Current Severe Weather Warning System

[mike57]

Not sure if this is the right place for this ...

Jo Farrow in her latest blog discusses the difficulties  regarding the current warning system.

I think that the public does not understand the current warning system, not just in respect of thunderstorms, but with respect to all severe weather.

The same weather event can move from say Yellow warning to Amber warning based on the time of day of the event because the impact is higher. I think most (non meteorological) people assume that the Yellow/Amber/Red progression is based purely on the severity of the expected weather event. It would be interesting to poll members of the public about what they understand by the current warnings, I think the results would point to there being a problem with understanding them.

Really there are 3 things to be taken into account, the big one, the severity of the expected weather, modified by probability of it occuring at a given location and increased or decreased impact based on timing, e.g. 09:00 is worse than 03:00 as there are less people out.

I think the current system is in danger of not being taken seriously and therefore people not heeding warnings. I would prefer to see the basic Yellow/Amber/Red system based mainly on severity. Probabilty could then be quoted as a %age I think most people understand the concept of a 20% chance of severe weather (at a given location). Discuss...

 

[Met4Cast]

It's very, very tricky and I don't envy the Met Office. There are pros and cons to both systems. 

The issue with basing it on severity instead of impact/disruption is that you can have severe weather that causes no impacts, i,e strong winds during the night, so say an amber warning was issued for strong winds because they met the severity criteria, but then perceptually, because there were no major impacts or issues, the public stop taking the warnings seriously. Then, next time an amber is issued and strong winds coincide with say rush hour, the public remember the last time there were no impacts under an amber, and take it less seriously. 

Conversely, warning about impacts to day to day life is arguably a better system because of that fact and I think by and large the current system works well for more predictable weather events, large scale lows, heavy rain, snowfall.

For less predictable events such as thunderstorms which tend to be much more localised, the warning system doesn't work. How many times have we seen the Met Office issue yellow thunderstorm warnings this year alone and completely flop? Heck, last year there was a run of 7 days of weather warnings for thunderstorms and every single day, storms failed to materialise. It happens, that's the nature of thunderstorms and we all understand that, but from a public point of view a yellow warning = something is going to happen, which doesn't work with thunderstorms. 

They also seem to bury the lead behind menus, the impact matrix is under 'further information', I wonder how many people bother to click that and check it? It adds so much context to the weather warning, yet it's hidden away. 

[Jo Farrow]

I'll add the blog link in 

https://www.netweather.tv/weather-forecasts/news/10924-thunderstorm-warnings---are-they-hitting-the-spot

and some examples from today . In this article, it was looking at how Storm forecasting didn't fit well into the red/amber/yellow system.

That system does have it's own issues realting to public understanding and response but the blanket yellow TS warnings that are now in place for 4 days (weds-Sat) just don't work. Which is why I included the US and Australian examples. Interesting topic though

 

Edited June 16, 2021 by Jo Farrow
spelling

[Met4Cast]

11 minutes ago, Jo Farrow said:

I'll add the blog link in 

https://www.netweather.tv/weather-forecasts/news/10924-thunderstorm-warnings---are-they-hitting-the-spot

and some examples from today . In this article, it was looking at how Storm forecasting didn't fit well into the red/amber/yellow system.

That system does have it's own issues for with public understanding and response but the blanket yellow TS warnings that are now in place for 4 days (weds-Sat) just don't work. Which is why I included the US and Australian examples. Interesting topic though

 

Thanks for posting, I tried to find it but couldn’t - I see you made the same point re: people bothering to check the warning matrix, in my opinion that should be front and centre to every warning, not hidden away. 

Myself, Nick, ConvectiveWx and anyone else who issues convective forecasts/outlooks tend to favour a % risk of thunderstorms in a given area, I wonder if this system would work in an official Met Office capacity though. 

Even now, there is confusion over the “30% risk of rain” forecasts you see on a majority of weather apps, so maybe that approach wouldn’t work. 

I’m not sure what the solution is personally, but I definitely think the current warning system just doesn’t work when it comes to thunderstorms, maybe it could if the Met Office started issuing live nowcast warnings/updates, but that seldom happens, largely I suspect due to the huge collaboration needed between Envionment agency & emergency responders before warnings are issued, it seems like a bottleneck that prevents live & frequent updates. 

[Jo Farrow]

Interestingly if you have looked at the @ MetOffice twitter feed today, there is no mention of Thunderstorms at all. You can name another type of storm (confusing- someone has already asked me that )  

[mike57]

I assume that ulimatley the purpose of the warnings is to encourage the public to change their plans in order to avoid unpleasent or dangerous outcomes. Could you put pie chart symbols on the map to show probility, would people understand it, something like to show likelyhood, or would this just confuse even more

[Jo Farrow]

At the moment the National weather warning, the yellow, gives same risk/impacts to east Wales/ northern England as SE England and East Anglia. All still at Unlikely

It was issued on Sunday night, although was time shortened yesterday to make way for the next broad brush TS yellow Fri/Sat.

Night-time seems to overrule, even in mid summer

[Met4Cast]

The lack of dynamic updates really is the major issue I think. It's been quite clear all day that Manchester for example wouldn't be seeing thunderstorms today, but no update has been issued to reflect this? The same broad brush warning has just been left live on the website.

The warning implies that there's an equal risk of thunderstorms in Manchester as there is in say, Kent, and that just isn't the case. Why couldn't the warning be refined? Even updated live to reflect the current situation on radars?

Does it just come down to a lack of people working?