Had a great time today visiting RLX (that is the ID code for the radar there). I got up about 8:30 and left at 9 for the hour long drive to Charleston. I missed my turn off the highway, but found my way there. It is at the very top of a hill on 119, just past the big shopping area of Corridor "G". I knew I was getting close when I saw the WSR-88D radome on the top of the hill. This isn't the one in Charleston, but this is what all the weather radars in the country look like:
I found the office, then the parking lot, but a sign said it was for employees only. I looked around for another place, but there weren't any around. "Oh, what the hell?" I'm allowed to be here. I parked in an empty place at the very end of the employee (the only) lot.
I rang the buzzer, and Chris let me in. He is the one I had emailed about coming. As it turns out, he is also the office's point of contact for aviation concerns in the area. He started talking to me and showing me around. The first thing I noticed was the technology in the room. One wall had 9 screens of various sized displaying some TV news/weather channels, and there were some more advanced weather products up as well. I think it important to note that the NWS forecasters do not sit around watching to see what Fox News says the weather will be. They keep those sorts of channels up in case something like a chemical spill occurs they can begin analyzing how it might spread right away. Wednesdays at the office are practice day for the weather radio alerts. While I was there, I saw each of the workers practice recording a message to be broadcast on the weather radio frequency. They do this so that if there is a real emergency, everybody remembers how to warn the public rather than having to contact the one guy who knows how to run the machine, or dig out the manual.
I sat down with Chris at his desk, and we began looking at some of the tools they have to work with. Geez, I though some of the imagery and weather products available online were good, but the stuff I have seen before is akin to putting a smiling sun over Dallas and an angry anthropomorphic rain cloud over Seattle compared to the sophisticated things these guys have access to.
I learned a great deal about their tools and the Weather Service in general from Chris. The RLX office, like most offices, is manned 24/7. Less at night, and fully staffed and very busy during thunderstorm season. I was primarily interested in learning about how the aviation weather products are created, and I was not disappointed. This is RLX's area of coverage, and there are 6 airports in the region that get their own unique forecast, a Terminal Aerodrome Forecast, or TAF:
The colored counties comprise the area that RLX is responsible for, and they issue 6 TAFs, for Beckley, Clarksburg, Elkins, Charleston, Parkersburg, and Huntington. A special program that runs on the short-term weather desk monitors current conditions, and alerts the forecaster if the TAFs are not corresponding to the real-world weather. The meteorologist can then look at them and issue amendments.
After about an hour, Chris sent me to work with Ken, who was assigned to the short term forecast desk today (there are 4 main workstations, short, medium, and long term forecasting, and another dealing with models and other data). Ken said he had been in the office there since 1977, so he had seen several waves of technology come and go. The workstations have 7 large computer screens so the meteorologists can see many things at once. Each screen can be setup to show what they want to see. Ken had one of his screens dedicated to TAF production, then radar, satellite, forecasting and model prediction, and the 5th was radiosonde data and forecasts.
The software they have access to is phenominal. As for collecting and analyzing data, they can overlay multiple images and model on top of one another. We looked at several combinations of satellite and upper air chart, multiple models can be imposed on top of one another to get a sense for how they compare, and they even have the computer power on site to run a high-resolution regional model tailored to the local area. Its too bad the local weather offices no longer are responsible for flight briefings!
All of the forecasting work is done with "grids." Basically, they have a different weather map for each hour of the day and each variable they forecast such as visibility, precipitation type, temperature high and low, etc. For the 7 day forecast, that is hundreds of maps, or grids! Each of the 3 above-mentioned forecasters has a time period to work on, short term does the next 24 to 30 hours, and new forecasts are released 4 times per day, sometimes with updates in between (especially when the weather is bad). When they have mapped everything to their satisfaction, the office uploads their grids to the national database, from whence the forecasts for the entire country are derived.
Apparently this grid system is rather new, because they said that not all offices have it yet, but it is really advanced and very powerful. In fact, it really aids in the production of very localized forecasts like TAFs. In the past, these were made by hand, but now they can just tell they system where to build one, and it comes up on the TAF editing computer.
The process works like this: we used the models, observations, radiosonde soundings, radar, satellite images, and yes, even pilot reports PIREPs, to work out the weather, and then Ken updated the grids to reflect the information. The grids are not tweaked models, they are totally programmed by the meteorologist based on all info available including local trends and personal experience. It was very helpful to me for Ken to explain to me what kinds of information he was pulling from each weather product, and I learned a great deal about advanced satellite and radar image interpretation - techniques I hadn't seen before, even in my forecasting course at Marshall. After the grids were updated and published, we pulled up the info in the TAF editing program. Since the TAF info comes from the grids, the Charleston office has the capability of showing detailed forecasts for any point in their area. They offer the same forecast data that makes up the 6 main TAFs for every airport in their area, even the grass strips. It is available on their page here. This data is not monitored or updated with temporary updates, as the official TAFs are, but it is potentially a very useful flight planning tool.
Also, I learned that all text-based forecasts come out of the grid system, including the Area Forecasts used by pilots. Something I did not know, however, is that each forecaster writes a discussion of their forecast once per shift. This talks about any error they think may come up, and includes an analysis of how good the forecast is. Sometimes the meteorologist may be very sure, sometimes not, but they have to issue one definitive forecast. I did not know this existed, but am very excited to start using it when making a tough go/no-go decision. These discussions are available under "Text Products," "Other Narrative Products," on the RLX web page, here.
They do, in fact, even go outside to get a sense of the weather. Sometimes people complain that forecasters "obviously haven't looked out the window," or "can't even get the first part of the forecast correct." Often times, they are issuing forecasts over large areas and long periods of time. In addition, they have to issue forecasts based on what is coming (obviously). If the sky is clear, but they can see a thick, low cloud shelf moving in, they may issue a forecast for the next hour for clouds, but the shelf may not arrive for 2 or 3 hours. It is a very tough business.
I asked about PIREPs, and everyone I talked to said they do look at them and wish there were more available. Reports are useful from any pilot (or even non-pilots, they complained about not having enough information about snowfall amounts in downtown Huntington.), about anything. Tops of clouds, or the top of the haze layer, or the temperature at any altitude, or even the fact that it is a clear day and the weather forecast was correct. The guys and gals there would be happy with any report, and I can now guarantee that any PIREP you guys may file will be looked at and used in increasing the accuracy of the forecast. I am going to try and make a habit of filing more PIREPs every flight. Not just on cross country trips, and not just when the weather is bad.
As I was leaving, Chris asked about any local pilot groups. He would like to make 3 or 4 presentations this spring to area pilots. I gave him the contact info for the FBO at the field, thanked them for their time, and went to lunch. By 1:30, I was starving. I walked outside into the snizzle (drizzle, only snow), as Ken has termed it, and started the hour drive home, wishing that I had access to that level of data and computer capacity every time I flew, it really makes the normal info available online look like kindergarten. Fortunately, the guys at RLX have pointed out a few more tools for me to use, I understand how the forecasts that I bet my safety on are created, and I have a lot more respect for the NWS. Based on the small facility size, small staff, and the huge benefits for all Americans; the NWS field offices are tax money well spent.