#rC3 - Angry weather ?

rC3 preroll music

Herald: It is with much pleasure that I
can now introduce our next speaker, so

it's just started raining outside, but
this heavy rain is not at all probably the

extreme weather effects that we will hear
about right now. The weather, the talk

that we are being presented next will deal
with extreme weather effects and how they

are linked with climate change and how we
even know about that. Our speaker today

is Fredi Otto. She's associate director of
the Environmental Change Institute of the

University of Oxford, and she's also the
lead author of the upcoming IPCC

assessment report, AR6. And without with
no further ado, I give you the stage

Fredi, please.

Frederike Otto: OK, thank you. Yeah. Hi.
It's stopped raining here in Oxford, just

about, but it's definitely flooded, so
that might actually be something to come

back to and talk about with respect to
climate change. So. Whenever we hear or

whenever today an extreme weather event
happens, we hear about hurricanes,

wildfires, droughts, etc., the question
that is immediately asked is, was this,

what is the role of climate change? And to
answer that, for quite a long time,

scientists gave an answer that we cannot
attribute individual weather events to

climate change. But… Sorry, OK. But this…
Because the first, the one answer that

people were giving were that, well, you
can't attribute individual weather events

or they were saying in a world where
climate change happens, of course, every

extreme weather event is somewhat affected
by climate change. And the latter is

attributed too, but that does not
obviously provide much information,

because it doesn't say anything about
whether the event was made more likely or

less likely or what the role of climate
change was. And the first answer that you

can't attribute individual events is not
true any longer. And this is... why that has

changed and how that has changed. And what
we can say is what the content of this

talk will be. So ultimately, every weather
event, extreme or not, is if you

absolutely boil down to it is unique and
they all have many different causes. So

there is always the role of just the
natural chaotic variability of the climate

system and climate and weather system that
plays a role. There's always a causal

factor in where the event
happens, whether it's over land, over a

desert, over a city or a forest, but also
man-made climate change can have an

influence on the likelihood and intensity
of extreme weather events to occur. And so

what we can say now, and what we mean when
we talk about attribution of extreme

weather events to climate change is how
the magnitude and likelihood of an event

to occur has changed because of man-made
climate change. And in order to do that,

we first of all need to know, what is
possible weather in the world we live in

today? So say we have a flooding event in
Oxford and the question is, was this

climate change or not? So the first
question is we need to find out what type

or what kind of event is the heavy
rainfall event that leads to the flooding.

So is it a 1 in 10 year event? Is it a 1
in 100 year event? And in order to do

that, you can't just look at the observed
weather records because that will tell you

what the actual weather that occurred is.
But it doesn't tell you what the possible

weather under the same current climate
conditions are. And so we need to find out

what is possible weather. And to do that,
we use different climate models. So we

simulate under the same climate conditions
that we have today, possible rainfall

events in December in Oxford. And we might
find out that the event that we have

observed today is a one in 10 year event.
And so if you do this, look at all the

possible weather events, you get a
distribution of possible weather under

certain conditions, which is shown in the
schematic on the slide here in the red

curve. And then you know that when it
rains above, say, 30 millimeters a day in

Oxford, then you have a real problem with
flooding. So you define that this is your

threshold from when you speak about an
extreme event. And so you have a

probability of this event to occur in the
world we live in today. Of course, that

does not tell you the role of climate
change, because in order to know that, you

would also you will also need to know what
would the likelihood of this event to

occur have been without man-made climate
change, and so. But because we know very

well how many greenhouse gases have been
introduced into the atmosphere since the

beginning of the industrial revolution, we
can actually remove these additional

greenhouse gases from the climate models
atmospheres that we use and simulate a

world that would have been exactly as it
is today, but without the greenhouse gases

from the burning of fossil fuels. And in
that world, we can then also ask the

question, what are possible heavy rainfall
events in December in Oxford? And we might

find that the event that we are interested
in is in that world, not a one in 10 year

