- 3 days ago
Stay informed with the latest weather forecast video, covering rain patterns, temperature outlooks, and severe weather warnings. Updated every few days with expert analysis, from Weather Scientist Alex Dunne.
Category
🗞
NewsTranscript
00:03Hello, it is Sunday, the 27th of September, and this is your National Weather Forecast.
00:07In this video, we're going to go through this week's weather, a little quick seasonal
00:10update, and then finally, a deep dive into how we actually make weather forecasts, what
00:15confidence means, and what we mean when we talk about models and ensembles.
00:17So let's start with the forecast.
00:19This week, we will see a dominated wet weather pattern as we have a trough that's slowly
00:24moving across the country.
00:26Here, looking at the weekly totals, we can see a lot adds up in Western Australia, and
00:30as well along the southeast coast here, with some areas seeing 25 to 50 mils.
00:35Current weather warnings that are associated from this system are along the southern coast,
00:39and then we also have some in the east coast, and these tend to ease on Monday.
00:43By Tuesday, we'll have a quick look into the bushfire risk ratings, but specifically for
00:48Wednesday, where we see this amplified risk throughout central Queensland.
00:51So let's start day by day.
00:53Okay, firstly, we're looking here, this is a little low system that's really bringing
00:58or driving a lot of this weather, and we're seeing some areas of 25 to 50 mils.
01:01On Monday, this low moves south, and we start to get these outward branches of low pressure,
01:06which is what's going to be moving across the country and delivering a lot of that rain.
01:10Tuesday moves across, same as Wednesday, and then Thursday is when it really starts to hit
01:15the southeast coast in Victoria, with some areas seeing potentially 15 to 25 mils, and that
01:20extends up into South Australia as well.
01:22Friday, the system really kind of splits as a high moves in.
01:25We've got a lot of rain over here in the east coast, and then another fair chunk of rain,
01:2925 to 50 mils in central WA.
01:32Saturday, both of these systems move across, and Sunday they move, but they also ease.
01:36So there's a bit of rain, a bit of moisture in the atmosphere there as well.
01:39Now, associated with these systems, we're also going to see an intrusion of warm weather
01:43down south, as this high pressure system, the downward descending arc, drives a lot of this tropical
01:48heat down into the southern regions across Tuesday, Wednesday, and Thursday, and that
01:52could see temperatures between 35 and 40 degrees Celsius in some areas.
01:56Now, this is also associated with looking again on Wednesday with some strong wind gusts,
02:0140 to 63 kilometres an hour, which I believe are contributing to that elevated fire risk
02:05towards the middle of the week, because we've got hot, windy weather.
02:09And we've also seen with those rain systems that they are likely to bring some thunder and
02:12lightning storms as well.
02:13So that could definitely be contributing to that elevated fire risk.
02:17OK, so that is our weather update this week.
02:20We can see quite a bit of rain on the way down the south.
02:22Let's get into the seasonal forecast.
02:25Checking in with our climate drivers now, we can see that El Nino continues to intensify
02:30as we go on throughout the year, with models suggesting it will peak November to December.
02:34If we head in and check in on the IOD, it's in a positive stage and looks to stay there
02:39for the next few months before returning back to neutral.
02:41The southern annular mode is a relatively neutral phase, and the Madden-Julian oscillation appears
02:46to be going into a slightly drier phase, but there's not much strength behind that.
02:49Let's check in with the Bureau's long-range forecasts.
02:52We know that these tend to sort of strengthen their confidence towards the end of the month
02:57as they run a few more models.
02:58So this is a good time to check in with them.
03:00So first off, we're having a look at the chance above median for the month of October.
03:04There is a likely chance that much of Western Australia will exceed their median rainfall
03:10for this time of year, whereas large parts of the northeast of the country and the southeast
03:14are unlikely to exceed this rainfall, and there's pretty strong confidence that in the southeast
03:20that there'll be below average rainfall.
03:22If we look at chance of extremes, unusually dry, so we're looking at the chance that it
03:27will be unusually dry, we can see in those same northeast and southeast areas that there
03:31is an amplified chance that there will be unusually dry conditions, whereas if we look
03:35at the unusually wet conditions, we can see that there is a more likely chance that there
03:40will be unusually wet conditions across WA.
03:42All right, let's check in with temperature now.
03:45We will look at maximum temperature, of course.
03:47Chance above median, not a great thing to look at because of our climate change signals.
03:51So instead, we'll look at our chance of extremes.
03:53Unusually cool weather.
03:55We might have an amplified risk of that in this very northeast corner, but it's the unusually
03:59warm temperatures that we are likely to see going into October.
04:02October definitely seems to be sort of the strongest compared to if we look at November, a little
04:07bit lower chance, and if we look October to December, higher chance than November to January.
04:13So definitely a likely chance that we're going to have quite a hot spring.
04:17Maybe that will ease towards the start of summer.
04:20All right, let's actually have a deep dive now into how we sort of get these chances and
04:25what we mean when we talk about confidence in these predictions and what on earth a model
04:29actually is.
04:30Let's go.
04:30So let's start off by chatting about what a weather model is.
04:34So unlike a lot of other scientific fields, in weather science, we can't just change a
04:40variable and manipulate something and then see what happens.
04:43So what we do is we build models.
04:45And these use the fundamental laws of physics, like gravity and Newton's laws and things like
04:50that, that enable us to make a simulated world.
04:53And then we will run an experiment within that modeled world and that will carry out through
05:00those mathematic equations and those physical equations what would happen to the weather
05:05if we ran it a certain way.
05:07So that's what a model is.
05:09Now, we have lots of different models that build their physics or the mathematics a little bit
05:14differently.
05:15So, you know, the Australians make their own model.
05:18The Europeans make their own model.
05:20The North Americans make their own model.
05:21The Japanese make their own model.
05:23So we have lots and lots of different models.
05:26And they all have their own sort of unique strengths and weaknesses.
05:29And so we try and look at all of them when we're making decisions about weather.
05:33That's why if you look at Jane's Weather, we'll go to the pro version.
05:36We show you all four, like we show you four of these top models.
05:41So right now I'm in my Jane's Weather app.
05:45I'm looking at the forecast and it just sort of tells me one number or one line.
05:50So it's saying at Thursday, the 1st of October, at, I don't know, what is this, the evening,
05:56about two mils of rain.
05:57Okay, that's what you'd get in sort of any weather app.
05:59It will sort of tell you how much rain is on the way per day or per hour, something like
06:02that.
06:03If I click on this pro version, I now get these different colors.
06:06And each of these different colors tells me what each of these different models is spinning up
06:11and how they differ.
06:12So at the same time, the Australian model is producing four mils.
06:18Somewhere there.
06:19The Canadian model is producing four mils.
06:21The US model, four mils.
06:22And the EU model, two mils.
06:24Okay.
06:24And then when we give you like one number, it's sort of the best estimate using all those
06:28different models to then give you that data.
06:31Okay.
06:31That's why if you flick open this pane on the side over there, it will tell you Thursday
06:37is 13 mils, but there's a range between 10 and 20 mils.
06:42And that's just to show you this range between the models.
06:44Okay.
06:44But what does that physically look like?
06:46Like what does that actually mean?
06:47So here I'm using one of my lovely supervisors, Michael Barnes, Dr. Michael Barnes has made
06:53this great tool, I suppose, of postage stamps to show you what actually goes within those
06:59models.
06:59Because we don't just have four models.
07:01Each of those models has a number of ensemble members.
07:06So in this one, we're looking at the European model and the European model has 50 ensemble
07:13members.
07:13So the European model doesn't just run one set of problems.
07:16It runs it 50 times with slight differences between them all to see what the big range
07:21of possibilities is.
07:22And so here we're looking at, you know, 50 different scenarios of the weather with slight
07:28perturbations.
07:29And then if I run it forward, you can start to see the models all become slightly different
07:38and they get more different the further away we are from when they were initially started
07:42or run computationally.
07:43So, for example, if we look at Ensemble Member 33 in the jet, there's this sort of nice peak
07:52of the wave over South Australia in Ensemble Member 33, whereas in Ensemble Member 41, the
07:57jet is like well below the Australian mainland.
