0:00

Brought to you by Toyota. Let's

0:02

go places. Welcome

0:07

to Forward Thinking. Hey

0:12

there, and welcome to Forward Thinking, the

0:15

podcast that looks at the future and says,

0:18

it's like rain on your wedding day.

0:20

I'm Jonathan Strickland, I'm La,

0:22

and I'm Joe McCormick. And today

0:25

you don't need a weatherman to know which way

0:27

the wind sucks, because we are going

0:29

to be talking about predictive

0:31

modeling of weather, weather forecasting.

0:34

Yeah, we've talked in

0:36

the past a lot about weather and

0:38

sometimes when I wasn't here. Yes,

0:41

we had a two parter about the potential

0:43

future of weather control with special

0:45

guest Julie Douglas back in February.

0:48

Yeah, And one of the interesting things about that episode

0:50

is I think in the end we decided, after

0:52

all of our research that really the

0:55

best avenue for humans to sort

0:57

of get a grip on the weather is not

1:00

to try to control it, because in many ways that

1:02

is a fool's are end, it's physically impossible,

1:04

Yeah, but to instead try

1:07

to understand it, just to have

1:09

a better better idea of what's coming

1:11

your way and win right.

1:13

The further out and the more accurately

1:15

you can forecast the weather, the better prepared

1:17

you are for the various eventualities

1:20

that will unfold, things like flooding.

1:23

Like if you know ahead of time that flooding is

1:25

is almost certainly going to affect

1:27

a certain region, you can start to take steps

1:30

to protect people and property in

1:32

that area. Sandbags are an amazingly

1:34

effective low tech solution to things or

1:36

maybe out there. They don't do much

1:38

good if they're not there, Yes that's true. If they're they're

1:41

somewhere else. If they're in a warehouse, that warehouse

1:43

maybe nice and dry, but the

1:45

area that you were hoping to say, will be

1:48

rather squishy. Uh same same

1:50

sort of thing that if you're talking about, like you're looking

1:52

ahead at a very long term forecast

1:54

and you were to say, oh, it looks like there's not going

1:57

to be any rainfall for quite some time,

1:59

you can start to make plans for that

2:01

so that you're not stuck in a

2:03

situation where it happened

2:06

but you weren't aware that that was

2:08

going to you know that was going to be the case. Uh

2:11

So, in other words, we don't necessarily try and

2:13

control it. We just get a better idea

2:16

of what is going to happen, so we're more

2:18

prepared for that. Yeah, And we glanced

2:20

across that topic in those Future of Weather Control

2:23

episodes. But uh yeah, so we

2:25

wanted to talk about that today. And we were also inspired

2:28

by a video episode that you did,

2:30

Jonathan about Bubble Yeah, the Bay

2:32

of Bengal Boundary Layer Experiment

2:34

or Bubble Yes. I

2:36

I said in the video that I consider

2:38

myself a bubblehead because I'm a huge

2:41

fan of this project. The

2:43

video and that just came out this week. You can check it out

2:45

on YouTube this very day if you would

2:47

like to, or on fw thinking

2:49

dot com. But specifically,

2:51

Bobble is a very particular

2:54

regional weather predicting project.

2:57

Yeah. It's a study of how a

3:00

number of complex factors in the Bay of

3:02

Bengal come together to create a

3:04

monsoon season of heavy rains in northern India

3:06

every year. And it's a particular

3:09

interest to researchers because that monsoon

3:11

season drives the agriculture and the

3:13

water supply and the energy supply for

3:15

about a billion people, so seven

3:18

of the world's population, no big uh,

3:20

And and clearly variations

3:22

in the seasonal norm of rainfall either too

3:25

wet or too dry reek havoc

3:27

on this region. So what if

3:29

we could predict those variations before

3:31

they happen, disaster could hypothetically

3:34

be if not prevented, then then perhaps

3:36

mitigated. Uh

3:38

Okay. And so besides being a

3:41

project that could hypothetically change the lives of

3:43

a billion people, Bobble is really

3:45

cool because it's kind of a microcosm of

3:47

weather prediction research in general, because it's

3:49

it's so multi disciplinary. You've got ships

3:51

and satellites making classic observations

3:53

in the bay. You've got robotic submarines

3:55

that are checking out the situation under the surface. You've

3:58

got researchers designing did little simulations

4:00

to crunch the data, and they'll

4:03

compare their models to the actual season's

4:05

results to see where they went right and

4:07

where they need to make improvements. Right. So we're going to

4:09

talk more about Bubble in detail

4:12

a little bit later, But first, as

4:14

per our usual m O, we like

4:16

to go back and look at how we got to

4:18

where we are now. Like, like, obviously,

4:21

when you look back to the ways

4:24

humans tried to forecast the weather

4:27

centuries ago, they're supercomputers

4:30

were sorely underperforming.

4:32

Yeah, so those advocacies

4:35

didn't process it quite

4:37

the same thing exactly. You have

4:39

all of your little scribes working in parallel,

4:42

attempting in vain to simulate

4:45

weather well, and the thing they were trying

4:47

to calculate was how angry the god was.

4:50

So there were several

4:52

steps along they were going, going off

4:54

the path in a few different ways. So let's

4:56

let's talk about let's talk about you know, kind

4:58

of the a shoudn't approach to

5:01

forecasting weather and work our way

5:03

up to what we tend to do today.

5:06

Okay, well, joking aside,

5:08

there were of course lots of just straight

5:10

up magical thoughts about

5:12

how to control the weather or predict the

5:14

weather originally, and so that's you know, that goes

5:17

that's a tradition that goes way way back into

5:19

the ancient world. Has to do a lot with that with astrology,

5:22

um and as kind of an offshoot

5:24

of astronomy, but mostly it was astrological

5:26

yeah, um, But those

5:29

those sort of magical predictive

5:31

interpretations. Aside, there were

5:33

actually throughout history plenty

5:36

of weather superstitions

5:38

and sort of rules of thumb that

5:41

actually do have grains of truth

5:43

to them. Yeah, there's a really

5:45

great article on how stuff

5:47

works Dot com about this, and I did a what the Stuff

5:50

video about it once and and it's

5:52

it's interesting, how many of them really do

5:55

hold water? Yeah? Yeah, Well it makes

5:57

sense because you figure people

6:00

are paying attention to what

6:02

has happened, and they realize that there's a

6:04

pattern where when

6:06

once a circumstances happened,

6:08

then typically you might get

6:10

a lot of rain. And so you start to make

6:12

a rule about that, and you know it's in

6:14

some cases it can be you can

6:17

be completely off base. It's just coincidence,

6:19

or you do what I like to call you. It's

6:21

it's called, you know, a confirmation bias, but

6:23

I would call it. I would say the van is always

6:25

at the corner, which is where whenever

6:28

there's a van parked at the corner, you notice it. Whenever

6:30

there's not a van parked at the corner, you don't

6:32

register it. So to you, the van

6:35

is always at the corner. Uh. In

6:37

those cases, obviously it may be that

6:39

you've made an observation, but it's a faulty

6:41

one. However, there's some that are

6:43

at least somewhat you know, reliable.

6:46

Yeah, here's one. You've probably heard

6:48

some version of this weather prediction before

6:51

and often in couplet form about

6:53

red sky at morning, sailor take

6:56

warning, red sky at night, sailor's

6:58

delight. I feel this. I feel badly for

7:01

that one sailor right like it's

7:03

just like like, oh man

7:05

is going to be awful because

7:07

it says sailor was

7:09

singular, so it's one guy. Well, that's the way

7:12

I always heard that. There are other versions. But

7:14

this is old, old, old, It goes way

7:17

back. People have been using this forecasting rule

7:19

for at least a couple of thousand years. We know

7:21

because it shows up in the Bible, shows

7:24

up in the Gospel of Matthew chapter

7:26

sixteen, where uh it says quote

7:29

the in RSV, the Pharisees

7:31

and sad Juicees came and to test Jesus.

7:33

They asked him to show them a sign from heaven,

7:36

and he answered them, when it is evening,

7:39

you say it will be fair weather, for

7:41

the sky is red, and in the morning

7:43

it will be stormy today for the sky is

7:45

red and threatening. You know how to

7:47

interpret the appearance of the sky, but you

7:50

cannot interpret the signs of the times.

7:52

So obviously that they're trying

7:54

to make a spiritual or religious point

7:56

there, but but just incidentally in the

7:58

narrative, at least we know that some people

8:01

back then we're saying this rule. Um

8:04

so, so the author of this passage had

8:07

heard of this before, and crazily enough,

8:09

it is partially true. So

8:11

what's the scientific basis for this? Why would the

8:13

color of the sky at sunset

8:15

or sunrise have anything to do with the weather?

8:18

Well, strongly tinted

8:21

red light at sunrise and sunset

8:24

actually tells you something about

8:26

the contents of the atmosphere

8:28

between you and the sun. So

8:30

specifically, it tends to indicate dry

8:32

air filled with dust and solid

8:35

particles which we would call aerosols.

8:37

So these particles in the air are the

8:40

cause of the reddening of the light because dust

8:42

and aerosols in the atmosphere scatter visible

8:44

light in a way that makes the light turn red.

8:47

Uh And in turn, this dry, dusty

8:49

air tends to indicate that you're

8:51

in a high pressure region, which means

8:53

less cloud formation and less

8:56

likelihood of a storm. A low pressure

8:58

region, on the other hand, would mean that

9:00

that they're tended to be more cloud formation

9:02

and more storms. So if you are looking

9:05

through red tinted atmosphere

9:08

to see the sun you're looking

9:10

through a high pressure region that's

9:12

less likely to rain on you. Uh

9:14

And and the thing about the atmosphere is that

9:16

it travels in

9:19

in the same direction. Well yeah, and that's

9:21

what this rule doesn't work everywhere, because while

9:23

the Sun's path is unidirectional

9:25

around the Earth, of course it's actually the Earth's rotation,

9:28

but metaphorically, the Sun's path

9:30

is unidirectional. I'm gonna need to see a site

9:32

for that. The weather

9:35

tends to travel in different directions

9:37

depending on where you live. So if you're

9:39

in the Arctic or the Antarctic, or in the

9:41

tropics, the sort of three extreme bands,

9:44

weather patterns more often move east to

9:46

west, and this rule doesn't

9:48

apply, or in fact, actually I guess the opposite

9:50

would apply, right, But for the mid

9:53

latitudes, you know, sort of the temperate zones

9:55

between the tropics and the Arctic or the

9:57

Antarctic, this is actually

9:59

more often true because the

10:01

weather patterns more often moved from west

10:04

to east. And what that means is, if you

10:06

look towards the sunset, you're

10:08

looking west at the weather that's

10:10

probably coming your way. And

10:13

if a red light scattering patch is

10:15

to the west of you, that's a high pressure area.

10:17

Probably that's probably headed your way,

10:20

meaning the weather will probably be fine. Uh.

10:22

And of course why would red sky at morning be a problem.

10:25

Well, that's because if you're in the mid latitudes again

10:27

looking east toward a sunrise, you're

10:29

seeing the weather that has probably already

10:32

passed by you. And high and low

10:34

pressure systems often do trade off in cycles.

10:36

But you may have noticed that I kept

10:38

saying the word probably over and over

10:41

again there, And that's because like all

10:43

weather prediction, this is probabilistic.

10:45

Using the system, you can predict

10:47

the weather better than random guessing,

10:50

meaning better than with fifty percent accuracy,

10:52

but still not anywhere near ad accuracy.

10:55

Right, So there could be some mornings

10:57

where you see a red sky and every it's perfectly

11:00

fine, it's beautiful weather. And there might be

11:02

some evenings where you see red sky and the next morning

11:05

you're soaking in it. Yeah, So the weather,

11:07

the weather is just very complex. It's it's

11:09

difficult to it's difficult to protict

11:12

with accuracy even now

11:14

using the supercomputers and everything

11:16

that we have involved in all the data we have,

11:19

but this one piece of folk

11:21

science and weather forecasting, it's

11:24

not the only one that turns out to have some basis

11:26

in truth. Right. Yeah, a few

11:28

others that I wanted to touch on because they're they're kind of

11:30

a favorite. Uh. Ring around the

11:32

moon rain real soon. Have you guys ever heard

11:35

this? This is the thing that you've heard. No, no,

11:37

not at all, but I believe you. Yeah. Uh,

11:39

there's there's another kind of version of it that

11:41

goes when a halo rings the moon or sun rains

11:44

approach and on the run. I

11:46

love that

11:49

sounds like something from one of their songs, and

11:52

and and the thing that's going on here it is

11:54

it does hold true

11:56

more than fifty percent at the time. I think it's

11:58

it's a similar probabilistic concept

12:00

to to the red red sky

12:02

at night Sailor's Delight sort of thing. But so,

12:05

so what's going on here is that, um,

12:07

when you've got a halo that

12:09

frames the moon or the sun, it's

12:12

produced by by moonlight or sunlight refracting

12:15

through high whispy clouds

12:17

that are made of ice crystals, and

12:19

uh and those those ice crystals. That

12:22

type of weather pattern typically occurs

12:24

in siro stratus clouds that often

12:27

move in ahead of weather fronts, where

12:29

where temperature differentials are going to cause

12:32

warm air to move upward, deensing moisture

12:34

and potentially forming rain clouds

12:37

potentially, So science

12:40

science thumbs up on that one. And

12:42

still not the only Moon

12:44

related weather, you

12:47

know, kind of folklore, right, sure,

12:49

Sure there's also clear moon frost

12:52

soon, Yeah, which which

12:54

makes perfect sense because because

12:57

clear nights do often mean that cold

12:59

weather is on the way, Because as far

13:01

as the planet is concerned, a cloudless

13:03

sky is sort of like having a bed

13:06

without blankets. Uh. You know, During

13:08

the day, the Earth absorbs sunlight

13:11

and and can converts it into into

13:13

heat that we all appreciate to

13:15

certain degrees um. When when the

13:17

sun sets, the surface begins radiating

13:19

that heat back out, and lacking

13:22

clouds to capture the heat and

13:24

snuggle it in all all tight and close, the

13:27

surface and the lower atmosphere grow increasingly

13:29

cold. In fact, I think in a tech stuff

13:31

episode I talked about this as

13:33

a means of creating ice in

13:36

certain regions, where you'd leave out a pan

13:38

a shallow pan of water outside

13:40

because the heat radiates out and it actually

13:42

becomes ice that way in certain regions

13:45

of the world. That's how it was done before refrigeration

13:47

reached those areas, so it's kind

13:49

of neat. Yeah. My favorite one though, has

13:52

to do with cows. Of

13:55

course, there is there's folklore about

13:58

uh or not, like a folk saying,

14:00

but yeah, that cows will lie

14:02

down when it's about to

14:04

rain, mm hmm. And

14:07

and I will, I will admit that cows

14:10

lie down for probably many reasons,

14:12

like they're tired. But

14:16

um, but but this one, but this one might be

14:18

due to to body heat. Okay,

14:21

cows tend to stand more often when

14:23

they're overheating, you know, in

14:25

order to breathe everything out right.

14:27

Sure, yeah, so so as seated

14:30

cow could arguably I

14:32

mean that the weather is cooling

14:34

down and therefore a storm

14:37

is a bruin. I also

14:39

like in the notes you have, this one may have a leg to stand

14:41

on. There. There are so many puns

14:44

in this in this house stuff works article. And

14:47

I yeah, I

14:49

didn't write it, no, oddly

14:52

enough, Yeah it was not it

14:54

was not, I but

14:56

but there. But there are definitely some some more

14:59

systematic approaches that people have come up

15:01

with over the years, sure, apart from just

15:03

sayings in folk wisdom. One big one

15:06

through in history is Aristotle's

15:08

Meteorologica. That's Aristotle's

15:11

hugely influential treatise on winds,

15:13

water, weather, and some other stuff like earthquakes.

15:16

Like much of Aristotle, it is

15:18

both startling lye intelligent and

15:21

hilariously wrong about lots

15:23

of things. I enjoyed the section

15:25

on how earthquakes are caused by evaporation

15:28

of rains that have soaked into the earth and

15:30

exhalations of breath from the ground.

15:33

But until a few hundred years ago,

15:35

I think the Aristotle's works were sort

15:37

of the Western world's gold standard

15:39

for knowledge about the causes of weather.

15:42

And it wasn't until you

15:44

know, fairly recent times that

15:46

we started being able to do much better. Yeah.

15:49

I mean generally speaking, you started

15:51

getting into like the mid

15:54

to late Renaissance, and you start seeing some other

15:56

thinkers propose alternatives

15:59

to some Aristotle's ideas. But

16:02

yeah, his his approach

16:04

or his his observations and his

16:06

his uh writings held sway

16:09

for centuries. Yeah yeah, um, And

16:11

and some of those new ideas came about

16:13

alongside changes in

16:15

concepts about physics and also about

16:18

astronomy, like like greater knowledge of astronomy

16:21

um up to and including the publication

16:23

of almanacs, which were very

16:25

very popular publications back in the day. Apparently

16:27

the only thing that outsold almanacs in the seventeenth

16:30

century in England was the Bible,

16:33

so lots of people were purchasing these things. Um.

16:35

And back in the late seventeen hundreds

16:38

and early eighteen hundreds, a couple different

16:40

mathematicians slash astronomers

16:42

started publishing yearly farmers almanacs

16:45

here in the in the States and what would

16:47

be the United States later on the

16:50

North America continent. Yes.

16:53

Um. The formulae, the formulas

16:56

that they use in order to make these predictions

16:58

are to this day guarded

17:00

as family or company secrets.

17:03

It turns out like it it could be something

17:05

like consulting the family cat. We

17:07

don't know, Yeah, and like

17:10

intensely guarded. I love I

17:12

I love stories about old farmers

17:14

are almanac and UH and the

17:16

Farmers Almanac, both of which are punctuated

17:19

slightly differently in terms of the possessive s, but

17:21

just the lower around all of this is is delightful.

17:24

In the case of one of the two almanacs,

17:26

I forget which one. UH, there

17:28

is a Caleb Weatherbe

17:31

who's sort of like the James Bond of

17:34

of this of this company. Because Caleb

17:36

Weatherby is not his real name, I'm

17:38

not sure if it's a dude. Uh I

17:42

there have been this series of Caleb Weatherby's

17:44

who have been the one entrusted with

17:47

the knowledge of how the of how the almanac

17:49

does it stuff. It's like cecil atoms, yes,

17:51

of straight of the straight dope. Yeah, there have

17:53

been many cecil atoms. Yeah

17:55

so, but so no one. No one knows

17:57

exactly how they make their predictions, but supposedly

18:00

take stuff like planetary positions

18:02

and sun spots and lunar cycles

18:05

and title patterns all into account,

18:07

and I get the distinct idea reading

18:10

stories about this that meteorologists

18:12

find find almanax like this rather

18:14

quaint. Uh what

18:16

One researcher who looked into the accuracy of these kind

18:18

of things found that they get their long ranging

18:20

predictions because they make predictions a year

18:22

or two out correct about

18:26

of the time. Is that a high

18:28

number or alone? Like how much variability

18:30

is there and what they could be predicting?

18:33

You can't because you wouldn't say, like is that

18:35

better than chance? Because it's hard to say without

18:37

knowing all the variables. Oh, sure, I'm not

18:39

sure. They claim to get it right about eight percent

18:41

of the time, and and that is that

18:44

is sore a

18:46

gap. Yes, but luckily

18:48

we didn't. We we haven't had to continue

18:51

relying just on stuff like

18:53

this forever because eventually, UH

18:56

physics, Yeah, people started

18:58

figuring out how hydro dynamics therm

19:00

thermodynamics both work, and once

19:03

humanity got a really good grip on

19:05

these concepts. Strangely enough, around

19:07

the same time that the American farmers almanacs

19:09

started publication, the science

19:12

of meteorology could take off, and by

19:14

the early nineteen hundreds, a Norwegian physicist

19:17

by the name Wilhelm Erknus devised

19:19

the first known seven equation

19:22

formula for for using observations

19:25

of existing weather conditions to solve

19:27

for future conditions. Taking taking

19:30

into consideration like like pressure and temperature

19:32

and humidity and then three aspects

19:34

of atmospheric motion. That forms the foundation.

19:37

Definitely, I mean, the more information we have, obviously,

19:40

the better picture picture we have what's going

19:42

on right now, and the more

19:44

um the more accurate we can

19:47

make a forecast for the future. Of

19:49

course, the further out you go from

19:52

the current UH scenario,

19:54

the current the current condition. Small

19:56

differences in in what you've predicted versus

19:59

what actually happened add up tremendous.

20:01

Yes, yeah, well, I mean it's a it's a sort

20:03

of principle of physics that you can extrapolate

20:05

on a very simple scale or on a very huge

20:08

scale. On the simple scale, imagine

20:10

aiming an arrow at a target. If you shift

20:13

your aim a millimeter over and the targets

20:15

a foot of way a foot away, it's not gonna make much

20:17

of difference. If the targets a hundred

20:19

feet away, it will make a difference, right,

20:22

So, same sort of idea is that you know the the

20:24

temporal distance as opposed to physical

20:27

distance, it does make a big difference.

20:29

But of course, once you get into the modern history

20:32

of our technological and scientific

20:34

capabilities for predicting whether one

20:37

big difference, of course is just going to be the scale

20:39

of of observation, increasing the

20:42

number and accuracy of observational

20:44

platforms to collect data about the

20:46

weather, so we have more information to work

20:48

with, uh, And that's pretty

20:51

easy. But another thing is that we can sometimes

20:53

overlook the simple ways that common

20:56

technological innovations help us in

20:58

specific ways, And one would be communication

21:00

technology such as the telegraph

21:03

originally and then like the telephone facts

21:05

and uh and the Internet, and these have allowed

21:07

people to better understand global weather

21:10

patterns in real time by rapidly sharing

21:12

and comparing information about

21:14

local weather. Yeah. Computer

21:16

science also allowed prediction

21:19

to to greatly advanced, starting in the fifties and

21:21

sixties and really ramping up over the past

21:23

say like twenty to thirty years, along

21:26

with the rate of our processing power. So

21:29

I mean, perhaps obviously, as our computational

21:31

ability and our observational ability

21:33

have increased, so has our forecast

21:35

accuracy. There was an analysis that was published

21:38

in Nature in and according

21:40

to that, the forecast accuracy for

21:42

the next three to ten days of weather has

21:45

improved by about a day per

21:47

decade um, meaning that

21:49

right now our ten day forecasts are

21:51

as accurate as nine day forecasts

21:54

were in the early odts. So,

21:56

in other words, every decade we go by, we're

21:58

getting one day better. Yeah.

22:01

I like it. So if I can figure out

22:03

whether or not I need to carry an umbrella with me

22:06

on Friday when it's Monday,

22:08

and and be reasonably certain that

22:10

that is in fact the right answer,

22:13

the better because I'm

22:15

not carrying it. If I don't have to write. In a

22:18

decade from now, you'll you'll be able to know

22:20

pretty well on Tuesday. I'm

22:22

looking forward to that. So

22:24

my suggestion, Jonathan, is that you need

22:26

to get a cooler umbrella that you feel

22:28

better about carrying all the time, Like maybe

22:30

like a penguin's umbrella, you know that

22:32

shoots machine machine gun fire

22:35

or has a big sword that comes out the end of it. I

22:39

have a blade runner umbrella

22:42

that's great glowing. Yeah,

22:45

I've got one of those. Um So,

22:48

who is really in charge

22:50

of gathering and crunching all this data?

22:52

I mean, I'm assuming when I turn

22:55

on the local news and I see

22:57

the local weather corresponded

23:00

on the news, that person hasn't

23:02

personally been responsible for gathering

23:04

and analyzing all that information. He

23:07

has no,

23:09

no, no, no. The guy I'm imagine

23:13

very specific, that guy launched

23:15

the satellite uh and has

23:18

collected the data. He built all of the computers

23:20

himself. Uh No. Modernly, weather

23:22

prediction is a joint public

23:24

like governmental and private industry

23:27

type of business because that the satellites,

23:30

the computers, the software, and the

23:32

the human compilation of all of this data

23:35

that go into it is each

23:38

each of those separately are huge expensive

23:41

arms of the venture. So and

23:44

and going into it, you know, like, of course you've got

23:46

local news stations, which are private companies

23:48

that are reporting on whether but

23:51

it's also a public service. It's it's

23:53

not just about personal convenience. It's absolutely

23:56

a very critical public service about getting

23:58

information about big

24:00

storms, danger, tornadoes, hurricane,

24:02

stuff like that out to the public um And

24:05

it's also partially a a tool for

24:07

commerce. The more that companies can learn

24:09

about what the weather is going to do, the better

24:11

that they can adjust whatever it is that they

24:13

need to adjust depending on what's sure. Like

24:15

if if you're part of the shipping company,

24:18

whether you're shipping stuff across land

24:20

or see you need to know

24:23

these sort of things because that can have a real

24:25

impact on everything from a delivery

24:28

date to the safety of the people and the products

24:30

that you're moving. Weather is

24:32

important, I mean, it's important to have this as

24:35

accurate a picture of what's going to happen. And

24:37

of course the further out you can do that, the

24:39

more beneficial it is for everybody.

