WMOD00508 - Laserfiche WebLink

Home Browse Search

306 JOURNAL OF APPLIED METEOROLOGY VOLUME 29 PPI scans with seedline positions at 1905 and 1930 are shown in Fig. 11. The seedlines were originally re- leased in a widespread cloud with weak echo. A hole in the echo then developed on the west side of S2 be- tween S 1 and S2, Fig. 11 a. Then as the original echoes continued to dissipate, a weak echo appeared, which coincided with and was oriented along S2, Fig. 11 b. The echo appeared in S2 31 min downwind and then advected with the seedline. The reflectivity remained < 10 dBZ. Note that the width of the line echo corre- sponds to the width of the seedline, which was allowed to disperse at 1 m S-I, conforming to the aircraft ob- servations. The aircraft penetrated the echo observed in S2 at 1921 and again at 1930. Both penetrations were at the position along S2 shown in Fig. 11 b. This echo then remained with S2 as it passed over KGY 58 min after seeding. A similar pattern was not seen with S 1 or S3; however, natural echoes were in close prox- imity to these seedlines and would have masked any radar reflectivity resulting from seeding. Recall the small particles produced by seeding illustrated in Fig. 9 and that the experimental design suggested that seed- ing signatures may be apparent in radar data only in cases without a preexistent echo at the seedline. The second method used in radar analysis was to construct time-height profiles, with analysis boxes 3 X 3 km, close to positions of seedline penetration by the research aircraft. Time-height profiles keying on S2, with the positions of S 1 and S3 also noted, are shown in Fig. 12. Note that SI-S3 were laid out in various intensities of echoing cloud; however, at 14 min downwind only S 1 remained in echo at this lo- cation along the line. The distinct radar feature asso- ciated with S2 can be seen at to + 30 extending from 2 to 3.6 km. The position of the research aircraft in S2 is shown at to + 30 and to + 40. At to + 40 the echo in S2 extended 860 m above the seeding altitude, im- plying a vertical velocity of 0.36 m S-I. This agrees with the terrain induced vertical motion of 0.4 m S-I during this period. Notice from Fig. 12 that, with the 14.6 m S-I ad- vection speed used, S 1 began trailing the echo in which it was initiated. Line 3, however, which was initiated just ahead of a natural echo, kept this same relative position throughout the period. These measurements indicate that the natural echo associated with S 1 was advecting faster than S2 and S3 in agreement with the aircraft sampling. The natural echo near S3, however, was found to advect at the same speed as S2 and S3. This provides additional evidence that the winds were slowing with time. The relationships between these natural echoes and the treated regions will be referred to later when discussing surface observations. The question arises as to why a detectable radar sig- nature did not develop in S3 even though a portion of the line existed ahead of the natural echo and S3 had a distinct aircraft seeding signature. Reviewing Fig. 8 it appears that 82 was the only seedline that developed particles > 1250 /-Lm as evidenced by the 2D-P cu- mulative size distribution at 1930. It was these larger particles, apparently aggregates, that were producing the radar echo. Using a simple mass-diameter rela- tionship, m = 0.27 d2 (Mason 1971), the 1 L -I of I mm particles observed in S I and S3 would yield a re- flectivity of 3 dBZ. Y olume filling problems would probably have reduced this value to near 0 dBZ at the range of these seedlines. For S2, with 2 L -I of 1.5 mm particles, a reflectivity of 13 dBZ could be expected. Why aggregation was not occurring in S I and S3 is not obvious. Their proximity to natural radar features may have limited the liquid water available within these lines. The observations demonstrate the variability of particle growth environments within these clouds. (iii) Precipitation measurements The seedlines from both fixed target experiments crossed the precipitation gauge network. The gauges in the network were designed to resolve precipitation rates down to 0.1 mm in 5 min, or 1.25 mm h -I. This, however, may not be sufficient to resolve increases due to seeding. With a Z- R relationship for snow from Gunn and Marshall ( 1958) and values observed for S2 from the second experiment (4-6 dBZ), the precipi- tation rate would be 0.1 mm h-I, a factor of 10 below the gauge resolution. To determine appropriate sampling periods for each precipitation gauge the OTM was used to determine an advection speed and direction since it includes the wind field down to the surface. The result was 1960 at 13.6 m S-I and is appropriate for both experiments. The duration of effects at the ground is a function of plume width and plume advection. For a spread rate of 1 m s -I and plume advection of 10 m s -I, the du- ration of effects at a gauge 45 km downwind would be 5 min. Precipitation rates for gauges predicted to have been in the path of the seedlines are shown in Fig. 13. The predicted period of effect (PPE) at each gauge is noted for PI-P3 and SI-S3. Based on radar data (not shown), the placebo ex- periment was conducted in a large area of precipitation, consisting mainly of dendrites falling from high ele- vations. This accounts for the more continuous pre- cipitation observed during passage of PI-P3. The in- termittent precipitation observed during the second experiment could be expected from the radar data of Fig. 11. Note that Pine Nut, 26 min downwind of seed- ing, showed no measurable precipitation during the PPE even though the radar indicated that naturally precipitating clouds passed overhead. The radar echo which developed along S2 passed directly over West- ville, 38 min downwind, and precipitation was mea- sured within the PPE. The OTM predicted initial fall- out of seeded crystals 44 min downwind of the seedline, making it plausible to observe seeding effects at West- ville. The natural precipitation regions which coincided