.084 DE 10 0.010 -0.055 3.49 G 0.065 0.065 FL 12 -0.071 -0.004 3.62 P2 -0.067 -0.067 IA 19 0.636 0.576 6.29 ?3 0.060 KS 20 0.411 0.198 4.27 ? 0.213 NV 32 -0.008 -0.009 3.14 P 0.001 0.001 NM 35 0.253 0.005 4.15 P 0.248 0.248 OH 39 2.682 0.125 0.37 X4 Site 0102 0111 AL 01 0.470 0.027 5.18 G 0.443 0.443 DE 10 0.116 -0.023 3.49 G 0.139 0.139 FL 12 0.002 -0.019 3.62 ? 0.021 IA 19 1.713 0.338 6.29 X KS 20 -0.136 0.137 4.27 ? -0.273 NV 32 0.210 -0.009 3.14 P 0.219 0.219 NM 35 0.333 0.128 4.15 P 0.205 0.205 OH 39 0.315 0.097 0.37 X Site 0113 0122 AZ 04 0.184 0.102 6.89 G 0.082 0.082 AR 05 0.162 0.230 5.54 P -0.068 -0.068 LA 22 ---5 --- --- --- MI 26 --- --- --- --- MT 30 0.034 -0.036 1.64 G 0.070 0.070 NE 31 0.082 0.056 3.93 G 0.026 0.026 OK 40 0.108 0.042 3.13 ? 0.066 TX 48 0.035 -0.002 3.63 P 0.037 0.037 VA 51 0.420 -0.050 3.62 G 0.470 0.470 WI 55 0.100 0.073 2.40 ? 0.027 Site 0114 0123 AZ 04 0.201 0.052 6.89 G 0.149 0.149 AR 05 0.150 0.157 5.54 P -0.007 -0.007 LA 22 --- --- --- --- MI 26 --- 0.057 3.30 --- MT 30 0.011 0.009 1.64 G 0.002 0.002 NE 31 -0.022 -0.131 5.08 G 0.109 0.109 OK 40 0.064 0.030 3.13 ? 0.034 TX 48 0.100 -0.024 3.63 P 0.124 0.124 VA 51 -0.027 0.018 3.62 X WI 55 0.155 0.184 2.40 ? -0.029 Mean difference 0.088 0.149 0.077 n 28 11 9 SD 0.148 0.158 0.125 t alpha/2, n-1 2.546 2.862 3.007 Confidence interval lower limit 0.016 0.013 -0.048 Confidence interval upper limit 0.159 0.285 0.202 Significant difference? (Overall confidence level = 95%) yes yes no 1 G = Drainage function rated as good. 2 P = Drainage function rated as poor. 3 ? = Drainage outlets not found. 4 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 5 --- = lacking measurement data for one or both test sections. TABLE 13 Change in International Roughness Index (IRI) in SPS-1 undrained dense-graded aggregate (AGG) base sections versus drained asphalt-treated base over drained permeable asphalt-treated base (ATB/PATB) sections
17 IRI ATB ATB/PATB Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0103 0107 AL 01 0.033 0.306 5.18 G1 -0.273 -0.273 DE 10 0.009 0.093 3.49 G -0.084 -0.084 FL 12 0.009 -0.004 3.62 P12 0.013 0.013 IA 19 0.544 ---3 6.29 X4 KS 20 0.619 0.152 4.27 ?5 0.467 NV 32 -0.001 -0.028 3.14 P 0.027 0.027 NM 35 0.244 0.215 4.15 P 0.029 0.029 OH 39 1.729 0.164 0.37 X Site 0104 0108 AL 01 0.054 -0.028 3.72 G 0.082 0.082 DE 10 -0.041 -0.041 3.49 G 0.000 0.000 FL 12 -0.012 -0.146 3.62 ? 0.134 IA 19 0.466 0.953 6.29 X KS 20 0.271 0.781 7.00 P -0.510 -0.510 NV 32 -0.026 -0.030 3.14 P 0.004 0.004 NM 35 0.156 0.022 4.15 P 0.134 0.134 OH 39 0.571 1.087 4.01 X Site 0115 0119 AZ 04 0.030 0.126 6.89 G -0.096 -0.096 AR 05 0.156 0.813 5.54 P -0.657 -0.657 LA 22 --- --- --- --- MI 26 0.141 --- 3.30 --- MT 30 0.023 -0.020 1.64 G 0.043 0.043 NE 31 --- -0.219 5.08 --- OK 40 0.032 -0.007 3.13 ? 0.039 TX 48 0.328 0.489 3.63 P -0.161 -0.161 VA 51 0.001 0.058 3.62 X WI 55 0.284 0.101 2.40 ? 0.183 Site 0116 0120 AZ 04 0.064 0.141 6.89 G -0.077 -0.077 AR 05 0.024 0.657 5.54 P -0.633 -0.633 LA 22 --- --- --- --- MI 26 0.104 0.150 3.30 ? -0.046 MT 30 -0.005 0.051 1.64 G -0.056 -0.056 NE 31 -0.104 -0.168 5.08 G 0.064 0.064 OK 40 -0.018 0.044 3.13 ? -0.062 TX 48 0.521 0.099 3.63 P 0.422 0.422 VA 51 0.004 0.009 3.62 X WI 55 0.226 0.012 2.40 ? 