U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-HRT-06-121
Date: November 2006

Long-Term Pavement Performance (LTPP) Data Analysis Support: National Pooled Fund Study Tpf-5(013)

Chapter 8. Local Adaptationsof Empirical Pavement Design Practices and Materials Standards

One of the primary study areas requested in the proposal was for the "Analysis of the extent to which local adaptations of materials standards and empirical pavement design practices have been effective at reducing the rate of pavement deterioration."

To gain an understanding of State design practices, a questionnaire was developed and sent to the PFS participants. The complete questionnaire is provided in appendix D. Basic information on standard roadway sections including structural design for given scenarios, standard specifications, and test procedures was requested. Questions regarding average unit bid prices and typical service life estimates were included for purposes of LCCA. A query on the knowledge of current treatment practices implemented by adjacent SHAs was also incorporated.

The pavement sections or specific pavement designs provided by the States are shown in tables 24 through 27.

Table 24. Primary highway flexible pavement design summary.
StatesAC (mm (inch))Treated Base (mm (inch))Untreated Base (mm (inch))Total Depth (mm (inch))Treated SG (mm (inch))Total FFa Depth (mm (inch))
Alaska 130 (5)NA480 (19)710 (24)915 (36) select1270 (50)
Idaho 165 (6.6)NA305 (12)470 (18.6)NA470 (18.6)
Illinois 355 (14)NA0 (0)355 (14)305 (12) lime660 (26)
Michigan 170 (6.5)NA710 (24)775 (30.5)var. gr. surf1065-1525 (42–
New York 150 (6)100 (4) ATPB305 (12)560 (22)NA560 (22)
North Carolina 265 (10.5)NA255 (10)NANA520 (20.5)
Ohio 203 (8)NA150 (6)355 (14)no A-4b for 915 (36)1270 (50)
Pennsylvania 445 (17.5)NA0 (0)445 (17.5)NA445 (17.5)
Design conditions: 30-year design; 5 million ESALs; frost-susceptible fine-grained soil with resilient modulus of 68,950 kilopascals (kPa) (10,000 pounds of force per square inch (lbf/in2) ).
Table 25. Primary highway rigid pavement design summary.
StatesPCC (mm (inch))Treated Base (mm (inch))Untreated Base (mm (inch))Treated SG (mm (inch))Total FFa Depth (mm (inch))
Idaho 230 (9)50 (2) ATPLC305 (12)NA585 (23)
Illinois 250 (9.75)100 (4)NA305 (12) lime655 (25.75)
Michigan 215 (8.5)150 (6) UTB255 (10)var. gran. mat1065–1525 (42–60)
New York 255 (10)100 (4) ATPB305 (12)NA660 (26)
North Carolina 205 (8)115 (4.5)NA205 (8) lime520 (20.5)
Ohio 205 (8)NA152 (6)no A-4b for 915 (36)1270 (50)
Pennsylvania 205 (8)NA205 (8)NA405 (16)
Design conditions: 30-year design; 5 million ESALs; frost-susceptible fine-grained soil with resilient modulus of 68,950 kPa (10,000 lbf/in2).
Table 26. Interstate highway flexible pavement design summary.
StatesAC (mm (inch))Treated Base (mm (inch))Untreated Base (mm (inch))Treated Subgrade (mm (inch))Total Depth (mm (inch))
Idaho 200 (7.8)NA710 (24)NA810 (31.8)
Illinois 515 (20.25)NA0 (0)305 (12) lime820 (32.25)
Michigan 185 (7.25)NA710 (24)1065 (42) granular1860 (73.25)
New York 180 (7)100 (4) ATPB305 (12)NA585 (23)
North Carolina 405 (16)NA255 (10)NA660 (26)
Ohio 290 (11.5)NA150 (6)no silt for 915 (36)1360 (53.5)
Pennsylvania 420 (16.5)NA255 (10)NA675 (26.5)
Design conditions: 30-year design; 10 million ESALs; frost-susceptible fine-grained soil with resilient modulus of 68,950 kPa (10,000 lbf/in2)..
Table 27. Interstate highway rigid pavement design summary.
StatesPCC (mm (inch))Treated Base (mm (inch))Untreated Base (mm (inch))Treated Subgrade (mm (inch))Total Depth (mm (inch))
Idaho 305 (12)50 (2) ATPLC305 (12)NA660 (26)
Illinois 265 (10.5)100 (4)0 (0)305 (12) lime675 (26.5)
Michigan 290 (11.5)NA405 (16)1065 (42) granular1765 (69.5)
New York 255 (10)100 (4) ATPB305 (12)NA660 (26)
North Carolina 280 (11)115 (4.5)0 (0)205 (8) lime600 (23.5)
Ohio 290 (11.5)NA150 (6)no silt for 915 (36)1360 (53.5)
Pennsylvania 330 (13)100 (4) ATPB100 (4)NA535 (21)
Design conditions: 30-year design; 10 million ESALs; frost-susceptible fine-grained soil with resilient modulus of 68,950 kPa(10,000 lbf/in2).


