1
00:00:00,600 --> 00:00:01,701
[light instrumental music]

2
00:00:01,735 --> 00:00:03,536
- [Narrator] NASA'S Jet
Propulsion Laboratory presents

3
00:00:04,403 --> 00:00:06,005
the von Karman Lecture,

4
00:00:06,038 --> 00:00:08,808
a series of talks by
scientists and engineers

5
00:00:08,841 --> 00:00:12,378
who are exploring our
planet, our solar system,

6
00:00:12,411 --> 00:00:19,986
and all that lies beyond.

7
00:00:24,824 --> 00:00:26,526
- Good evening
ladies and gentlemen.

8
00:00:26,559 --> 00:00:28,594
How's everyone tonight?
- Good.

9
00:00:28,627 --> 00:00:30,063
- Good, well thanks
again as always

10
00:00:30,096 --> 00:00:32,765
for coming out to
join us this evening.

11
00:00:32,798 --> 00:00:34,500
One of the greatest
uncertainties

12
00:00:34,533 --> 00:00:36,669
in projections of
future climate change

13
00:00:36,702 --> 00:00:39,572
is how terrestrial
ecosystems contribute to

14
00:00:39,605 --> 00:00:43,576
or help counteract the rise
in atmospheric carbon dioxide.

15
00:00:44,443 --> 00:00:46,245
This is because these
systems can both

16
00:00:46,278 --> 00:00:48,514
absorb carbon and emit it.

17
00:00:48,547 --> 00:00:51,417
Here at JPL we are using
satellite remote sensing

18
00:00:51,450 --> 00:00:53,719
and sophisticated
modeling to understand

19
00:00:53,752 --> 00:00:57,290
how Earth's carbon, water,
and nutrient cycles are linked

20
00:00:57,323 --> 00:01:00,593
and how they impact the
Earth's system as a whole.

21
00:01:00,626 --> 00:01:02,462
Tonight our guest
will give an overview

22
00:01:02,495 --> 00:01:05,631
of the latest data sets and
model developments from JPL,

23
00:01:05,664 --> 00:01:08,768
and discuss new insights into
the behavior and understanding

24
00:01:08,801 --> 00:01:12,138
of terrestrial ecosystems
in a changing climate.

25
00:01:12,171 --> 00:01:14,907
Our guest tonight is a
research scientist here at JPL,

26
00:01:14,940 --> 00:01:17,610
and the science lead of a new
instrument about to launch

27
00:01:17,643 --> 00:01:20,880
to the International Space
Station called ECOSTRESS.

28
00:01:20,913 --> 00:01:22,582
He is originally from LA,

29
00:01:22,615 --> 00:01:24,383
got his undergraduate
and graduate degrees

30
00:01:24,416 --> 00:01:25,751
from UC Berkeley,

31
00:01:25,784 --> 00:01:28,921
and then did a postdoc
at Oxford University

32
00:01:28,954 --> 00:01:33,759
where he taught for a few years
before joining JPL in 2010.

33
00:01:33,792 --> 00:01:36,395
His research focuses on
terrestrial ecosystems,

34
00:01:36,428 --> 00:01:39,565
water, carbon, and nutrients
using a combination

35
00:01:39,598 --> 00:01:42,034
of supercomputer
models, remote sensing,

36
00:01:42,067 --> 00:01:45,505
and field campaigns from
the Amazon to the Arctic.

37
00:01:45,538 --> 00:01:48,474
When he's not sciencing, he's
juggling his six year old

38
00:01:48,507 --> 00:01:51,177
and three year old kids,
playing basketball,

39
00:01:51,210 --> 00:01:53,613
snowboarding, breakdancing,

40
00:01:53,646 --> 00:01:55,415
and doing acrobatic
yoga [laughing].

41
00:01:56,415 --> 00:01:57,950
Ladies and gentlemen,
please help me welcome

42
00:01:57,983 --> 00:02:00,219
tonight's guest,
Dr. Josh Fisher.

43
00:02:00,252 --> 00:02:03,923
[applauding]

44
00:02:08,794 --> 00:02:10,863
- Hello, I'm Josh,

45
00:02:10,896 --> 00:02:13,966
and I will not be doing
acrobatic yoga today.

46
00:02:13,999 --> 00:02:16,369
That will be another
von Karman Series,

47
00:02:16,402 --> 00:02:18,304
I suppose, they'll
have to start up.

48
00:02:19,338 --> 00:02:22,441
So I'm a scientist here at JPL,

49
00:02:22,474 --> 00:02:26,312
and my focus is on plants,

50
00:02:26,345 --> 00:02:30,617
on vegetation and everything
that impacts plants on land.

51
00:02:32,384 --> 00:02:34,020
So I'm gonna take you
through a little bit

52
00:02:34,053 --> 00:02:38,391
of a whirlwind journey
through the types of questions

53
00:02:38,424 --> 00:02:43,396
and puzzles that we try to
solve as ecosystem scientists.

54
00:02:45,898 --> 00:02:50,469
Our story begins in 1835
when Hans Christian Anderson

55
00:02:50,502 --> 00:02:53,072
wrote a children's story
about a young woman

56
00:02:53,105 --> 00:02:56,776
whose royal identity is
established by her test

57
00:02:56,809 --> 00:02:58,744
of her physical sensitivity.

58
00:02:58,777 --> 00:03:01,614
The test, unbeknownst
to her, was that a pea

59
00:03:01,647 --> 00:03:03,950
was placed in her bed
covered by 20 mattresses

60
00:03:03,983 --> 00:03:05,851
and 20 feather beds.

61
00:03:05,884 --> 00:03:07,853
In the morning
she lamented about

62
00:03:07,886 --> 00:03:09,822
the discomfort from the pea.

63
00:03:09,855 --> 00:03:13,726
Only a real princess would
have such sensitivity,

64
00:03:13,759 --> 00:03:15,595
thus verifying her
claims to royalty.

65
00:03:17,863 --> 00:03:21,434
The Earth's land surface,
the plants on Earth,

66
00:03:21,467 --> 00:03:25,071
represent a very small
portion of the total Earth,

67
00:03:25,104 --> 00:03:29,775
yet exert an enormous influence
on the Earth's climate

68
00:03:29,808 --> 00:03:31,811
and the fate of the Earth.

69
00:03:31,844 --> 00:03:35,248
It's the figurative
pea in Earth's bed,

70
00:03:36,248 --> 00:03:39,652
and what we're trying to
establish is the true future,

71
00:03:39,685 --> 00:03:42,855
what is the future,
the royal identity?

72
00:03:44,356 --> 00:03:45,458
How is this done?

73
00:03:45,491 --> 00:03:47,526
It's done in models.

74
00:03:47,559 --> 00:03:48,560
We start with models.

75
00:03:48,594 --> 00:03:51,998
Models that start with
essentially land or soil,

76
00:03:52,031 --> 00:03:54,900
and you kind of throw a
bunch of weather at it,

77
00:03:54,933 --> 00:03:57,770
some CO2, and up
pops and ecosystem.

78
00:03:57,803 --> 00:04:01,240
If you change the
weather around

79
00:04:01,273 --> 00:04:02,508
or any of these
conditions,

80
00:04:02,541 --> 00:04:04,510
then the ecosystem changes.

81
00:04:06,412 --> 00:04:07,913
A very simple way
of representing

82
00:04:07,946 --> 00:04:09,615
how ecosystems behave.

83
00:04:09,648 --> 00:04:13,152
But these ecosystems and
these plants are made up of

84
00:04:14,253 --> 00:04:17,657
a huge amount of processes
that we're trying to understand

85
00:04:17,690 --> 00:04:20,726
at the biological
to the global scale.

86
00:04:20,759 --> 00:04:23,963
Starting with
microns large pores

87
00:04:23,996 --> 00:04:27,366
we can't see with our eyes
on leaves that when open

88
00:04:27,399 --> 00:04:32,305
allow plants to take up
CO2 and release water.

89
00:04:32,338 --> 00:04:35,474
But when closed, plants
stop taking up CO2.

90
00:04:35,507 --> 00:04:40,379
So these microns large
pores exert influence

91
00:04:40,412 --> 00:04:42,115
on the entire Earth.

92
00:04:43,615 --> 00:04:45,951
So they take up CO2,
they also release water.

93
00:04:45,984 --> 00:04:48,287
It's hard to kind of picture
water coming out of the leaves

94
00:04:48,320 --> 00:04:50,856
so I put this little geyser
coming out of the leaf

95
00:04:50,889 --> 00:04:53,192
so you can imagine water
coming out of leaves.

96
00:04:56,662 --> 00:04:59,966
Within the leaf is the
photosynthetic machinery,

97
00:05:01,834 --> 00:05:04,570
very, very complex
chemical process

98
00:05:04,603 --> 00:05:07,973
that we have to somehow
extend to the global scale,

99
00:05:08,006 --> 00:05:09,909
and we have to make
a lot of assumptions

100
00:05:09,942 --> 00:05:13,379
because we can't model
every single electron

101
00:05:13,412 --> 00:05:15,348
or chemical process
in these leaves.

102
00:05:16,515 --> 00:05:18,518
Not every leaf is
created equally.

103
00:05:19,518 --> 00:05:21,921
Even within a given tree,
there's leaves that are shaded,

104
00:05:21,954 --> 00:05:23,255
there's leaves that are bigger,

105
00:05:23,288 --> 00:05:25,424
there's leaves that are older.

106
00:05:25,457 --> 00:05:28,494
So how do you represent
the diversity of leaves

107
00:05:28,527 --> 00:05:31,364
within a single tree, let
alone across all the trees

108
00:05:31,397 --> 00:05:33,299
and plants of the planet?

109
00:05:35,567 --> 00:05:38,938
We need to understand
how radiation or light

110
00:05:38,971 --> 00:05:42,375
diffuses through canopies
turning on and off

111
00:05:42,408 --> 00:05:44,777
these leaf processes,

112
00:05:44,810 --> 00:05:48,448
and how that light
heats up the planet,

113
00:05:49,348 --> 00:05:52,618
and this is something that we
can see from space as well.

114
00:05:52,651 --> 00:05:54,387
Plants intercept rainfall.

115
00:05:54,420 --> 00:05:56,889
This is water that does
not go to the soil,

116
00:05:56,922 --> 00:05:59,558
and the soil processes are
not turned on with this water.

117
00:05:59,591 --> 00:06:03,462
But a lot of water does
come through the canopy

118
00:06:03,495 --> 00:06:06,165
into the soil, so we
have to understand

119
00:06:06,198 --> 00:06:10,236
and be able to predict
what that difference is.

120
00:06:10,269 --> 00:06:13,038
Plants were actually first
included in climate models

121
00:06:13,071 --> 00:06:14,974
not because of the biology,

122
00:06:15,007 --> 00:06:16,809
but because of the
physical structure.

123
00:06:16,842 --> 00:06:19,845
They got in the way of the
winds of atmospheric models,

124
00:06:19,878 --> 00:06:23,416
so they also have this
not only carbon cycle,

125
00:06:23,449 --> 00:06:25,518
but just physical structure

126
00:06:25,551 --> 00:06:27,387
influence on the
Earth's system.

127
00:06:29,054 --> 00:06:32,858
We need to understand
how leaves drop

128
00:06:32,891 --> 00:06:34,393
in deciduous systems.

129
00:06:34,426 --> 00:06:36,629
Our models shouldn't drop
all the leaves at once,

130
00:06:36,662 --> 00:06:38,697
or keep all the leaves
on at the same time.

131
00:06:38,730 --> 00:06:41,233
So we need to be able to
predict that delicate balance

132
00:06:41,266 --> 00:06:44,537
of leaf phenology.

133
00:06:46,939 --> 00:06:49,308
We know that plants take
up carbon and make sugars

134
00:06:49,341 --> 00:06:51,210
and make wood, but they
also burn that sugar

135
00:06:51,243 --> 00:06:55,314
for metabolism just like we
burn sugar for metabolism.