0.214 Mean difference -0.031 -0.044 -0.133 N 26 9 10 SD 0.263 0.109 0.356 t alpha/2, n-1 2.560 3.007 2.925 Confidence interval lower limit -0.163 -0.153 -0.462 Confidence interval upper limit 0.101 0.065 0.196 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 P = Drainage function rated as poor. 3 --- = lacking measurement data for one or both test sections. 4 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 5 ? = Drainage outlets not found. TABLE 14 Change in International Roughness Index (IRI) in SPS-1 undrained asphalt-treated base (ATB) sections versus drained permeable asphalt-treated base (ATB/PATB) sections
18 IRI ATB/AGG PATB/AGG Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0105 0107 AL 01 0.023 0.306 5.18 G1 -0.283 -0.283 DE 10 0.040 0.093 3.49 G -0.053 -0.053 FL 12 0.007 -0.004 3.62 P2 0.011 0.011 IA 19 0.538 ---3 6.29 X4 KS 20 1.167 0.152 4.27 ? 5 1.015 NV 32 -0.026 -0.028 3.14 P 0.002 0.002 NM 35 0.133 0.215 4.15 P -0.082 -0.082 OH 39 0.688 0.164 0.37 G 0.524 0.524 Site 0106 0108 AL 01 0.107 -0.017 5.18 G 0.124 0.124 DE 10 -0.034 -0.041 3.49 G 0.007 0.007 FL 12 -0.011 -0.146 3.62 ? 0.135 IA 19 0.357 0.953 6.29 X KS 20 0.771 0.781 7.00 P -0.010 -0.010 NV 32 -0.010 -0.030 3.14 P 0.020 0.020 NM 35 0.269 0.022 4.15 P 0.247 0.247 OH 39 0.646 1.087 4.01 X Site 0117 0119 AZ 04 0.001 0.126 6.89 G -0.125 -0.125 AR 05 0.126 0.813 5.54 P -0.687 -0.687 LA 22 --- --- --- --- MI 26 0.546 --- 3.30 --- MT 30 -0.014 -0.020 1.64 G 0.006 0.006 NE 31 -0.109 -0.219 5.08 G 0.110 0.110 OK 40 0.048 -0.007 3.13 ? 0.055 TX 48 0.036 0.489 3.63 P -0.453 -0.453 VA 51 0.002 0.058 3.62 ? -0.056 WI 55 0.166 0.101 2.40 ? 0.065 Site 0118 0120 AZ 04 0.055 0.141 6.89 G -0.086 -0.086 AR 05 0.112 0.657 5.54 P -0.545 -0.545 LA 22 --- --- --- --- MI 26 0.279 0.150 3.30 X MT 30 0.019 0.051 1.64 G -0.032 -0.032 NE 31 -0.163 -0.168 5.08 G 0.005 0.005 OK 40 0.082 0.044 3.13 ? 0.038 TX 48 0.038 0.099 3.63 P -0.061 -0.061 VA 51 0.047 0.009 3.62 G 0.038 0.038 WI 55 0.150 0.012 2.40 ? 0.138 Mean difference 0.002 0.020 -0.156 N 29 12 10 SD 0.301 0.192 0.299 t alpha/2, n-1 2.540 2.812 2.925 Confidence interval lower limit -0.140 -0.136 -0.432 Confidence interval upper limit 0.144 0.176 0.121 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 P = Drainage function rated as poor. 3 --- = lacking measurement data for one or both test sections. 4 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 5 ? = Drainage outlets not found. TABLE 15 Change in International Roughness Index (IRI) in SPS-1 undrained asphalt-treated base over dense-graded aggregate (ATB/AGG) sections versus drained permeable asphalt-treated base over aggregate (PATB/AGG) sections
19 IRI AGG ATB/PATB Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0101 0110 AL 01 2 -1 6.11 G1 3 3 DE 10 ---2 2 3.68 --- FL 12 --- --- --- --- IA 19 0 -1 4.88 ? 3 1 KS 20 -5 3 6.36 ? -8 NV 32 0 2 4.35 P4 -2 -2 NM 35 3 1 3.01 P 2 2 OH 39 -- 1 3.73 X5 Site 0102 0111 AL 01 8 3 6.11 G 5 5 DE 10 7 1 3.68 G 6 6 FL 12 --- --- --- --- IA 19 7 -1 4.88 X KS 20 -11 -1 6.36 ? -10 NV 32 5 2 4.35 P 3 3 NM 35 1 1 3.01 P 0 0 OH 39 --- 2 3.73 X Site 0113 0122 AZ 04 2 7 4.79 G -5 -5 AR 05 -4 -1 4.56 P -3 -3 LA 22 3 1 1.51 ? 2 MI 26 --- --- --- --- MT 30 1 2 1.08 G -1 -1 NE 31 23 9 4.26 G 14 14 OK 40 2 0 0.96 ? 2 TX 48 1 1 2.17 P 0 0 VA 51 4 0 2.14 G 4 4 WI 55 5 -24 2.13 ? 29 Site 0114 0123 AZ 04 4 -3 3.88 G 7 7 AR 05 1 0 4.56 P 1 1 LA 22 4 2 1.51 ? 2 MI 26 --- 2 4.18 --- MT 30 1 0 1.08 G 1 1 NE 31 20 17 4.26 G 3 3 OK 40 1 1 0.96 ? 0 TX 48 7 2 2.17 P 5 5 VA 51 0 3 2.14 X WI 55 5 5 2.13 ? 0 Mean difference 2.3 3.7 0.8 N 27 10 8 SD 7.1 5.1 2.6 t alpha/2, n-1 2.553 2.925 3.118 Confidence interval lower limit -1.2 -1.0 -2.1 Confidence interval upper limit 5.7 8.4 3.6 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 --- = lacking measurement data for one or both test sections. 3 ? = Drainage outlets not found. 4 P = Drainage function rated as poor. 5 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. TABLE 16 Change in rutting in SPS-1 undrained dense-graded aggregate (AGG) sections versus drained asphalt-treated base over permeable asphalt-treated base (ATB/PATB) sections
20 Base Type ATB PATB/AGG D ifference Drained N Y Age, years Drainage functioning All Good Poor Site 0103 0107 AL 01 3 7 6.11 G1 -4 -4 DE 10 3 2 3.68 G 1 1 FL 12 --- ---2 --- --- IA 19 -2 --- 4.88 X3 KS 20 3 -8 6.36 ?4 11 NV 32 2 0 4.35 P5 2 2 NM 35 2 2 3.01 P 0 0 OH 39 12 --- 3.73 X Site 0104 0108 AL 01 4 2 6.11 G 2 2 DE 10 3 2 3.68 G 1 1 FL 12 --- --- --- --- IA 19 2 4 5.43 X KS 20 0 3 6.36 P -3 -3 NV 32 1 2 4.35 P -1 -1 NM 35 1 1 3.01 P 0 0 OH 39 5 14 3.73 X Site 0115 0119 AZ 04 1 8 4.79 G -7 -7 AR 05 3 -1 4.56 P 4 4 LA 22 5 2 1.51 G 3 3 MI 26 5 --- 3.30 --- MT 30 0 0 1.08 G 0 0 NE 31 11 6 4.26 G 5 5 OK 40 0 0 0.96 ? 0 TX 48 8 1 2.17 P 7 7 VA 51 0 2 2.14 X WI 55 7 7 2.13 ? 0 Site 0116 0120 AZ 04 3 5 4.79 G -2 -2 AR 05 0 -1 4.56 P 1 1 LA 22 4 -3 1.51 G 7 7 MI 26 2 --- 4.18 --- MT 30 0 1 1.08 G -1 -1 NE 31 10 8 4.26 G 2 2 OK 40 0 0 0.96 ? 0 TX 48 10 2 2.17 P 8 8 VA 51 1 3 2.14 X WI 55 5 2 2.13 ? 3 Mean difference 1.5 0.6 2.0 N 26 12 9 SD 3.9 3.8 3.7 t alpha/2, n-1 2.560 2.812 3.007 Confidence interval lower limit -0.5 -2.5 -1.7 Confidence interval upper limit 3.5 3.7 5.7 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 --- = lacking measurement data for one or both test sections. 3 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 4 ? = Drainage outlets not found. 5 P = Drainage function rated as poor. TABLE 17 Change in rutting in SPS-1 undrained asphalt-treated base (ATB) sections versus drained permeable asphalt-treated base over aggregate (PATB/AGG) sections
21 Base Type ATB/AGG PATB/AGG Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0105 0107 AL 01 4 7 6.11 G1 -3 -3 DE 10 5 2 3.68 G 3 3 FL 12 ---2 --- --- --- IA 19 2 --- 4.88 X3 KS 20 21 -8 6.36 ?4 29 NV 32 3 0 4.35 P5 3 3 NM 35 1 2 3.01 P -1 -1 OH 39 5 --- 1.12 --- Site 0106 0108 AL 01 1 2 6.11 G -1 -1 DE 10 1 2 3.68 G -1 -1 FL 12 --- --- --- --- IA 19 1 4 5.43 X KS 20 1 3 6.36 P -2 -2 NV 32 0 2 4.35 P -2 -2 NM 35 -3 1 0.58 P -4 -4 OH 39 2 14 3.73 X Site 0117 0119 AZ 04 3 8 4.79 G -5 -5 AR 05 2 -1 4.56 P 3 3 LA 22 5 2 1.51 G 3 3 MI 26 5 --- 4.18 --- MT 30 0 0 1.08 G 0 0 NE 31 11 6 4.26 G 5 5 OK 40 2 0 0.96 ? 