Dear Editor,

Greetings from the Jazz City.

This is Stephen J. Ahn working for the National Weather Service (NWS) in New Orleans/Baton Rouge, Louisiana. I am a meteorologist (forecaster) and have stationed in Los Angeles, CA.. San Francisco, CA.. HQs in Silver Spring, MD.. San Juan, PR.. and now in Slidell, LA.

I am not a research scientist, nor any scientific leader. However, I have supplemental field experience from the National Weather Service ranging from the Local Climatology to the Tropical Operations. I started my Federal Career in 1989. This was a couple of years prior to beginning NWS modernization process across the west coasts. Automation Field Operation System (AFOS) was the main computer system in network and several PCs were just installed in the Los Angeles office by that time. We even had XT and AT PCs with few local applications. Through the modernization, however, I am now capable to use Advanced Weather Interactive Processing System (AWIPS) and serving as an UNIX Systems Administrator. This has been possible under extensive training programs.

Numerical Modeling is an essential tool to predict upcoming weather events. Currently the NWS has several different models running under National Center for Environmental Prediction (NCEP) operations. For few example, Aviation model (AVN) runs for 72 hours. ETA and NGM models run for 48 hours. Medium Range (MRF) runs for 240 hours. However, all these models have a great amount of biases that individual forecasters require deep understanding of local region climatology and interactions of other aspects. Please note that the NWS forecasters often called the Nested Grids Model (NGM) as ``No Good Model'' (NGM). For the Hurricane Georges, for instance, AVN brings it too far north way up to the Bermuda(?) while NGM hanging around north of Cuba and at the same time ETA is out of business... dealing with unrealistic low pressure developing across the Gulf of Mexico.

Yes, timely operation of Numerical Model process is very important, but not mandatory requirement for forecasting. Observation is the mandatory requirement. Sometimes the NCEP Cray crashes and model outputs are delayed or not available until XX hours. On the other hand, one particular observation is not the only item to look at. Anyhow... The public forecast package provides up to 48 hours due to the available model outputs, but the accuracy is obviously getting lower with time. Such a small scale features cannot be detected by models. Thus we issue NOWcast dealing the weather events during the next 2 to 3 hours which the routine forecast products do not carry. Plus the NOWcast can provide more details in the specific areas of concern.

The key point here should be the COMMUNICATION. Even though the NWS issues watches and warnings, when the communication fails to reach our customer, To Whom The Bell Toll???

According to the recent report, our (NWS) warning accuracy for tornadoes has been improved from less than 40% to nearly 70%. The lead time for tornado warnings have more than doubled in the past 10 years - from less than 5 minutes to nearly 12 minutes in 1998. In addition, the false alarms for all severe weather events have dropped from about 65% to below 50%. The below 50% is an all-time low record.

In some offices, during a widespread tornado outbreak, more than 40 warnings with average lead times of 16 minutes are issued in about 5 to 7 hour period. Well... damages are extensive, but no lives are lost. The results came out from

These included but not limited to WSR-88D Doppler Radar, AWIPS, New generation of weather satellites... capable of multiple and detailed probings of the upper atmosphere. Scientific workforces more locally focused for forecasters. Partnerships with the academic and research communities. And finally the State-of-the-art workstations and communications systems.

In February this year in central Florida, 7 tornadoes ripped through residential areas between 11:00 pm and 2:30 am... killing 42 and injuring 260. As far as the hours of occurrence are concerned, few were watching TV. No outdoor sirens. No weather radios scarce. Just few received warnings. In March 1994 in Alabama, 20 killed during the congregation. Note that the warning was issued 12 minutes beforehand, but the people in the church never knew that a storm was coming. As a result, the Vice President jumped in a community outreach program. Mr. Gore has campaigned to expand the NOAA Weather Radio as common in homes and public places as smoke detectors. This expansion effort is to be funded by a public/private partnership, wherein the public would finance the purchase and installation of the transmitters with associated equipments and the NWS would provide operations and maintenance supports.

We, the NWS, anticipate an increase in tornado warning accuracy to at least 75 percent with lead time of at least 15 minutes by the year 2004. By 2004, severe storm warning accuracy will also increase to near 90 percent with lead times approaching to 25 minutes. However...once again... improved warnings are ONLY the first part of an effective life-saving mission. Earlier warnings never guarantee to save lives. THE JOB IS NOT REALLY DONE UNTIL THE WARNINGS ARE RECEIVED BY THE PUBLIC AND PEOPLE TAKE ACTION THAT SAVES LIVES.

On the other hand, I have to mention the implementation of a Local Office Team concept. In the local NWS Forecast Offices, the operations staff meet together as needed bases and discuss about the local office improvements. Implementations such as eliminating extra paper works could be a good example. We have reduced tremendous amount of paper works...many of them are duplicated works from the bottom line of so called beauracracy. Weather charts hanging on the wall have been gradually disappeared. Number of forecast products are eliminated or absorbed (or combined) by other products. Forecasters on duty are the sole responsibility to decide go or no go... and so on.

Finally, I would like to introduce the NWS Duty Priorities. When the workload in any forecast offices exceeds the capability of the staff, weather and flood warnings take TOP Priority over other duties. However, it does not replace the exercise of professional judgment. The duty priorities as follow...

  1. Warnings, Watches, and Advisories
  2. Mission-Critical Observations
  3. Forecasts, Observations, and Basic Weather Watch
  4. Non-Critical Public Services
  5. Training, Development, and Focal Point Duties

Above all, these are just my 2 cents worth as a forecaster of the U.S. NWS field offices. This does not imply for any other impacts at all.

Stephen J. Ahn, Meteorologist, NWSFO New Orleans/Baton Rouge
Mr. Stephan J. Ahn is also a member of the KASA (Korean Atmospheric Scientists in America).