The 22nd Conference on Hurricanes and Tropical Meteorology, held in Ft. Collins, Colorado, 19-23 May, 1997, provided the most recent opportunity for a meeting of the Working Group.

The meeting convened at 8 pm in the Kansas State Room of the University Park Holiday Inn. The meeting was co-chaired by Frank Marks (AOML/HRD) and Colin McAdie (NWS/TPC). Attendees are given below.

Frank Marks gave a brief history of the Working Group for new attendees. He pointed out that both a statement of operational need and two hours of digital data (Archive II) recorded during H. Erin had been made available on the TPC ftp site in an initial effort to encourage developmental work. Progress has been slow due to a lack of funding; however, it was noted that Paul Harasti, U Toronto, had been able to use the H. Erin data in a test of his PCA technique, which shows promising results. Considerable time was spent in dealiasing, according to Paul. See p. 651 of the preprint volume.

Colin McAdie continued with a review of Working Group objectives, the primary of these being: 1) to facilitate the migration of needed algorithms into the operational environment, and 2) to enhance the utility of existing algorithms in diagnosing meso-scale features found within tropical cyclones. McAdie pointed out that the OFCM National Plan for Tropical Cyclone Research and Reconnaissance (1997-2002), in its research objective priority rating for warning centers, lists a high priority for Doppler radar algorithms by all three operational centers (TPC, CPHC, and JTWC) (p 3-2, FCM-P25-1997).

As far as the implementation of these algorithms, the ‘re-host' (the migration of WSR-88D software to a UNIX platform) has become central to the objectives of the Working Group. The re-host is motivated by the need for broader data access and by the need for local in-house ability to develop, modify and test existing or new algorithms. McAdie also presented the concept of a WDSS at TPC with suitable ("switchable") T1 connections to a relevant subset of remote sites, via RIDDS. TC algorithms would run on this platform.

Scott Spratt of the Melbourne Forecast Office presented an overview of the WDSS, based on a year of usage in that office. RUDDS does the actual algorithm processing after taking the data from the RIDDS. Their office is already running the improved NSSL algorithms on RUDDS. Some of the new (and desirable) capabilities are: the ability to obtain a data read-out from the cursor, display of trends, and time-height cross-sections. At the close of Scott's presentation, there was discussion about the fact that TPC would take a wide-band feed from the Melbourne RIDDS, which would not affect whatever they were running locally.

Mark Fresch, of the OSF Algorithm Branch then presented the OSF perspective on algorithm implementation and system change. He noted that priorities for development work are determined by the NEXRAD Technical Advisory Committee. The TAC maintains a priority list which now places tropical cyclone techniques 12th on a list of 14 priorities. This low priority has been a hindrance to progress. The next TAC meeting will be held in Anchorage in July; this represents an opportunity to present a specific, low-cost plan of action to the TAC, the goal being to allow OSF to release limited funding for tropical cyclone algorithm evaluation.

The concept presented by Fresch, and endorsed by the WG, was that a Request for Proposal be constructed. (The following contains input from the group) The RFP would specify needed algorithms, define a test data set, specify a UNIX platform, and a point-rating system. Algorithm results are to be verified against aircraft reconnaissance data. The RFP would specify a dollar amount for work to be done, the deliverable being the algorithm numerical/graphical output derived from the test data set. Respondents would be assured on an objective evaluation. Based upon the evaluation results, the OSF would then be able to recommend that winning algorithms be turned over to (most likely) NSSL for fitting out and implementation on the WDSS. TPC would then be able to run them operationally.

The group agreed that Marks, McAdie, and Fresch should write the RFP. Although a draft process can commence at once, the actual issuance of the RFP will hinge upon approval of concept by the TAC, which in turn allows release of OSF funds.

There was discussion about the need for VCP 22. The Working Group gives strong, unanimous endorsement for the earliest possible implementation of VCP 22. This can be an invaluable tool in the diagnosis of (for example) tornadic spin-ups within hurricane rainbands. Mark Fresch has action item to check on status of VCP 22. (His report, 6/6/97: VCP 22 will not be implemented for Build 10. It will probably not be implemented until Open Build 2. Contents of Open Build 2 will be determined in an August PMC meeting. He recommends that regional and program focal points be contacted now for a hearing at the August meeting.) (CJM..will draft note to Randy Racer from WG 6/9/97) (done: 7/16/97 CJM)

The group requested that all of the acronyms used during the meeting be defined on the WG homepage. McAdie agreed to carry this out.

In closing comments, Frank Marks suggested that the TPC-proposed model (processing wideband at a center for specific applications) represented a departure from the single-site approach that has predominated the 88D program thus far.

The meeting was adjourned at approximately 10 pm.

return to text of minutes

Attendees: