Subscribe
Share
Share
Embed
Documenting the monumental discoveries of researchers, and the role Pawsey plays in supporting their breakthroughs.
Welcome to HPC Hearts & Minds, where we talk to the people behind some of today’s most fascinating discoveries.
High Performance Computing, or HPC, allows researchers to tackle incredibly complex questions. This series is all about finding out the real-world impact that comes from using this large-scale computing.
We want to showcase the hearts and minds behind the technology.
Pawsey sits on the Whadjuk
Country of the Noongar nation,
and we'd like to pay our respects
to their elders, past,
present and emerging.
The swimming changed
my personality, made me channel
that energy into challenges
and things like that.
So you’re always striving to be better
and things like that,
and that mindset
basically put me along the whole career.
What does swimming and supercomputing
have in common?
Welcome to HPC Hearts and minds.
Where we talk to the people behind some of
today's most fascinating discoveries.
High Performance Computing, or HPC,
allows researchers
to tackle incredibly complex questions.
This series is all about
finding out what can be done
using this kind of large scale tech.
More importantly,
it's the hearts and minds behind that tech
that make it special.
We've been lucky enough to speak with
a handful of amazing people for this show.
First on our list is Professor
Charitha Pattiaratchi,
an oceanographer
at The University of Western Australia.
For over 25 years,
Chari has been instrumental
in showcasing the power behind the waters
that surround us,
including discoveries in tsunami
impacts and dense shelf water transport.
Cheri received the People's Choice
Award at the Premier’s Scientist
of the Year for 2025 for his contributions
to Australian oceanography.
His work has also helped discover
locations of debris from the missing MH
370 flight back in 2014,
which we talk about as well.
Chari sees what he does
as if it's a jigsaw puzzle.
Taking each piece
like supercomputing, field work
and satellite imagery and putting it
together to see the whole picture.
Studying the Indian Ocean and why
it vastly differs from other oceans.
Chari, thank you so much for joining me
today.
You're welcome.
So what got you into oceanography
before you got into it?
Oh, Interesting. So,
I wanted to be an oceanographer
from the time I was 13 or 14.
All right.
My father was a geologist.
And then so we had, you know,
the family had a science background,
and I used to go on field trips
with him around Sri Lanka.
And I got interested in science
and things like that.
And then I also started swimming.
I was reading in those days,
of course, you know that the age group,
you always read comics.
Not Marvel superheroes, but, funny ones.
And I saw this little piece on this comic
which says,
you know, a career in oceanography.
An oceanographer
basically studies the ocean.
And there's different components
for oceanography.
So there's physical oceanography,
which I do,
which looks at the physics of the ocean.
So waves, currents, tsunamis,
how the ocean works, how the ocean moves.
That's mixing.
So there's a physical oceanographer.
Then there's a biological oceanographer,
which looks like what happens
in the biology of the ocean,
this chemical oceanographers,
there's geological oceanographers.
So, you know, every single, part of
the ocean is studied by different groups.
So it's a it's a multidisciplinary area.
So I build the instruments,
I do all the electronics,
I do everything,
and then I take the student out on a boat.
I drive the boat, we collect the data,
and then we download the data.
Then I write the Matlab scripts
to actually analyse the data.
So from the time from a drawings
on a piece of paper,
this is the instrument that I want
all the way to actually writing a paper.
That's the whole stage
that you go through.
A lot of the things that I did was
basically because of the team that I had.
Fantastic students, fantastic staff
to be able to do that.
I guess it's really hard
when you have to explain that to somebody,
because there are so many factors
that lead into what makes an ocean
an ocean really.
So I said: “Oh, hang on.
This is something for me”.
And interestingly, I wrote it down
in a little blue book,
and I still got the blue book.
And when I read it,
I actually said, oh, hang on,
this is what it said 13 years ago.
What you would do as an oceanographer.
Now look back and say “Wait
a minute, I've done all of that”.
That would
have been so great at 13 years of age,
because I know a lot of the times
when you're that age,
you have no clue
about what you want to go into.
But it was a little bit forceful as well
because I wanted to become a geologist.
All right.
But my mother said, “No,
I'm not going to have two people
with the same profession in the family,”
because my father's family are basically
professional musicians, and whenever
they got together, they would argue.
“Just look at your father's family.
Nope, that's not what I want in my family.
So you're not becoming a geologist”.
I mean, that's that's such a strange,
profession as well, to be banned from it.
So usually it's an artist or an actor,
like.
“No family is ever going to go
into those kind of fields”,
but I mean, oceanography or at least,
you know, science and physics
and these sorts of things are still
I mean, as a geologist,
I'm sure your dad has many ideas.
The interesting thing
is, you know, I, you know, now I keep,
you know, people ask me and say “Oh,
you know, what do you think?”
You know, Sri Lanka is an island nation
and things like that.
But I actually tell people,
what do the general population of Sri
Lanka think about the ocean, right.
In one word: scared.
Right?
People are really scared about the ocean.
So this is the interesting part
for me as well
because, you know,
I was Sri Lanka's best swimmer, right?
But my father would not allow me
to go to the swimming the ocean.
He was too scared.
Right.
“You can swim in the pool,
not in the ocean”.
So he also had a problem in terms of me
becoming an oceanographer,
going out in boats and doing things.
So he had to come over it.
When I actually started studying,
I remember he came and visited me
while I was studying
and I said: “I'm going on the ship.”
And you could see his expression changed.
Didn't
really--It was good to support,
to become an oceanographer,
but I don't think he really realised
what it entails.
Oh, it's too bad you couldn't say,
you know, study rivers and streams.
Please stay away from the oceans.
What I find really fascinating,
the fact that
because you were a professional swimmer.
You had a lot of,
high quality titles for a long, long time.
right, in Sri Lanka. Yeah.
And also when I was at university,
the time that I was
an undergraduate,
I was the fastest swimmer in English.
Not a bad title to have.
Like,
do you feel like the love in the ocean
helped you in terms of swimming as well?
Do you think that they both...
I mean,
not the, they’re sort of two different
things, but, you know, in
when I was born, I was very sick.
My life expectancy was two years, really.
So, very sick child and things like that.
So, and one of the part of that was that
I actually did not speak
or communicate
until I actually started going to school.
Wow, okay.
So that was, you know, and then while
in school, I was basically hyperactive.
So going from one to another and whatever.
The thing is that the swimming changed
my personality and whatever,
but also made you know, of course,
that ADHD and whatever, athletics
and all of that
made you a competitive person.
But swimming made me channel
that energy into challenges
and things like that.
So you always striving to be better
and things like that.
And that mindset is basically put me along
the whole career, you know.
You basically have to be best
and you want to, you know, and,
and do the best you can.
So that's the sort of the training
that actually got into it.
So a few years ago I did the Rottnest Swim
because that was a challenge, right?
Because for 20 years
I have been predicting the weather
and which where the currents are going
and things like that.
And I said, oh, okay,
maybe this is something I should do.
I always wanted
to do it, but I never had the opportunity.
But then you do it.
So there are different challenges
that you want to do.
And day to day.
It's a beautiful thought to think
that something like, like a sport,
like swimming, for example,
can have a really good place in the fields
of science and study and research,
because you are, in a sense, having a bit
of a challenge in whether it be a distance
or endurance for, for sport.
But then, you know, trying to find
its really scientific challenges.
You know, you're
either swimming in a vast ocean
and trying to find a very, very small,
minute thing.
But you know that,
you know, with hard work and persistence,
you'll hopefully find answers.
So it's the same kind of, I guess, drive.
But I mean, it's the mindset.
I mean, you know, so,
you know, in swimming you set a goal
and you know, you want to do okay, here's
the record that you want to break.
So you actually then
but it's not only that you have a goal.
You also have a plan to achieve that.
How you train, how you do, you know,
all of that is stepped in your mind.
You do it.
Sometimes you actually don't
even think you're doing it right.
But that's, that's why you go and train.
That's why you spend six hours in a pool
going up and down a black line.
Right.
And the most boring thing you can do,
but you're trying to strive that
you concentrating, you have perseverance,
you've got determination.
So when you actually come,
you know, in the same way
we're talking about computing, right?
You're writing a code.
You know, you want to get that code right.
You want to do
you want to get that code to be able
to do what you want it to do, right?
It's the same analogy.
You persevere.
You actually say, I want that. Yeah.
And you actually keep going.
So you can see it's the same mindset.
Yeah.
It's a great, I guess, lesson to be able
to take into future, endeavours
without even realizing, like you say,
sometimes you're just doing it
in the heat of the moment
without realizing you, you know, again
and again trying to focus on that.
So my advice for parents and things
like that, I mean, like even my kids:
go and do sport.
Forget about studies, you know sport.
It's a good outlet. Yeah.
Sometimes you know
it's always good to have a goal.
But I think after a while
if you're stuck in it
and without having something
to take your mind off,
like swimming is very much
a meditative thing as well.
You just you're constantly thinking,
breathing and that sort of stuff.
But it does take your head out
for a little bit to be able to relax
as much as you are
physically working yourself out,
but you get to like have a little bit of a
think as well.
Oh, that's excellent.
I love the way in which you were able
to pull that into your work.
I would love to find out how you went
from, Sri Lanka to start
studying in oceanography, I guess.
What was the beginning?
What brought you into that field?
Well, I told you
I wanted to become an oceanographer.
Right.
And there was no, universities
or any expertise
in Sri Lanka,
so basically have to look overseas.
And then I looked at,
I remember applying for universities.
I went on a, on a big
swimming tour to China,
and then when I came back,
I had all these prospects
from the universities I wanted to go to.
And there was only one university
undergraduate degree
in oceanography in the UK.
And I said,
as soon as I looked at prospectus,
I said, this is where I wanted to go.
So I went there
and, I lived there for 12 years.
Oh, wow.
The bachelor's, master's,
PhD and a postdoc.
Do you feel like living there
for 12 years?
You started to love
a different type of ocean.
I guess the landscapes
and things like that. It's.
I mean, again, you know, it's.
You don't actually think about that again.
You know, your mindset is to actually do
what you want you to do.
Everything else,
the environment, you get used to it.
If it's cold, I mean, you know,
Swansea
has one of the largest tides in the world.
So ten meters,
which means that you can only go out,
you know, here
we actually just go, oh, we go for a swim.
Now, you can't do that.
If you go at low tide,
the water is 2 or 3 kilometres outside.
So, you know, when you go out in a boat,
you have you know, everything is so often
you would get up at 2 or 3 in the morning
in the middle of the winter
and go out to the boat.
But you have to do,
you know, actually think about, oh,
I don't like this small one.
It's something you have to do.
Yeah.
As you start studying
these different fields, like like you
said, it's just one of those things
you would go at that at 2am,
and you would get all this data
and that sort of thing.
Back then, did you feel like the way
in which you were collecting this data?
Was there any inkling
that there was any kind of technology
at that time that would help assist
when you were collecting this data?
So you'd think, oh,
it's, you know, when
you actually look back in some of these,
some of the technology that I use
that I'm still using, right?
So, you know, in different parts, but
you actually never looked into the future.
The other thing I should also say is that,
you know,
I made a sacrifice to actually go
to the UK.
You know,
my father being a geologist scientist
didn't get enough money,
so we have to pay.
So he basically had to resign his job,
right?
Moved the whole family.
So they.
He got a job in Cyprus.
So my family and everybody basically moved
to Cyprus for me to go to university.
Wow. Good dad.
Good family.
Really. It's the whole family, right?
It's a whole feature of that.
And then then when, you know.
But the good thing is that my, it was,
one of these similar
to United Nations type of a job.
And so they pay for the kids
to the 21 to go to university.
So once you after you got 21,
you're on your own again.
So, then I didn't
want to put the burden again
on my parents and family,
so I decided I would work.
So now you actually work and study.
So now I'm trying to figure out
what is the highest paid job
that you could work at night,
and study during the day.
Yeah.
So it works out to be a bouncer
in a nightclub, right?
You are a bouncer at a nightclub. Really?
So for three years
I was a bouncer in a nightclub.
So in the same thing,
I mean, I used to go to work, you
know, work finishes 2 or 3 in the morning
and often straight from work.
Go home, change down to the docks
to go and do fieldwork. Wow.
That's I mean,
yeah, obviously, like you said,
it was a job that came with great money.
But at the time, I don't think you were
thought this is a dangerous job.
It's like, oh, it's a great opportunity
to get things going.
I still remember
the professor, you know, typical.
Very English professor.
You know, I used to come,
you know, come to the pub and,
and you know, maybe I had few stitches
on my head and, things like that.
And he would actually say, you know,
“When Sir Arthur J.
Rank established his empire,
I don't think he ever thought
he was going to support oceanography”.
So Arthur Rank is basically
somebody like Rupert Murdoch at that time.
Okay. Yeah.
He had films.
He had all the movie theatres, bingo
halls, he was in and films and whatever.
So that was the big empire at that time.
So he was actually laughing and saying,
you know, you never know.
You never know what your work ends up
leading to doing. Spy.
So the comics, that you read,
I'm sure the person
that put that information,
that was just to give you
a bit more information,
I think he was just doing
a, Well, it's not for me.
He was basically,
trying to motivate people to to do that.
Yeah.
Probably don't realize how much influence
you have on somebody
until someone tells you, right?
You know, that sort of thing.
So when did you first get to experience
any kind of supercomputing power, I guess,
and what was that transition like going.
Well...maybe
I should go back how I actually start,
you know, as an undergraduate.
So computers were just starting there
were,
you know, all from the time
I was doing my undergraduate project.
It's about computers.
I know everything that I do.
So you actually go from there Fortran
and then just the normal mainframe.
Then in the UK
they brought out this laptop.
It's called the BBC. Okay. Right.
So why is it called the BBC's
that they actually have a TV program
and they designed from the TV program
this computer.
Okay. Laptop.
And basically that was the
one which was going to be
go to the schools and things like that.
And you actually did that.
So and started programming with that,
but in those days
storage was on cassette tapes.
And there's also the Commodore
PET computer.
I think actually, you know, in,
in, in the end what you actually doing,
you know, even if you do have a computer,
you actually have to visualise, right.
So then you actually had basically
had the HP Flatbed Plotter,
you had a pen, okay.
And every stroke of the pen
had to be controlled by the computer.
So you basically have to write your code,
lift the pen,
draw this line, lift
the pen, go here with the pen. Go.
And so that's how I basically started
doing my programming.
I mean yeah it goes back to the time
of trying to find out how this can help.
And at the time it probably
would have seen such a headache to do so.
Now actually, you know, in some of my
talks, I actually put a old computer card.
I don't know whetere--do you know what--
I think it looks like a punch card
kind of thing. Oh, yeah. Yeah.
So you actually type it in,
and then you actually a whole lot of this,
and then you go to the computer.
So you might have a stack
and you put it in there,
press the button, it runs your program.
That's how it sort of started.
I mean imagine
dropping one of those piles.
Oh you have to go back. Nightmare.
Yeah. I'm absolute nightmare
to be able to do that.
So that's you know, so there was
no supercomputing or parallel computing.
So when I came to Australia,
that was what I actually
was striving to do.
So it took me like two years
to raise enough funds
to actually buy a Sun Workstation.
Okay, this is a just a standalone, but
it had more computing power than anything.
So that basically was the main part.
And then we had
I can't remember what company it was.
They gave a good deal.
Ultra super fast computer.
Comes back in good days. Perfect.
You got that then?
There was a company in, in California
who actually gave us a.
They call it a supercomputer,
but it wasn’t that much faster,
so we
had that, but I never actually used it.
So it was so up until,
I came along,
it was mainly running on the Sun
Workstation.
Okay.
Your go-to machine was a Sun Workstation
that you've been used to, probably for,
I guess a few years.
You were using this? Yes.
Yeah, maybe 5 or 6 years.
And then iVEC comes along.
What's interesting to me is the fact
that you have this mode of study,
but then you also have to start learning,
programming and understanding
how to use these machines to do the tasks
you're hoping for them to do.
You should know this.
The interesting thing is
I never, ever have any
formal training in coding.
So all all self-taught.
So it's basically
running Fortran programs.
And then the next one
that basically I had the first week
I was at UWA I had there was a postdoc
and she came and said “Oh, Chari,
you know,
I'm so-and-so” and said “Look,
my brother is developing some
software
and he's got this little company.”
But would you actually consider buying.
Okay. Software.
And, so she gave me two, five
and a quarter floppy disk.
Yeah.
And, I said point,
you know, so well, this is Matlab.
That's for visualisation. Okay.
It's in there.
So you have different
components in a sense.
So if you want to if you're running
a Fortran program to actually do anything,
you've got to compile it
and then you run it.
There's a whole lot of steps.
But in Matlab
you can do this straight away.
It is very limited in terms of,
memory and how big
that you can do things in that.
But that would have been
a big change for you,
going from one way of processing what
you would hope to do to this Matlab,
which was visualising that really quickly.
So to improve your work, the yeah,
when I came here I started doing that.
Yeah.
So but you know that was till
I actually got used to Matlab,
I was because you know in in program
you use what you actually know
and what you're comfortable with
and what you knowledge is.
So the first few plots,
the first few papers, I wrote
was basically this line plots
with a flatbed plotter.
Right.
And then went to Matlab and then started
changing the way you create the plots. And
so when you
go to supercomputing as well, right.
The way that I have worked
is that you do the big number crunching.
Okay. On the supercomputer.
But you do all your visualisations
on Matlab okay.
Right. Yep.
So it's, it's very different to
some other people will do.
But the way that we have worked
and I have worked is that way.
Yeah.
