aurbo
New Member
Posts: 43
|
Post by aurbo on Apr 26, 2013 22:36:31 GMT -5
|
|
|
Post by kcfan on Apr 29, 2013 12:29:41 GMT -5
What I've been hearing lately is that someone (Navy/NWS?) is working on a forecast model that has the capability to provide a TAF without human intervention. This could be the waive of the future towards reducing the signature of the NWS. I see a couple glaring problems with this idea:
1. The model accuracy is entirely based on the accuracy of the data being ingested into the model runs. If CWO's are removed, then the quality of the forecasts will diminish exponentially.
2. Most of the models don't reflect local topography and micro scale climatology and those that do have serious limitations.
|
|
|
Post by weatheri on Apr 29, 2013 12:39:44 GMT -5
kcfan, Someone at Willow Grove, NAS (KNXX) had software years ago that did that and they took that info into account before issuing a TAF. It gave them a look via TAF what the model was forecasting. It was developed by one of their forecasters.
|
|
|
Post by kcfan on Apr 29, 2013 12:53:27 GMT -5
Weatheri, thanks for the info. I do know the technology has been around for awhile, but it was never accurate or useful enough to replace human forecasters. I'm hearing there is a push to improve the accuracy of the products to be more useful. Either way, I don't see how they can "safely" cut both the CWO program and implement a fully automated forecasting program. One without the other will be a complete mess. Cutting the CWO program will ultimately reduce the accuracy of forecasts no matter how they look at it because observational data is the foundation from which all forecasts are derived.
|
|
|
Post by alstein on Apr 29, 2013 16:01:05 GMT -5
What I've been hearing lately is that someone (Navy/NWS?) is working on a forecast model that has the capability to provide a TAF without human intervention. This could be the waive of the future towards reducing the signature of the NWS. I see a couple glaring problems with this idea: 1. The model accuracy is entirely based on the accuracy of the data being ingested into the model runs. If CWO's are removed, then the quality of the forecasts will diminish exponentially. 2. Most of the models don't reflect local topography and micro scale climatology and those that do have serious limitations. Most military forecasters who do TAF's/ (E-3's with little experience) go heavily off the model- especially if they're not talented at the job. Much of that is the flack you get if you try to predict something off the model and are wrong.
|
|
aero0
New Member
Posts: 29
|
Post by aero0 on Sept 7, 2014 10:51:12 GMT -5
The General plain language forecast is currently made by a nws computer along with the TAF, but there is a forecaster to make any changes necessary before the forecast is officially issued. The voice on NOAA WX radio is a computer voice. Computers can make all forecasts including Fire Wx, Marine, Ag, general and aviation with occasional human changes. There actually is no reason to have television personalities on TV, the computer could make an accurate visual and audio of the forecast in detail. In some countries of the world weather offices have reduced personnel from 4 employees on shift to one and let their automated aviation observation run on its own most of the time, including the U.S. Air Force with their FMQ-19 observation computer which often reports precipitation that doesn't exist with no apparent correction made by the observer on duty. The latest I've heard on the FMQ-19 is that the USAF observer/forecaster doesn't edit the obs at all no matter how poor they are.
An ATC Tower in a major city in the West recently had 20 hours of aviation weather obs missing because the Tower personnel didn't know they were suppose to call the FSS or use AISR to transmit the ob when ASOS comms go down, or so they said.. and very few people even noticed because of good weather.. but the nws and Military still needs all the observational parameters for their models.
my 2 cents worth
|
|
|
Post by alstein on Sept 7, 2014 11:29:21 GMT -5
I don't know how it is today, but in 2007 the AF had a mixed bag in their weather offices, and much of the reason for increased automation was due to deployments of CONUS personnel to Iraq, which heavily hit the base weather stations, and they had hubs to do their forecasting. Also most bases aren't 24-hr aviation, they close on holidays and weekends. Only about 1/4 the bases in SECONUS stayed open 24/7.
An E-3 fresh out of weather school had to write 8 TAFs for 4 bases a night on average, and amend them , and the AF had stricter amend status than NOAA does.
No one notices breakdown in the obs when things are good, it's only when things go wrong that it gets noticed. In that sense, we are emergency personnel.
|
|
|
Post by vortlobes on Aug 1, 2015 10:24:43 GMT -5
The NWS and the Navy have had the Automated TAF in use for years. The Plain Language Forecasts are now also put out by the computer with Human oversight, based on the Grid system that was put in place around 2003-2004.
|
|
|
Post by hlsto2 on Aug 22, 2015 1:04:44 GMT -5
and if the FAA gets it's way...closing 57 CWO's and firing hundreds of professional observers (that know what they are doing) will also affect forecasts
|
|
|
Post by tamrich on Sept 9, 2015 0:09:20 GMT -5
The federal govt, both the FAA//FSS and the NWS, make up their own rules and even change them when it fits their current narrative. I'm in weather, at a remote site in Alaska, and work for the feds (I won't say who as management may be members on your forum). For example, the airport I work at now is a Level D field but was a higher level several years ago. This means that ASOS can run by itself (we are told explicitely) and no human needs to augment it. Recently we had aviation obs taken away from us and given to FSS. When we question what a poor job the current FSS personnel do with the wx we are told that they do not even have to edit the wx, that we are wrong to criticize FSS as they are supposed to leave ASOS alone and only augment when a sensor is msg or has failed. That we (the former certified observers) are wrong and are the problem. The feds create their own bureaucracy, rules and regulations, then modify them as needed, to fit their new rules. Another example, now the NWS no longer warns the public about any new watches, warnings or advisories. Doing so, even after decades of past practice when they did, is now against their mission. Call the new rules with new regulations "Weather-Ready Nation." Now the public must look to see if there are any warnings in their area. One cannot trust the NWS or FAA/FSS. Even though ASOS really does not depict the current wx as well as a human taking manual observations, the feds will 'define' the weather in such a way that one cannot argue with them. By saying that ASOS is correct and the only accepted wx at any particular field with an ASOS or even AWOS, they are closing debate on the subject. Now that wx observing has pretty much become automated, what is next is forecasting. In a generation or two there will be far less humans involved in wx forecasting. It will go the way of the weather observer/meteorological technician just not as soon as the former. What I believe is going to hasten the process is the notion that the federal govt is too big and needs to be cut back. The federal govt, and expecially the NWS, will nearly completely automate weather to a point where few, if any, of us will be in wx much longer. It's all really sad......
|
|
|
Post by lostsheep on Sept 9, 2015 10:25:05 GMT -5
Thanks for sharing this tamrich. Your spot on about the bureaucracy and the bull crap within!!! My gut still tells me we are working on our final contract and it will be over in a few years!!! Good luck up in your neck of the woods tamrich!!!
|
|