Expanding warning times for tornadoes....

[rdale]

Sorry - you lost me there. I'm not sure any of that team actively posts here during events, my guess is you're more likely to see results later on.

http://ewp.nssl.noaa.gov/

 

[Mike Johnston]

Sorry - you lost me there. I'm not sure any of that team actively posts here during events, my guess is you're more likely to see results later on.

http://ewp.nssl.noaa.gov/

No, a lot of members post forecast and NOW comments here during events, and some in the post-chase discussion threads. I've never seen a storm-scale numerical forecast model cited...maybe I've missed it. So, if there really are elves behind the curtains with better tools, then let's see it! Or, as the people of Missouri would say: SHOW ME!

 

[rdale]

Unless Greg is ready to say otherwise - none of that experimental stuff goes to the public in realtime (unless you want to run your own model, and feel like disseminating that?)

 

[Patrick Marsh]

During our (NSSL) yearly spring program, one of the activities is analyzing and evaluating high resolution model output from assimilated observations. This analyses are created every 5 minutes. There is also the HRRR project which is generating storm scale forecasts every hour.

 

[Mike Johnston]

Unless Greg is ready to say otherwise/ - none of that experimental stuff goes to the public in realtime (unless you want to run your own model, and feel like disseminating that?)

Oh, so you challenge me with "have you looked at their work?" then tell me it's not availabe to the public. That's classic bureaucratic two-step. I keep up fairly well with the academic papers on tornadogenesis, and understand that, while there are some promising hypothetical suggestions, we really don't understand it yet. Yet now, you come out as an advocate for completely changing the tornado warning system - with the keystone of such a change being an experimental numerical weather model that isn't publicaly available? And - it's supposed to be based upon sociological research, yet you lampoon Joe and Jane public (and those mad radar operators, as if they were clowns)?

Again, tell me about how an operational forecast - equal or superior to the current system - of a tornado would work. At least give me a hint, as I assume you're more privy to these secret numerical models than I am. What kind of spatial and temporal resolution at the surface and above is required? Just a ballpark, nothing specific. What insights into tornadogenesis are incorporated into these storm-scale models?

 

[Patrick Marsh]

Again, tell me about how an operational forecast - equal or superior to the current system - of a tornado would work. At least give me a hint, as I assume you're more privy to these secret numerical models than I am. What kind of spatial and temporal resolution at the surface and above is required? Just a ballpark, nothing specific. What insights into tornadogenesis are incorporated into these storm-scale models?

No one is going to change the current warning system overnight. The WoF paradigm is "a vision for 2020".

I encourage you to read the information found here: http://www.nssl.noaa.gov/projects/wof/documents/. You'll see the proposed path forward. I can assure you, social scientists are involved in the process.

Again, nothing is going to change overnight.

 

[rdale]

Oh, so you challenge me with "have you looked at their work?" then tell me it's not availabe to the public.

My apologies - the wording should have been "have you looked at what they are trying to do in their operational guide?"

- To evaluate the accuracy and the operational utility of new science, technology, and products in a testbed setting in order to gain feedback for improvements prior to their potential implementation into NWS severe convective weather warning operations.

- To foster collaboration between research scientists and operational meteorologists.

- To evaluate operational utility of Phased Array Radar (PAR) technology during real-time operational warning situations as well as through playback of archived cases.

- To evaluate operational utility of Collaborative Adaptive Sensing of the Atmosphere (CASA) dense radar networks for severe weather decision-making through real time warning situations and structured experiments with archived data.

- To evaluate operational utility of experimental applications intended for the GOES-R satellite platform, including total lightning data, during real-time operational warning situations as well as through playback of archived cases.

- To evaluate the operational utility of multiple-radar/multiple-sensor severe weather algorithms in supporting NWS hazardous convective weather warning decision making.

- Forecasters may have an opportunity to evaluate various radar data assimilation algorithms that may be used within future Warn-On-Forecast systems.

Now to be honest, those all seem like great goals. Yet you want the group to stop their work, because we don't know everything we need to know about tornadogenesis. Color me confused.

Yet now, you come out as an advocate for completely changing the tornado warning system

Hmmm... Did I do that in a different thread? "Completely changing" would imply getting rid of the current watch/warning system. That never came out of my mouth, nor any other poster here. Just something that Mike Smith said he heard a few years back. I don't see that in the above mission statement list.

with the keystone of such a change being an experimental numerical weather model that isn't publicaly available?

Not sure I said that storm-scale modeling is the keystone either, but maybe we're just misinterpreting each other.

And - it's supposed to be based upon sociological research, yet you lampoon Joe and Jane public (and those mad radar operators, as if they were clowns)?

You've lost me completely now. Thanks for the discussion, it's good to see all viewpoints on the issue.

 

[Greg Stumpf]

As for the comments about the hypothetical person out on a hiking journey, I guess that comes back to the basic mission of the NWS. I don't think it's mission implies being all things to all people. As an individual, if you undertake an endeavor out on the edge of normal domestic life it's up to you to take that extra step to research and keep abreast of the hazards involved. The NWS is supposed to serve the general public, not necessarily every outlier.

