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-08-057
Date: November 2008

Long-Term Pavement Performance Computed Parameter: Frost Penetration



Importance of Frost Penetration Information

Knowledge of frost penetration beneath the pavement structure is critical for many pavement design, analysis, and management applications. Problems caused by frost include the seasonal change in the bearing capacity of soils brought by freezing and thawing. As subsurface temperatures decrease, the moisture in the unbound pavement layers freezes into ice that binds the aggregate particles together. Frost penetration leads to an increase in the strength and stiffness of the unbound pavement layers and subgrade soil. The process of ice formation also draws moisture into the freezing zone. When the frost thaws in the spring, the moisture increase in the soil can lead to weakened support for the pavement structure.

Another mechanical process associated with frost is the volumetric change in frost-susceptible soils, referred to as frost heave, which can lead to vertical differential movements of the road and subsequent poor performance. This heaving of roadbeds out of vertical alignment and breaking of the pavement surface often complicates highway maintenance.

Over the years, the National Oceanic and Atmospheric Administration (NOAA)(1) and Environment Canada(2) have developed and published the climatic maps containing historical frost penetration values, as well as the number of freeze-thaw cycles in the form of contour maps. These maps provide frost depth estimates for natural (uncovered) land in the United States and Canada. The frost penetration conditions under pavements may be different from that of exposed land surfaces. In addition, deicing salts may have an effect on frost penetration as they eventually dissipate into the soil.

Seasonal Monitoring Program

To provide the transportation community with the data needed to understand the magnitude and impact of diurnal, seasonal, and annual variations in pavement properties and responses, including the effects of frost penetration beneath pavement section, the Federal Highway Administration (FHWA) Long Term Pavement Performance (LTPP) program selected a number of test sites throughout the United States and Canada for the Seasonal Monitoring Program (SMP).

The original SMP (hereto referred as SMP I) included a total of 65 test sections and lasted from 1992 to 1999. As a part of the SMP I experiment, 37 pavement test sections were instrumented with electrical resistivity (ER) probes to monitor the frost penetration in unbound pavement layers. In addition, these sections were instrumented with time domain reflectometry (TDR) and temperature probes.

At the conclusion of SMP I, the LTPP team realized the need for additional monitoring of these sites and initiated the SMP II program. The objective of the SMP II monitoring was to continue providing the data needed to attain a fundamental understanding of the magnitude and impact of variations in pavement response and properties due to the separate and combined effects of temperature, moisture, and frost penetration. The SMP II included a total of 22 test sections and lasted from 2000 to October 2004. LTPP continued monitoring the ER trend as a part of the SMP II experiment at 12 test sites.

To aid in the interpretation of the ER data, an interactive computer program called FROST was developed in the late 1990s, and the available data were analyzed (see FHWA-RD-99-088 for more information on FROST).(3) FROST used ER data (voltage, contact resistance, and resistivity) in conjunction with soil temperature data to determine the depth of frost penetration in unbound layers for the SMP sections.

The results of frost penetration analysis are stored in two computed parameter tables in the LTPP database as follows:

The SMP_FREEZE_STATE table characterizes the freeze state as frozen or nonfrozen at each ER measurement depth. This information is useful for understanding or reevaluating the process by which the results presented in table SMP_FROST_PENETRATION were derived. The data in table SMP_FROST_PENETRATION translate the freeze state at each measurement depth into starting and ending depths of frozen layer(s). The SMP_FROST_PENETRATION table is the end product of the data analysis to determine the boundaries of frozen layers within the pavement cross section. These computed parameters tables contain information necessary analyzing the changes in pavement structural responses due to the seasonal changes in pavement layer properties.

These tables were updated twice with the new batches of the processed data: the first upload was based on the July 1999 version of the LTPP data for SMP I sections, and the second upload included SMP II sections based on the July 2001 version of the LTPP data. With the completion of monitoring measurements on the SMP sections in October 2004, there was a need to complete the interpretation of measurements not previously interpreted and to add the results to the database. In addition, through previous interpretation of SMP ER and soil temperature data, it became evident that the accuracy of the LTPP frost predictions could be improved by adding thermodynamic analysis capability to estimate missing temperature readings and by cross-referencing ER trends with moisture and temperature changes.


The objective of the current project was to update and complete the interpretations of frost penetration using measurements collected at the instrumented SMP sites. To achieve this objective, the project team was charged to review and enhance LTPP procedures for frost penetration determination. The enhanced procedures were subsequently used to critically review and flag questionable previous frost estimates and to complete the interpretation of frost depth in the unbound layers for new data and to update previous estimates.

The project was divided into two phases. The objectives of Phase I were to assess the existing LTPP frost penetration analysis methodology and frost penetration results, propose enhancements to the LTPP frost penetration analysis process, and develop a detailed research plan that would include the proposed enhancements for the frost penetration determination procedures. The objectives of Phase II were to implement these procedures in a software research tool, conduct frost penetration analysis, and develop new LTPP frost predictions using SMP data.


This final report documents the investigations performed during the study, describes the enhancements to the frost penetration analysis methodology developed, and summarizes the results of frost penetration estimates for LTPP SMP sections. Activities performed throughout the remainder of this report are discussed as follows:

In addition, the following two appendices are provided with the report:

 ContentsNext >>


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