As can be seen from these tables, there was a large variation in the roadway design sections reported for each SHA. There was no specific trend in the pavement designs between those SHAs that experience deep frost and those that experience moderate to no frost penetration. No SHA reported any specific design thickness requirement based on frost depth. One SHA, Pennsylvania, noted design consideration for frost heave in accordance with the 1993 AASHTO Design Guide. In particular, the design is based on the loss in serviceability because of frost heave as described in appendix G of the 1993 AASHTO Guide for Design of Pavement Structures.(1)

All of the SHAs provided a pavement design based on the subgrade soils resilient modulus (MR) value noted in the questionnaire without accounting for frost effects; however, one SHA does not design pavement structures for MR values greater than 41,370 kPa (6,000 lbf/in2). This may reflect either the weaker soils that are encountered in the State or a method to partially account for thaw weakening in the environment of the State. Two SHAs indicated that their standard construction procedures require that frost-susceptible subgrade soils be removed for a depth of 1m (3 ft) or more. This additional provision is shown as treated subgrade in tables 24 through 26 (Michigan and Ohio).

Alaska traditionally places roadways on embankment sections that consist of a minimum height of 1m (3 ft) of select frost-free material. This practice provides for a minimum of 1 m (3 ft) of frost-free material, in addition to the normal pavement design thickness, to protect the pavement against weakening from frost heave and thaw.

To be consistent with the pavement performance indicated from the LTPP database, Idaho reported design sections that generally represent design procedures that were in place during the development of the LTPP test sections. The SHA is now incorporating a rock cap in much of its new construction that consists of 0.3 to 0.6 m (1 to 2 ft) of rock ballast placed between the subgrade and the surfacing layers, which has been shown to provide long-term benefits in the reduction of frost and moisture-related pavement damage. Nordic countries such as Sweden typically use a similar ballast layer that is usually placed at depths of a little more than 1 m (3 ft).(17) Figure 66 provides a photograph of a deep-base section in Sweden.

In earlier studies, many northern SHAs were found to require a minimum depth of frostfree material ranging from 50 to 100 percent of the maximum measured frost depth(4) for the specific design area. SHAs such as Utah require 100 percent of the measured frost depth, which ranges from 0.6 m (2 ft) to more than 1.2 m (4 ft) in that region. The recent reconstruction of I-15 through Salt Lake City used a 330-mm (13-inch) thick PCC over a 205-mm (8-inch) open graded drainage layer with an additional 380 mm (15 inches) of granular subbase to provide for the required 915 mm (36 inches) of frost measured in the area.(18)

The effect of these treatments is difficult to quantify because the treatments may or may not be incorporated into the existing LTPP sites depending on individual SHA practice. This may be accounted for in the models developed from the LTPP database because these practices are part of the roadway sections in the GPS experiment. The consultant did not directly separate sites with extra surfacing depths based on individual agency policy to either subexcavate frost-susceptible soils or add additional granular base based on local frost depths from the remaining sites. To a limited extent, additional surfacing beyond normal AASHTO design depths is represented and accounted for in the LESN term that was found to have a significant contribution as an explanatory variable in the models developed in this research study.

In general, it appears that many SHAs deal with greater frost depths by adding additional granular surfacing to reduce or prevent frost heaving. Some SHAs accomplish this by replacing the frost-susceptible subgrade with an acceptable frost-free material at fairly specific depths or at variable depths depending on the environmental range in that State. Other SHAs increase the pavement depth using additional gravel surfacing consistent with measured frost depths. Not all SHAs follow these practices, and the LTPP database incorporates pavement deterioration trends from all SHAs including northern SHAs that increase the surfacing and those who do not.

Many SHAs increase the depth of frost-free material over that required by the AASHTO Guide for Design of Pavement Structures (1) to minimize frost heaving effects where significant frost depths occur. The literature review and SHA research review did not find any specific examples where a difference in pavement performance was quantified relative to increased surfacing depths or removal of frost-susceptible soils. Considering the prevalence of the practice across the northern tier SHAs, it appears to be a general, but not universal, practice.

For years, Washington has minimized frost heaving effects and thaw weakening on lowvolume roadways by adding gravel surfacing beyond that designated by the AASHTO pavement design procedure for frost depth. The use of additional surfacing significantly increased the service life of the roadway and minimized or eliminated the need for spring load restrictions to protect the roadway. Washington State, as part of its Highway System Plan, includes an Economic Initiative Strategy and subprogram for "All Weather Roadways (Freeze/Thaw)," which is directed to improving low-volume roadways that should be open throughout the winter for the transportation of goods and services.(19) That program consists largely of rebuilding low-volume roadways by increasing the surfacing depth to minimize the effect of frost heaving and thaw weakening. The ultimate goal is to minimize or eliminate the need for spring load closures to protect the roadway, which often causes an economic hardship to the local community. That program has been quite successful.