24:42

So that kind of leads us over into the discussion

24:45

of some of the current attempts

24:47

to get an even deeper, more

24:49

keen understanding of the factors

24:52

that influence whether UM

24:55

and that kind of brings us also to Bobble,

24:57

to that project we were talking about off

25:00

the coast of India. So Bobble

25:02

is pretty cool in that it's it's relying

25:05

upon multiple sources

25:08

to gather information UM

25:10

also that we can get a better understanding

25:13

of the monsoon season in India. So

25:15

that includes satellite data, atmospheric

25:18

measurements courtesy of an f A a M

25:20

aircraft and I'll go into that in a second,

25:23

and some floats that are carrying

25:25

scientific equipment, as well as those underwater

25:28

robots that Lauren mentioned that are incredibly

25:31

cool. I was so interested

25:33

to hear, mostly just about how they move

25:35

through the water because it's a

25:37

brilliant and simple means

25:40

of propulsion. But first of all, the

25:42

project has a collaboration between India

25:44

researchers and scientists from the UK, specifically

25:47

the University of East Anglia and the University of Reading,

25:50

and the research will take place during

25:52

the two thousand sixteen monsoon season, which

25:55

has technically started as we record this

25:57

podcast. It's June and July. So

26:00

the monsoon season is India's rainy season.

26:03

India gets a lot of its

26:05

rain during the season. Of the

26:07

rain that falls in India falls during

26:09

the monsoon season, and there is a lot of Yeah,

26:12

we're talking ten ms annually of

26:14

rain. Ten ms, it's thirty three ft

26:16

or so. In some places it's up

26:18

to eleven ms. It depends on the region

26:21

of India. Um.

26:23

So the project's goal is to gain a deeper understanding

26:25

of the factors that influence this monsoon season

26:27

and that way we can make better predictive models

26:30

of what areas of India are going to get

26:32

what amount of rain, and that will help

26:34

subsistence farmers plan out there they're

26:36

farming to make certain that they

26:38

take the best advantage of that. It also will help

26:41

in the case of figuring out

26:43

this particular region might be very

26:45

susceptible to flooding and we

26:47

need to take measures to protect the people

26:49

who live there. Right, So there's

26:52

there stands to be a really incredible

26:55

benefit too. Like we said

26:57

earlier, up to a billion people

27:00

to to cracking this code, to figuring out better

27:02

how it works and therefore how to predict it. Right.

27:05

So first step of course is you

27:07

gotta get the data right. You have to collect the data

27:09

before you can do anything with it, and that's where

27:11

all of that equipment I mentioned comes

27:13

into play. So first we have the

27:15

f A a M aircraft. F a

27:17

a M stands for a Facility for Airborne

27:20

Atmospheric Measurements, So it's

27:22

flying through the atmosphere gathering

27:24

data on the atmosphere as it moves

27:26

through. It's pretty uh

27:29

interesting. You need to take a little look

27:31

at the picture of of these things as

27:33

a special refitted B a E Systems

27:36

aircraft out of the UK and

27:38

uh it's the result of a collaboration

27:40

between the Natural Environmental Research

27:43

Council and the Met Office in the United Kingdom.

27:45

Now, the f a a M has a collection of sophisticated

27:48

instrumentation aboard it. They can those

27:50

instruments can measure everything from radiative

27:52

transfer so essentially the way heat is

27:54

moving through the troposphere, the

27:57

chemical composition of the atmosphere, humidity,

27:59

tem sure turbulence, cloud physics

28:02

and more that turbulence in the cloud physics

28:04

that's really important. Things like vertical sheer

28:06

that has a huge impact on

28:08

weather patterns and it's one of those things

28:10

that we need to have a lot of data on in

28:13

order to really understand what's happening, and

28:15

the team will actually compare the data

28:17

gathered by the aircraft to that

28:20

from the other sources the floats, the weather

28:22

satellites and underwater robots to get a complete

28:24

picture of what's happening in the bay during

28:27

the monsoon season. Uh So

28:29

some of that other equipment that the ARGO floats.

28:31

Now, ARGO floats are deployed

28:34

all around the world, not just off the coast of

28:36

India. In fact, there are more than three thousand

28:39

of them floating in the oceans, and

28:41

they measure temperature, ocean

28:43

velocity, so the actual velocity

28:46

of the water, the salinity

28:48

of the upper two thousand meters of

28:50

the ocean. Scientists primarily

28:52

use ARGO to monitor climate change,

28:55

so they're doing it to see how conditions

28:57

are changing over time to get a better idea of

28:59

what is the x will practical effect of

29:02

climate change. The data data

29:04

gathered by ARGO is publicly available within

29:06

a few hours of its collection, so

29:08

um, the scientists on

29:10

this project are going to rely on obviously on the ones

29:13

that are specifically off the coast of India.

29:16

Then you've got those underwater robots, they're called sea

29:18

gliders. They look kind

29:20

of like um, almost like a

29:22

torpedo shape. Some sometimes

29:25

they're referred to as like an robotic

29:27

dolphin, which is odd because they don't

29:29

really have like they're not jointed where you

29:31

have a t now they've got

29:34

they've got a pair of wings that can tilt.

29:36

But they use changes in buoyancy

29:39

and those wings to create forward

29:41

momentum so they can move through the water. And

29:43

they have a battery inside of them that can

29:45

actually shift around as

29:48

ballast, and that will allow them to change their

29:50

pitch and roll so they can dive

29:52

down. They can they can move through the water. They

29:54

do so very slowly compared to say

29:56

a propeller, But unlike

29:59

a propeller, it's in incredibly energy

30:01

efficient. Yeah, it doesn't have to use a

30:03

lot of energy to change. Uh, it's it's

30:05

position because of the buoyancy and use

30:07

of its own battery is ballast. So

30:10

therefore, if it's energy efficient, that means that it can

30:12

travel quite a great distance, probably on a single

30:14

charge, without having to go back to home base and

30:16

h and be juice up again. Exactly, it can

30:18

stay under water for a long time and can travel a

30:21

great distance. Really essentially

30:23

only has to surface if you do have to

30:26

recharge it or for it to beam

30:28

the data back. It's got a radio antenna at the

30:31

tip of it that will poke

30:33

out the water beams that information

30:35

and the team can gather it. Uh.

30:37

It's really a neat looking device

30:40

and there are videos online that you can watch

30:42

of it in action. Um. They're

30:44

they're a little expensive there, about a hundred fifty pounds

30:47

sterling each. Uh.

30:50

The University of East Anglia used

30:52

to have six of them and then lost two

30:54

of them. One of them got run over by

30:56

a boat. What a ship

30:59

really well, because these things tend to

31:01

stay fairly close to the surface

31:03

in order to beam information back and

31:06

one and they don't move very quickly and

31:08

they're hard to see. They're not huge right

31:10

there, about the size of a person, but

31:13

if you're operating a large ship

31:15

like a cargo vessel, you

31:17

may not see it. And a cargo vessel collided

31:20

with one and destroyed it. The second one was lost

31:22

in Arctic ice. I believe so.

31:26

But there are actually seven of

31:28

them in operation for the Bubble project.

31:31

UM, so really interesting.

31:34

They also can hold lots of different types of sensors,

31:36

not just ones to measure

31:38

the various factors

31:41

in the ocean, but others as

31:43

well, for for things like marine biology. Now,

31:45

of course, in the case of bubble marine biology

31:47

was not really one of the things they

31:49

were necessarily concerned with. So that's not the

31:52

that's not in the instrumentation

31:55

um for those particular seed gliders.

31:57

Instead, they're looking at sensors that are going

31:59

to measure stuff like the turbidity

32:02

of the water, the temperature, salinity,

32:04

and the oxygen content. Now

32:08

you collect all this data with the floats, the

32:10

robots, the satellites, the aircraft,

32:13

and now you know everything. Now you gotta do stuff

32:15

with it. That's the problem is like

32:17

like for one thing, like you

32:19

know, just just that information

32:21

alone is incredibly valuable, but without

32:23

knowing how it all interacts

32:26

with one another, which factors are

32:28

more important, which ones are really

32:30

impacting the monsoon season the most,

32:33

which are causitive versus

32:35

just correlative, Right, Like, there

32:37

may be some things that change. Maybe

32:39

they're changed because the monsoons are moving

32:42

through, not because they change

32:44

and then cause the monsoon. Right, So

32:46

you've got you've got to determine all this. You have to

32:48

crunch all that information, and

32:51

that's gonna be the next big challenges

32:54

grabbing all that data and doing something

32:56

useful with it so that then you

32:58

can take that knowledge and communicate

33:01

it to people, so that you can make actual,

33:04

uh, real world actions based

33:07

upon that data. And

33:09

this is where we start to shift over to a very

33:11

important tool in weather forecasting

33:14

and weather modeling and climate

33:16

science supercomputers.

33:19

Yeah, because if you haven't cotton on yet,

33:22

the problem of weather is is a

33:24

big data problem. Yes, it's a

33:26

it's a huge data problem because

33:28

we know lots of different variables

33:31

affect weather. We know those variables

33:34

change greatly over spans of

33:36

time. Right, So you've got a

33:38

lot of information and that information is

33:40

constantly in flux. So

33:43

how do you process that in a reasonable

33:46

way. Supercomputers have proven

33:48

to be a really important element

33:51

of this analysis. So part

33:54

of understanding this is knowing what a super

33:56

supercomputer really is. It's not just

33:59

a really beefed up PC.

34:01

Right, It's not a beefed up Mac.

34:04

It's not a beefed up Mac either um

34:07

also known as big Mac.

34:11

Yeah, it's none of those things.

34:14

Although I mean, Mr

34:17

Hodgeman, if you're listening, you don't need to beef up. We

34:19

like you the way you are. So supercomputers

34:22

tend to be organized in a way

34:24

where you've got nodes, which

34:26

is essentially either a CPU or a GPU

34:29

um and those are organized into blades.

34:32

Those blades are further organized into racks,

34:34

which are cooled in some interesting

34:36

way, usually water cooled. Because

34:39

you get that many processors in a place

34:42

together, they generate a lot of heat. Heat

34:44

and electronics over the long term are

34:46

not good friends with one another. So

34:49

the in effect is you've got a supercomputer that acts

34:52

kind of like a multi core processor.

34:54

So if you have a multi core processor, you

34:57

might wonder, well, how does this make my computer faster?

35:00

Well, it works really well for certain types

35:02

of computational problems. Those

35:04

will be problems that could be broken up into smaller

35:07

bits. It works less well for problems

35:09

where you have to solve one problem

35:12

before you can start on the next problem.

35:14

Right, So if you were to have the

35:17

first type where you have a problem, you can split up into

35:19

little bit. So you can think of that as imagine

35:21

you've got uh A, I like

35:23

to use this analogy. You've got a math class, and

35:26

in that math class is a math genius,

35:28

and then you've got a bunch of decent

35:31

math students, but they're not of genius

35:33

level. You've got a math problem

35:35

that's that first type one that could be broken

35:37

up into several smaller problems, and

35:40

you give the math genius the full thing, and

35:42

you give each of the math the

35:44

good math students part of that problem.

35:47

The group of good math students are

35:49

more likely going to finish it before

35:51

the math genius, even though the math genius has

35:54

a grasp of mathematics that far

35:57

outpaces that the rest of the class. If

36:00

the second type of problem, like you were talking Joe, then

36:02

the math genius is more likely to finish it

36:04

because you can't divide that problem

36:07

up and and give each little piece to all

36:09

the different math students. So the math students

36:11

represent that multi core processor,

36:13

right with a supercomputer.

36:16

You've just got thousands of these

36:18

processors, like more than eighty

36:20

thousand for some big supercomputers.

36:23

Right, And so you

36:26

take this problem, the problem being, here

36:28

are all these variables in weather,

36:31

and I want My solution is I want to create

36:33

a weather simulation so that I

36:35

can forecast what will happen

36:37

in the future based upon the

36:39

current situation. Now, so

36:42

that's your first step. You create your model,

36:44

then you look and see if your model is any good. One

36:48

way you can do this actually is

36:50

to feed in data from the past. So

36:52

let's say that you have collected a huge amount

36:54

of information from two weeks ago.

36:57

Well, you already know what happened after that

36:59

because it's in the past. Sure, so you

37:01

can you can feed all of the information from

37:04

two weeks in the past into the computer and say,

37:07

h if I modeled this a certain way,

37:09

then do I get like, like, how close

37:11

do I get to what actually occurred after

37:14

that first week? Right? And if and if it turns

37:16

out that it didn't come very close, you

37:18

start making adjustments. You start saying, all

37:20

right, this one factor that I thought was

37:22

really important turns out maybe it's not so

37:24

important. And this other thing that I kind of overlooked

37:27

turns out as much more instrumental than I had anticipated.

37:30

And and this is a long process,

37:32

but you you refine that simulation.