136
00:06:55,347 --> 00:06:58,317
And we need to understand
how much sugar they store,

137
00:06:58,350 --> 00:07:01,921
how much sugar they burn, and
when they run out of sugar

138
00:07:01,954 --> 00:07:02,955
are they gonna die?

139
00:07:05,424 --> 00:07:08,294
There's a huge diversity of
plants, thousands of species.

140
00:07:08,327 --> 00:07:10,629
We cannot model
every single species,

141
00:07:10,662 --> 00:07:13,132
or even see every
single plant or species,

142
00:07:13,165 --> 00:07:15,568
so we have to make
these assumptions about

143
00:07:15,601 --> 00:07:17,169
groupings of trees.

144
00:07:17,202 --> 00:07:19,805
Sometimes we call them
plan functional types

145
00:07:19,838 --> 00:07:24,310
or other types of
groupings such as grass

146
00:07:24,343 --> 00:07:28,514
or broadleaf trees
or needle leaf trees.

147
00:07:28,547 --> 00:07:31,350
So the amount of plant
types that we define

148
00:07:31,383 --> 00:07:34,987
influences how we
predict the outcomes

149
00:07:35,020 --> 00:07:37,256
to changing climate or CO2.

150
00:07:38,190 --> 00:07:40,159
The more plant functional
types we have in there,

151
00:07:40,192 --> 00:07:42,394
the more computational
demand there is.

152
00:07:42,427 --> 00:07:44,663
So there's this trade
off between realism

153
00:07:44,696 --> 00:07:47,600
and our ability to
conduct these experiments.

154
00:07:49,735 --> 00:07:52,538
Some plants grow
faster than others.

155
00:07:52,571 --> 00:07:57,210
Some plants put their
carbon into wood.

156
00:07:58,310 --> 00:08:00,813
Some plants put their carbon
into roots more than others

157
00:08:00,846 --> 00:08:02,615
and into leaves, so
we need to be able

158
00:08:02,648 --> 00:08:04,116
to understand that balance.

159
00:08:05,684 --> 00:08:10,089
How much nutrients do plants
take up, and how do they do it?

160
00:08:10,122 --> 00:08:11,524
Again, questions we ask.

161
00:08:12,824 --> 00:08:15,961
The layers in the soil
are really important

162
00:08:15,994 --> 00:08:19,131
for understanding the
terrestrial ecosystems

163
00:08:19,164 --> 00:08:20,699
in the Earth's climate.

164
00:08:20,732 --> 00:08:25,671
Soil stores a lot of
carbon and nutrients.

165
00:08:25,704 --> 00:08:28,941
The more soil layers we
include in these models,

166
00:08:28,974 --> 00:08:33,079
again, the more computational
power required, but

167
00:08:34,713 --> 00:08:36,449
the more realism as well.

168
00:08:37,416 --> 00:08:39,818
We need to understand
how leaves decompose,

169
00:08:39,851 --> 00:08:41,487
how fast they decompose.

170
00:08:41,520 --> 00:08:43,656
We don't want our models
just dropping leaves

171
00:08:43,689 --> 00:08:45,057
and instantly decomposing,

172
00:08:45,090 --> 00:08:47,259
or sitting there and piling up,

173
00:08:47,292 --> 00:08:49,361
so again getting
that balance right.

174
00:08:49,394 --> 00:08:51,697
And what about all
the worms and termites

175
00:08:51,730 --> 00:08:55,067
and microbes that eat
these dead leaves and wood,

176
00:08:55,100 --> 00:08:56,969
and then respire back that CO2?

177
00:08:57,970 --> 00:09:00,339
We have to be able to
understand how that's changing

178
00:09:01,273 --> 00:09:04,744
as temperatures and
water cycling changes.

179
00:09:05,611 --> 00:09:07,846
At the end of the day we're
interested not necessarily

180
00:09:07,879 --> 00:09:10,349
in single trees, but
the entire ecosystems.

181
00:09:10,382 --> 00:09:13,652
How all these different
trees and grass and systems

182
00:09:13,685 --> 00:09:16,556
integrate across
the larger picture.

183
00:09:18,290 --> 00:09:21,527
Hydrologically, snow
melt is a very difficult

184
00:09:21,560 --> 00:09:25,130
process to model, a
process that involves

185
00:09:25,163 --> 00:09:28,133
a lot of radiative
and heat transfer,

186
00:09:28,166 --> 00:09:29,635
but very important
for ecosystems

187
00:09:29,668 --> 00:09:32,738
because when there's snow a
lot of things are turned off,

188
00:09:32,771 --> 00:09:34,840
and when they're wet a lot
of things are turned on,

189
00:09:34,873 --> 00:09:37,410
so we have to get that
balance right as well.

190
00:09:38,477 --> 00:09:40,846
How the water moves
through this top surface

191
00:09:40,879 --> 00:09:44,149
into the deeper ground waters
is also very important,

192
00:09:44,182 --> 00:09:47,153
and how it evaporates
back up to the atmosphere.

193
00:09:48,120 --> 00:09:50,623
And of course the larger
evapotranspiration,

194
00:09:50,656 --> 00:09:51,890
the evaporation off the soil

195
00:09:51,923 --> 00:09:54,760
and the transpiration
out of leaves.

196
00:09:54,793 --> 00:09:56,795
This is not a picture
of evapotranspiration.

197
00:09:56,828 --> 00:09:58,063
You can see this
is just clouds,

198
00:09:58,096 --> 00:10:02,535
but you can imagine water
coming off these trees.

199
00:10:03,635 --> 00:10:05,938
We need to understand
how water runs off

200
00:10:05,971 --> 00:10:08,073
into our rivers and oceans,

201
00:10:08,106 --> 00:10:10,476
and how it's routed
over the landscape.

202
00:10:11,510 --> 00:10:13,312
All these different
water components,

203
00:10:13,345 --> 00:10:15,648
the snow, the ground
water, the soil moisture,

204
00:10:15,681 --> 00:10:18,484
they need to keep a delicate
balance with each other;

205
00:10:18,517 --> 00:10:21,387
otherwise, our models might
build up too much snow

206
00:10:21,420 --> 00:10:22,921
or too much ground water,

207
00:10:22,954 --> 00:10:25,858
so we have to keep a
delicate balance of water.