event, but a one in 20 year event. And
because everything else is held the same,

we can then attribute the difference
between these two likelihoods of

occurrence of the extreme event in
question to man-made climate change. And

so with this fake example that I've just
used, we would then say climate change has

doubled the likelihood of the event to
occur because one that was one in 20 year

event is now one in 10 years. So that is
basically the whole theoretical idea

behind attributing extreme events and this
method can be used. And so, for example,

with our initiative that's called World
Weather Attribution, we have looked this

year at the extreme heat in Siberia, the
beginning of this year that, among other

things, led to temperatures above 38
degrees in the city of Verkhoyansk, but

also let to permafrost thawing and large
wildfires. And that event was made so much

more likely because of climate change that
it's almost would have been impossible

without climate change. So when we did the
experiments that the models it's a one in

80 million year event in a world without
climate change. And it's still a

relatively extreme event in today's world,
but it is possible. So this is the type of

event where climate change really is a
game changer. Another event that we have

looked at is Hurricane Harvey that hit the
Houston and Texas in 2017 and caused huge

amounts of damage with the rainfall
amounts it brought. And several attribution

studies doing exactly what I've just
described found that this type of, so this

extreme rainfall associated with a
hurricane like Harvey has been made three

times more likely because of climate
change. And colleagues of mine, Dave Frame

and his team, have then used these studies
to figure out how much of the economic

costs this hurricane can be attributed to
climate change, and found that of the 90

billion US dollars that were associated,
that were associated with the flood damage

from Harvey, 67 billion can be attributed
to climate change, which is in particular

interesting when you compare that to the
state of the art economic cost estimations

of climate change in general, which had
estimated only 20 billion US dollars for

2017 in the US from climate change. And of
course, not every year is an event like

Harvey, but it shows that when you look at
the impact of climate change in a more

bottom up approach, so looking at the
extreme events, which are how climate

change manifests and affect people, you get
very different numbers, as if you just

look at large scale changes in temperature
and precipitation. But of course, not

every extreme event that occurs today has
been made worse because of climate change.

So this is an example of a drought in
southeast Brazil that happened in 2014,

2015, where we found that Climate change
did not change the likelihood of this

drought to occur, so it was a one in 10
year event in 2014, 2015, and also without

climate change, it has a very similar
likelihood of occurrence. However, what we

did find when we looked at, OK, what else
has changed? Why has this drought that has

occurred in a very similar way earlier in
the 2000s and also in the 1970s with much

less impacts. We looked at other factors
and found that the population has

increased a lot over the last or over the
beginning of the 21st century, but in

particular, the water consumption in in
the area and the water usage has increased

almost exponentially. And that explains
why the impacts were so large. So this is

what I've just said is sort of basically
the the very basic idea and and how in

theory these studies work and how and some
results that we find. In practice, it is

usually not quite as straightforward,
because while the idea is still the

same, we need to use climate models and
statistical models for observational data

to simulate possible weather in the world
we live in and possible weather in the

world that might have been. That is, in
theory, straight forward, in practice,

it's often relatively difficult, and what
you see here is how the results of these

studies look when you don't use schematic
and if you're not a hydrologist, this

might be a bit of an unfriendly plot. But
it's it's basically the same as the

schematic that I've showed at the
beginning, but just plotted in a way that

you can see the tails of the distribution
particularly well, so where the extreme

events are. So on the X-axis, we have the
return time of the event in years on a

logarithmic scale and on the Y-axis, you
see the magnitude of the event and that

defines what our extreme event is. And
this is actually a real example from heavy

rainfall in the south of the U.K. And you
can see here in red, each of these red

dots that that you see on the red curve is
a simulation of one possible rainfall

event in the South of the U.K. in the year
2015 in the world we live in today with

climate change and the dashed line
indicates the threshold that led to to

flooding in in that year. And on the
X-axis, when you go down from the dashed

line, you can then see that this is
roughly a one in 20 year event in the

world we live in today. And all the blue
dots on the blue curve are simulations of

possible heavy rainfall in the South of
the U.K. in 2015, in a world without man-

