WMOD00430 - Laserfiche WebLink

Home Browse Search

JULY 1988 RAUBER ET AL. 813 'St ,=,3 '" 1;; w o / / ",~:ft~~" DRV ----:::z: ~ ~::'~Y:I;:II::.E:;alion D ot] (lr00.~r:::..0)~2:~I:II~I:II:III\.\ I :(l'lA~RY {::1.::/::::JII<.:::;J~:.:.?:''':'..''\Jlrl'I::IJ :'111' ~ '--------v----- Shallow moist zone ~ Time Seeding Opportunity FIG. 2. Conceptual model of a cross section through many Sierran storms. Stippled areas denote relative humidities above 90%. Most seedable regions determined from radiometric and aircraft observa- tions are annotated. The time of passage of a complete storm can vary significantly from as short as 12 to >48 h (from Reynolds and Dennis 1986). mountain width (-100 km). These features were not well represented by a single vertical sounding at the base of the mountain normally used to initialize models. 2) Sierra Nevada storms were seldom, if ever, steady state. Wind fields over the Sierra Nevada continually changed in storms in response to the storm's mesoscale and synoptic scale evolution. Significant changes often were observed in time scales as short as 3 h. By the 1984/85 season, it was recognized that si- multaneous rawinsondes were required at least every 3 h at two sites, one near the base of the Sierra Nevada Range and one near the Sierra Nevada Crest to monitor changes in the wind fields associated with mesosynop- tic-scale storm evolution. The valley launch sites were either at Sheridan, California (60 m MSL, collocated with the radar) or Lincoln, California (57 m MSL, 1986-87 season only). The high elevation launch site was at Kingvale (1800 m MSL), which was collocated with the ground microphysical measurements and within the fixed target area. These soundings provided measurements of the wind fields at the boundaries of the seeded volume at a frequency necessary to resolve changes due to the passage of synoptic-scale and me- soscale features. Additional soundings were provided by the research aircraft during takeoff over the valley and by supplemental soundings launched at Kingvale. The dual-sounding system was incorporated opera- tionally for targeting in SCPP beginning in February 1986. In this paper, the diagnostic technique developed to guide aircraft to the appropriate location to target seeding effects within the ARB using this dual-sounding system is described. Evaluations of the parameteriza- tions used to recover the wind fields across the Sierra Nevada and simulate ice particle growth are presented. Finally, we present a case study comparing radar echo evolution within a seeded cloud region with predicted particle trajectories. 3. Targeting methodology In this section, we describe the components of the targeting method used during the last 2 yr of SCPP. This targeting method was designed for operational ex- pediency. The program was developed for use on a minicomputer, but can operate on a personal com- puter. The total time required to run the program on either system and relay information to the aircraft was usually under 3 min. Rawinsonde data used in the computations were normally available from the field within 30 min after the termination of the sounding. Aircraft and radiometer data were available in real time. The methods presented here focus only on the muItiple-rawinsonde technique that was utilized suc- cessfully later in the program. a. Kinematics Wind speed and direction were measured in vertical soundings at two locations, Sheridan or Lincoln, in the Sacramento Valley, and Kingvale, near the Sierra Crest. These sites were separated by 85 km. Seeding was nor- mally conducted at or below the 3500 m level ( -650 mb). The "target" in all cases was Kingvale. The primary region where accurate winds were re- quired for targeting, designated as region A on Fig. 3, was bounded on the west by Sheridan, the east by Kingvale and aloft approximately by 650 mb (surface "D" on Fig. 3). Regions B, downwind of the crest, and C, in the upper troposphere, were normally outside the primary region of interest during airborne seeding ex- periments. However, it was possible that particle tra- jectories occasionally would fall into region B (trajec- tory 2 on Fig. 3) or initiate or pass briefly into the lower part of region C (trajectory 3 on Fig. 3). The procedure to determine winds in each region differed because of data availability. In region A, the primary region of interest for the experiment, winds were determined in the following way: The barrier-perpendicular (u) component was de- termined by assuming conservation of mass flux throughout the domain A. The mass flux perpendicular to the crestline below the 650 mb le:vel at Sheridan was measured directly from the sounding by calculating JP650 Ms = g-I u(P)dp. (1) Psh Here, Ms is the mass flux below 650 mb at Sheridan, P pressure, g gravity, and the subscripts sh and 650 denott~ the surface pressure and the 650 mb pressure at Sheridan. A similar integration was then performed at Kingvale to determine the pressure level over King- vale where the mass flux at Kingvalle, Mk, equaled Ms: Ms = Mk = g-I iPt u(P)dp. (2) Pkv Here PI is the pressure level at Kingvale below which Mk = M" and Pkv is the surface pressure at Kingvale.