208
00:10:25,891 --> 00:10:28,160
Hard to visualize so I just
put a guy balancing over water,

209
00:10:28,193 --> 00:10:30,429
but water balance.

210
00:10:31,430 --> 00:10:34,600
Then there's dynamic
components of ecosystems:

211
00:10:34,633 --> 00:10:36,869
How trees compete
with one another.

212
00:10:39,438 --> 00:10:41,807
They compete for
space, for light,

213
00:10:41,840 --> 00:10:43,709
for water, for nutrients.

214
00:10:44,876 --> 00:10:48,180
Which plant wins when
there's that kind of a fight?

215
00:10:48,213 --> 00:10:50,282
And when there's a disturbance,

216
00:10:50,315 --> 00:10:54,319
which plants go colonize
and establish first?

217
00:10:54,352 --> 00:10:56,388
Or when there's a
change in climate,

218
00:10:56,421 --> 00:10:58,957
are there new bioclimatic
envelopes for plants

219
00:10:58,990 --> 00:11:03,229
to move into as other
plants die out from?

220
00:11:05,130 --> 00:11:07,700
Speaking of death, how
do you kill a tree?

221
00:11:07,733 --> 00:11:09,702
This is actually a
harder question to answer

222
00:11:09,735 --> 00:11:11,570
than one may realize.

223
00:11:11,603 --> 00:11:13,505
There's a lot of different
ways trees can die,

224
00:11:13,538 --> 00:11:15,941
and again these models
have to kill the trees

225
00:11:15,974 --> 00:11:17,876
at the right rates
for the right reasons.

226
00:11:17,909 --> 00:11:19,578
Is it because they overheated?

227
00:11:19,611 --> 00:11:22,748
Is it because they ran out
of carbon for metabolism?

228
00:11:22,781 --> 00:11:24,216
Is it because of a wind throw?

229
00:11:24,249 --> 00:11:26,652
Is it because of a
disturbance like fire,

230
00:11:26,685 --> 00:11:29,755
the largest disturbance
in Earth's system?

231
00:11:29,788 --> 00:11:30,890
How do you model fire?

232
00:11:32,224 --> 00:11:33,459
And what comes off of fire?

233
00:11:33,492 --> 00:11:37,229
Smoke and other
gases and aerosols.

234
00:11:37,262 --> 00:11:39,932
So all these processes
get wrapped up

235
00:11:39,965 --> 00:11:44,870
into this simple diagram
of a modeled ecosystem.

236
00:11:44,903 --> 00:11:46,371
Of course we know
that ecosystems

237
00:11:46,404 --> 00:11:48,107
are a lot more complex
than this, right?

238
00:11:48,140 --> 00:11:49,942
They look a lot more like this

239
00:11:49,975 --> 00:11:53,512
with all these processes
embedded within.

240
00:11:54,813 --> 00:11:58,017
So that's what ecosystem
scientists try to understand.

241
00:11:59,684 --> 00:12:00,652
That's all well and good,

242
00:12:00,685 --> 00:12:02,588
but we've thrown another
kind of monkey wrench

243
00:12:02,621 --> 00:12:05,724
into this machinery,
and that's this.

244
00:12:05,757 --> 00:12:09,161
Without even putting
axes or labels on this,

245
00:12:09,194 --> 00:12:12,564
a lot of people can recognize
this; it's becoming iconic.

246
00:12:12,597 --> 00:12:15,501
It's the rise in CO2
in the atmosphere

247
00:12:15,534 --> 00:12:18,504
influencing plants and
influencing our understanding

248
00:12:18,537 --> 00:12:20,473
of how plants respond.

249
00:12:22,207 --> 00:12:25,644
So all these complexities
and uncertainties

250
00:12:25,677 --> 00:12:27,446
are wrapped together
in these climate models

251
00:12:27,479 --> 00:12:29,915
run by different institutions,

252
00:12:29,948 --> 00:12:33,085
some out of the
US, Japan, France,

253
00:12:33,118 --> 00:12:36,122
and so there's a lot of
disagreement about how to model

254
00:12:37,556 --> 00:12:41,026
carbon metabolism, or
how to model evaporation.

255
00:12:41,059 --> 00:12:43,295
So that's why we get a
lot of these differences

256
00:12:43,328 --> 00:12:46,131
with what we project
into the future.

257
00:12:46,164 --> 00:12:49,368
How a plant behaves
to my French colleague

258
00:12:49,401 --> 00:12:51,236
is different than
how a plant behaves

259
00:12:51,269 --> 00:12:52,771
to my Australian colleague.

260
00:12:52,804 --> 00:12:54,206
So we have to come
together as scientists

261
00:12:54,239 --> 00:12:58,110
to really understand the
world beyond our backyards.

262
00:12:58,143 --> 00:13:01,446
So this is a kind of a classic
picture of these models

263
00:13:01,479 --> 00:13:04,483
shooting out into the future
and diverging heavily.

264
00:13:04,516 --> 00:13:06,118
The models that
go up are saying

265
00:13:06,151 --> 00:13:08,120
ecosystems are
gonna do just fine,

266
00:13:08,153 --> 00:13:10,022
and the models
that are going down

267
00:13:10,055 --> 00:13:13,192
are saying ecosystems
are gonna crash.

268
00:13:13,225 --> 00:13:17,830
Now this paper was put out in
2006, a number of years ago,

269
00:13:17,863 --> 00:13:19,731
so a lot of development
has happened

270
00:13:19,764 --> 00:13:21,200
in these models since then.

271
00:13:21,233 --> 00:13:22,968
The authors put
out another paper

272
00:13:23,001 --> 00:13:25,170
just a couple of years
ago, about 10 years later,

273
00:13:25,203 --> 00:13:28,207
and this is what the models
look more like today,

274
00:13:29,241 --> 00:13:31,910
which is not too different.

