Deep Frost Depths - March 2003

The relatively deep frost of the 2002-2003 winter is something that Minnesotans have not dealt with in about 10 years. The depth of frost depends on many factors. The frost depth depends on the winter air temperature or the structure of soil itself. Another factor is whether the frost is measured under sod or under bare ground. The frost is deeper under bare ground. At the University of Minnesota St. Paul Campus Climate Observatory the maximum frost depth was reached around March 11-15, 2003 with about 31-32 inches under sod compared with 52 to 53 inches under bare ground (but still covered with snow.) The depth is deeper where the snow has been cleared in areas such as roadways and sidewalks. Thermocouples (temperature sensors) have also been inserted in the ground at a few locations such as the University of St. Paul Campus Climate Observatory, and at the University of Minnesota Southern Research and Outreach Center at Waseca.

Another method of measuring the depth of ground frost is the use of a frost tube. A hole is bored or drilled into the ground perpendicular to the surface. Most frost tubes have a depth of 60 inches. A plastic PVC pipe is inserted in the ground and a clear tube is placed inside the plastic pipe. A liquid is then placed inside the tube that will change color and freeze when the temperature drops to 32 degrees. The tube is sealed at both ends and has increments in inches or centimeters marked on it. During the winter, the observer simply pulls the tube out of the PVC pipe to estimate where the liquid has frozen and changed color. Some places in northern Minnesota reached the maximum depth measured by frost tubes with 60 inches at the Duluth National Weather Service and Grand Forks National Weather Service during the first half of March.

frost depth map There are only a handful of sites that measure frost depth in Minnesota. An example of these are on the map to the right. The majority of these sites are monitored at the major dam sites by the U.S. Army Corps of Engineers St. Paul District Water Control Center. Others are measured by the University of Minnesota and the North Central River Forecast Center. The depths shown here are not the deepest frost seen this season, but does give a good idea of how many places record frost depth.

Checking the historical data at St. Paul, the 31 inches of frost depth in 2003 is similar to the depth observed in March, 1991. In 1990, the frost depth reached a maximum around 41 inches in early March. A study at the St. Paul campus conducted by University of Minnesota Graduate Student Dave Wall looked at frost under bare soil from the winters of 1964-1965 to 1982-1983 (excluding 1973-1974). The average deepest frost penetration was 38.2 inches with the maximum depth reached around February 25.

In the spring, the ground thaws from the top and bottom and the last frost to leave the ground is in the middle. The average last frost was around April 5th at a depth of about 20 inches.

Another study conducted by University of Minnesota Professors Donald Baker and James Swan looked at the maximum depths of frozen soil in inches from 1959-1965. They estimated probable maximum frost depths for the middle of February from sometimes unusual sources such as grave diggers at cemeteries. The measured maximum depth under sod for St. Paul from 1961 to 1965 was 60 inches, and under bare soil the deepest measurement was 64 inches. The estimated maximum frost depth was between 60 and 72 inches for most of the state, with the exception of along the Iowa border, where the estimates were around 48 inches.

An investigation by St. Cloud State undergraduate Dan Pokorny in 1993. This study looked at frost depth in the 1970's and 1980's. The deepest frost found at the St. Paul site that was not cleared of snow during these years was around 45 inches in 1977, 1978, and 1980. Farther north some of the more deeper frosts reported were at Northome and Brainerd with frost depths around 57 inches in 1978.

The Minnesota Department of Transportation has some excellent frozen soil profiles that are used for seasonal load limits on roads.

SCOLIN5.GIF (2527 bytes)

- Return to the Climate Journal

Last modified: March 21, 2003