Sunday, December 28, 2008

The Human Element: Part I

The urban and local flooding prediction was dropped and for good reason. Other than some stopped-up drains there was little threat in the central Puget Sound lowlands since the temperatures were only warming up slowly and heavy rain was not forecast to occur.

This is a good example of over-warning...something that does happen occasionally. Why does it occur? Well sometimes it is due to faulty forecasts...not the case this time. Other times there is a tendency to try to cover a low-probability dangerous situation...better safe than sorry kind of philosophy. And then there are the psychological elements. Perhaps meteorologists are the last people you want forecasting the weather, because many of us love severe weather and subconsciously want it to happen. When you want something to happen....well you can guess the result. And lets face it...when one goes for a severe forecast, the media takes notice...another positive feedback for some. When I give my forecasting class I tell students they have to think like Spock or Data of Star Trek...put the emotions aside..put this is more easily said then done. Anyway, meteorologists are human and you have to factor that in. Certainlywe are no worst than economists...and I think you can argue that our track record is a lot better. I mean A LOT better.
How can you get into the head of a meteorologist to decide whether you believe their forecast? Well, for me you have this blog and my Friday 9:50 AM segment on KUOW. For the NWS, they have very useful forecast discussions...something a few of you mentioned in the comments. The lead NWS forecaster on each shift pens a description of what she or he thinks is happening and will happen and why. A great product. The only issue is that there is often some technical jargon in there (like CWA--Coordinated Warning Area--and some weather terminology you might not be familiar with like 1000-850 mb thickness). But 50-75% should be understandable. The forecasters also identify themselves and if you read them frequently enough you learn about their "quirks." (I won't say any more or will certainly get into trouble) You can find the forecast discussion on the web....for the Seattle office: http://www.wrh.noaa.gov/sew/get.php?wfo=sew&pil=AFD&sid=SEW
I find it useful to read the discussions of neighboring offices (e.g., Portland) as well.

There are other useful discussions to look at. The Northwest Avalanche Center does a mountain weather forecast and discussion...and these individuals are very skillful and knowledgeable about mountain weather: http://www.nwac.us/forecasts.htm

Scott Sistek of KOMO TV--a past UW Atmos graduate and the chief behind the scenes weather staffer at that station--has a wonderful weather blog (http://www.komonews.com/weather/blog). Great information from a very insightful guy. King5 weathercasters also do blogs and some of them have substantial backgrounds (Jeff Renner, Sharon O'Donnell, and Rich Marriott all have professional degrees in the field).

And of course you can watch the TV weather broadcasts--where you often hear some of the logic and uncertainty underlying the forecast.

In any case, there is valuable information regarding the confidence and nature of the forecast that you can glean from the above sources. It pays to be an intelligent consumer of weather information.

PS: A great example of a Puget Sound convergence zone on the latest radar:

31 comments:

Joseph Ratliff said...

Hey Cliff...when you make a post in Blogger...you can make your links live:

1) Highlight the entire link. You might want to copy it as well for step 3.

2) Click on the little chain link in the post writing area.

3) Then paste the link again in the box that comes up...or re-type it if you didn't copy it in step one.

4) Click "ok" or whatever to get you back to your post in the posting area.

Your link will be active in your post, so your readers can just click on them :)

Thanks for the excellent book and blog Cliff.

Joseph Ratliff

Anonymous said...

I read NWS Portland's discussions last week while awaiting the Christmas Eve/Day event. The sad thing was that they didn't change their discussion in three or four days and used the exact same wording (i.e. copy and paste) the entire time. It became frustrating - what good is reading the discussions if they don't update it with their most current thoughts?

I also found it hard to trust when their point and click forecasts mentioned things like "rain likely, high around 31."

Anonymous said...

I really miss Shannon O's wacked-out, ill-fitting, and wrinkled outfits. Someone must have gotten to her.

mainstreeter said...

Cliff, I noticed some locations in E WA, like Deer Park reporting 54 Felts 56, while Geiger is still 36 degrees. Are we seeing some very localized "chinook' event or are their equipment values wrong?

Kevin Purcell said...

See the previous blog post comments for commentary on the Enhanced AFD. Very nice graphics included so you can follow the discussion (without digging in the models).

http://www.wrh.noaa.gov/sew/gafd/latest_webafd.html

The FAQ associated with the Enhanced is a minor tutorial on weather prediction and reading model output (including reading wind profiler output!).

http://www.wrh.noaa.gov/sew/FAQs/EG-AFD_FAQs.php

Not sure why they "bury" the enhanced AFD ... it should be the default though I don't think it comes with a glossary pop-up.

As Cliff seems to be acting as a "condensation center" for amateur weather geeks perhaps we could use a "Cliff Mass Weather Blog" Google Group alongside the blog for more "interactive" discussions. Blog comments tend to die so people put less effort into them knowing they have a short life (especially when the same question is asked over and over).

Anonymous said...

Thanks for giving some of the local TV meteorologists some credit. They are in a tough position being on commercial stations that have to keep the public engaged (and not just informed) between commercials. Some here have derided them but if you understand the TV news context you can ignore the hype from the "newscasters" and focus on what people like Renner and Marriott have to say.

Joseph Ratliff said...

I like Kevin Purcell's idea.

Good call Kevin!

Timothy Hogg said...

I have been reading this blog for the last couple of weeks while on vacation with my wife here. I enjoy it about as much as I enjoy away being away from Seattle during the snow!
I think you hit on something very interesting today that I have been thinking about alot with journalism in general-- and is the theory behind "if it bleeds, it leads"... Yes, forecasters and journalists are cut from the same cloth-- one chases weather while the other chases news-- both can be tedious and not very rewarding in many aspects, so at times there is a tendancy to over embellish the facts and instead, editorialize.. I dont mind this in the forecasting model, because if you're wrong, I might get a little wet (or usually wear too little clothing) but for the most part, all if fair in love and weather.
Journalists on the other hand, can do some serious damage while thinking they are being responsible. Long gone seem to be the days or fair journalism, but thats for another post :).
Keep on keeping on with this new medium, brotherman-- its by far the most rewarding way to deliver this kind of "news".

JewelyaZ said...

Kevin,

I agree. I'd love to have a place where we could have a more continuous discussion. I'm pretty amateur at weather stuff in the PNW but I've been owning/running Yahoogroups since they were eGroups (a transition that happened in 1999 -- I've been running groups since 1995) and would be glad to set up/volunteer moderate a Yahoogroup for us. Cliff, give it your blessing if you will, and I'll go set it up.

JewelyaZ said...
This comment has been removed by the author.
Kevin Purcell said...

The reason I suggest a google group is that Blogger is google so people wouldn't need yet another account (even though I have one).

Plus, a better reason, the Google Groups search tends to be a bit better (I find) and a lot easier to find throughout Google. Data in a Yahoo Group tends to stay there and not "leak out". When I make an effort to post I'd like to share that with the biggest number of people. Yahoo doesn't quite do that.

But before using Cliff's name I think we need some authorization (and I'd rather he'ed own it ... it's his name!).

Cliff: my email is exposed in my profile if you need assistance to do this.

JewelyaZ said...

Better to beg forgiveness than ask permission... here's the weather discussion group that I started. I hope it will become an active discussion place and a repository for FAQs and other educational pointers.

http://tech.groups.yahoo.com/group/PNWeather/

JewelyaZ said...

Kevin,

I love all things Google... I've been using Gmail since one week after the very start of the beta and I preordered and love my T-mobile G1 Google phone... but I *hate* Google groups. If that's what everyone wants, I'll go there, but I set up a Yahoo group and maybe folks will use it. I find the Yahoo interface much better in general for group activity.

Anonymous said...

Seems to me it would be more appropriate to have a UW mailman discussion list (no advertisements!) UW policy definitely allows for general public mailing lists (e.g. tweeters). Archives can be set up so that it's easy to to get back messages, and messages could also be archived at a web site (UW or other) for web searches.

If Cliff doesn't want to set it up and moderate it, maybe he could assign one of his students:-). It isn't much work in any case.

http://www.washington.edu/computing/mailman/

Anonymous said...

From this morning's post about flooding issues...

"This is a good example of over-warning..."

"(At) Other times there is a tendency to try to cover a low-probability dangerous situation..."

"Perhaps meteorologists are the last people you want forecasting the weather, because many of us love severe weather..."

Ya think??? On the heels of your forecast earlier today for another Chanukah-type event next weekend (maybe, sorta), this really seemed like an ironic time to be discussing over-forecasting.

Cliff Mass Weather Blog said...

I never said I was immune to the temptation of overforecasting! Anyway, any extreme event is by its very nature unlikely...there are many ways for the severe event not to happen. But I do think it is useful to have a heads up about potential, low-probability events...as long as one is clear about the unlikely nature of the occurrence.

And Kevin is right..the enhanced (graphical) forecast discussion, produced by the NWS, is very nice.

Matt said...

Here is something I have been wondering about for years - why are the National Weather Service forecast discussions IN ALL CAPS? Does somebody send them in via telegraph? It seems a trifle ridiculous in this day and age.

Anonymous said...

Matt, the National Weather Service software converts discussions to all caps, regardless of whether the individual forecaster types it in all lowercase, all uppercase, or a mix. As a former NWS forecaster, it drove me nuts. I think it has something to do with how it did used to be sent via teletype and they still have it like that for whatever reason. I never really figured it out when I was forecasting.

climo man said...

Regarding the NWS weather discussions: I highly recommend reading the daily afternoon discussion from the Spokane NWS office. Over the years they have done, in my opinion, the most consistently best job when it comes down to describing weather features and potential stengths and trajectories of winter systems.This is not suprising since the Spokane area is on the brink of snow during much of the winter, so they have to be precise.I find the Seattle forecast discussion--the detailed part, not the summary--often vague and lacking in detail. And their occasional flippancy is quite irksome.( Furthermore, their spelling is awful,too.BEHRING Sea? Come on people, at least proofread your work, before you publish.)Those folks need to realize that not everyone who reads the discussions are naieve novices.

mb in Port Angeles said...

Right now the regular Area Forecast Discussion is from 9:00PM, but the Enhanced Area Forecast Discussion (the one with the links to graphics) is still from 3PM. It would be nice if both were kept current. Otherwise the fancy one isn't as useful.

Interested Aviator said...

Climo Man,

The correct spelling is naive, not naieve. "Come on people, at least proofread your work, before you publish." (Snicker)

Also, a little humor in an AFD is a nice bit of relief, and not unprofessional if clever and occasional. My favorite was a KBFL AFD from 20 years ago that said something like "given the high QPF and tight gradients there's a good chance of blowing mud".

The variation between the content of AFD's between the various offices is interesting and suggests that the NWS has no strict guidelines for them. This makes sense as there appears to be no intended public use for AFD's.

If you're looking for detail and precision, look at the aviation synoptic products or the NWAC mountain forecasts. BTW, the NWAC forecasts are sometimes done as quite funny poems!

Hope I spelt everythin wright!

;)

Anonymous said...

From a forecaster at Seattle: I could not find the misspelling quoted in climo man's comment. Things can slip through when many products are being composed and the phones are ringing off the hook.

By the way mb ... the graphical forecast discussion takes a while to prepare and has lower priority than marine forecasts, watches/warnings, terminal forecasts, weather radio products, and calls from the press.

Anonymous said...

I experienced the "blowing mud" referred to by a previous commenter near Pocatello, ID, last March. All the way across Idaho we'd been watching lightning activity in the mountains to the north. As we got close to Pocatello we drove into a localized zone of 30-40-mph winds with spatterings of rain. That was more than enough wind to pick up dust, and enough rain to mud-ify it. It was over again in just a mile or two' maybe it was just a big dust devil.

- Pete

fjblau said...

I met one of the forecast discussion contributing authors at the PNW Weather Conference a few years ago... I got a chance to gush at the beautiful descriptions and meteorological poetry these people create!

The discussions are always the first place I go in the morning!

Anonymous said...

As to misspellings in the discussion -- I interned at the Seattle NWS in 1993 -- the NWS AFD is the original text messaging language before texting was invented. For a long time, when it was broadcast via ticker, words were required to be 4 letters or less, and no punctuation was allowed besides a period. Everything was abbreviated and I'm sure a lot of those historical abbreviations live on (like HI RES MODL for "high resolution model" or EXTD for "extended", even though they now have a longer leash to type the AFD more conversational now that the ticker is outdated.

Bruce said...

You can find more information about the terminology used in the Forecast Discussions at NWS Forecast Discussion Jargon. It's a handy guide to the abbreviations and technical language.

As for the annoying ALL CAPS: I've exchanged emails on this topic with the folks at the Seattle office. It seems the format is dictated by international treaties. Changing to a system appropriate even to late-20th century technology apparently requires "an act of Congress."

MarkM said...

I appreciate the educational aspects of this blog e.g. pointing out the great example of the CZ. Cliff, is it possible for you to highlight or label things with arrows so that the less sophisticated of us can learn a little more? Thanks.

Anonymous said...

Cliff, what do the latest models show regarding the huge potential windstorm the end of this week? I'm waiting on pins and needles!

Fleetwood said...

In the spirit of providing useful weather links. Here are a few I like. Apologies if some of these were provided earlier by others:

Radar map, with animation
Radar Map

Another cool radar map, with animation:
Radar Map

Weather discussion from the NOAA duty meteororologist:
Area Forecast Discussion

Tabular Weather observations:
SeaTac Weather Observations

Anonymous said...

Here is a link to the 4 area NWS discussions (Seattle, Portland, Spokane, and Pendleton.

http://www.atmos.washington.edu/data/disc_report.html

bud said...

Hi Cliff,
I follow your blog and other sources to try and marry-up: wind strength and location; specific timing and location of the low; local tide tables to try and predict storm tides. Any help along these lines would be appreciated.
Your book by the fire on a "snow day"...I'd recommend it!
Bud Searle
Camano Island