275
00:13:31,943 --> 00:13:34,580
We're still faced with
a lot of uncertainty.

276
00:13:34,613 --> 00:13:36,782
I think the only thing
that's gotten better

277
00:13:36,815 --> 00:13:38,383
is that we've
picked a more like

278
00:13:38,416 --> 00:13:40,586
modern color scheme
with the lines,

279
00:13:40,619 --> 00:13:45,191
but you know I don't think
that is useful in science.

280
00:13:46,424 --> 00:13:50,062
So what we have
here is essentially

281
00:13:50,095 --> 00:13:53,065
like that fable the Blind
Men and the Elephant,

282
00:13:53,098 --> 00:13:54,867
which I kind of
think of as just

283
00:13:54,900 --> 00:13:56,768
terrestrial ecosystem modelers.

284
00:13:56,801 --> 00:13:59,037
Everyone is saying
something different

285
00:13:59,070 --> 00:14:01,340
about the world
in front of them,

286
00:14:01,373 --> 00:14:03,576
but they're all kinda
saying the same thing.

287
00:14:05,076 --> 00:14:08,247
They're all wrong,
but they're all right.

288
00:14:08,280 --> 00:14:10,749
So how do we put this
knowledge together

289
00:14:10,782 --> 00:14:13,352
to really understand the
elephant in front of us,

290
00:14:13,385 --> 00:14:16,488
the fate of Earth's
ecosystems in the biosphere?

291
00:14:18,623 --> 00:14:20,692
Now a lot of these
models as you can imagine

292
00:14:20,725 --> 00:14:24,163
are built in the US, in Europe,

293
00:14:25,030 --> 00:14:27,566
and so I've put
together this cartogram

294
00:14:27,599 --> 00:14:31,370
which blows up the
size of the country

295
00:14:31,403 --> 00:14:32,604
depending on how
much investment

296
00:14:32,637 --> 00:14:34,306
there is in the models.

297
00:14:34,339 --> 00:14:35,607
So you can see if we're
trying to come up with

298
00:14:35,640 --> 00:14:37,843
a global picture, but
we're developing models

299
00:14:37,876 --> 00:14:39,711
based on our own
inherent biases

300
00:14:39,744 --> 00:14:42,948
from how we understand how
plants work from our backyards,

301
00:14:42,981 --> 00:14:45,450
we're gonna have
these inherent biases

302
00:14:45,483 --> 00:14:46,986
about the world at large.

303
00:14:49,521 --> 00:14:54,326
The way I approach this
problem and this challenge

304
00:14:54,359 --> 00:14:59,298
is through a triangulation of
carbon, water, and nutrients

305
00:14:59,331 --> 00:15:03,035
as they impact climate and as
they're impacted by climate.

306
00:15:03,068 --> 00:15:05,203
You can adjust this from an
ocean, and an atmosphere,

307
00:15:05,236 --> 00:15:06,705
and a land perspective,

308
00:15:06,738 --> 00:15:10,442
and I address it from a land
and ecosystem perspective

309
00:15:10,475 --> 00:15:12,511
but working hand-in-hand
with my atmospheric

310
00:15:12,544 --> 00:15:16,515
and ocean and ice colleagues
at JPL and worldwide.

311
00:15:17,482 --> 00:15:18,917
So one of the first
things that I did

312
00:15:18,950 --> 00:15:21,720
is I started taking all these
models from all over the world

313
00:15:21,753 --> 00:15:25,857
and putting them here in
house on JPL's supercomputers

314
00:15:25,890 --> 00:15:27,993
to run with common conditions.

315
00:15:28,026 --> 00:15:31,763
I didn't want the French
guys running it one way

316
00:15:31,796 --> 00:15:35,233
and the British guys
running it a different way.

317
00:15:35,266 --> 00:15:37,269
We had to run them the same way

318
00:15:37,302 --> 00:15:39,004
so that we can understand
that the differences

319
00:15:39,037 --> 00:15:41,506
are not due to the fact that
somebody started their model

320
00:15:41,539 --> 00:15:45,644
in 1801 and the other
one started in 1850.

321
00:15:45,677 --> 00:15:50,315
So we've been using NASA's
supercomputing infrastructure

322
00:15:50,348 --> 00:15:52,217
to help solve that
part of the problem,

323
00:15:52,250 --> 00:15:53,952
but that's really
not good enough.

324
00:15:53,985 --> 00:15:57,289
We still need to understand
how well the models are doing.

325
00:15:58,323 --> 00:16:02,060
So here's a plot of all the
land on Earth by latitude.

326
00:16:02,093 --> 00:16:05,664
So you can see there's more
land in the higher latitudes.

327
00:16:05,697 --> 00:16:09,201
This blue line is the
breathing of the biosphere,

328
00:16:09,234 --> 00:16:12,137
how much breathing in and
out ecosystems do globally,

329
00:16:12,170 --> 00:16:14,440
and most of it occurs
in the tropics.

330
00:16:15,407 --> 00:16:17,242
But there's actually
what we call two poles

331
00:16:17,275 --> 00:16:18,343
of the carbon cycle.

332
00:16:18,376 --> 00:16:21,847
The amount of carbon
stored in ecosystems is--

333
00:16:21,880 --> 00:16:23,715
maximizes both in the
tropics where you have

334
00:16:23,748 --> 00:16:26,885
all these big trees and in
the Arctic where you have

335
00:16:26,918 --> 00:16:31,924
millennia of carbon locked in
the soils and the permafrost.

336
00:16:33,258 --> 00:16:35,761
Now how well are these
two poles sampled?

337
00:16:35,794 --> 00:16:37,729
As you can imagine
on the ground

338
00:16:37,762 --> 00:16:39,698
we sample
in our backyards:

339
00:16:39,731 --> 00:16:42,300
We sample in the US,
we sample in Europe;

340
00:16:42,333 --> 00:16:45,370
the places that have
the money to do this.

341
00:16:45,403 --> 00:16:48,507
So we are undersampling
the most important

342
00:16:48,540 --> 00:16:50,609
regions on the planet.