37:35

This is a cool way in which weather prediction,

37:37

I think has the potential to be a

37:40

constantly improving science because unlike

37:42

some disciplines, uh, this is

37:44

not a field in which testing the predictive

37:47

power of your theory or in this case, your algorithm

37:49

is difficult because compared

37:52

to something like psychology, where the results

37:54

of your experiment might often be very fuzzy

37:56

and indeterminate, or like particle

37:59

physics, where you might have to test the predictions

38:01

of your theory by building

38:04

some giant experimental instrument that operates

38:06

at the giga electron volt scale or something

38:08

like that, the weather is not like that. We

38:11

have tons of data on it, always

38:13

new data coming in. We've got plenty already,

38:15

and we have lots of good ways of measuring it already.

38:17

Yeah, And the problem

38:20

is really that we have a wealth. We have to we have we

38:22

are we are befuddled by our wealth of information,

38:24

right, Yeah, No, I just like I just like the like we have

38:26

plenty of it already. Like I was just thinking, like, well,

38:29

not much weather today. I

38:31

had a lot of weather yesterday,

38:34

which uh oh there

38:36

it's from the Mystery Science Theater episode

38:38

of pod People Were the Best.

38:41

One of the characters asks, uh,

38:43

do you think the weather all hold in? One of the

38:45

viewers comments, no, I think it's just gonna

38:47

stop. That was Tom Servo who said

38:50

that. I remember that. Yeah, No, that's a

38:52

fantastic episode. So Tangent

38:55

go watch that episode of MST three K. It's one of the

38:57

best ones they ever did. Back to back to

38:59

the their forecasting. So, according

39:01

to Science Daily, supercomputers

39:04

spend about equal amount of time running their

39:06

simulations to

39:08

assimilating new real world data into

39:11

the models. So, in other words, half

39:13

the time you're simulating whether the other

39:16

halftime you're adjusting that simulation

39:18

so that it more accurately reflects

39:20

the real world. And as we get a better

39:22

understanding of the things that affect

39:25

whether, we can refine that um.

39:27

A study in Japan ran

39:29

a global atmosphere simulation

39:32

and found that a weather event event

39:34

in one part of the world can affect other

39:36

weather events thousands of kilometers away.

39:39

And so it starts to dawn

39:41

on you that in order for you to accurately

39:44

forecast a local weather

39:48

system, you have to actually look well

39:50

beyond the immediate region,

39:52

because there are factors that will affect

39:55

that local weather system that are happening

39:58

really far away. And it may be that it's it's

40:00

something that's not i mean, gonna

40:02

instantaneously affect your local weather,

40:04

but it will have an impact. So maybe something

40:07

that would have normally been a

40:09

rainstorm, but that's it could

40:12

potentially turn into something much more severe like

40:14

tornadoes. So it

40:17

was really interesting. And the study included ten thousand,

40:19

two hundred forty simulations, and

40:21

they divided the global model into twelve

40:24

kilometer sectors, so like a grid

40:27

of a hundred twelve kilometers. Now that's

40:29

also important because the smaller

40:31

those squares are in the grid, the

40:34

more data you're feeding into the

40:36

simulation, and the more powerful

40:38

the supercomputer has to be. Yeah,

40:40

and of course we're always expanding our hardware

40:43

and software capability. So in January,

40:47

the n o a A announced a major upgrade

40:49

and its Weather and Climate Operational

40:51

supercomputer system. Uh,

40:54

and this was interesting. The two computers

40:56

they have are called Luna and Surge

41:00

Urge not like the soda, like a

41:02

wave. Well, yeah, like the soda.

41:06

Sorry. The Luna and

41:09

Surge are based in Florida and Virginia

41:11

and each one runs at two point

41:13

eight nine pedal flops for a combined

41:16

five point seven eight pedal flops of

41:18

computing capacity. And that is up from

41:20

the system's capacity of just seven seventy

41:22

six terra flops. Nothing to sniff at, but

41:25

significantly lower last year. Flops,

41:28

by the way, stands for floating floating

41:30

point operations per second exactly. So

41:33

in the press release, the you

41:35

know, administrator Dr Catherine Sullivan

41:38

said that this upgrade would help the organization

41:40

deal with quote the tidal wave

41:42

of data that new observing platforms

41:45

will generate. Just once again, I think we've

41:47

sort of said this before, but uh, indicating

41:49

that the problem in weather prediction these

41:52

days is not a data problem, but

41:54

it's an analysis problem. It's the

41:56

what we do with the data that's where the

41:58

bottleneck is. Right.

42:01

So also from n

42:03

o A A NOAH, the National Oceanic

42:05

and Atmosphere Administration. In other words, uh,

42:07

they're running fifteen hour forecasts using

42:10

something called the high resolution Rapid Refresh

42:12

model, also known as the H triple

42:14

R in meteorological circles.

42:17

So if you have a meteorologist in your family, just

42:19

ask them how the H triple R is going her

42:23

her or if you want to put model

42:25

in there, it's the HERB. Anyway,

42:28

the model divides the map, the global

42:30

map up into three kilometer sections.

42:33

So you remember I was talking about the Japanese

42:35

study that was a hunter and twelve kilometers,

42:37

So this one's more precise. It's divided

42:39

the the entire world into smaller

42:41

sections, which increases

42:44

the amount of data significantly that

42:46

they have to handle in order to make this fifteen hour

42:48

forecast. That's also why it's

42:50

only fifteen hours out, because

42:53

to to extend the forecast

42:55

further would require even greater processing

42:58

challenges, which they're working to overcome

43:00

and slowly push that number further

43:03

and further out. Um.

43:05

But it's really interesting that they

43:07

are looking at the world in three kilometer sections.

43:10

It blows my mind because you think how

43:12

huge uh an amount of data

43:15

that must be that they're dealing with consistently,

43:18

and they're refreshing this hour by hour to look

43:20

another fifteen hours ahead. UM.

43:23

So, in Europe weather satellites are

43:25

actually more advanced than the ones that we're using

43:27

here in the United States right now,

43:30

but that will change. The

43:32

US has plans to launch the Geo Stationary

43:35

Operational Environmental Satellite are

43:37

also known as GOES ER. Did

43:41

they come in the form of a giant slore as

43:44

as GOES are the destructor or? Um?

43:47

Yeah, I'm having Ghostbusters flashbacks on that.

43:49

But it's scheduled to launch in the fall of this year.