2 TX 48 0 1 2.17 P -1 -1 VA 51 1 2 2.14 ? -1 WI 55 5 7 2.13 ? -2 Site 0118 0120 AZ 04 2 5 4.79 G -3 -3 AR 05 -1 -1 4.56 P 0 0 LA 22 4 -3 1.51 G 7 7 MI 26 --- --- --- X MT 30 2 1 1.08 G 1 1 NE 31 13 8 4.26 G 5 5 OK 40 1 0 0.96 ? 1 TX 48 1 2 2.17 P -1 -1 VA 51 3 3 2.14 G 0 0 WI 55 7 2 2.13 ? 5 Mean difference 1.4 0.8 -0.6 N 28 13 9 SD 6.2 3.6 2.3 t alpha/2, n-1 2.546 2.772 3.007 Confidence interval lower limit -1.5 -1.9 -2.9 Confidence interval upper limit 4.4 3.6 1.7 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 --- = lacking measurement data for one or both test sections. 3 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 4 ? = Drainage outlets not found. 5 P = Drainage function rated as poor. TABLE 18 Change in rutting in SPS-1 undrained asphalt-treated base over dense-graded aggregate (ATB/AGG) sections versus drained permeable asphalt-treated base over aggregate (PATB/AGG) sections
22 Base Type AGG ATB/PATB Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0101 0110 AL 01 8 9 7.21 G1 -1 -1 DE 10 0 0 4.50 G 0 0 FL 12 ---2 --- --- --- IA 19 19 12 7.70 ? 3 7 KS 20 0 5 3.40 ? -5 NV 32 2 1 5.63 P4 1 1 NM 35 --- --- --- --- OH 39 0 0 1.01 X5 Site 0102 0111 AL 01 18 13 7.21 G 5 5 DE 10 15 0 4.50 G 15 15 FL 12 --- --- --- --- IA 19 32 22 7.70 X KS 20 0 22 3.40 ? -22 NV 32 1 0 5.63 P 1 1 NM 35 --- --- --- --- OH 39 --- 23 5.44 X Site 0113 0122 AZ 04 12 7 7.75 G 5 5 AR 05 15 13 5.82 P 2 2 LA 22 --- --- --- --- MI 26 --- --- --- --- MT 30 13 16 1.78 G -3 -3 NE 31 --- --- --- --- OK 40 --- --- --- --- TX 48 4 2 3.89 P 2 2 VA 51 0 0 1.44 G 0 0 WI 55 --- --- --- --- Site 0114 0123 AZ 04 13 16 7.75 G -3 -3 AR 05 14 13 5.82 P 1 1 LA 22 --- --- --- --- MI 26 --- 4 4.87 --- MT 30 13 13 1.78 G 0 0 NE 31 --- --- --- --- OK 40 --- --- --- --- TX 48 0 0 3.89 P 0 0 VA 51 0 0 1.44 X WI 55 --- --- --- --- Mean difference 0.3 2.0 1.2 N 18 9 6 SD 7.1 5.7 0.8 t alpha/2, n-1 2.648 3.007 3.521 Confidence interval lower limit -4.2 -3.7 0.1 Confidence interval upper limit 4.7 7.7 2.2 Significant difference (Overall confidence level = 95%) no no yes 1 G = Drainage function rated as good. 2 --- = lacking measurement data for one or both test sections. 3 ? = Drainage outlets not found. 4 P = Drainage function rated as poor. 5 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. TABLE 19 Change in cracking in meters in SPS-1 undrained dense-graded aggregate (AGG) sections versus drained asphalt-treated base over permeable asphalt-treated base (ATB/PATB) sections
23 Base Type ATB PATB/AGG Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0103 0107 AL 01 11 9 7.21 G1 2 2 DE 10 0 0 4.50 G 0 0 FL 12 ---2 --- --- --- IA 19 11 16 7.70 X3 KS 20 18 0 3.40 ?4 18 NV 32 3 0 5.63 P5 3 3 NM 35 --- --- --- --- OH 39 42 --- 5.44 X Site 0104 0108 AL 01 0 21 7.21 G -21 -21 DE 10 0 0 4.50 G 0 0 FL 12 --- --- --- --- IA 19 23 17 7.70 X KS 20 31 60 6.10 P -29 -29 NV 32 1 1 5.63 P 0 0 NM 35 --- --- --- --- OH 39 24 46 5.44 X Site 0115 0119 AZ 04 9 9 7.75 G 0 0 AR 05 14 28 5.82 P -14 -14 LA 22 --- --- --- --- MI 26 12 --- 4.87 --- MT 30 12 14 1.78 G -2 -2 NE 31 --- --- --- --- OK 40 --- --- --- --- TX 48 0 0 3.89 P 0 0 VA 51 0 0 1.44 X WI 55 --- --- --- --- Site 0116 0120 AZ 04 5 10 7.75 G -5 -5 AR 05 13 20 5.82 P -7 -7 LA 22 --- --- --- --- MI 26 0 4 0.93 ? -4 MT 30 12 21 1.78 G -9 -9 NE 31 --- --- --- --- OK 40 --- --- --- --- TX 48 0 0 3.89 P 0 0 VA 51 0 0 1.44 X WI 55 --- --- --- --- Mean difference -4.0 -4.4 -6.7 N 17 8 7 SD 10.4 7.6 11.4 t alpha/2, n-1 2.666 3.118 3.276 Confidence interval lower limit -10.7 -12.7 -20.8 Confidence interval upper limit 2.7 4.0 7.4 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 --- = lacking measurement data for one or both test sections. 