343
00:16:50,642 --> 00:16:54,279
That's where NASA and JPL's
satellite remote sensing

344
00:16:54,312 --> 00:16:55,847
can really come into play

345
00:16:55,880 --> 00:16:58,617
because we now have a
global picture across

346
00:16:58,650 --> 00:17:01,186
multiple dimensions of
ecosystem properties,

347
00:17:01,219 --> 00:17:03,589
and I'll go through
some of that now.

348
00:17:04,456 --> 00:17:06,825
Coming into a case
study in Amazonia.

349
00:17:06,858 --> 00:17:08,960
It's our crown
jewel of ecosystems.

350
00:17:08,993 --> 00:17:09,828
This is where

351
00:17:11,663 --> 00:17:13,765
the enormous amounts of,

352
00:17:13,798 --> 00:17:16,935
the largest biodiversity
on the planet exists.

353
00:17:16,968 --> 00:17:19,037
This exerts a huge influence

354
00:17:19,070 --> 00:17:22,007
on the Earth's
climate as a whole.

355
00:17:22,040 --> 00:17:24,543
It breaths in the CO2
that we breathe out,

356
00:17:24,576 --> 00:17:27,512
and it breathes out the
oxygen that we breathe in.

357
00:17:27,545 --> 00:17:29,514
It is the lungs of our planet.

358
00:17:29,547 --> 00:17:31,516
So we are very concerned with

359
00:17:31,549 --> 00:17:33,785
what is gonna
become of our lungs.

360
00:17:33,818 --> 00:17:36,888
Are we gonna be breathing
fine, be breathing fine,

361
00:17:36,921 --> 00:17:38,424
or are we gonna be choking?

362
00:17:40,024 --> 00:17:43,295
Looking at these models we
see that some of the models

363
00:17:43,328 --> 00:17:45,564
are predicting that
the Amazon forest

364
00:17:45,597 --> 00:17:47,499
is gonna die back
into the future.

365
00:17:48,433 --> 00:17:50,235
This is quite worrisome.

366
00:17:50,268 --> 00:17:51,336
How are they doing this?

367
00:17:51,369 --> 00:17:55,006
It's because of a predicted
increase in droughts:

368
00:17:55,039 --> 00:17:58,143
drought intensity,
drought magnitude,

369
00:17:58,176 --> 00:17:59,511
and drought frequency.

370
00:18:02,780 --> 00:18:05,283
The largest drought in
the history of our records

371
00:18:05,316 --> 00:18:08,186
in the Amazon occurred in 2005.

372
00:18:08,219 --> 00:18:10,522
So scientists thought, great!

373
00:18:10,555 --> 00:18:12,390
I mean this is not
great, but this is great

374
00:18:12,423 --> 00:18:14,359
because it's kind
of a test to see

375
00:18:14,392 --> 00:18:17,729
how resilient is the Amazon.

376
00:18:17,762 --> 00:18:20,799
So when this drought
hit in '05 we said,

377
00:18:20,832 --> 00:18:22,234
okay, what did the Amazon do?

378
00:18:22,267 --> 00:18:26,371
Did it choke or did
it do all right?

379
00:18:26,404 --> 00:18:29,641
One of the first studies that
came out after this drought

380
00:18:29,674 --> 00:18:33,512
was very curious, a bit
of a head scratcher.

381
00:18:33,545 --> 00:18:36,815
It showed and suggested
that the Amazon greened-up

382
00:18:36,848 --> 00:18:37,983
during the drought.

383
00:18:38,016 --> 00:18:40,886
This didn't really make much
sense to a lot of people.

384
00:18:40,919 --> 00:18:43,288
But a lot of high
profile papers came out

385
00:18:43,321 --> 00:18:44,689
saying that the Amazon
forest greened-up,

386
00:18:44,722 --> 00:18:47,359
and their argument was that,
yes, there was less rainfall,

387
00:18:47,392 --> 00:18:50,128
but there was still plenty
of water in the soil

388
00:18:50,161 --> 00:18:52,030
and there was actually
more sunlight now

389
00:18:52,063 --> 00:18:54,733
so the plants could
photosynthesize more.

390
00:18:54,766 --> 00:18:57,903
But they didn't actually have
on-the-ground measurements

391
00:18:57,936 --> 00:19:00,172
to substantiate
this hypothesis,

392
00:19:00,205 --> 00:19:02,341
but it was very
intriguing indeed.

393
00:19:03,908 --> 00:19:04,910
So what we did was we took

394
00:19:04,943 --> 00:19:06,811
some of the satellite
measurements.

395
00:19:06,844 --> 00:19:07,879
There as a mission called

396
00:19:07,912 --> 00:19:09,414
the Tropical Rainfall
Measuring Mission

397
00:19:09,447 --> 00:19:11,350
which gave us the
precipitation.

398
00:19:12,450 --> 00:19:15,086
I had actually done
a lot of my PhD work

399
00:19:15,119 --> 00:19:17,989
on evapotranspiration, so
this was able to give us

400
00:19:18,022 --> 00:19:20,692
the amount of water
that actually evaporated

401
00:19:20,725 --> 00:19:21,593
out of the soil.

402
00:19:21,626 --> 00:19:22,594
You need both.

403
00:19:22,627 --> 00:19:24,496
You can't just know that
there was less rainfall

404
00:19:24,529 --> 00:19:27,032
without knowing how much
evaporation there was

405
00:19:27,065 --> 00:19:29,501
to determine how much
water there is in the soil.

406
00:19:30,635 --> 00:19:34,239
So if we look at a record
of rain in the Amazon,

407
00:19:34,272 --> 00:19:36,274
it kind of goes up
and down, up and down.

408
00:19:36,307 --> 00:19:38,109
In '05 during this drought
there was a little dip,

409
00:19:38,142 --> 00:19:41,546
but it was hard to tell
if the soils dried out.

410
00:19:41,579 --> 00:19:44,516
So when we put this
evaporation drought on it,

411
00:19:44,549 --> 00:19:47,119
we can see how much
the Amazon dried.

412
00:19:48,186 --> 00:19:50,055
So that really gave
us a better indicator

413
00:19:50,088 --> 00:19:53,124
of how dry the Amazon was.