Is that is that how you felt comfortable
I guess using visualisation
is because Matlab had its way
of being able to present this data
the way he was familiar with it.
Use your comfort space.
I can do anything with Matlab.
Yeah, but the I have a picture in my mind.
I can actually get it done in my lab.
Yeah, but in the computing power
that I have, maybe I don't actually have
the space and the memory to be able to do
the big number crunching.
So you do the number
crunching in the supercomputers and then
you basically do the visualisation
and the summary plots and things like.
Yeah, technology
always seems to be a way for us
to expand our sort of a view on
how we want to capture something.
So you might be limited to say
data storage or these sorts of things.
And then along comes a new piece of tech
that says, you can do
30 times the amount of processing
that you would want to.
Does that make you
when you're trying to work something out,
have to step back and go, okay, maybe
I should be asking a different question
or a bigger question.
So I mean, it's also have you evolve
I mean, you know, the, the
a lot of people really didn't like Matlab
because it's
commercial software,
but you need to pay a big licensing fee.
And so basically that's
how Python had started.
Right.
And you know, it's
the same as Windows and Unix.
I don't want to learn Python.
Well, it's a new language. Why?
Like I don't really want to do something
and spend my time learning something.
You know, I have, you know,
people who work for me and, and said, oh,
Python is more efficient to do that.
I said, can you please do this to me?
No. And, and and then it can be done.
So I, I still do a lot of Matlab
programming.
I mean, people still get amazed.
They walk into my office.
Hey, are you programming?
And I said, yes, of course.
You know, that's that's
sort of the thing you like to do, right?
Yeah.
So, I'm still a lot of hands on
and virtually in, you know,
80 to 90% of the figures that I use
in my papers, I make it from scratch.
But it's so not a lot of people
at my sort of
level will not do anything like that
in the sense.
So the technology does improve.
But, you know, in my case,
I stay with what I'm comfortable with.
Yeah.
That's it.
You don't have to join the crew
enjoying every single new flashy
bells and whistles thing.
If you know how to get an answer
that you see in your head
sometimes it's
good to go back to the machines at work.
Well, okay.
I want to check to see
to how you went from using Matlab
and then going into supercomputing.
How did that help your work when you
started to find out that you could use
this, power to do all the number
crunching before you would visualise it
in all of this, you basically,
so what I do,
you know,
to give you, a set of an example,
we would have an
area of the ocean, like,
let's say the coastal waters
of Perth
from Fremantle to Yanchep or whatever.
And then you run this,
you know, model, computer model.
So you basically, you know,
these days they call it the digital twins.
Okay.
So basically you're simulating the ocean.
You are limited by two things.
How fast is your computer
to actually do the scaling?
So what are we doing?
You know, just like you do on a, on a TV.
It's a pixel. Well,
so we are actually dividing the ocean
into little squares or rectangles.
Okay.
So how many rectangles
you can have in that system
and how long can you run the system?
Okay.
It's the ones which actually depends
on your computing right.
So it's always a balance.
So you pick your
resolution.
We call that based on
how long you want to run.
I remember in one of the students
who was actually running it,
he was looking at rip currents.
Okay?
And, we were selected
two meters high resolution.
But when we actually put it
on supercomputer,
it took four times
real time to run one hour.
Okay, one hour of simulation.
Would take four hours of supercomputing
time.
Right? Okay.
Right. So.
But that's what we want that time.
Nobody's ever done that before work.
Right.
So that high resolution system,
and so that's basically,
an example.
Setonix now probably now do it
a few seconds, probably.
So, you know,
that's just sort of the, the advances.
So that allowed, you know,
so those days when we started even with,
Carlin and Cognac,
we could basically run for a few weeks.
Now we're running 20 years.
Yeah, we're running the whole of Australia
for 60 years.
Right.
So we're reducing that resolution
but also extending
and that we can actually do that
because of the supercomputers.
Yeah.
Does that continually blow your mind
knowing that you can extend these sorts
of, of projections way further than you
ever did, say, five, ten, 15 years ago?
You know, the thing is that what you do
is that rise to the challenge.
So rise to the challenge. You decide.
Okay, I have this computer. Right.
So those days, one model run on my Sun
Workstation for, let's say, ten days
would take about five days, right?
So you actually run it in the background
in the back,
you know, in your office in the corner.
It's turning away.
You're doing something else
till your simulation finishes.
Right. So but that's only the thing.
And you say, well,
I would like to do higher resolution
because the higher the resolution,
the more things that you learn.
Right.
So this is what basically brought
with the supercomputing.
And as the computing power increased,
we also increased.
We reduce our resolution.
We simulate a bigger area.
We can run longer simulations.
Do you remember the first time
you were able to increase that resolution,
or just increase the way
you were studying something, and just
how that felt to sort of
see your work, the potential?
Oh, absolutely.
Because we then started learning
about more about the ocean,
how it works,
because we have the higher resolution
and, yeah, definitely supercomputing
in the way that
these machines continually evolve
and get bigger and that sort of thing.
But I'd love to find out from you
is because we have these sorts of,
processing powers
with, with supercomputers is a
sort of a worry
that if we didn't have access to those,
that we wouldn't be able to do the kinds
of, challenges that we need to do.
Definitely.
It's not only the, you know,
we're talking about the supercomputing,
but it's about,
what we call parallel computing.
So we actually take the piece of the ocean
and we run it.
That's what I did
with, with a single processor,
some type of things that we're looking at.
How small, how high resolution can we go,
what big domain can we have
and how long can we run?
So those are the three
main things in the ocean that we can do.
So now imagine that we actually have
the Sun Workstation would run
very low
resolution, let's say for the, the area,
you're
running that resolution to the ocean,
which extends from Singapore
to beyond New Zealand.
So that big space, that big space,
for 20 years or longer. Yeah.
When we could only run for about ten days,
which would take five days.
So, so that is the advances
and the advantages
we basically have and,
and it allows us to do so many things,
find so many new things
about how the ocean works,
which is basically what I do.
So, you know,
I can look at the satellite image
and say, well, that's interesting.
Yeah.
How can we explain?
So, you know, we said that
I was a physical oceanographer,
but I also talk about, the process.
So I look at processes.
So how does the ocean work?
But computing is not the only thing.
Right?
So the computing also has to be,
supplemented
by going out on ships
and collecting data on the real ocean.
So that's another completely
different part.
So my research in the ocean
basically has three components in there.
So I call them sort of the field
measurements,
numerical simulation
which is the computing.
And then satellite remote sensing.
Okay.
So think of the ocean as a jigsaw puzzle
okay.
But those pieces are what I just said.
Right.
So one part of information
that will come from the computing,
some part will come from the ocean,
some part would come from the satellite.
They're all complementary
because they give different things.
And then my idea is to my job
is to actually put them all together
and tell the story.
Just to so puzzling every day.
Yeah, exactly.
Exactly.
Yeah.
Do you,
do you feel like it's been really help.
It's just out of curiosity as well,
because of all the data
you've been able to collect, say, around
Australia for 60 plus years.
Right.
How helpful
that's been to have that access as well.
In that continual collection I've done.
Well, it's this is the 60 years of not,
not actually feel that this is computing,
simulation know so 60 years,
simulations solely
because the tidal cycles,
some of the tidal cycle,
the sun are about 20 years.
So you need to have three of them
to actually do that.
Okay.
So we run that and we ran it to actually
looking at the actually water levels.
So anywhere around Australia
we can go to the website and actually say,
well, okay,
this is the maximum water level
we will expect
at this particular location.
Yes. And the other part
that, you know, in the field data
and all of the thing is
that we make that data available
freely to anyone.
So we'll have a website
which will deliver the data and off we go.
What really makes
tackling those really tough challenges
is being able
to have access
from different players as well, I imagine.
I would love to talk about your gliders
and, and the kind of work that comes from
that as well, because like you said,
there are some components is the computing
and then this is satellite as well.
And then there's the physical
having to go out there again and,
and how I guess
the technology in that space has made it,
I guess, safer for you
to also get this information
because as you said,
the ocean can be quite a
a rough place to be before iMOS,
which is the Integrated
Marine Observing System, which actually
funds the gliders and things like that,
just go out in the boat,
go out and ship and, you know, in fact,
this week, this month, now, 25 years ago,
we did the first ever research voyage.
Really? 25 years ago.
You know, Australia
even now has only one ship.
Okay.
And it basically goes around.
So you could once in two or three years
or four years chance of getting ship time.
So it's competitive.
You write a proposal and you get it.
So had 12 days.
And we basically collected data.
So that was the first time
we collected we did sort of
we say let's say 230 stations
that means the ship would go stop,
and we load the instrument
2,000 metres, comes back.
That's collecting the data.
Two hours sailing to the next one,
do the same thing.
So you do that and you said, now
the gliders are autonomous.
They actually do that will actually do
in that same 13 days,
it will cover 500 kilometers
or something like that.
Wow. And also will collect
something like 3,000 dives.
So you so remember I told you about
the resolution,
so you can have high resolution
and temporal.
Gliders do exactly the same compared
to a ship, right.
Much higher resolution, much more data
and a lot more data.
And the other thing also is that in ship,
when the weather gets bad,
you get to come back to shore.
Gliders are out there.
So it's collecting
very different types of data. Wow.
So there's a parallel in them.
In terms of before and after.
Yeah, I think that's the
what's so exciting to hear
about that kind of evolution
of, of these technologies
as well, is because you are getting data
that you probably have never seen before.
It's probably going through certain places
that you've never seen it.
I mean, we're collecting at the moment.
We're actually collecting data
from the Gulf of Carpentaria.
What's that?
That's up in the North Cape
York Peninsula. Okay.
So there's a town called Reaper
that's we've actually deployed.
And so we're collecting data.
And the thing with the gliders, you know,
in the same way, you know, in the
in the very high resolution,
both the time and space,
we're finding things that we never, ever
even thought about really,
because we missed them by having that ship
based, very low resolution data.
So that's what's amazing.
It's incredible to hear that.
I love hearing the fact
that there are certain ways in which,
you know, you're so used to a way
in which you can collect data.
And now as things change, you start going,
wow, what we didn't realize about this,
maybe we can bring that in
to our next project or, continue.
Yeah.
But so we find things from the gliders
and then
the gliders are not there all the time.
So how do you fill the gaps?
Supercomputing.
I would say it's supercomputing. Exactly.
And then
so you understand these processes
and said, oh look, this is happening.
So one of the main things that we found
with the gliders is what we call dense
shelf water transport. Okay.
So what happens in Australia
mainly is that during the winter months
there's a lot of heat loss from the ocean.
Okay.
Not land as well
but from like so the coastal waters
becomes cooler than the offshore waters.
Okay. But the cooler water is more dense.
So it sinks and goes along the seabed.
So it takes that's
how it actually takes the water away.
All right. Well, we never knew
this was happening.
So everywhere we put a glider in in way
in Australia we found this process.
So this is so unique in that sense.
So now we actually want to know.
Well okay.
Now can we simulate can we put it
in a computer and we actually take that.
Yes we can.
And then we actually say well you know,
how does it change.
You know,
how does it change to year to year.
Right.
So we kind of linear and not linear.
How does these systems change
in different locations
for these different forcings?
Once you discovered something like that,
did it really shift the way
in which you were hoping to,
I guess, study a particular project
and thought, well, this moving the gliders
a bit too much to the
so this is the set of
it's a different paradigm.
Okay.
So with iMOS,
which I sort of was part of the team
which set it up, you know, in terms
of like the proposal and things like that.
So the idea is that we will have global,
an observing system.
Okay.
So the ocean observing system,
so you actually put instruments there
and you actually make measurements
while doing those
measurements, you may find
lots of very interesting processes.
And then you actually write about,
okay, it's very different to say, okay,
here I'm going to
I never actually wrote a proposal
to say I'm going to look at dense water
transport.
This is my aim.
This is what I'm going to do.
This is, you know, the idea
of writing a research proposal there.
We did that right.
The data is there,
and it's always coming in.
Coming it just, 20 years later,
we're still doing
new things happening.
So that's sort of the different parts
on, on how things have changed.
So in the before,
we would write a proposal on the ship.
Right.
These are the processes
we want to look at sometimes.
You didn't know they exist.
So it was actually sort of you had a bet,
you know, if the conditions were right,
you would get this, that and the other,
but now it's there all the time.
Can you remember, like,
I guess one of the most,
some of
the really surprises you about the ocean
since you've started that I guess over
time, you've been like, that is something
that I don't think that we've
really realized about our oceans up to
this point was I talk about
the dense water,
we are unique in that way.
So that's the other thing
about the oceanography.
You know, if you, in Britain,
people say, are you from Australia?
Are you upside down?
So my phrase is that, yes,
we're upside down.
But in Western Australia
our oceanography is upside down as well.
That's completely opposite
to anywhere else in the world.
So a lot of things that you would actually
read in a textbook and generalize
doesn't happen in.
Right.
So that's been part of a challenge
to actually get some of the findings
that we do to be accepted
by the people who have very narrow
buoyancy in a way.
No, no, no, that can't happen. Right.
So you have to it's been a challenge
in that particular case to be able to.
I think I've seen you say at one point
in one of the talks online about how
the Indian Ocean
is the most misunderstood of all oceans.
You know, it's different.
Yeah, it's different
because it hasn't got a northern section,
in, in lots of these things, in a way.
I also say, is it
being in the wrong place at the right time
or the right place at the wrong time?
Right.
And I mean, I can actually say many,
many examples
in, in, you know, it's in there.
So in 2004, I was on holiday
in Sri Lanka and going to the beach
when the tsunami happened.
Wow. You were there.
So I got wet. Okay.
I was the only one of the few people
in the world,
I mean, actually only person maybe now,
but at that time
when after Japanese tsunamis, nobody's
actually somebody who works in tsunamis.
They’ve never experienced a tsunami.
Oh, right. Okay. Yeah.
So they're sitting in
labs and things like that
and do basically computer modelling.
So, so I was actually in that
system and then, you know, I was
then nominated by Sri
Lanka to be in the team
which developed the tsunami
warning system.
Oh right. Okay. Yeah.
So I was in charge of guess what?
Computer modelling.
Is it because you knew it?
Oh, was just like,
we need somebody to be able to do this.
Well,
I mean, it was that committees, right?
So you actually chair committees and you
direct the committee and things like that.
And, you know,
how can we actually develop the software
which is required for the thing?
And so you actually get to
I didn't actually do the grunting,
but also supercomputing facilities
and I can run.
So now my models much higher resolution
than anyone else can do.
So that you do as well.
So in all of this to be able to
I then not identify but take advantage
of what you have.
And build on that in the same way,
you know,
why should the plane just crash off
Western Australia basically.
And I'm the only oceanographer.
So everyone comes to you
to ask things about it.
And then you said, oh, yeah, it's cold.
It's in there.
And then you actually say,
well, I've got a minute.
We have much more skills than just saying
it's cold and rough and whatever.
We can actually use computing power
to actually do this.
So we did not, but nobody listened to us
because what we were saying
was far, too far fetched.
Right? Right.
Was we said, or any of this debris
will end up in the, western Indian Ocean.
So Mauritius or Madagascar
and all of that, nobody would say no
Australian.
The ATSB, the official, they said, oh,
the debris will end up in Sumatra.
So they actually told the Indonesian
government to go and look for debris.
All right.
I said hang on, that can’t happen.
We have these current systems
and whatever.
And of course
when the flaperon ended up in
Reunion Island,
everybody started listening to us.
Pawsey rang me up and said, “Oh,
would you like some supercomputing time?”
And we will actually--
So we were running it on a desktop again,
the same we can only do certain things.
So yeah.
So they, posted
gave us programmers and computing times
and we do all of these
pretty cool simulations.
And then when we were just
about finishing those simulations,
this person comes,
you know, and visited me from the US.
It was a lawyer, and he said, “Where
should I go
to find debris?” And I said, oh, look,
this is where you should go.
And, so off he went.
And he found debris
in the places that we predicted.
Oh, he's found 54 pieces.
54 pieces, and most of them from the areas
I told him to go.
And that comes from the simulations
that we did.
It's like late, late term vindication.
But still,
you just hope
at the time that we were like,
we understand this space,
we understand these waters.
Please have a listen to what
we're trying to say about it. Yeah.
I mean, yeah, it's sort of, interesting.
In the next two months,
they're going to go out to look for it.
Oh, they're going to go back
and look for it again. Really.
That's right. So fingers crossed
you please take
this PDF and read it for a little bit.
So is there any kind of like
I guess, collaboration with the
what you've done in the past
for something like that?
If they were going to go look out
or there they go.
I mean,
the thing is that, you know, is happen.
Nothing's changed.
Right.
Change
anything which happened in 11 years ago.
Yeah. 2014.
So I mean,
we haven't done anything since 2015
when we did supercomputing stuff
because there's nothing else to do.
You did all of that going
physically go and finding things.
Yeah.
It's incredible
to think that there was a time in which,
you know, the kind of searches.
I think I've seen it in
some of the videos that you've talked
about, the way in which
they were looking at particular zones,
like zone one, two, three
and then go back up.
And what you were able to do
was a lot more detailed.
Me, I'm going to go like, well,
this is the sort of two years of
is it current flow that you were able
to take the current flow
and then the current flow
will actually use the debris.
Then the learnings you did from, from the,
the currency did there
were you able to take that kind of studies
that we were using
and put it into different projects
you were working outside of that.
The other way
around. All the way around, really
completely the other way around.
That's what I'm saying.
We, you know, rather than saying, hang
on, it's
rough and it's windy and it's big waves
and it's cold out there.
We have the skills to actually trace
some of this stuff.
So that's basically
what we had been doing,
and we had all that expertise
long before imagery.
So it became a little
you already in the business of that.
You were also
it was basically taking that expertise.
So you know for example,
so in all of this, what we're doing
is that we have basically in the computer
tracking parcels of water.
Right.
That parcel of water may have, fish eggs,
it might have lobster larvae,
it may have dead seaweed.
It may have aircraft debris.
It might have plastics, may have turtles.
So anything is the same.
But we are actually tracking.
So how you track it
depends on what you're tracking.
But in terms of the principle
it's basically the same.
So we had a big project down in Port
Geographe in Basselton.
Okay.
Where there was a problem with seaweed,
which was basically accumulating.
It was smelly.
It was horrible environmental disaster.
So the government
came to us to actually redesign
and to solve this problem, which we did.
And, the government
spent $27 million, which actually,
based on our recommendation
to remodel the system. Wow.
So we had all of that expertise. Right.
And so MH 370? Yeah. No problem.
We can do that.
The work that you get to do in the future,
you have no idea how the applications can,
can, can be used for it.
But it's the same thing as people say,
you know,
sort of they say,
oh, you have an uncanny thing to actually
look at that and say, we can do that,
or this is what happens
and things like that.
And is there anything
you've been working on lately
or something that you're hoping to do
in the future
that you're really excited about?
My idea is to see how I can consolidate
a lot of knowledge
in here, and how can actually
convert
it, but I still haven't got the mindset
to actually be able to do that.
The way, as I said,
because the oceanography is upside down,
I have the opportunity to find new things,
you know, pioneering
things I can discover in a lot of them.
It's actually
the unique nature of WA
that what's actually allows you
to do things that,
you know, again, it's part of mentoring,
teaching, giving part of that knowledge
into the system.
Well I would I'd love to be able
to hear as well, because I know
you've done work in terms of
not just predicting
extreme weather patterns around Australia,
but also flooding.
I believe it was you have some, some
pattern of some, it's coastal flooding.
Coastal flooding. Yeah.
So we you know, it's basically
the same thing with the sea level.
Right.
So in terms of the extreme sea levels
that we've been talking about.
So the other thing to
to actually put the two things together.
So so we talk about tsunamis and,
and most people think of tsunamis
as originating from earthquakes.
Okay. Right.
It's another different types of tsunamis
which is called meteorological tsunamis.
And they're generated
by atmospheric effects mainly.
Okay.
Thunderstorms.
Oh, really? Okay.
So again guess what.
Southwest Australia is a hotspot.
So we get many, many of them a lot more.
And they're, we call them and salamis
are much bigger than the seismic.
So now and we may have 10 or 20 in a year
really that we don't realize
So these things are common.
And we've had problems with in Fremantle
Harbor where boats actually, you know,
ships have come ashore, you know, broken
their moorings and things like that.
Being in the right
place, everything happens.
So and I said that
during the seismic tsunami
I was in Sri Lanka.
So the day before
the Rottnest Swim that I was doing,
I've obviously I've had to have a look
at the weather forecast.
And they were predicting thunderstorms.
Oh. hang on a minute.
There's going to be a meteor tsunami.
So sure enough
there was while I was swimming. Wow.
So I'm the only person in the world
who has swum in a seismic tsunami
and a meteor tsunami.
I mean, that's a there's a hefty
CV of tsunamis you're collecting.
Yeah. Did did you.
Oh, sorry. Do you see what I mean? Yeah.
Right place, wrong time,
whatever this is, being able
to do that challenge of the swim
in itself is quite big.
But, then thinking this, you're
going into the water
that the weirs might be able
to, do I mean, in, in deep water, it's
not, you know, because I did it in a,
in a sort of a relay in a team.
So the only way I wanted to do
it was in a, meteor.
You know, the tsunami was there
because I saw the thunderstorm.
But then because you, you know, have
to get into the boats to have your rest.
Yeah, but getting on to the boats to just
the boat was growing up and down
and whatever it was, but it was only for
about 10 or 15 minutes and you know that.
Yeah.
So you can tell the others, don't worry.
They sit and, come down.
I was about to ask,
did you tell all the swimmers as well?
I was like, something's coming up.
And, you're just trying to get me riled
up, but, but then again,
something you learn rights.
And so I mean, in, in Sri Lanka,
I kept telling people
and said, hang on a minute,
you know, it's come and gone.
The tsunami lasts ten, 15 minutes.
That's it. It's come and gone.
And then I went to this beach
and then this huge tsunami comes.
So I was not thinking
then I'm actually driving back
and people were running.
What is going on?
This is not the tsunami.
But I have learned, and researched
and things like that.
So I got and then I got the,
I managed to get a tide
gauge established because once visiting
Sri Lanka, I was telling everybody
straight up is one of the luckiest places,
because they don't have global warming.
They say “What you mean?” I said
well you don't have any sea level rise.
And people say, “What you mean?” I said,
well you don't have a tide gauge,
you're not measuring. So how do you know?
That's fair.
So and then when I came back,
I basically contacted
some of my colleagues,
colleagues in the US,
and they agreed to put ice tide
gauge there.
Yeah.
And that was established sort of
two months before the tsunami happen. It's
one of
the few tide gauges which is operating
to record the tsunami.
Right. What timing.
What timing again, you know,
initiated by me again, you know,
anyone actually found out
was that those big waves were coming.
The waves had gone past Sri Lanka
hit, Maldives
got reflected and coming back.
Oh, okay.
So that's the first time
that people have shown how these.
And then we looked at Western Australia
that the biggest waves in Western
Australia came 18 hours afterwards because
they went all the way to Madagascar.
Okay.
But wow, it's interesting to be able
to tell people that to you is like,
where did this come from?
Well, actually, we can show you
this is how it's exactly.
The tide gauge is basically is a tube.
Okay.
So what you want to do
is to actually not have waves.
Right.
