WMOD00409 - Laserfiche WebLink

Home Browse Search

5. GAPF plots were re-examined for questionable stations. Stations confirmed as having offsets larger than the maximum difference across the 48-hr average analysis were "turned off." Wind directions were not turned off because more tolerance due to the large local effects was possible. Wind speeds were corrected by adding a bias to the recalibration file. 6. Some final adjustments may be required after examining production runs. The above steps only ferreted out large biases and obvious technical problems. Small biases are undistinguishable from natural variability. These problems (if they do exist) must be looked for on a plot-by-plot basis and it will be up to the analyst to decide on the source; whether instrument error, data recording error, or actual meteorological signal. The periods used to determine offsets and periods providing candidate covariates are compared in Table 11.10. Table 11.10 Complexes. Periods Used to Recalibrate "PROBE" Data for Candidate Convective 12 June 1981 48 Hour Periods Used to Derive Offsets (GMT) 19'"20 June 1981 10'"11 July 10'"11 July Periods with Candidate Convective Complexes (GMT) 20 June 11 July 19 July 21 July 22 July 1 August 2 August 20"'21 July 31 July '" 1 August 3. Radar Convective Complex Definition Use was made of radar data analysis software developed earlier in HIPLEX by Reclamation and contracted groups. The SWR-75 radar located at Miles City provided data necessary to CCls. Guidelines used to define a CC are: (1) highest echo greater than 8.5 km msl; (2) at lest 30 dBz reflectivity some~time during its life, not necessarily at the same time of the 8.5 km echo height; (3) at least a 30 minute lifespan. Several years ago there was a separation clause as well, but it was dropped for this analysis. These criteria were applied to the composite PPI depictions produced by the University of North Dakota. 32