3 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 3 ? = Drainage outlets not found. 5 P = Drainage function rated as poor. TABLE 20 Change in cracking in meters in SPS-1 undrained asphalt-treated base (ATB) sections versus drained permeable asphalt-treated base over aggregate (PATB/AGG) sections
24 Base Type ATB/AGG ATB/PATB Difference Drained N Y Age, years Drainage functioning All Good Poor Site 0105 0107 AL 01 25 9 7.21 G1 16 16 DE 10 0 0 4.50 G 0 0 FL 12 ---2 --- --- --- IA 19 27 16 7.70 X3 KS 20 30 0 3.40 ?4 30 NV 32 2 0 5.63 P5 2 2 NM 35 --- --- --- --- OH 39 0 --- 2.13 --- Site 0106 0108 AL 01 4 21 7.21 G -17 -17 DE 10 0 0 4.50 G 0 0 FL 12 --- --- --- --- IA 19 17 17 7.70 X KS 20 23 60 6.10 P -37 -37 NV 32 0 1 5.63 P -1 -1 NM 35 --- --- --- --- OH 39 32 46 5.44 X Site 0117 0119 AZ 04 5 9 7.75 G -4 -4 AR 05 13 28 5.82 P -15 -15 LA 22 --- --- --- --- MI 26 5 --- 4.87 --- MT 30 14 14 1.78 G 0 0 NE 31 --- --- --- --- OK 40 --- --- --- --- TX 48 3 0 3.89 P 3 3 VA 51 0 0 1.44 ? 0 WI 55 --- --- --- --- Site 0118 0120 AZ 04 9 10 7.75 G -1 -1 AR 05 14 20 5.82 P -6 -6 LA 22 --- --- --- --- MI 26 0 4 0.93 X MT 30 12 21 1.78 G -9 -9 NE 31 --- --- --- --- OK 40 --- --- --- --- TX 48 13 0 3.89 P 13 13 VA 51 0 0 1.44 G 0 0 WI 55 --- --- --- --- Mean difference -1.4 -1.7 -5.9 N 18 9 7 SD 14.0 8.8 16.2 t alpha/2, n-1 2.648 3.007 3.276 Confidence interval lower limit -10.2 -10.5 -25.9 Confidence interval upper limit 7.3 7.1 14.2 Significant difference (Overall confidence level = 95%) no no no 1 G = Drainage function rated as good. 2 --- = lacking measurement data for one or both test sections. 3 X = Subdrainage outlets located and inspected in the section that was designed to be undrained. 3 ? = Drainage outlets not found. 5 P = Drainage function rated as poor. TABLE 21 Change in cracking in meters in SPS-1 undrained asphalt-treated base over dense-graded aggregate (ATB/AGG) sections versus drained permeable asphalt-treated base over aggregate (PATB/AGG) sections
25 Table 18 shows the comparisons of undrained dense asphalt- treated base over dense-graded aggregate (ATB/AGG) versus drained permeable asphalt-treated base over aggregate (PATB/ AGG). For drained PATB/AGG sections with drainage func- tioning rated as good, the change in rutting was slightly, but not significantly, less than in the corresponding undrained ATB/AGG sections. The same was true for all drained PATB/ AGG sections considered together, regardless of drainage functioning. For drained PATB/AGG sections with drainage functioning rated as poor, the change in rutting was slightly but not significantly greater than in the corresponding undrained ATB/AGG sections. Effect of Drainage on Asphalt Pavement Cracking Development