43:52

It will actually become the most

43:54

advanced meteorological satellite

43:57

in orbit for at least

43:59

a short time, finally outpacing the ones

44:01

that are are currently over Japan

44:03

and Europe. Other other recent news

44:06

involved IBM

44:08

spent about two billion dollars acquiring

44:11

basically everything in the Weather Company

44:13

except for the Weather Channel itself. And

44:16

uh, and so they're apparently gonna pitt

44:18

Watson against all that data and just kind of see

44:20

what they can do. Interesting Watson takes it

44:23

down? What Watson Watson

44:25

will take all that data and make yet another

44:28

bizarre and unimaginable

44:31

recipe that involves pot stickers

44:33

that don't have any of the ingredients in them

44:35

that they claimed that is

44:38

it going to rain next year? First? Grill you r

44:40

let us? Oh

44:42

man, I still think we have to each

44:44

take one of those recipes, make it and bring it in. We never

44:47

did do that. We

44:49

should do a live show where we subject each other

44:51

to cooking grill your perade

44:54

olives. I think we all we all will

44:56

need to have a chef hats and

44:59

and aprons with humorous

45:01

sayings on them. Uh,

45:03

that's that's what I suggest. All right, Well, well well we'll

45:05

work on that at any rate. Let's look at again

45:08

kind of further off, like what was the future going

45:10

to bring. So once we have these more advanced

45:12

satellites, we're constantly working

45:14

on building better supercomputers,

45:17

which often are used for this kind of thing,

45:19

as well as other branches of science as

45:21

well. Uh So, for

45:23

one thing, as we get this greater understanding of the global

45:26

influences of whether we can we

45:28

can improve our forecasting when we understand

45:31

that an event happening thousands of

45:33

miles away will have an impact

45:35

on the weather in our area and

45:38

we have a better way of of predicting

45:40

what that impact will be. That's

45:42

gonna benefit people in ways that we can't even

45:44

really get a grip on right now. Um.

45:48

One of the other things we have to remember is that it's

45:50

a lot easier to predict weather

45:52

in general, that is severe weather.

45:55

Um. So you'll see this

45:58

on lots of different sites that are talking

46:00

about meteorology. They'll

46:02

say like, oh, you know, we can predict general

46:04

weather systems out maybe as far as

46:07

a couple of weeks or further. But when you

46:09

start getting into the

46:11

the the possibility of

46:13

severe weather, it's closer to like

46:15

five days, and each day out

46:18

is less accurate than the day before, which

46:21

means that when you're looking at the tail end

46:23

of that forecast, you have to keep that in mind.

46:26

Um. I tried to do that all the time when I'm thinking, like,

46:28

oh, I'm going on vacation in two weeks, let me see

46:30

what the weather is gonna be like in ten days, And

46:32

and often I go

46:35

in with a false sense of security, or

46:37

I'm end up preparing for

46:39

a rainstorm that had just doesn't happen.

46:42

But as we get more information,

46:46

we get better at anticipating

46:48

these things and predicting them accurately. Obviously,

46:51

this could help lots

46:53

in lots of ways, like in that

46:56

commerce that we were talking about, or in travel.

46:58

Absolutely having better weather prediction

47:01

could have all kinds of commercial and

47:03

environmental bonuses, like imagine

47:05

being able to reboot flights

47:07

around bad weather systems before storms

47:10

hit, thus preventing having to sit

47:12

around at the airport all day, or or having to

47:14

have your flight canceled, or even

47:16

allowing pilots to save on fuel by

47:18

plotting better courses. Also,

47:21

as as Julie brought up, in our prior weather episodes,

47:24

changes in whether change our buying habits,

47:26

supermarkets could plan to stock up on

47:28

those frier chickens or whatever it is, way

47:31

more in advance. Apparently, apparently

47:33

during certain disasters, fried

47:36

chicken just flies off the shelves unless

47:38

which is weird because chicken rarely flies even

47:40

when it's not fried. But also

47:43

there's the issue here in Atlanta.

47:45

I made the joke in our notes that it's

47:47

not really a joke. It's actually just a fact that

47:50

if there's even the hint of snow, you

47:52

can expect a run on supermarkets for all the

47:54

milk, bread, sometimes bleach

47:57

bleaches, big yeah, and then people

47:59

get home toilet paper. What do you do with

48:01

this? Yeah, I never buy

48:04

this to begin with, exactly

48:07

lots of French toast. That's what we're

48:09

gonna be having kids. So yeah,

48:11

But they having those predicting

48:14

those better forecast means that you know, you

48:16

can actually prepare for that sort of stuff and

48:18

uh and hopefully not encounter things like

48:21

shortages or or

48:23

or you know, where people go to a

48:26

store and then they realize that they're out of luck

48:28

because everybody has rushed it. If

48:30

you've got more time to prepare for that, then you can

48:32

build up your inventory and make

48:34

better profit and people can be happy that they

48:37

can you know, get their bread and milk and eggs and make that French

48:39

toast and then when it doesn't snow, everyone complains

48:42

about it the bread and milk and eggs go bad,

48:44

but you don't care. You sold them already.

48:46

Yeah, yeah, capitalism. So,

48:49

uh, it was fun to kind of look into

48:51

this. I always, I always really enjoy discussing,

48:54

uh, the idea behind weather science. I'm

48:57

not big on talking about the weather in general, but

48:59

whether science to me, is really neat because you

49:01

start to realize how incredibly

49:04

complicated it is and how much energy

49:07

are the the energy that are that happens

49:09

to be in these big weather systems like you

49:11

know, we we if you talk about hurricanes,

49:14

the amount of energy and a hurricane is phenomenal.

49:17

Right, as Lauren has so succinctly

49:20

put it before, there's more wind than truck

49:22

fair enough. So to me, that's

49:24

why I love talking about these things and why I felt

49:27

that it was fun to to come back and revisit

49:29

this. Plus I wasn't in the last couple, so I really

49:32

wanted to kind of jump into it.

49:34

But guys, if you have any suggestions for future episodes

49:37

of our podcast, let us know,

49:39

send us an email. That address is FW

49:41

Thinking at how Stuff Works dot com,

49:44

or you can drop us a line on Twitter. The handle there

49:46

is f W Thinking, or search f W Thinking

49:49

and Facebook. Our profile should pop right

49:51

up. You can leave us a message there and we

49:53

look forward to hearing from you, and we'll talk to you again

49:56

really soon. For

50:02

more on this topic and the future of technology,

50:05

visit forward Thinking dot Com

50:08

problem brought

50:17

to you by Toyota. Let's Go Places,