made climate change. And you can see that
these two curves are different and

significantly different, but they are
still relatively close together. And so

the event in the world without climate
change would have been a bit less likely,

so we have roughly a 40 percent increase
in the likelihood. But still other factors

like, yeah, just the chaotic variability
of the weather and also, of course, than

other factors on the ground where houses
build in floodplains and so on play an

important role. So this is the
actual attribution step. So when we find

out what the role of climate change is,
but of course, in order to do that, there

are a few steps before that are crucially
important and absolutely determine the

outcome. And the first step, the first
thing to find out is what has actually

happened, because usually when we read
about extreme weather events or when we

hear about extreme weather events, you see
pictures in newspapers of flooded parts of

the world. And so you don't usually have
observed weather recordings reported in

the media. And the same actually is
true for us. So when we are, so we work a

lot with the Red Cross and they ask us:
OK, we have this large flooding event, can

you do an attribution study? Can you tell
us what the role of climate change is?

Then we also just know: OK, there is
flooding. And so the first step is we need

to find out what is the weather event that
actually caused that flooding. And that is

not always that straightforward. And this
is what you see here on this map, on this

slide is a relatively stark example, but
not an untypical. So it's of an extreme

rainfall event on the 10th of November
2018 in Kenya. And on the left hand side

is one data product of observational data,
of observational rainfall data that is

available and on the right hand side is
another showing the same event. And the

scale which I failed to to say on the
slide in millimeters per day. And so on

the left hand side, you have extreme
rainfall of above 50 millimeters per day,

which is considering that, for example, in
in my home town of Kiel in Schleswig-

Holstein, there is about 700 millimeters
of rainfall per year. You can see that 50

millimeters in a single day is very heavy
rainfall, whereas in the other data

product, you don't see as much rain. You
still see large rain, but it's not in

the same magnitude, and it's also not
exactly in the same place. And so given

that most countries in the world do not
have an open data policy, so you can't

actually get access to the observed
station data, but you have to use

available, publicly available products
like the two have shown here, you have to

know and you have to work with experts in
the region to actually know who hopefully

has access to the data to actually find
out what has happened in the first place.

But of course, if you don't know that or
there is not always a perfect answer, then

if you don't know what event that is. It's
very difficult to do an attribution study.

Assuming though you have found a data
product that you trust, the next question

then is what is actually the right
threshold to use for the event? So if you

have flooding that was pretty obviously
caused by one day extreme rainfall event,

then that would be your definition of the
event. But it could also be that the

flooding has been caused by a very soggy,
rainy season. So actually, the really the

real event you would want to look at is
over a much longer time scale or if the

flooding occurred mainly because of some
water management in the rivers and has

actually flooded further upstream, your
spatial definition of the event would be

very different. And so and what you see
here on this plot is an example of a heat

wave in Europe in 2019. And there, what
usually makes the headlines is the maximum

daily temperature. So if records are
broken, so you could use that as a

definition of the event that you're
interested in. But of course, what really

causes the losses and damages from extreme
events is not necessarily the one day

maximum temperature, but it is when heat
waves last for longer, and especially when

the night temperatures are also high and
not just the daytime temperatures. So you

might want to look at an event over five
day period instead of just the maximum

daily temperatures. Or, and this is sort
of why I have shown the pressure plot on

the right hand side, which is really just
an illustration, it's not terribly

important what's on there. But there are,
of course, different weather systems that

can cause heat waves, especially in the
area here in the south of France. It could

be a relatively short lived high
pressure system bringing hot air from the

Mediterranean. Or it could be something
that is caused from a long time stationary

high pressure system over all of Europe.
If you want to take that into account,

obviously also your event is different.
And there is no right or wrong way to

define the event because there are
legitimate interests in the maximum

daily temperatures, legitimate interest in
just a specific type of pressure system

and interest in what actually causes more
excess mortality on people, what would be

the three day or longer type of heat
waves. But whichever definition you

choose, it will determine the outcome of
the study. And here are some typical

results of attribution studies when you
look at them in a slightly more scientific

way and slightly less just the headline
way as the ones that I've shown earlier.