414
00:19:53,157 --> 00:19:55,093
But that's only half the coin.

415
00:19:55,126 --> 00:19:57,663
We also need to know if
the trees were dead or not.

416
00:19:58,696 --> 00:20:00,665
Now we couldn't get
this from space.

417
00:20:00,698 --> 00:20:02,567
There was no spaceborne asset.

418
00:20:02,600 --> 00:20:05,670
So we had to go
into the Amazon.

419
00:20:05,703 --> 00:20:08,940
We spent a lot of
time and energy

420
00:20:08,973 --> 00:20:13,345
censusing these longterm
monitoring plots to see

421
00:20:14,646 --> 00:20:17,182
if the trees died or not;
very challenging to do.

422
00:20:18,416 --> 00:20:22,420
The bottom line was that
at the end of the day when

423
00:20:22,453 --> 00:20:23,888
areas that were more dry

424
00:20:23,921 --> 00:20:27,259
according to our rain
and evaporation index,

425
00:20:27,292 --> 00:20:31,930
we also saw a very tight
correlation with tree death.

426
00:20:31,963 --> 00:20:33,865
Drought equals tree
death essentially.

427
00:20:33,898 --> 00:20:34,732
Not too surprising,

428
00:20:34,766 --> 00:20:37,168
but remember those
first papers said

429
00:20:37,201 --> 00:20:39,504
that the Amazon
rainforest greened-up

430
00:20:39,537 --> 00:20:42,173
during this drought, so
we had a bit of conflict

431
00:20:42,206 --> 00:20:43,642
in the scientific literature.

432
00:20:44,575 --> 00:20:47,445
When our paper came out,
some of the original authors

433
00:20:47,478 --> 00:20:50,348
of the Green-Up
paper broke ranks

434
00:20:50,381 --> 00:20:53,918
and they kind of reanalyzed
their data and found that

435
00:20:53,951 --> 00:20:57,255
they had not quite properly
accounted for fire.

436
00:20:57,288 --> 00:20:58,790
There was more fire
in this drought

437
00:20:58,823 --> 00:21:00,258
and the smoke was kind
of getting in the way

438
00:21:00,291 --> 00:21:01,393
of their signals.

439
00:21:01,426 --> 00:21:04,696
So they came up with
a new paper that said

440
00:21:04,729 --> 00:21:06,698
Amazon forest did not green-up.

441
00:21:09,567 --> 00:21:13,505
Yeah, it's kinda like gang
warfare, but nerd style.

442
00:21:13,538 --> 00:21:14,973
[laughing]

443
00:21:15,006 --> 00:21:15,974
You don't want
to see these guys

444
00:21:16,007 --> 00:21:17,976
in the science
conference hallways.

445
00:21:18,009 --> 00:21:19,177
It gets tense.

446
00:21:20,678 --> 00:21:23,415
You know like pocket
protectors get thrown down.

447
00:21:23,448 --> 00:21:24,983
I'm kidding.

448
00:21:25,016 --> 00:21:29,988
But, again, so there's been
scores of papers on this topic

449
00:21:32,023 --> 00:21:33,258
because it's important,

450
00:21:33,291 --> 00:21:35,560
this is really important
for us to understand.

451
00:21:35,593 --> 00:21:39,097
What's gonna happen to
the tropics, the Amazon,

452
00:21:39,130 --> 00:21:42,834
and ecosystems under this
projection of increasing

453
00:21:43,801 --> 00:21:46,305
frequency and
magnitude droughts?

454
00:21:47,705 --> 00:21:49,708
I mentioned that the '05
drought was the biggest

455
00:21:49,741 --> 00:21:51,509
in the history of our records,

456
00:21:51,542 --> 00:21:53,945
and these models are projecting

457
00:21:53,978 --> 00:21:55,180
more frequent, bigger droughts.

458
00:21:55,213 --> 00:21:56,948
In 2010 and even
bigger drought hit.

459
00:21:56,981 --> 00:21:59,884
Just five years later in
the history of our records

460
00:21:59,917 --> 00:22:01,953
a bigger drought
hit the Amazon.

461
00:22:03,554 --> 00:22:06,891
Five years after that an even
bigger drought hit the Amazon.

462
00:22:06,924 --> 00:22:09,394
So these projections
and the models

463
00:22:09,427 --> 00:22:13,565
are actually starting to play
out in front of our eyes.

464
00:22:14,932 --> 00:22:17,268
What's different in
the recent droughts,

465
00:22:17,301 --> 00:22:20,739
unlike the '05 drought, is
that we have new capabilities,

466
00:22:20,772 --> 00:22:22,707
we have new technical
capabilities from space

467
00:22:22,740 --> 00:22:24,309
and from models.

468
00:22:25,910 --> 00:22:30,281
We can now observe the glow of
plants, called fluorescence.

469
00:22:30,314 --> 00:22:32,450
This is actually a
bit of a mistake.

470
00:22:32,483 --> 00:22:35,387
We didn't intend to
do these measurements.

471
00:22:35,420 --> 00:22:37,455
These are measurements from
three different satellites:

472
00:22:37,488 --> 00:22:42,060
OCO-2 is out of JPL, there's
one called GOSAT out of Japan,

473
00:22:42,093 --> 00:22:46,765
and one, a satellite
called GOME, out of Europe.

474
00:22:47,665 --> 00:22:48,933
These satellites
were not intending

475
00:22:48,966 --> 00:22:49,934
to measure fluorescence;

476
00:22:49,967 --> 00:22:52,036
they were intending to
measure something else.

477
00:22:52,069 --> 00:22:55,607
But they happened to have this
measurement of fluorescence

478
00:22:55,640 --> 00:22:59,077
that we started scratching
our heads and thinking,

479
00:22:59,110 --> 00:23:02,313
what is this measurement,
this is quite weird.

480
00:23:02,346 --> 00:23:04,215
We dug into it and
we found that

481
00:23:04,248 --> 00:23:06,484
it was this
glow of plants.