For the purpose of this analysis, the area of alligator crack- ing (all severities) was added to the area affected by longitu- dinal cracking (sealed and unsealed, all severities, wheelpath and nonwheelpath, length times 18 inches or 0.45 m), and this area was divided by the area of the pavement section (typically 556 m2). Both alligator cracking and longitudinal cracking were considered together, because examination of the survey data indicated that the trends in each may be very erratic from year to year whereas the sum of the two tends to have a more stable trend. This is believed to be due to variation from year to year in the survey techniciansâ classification of the distress observed. It is conceivable that this summation method produces some overestimates of the percent area cracked in cases when both alligator cracking and longitudinal cracking are located in the same area. Also, the selection of 18 inches as a typical wheel- path width is arbitrary, and different cracked area percentages would be obtained if some other width were assumed. The cracking histories for eleven SPS-1 sites are shown in Appendix A. Cracking histories are not shown for the Florida, Louisiana, Nebraska, Nevada, New Mexico, Oklahoma, and Wisconsin sites because of very low calculated cracking levels. The comparisons of cracking between drained and undrained sections of matching designs are shown in Tables 19, 20, and 21. Which comparisons were deemed possible was assessed on the basis of the drainage detection and drainage functioning information summarized in Tables 11 and 12. Table 19 shows the comparisons of undrained dense-graded aggregate base (AGG) versus dense asphalt-treated base over drained permeable asphalt-treated base (ATB/PATB). For drained ATB/PATB sections with drainage functioning sub- jectively rated as good, cracking was slightly, but not signif- icantly, less than in the undrained AGG sections of corre- sponding design. The same was true for drained ATB/PATB sections with drainage functioning rated as poor and for all drained ATB/PATB sections combined (good, poor, and unknown drainage functioning). Table 20 shows the comparisons of undrained dense asphalt- treated base (ATB) versus drained permeable asphalt-treated base over aggregate (PATB/AGG). For drained PATB/AGG sections with drainage functioning rated as good, cracking was slightly, but not significantly, greater than in the corre- sponding undrained ATB sections. The same was true for drained PATB/AGG sections with drainage functioning rated as poor, and for all drained PATB/AGG sections combined, regardless of drainage functioning. Table 21 shows the comparisons of undrained dense asphalt- treated base over dense-graded aggregate (ATB/AGG) versus drained permeable asphalt-treated base over aggregate (PATB/ AGG). For drained PATB/AGG sections with drainage func- tioning rated as good, cracking was slightly, but not signifi- cantly, greater than in the corresponding undrained ATB/ AGG sections. The same was true for drained PATB/AGG sections with drainage functioning rated as poor, and for all drained PATB/AGG sections considered together, regardless of drainage functioning.