Because, of course, what also is important
is not only how you define the event,

depending on the impacts and depending on
what you're interested in. The extreme

event and what observational data you have
available. But of course, there's also

then the question of what models, what
climate models do we have available? And

there's always some tradeoff between what
exactly caused the event and what we can

meaningfully simulate in a climate model.
And then all climate models are good for

something and bad for other things. So
there always need to be a model evaluation

stage. So where you test if the models
that you have available are actually able

to simulate in a reliable way the event
that you're interested in. But even if you

have done all this, it can sometimes be
that the models and the observations that

you have show very different things. And
so the heat wave in Germany in 2019, which

was also on the slide before,
is an example of that. When we

look at the long term observations of
extreme, of high temperatures and see how

they have changed over time, we find that,
because of the change in climate, we have

observed, the likelihood of this type of
heat wave has increased more, yeah, about

300 times. So you see this
in the black bar, the black bar in the

middle of the blue bar, on the left hand
side, at the very top where it says DWD

obs, that's the Deutscher Wetterdienst
observations and we see that where this

black bar is about, again, a logarithmic
scale, about 300 hundred times more

likely. But of course, because we have
only 100 years worth of

observations and summer temperatures
are extremely variable, there is a large

uncertainty around this change. And so we
cannot, from the observations alone, we

cannot exclude 100.000 times change in the
likelihood of this heat wave. But

similarly, also not a 20 times heat wave.
But what the main point is, that in all

the climate models and all the red bars
that you see on there are the same

results, but for climate models where we
have compared today's likelihood of the

event to occur with the likelihood in the
world without climate change, and you see

that the change is much lower. And of
course, climate change is not the only

thing that has changed and that has
affected observed temperatures. But other

factors like land use change and things
like that are much smaller in the size

than the climate signal. So they cannot
explain this discrepancy. So this means

that the climate models we have available
for this type of study have obviously a

problem with the extreme temperatures in a
small scale. And there are effects that we

don't yet understand. And so we can't say:
OK, this heat wave was made 10 times more

likely. But we can only say, that with our
current knowledge and understanding, we

can say that climate change was an
absolute game changer for this type of

heat wave, but we can't really quantify
it. On the right hand side is a much nicer

result on the top one, which is for
extreme rainfall, in Texas 2019 and nicer

result I mean now for a scientist and
in a scientific way. So we have in blue

two different types of observations from
the heavy rainfall event, and they both

show pretty much exactly the same result.
And also the two climate models that we

had available that passed the model
evaluation tests show an increase in the

likelihood of this event to occur. That is
very similar to that in the observations

in terms of order of magnitude. And so in
that case, we can just synthesize the

results and give an overarching answer,
which is that the likelihood of this event

to occur has about doubled because of man-
made climate change. And the last example

that I, that is here is for drought
in Somalia in 2010, where not only the

observations are extremely uncertain. So
from the observations, you could say we

could have both an increase in likelihood
or a decrease in likelihood by a factor of

10. But also the climate models show a
very, very mixed picture where you can't

even see a sign that that is conclusive.
So in that case, you can say, we can

exclude that climate change made this
event more or less than 10 times, more

than 10 times or less than 10 times more
likely. But we can't say anything more. So

we can exclude that it's a complete game-
changer like we have for heat waves, for

example. But that's about the only
thing that you can say for a result

like this. So this was sort of the
most detailed scientific stuff that I

would like to show, because I think it's
important to get some background behind

the headline results that would just
be climate change doubled the likelihood

of this event. So there are always four
possible outcomes of an attribution study

a priori. And that is because
climate change affects extreme weather in

two ways basically. One way is what we
would call the thermodynamic way, which

means that because we have more greenhouse
gases in the atmosphere, the atmosphere

overall gets warmer. So you have, on
average, an increase in the likelihood of

heat waves decrease in the likelihood of
cold waves. A warmer atmosphere can also

hold more water vapor that needs
to get out of the atmosphere as rainfall.

So on average, from the warming alone, we
would also have more extreme rainfall. But

then there's the second effect, which I
call the dynamic effect, and that is

because we've changed the composition of the
atmosphere, that affects the atmospheric

circulation. So where weather systems
develop, how they develop and and how they

move. And this effect can either be in the
same direction as the warming effect. So it

can be that we expect more extreme rainfall,
but we also get more low pressure systems

bring rain to get even more extreme
rainfall. But these two effects can also

counteract each other. And so you
can expect more rainfall on

average. But if you don't get the weather
systems that bring rain, you either have

no change in likelihood and intensity or,
if the dynamics win, you have actually

decrease in the likelihood of extreme
rainfall in a particular season or region.

And so this is why a priori, that can
always be four outcomes: It can be that

the event was made more likely. It can be
that it was made less likely. It can be

there's no change. Or it can be that with
our current understanding and tools, we

can't actually answer the question. And so
this has been possible to do now for

about a decade, but only in the last five
years really have many, many people or

many scientists started to do these
studies. And so there is actually a

relatively large, there are
lots of attribution studies on different

kinds of extreme events. And what you can
see on this map here is what the news and

energy outlet CarbonBrief has put all
these studies together. And you can see in

red where climate change played an
important role, and blue where climate

change did not play a role. And in gray,
that was an inconclusive result. This is

very important, though, that this is not
representative of the extreme events that

have happened. This is just represents the
studies that have been done by scientists

and they are, of course biased towards
where scientists live

and also towards extreme events that are
relatively easy to simulate with climate

models. So there are lots of heat waves in
Europe, Australia and North America

because that is where people live. And on
this next map, I have tried to

show the discrepancy between the extreme
events that have happened and those for

which we actually do know the role of
climate change. So here in red are deaths

associated with extreme events since 2003.
So since the first event attribution

study. And it's death from heat waves,
storms, heavy rainfall events and droughts

primarily in different parts of the world,
the bubble is always on the capital of the

country. And the larger the bubble, the
more deaths due to extreme events in those

years. And in black overlaying that are
those deaths for which we know the role of

climate change. So that doesn't mean that
the deaths are attributed to

climate change, but it means that there
we do know whether or not to what

extent climate change played a role. And
you can see that most of the European

countries, the black circle is almost as
large as the red one. So for most of the

extremes or most of the deaths associated
with extreme events, we do know the role

of climate change. But for many
other parts of the world,

there are no or very small black circles.
So for most of the events and the deaths

associated with them, we don't know what
the role of climate change is. And I've

used death here not because I'm
particularly morbid, but because it's

an indicator of the impacts of
extreme weather that is relatively good

comparable between countries. So this
means that with event attribution methods

that we have developed over the last
decade, we now have the tools available to

do, to provide an inventory of the impacts
of climate change on our livelihoods. But

we are very far from having such an
inventory at the moment because most of

the events that have happened, we actually
don't know what the role of climate change

is. And so we don't know in detail on
country scale and on the scale where

people live and make decisions, what the
role of climate change is today. There's

another slightly related issue with that
is, that the extreme events that I've used

to create the map are shown before with
the death of climate change, with the

death of extreme weather events. They are
from a database called EM-DAT, which is a

publicly available database where losses
and damages associated with disasters

technological disasters, but also
disasters associated with weather are

recorded. But, of course, they only can
record losses and damages if these losses

and damages are recorded in the first
place. And so what you see on this map is

in grey and then overlayed with different
with different circles are heat waves that

have occurred, they have occurred between
1986 and 2015 on this map. But you could

draw a map from 1900 to today, and it
would look very similar. And that shows

lots and lots of heat waves reported in
Europe and in the US, India, but there are

no heat waves reported in most of sub-
Saharan Africa. However, when you look at

observations, and also we see that extreme
heat has increased quite dramatically in

most parts of the world and a particular
hotspot is sub-Saharan Africa. So, we know

from when we look at the weather that heat
waves are happening, but it's not

registered and it's not recorded. So we
have no idea how many people are actually

affected by these heat waves. And so we
then, of course, don't do attribution

studies and don't find out what the role
of climate change in these heat waves is.

So in order to really understand the
whole picture, we would also need to start

recording these type of events in other
parts of the world. And so my very last

point, before, I hope that you have
questions for me, is: Of course,

everything I've said so far was talking
about the hazards, so talking about the

weather event and how climate change
affects the hazard. But of course that is

not the same or translates immediately
into losses and damages, because whether

or not a weather event actually has any
impact at all is completely driven by

exposure and vulnerability. So who and
what is in harm's way. And I've already

shown, I've already mentioned the example
early on with the drought in Brazil, where

the huge losses and damages were to a
large degree attributable to the increase

in water consumption. And thus,
therefore, in order to really find out how

climate change is affecting us today, we
not only need to define the extreme events

so that it connects to the impacts, but
also look into vulnerability and exposure:

What is changing, what's there and what
are the important factors. But we can

do that. And so we have really made a lot
of progress in understanding of how

climate change not only affects global
mean temperature, which we have known for

centuries, and how it affects large
scale changes in temperature and

precipitation, which we have also known
for a very long time. But we now have

actually all the puzzle pieces together to
really understand what climate change

means on the scale where people live and
where decisions are made. We just need to

put them together. And one lens or one way
of where they are currently put together

is, for example, in courts. And so because
it's obviously people who experience

losses and damages from climate change.
And so one way to address that is going

through national governments, local
governments, hoping for adaptation

measures to be put in place. But if that's
not forthcoming quickly enough, there is

the option to sue. And so this is one
example which is currently happening in

Germany where a peruvian farmer is suing
RWE to basically pay their share of a

adaptation because of largely increased
flood risk from glacier melt in the area.

And they want RWE to pay from their
contribution to climate change, where

their emissions and then have some funding
for the adaptation measures from them. And

that is one example of where these kind of
attribution studies can be used in a very

direct way to hopefully change
something in the real world. And with

this, I would like to end and yeah, leave
you with some references, and hope you

have some questions for me.

Herald: Sind wir durch? So, ja. Herzlichen
Dank für den Vortrag. Ich hab, bevor wir

zum Q&A kommen muss ich einmal mich im
Namen der Produktion bei den Zuschauern

entschuldigen, ich glaube ihr hattet etwas
Produktionssound auf den Ohren, das sollte

natürlich nicht so sein. Gut, wir haben
jetzt keine Fragen aus dem Chat bisher.

Aber vielleicht eine Frage von mir, das
letzte Beispiel war ja ein Fall

einer Klage über Ländergrenzen hinaus
quasi, ist das ein Ansatz, den man, den

wir in Zukunft öfter sehen würden, das
heißt, dass über Ländergrenzen hinweg

Menschen oder Organisationen sich
gegenseitig versuchen quasi über den

Klageweg auf den richtigen Weg zu bringen.
FO: Also es ist tatsächlich ein, eine

Ausnahme, dass das im Fall RWE und Lliuya
funktioniert, denn das deutsche Recht

sieht vor, dass Firmen, die in Deutschland
ansässig sind auch verschieden

verantworlich sind, die nicht in
Deutschland stattfinden.

H: So sorry to interrupt. I just realized
that we are still in English talk. Sorry

for that.
FO: OK. No worries. So your question was

if we're going to see more
international court cases where across

countries, across nation states we have
climate litigation. And this type of

litigation that I've just shown as
the example is in so far an

exception, as in German law, a company is
also responsible for the damages caused

outside of Germany. Which is not the case,
for example, for companies in the US

or so. So, and this is why Lliuya sued RWE
and not, for example, ExxonMobil. But

these type of cases, where this
Lliuya case is an example. We see a lot of

a lot of them, an increasing number of
them each year. And they are difficult to

do across nations because this, the German
law is exceptional on that case. But there

are other ways, like, for example, why are
human rights courts that can be done

across nation states and that is also
happening. So it's at the moment, it is

still legally not super straightforward to
to actually win these cases, but

increasingly a lot of lawyers working on
that so that we will see a lot of

change in that in the coming years.
H: OK, thank you. In the meantime, there

appeared some questions from the chat and
from the Internet. I will go through them.