Figure 66. Photo. Road construction in Sweden with deep base section.

View Alternate Text

There was a very limited amount of literature available from the participating and adjacent SHAs regarding frost penetration and FTC mitigation. A study conducted for Ohio found that overlays in areas with high snowfall deteriorate faster than those in other areas.(20) Two studies performed in Alaska considered differential thaw settlement when considering pavement structural performance and the statewide pavement management system.(21,22) Soil characteristics were investigated and empirically linked to pavement performance.

One source of information on frost heaving and thaw weakening was the WSDOT Pavement Guide, which includes a description of frost action as well as several reviews of other SHA practices.(4) It provides a general description of frost heaving and thaw weakening as well as a discussion on other SHA practices including added surfacing based on depth of frost. Drainage and the use of capillary blankets are also addressed.

In special cases, such as in the Western States and Northeastern States, where rolling to mountainous terrain are present, moisture enters the roadway prism flowing along lateral soils deposits. This happens in varied silt and sand layers or on top of rock contact zones. Washington State has this terrain, where much of the moisture enters the roadway prism laterally instead of vertically from capillary tension, which also produces ice lenses and causes frost heaving. WSDOT has significantly reduced frost heaving in these areas through the use of longitudinal drainage where the water can be intercepted before it enters the roadway prism, as shown in the photograph in figure 67. Most of the worst frost heaving areas in the State highway system exists in the northeastern slopes of the Cascade Mountain Range. This has been corrected either by digging out and increasing the surfacing depths, adding more surfacing depth, or installing longitudinal drainage along the inslope of the ditch line.(23)

Similar treatments have been used by the New York State Department of Transportation (NYSDOT).(24) In addition, NYSDOT has required the use of rock subgrade fragmentation where the rock is fragmented for a depth of .9 to 1.2 meters (3 to 4 ft) below subgrade during rock excavation to provide adequate drainage in areas of rock cuts.

Figure 67. Photo. Installation of longitudinal drainage to reduce frost heaving.

View Alternate Text

The photograph in figure 67 shows the installation of a longitudinal drain in the in-slope of the ditch line. A geotechnical study revealed that considerable moisture was entering the roadway prism from lateral flow in and below the cut slope, which caused the frost heaving in the limits of the cut section. Longitudinal drains were installed to intercept this water and reduce or eliminate the frost heaving. It should be noted that this application was successful but its potential application is limited by the nature of the area and local environment. In this particular case, the frost penetration is approximately 1 m (3 ft) deep, and the ditch line is covered with snow through most of the winter. This significantly reduces the frost depth along the ditch line and does not allow the drain line to freeze.

There was, for a period, some application of foam board to reduce or eliminate frost heaving. There were several installations across the country from Colorado to Maine as well as in Canada and the Nordic countries.(25) The use of foam provided favorable results as far as reducing frost heaving; however, several States experienced a problem with the roadway frosting in the area where the foam was used. There is currently very little or no use of foam insulation for this application.

Maintaining the same frost-free surfacing depth throughout the roadway prism is an additional design consideration to reduce differential frost heave. This was not mentioned in the SHA responses to the questionnaire, but it was recognized in the literature of one adjacent SHA as having contributed to shoulder heave,(26) which resulted in early distress initiation. This appears to be a treatment required more by Western States than those in the Midwest or East.

Most western SHAs maintain the same depth of frost-free surfacing, be it bound pavement or untreated surfacing, throughout the entire roadway prism. This is usually done to prevent differential frost heaving over the full width of the roadway section.

Some of the roadway sections that were provided in response to the questionnaire indicated that the pavement section would not have been designed with a uniform depth of frost-free material across the full roadway prism. In States where a significant depth of frost-susceptible subgrade is removed and replaced (such as in Michigan and Ohio), there would be no frost-susceptible material within 1 m (3 ft) or more of the subgrade surface. In other States where frost-susceptible soils can be placed in the roadway prism, such as that shown in figure 68, there is a potential for differential frost heaving across the roadway prism.

Figure 68. Diagram. Standard pavement section from a Midwestern State.