Wrong. The NWS serves "the public". That includes all the supposed "outliers", the entire spectrum.

 

[Greg Stumpf]

I've never seen a storm-scale numerical forecast model cited...maybe I've missed it.

You've not missed it because there aren't any storm-scale resolving models being run in real-time. The computational expense to do these runs in real-time is prohibitive. All of the data are being run on case studies and currently take a very long time (in some cases, days) to complete the runs.

 

[Mike Johnston]

Wrong. The NWS serves "the public". That includes all the supposed "outliers", the entire spectrum.

True, but does that necessarily include every individual, every enterprise in whatever endeavor they may undertake? I mean, someone who wants to climb to the top of Pike's Peak in December may encounter some particularly advserse conditions compared to a citizen going about his normal business in Goodland, KS.

And how much time and space should be devoted to covering the outliers to the detriment of the norm?

 

[Greg Stumpf]

True, but does that necessarily include every individual, every enterprise in whatever endeavor they may undertake? I mean, someone who wants to climb to the top of Pike's Peak in December may encounter some particularly advserse conditions compared to a citizen going about his normal business in Goodland, KS.

And how much time and space should be devoted to covering the outliers to the detriment of the norm?

Yes, it means *every* individual. So, the challenge...how do we devise a hazard information system that covers the spectrum of users? I believe that's what we're attempting to do with digital probabilistic hazard grids. Some of the "outliers" may need information at a very tail end of the probability distributions. But they will probably not get their info straight from the grids (unless they were extremely savvy), but rather from a specialized data content provider.

And BTW - what your citizen might be doing in Goodland KS is completely different to what I do during my day. So, what is "normal"?

 

[Mike Smith]

[replacing the current system is] Just something that Mike Smith said he heard a few years back. I don't see that in the above mission statement list.

Then you are not looking very hard, http://www.nssl.noaa.gov/primer/tstorm/tst_predicting.html :

A move from "warn on detection" to "warn on forecast" paradigm will extend warning lead times. NSSL is beginning to study ensembles for very short-range (0 to 1 h) forecasts of severe weather events. These ensembles assimilate Doppler radar data into cloud-scale numerical models to provide improved predictions of thunderstorms and their associated severe weather. While still in a research mode, initial results suggest that it may be possible to use these forecasts in warning operations, leading to a shift from the present "warn on detection" strategy to a "warn on forecast" strategy that would provide longer lead times for severe weather events.

Given that zero to 1 hour is the current lead time of convective warnings, the clear meaning of the above paragraph, copied moments ago from NSSL's web site, is that the program is going to replace the current system. If you read David Andra's paper (that Greg cited) there is no evidence of a polygon warning (the current system), all examples are probabilistic. That paper is dated March of 2010. Are you saying everything has changed since then?

So, let me say it again: In emails from participants, presentations given in Norman, and the clear wording here and in other documents, the explicit stated goal is the replacement of the current system. I don't understand why you dismiss this concern.

Mike

 

[Greg Stumpf]

I don't understand how a shift in warning decision making strategy equates to a replacement of the warning system. As I've stated before, what WoF really is providing is augmentation of the present-day detection and extrapolation techniques with output from storm-scale NWP. One example of what this will provide guidance for are short-term forecasts of convection to be made where storms do not yet presently exist upstream. So, a polygon would be issued, seemingly, where there are no radar echoes. Wait 60 minutes, and there's now a storm there!

If you read David Andra's paper (that Greg cited)

Dave Stensrud (NSSL) authored the paper.

 

[Patrick Marsh]

Mike,

You can try and spin this however you want. No one has said that changes would be made to the binary warning system. No where in the paper (from which you cite the image of probabilistic information) does it say the probabilistic information will be the ONLY information being disseminated to the public. In fact, for all we know, the probabilistic information might only go to the forecaster who uses it as yet another piece of information in issuing his or her (binary) warning. The simple fact is, none of how the implementation and dissemination of WoF into the NWS has been heavily discussed, yet alone finalized. We're still trying to improve our observing, assimilating, and modeling capabilities (not to mention waiting on the computing resources) to be able to do this kind of technique. At this point, we're just trying to get the science down. That's the focus of the research; that's what people should be excited about. We have a long way to go before we start worrying about changing the warning products. If NOAA starts tinkering with the warning process before the WoF science is understood, I'll be right beside you in criticizing. If NOAA tries to implement WoF before a robust discussion about how to do so with minimal impacts to (at that time) current warning services, I'll be a louder critic than you.

 

[Mike Smith]

If NOAA starts tinkering with the warning process before the WoF science is understood, I'll be right beside you in criticizing. If NOAA tries to implement WoF before a robust discussion about how to do so with minimal impacts to (at that time) current warning services, I'll be a louder critic than you.

Fair enough, thank you.

Mike