First question is: are the results of the
individual attribution studies published

as open data in a machine readable format?
FO: laughter So all the studies that

we do that that I've done with my
team, with world weather attribution. So

there all the data is
available, and it's available on a

platform that's called Climate Explorer.
So that should be machine readable. So and

this is deliberately because yeah, because
we want to make it as transparent as

possible so everyone can go away, use our
data, and redo our studies, and find out

if we made any mistakes. But this is not
the case for all the studies that exist,

because most of them or many of them are
published in peer reviewed journals and

not all peer reviewed journals have
open data and open access policies.

But increasingly, journals have.
So if you, for example, go to the

CarbonBrief website and look at the map of
studies, there you have links to all

the studies. And a lot of them have the
data available.

H: OK, maybe a follow up to this one. The
next question is, are the models somehow

available or usable for a wider interest
public or is APC required? I'm not quite

sure what APC means.
FO: So the model data is publicly

available from–and this is one reason why
we have been able to do these studies

because until relatively recently, model
data was not publicly available and only

scientist working in a specific country
could use the model developed in that

country–but now all the model data is
shared publicly and people can use it. So

it's definitely there and usable. It just
requires some expertise to make sense of

it. But it's, yeah, people can use it.
H: OK, the next question is: to what

certainty can you set up counterfactual
models, which are an important reference

to your percentage value, and what
data are the basis for these models?

FO: So the counterfactual simulations are-
the climate models we use are basically the

same models that are used also for the
weather forecast. They are just run in

lower resolution. So, which I guess most
of this audience knows what that means. So

the data points for the part, so that it's
not so computing intensive. And these

models, they are tested against observed
data. And so that is how we do the model

evaluation. So that is some simulations of
the present day. And for the

counterfactual, we know extremely well how
many greenhouse gases have been included

into the atmosphere since the beginning of
the Industrial Revolution, so that there

is some very large certainty with that
number and we remove that from the models'

atmospheres. So the models have exactly
the same set up, but the lower

greenhouse gases, lower amount of
greenhouse gases in the atmosphere, and

then are spun up and run in exactly the
same way. So, they, but of course, we

can't test the counterfactual. And so that
means that we assume that the sort of the

the weather was still the same, physics
will still hold in the counterfactual. And

that the models that are developed
using present day represent the

counterfactual. Which is, which is an
assumption.

But it is not a completely
unreasonable assumption, because of

course, we have now decades of model
development and have seen that, in fact,

that indeed climate model projections that
have been made 30 years ago have actually

come… come to… have been realized, and
pretty much the same way on a large scale

that they have, as they had been predicted
30 years ago. And so that assumption

is not, yeah, it's not a big assumption.
So the counterfactual itself is not a

problem. But of course, also the present
day model simulations, they are

not… they are very far from perfect. And
there are some types of events which state

of the art climate models just can't
simulate. And so, where we can- what

we can say very little. So well, for
example, for hurricanes, we can say

with high certainty about the
rainfall associated with hurricanes, the

hurricane strength itself and the
frequency of hurricanes is something

which is very difficult to simulate with
state of the art models. So our

uncertainty there is much higher.
H: OK. And then, well, some, one question

that emerges from all of this is,
of course, if we know this much and way

more than in the past, how are
politicians still ignoring that

information? And how can we
convey that into their minds?

