okm: Oklahoma Mesonet

General Data Description

The Oklahoma Mesonet (sponsored by University of Oklahoma and Oklahoma State) provide 05, and 15 minute averaged surface meteorological data daily in near real-time and quality assured data files with about a monthly delay. In addition, soil moisture sensor information is available in 30 minute averaged quality assured files.

The Oklahoma Mesonet network consists of over 100 automated observing stations located throughout the state. Station names, latitudes and longitudes can be found at the following URL : http://www.mesonet.org/index.php/site/

See http://www.mesonet.org/index.php/site/about/instruments for sensor specifications.

The State of Oklahoma restricts the sharing of these data with outside agencies or with research scientists who are not DOE/ARM employees or members of the Atmospheric System Research (ASR) community.

Measurement Description

Each file gives the station names, latitude, longitude, and altitude about means sea level for each station.


Parameters are sampled every 3 seconds except barametric pressure, which is sampled every 12 seconds. The averaging interval is 5 minutes. The parameter value for the average represents the time at the end of the averaging interval. The parameters measured at each station are:

  • relative humidity (percent)
  • air temperature at 1.5 meters (C)
  • average wind speed at 10 meters (m/s)
  • vector average wind magnitude (m/s)
  • vector average wind direction at 10 meters (degrees)
  • standard deviation of wind direction (degrees)
  • standard deviation of wind speed (degrees)
  • maximum 3-second wind speed (m/s)
  • accumulated rainfall since 00GMT except at 00GMT the number given is the total accumulation for the past 24 hours(mm)
  • pressure (hPa)
  • solar radiation (watts/m^2)
  • air temperature at 9 meters (C)
  • average wind speed at 2 meters (m/s)

Parameters are sampled every 30 seconds. The averaging interval is 15 minutes. The value for the average represents the time at the end of the averaging interval. The parameters measured are:

  • soil temperature at 10cm beneath natural sod cover (C)
  • soil temperature at 10cm beneath bare soil (C)
  • soil temperature at 5 cm beneath natural sod cover (C)
  • soil temperature at 5 cm beneath bare soil (C)
  • soil temperature at 30 cm beneath natural sod cover (C)
  • leaf wetness index (percent) **
  • site battery voltage (volts)
** the leaf wetness index appears in the .a1 files until 10/10/99 but the Oklahoma Mesonet scientists report that the sensor never worked properly and that the data should not be used. This measurement is not included in the .b1 level data.


Sensor temperature rise at 4 depths (C)

".b1 levels"

Quality control flags (values from 0-9) are available for each measurement in the 5 and 15 minute averaged data in the .b1 data streams. The flag values and their meaning are as follows:

  • 1 Suspect (there is concern about accuracy of datum)
  • 2 Warning (datum is very questionable but information can be extracted)
  • 3 Failure (datum is unusable)
  • 4 Not Installed Yet (station awaiting installtation of sensor)
  • 5 - (internal use only)
  • 6 - (internal use only)
  • 7 parameter not checked
  • 8 Never Installed (station is not intended to measure this parameter
  • 9 Not Applied / Missing Datum (datum is missing for this station and parameter or QA not applied)

Missing Data Values:

-99999 denotes invalid or missing value

Temporal Coverage

All data streams begin January 1, 1994 except for sgp30okmX1.b1 which begin January 1, 1998 to December 31, 1999 and October 2002.

Level a1 data are available at the ARM Archive approximately 4 days after they are generated. Level b1 data are available approximately one month after they are generated.

Area Covered

The area covered is the state of Oklahoma. Station names, latitudes and longitudes can be found at the following URL : http://www.mesonet.org/index.php/site/sites

There are more than 100 stations in the data. But the stations have been changed over the years. Below is the change log of stations from 1999-present.
GRA2 was changed from GRAN on 19990316
PORT was added on 11/5/1999 (replaced TULL)
PRES was taken out 1/20/2000
BEEX was added on 4/1/2001
CATO was removed on 4/19/2001
INOL was added on 12/1/2001
NRMN was added on 07/02/2002
CLRM was added on 07/10/2002
NEWP was added on 10/08/2002
BROK was added on 20030414
MRSH was added on 09/29/2003 as a replacement of MARS.
and MARS was taken out on 09/29/2003
ARD2 was added on 02/22/2004
FITT was added on 05/27/2005
OKCN was added on 03/13/2007
OKCW and OKCE were added on 04/22/2007.
CARL was added on 03/13/2008
WEBR was added on 04/28/2008 as a new station
KIN2 was added on 03/08/2009 as a new station to replace KING
HOLD was added on 05/29/2009 as a new station to replace CALV.
ANTL was replaced by ANT2 on 04/15/2011

Data Stream Inputs

sgp05okmX1.00 (01/01/1994-10/10/1999)
sgp15okmX1.00 (01/01/1994-10/10/1999)
sgp30okmX1.01 (01/01/1998 -12/31/1999, October 2002- )
sgpokmX1.00 (12/01/1999-)
sgpokmX1.01 (01/01/1994-)

Notification Form Link

sgpokmX1 Notification Form


Laurie Gregory
Brookhaven National Laboratory

Data Source

The Oklahoma Climatological Survey at the University of Oklahoma.

