Thunderstorms; So-So Forecasts, but Lack of Communication

What happened the past two days (especially Saturday) is a good example of a growing problem and a dirty little secret in the meteorological world. We often know more about upcoming weather patterns than we share with you in a simplistic “high-low-no rain/rain” forecast.

The slide above is from the Dr. Cliff Mass presentation to our local AMS chapter just 2 weeks ago. I hadn’t really given it much thought until he made it the cornerstone of a push towards using more “uncertainty language” in a forecast.

1. Regional models were clearly showing  the possibility for afternoon showers or thunderstorms somewhere in the western valleys of northern Oregon or SW Washington both Friday and Saturday afternoons.  And the Portland/Vancouver Metro Area seemed to be a good place for that to occur. 

Here are the 12z RPM and WRF-GFS models from Friday morning, showing later Friday afternoon/evening…looks wet to me on both, although the placement was off, especially on the WRF-GFS (2nd image).  We actually did okay with Friday here at FOX-12, mentioning the chance for afternoon rain or thunder.

Now on Saturday morning, our 12z RPM showed almost exactly what occurred just 14 hours later; although it was a few hours earlier than reality in the end:

The WRF-GFS wasn’t quite as good, but it was hinting at some sort of action coming off the Coast Range (not correct):

2. Yet all the forecasts (including mine) downplayed the threat enough Saturday that the thunderstorms were probably a surprise for most of the public.  We didn’t make it abundantly clear that an evening thunderstorm in the city was quite possible.  Yes, a relatively small chance, but the chance was there.  On Friday night at 8/10/11pm I showed our RPM and mentioned that I thought thunderstorms would probably stay on the far east side of the metro area, close to the Cascades.  I thought there would be a slightly shift to more westerly upper-level flow later in the day keeping the developing storms out there.  Clearly location was wrong on that forecast.  But when I saw the skies darken to my northwest and heard a few rumbles, it definitely wasn’t a surprise.  It was more like “I wish I would have included a chance for the entire metro area.”

 The amount of weather information we receive nowadays is incredible, but breaking that down into a forecast that goes just a few minutes on TV (and for a whole state) is a challenge.  Maybe more important, forcing that info into a pretty little cloud/sun/rain graphic is even tricker.  We are sure able to bring uncertainty information verbally into a weathercast and through text on a website.  We need to work harder getting that information out.

Now Dr. Mass also suggested we not show a 7 Day forecast like this:

His point being that we can’t REALLY nail a forecast within a degree or two (regularly) more than a couple days out, so we should be giving temperature ranges instead of specific numbers.  But this ship sailed a long time ago…television 7 day forecasts have specific numbers and that won’t change. 

These 7 Day forecasts also don’t give the public any idea about our confidence in certain patterns either.  For example, models are showing some variance (right now) on Thursday and Friday.  Some are pushing the rain in a bit quicker on Thursday, but others waiting until late Friday.  So I’m quite confident on the Monday-Wednesday forecast, and pretty confident on the general pattern change to wet and cool next weekend.  But in between is a period with significant uncertainty.  You sure don’t see it on the graphic; and there are plenty of numbers, text, and logos on the graphic already.  So where would you show lower confidence on this graphic?  There is no easy way, for now it has to be verbal on television or webcasts. 

Weather forecasts are getting better and better, but we have work to do when conveying that increased forecast accuracy don’t we?

Chief Meteorologist Mark Nelsen

20 Responses to Thunderstorms; So-So Forecasts, but Lack of Communication

  1. mkd says:

    NWS handles the question of uncertainty by additional verbiage, but you have to link to their “Forecast Discussion” to read it. See one here: http://forecast.weather.gov/product.php?site=NWS&issuedby=PQR&product=AFD&format=CI&version=1&glossary=1.

    In my eyes, it boils down to two pieces of information: actual words about confidence/probability; and where the forecast comes from (specific models and their agreement vs. just climatology). Between them, I find I can make reasonably guesses about what the forecast means to me and any actions I should take.

  2. RoyalPain83 says:

    You got my vote for most irritating comment of this post so far….congradulations.

  3. keebler1175 says:

    I think you should use grey-scale to show uncertainty. Have the days that you are sure of in full color, and fade to grey-scale as the week progresses, only moving the cells to color as you gain certainty. The more certain you are, the more color and less grey.

    • Mr Data says:

      I second that! While it won’t please everyone it will be a major step in the right direction.

  4. This is one of the hardest locations in the country to forecast, and you’re doing your best, keep it up like you’ve been doing, and things will still go great!

  5. bgb41 says:

    5/27/2012 Oregon (All) Temperature Summary

    Warmest:
    High:79 at Castle( 283 ft)
    Low: 57 at John Day River B(305 ft) & Blalock(280 ft)

    Coldest:
    High:37 at HOWARD Mt Howard(8150 ft)
    Low: 21 at Sand Creek (US 9 (4525 ft )

    Largest Diurnal Change: 38 degrees
    KLAMATH NWR (62/24 ) (4531 ft )
    Beatty (62/24) (4320 ft)
    Sand Creek (US 9 (59/21) (4525 ft)

    Heaviest Rainfall:
    0.32″ at YELLOWPINE(4600ft)
    0.32″ at Aurora State Air(194ft)

  6. WestVan says:

    Mark,
    If you remember. I made this same comment back when you were being praised for accurately predicting a snowfall this year which you felt like you had missed. The models were very uncertain and you put that in your forecast. When it did snow the public felt like you did a great job but you felt like you hadn’t nailed it down well enough because you didn’t really expect it to snow.

