New Met Eireann rainfall radar website

kumate_champ07

not sure how many are aware of the new upgraded website, you can now zoom in on the map


https://beta.met.ie/

cdaly_

You'll see that you can have a wind or rain forecast on the map also (links to right).
There's a link to a feedback form at the top. Add your comments please...

recedite

The rainfall and wind forecasts are a nice touch. Also the customised location.

magicbastarder

i left a comment pointing out that the new site shows 90 minutes of rainfall on the radar, but the old site show six hours. not a major issue, but would be nice to see further back (unless there's a toggle somewhere that i missed).

recedite

I think the only reason for looking back was so that you could mentally project forward, but now you don't have to, because the rainfall radar forecast does it for you.

On that, it seems to project forward for nearly a week, which is probably too far. I remember yer wan from Met Eireann on the RTE news saying they didn't go more than 5 days ahead because it gets so inaccurate.
Still, with everyone else now doing long range fake forecasts, you gotta keep up with the competition...

cdaly_

recedite wrote: »

I think the only reason for looking back was so that you could mentally project forward, but now you don't have to, because the rainfall radar forecast does it for you.

There's a big difference between the radar and the forecast. The radar is showing what's actually falling* while the forecast is showing expected on the ground precipitation over a period. In the very short term ("will I get wet on the commute?") projecting the radar is likely more accurate.

*The radar is looking at the sky, not the ground and, due to the curvature of the earth, what it shows may be quite a height above ground and may not actually fall. The radar also shows a rainfall rate while the forecast shows a projected accumulation.

magicbastarder

cdaly_ wrote: »

*The radar is looking at the sky, not the ground and, due to the curvature of the earth, what it shows may be quite a height above ground and may not actually fall.

that's showing rain that's not there (or exaggerating it), yeah? a couple of times - not often - i've been out and it's it's been raining, usually a mizzle, which doesn't show on the radar. is that low lying cloud which the radar treats differently to rain? or can it distinguish?

learn

recedite wrote: »

On that, it seems to project forward for nearly a week, which is probably too far. I remember yer wan from Met Eireann on the RTE news saying they didn't go more than 5 days ahead because it gets so inaccurate.
Still, with everyone else now doing long range fake forecasts, you gotta keep up with the competition...

If correct, I hope this 9 day forecast high pressure wins the battle over the low.

https://www.weather-forecast.com/maps/Germany

cdaly_

magicbastarder wrote: »

that's showing rain that's not there (or exaggerating it), yeah? a couple of times - not often - i've been out and it's it's been raining, usually a mizzle, which doesn't show on the radar. is that low lying cloud which the radar treats differently to rain? or can it distinguish?

There's a couple of things to consider here.

Radar is pointing above ground so it won't see the stuff at ground level.
Reflectivity of rain is proportional to something like the sixth power of droplet size which means that drizzle is a lot harder to detect.
There can be issues with mountain shadow so that much less is shown in an area to the far side of mountains from the radar itself.
There's also a bunch of processing required to filter out effects of domestic wifi.

eeeee

cdaly_ wrote: »

There's also a bunch of processing required to filter out effects of domestic wifi.

That's savage intetesting. Does it muddy the image? I never knew WiFi was visible to radar.
I have heard that it confuses bees, and disorientates them. Seeing this (the cloudy, muddy-ing effects of WiFi) tangibly would be super interesting. Is there an image of this anywhere?

magicbastarder

i remember there used to be a comment on the rainfall radar page about a line appearing due to wi-fi at the airport? or am i imagining that?

does wi-fi also interfere with navigational radar, is what i am now wondering...

Cee-Jay-Cee

I've been using the rainfall radar on the Weather Live app for iphones for the last 2 or 3 years. Its incredibly accurate for the day ahead and gets less so the further into the future you go. Last year I was going a charity cycle from Sligo to Letterkenny. I said to the lads with me that it looked like we were in for a soaking somewhere in or around Donegal Town and sure enough as soon as we passed Donegal town, the skies opened and we were washed out of it for about 20 minutes.

I used the 'Windy' app also, its very very good too and very accurate for the following 48hrs or so.

Lumen

nee wrote: »

I never knew WiFi was visible to radar. I have heard that it confuses bees, and disorientates them.

The challenge in dealing with this problem is that the tiny tinfoil hats get in the way of the bee's antennae.

http://apistech.eu/bees-and-cellphones-not-another-horror-story/

cdaly_

nee wrote: »

That's savage intetesting. Does it muddy the image? I never knew WiFi was visible to radar.
I have heard that it confuses bees, and disorientates them. Seeing this (the cloudy, muddy-ing effects of WiFi) tangibly would be super interesting. Is there an image of this anywhere?

magicbastarder wrote: »

i remember there used to be a comment on the rainfall radar page about a line appearing due to wi-fi at the airport? or am i imagining that?

does wi-fi also interfere with navigational radar, is what i am now wondering...

As I understand it, rainfall radar operates on a particular frequency and it's this which conflicts with domestic wifi. I'm not sure if the frequency is a physical thing or a regulated one. Nav radar will use different frequency.

The effect of wifi is not to muddy the signal. Instead the radar sees a signal emanating from a particular bearing which results in a radial (expanding) sector of false echo. It's particularly annoying because it obscures a significant distance. In the images below you can see a pair of such echoes pointing at Shannon Airport (one through Cork City, the other through Kilkenny and Wexford counties). The point of the echo is not necessarily the location of the wifi device. The second image shows another set of echoes pointing at Dublin Airport. Quite a broad sector over Co. Laois with a set more in north Co. Wexford. These images have already been somewhat filtered. The unfiltered versions can be a lot worse.

magicbastarder

can you not just send loads of rain at the offenders to punish them?

Lumen

cdaly_ wrote: »

As I understand it, rainfall radar operates on a particular frequency and it's this which conflicts with domestic wifi. I'm not sure if the frequency is a physical thing or a regulated one.

You would expect someone called Nigel to have the explanation.

http://wifinigel.blogspot.ie/2013/05/5ghz-missing-3-channels-in-europe.html

Maybe unregulated equipment. Like Black Helicopters maybe.

eeeee

Lumen wrote: »

The challenge in dealing with this problem is that the tiny tinfoil hats get in the way of the bee's antennae.

http://apistech.eu/bees-and-cellphones-not-another-horror-story/

Oh I'm glad to read it! I can stop making tiny tinfoil hats now. I was getting really good at them too boo :pac:

cdaly_ wrote: »

As I understand it, rainfall radar operates on a particular frequency and it's this which conflicts with domestic wifi. I'm not sure if the frequency is a physical thing or a regulated one. Nav radar will use different frequency.

The effect of wifi is not to muddy the signal. Instead the radar sees a signal emanating from a particular bearing which results in a radial (expanding) sector of false echo. It's particularly annoying because it obscures a significant distance. In the images below you can see a pair of such echoes pointing at Shannon Airport (one through Cork City, the other through Kilkenny and Wexford counties). The point of the echo is not necessarily the location of the wifi device. The second image shows another set of echoes pointing at Dublin Airport. Quite a broad sector over Co. Laois with a set more in north Co. Wexford. These images have already been somewhat filtered. The unfiltered versions can be a lot worse.

This is absolutely fascinating. I wonder could you use it to trace and map wifi signal across the country. Obviously it only sees a particular kind/frequency of signal. Ooh mind buzz.

cdaly_

nee wrote: »

This is absolutely fascinating. I wonder could you use it to trace and map wifi signal across the country. Obviously it only sees a particular kind/frequency of signal. Ooh mind buzz.

If you think that's a mind buzz, how about the talk I was at from Ericsson about using microwave signal attenuation as a rainfall detection network. The premise was that there are such dense networks of microwave links and that the attenuation due to rainfall is known to engineers (so that they can boost signal) so you should be able to get pretty high resolution rainfall detection...

Lumen

I have to say that the iconographic representation of daily weather is a triumph of simplicity.



That tells me exactly what I need to know, and no more.

velo.2010

So our older viewers won't feel left out....

cdaly_

Lumen wrote: »

I have to say that the iconographic representation of daily weather is a triumph of simplicity.



That tells me exactly what I need to know, and no more.

Don't forget the 7-day forecast below it...

eeeee

cdaly_ wrote: »

If you think that's a mind buzz, how about the talk I was at from Ericsson about using microwave signal attenuation as a rainfall detection network. The premise was that there are such dense networks of microwave links and that the attenuation due to rainfall is known to engineers (so that they can boost signal) so you should be able to get pretty high resolution rainfall detection...

Mind. Blown.

Lumen

cdaly_ wrote: »

Don't forget the 7-day forecast below it...

Don't be ridiculous, nobody can forecast the weather. :pac:

devonp

the beta version was great with a rainfall forecast etc
but since the full new version i don't see the same forecasting links ?? to see rain forecast over 3-4 days ?? etc

(on a map that is , not just symbols)

magicbastarder

are you browsing on a smartphone or PC?
on a 'fat' browser, it's the 'Rainfall Forecast (7 day)' to the right of the main map on the front page.

if you're using the app on a smartphone, click the '>' beside 'maps - recent rainfall'; this opens a new page, and it's the third icon down on the right.

devonp

cheers

working ok on a PC (work) but not on my old mac , running a "too embarrassed to say" version of OSX

tomasrojo

devonp wrote: »

cheers

working ok on a PC (work) but not on my old mac , running a "too embarrassed to say" version of OSX

The computer I'm writing this on has a floppy drive.

DirkVoodoo

This thread is awesome. Thanks cdaly, keep it coming!

Lumen

I woke up early today and amongst other morning chores I imported the hourly Casement Aerodrome weather dataset from met.ie into a MySQL database so I could run queries on data all the way back to the beginning of time.

One of the surprising facts I found is that 1/8 of the hours have measurable rainfall (0.1mm or over). And there's no large bias towards rainfall at different times of day - I always thought it rained more at night time.

There have been discussions on this forum before where people (me included) advanced the idea that on a typical (say <30 minute commute) you had to be quite unlucky to be rained on, but this appears not to be supported by the data.

Dublin doesn't feel like a place that is raining one hour in eight. Well, apart from in 2018 when it feels more like 50/50.

magicbastarder

Lumen wrote: »

There have been discussions on this forum before where people (me included) advanced the idea that on a typical (say <30 minute commute) you had to be quite unlucky to be rained on, but this appears not to be supported by the data.

well, if there's a 1 in 8 chance within one hour, it's going to be less than that in a half an hour - say 1 in 10. which might imply that in a week of commuting, you'd get rained on once?

Lumen

magicbastarder wrote: »

well, if there's a 1 in 8 chance within one hour, it's going to be less than that in a half an hour - say 1 in 10. which might imply that in a week of commuting, you'd get rained on once?

If we take the simple case of a commute one hour each way, with a probability of staying dry of 88%, then the probability of staying dry over one 5-day week is 0.88^10 = 28%. Conversely, the probability of getting wet at least once is 72%.

I'm too stupid to work out the probability for half-hour intervals.

magicbastarder

christ. it's been over 20 years since i studied probability. i've forgotten more than i ever knew.
it took me a few minutes to cop that your reply does not necessarily contradict my (superficial) point.

Lumen

magicbastarder wrote: »

it took me a few minutes to cop that your reply does not necessarily contradict my (superficial) point.

Hmmm, I think it probably does.

magicbastarder

anyway, with a 1 in 10 chance of it raining on any one trip, you'll *probably* get rained on once per week*

*assuming you are cycling a spherical blackbody in a vacuum.

cdaly_

Lumen wrote: »

If we take the simple case of a commute one hour each way, with a probability of staying dry of 88%, then the probability of staying dry over one 5-day week is 0.88^10 = 28%. Conversely, the probability of getting wet at least once is 72%.

I'm too stupid to work out the probability for half-hour intervals.

Don't forget that the day to day probability of rain is not independent so the calculations would get more complex. Also, having rain in 1/8 of hours is not evenly distributed. You might get 7 hours of rain (while you're at work), ride in and home dry and see nothing for the rest of the week.

The other thing to note is that the rainfall is measured (these days*) using a tipping bucket rain gauge and, because the collected rain may sit in the bucket waiting for the last drop, that the 0.1mm of rain recorded in a tip may have fallen over multiple hours (as drizzle) or even days (end of a shower one day, start of a shower another).

You could consider dowloading some of the minute data (non-airport) to get into the real nitty gritty of getting rained on...







*pre-automation, a 5-inch manual gauge was used.

tomasrojo

I have rain gear packed into my satchel, which I carry whenever I leave the house -- by foot or bike overwhelmingly. After being rained on a few times, I have to wash the rain gear. I have to wash it every few weeks at a minimum most times of the year, with occasional stretches of a month or two without washing.

So, not very scientific, but I have to take out and wear the rain gear quite often, but not constantly.

So, I agree, you don't have to be quite unlucky to be rained on in an average week in Dublin(*). When I lived in Galway it rained about three or four times as often, I think.

(*)EDIT: Now I think about it, I have cycled or walked more than an hour a day during the week for the past few years. Probably more like 2.5 hours a day. So my exposure is higher than in the half-hour commute case that people usually put forward when arguing that you'd have to be unlucky to be rained on very often on your commute.

magicbastarder

cdaly_ wrote: »

The other thing to note is that the rainfall is measured (these days*) using a tipping bucket rain gauge and, because the collected rain may sit in the bucket waiting for the last drop, that the 0.1mm of rain recorded in a tip may have fallen over multiple hours (as drizzle) or even days (end of a shower one day, start of a shower another).

so the rain fallen will only register when the bucket has reached a certain level, no matter how long that has taken? how do you account for evaporation, if so?

tomasrojo

cdaly_ wrote: »

Don't forget that the day to day probability of rain is not independent so the calculations would get more complex.

I think it rains more often in the early afternoon than in the morning in the winter in Dublin. Maybe not.

tomasrojo

magicbastarder wrote: »

so the rain fallen will only register when the bucket has reached a certain level, no matter how long that has taken? how do you account for evaporation, if so?

I guess so:

The tipping bucket rain gauge is not as accurate as the standard rain gauge because the rainfall may stop before the lever has tipped

https://en.wikipedia.org/wiki/Rain_gauge#Tipping_bucket_rain_gauge

Mr. Grieves

Good analysis here, and there's a tutorial if you want to do some analysis yourself on another post.

https://www.shanelynn.ie/wet-rainy-cyling-commute-in-ireland-with-wunderground-and-python/

tomasrojo

Mr. Grieves wrote: »

Good analysis here, and there's a tutorial if you want to do some analysis yourself on another post.

https://www.shanelynn.ie/wet-rainy-cyling-commute-in-ireland-with-wunderground-and-python/

My estimate of Galway being 3-4 times wetter than Dublin is confirmed, to my great surprise. (Not surprised it's wetter, just that my guess was very reasonable.)

cdaly_

magicbastarder wrote: »

so the rain fallen will only register when the bucket has reached a certain level, no matter how long that has taken? how do you account for evaporation, if so?

The bucket itself is quite small and is enclosed in a white housing. It's hard to quantify evaporation in that enclosure between things like sunshine warming the housing, wind blowing through (hopefully minimised by the exposure of the instrument) and the humidity in general.

I would say that evaporation losses are fairly low except between showers in summer time but that dual gauges help to offset this as they are unlikely to tip at the same time and so there's a decent chance that one is empty while the other one is waiting for its moment to tip.

tomasrojo wrote: »

I think it rains more often in the early afternoon than in the morning in the winter in Dublin. Maybe not.

I look forward to the results of your study. Data available from Historical data on met.ie

magicbastarder

actually, my next door neighbour was telling us a few years ago that her dad used to be head of met eireann. she was trying to get them to come collect a load of his papers, but was having difficulty getting them to do so. maybe the required data could be in there...

that said, there's a strong chance she's binned the papers if they didn't collect them.

Zen0

I know it’s a minor gripe, but I wish they’d put the latest station reports in the app rather than requiring a click through to the web. I like to check the temperature at Casement before heading out on a club spin; I find it a good indicator of what I should wear.

magicbastarder

you can get similar info at https://www.tiitraffic.ie/weather/ too (though that doesn't address your complaint).

andy69

and you can get data from local weather stations (personal ones, usually installed in someone's garden) which, like my own one, report data live up to wunderground.

There seems to be one there in Rathcoole for example:

https://www.wunderground.com/personal-weather-station/dashboard?ID=IRATHCOO4&cm_ven=localwx_pwsdash

Won't give road temp's like the TII website, but wind speed/chill-factor/direction etc should be all there

TGD

The site asks you to accept cookies but then gives Dublin as the default location each time. With cookies should it not have copped on that I live in the other end of the country after numerous entries?

magicbastarder

a cookie just tells them that you've been on the site before, and what info you may have looked at before. it wouldn't necessarily store location info (until after you've OKed that)
generally, a location lookup is done by geolocating your IP address. your IP address may only resolve to 'ireland', or be registered in dublin (even if you're in cork).

cdaly_

Zen0 wrote: »

I know it’s a minor gripe, but I wish they’d put the latest station reports in the app rather than requiring a click through to the web. I like to check the temperature at Casement before heading out on a club spin; I find it a good indicator of what I should wear.

TGD wrote: »

The site asks you to accept cookies but then gives Dublin as the default location each time. With cookies should it not have copped on that I live in the other end of the country after numerous entries?

Did you give any feedback to Met?

@Zen0: have you looked at the meteogram on the app? Just click on the icon for today's weather and you should get an hour-by-hour forecast for your location.

silverharp

I preferred the old rain radar, colours were more distinctive, the new one is too blue

cdaly_

silverharp wrote: »

I preferred the old rain radar, colours were more distinctive, the new one is too blue

You can see the old one for the moment on archive.met.ie but right now, the old one is bluer than the new.

The rain layer in the new website is semi-transparent so that the map shows through below. That will result is somewhat muted colours.