08:00So that's what we're looking at when we're looking at Ensemble Members.
08:03But what does this mean for you as someone that's consuming weather?
08:06Well, this is where I really like this other visualisation that Michael's made.
08:10And this is to show you precipitation for Australian capital cities.
08:14It's just a visualisation, but it happens everywhere.
08:16OK, so we will look into Melbourne because that's where we were looking at before.
08:20So we're looking at Melbourne, Victoria.
08:23Each of these columns is a date.
08:25So we've got Monday, Tuesday, Wednesday, Thursday, Friday onwards.
08:28And each of the rows is, I can point here, each of these different rows is a different
08:34ensemble member.
08:35OK, so they've all spun up their own sort of thing.
08:38And each of these rows here or each of these colours is to show you what the ensemble member
08:44shows for that day.
08:45OK, so ensemble member.
08:47Oh, so actually, this is a good one.
08:49All the ensemble members for Monday are showing that there is not going to be any rain in Melbourne.
08:54And so that would mean as a meteorologist, I'd say we have very strong confidence that
08:58it will not rain in Melbourne on Monday, the 28th of September.
09:03OK, cool.
09:04If we look at Tuesday, I've got one ensemble member showing 0.5 mils, another 0.1, another
09:090.2, another 0.2.
09:10But the majority of them are not showing a lot of rain.
09:14OK, so I would say there's pretty strong confidence there's not going to be rain on
09:17Tuesday.
09:17Or I would say maybe there's low confidence of below or less than one mil on Monday.
09:23OK, whereas if we go to Thursday, when we look down here, this is the mean.
09:27So the average amount of rainfall that the ensembles are showing in Melbourne on Thursday
09:32is 12 mils.
09:33Cool.
09:33But if we actually go through these, we can see there's a range.
09:37So ensemble member 46 is showing one mil and ensemble member 47 is showing 30 mils.
09:42So there's obviously a very big difference between them.
09:44But if I look across all of them, there's relatively good consistency from the 50 models that I'm
09:52going to get somewhere between sort of five mils and maybe like, I don't know, 15 mils,
09:57something like that.
09:58So I'd say pretty strong confidence somewhere between five and 15 mils is on the way for
10:01Melbourne.
10:02I would say with much less confidence, because only one ensemble member is showing 30 mils,
10:07I'd say there's a very low chance of 30 mils.
10:09But a low chance means that it still could happen.
10:11So that's how we're talking about weather.
10:13So that's a little sneak peek behind how weather forecasts are made and why we talk about
10:19confidence.
10:20Now, when we look at seasonal forecasts, it's the same deal.
10:25They're still spun up by models.
10:27All those models have ensemble members within them.
10:30And we're talking about how well do the ensemble members agree on certain scenarios?
10:36OK, and that's why when we look at long range forecasts, we're talking about the chance of
10:41something.
10:41So there is a high chance that Western Australia will receive above average rainfall for October,
10:48which means that probably a very high number of ensemble members have shown greater than
10:55average rainfall conditions.
10:56That's that's where we're coming from with that.
10:58OK, I hope that little deep dive helps you understand a little bit more about how we make
11:03these decisions and how we communicate the weather, as well as sometimes why it's wrong
11:08or the weather that you receive is maybe a little bit different from your forecast or things
11:12like that.
11:12OK, what this inherently shows you, this one is a very good example, is that weather is very
11:19we call it stochastic, which means random and we can predict it fairly well.
11:24But the further out from an event we are, the less what like the less confidence in the
11:30predictions that we have.
11:31But also like just inherently rain can fall, you know, the suburb over from Melbourne and
11:36that could change the millimetres that are received, you know, by five or ten mils.
11:40So that's that's just another variable in there as well.
11:42OK, if you have any questions, reach out and let us know.
11:45I thought that was a fun deep dive for you.
11:46Something I wanted to make for a while, especially as we're coming up into a season where knowing
11:54uncertainty and knowing confidence is really, really important.
11:57OK, that's it for this week.
11:59Do have a lovely week.
12:00Stay safe and we'll see you next time.
12:01Bye.