Degradation of United States upper air network

[Tim Vasquez]

I apologize if this has been posted already, but I haven't been able to keep track of all the threads lately due to other ongoing responsibilities. Anyhow this item crossed my mailbox and my conclusion is that the United States upper air network is in grave danger of being rendered useless by poorly-tested radiosonde systems. This news is of astounding importance to anyone who uses upper air data in any form.

Robert Maddox, a veteran storm forecaster and NSSL director from 1986-1996, has independently studied the RRS (Radiosonde Replacement System) that the NWS is currently phasing in and has found major problems. Apparently these systems (built by Sippican, a Lockheed-Martin company) were brought in with almost no input from the meteorological community and they are reportedly producing significant amounts of erroneous data, including strange superadiabatic layers and frequent evaporational cooling problems when departing cloud layers. Furthermore Maddox has identified a lack of communication between NWS head offices and the meteorological community that has led to erroneous use of the data.

The full report can be found here:
http://www.squidinkbooks.com/madweather/rrs.htm
and newsworthy examples of problems are being reported here:
http://madweather.blogspot.com/

If this installation continues without further review, we are soon going to be facing very significant and unpredictable effects on everything from mesoscale model errors to problems with daily severe weather forecasting. I think this is serious enough where we may need to address this with the trade organizations (AMS and NWA) and perhaps start contacting our elected officials if these kinds of problems are validated.

Tim

 

[Tim Vasquez]

Example from Thursday night showing evaporation problem:

 

[Zac Flamig]

This was posted on here before, but surely it needs more attention.
Of course my thoughts on the matter then are much the same as they are now. I can't help but wonder if they're just trying to pressure the manufacturer to fix these problems by going operational with the system?
Its also a matter of how severe are these problems? Are they something that is fixable or are they a design flaw in the new system?
I'm of the general opinion that these are probably just growing pains with the new system and that they will be resolved eventually.

Of course, I could be completely wrong and theres a very real but seemingly unlikely chance that this is a systematic failure and that the NWS is headed down the wrong road.

 

[Rich Thompson]

I can tell you that I'm not aware of *any* discussions within the NWS, that involved forecasters, regarding the rawinsonde replacement system. We were told it would happen, but that's about it.

Not only are the measurements questionable, but the timing of the installations could not have been worse in many areas. I'm sure that somebody would get upset no matter when a sounding is replaced. Still, most of the replacement systems have been installed near the peak in severe weather frequency for each region (March across the Gulf states, May in parts of the Plains, June across the northern Plains, etc.).

Rich T.

 

[Doug Lee]

I knew a thread had been started on this before, because I remember replying to it:
http://stormtrack.org/forum/showthread.php?t=13051.