In Alaska, the Canadian Forest Fire Weather Index (FWI) system is used to track the effect of weather of fuel moisture conditions. The prescription of the FrostFire experimental burn (near Fairbanks) was based on the basic weather conditions as well as the FWI codes, which represent the moisture content of forest fuels. To determine the likelihood of the weather and fuels coming into prescription, the FWI codes for 31 fire seasons (May through August) were computed from the Fairbanks meteorological observations (Fairbanks was the closest station with a long record of weather observations). Specifically, the prescription called for the Duff Moisture Code (DMC) to the within the range of 59 to 71 with a desired value of 63, and the Drought Code (DC) to range from 300 to 475 with a desired value of 350. The Fine Fuels Moisture Code (FFMC) could range from 88.5 to 92, with an optimum value of 91. The FFMC was in the prescribed range and average of 7 of the 25 years for which data were available. There was no trend in frequency from May through the middle of August, with a decline in the last half of August. The DMC, however, never reached the acceptable range until 11 May, and after this date was within that range an average of 3 of the 31 years. After 1 August the DMC was in prescription much less frequently than earlier in the summer. The frequency of DC values exceeding 350 steadily increased from the beginning of the fire season through the middle of July. For the month of August, the number of years when the DC was in prescription were less then for July, yet still significant. Based on the these and other considerations, the window for the experimental burn was set for 23 June to 31 July 1999. Conditions came into prescription on 8 July 1999, and the fire was successfully ignited.