HYSPLIT Verification Results Using DATEM

Results Overview

The statistical results from running HYSPLIT for each of the tracer experiments in the DATEM archive are summarized in this section. If available, the 32 km North American Regional Reanalysis meteorological data were used for the computations, otherwise the 2.5 degree global reanalysis data were used. Two RANK statistics are tabulated for each experiment, a comparision with the paired in space-and-time data and the temporally averaged data which are only paired in space. The link to the RANK gives the complete statistical summary for that experiment. See the DATEM report on computational details. A perfect model would have a rank of four.

  • ACURATE – consisted of measuring the Kr 85 air concentrations from emissions of the Savannah River Plant, SC. Twelve and 24 hour average air concentrations were collected for 19 months (March 1982 – September 1983) at 5 locations along the United States east coast from 300 to 1000 km from the plant (Fayetteville, NC to Murray Hill, NJ).
  • INEL74 – consisted of a little over two months of Kr-85 releases (February 27th to May 5th, 1974) from Idaho and continuous 12 h sampling (February 27th to May 4th, 1974) at 11 locations in a line about 1500 km downwind (Oklahoma City, OK to Minneapolis, MN).
  • ANATEX – consisted of 66 PerFluorocarbon Tracer (PFT) releases (33 each from two different locations – January 5th to March 26th, 1987) every two and one half days. Air samples were collected for 3 months (January 5th to March 29th, 1987) over 24 h periods at 75 sites covering most of the eastern US and southeastern Canada. PTCH (perfluro-trimethylcyclohexane) was released from Glasgow, Montana (GGW), and PDCH (perfluro-dimethylcyclohexane) and PMCH (perfluro-monomethylcyclohexane) from St. Cloud, Minnesota (STC).
  • CAPTEX – during September and October of 1983, consisted of six 3 h PFT releases (September 18th to October 29th, 1983), four from Dayton, Ohio and two from Sudbury, Ontario, Canada with samples collected at 84 sites, 300 to 800 km from the source, at 3 h and 6 h averages for about a 48 hour duration after each release (September 18th to October 30th, 1983).
  • OKC80 – consisted of single release of two different PFT tracers (July 8th, 1980) over a 3 hour duration with samples of 3 hour duration collected at 10 sites 100 km and 35 sites 600 km downwind from the Oklahoma City release point from July 8th to July 11th, 1980.
  • ETEX – although there were two releases, the primary evaluation data set is from the first release on October 23, 1994 from western France. Three hour duration sampling was conducted in western europe from October 23 through October 27th.
  • METREX – consisted of 6 hour emissions of perfluorocarbons simulaneously, from two different locations, every 36 hours. The tracer release locations were in suburban Washington, D.C, while 8 hour air samples were colleced at three locations within the urban area. The experiment ran for one full year. In addition monthly air concentration samples were collected at about 60 locations throughout the region.
  • SRP76 – consisted of continuous emissions of Krypton-85 from the Savannah River Plant near Aiken, SC. About 15 sampling locations were placed around the plant at varying distances (15 – 150 km) and directions. Weekly samples were collected between September 1975 and August 1977. Twice-daily samples were collected over four months (November 1976, February, April, July of 1977).


HYSPLIT4 (Revision 430)
EXPERIMENT Average Paired
ACURATE 3.44 1.96
ANATEX GGW 3.47 2.03
ANATEX STC 2.48 1.73
CAPTEX 3.03 2.23
ETEX 2.74 2.34
INEL74 2.35 1.97
METREX (pmch) 2.91 1.90
METREX (pdch) 2.64 1.83
OKC80 2.51 1.94
SRP76 2.15 1.05