About Soil Temperature Climatology


Description:

The Soil Temperature Climatology Tool is a product developed by the Midwestern Regional Climate Center (MRCC) in collaboration with the USDA Midwest Climate Hub to provide historical soil temperature statistics based on the period 1991 to 2020 for the north-central United States. This tool can be used to answer such questions as:

  • On average, on what day of the year does a location’s soil temperature warm above or cool below 50°F?
  • What would be considered an early date for a location’s soil temperature to warm above 50°F?
  • What is the record-latest or record-earliest date when a location’s soil temperature has cooled below 50°F?
  • What is the distribution of dates when a location’s soil temperature has warmed above 50°F?
  • On what date in a given year in the period 1991 to 2020 did a location’s soil temperature warm above or cool below 50°F?
  • What is a location's average soil temperature on a selected date?
  • What would be considered a location's low or high soil temperature on a selected date?
  • On a selected date, what is a location's record-lowest and record-highest soil temperatures, and when did they occur?

Definitions:

  • Soil Temperature Threshold: Climatological information for several soil temperature thresholds is displayed by the tool. These thresholds include 30°F, 32°F, 35°F, 40°F, 45°F, 50°F, 55°F, and 60°F.
  • Period of Record: The tool covers the 30-year period from 1991 to 2020.
  • Location: Soil temperature data are provided for polygon-shaped areas that are roughly spaced on a 32-km (20-mi) resolution grid based on the North American Regional Analysis (NARR) model. More information about NARR is provided in the Data section below.
  • Date When Temperature Warms Above (“warm above”): This is the date when a location’s soil temperature warms above a selected temperature threshold and remains above that threshold until the next cold season. It cannot be any later than September 30th.
    • Earliest: The earliest date in the period of record when a location’s soil temperature warmed above the selected threshold.
    • Early: Statistically, “early” is defined as the 10th percentile of all “warm above” dates in the period of record at a location for the selected threshold. In other words, only 10% of all “warm above" dates in the period of record occurred earlier than the “early” “warm above" date.
    • Average: The statistical average of all “warm above” dates in the period of record at a location for the selected threshold.
    • Late: Statistically, “late” is defined as the 90th percentile of all “warm above” dates in the period of record at a location for the selected threshold. In other words, only 10% of all “warm above” dates in the period of record occurred later than the “late” “warm above” date.
    • Latest: The latest date in the period of record when a location’s soil temperature warmed above the selected threshold.
  • Date When Temperature Cools Below (“cool below”): This is the date when a location’s soil temperature cools below the selected temperature threshold and remains below that threshold until the next warm season. It cannot be any later than March 31st.
    • Earliest: The earliest date in the period of record when a location’s soil temperature cooled below the selected threshold.
    • Early: Statistically, “early” is defined as the 10th percentile of all “cool below” dates in the period of record at a location for the selected threshold. In other words, only 10% of all “cool below” dates in the period of record occurred earlier than the “early” “cool below” date.
    • Average: The statistical average of all “cool below” dates in the period of record at a location for the selected threshold.
    • Late: Statistically, “late” is defined as the 90th percentile of all “cool below” dates in the period of record at a location for the selected threshold. In other words, only 10% of all “cool below” dates in the period of record occurred later than the “late” “cool below” date.
    • Latest: The latest date in the period of record when a location’s soil temperature cooled below the selected threshold.
  • Temperatures on Selected Calendar Date: This is a location's statistical average soil temperature on a selected date in the period of record.
    • Lowest: The lowest soil temperature on a selected calendar date in the period of record at a location.
    • Low: Statistically, “low” is defined as the 10th percentile of all a selected calendar date’s soil temperature values in the period of record at a location. In other words, only 10% of all soil temperature values in the period of record on the selected calendar date were lower than the “low” soil temperature value at the location.
    • Average: The statistical average of all of a selected calendar date’s soil temperature values in the period of record at a location.
    • High: Statistically, “high” is defined as the 90th percentile of all a selected calendar date’s soil temperature values in the period of record at a location. In other words, only 10% of all soil temperature values in the period of record on the selected calendar date were higher than the “high” soil temperature value at the location.
    • Highest:The highest soil temperature on a selected calendar date in the period of record at a location.

Data Source:

The tool uses daily average 10-cm (4-in) soil temperature data available on a grid that is roughly spaced at 32-km (20-mi) from the North American Regional Reanalysis (NARR), produced by the Physical Sciences Laboratory (PSL) in the National Oceanic and Atmospheric Administration (NOAA).

NARR simulates soil temperature using a coupled Noah Land Surface Model (LSM), which is a community model developed in collaboration with the National Centers for Environmental Prediction, Oregon State University, the U.S. Air Force, and the Hydrologic Research Laboratory. The Noah LSM simulates land-atmosphere interactions, including surface evaporation and energy, and soil conditions. Simulated soil data represents conditions under a vegetative land cover. More detailed information about how soil temperature is simulated by NARR is given by Leeper et al. (2017), Mitchell (2005), and Chen and Dudhia (2001).

Data Processing

Files containing the 10-cm daily average soil temperature data in monthly increments from January 1991 to December 2020 were downloaded from the NARR website. The daily values were composited into seven-day running mean values. For example, the seven-day running mean value corresponding to 02/10/2020 included the seven daily values from 02/04/2020 to 02/10/2020.

From the calculated seven-day running mean values, only those values for grid locations covering the north-central United States were extracted, corresponding to a region stretching from 77°W to 102°W longitude and from 35°N to 49°N latitude. The resulting values were subjected to a bias analysis in which 10-cm seven-day running mean soil temperature values from nearby weather stations were compared to nearby NARR grid locations for the 2018 to 2022 period. Weather station networks used in the analysis include Enviroweather, Missouri Mesonet, Iowa Environmental Network, Hennepin West, Purdue Mesonet, South Dakota Mesonet, North Dakota Automated Weather Network, Illinois State Water Survey, Kansas Mesonet, Nebraska Mesonet, and Indiana Water Balance Network. The analysis led to a domain-wide bias value being calculated for each day of the year. These domain-wide bias values were then added to each respective day’s NARR grid values, resulting in bias-adjusted NARR soil temperature values.

The bias-adjusted NARR soil temperatures were then used to calculate annual “warm above” and “cool below” dates and other statistics (see Definitions section above).

At some locations in the southern part of the domain, the soil temperature does not regularly cool below the 32°F or 30°F thresholds, so no shading will appear on the map in those locations. Additionally, there are a small number of coastal locations around the Great Lakes where soil temperature data are not available from NARR. For those locations, data from an adjacent location farther inland from the coast is used.

References:

Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface-hydrology model with the Penn State-NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev., 129, 569–585.

Leeper, R.D., J.E. Bell, C. Vines, and M. Palecki, 2017: An Evaluation of the North American Regional Reanalysis Simulation Soil Moisture Conditions during the 2011-13 Drought Period. J. Hydrometeorology, 18, 515-527.

Mitchell, K., 2005: The community Noah land-surface model (LSM) user’s guide, version 2.7.1. NOAA/NCEP Doc., 26 pp. [Available online at https://ral.ucar.edu/document-or-file/noah-lsm-users-guide.]