WMOD00282 - Laserfiche WebLink

Home Browse Search

both raw and decoded format. The menu in the comment section identifies all variables in the named common lists. This style of having an unpacker of Level II data that loads the named common arrays with full radial data for each call is very convenient. It only takes a CALL GETRADAR to get whatever one might want to know about the next radial's data. PROGRAM RADAR7 Mter initiating the adaptable parameters, the Snow Algorithm calls up three files specific to the sites being considered. The terrain file was generated by "nearest neighbor" sampling from 30-arc-second data prepared by the Defense Mapping Agency, which has an original resolution near 1 km, similar to the radar resolution. A file could be prepared from similar 3-arc-second data if greater precision is desired, but that task is not a part of the work described here. The occultation file is the same as that used in the PPS Algorithm except that the first four bytes and all trailing bytes have been stripped to make a file that is easy to exaIiline under direct access. The hybrid scan. file is related to that produced by Tim O'Bannon for having the bottom of the beam near the terrain but at least 500 feet above it. The expanded sector portion is being ignored; this version of the Snow Algorithm does not try to substitute for reflectivity data in range bins at the edges of cluttered regions. The remaining hybrid sector data for the four lowest tilts were merged into one 230- by 360-array with numbers 1,2,3, and 4 indicating which tilt (0.5, 1.5, 2.5, or 3.50) to use for each range bin. All three site files (terrain, occultation, hybrid scan) are then easily examined by direct access when needed, avoiding the need to reserve program memory for their full arrays. The Snow Algorithm uses a file name convention that expects an increment of the file name extension from 2 to 401, representing the file numbers on a Level II tape that are found after the header file. The Snow Algorithm expects a set of files available on a removable or fixed hard disk. The Snow Algorithm calls each file in sequence until the series has no more files left. The algorithm then asks for a new file name for a continuation of the processing. The operator can thereby step over a gap in the series or insert a different removable disk or Level II tape when the next expected file is not found. The rewrite of the Snow Algorithm can use whatever scheme is convenient for obtaining the next file of the series. After loading and initializing adaptable parameters and site-specific files, the Snow, Algorithm first calculates the vertical wind profile, needed for the advection scheme. The Snow Algorithm operates basically the same as the current version used by all WSR-88Ds. The software modifications are merely a condensation by removing white space and trivial comment lines and a substitution of numbers for named constants so that a reader might better understand the processes. The functional modifications allow 4 velocity measurements per radial rather than 1 and use 10 km rather than 30 km for the default range. The Level II tape data file on disk is then rewound. The entire file must be checked to determine the wind profile and the ending time for the scan, all before considering the reflectivities. Both the wind data and the reflectivity data occupy the same /STORE/ named common and therefore must be loaded sequentially. The process could probably be speeded up in the rewritten version that does not have memory limits by loading the four reflectivity arrays while the wind data are being examined. That procedure would prevent the necessity for a rewind and a rereading could be avoided. The Snow Algorithm then calls GETARI to load the reflectivity arrays. Notice that the dimensions are to 3700, of which 367 are typically needed. Unlike the PPS version, this Snow 42