FO: Well, if I knew the answer to that, I
would probably not be standing here,

but actually doing politics. But I
think it takes a frustratingly long time

for things to change and things should
change much faster. But we actually- the

last two years have shown huge progress, I
think, in terms of also putting climate

change on the agenda of every politician.
Because, and that's largely due to the

Fridays For Future movement, but also to a
degree, I think, due to the fact that we

now actually know that the weather that
people experience in their backyard–and

pretty much independent of where their
backyard is–is not the same as it used to

be. And so people do experience today
climate change. And I think that

does help to bring a bit more urgency.
And, of course, I would have said everyone

has climate change on their agenda, which
was very different even two years ago,

where there were lots of people who
would never talk about climate change and

their political agendas has played no
role. It doesn't mean that it

has the right priority on that agenda,
but it's still a huge step forward that

has been made. And so I think we do know
some things that do work, but we just have

to just keep doing that. Yeah, I don't
think I can say more. I don't have a magic

wand to change it otherwise.
H: Maybe some other point of impact.

One of the question is, is it possible to
turn the results of attribution studies

into recommendations for farmers and
people who are affected in a financial way

by extreme weather and how to change
agriculture to reduce losses from extreme

weather effects?
FO: Yes, absolutely. So that is

one of the most useful things of these
studies is well, on the one hand, to raise

awareness. But on the other hand, if you
know that a drought that you have

experienced that has led to losses is a
harbinger of what is to come, then that is

incredibly helpful to know how
agricultural practices might need to be

changed. Or that insurance for losses from
agriculture might need to be changed. And

so this is exactly why we do these
attribution studies. Because not

every extreme event has always
shows the fingerprints of

climate change. And if you know
which of the events are the ones where

climate change is a real game changer, you
also do know where to put your efforts and

resources to be more resilient in the
future. And for financial losses, it

is on the one hand, yeah, you can use
these studies to find out what your

physical risks are for your assets. And
how they, and of course, everything that

I've said, comparing the counterfactual
with the present we can do, and we do this

also with the future. So you can also see
how in a two degree world, the events,

the likelihood and intensities are
changing. And of course, you can then

also, in a less direct way, use this kind
of information to see, to assess what

might be other risks from- where might be
stranded assets, what are other risks

for the financial sector,
for the financial planning.

Where could liability risks be and how
could they look like. So there is, because

extreme weather events and their changes
in intensity and magnitude is how climate

change is manifesting, it really connects
all these aspects of where the

impacts of climate change are.
H: OK, last question for today. I hope I

can get that right. I think the question
is if there are study, if there are

studies on how we cultivates fields
and agriculture. How does this impact the

overall climate in that area? The example
here is that only an increase in water

consumption was directed to São Paulo. Or
might there also be a warm world created

by monoculture in central Brazil?
FO: So, yeah, I don't know details, but

there are, but land use changes and land
use does play a role. On the one hand, it

affects the climate. So if you have, if
you have a rainforest, you have a very

different climate in that location as if
there is a savanna or plantation. And

also, of course, if you have monocultures,
you are much more, your losses are

larger usually as if you have different
types of agriculture. Because

in a monoculture everything is in
exactly the same way vulnerable and

so that, yeah. So that does,
land use change plays a hugely important

role with respect to the impacts of
extreme weather. And that is one thing to

look at. When I was saying, talking about
looking at vulnerability and exposure, and

of course also changes in the hazard are
not just because of climate change, but

also because of land use change. And you
can use exactly the same methods, but

instead of changing the CO2 or the
greenhouse gases in the atmosphere of your

model, you can change the land use and
then disentangle these different drivers

in and hazards.
H: OK, Fredi Otto thank you very much for

your presentation and for the Q&A. It was
a pleasure to have you with us. And yeah,

if you have any questions, any more
questions, I guess there are ways to

contact you.
FO: laughter

H: I think your email address and contact
details are in the Fahrplan for all the

viewers that have way more questions. And,
I don't know, do you have access to the 2D

world and do you explore that?
FO: Given that I don't know what you mean,

probably not, but…
laughter

H: OK.
FO: That can also be changed.

H: Yeah, it's the the replacement for
the congress place itself. But anyway,

if you viewers and you people out there
have any more questions, contact Fredi

Otto. And thank you again very much for
your talk. And, yeah. Have a

nice congress, all of you.

rc3 postroll music

Subtitles created by c3subtitles.de
in the year 2020. Join, and help us!