Because you're actually measuring
very the mean sea level okay.
So you basically have a tube
with a little hole in the bottom
and you have a float okay.
So when the water level goes up
the float goes up and down.
And then is the old way was to actually
connected that to a pulley.
And it'll actually do an analog trace.
Oh okay. So that's the old one. Yeah.
So now they rather than the float,
they actually have a, radar system.
So basically measures just like an echo
sounder, the height of the water level
inside the tube.
Again, another change in the technology.
That just makes it a bit more accurate.
Yeah.
So again, you know, before the analog,
traces were digitized every
hours, which missed everything
which happens in between.
Right.
So now we are measuring at one minute
or twice a second.
So you're saying
look for different things.
That's a huge change in resolution. Right.
But it's all into
you know it's all development.
Because of basically technology mainly
memories.
Right. Batteries
and different way of measuring things.
Is there any other projects you'd love
to talk about that you're in at the moment
that has utilised supercomputing
in any particular way when I recommend.
All right.
So there is one part
that we talked about in
terms of the rip currents.
It's a fantastic study in terms of,
we could never have done that
to understand how these currents form.
And where they form and how all of that
was actually done with these
super study in time.
So. Right.
People are still doing that
same experiments now, 25 years later.
But it's, because they, you know,
not many people
have the access
that we've had to supercomputing.
But you can see that that's the same thing
that I was actually saying.
I’m lucky
to have Pawsey and iVEC.
But if I were somewhere else in Australia,
I wouldn't have that right.
Or in any other country,
it's having that facility and then it's
not only having the facility,
but to actually take advantage of it.
So that and then we actually had
so we were starting.
So Western Australia we're
really interested in the Leeuwin Current.
Okay.
Which is and what we call
an anomalous current, which actually goes
brings warm water from the equator
down to the southern latitudes.
Okay. Other countries,
it's the other way around.
Oh, right. Got cold water going.
So that's why it's different.
So we're under trying to understand.
So I had students
who are basically looking at,
let's say between Geraldton and Perth.
And we were simulating that.
And to understand how this is happening
and when the you know
and then we actually expanded
to Western Australia,
then we expanded to whole of Australia,
and then we expanded it all the way
from Singapore to New Zealand
now and then, rather than running it
maybe for a few weeks, a couple of months.
So those days
we could run summer and winter,
in between we could not
because we didn't have the ability,
even with the supercomputers
with like Cognac and
Carlin,
the initial supercomputers,
they still didn't have the power.
But now with Setonix and things like that,
we can run for 20 years
longer in terms of the system.
So all of those actually will develop
in that system.
And so huge.
Just to think like how the, the,
the potential
just keeps growing and growing.
And it just helps you to sort of reframe
what do you what do you want.
So you start, you know, again,
you know, running
tsunami models to see
what is the risk for Western Australia.
Right.
So we're lucky to a certain extent,
you know, South of Geraldton
because all the waves coming from
the north, it's deflected
because there's a big underwater
topography okay.
Right.
Which actually protects us to actually
to do the computing and actually show
this is what really happens.
It's good.
It's quite comes in handy, right?
Especially when you're trying to predict
things that are happening in the future.
But the thinking also is that, you know,
would, you know, with tsunamis
and storm surges
and all of that is human safety
infrastructure, right?
So a lot of that is also got,
sort of satisfaction
in terms of, you know, tsunami warning
if you can save one life.
For sure.
It's fantastic in a way.
But it's, you know, when we're actually
doing the tsunami thing because, you
know, again, with all of these fundings,
it's a hard part.
Many times we've been asked and says,
can you put the cost
on a human life?
A dollar amount sort of thing.
It's really interesting.
Nobody ever wanted to when we were doing
tsunamis wanted to put a dollar value.
No! That's on the price of a life.
Even if it was this--on
COVID because it was so
important, to
say how much money are you going
to spend to contain this.
Yeah.
You needed to prove the reason
why you were having such a focus on it.
Yeah, it's fascinating in that way.
Yeah.
I would hate to be the person
who put that together.
And then people thinking, how could you?
How dare you?
But the understanding is
we need to be able to to justify
why we need to do this study, for example.
I mean, yeah.
So in our case, you know,
zero deaths,
that's that's the bottom line.
Yeah.
There is a continual drive for not just
information, but hitting these challenges.
Right.
There's always a challenge to be solved.
And, you know, in often
I don't think people realize
to a certain extent
how much effort that goes in.
Still this challenge.
So the, the challenges, you know,
whatever supercomputers we have,
we still can't line climate models,
okay.
For what happens in 2050
or even in ten years
time at the resolution that we want
people talk about.
Oh, yeah, the cyclones are, you know, be
going to big storms are going to
but we can simulate that
because we don't have the computing power
to be able to do that. Right? Yeah.
So there's still challenges ahead
to be able to
to run these models
at the resolution to actually resolve
the different processes.
So none of the climate models resolve
a tropical cyclone.
None of the climate models
resolves the La Niña and El Niño events.
We have no idea. Right.
All of this. So at the moment
it's if I give you the analogy,
the climate models are my Son Micro
Workstation.
Right.
That's the level that we are in.
We’re
just in the place to actually do that.
And this is big groups around the world,
trying to do things,
but we're not there, you know.
Yeah.
From a computing pod from the ocean
and the atmosphere.
That is the big challenge
because it's I'm not going to do that.
You're supposed to do it now.
I guess I just
it proves that
I guess it's really important for us
to continually invest in these sources.
Absolutely, absolutely.
It's provides
societal answers, societal questions.
Definitely.
I really appreciate your time with me.
Chari.
It's good to hear
about all your wonderful adventures.
I'm sure there will be many more as well.
Yeah, but I can't thank you enough. Nope.
You're welcome. Thank you.
Thanks for listening
into a chat with Chari.
There were a lot more stories
for us to share with you, and
we can't wait to release our next episode.
If you are interested
in what supercomputing can do to research,
visit our website for more stories
like this.
Until then, we'll talk soon.