    I really believe the public craves more uncertainty in their forecast. It is why I read this blog even though I can’t read a weather map. Here I find out about the uncertainties which help me interpret the forecasts more accurately.

    As far as temperature ranges. We all know you can’t get a temp spot on. That I don’t think bothers the public. Much more important to me is expressing the chance for rain as a percentage chance. In _every_ other place I have lived, TV mets. would say there is X% chance of rain tomorrow. In PDX the tendency seems to be to say. It will rain or it should be dry, hiding the possibilities which exist.

    Frankly, if I were making the graphics I would either, put a percentage such as 40% on top of your clouds or I would get rid of the clouds and just go with a percentage of precipitation. That would be a more informative graphic to me. If you wanted to say sun breaks, you could have the sun poke out from behind the percentage rather than from behind clouds. If you wanted to say it is likely to rain in the morning you could put rain drops which come out underneath the percentage number, left for morning, right for afternoon. (Like you do with the clouds). A percentage number would give me a really key piece of information which I am missing right now. I bet the public would respond very positively to that change.

    I think chance of precipitation is really important when it comes to snow. I love it when you give us a range of forecasts, ’cause we all know that you can’t nail a snow forecast in PDX. They are just too tricky. So if you say 20% chance of snow on the valley floor, then we don’t get mad when we see flakes. Too often forecasters tend to say it won’t or it will snow on the valley floor. Even saying 90% chance of snow still gives you an out when something unexpected happens. I doubt you ever could truly be more than 90% sure of any snow forecast in PDX.

    Thanks for all your work. You truly are the best. I enjoy watching you. Personally, I would love more uncertainty in the forecast.

  7. karlbonner1982 says:

    I have an idea! If there’s uncertainty as to the temps, OR the timing of a warmup/cooldown, put an asterisk next to the number and explain the uncertainty in a few words, like “Timing of cooldown somewhat uncertain” or “Intensity of warm spell uncertain”.

    Honestly, the best solution to the temp accuracy problem is one that probably won’t get anywhere in the immediate future, at least in the US: switching to Celsius. Larger degrees would be perfect for dealing with limits on precision!

  8. W7ENK says:

    It’s okay Mark, I don’t judge. Besides, if you’re going to blow a forecast (or appear to blow a forecast), better to blow it with something exciting! Just like, I’d rather you call for no snow, and wake up to a surprise 6 inches! 😆

    Thunderstorms are a nice surprise… unless you’re trying to have a BBQ. Sorry Brian. 😥

    • I think I remember Mark saying on this blog (the 24th)something about ” The screaming message would be that any of us in the western valleys MIGHT see a late day shower or thunderstorm.Saturday should be similar, then a dying front moves onshore to give us lots of clouds and a few light showers.” …so I guess you hit it on the mark, Mark 🙂 but weren’t quite ready to go prime time with the “screaming message” !

  9. stormblown says:

    I have to say watching the forecast last Wednesday and even Thursday and everyday before there wasnt indication of t-storm or any shower activity on Friday or Saturday.

    What I would have liked to see in the 7 day graphic is the word (chance) written in and mb a little lighting bolt cloud at the end of those days. Up until just a few days there was no indication on the forecasts maps local mets made. If the the chance is 10% or greater that a rain shower, especially and thundershower may occur then there should be that chance wording.

  10. Jane says:

    I’m not a weather guru, but a medical researcher by profession (and an all-around natural science nerd, which is why I’m here!). I live with uncertainty in my research every day and get used to speaking about “possible”‘ this and “maybe” that. I cheered through this blog entry from beginning to end, because even as a non-meteorologist I can tell that TV forecasts are gross oversimplifications, due to time constraints. What is hard to tell is to what degree SOME of the forecasters themselves have paid attention to the nuances. I love it when you all throw in just a few specifics here and there (“see this line here, it’s showing the front coming over sometime between 4 and 7”)–even if you can’t explain everything you’d like to, just showing a little bit of your work in detail raises credibility a great deal in my eyes–and it can go a long way toward dispelling the false idea that you’re 100% sure about everything you say. Mother Nature loves to keep us guessing, no matter what branch of science we’re in!

  11. weathercrazy82 says:

    Typically when storms move off of the Cascades they weaken. This was not the case this time.

  12. What about adding a confidence % on each day on the graphic? That might at least let people know how certain you are about what will happen 4-7 days from now.

    I LOVE this blog and share it with my friends all the time! I’m not a weather geek so some of this goes way over my head, but it’s nice to get more details about what’s going on in the weather world, and what the weather people are thinking. So thanks very much for this!!

  13. Larry-Beaverton, Oregon USA says:

    You’re still the best Mark, no worries!!

  14. bgb41 says:

    Yesterdays evenings radar looked like the TS activity would skirt the eastern foothills but instead convection exploded as it descended south through Clark Co. Situations like this are rare and cannot be predicted ahead of time (like 05-24-2008). I am truly amazed that such a rare event happened just 4 yrs apart on literally the same date and day of the year. I got .57 on friday and .61 on saturday here at BG Lake. Way more lightning on friday but way heavier precip rates on Saturday. Certainly a two day period I will remember for the rest of my life.

%d bloggers like this: