"Once in 100 year" weather events

Forums › Off topic › "Once in 100 year" weather events

timmmay

20575 posts

Uber Geek

Trusted

Lifetime subscriber

#248457 27-Mar-2019 07:28

From this NZHerald article (from the Mayor). "Mayor: "This is a 100-year event, there is a huge amount of rain".

We were down there in about 2012, a bridge washed out because of the amount of rain. I think it may have been the same bridge, just north of Franz Joseph. It took out the power to the town, but fortunately many places have generators.

Seems like those "one in 100 years" weather events are coming a lot more often these days.

1 | 2

30853 posts

Uber Geek

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#2205465 27-Mar-2019 07:52

That's not quite how probability works.

A one in 100 year event is something based around rainfall or river flow, with the probability of a 1 in 100 year event happening is 1% each year. A bit like rolling a die, if you get a 6 the odds of getting a 6 on the next throw is still the same.

After a one in 100 year event the odds of getting a one in 100 year event the very next year is still the exact same 1%

ripdog

548 posts

Ultimate Geek
Inactive user

#2205469 27-Mar-2019 08:02

sbiddle:

That's not quite how probability works.

A one in 100 year event is something based around rainfall or river flow, with the probability of a 1 in 100 year event happening is 1% each year. A bit like rolling a die, if you get a 6 the odds of getting a 6 on the next throw is still the same.

After a one in 100 year event the odds of getting a one in 100 year event the very next year is still the exact same 1%

You're talking about rolling a die, and you're right in that scenario - the shape of the die does not change.

In this case, we're talking about the climate, where the shape of the die is ABSOLUTELY changing. The probability of these events is estimated via the historical record of such events, and as new major weather events are added to the record, the probability must be re-estimated frequently.

surfisup1000

5288 posts

Uber Geek

#2205471 27-Mar-2019 08:04

sbiddle:

That's not quite how probability works.

A one in 100 year event is something based around rainfall or river flow, with the probability of a 1 in 100 year event happening is 1% each year. A bit like rolling a die, if you get a 6 the odds of getting a 6 on the next throw is still the same.

After a one in 100 year event the odds of getting a one in 100 year event the very next year is still the exact same 1%

It helps to ask the question, what is the probably of two 1/100 year events in the space of 7 years? It will be low, but still high enough to be plausible. Another 1/100 year flood next year and you'd be asking questions about why.

To use your dice example, if you throw it 10000 times and roll 2000 1's, then it is likely something is wrong with the dice.

Same with too many 1/100 year floods. If too many are occurring then either the baseline odds were set incorrectly from the start, or the surrounding weather patterns are changing.

nathan

5695 posts

Uber Geek
Inactive user

#2205474 27-Mar-2019 08:07

These 1:100 events, seem to be more used as a figure of speech, not a statistical measurement

Batman

Mad Scientist
29760 posts

Uber Geek

Trusted

Lifetime subscriber

#2205479 27-Mar-2019 08:30

timmmay:

From this NZHerald article (from the Mayor). "Mayor: "This is a 100-year event, there is a huge amount of rain".

We were down there in about 2012, a bridge washed out because of the amount of rain. I think it may have been the same bridge, just north of Franz Joseph. It took out the power to the town, but fortunately many places have generators.

Seems like those "one in 100 years" weather events are coming a lot more often these days.

these 100 year events seem to happen every 1-2 years in the last 3-4 years

MikeB4

18435 posts

Uber Geek

ID Verified

Trusted

#2205480 27-Mar-2019 08:31

Climate change is making these stats and predictions a lottery. These events are going to be one in X number of months.

MileHighKiwi

782 posts

Ultimate Geek

ID Verified

Lifetime subscriber

#2205481 27-Mar-2019 08:38

The most rain I've ever experienced was on the West Coast near Franz Josef. Bucketed down for two days, couldn't even leave the motel. The media love a good weather event.

Equinox IT

Cloud spending continues to surge globally, but most organisations haven’t made the changes necessary to maximise the value and cost-efficiency benefits of their cloud investments. Download the whitepaper From Overspend to Advantage now.

frankv

5680 posts

Uber Geek

Lifetime subscriber

#2205519 27-Mar-2019 10:27

Batman:

these 100 year events seem to happen every 1-2 years in the last 3-4 years

If you have 50-100 independent weather event locations (yes, I know that weather events aren't independent), then you would expect one 100-year-event every 1-2 years.

chevrolux

4962 posts

Uber Geek
Inactive user

#2205535 27-Mar-2019 10:56

My partner is a hydrologist that deals with this stuff daily, I always get told off for saying things like "but it hasn't been 100 years". The "1 in 100 year" thing is just a term they use to represent the probability for the layman to understand the significance, as Steve mentioned. It doesn't mean an event like this will only happen every 100 years.

For example, new bridges built on main state highways get spec'd for "1 in 2500 year" events.

But this is all based off historical data and modelling, the biggest issue being data collection has varied massively across the councils that collect it. There is no centralised system for storage or even a standard for how the data should get collected. There is a standard now that some councils have implemented to give QC ratings to data, but it is by no means a requirement so many councils simply don't bother.

Batman

Mad Scientist
29760 posts

Uber Geek

Trusted

Lifetime subscriber

#2205558 27-Mar-2019 11:42

frankv:

Batman:

these 100 year events seem to happen every 1-2 years in the last 3-4 years

If you have 50-100 independent weather event locations (yes, I know that weather events aren't independent), then you would expect one 100-year-event every 1-2 years.

No same event same place

For example, Dunedin had 2 100year rainfalls in 2 years

clevedon

1059 posts

Uber Geek

#2205596 27-Mar-2019 12:29

Batman:

frankv:

Batman:

these 100 year events seem to happen every 1-2 years in the last 3-4 years

If you have 50-100 independent weather event locations (yes, I know that weather events aren't independent), then you would expect one 100-year-event every 1-2 years.

No same event same place

For example, Dunedin had 2 100year rainfalls in 2 years

Yeah, but they might not get another one for another 99 years - so it's still 1 in a 100 though.

Geektastic

17942 posts

Uber Geek

Trusted

Lifetime subscriber

#2205944 27-Mar-2019 22:32

timmmay:

From this NZHerald article (from the Mayor). "Mayor: "This is a 100-year event, there is a huge amount of rain".

We were down there in about 2012, a bridge washed out because of the amount of rain. I think it may have been the same bridge, just north of Franz Joseph. It took out the power to the town, but fortunately many places have generators.

Seems like those "one in 100 years" weather events are coming a lot more often these days.

Don't look past people adding descriptors like that for impact. It makes it sound more serious in the press and perhaps benefits the mayor by getting his problem more attention and, potentially, funding.

robbon44

252 posts

Ultimate Geek

#2206056 28-Mar-2019 08:02

The discussion is a good one and the term 1 in 100 yr event is correctly interpreted as a 1% chance however..here is my take... the event also relates to a level of impact or an amount of rainfall, eg the 1in 100 yr flood level or the 1-1000 yr event (the alpine fault rupture is a 1-300yr event by the way and it’s been about 350yrs since the last rupture ! Just saying...).

Whilst it’s quite right to state that these events seem to be occurring more often, ie climate change, the level of impact is still the same. This is why we have a 1-500 event level or 1-1000 event level.

There is no reason why a flood event could not occur every year (or every day) however the event level is not determinable until after the event has occurred. That was a 1-100 event level.

The probability is that we will have more 1-100 events than 1-500 events and so on...

The frequency of 1-100 yr events may seem to be increasing or more frequent than you remember but media coverage and means of monitoring and recording the events is now better than it has ever been. As such we are recording the event levels better and thus appear to have more 1-100 events.

Another point to consider is that with population growth, urbanisation and development of land then the environment reacts as we change it to suit our demand. Thus we are creating or rather modifying the environment and creating a world that cannot necessarily deal with the events that the natural world throws at it.

We try to design our world to accomodate the 1-100 event or even the 1-1000 event but as our urbanisation sprawl increases the events become more focused and more intense and the impact of the event becomes more severe.

Climate change is real. It is affecting our weather patterns and we are beginning to notice this more and more. However our lifetime is a relatively short period of time in the grade scheme of things and whilst these changes are beginning to be a bit of a concern (lol) you do need to consider the larger longer timeframe of the earths existence and the bigger longer term picture... that said I am firm believer in the fact that we are growing as a population faster than every before and our impact on the planet and the environment has never been seen before... so we really should try and understand our role in the process and reduce our overall impact as best as possible...

That got deep eh !

robbon44

252 posts

Ultimate Geek

#2206057 28-Mar-2019 08:02

tdgeek

29743 posts

Uber Geek

Trusted

Lifetime subscriber

#2206596 28-Mar-2019 19:41

Its probability, pure and simple. If its 1 in 100 that's 1% probability.

There is merit in we are watching and recording more data, but the bottom line is we have climate change. So an event that was 1 in 100 years, will now be 1 in 10 years, soon to become every year.

This year and last year we had high marine temps. So more evaporation, more volatile weather.

Its already here.

1 | 2