Reprocessing History

Reprocessing Date


Start End Reprocessing


Reason Reprocessing

sgp05okmX.b1 and sgp15okmX1.b1 covering 19940101-19971231were reprocessed on 05/02/2011. Below is the explanation on the reprocess from Nathan Bain at the Oklahoma Mesonet .
" The biggest change for this batch of data is all air temperature readings at 1.5 meters now have a flag of 2 applied to them for 1994 through early 1997. The reason for this change is as follows: The way we sampled the temperature/relative humidity sensors caused a residual voltage that impacted the temperature measurements. The residual voltage generally made the temperature readings a few tenths of a degree too warm. Sometimes the error was as great as 1.5 C, while other times it was a cool bias of up to 0.5 C. This problem was unknown to the engineering and scientific community at the time. We downloaded new logger programs in March of 1997 to remedy the error. This manuscript provides more details. Errors of a degree or so won't necessarily impact many of our users and products but we're concerned that someone doing detailed climate work will be confused by the early temperature data. In retrospect, we have a new perspective on what "climate quality" data means, and we have decided to flag those early temperature observations as erroneous.
The manuscript mentioned above is located at: http://ars.mesonet.org/news/OK_Mesonet_Document_Chris_Fiebrich_1.4.2011.pdf "

Reprocessing Date: 20130530
sgp30okmX1.00 for 19980101-1991231 were reprocessed. The quality assurance Meteorologists have re-analyzed the data and released them to the public. The data between 20000101-20020930 will be reprocessed later.
Reprocessing Date: 20130806
The sgp30okmX1.01 data between 20000101-20020930 were reprocessed.

Reprocessing Date: 20140218

The OKM has completed reprocessing data for several periods of time. These include the previous year's rerun that they do every year, completed re-analysis of 2004, and several dates where they recently discovered some data quality problems. In most cases, only one variable for one station has changed. The full list of dates with updated files are:
1994: April and May
1995: Entire year
1996: Entire year
1997: January, February, March 1-2, June 18-19
1999: August 6
2000: June 12 through July 7
2004: Entire year
2006: August 22-23
2007: April 11-12, June 18-19
2008: June 27-28, August 12-13, October 7-8
2009: October 16-17
2010: September 9-10
2013: Entire year

Reprocessing date: 20140922
The data of 2005 were reprocessed with QA reanalysis. The data of 201303-201408 were reprocessed to fix the missing observations cause by the ingest bug.
2005: January-December
2013: March-December
2014: January-December

Data User Notes

Users of these data should use the level "b1" data rather than the "a1" data.

There are DQRs available for each month of data beginning with January 1995. These monthly DQRs give an overview of the data quality. Users should review the contents of these reports. There are other DQRs available as well which document changes in station names and locations. All DQRs can be found at: the PIFCARDQR web site.

The soil moisture sensors were not commissioned until October 2002, so data prior to then are considered experimental. Oklahoma Mesonet will provide data from prior to October 2002 on a case-by-case basis, giving their QA meteorologists a chance to review it before sending it out. Oklahoma Mesonet personnel are working on a full QA evaluation of soil moisture prior to the commissioning date, but it is still a work-in-progress (2005 March).

The measurement "sensor temperature rise" in the sgp30okmX1.b1 files is the temperature dissipation from the heating pulse. Soil water potential, fractional water index and volumetric water can be derived. It is more fully described at

Matric Potential can be caculated from trise 
 MP = -(c*exp(a*TR))/100
where MP = Matric Potential(bars)
      TR = DeltaT reference, or 'trise'(degree C)
       a = 1.788
   and c = 0.717
Soil water content may also be calculated from TR. User should contact the Oklahoma Mesonet Manager for suggested equations and site-specific parameters.

On 07/12/2011, entries/columns associated with Oklahoma Mesonet soil moisture sensor at 75 cm were removed from the data.


DQRData Quality Report
OKMOklahoma Mesonet
SGPSouthern Great Plains

Citable References

Brock, F.V., K.C. Crawford, R.L. Elliott, G.W. Cuperus, S.J. Stadler, H.L. Johnson and M.D. Eilts. 1995. The Oklahoma Mesonet: A Technical Overview. Journal of Atmospheric and Oceanic Technology 12(1):5-19.

Arndt, D.S., M.A. Shafer, S.E. Fredrickson and J.P. Bostic. 1998. Quality Assurance at the Oklahoma Mesonet: A Systems Viewpoint. Preprints, 10th Symposium on Meteorological Observations and Instrumentation; American Meteorological Society, Phoenix, Arizona, January 11-16.