View Alternate Text
1 inch = 2.54 cm; 1 ft = 0.3 m

Using a uniform depth of frost-free material across the entire roadway prism appears to represent standard practice by some SHAs, but it may not represent universal or even general practice. The amount of differential frost heaving that would be experienced in a roadway section such as that shown in figure 68 is likely dependent on the environment where it is used. In milder environments, where the freezing front advances slowly through the upper portion of the roadway prism, the amount of moisture that is brought into the frost-susceptible material in the outer edges of the prism might be enough to cause greater heaving in that area, causing differential heaving of the roadway prism. On the other hand, a rapidly advancing freezing front in a more extreme climate may plunge below the surfacing layers quickly enough to minimize any increase in moisture and heaving in the outer sections of the roadway prism. SHAs experiencing differential heaving of the outer edges of the roadway prism could benefit by investigating the use of uniform depths of frost-free material across the entire roadway section.

Applying chip seals on hot-mix asphalt pavements is standard practice in many northern tier SHAs. Idaho, Montana, Wyoming, North Dakota, and South Dakota chip seal hotmix pavements just after construction, and continue chip seal applications at 6 to 8 year intervals until the next resurfacing project, which might be 25 years or more after initial construction.(10)

A limited amount of research is available that addresses the use of chip seals to improve pavement performance in deep frost areas. North Dakota has documented the used of sand and chip seals to protect pavements from FTCs and expansive soil problems.(10) It is quite likely that the use of chip seals on hot-mix asphalt pavement reduces the amount of surface raveling in areas with deep frost penetration and numerous FTCs. While the treatment appears to provide some protection against raveling, the relative advantage in improving pavement performance was not quantified. It should be noted that the use of chip seals on hot-mix asphalt pavement was confined to SHAs that generally have lower traffic volumes, where the application of chip seals are usually more acceptable.


The responses from SHAs on material standards appear in appendix E, as well as in data gathered from standard specifications available on individual State Web sites.

Most SHAs indicated that they have adopted the use of the Superpave PG binder specification and the Superpave mix design procedures. An abbreviated summary of the responses to the questionnaire and the review of standard specifications on the web sites are shown in table 28.

Table 28. Hot-mix asphalt concrete binder grading and mix designs used by the PFS for surfacing courses.
StateHMAC ReferenceAC GradeMix Design
Alaska Hot-Mix Asphalt (HMA), Type II, Class BPG 58/64–28Marshall
Idaho Plant Mix PavementPG 64–28Marshall
Illinois SP HMA Surface CoursePG 58/64–22Superpave/Marshall
Michigan Gap Graded SPPG 70–22PSuperpave
New York 12.5 mm Superpave HMAPG 64–28/PG 70–22Superpave
North Carolina S-12.5CPG 70–22Superpave
Ohio Item 880 (7 yr warranty)PG 64–22/PG 70–22MSuperpave/Marshall
Pennsylvania Superpave HMA Wearing CrPG 64–22/PG 58–22Superpave/Marshall

Appendix F contains the full set of specification comparisons for wearing course and base course, as well as treated and untreated surfacing. An overview of standard design practices for each SHA is also provided in the appendix.

Because cold weather performance was a major consideration in the development of the Superpave binder specifications, it is logical that their use will lead to improved performance; however, the use of Superpave mix design procedures has, to a large extent, eliminated local adaptations in mix design procedures and specifications that might have provided improved performance in areas of deep frost penetration and numerous FTCs. The Superpave mix design procedure does not differentiate between mix designs where mixes will be exposed to numerous FTCs and those that will experience little or no FTCs.

A review of mix specifications shows there are still some local adaptations in acceptable Los Angeles (L.A.) wear values (35 to 55 maximum) and sulfide soundness values (9 to 18 percent maximum). These differences probably represent material availability issues more than environmental issues in the respective SHAs. In addition, minor differences exist in the requirements for antistrip agents. Most agencies required a minimum retained strength of 70 to 80 percent after Lottman conditioning of the test samples to eliminate the need to add antistrip to the mix design. North Carolina requires a minimum of 85 percent of the retained strength after a modified Lottman conditioning. The modified test procedure excludes freezing the sample before the tensile strength ratio (TSR) tests were run. That requirement may not be as demanding as a 70-percent requirement based on the normal Lottman conditioning (with the FTC). It is interesting to note that North Carolina also has the mildest environment as far as frost depth and freeze-thaw cycling is concerned when compared with the other SHAs evaluated in the study. As such, SHAs with more aggressive winter climates should not consider the same modification to antistrip requirements.

Because many SHAs are in the process of adopting the Superpave binder specifications as well as the mix design procedures, local adaptations in mix designs and specifications do not indicate improved pavement performance in areas with either deep frost penetration or numerous FTCs. Many other SHA specifications were reviewed to determine if differences in grading and density requirements existed between the more northern SHAs as compared to the more southern SHAs. Most agencies showed control points and density requirements similar to those contained in the Superpave mix guidelines. Some SHAs have maintained their grading requirements, but there was no consistent pattern to lead to strong conclusions.

Previous | Contents | Next

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101