482
00:23:06,517 --> 00:23:10,855
So now for the first time
in the history of humankind

483
00:23:10,888 --> 00:23:15,560
we can see photosynthetic
activity as the glow of plants,

484
00:23:15,593 --> 00:23:18,096
when before we were just seeing
if they were green or not.

485
00:23:18,129 --> 00:23:21,032
We saw that the Amazon
greened-up or it was not green.

486
00:23:21,065 --> 00:23:24,235
This is like going to your
doctor and your doctor saying,

487
00:23:24,268 --> 00:23:26,371
"I can't help you
until you dead."

488
00:23:26,404 --> 00:23:29,007
I can't tell until
you're not green anymore.

489
00:23:29,040 --> 00:23:34,012
Now we can see activity is
slowing down or changing

490
00:23:34,045 --> 00:23:38,483
before plants, crops,
drop their leaves or die.

491
00:23:42,920 --> 00:23:45,523
NASA also established the
Carbon Monitoring System

492
00:23:45,556 --> 00:23:47,592
where we started
integrating our observations

493
00:23:47,625 --> 00:23:51,196
across ocean, land,
and atmosphere,

494
00:23:51,229 --> 00:23:55,167
and anthropogenic
carbon cycling.

495
00:23:56,067 --> 00:23:58,903
This CMS program
has been established

496
00:23:58,936 --> 00:24:02,974
and is not moving into new
applications and new venues.

497
00:24:03,007 --> 00:24:06,444
Another really exciting avenue
that is just on the horizon

498
00:24:06,477 --> 00:24:08,079
is we're starting to look at

499
00:24:08,112 --> 00:24:09,681
the International
Space Station.

500
00:24:11,349 --> 00:24:13,952
We don't see a lot out
of the space station

501
00:24:13,985 --> 00:24:18,122
in terms of Earth observation
for science inquiry

502
00:24:18,155 --> 00:24:21,092
or for, especially
for, ecosystems;

503
00:24:22,527 --> 00:24:26,631
we see astronauts doing their
thing on a space station.

504
00:24:29,066 --> 00:24:32,570
If we look at this year's
calendar starting in March,

505
00:24:32,603 --> 00:24:34,739
there's a series of launches
going to the space station

506
00:24:34,772 --> 00:24:38,043
on SpaceX rockets, cargo
resupplies for the astronauts.

507
00:24:39,043 --> 00:24:42,214
They each tell us something
different about ecosystems.

508
00:24:43,447 --> 00:24:44,916
I'm gonna tell you a
little bit about ECOSTRESS,

509
00:24:44,949 --> 00:24:48,853
which is the first one
coming up out of JPL in June.

510
00:24:48,886 --> 00:24:50,855
It's right on the horizon,

511
00:24:50,888 --> 00:24:52,991
and I happen to be
the science lead of it

512
00:24:53,024 --> 00:24:55,527
so of course I'll
tell you about it.

513
00:24:57,628 --> 00:25:00,765
It starts with this
premise that the landscape

514
00:25:00,798 --> 00:25:03,234
is very heterogeneous.

515
00:25:03,267 --> 00:25:06,137
Lots of plants are doing lots
of things at very small scales

516
00:25:06,170 --> 00:25:09,207
so we need to know what's
going on everywhere

517
00:25:09,240 --> 00:25:10,208
but at small scales.

518
00:25:10,241 --> 00:25:13,244
It's kind of a
conundrum, if you will.

519
00:25:13,277 --> 00:25:16,047
We can measure stuff on
the ground with sensors.

520
00:25:16,080 --> 00:25:20,018
We can run drones around the
landscape, or even aircraft.

521
00:25:20,051 --> 00:25:22,620
Then of course we have
our space satellites.

522
00:25:22,653 --> 00:25:26,624
I've built sensors you
can stick into trees,

523
00:25:26,657 --> 00:25:29,494
and it will tell you how
much water they're taking up.

524
00:25:29,527 --> 00:25:31,195
We put them at these
towers that measure

525
00:25:31,228 --> 00:25:33,498
the water fluxes
and carbon fluxes

526
00:25:33,531 --> 00:25:35,767
in and out of ecosystems.

527
00:25:35,800 --> 00:25:38,269
We have drones that
can fly back and forth

528
00:25:38,302 --> 00:25:41,239
across crops seeing how
much water is needed

529
00:25:41,272 --> 00:25:45,043
or how much is used, as well
as through forests determining

530
00:25:48,346 --> 00:25:49,481
tree inventories.

531
00:25:52,283 --> 00:25:55,587
But we really need something
that does everything in one.

532
00:25:55,620 --> 00:26:00,224
We can't put drones and towers
and aircraft everywhere.

533
00:26:00,257 --> 00:26:02,627
So if we look at our current
satellite capabilities,

534
00:26:02,660 --> 00:26:04,963
this is MODIS, and it
shows this landscape

535
00:26:04,996 --> 00:26:07,231
at one kilometer resolution.

536
00:26:07,264 --> 00:26:09,667
It's kinda picking up that
heterogeneity, but not quite.

537
00:26:09,700 --> 00:26:12,971
This is Landsat at
60 meters resolution,

538
00:26:13,004 --> 00:26:14,606
much sharper picture.

539
00:26:15,506 --> 00:26:17,308
And you can see it's
starting to pick up

540
00:26:17,341 --> 00:26:20,278
those differences
across the landscape.

541
00:26:20,311 --> 00:26:22,714
This is our managed
landscape with agriculture.

542
00:26:22,747 --> 00:26:26,084
If we look at a
natural landscape

543
00:26:26,117 --> 00:26:28,586
and we paint on the
vegetation here,

544
00:26:28,619 --> 00:26:31,889
we can see that the
riparian corridors

545
00:26:31,922 --> 00:26:34,092
require very fine
spatial skills.

546
00:26:34,125 --> 00:26:36,895
MODIS is not gonna pick
that up; Landsat will.

547
00:26:38,362 --> 00:26:40,865
So that's space,
what about time?

548
00:26:40,898 --> 00:26:44,636
This is the evaporation
measured at one of the sites

549
00:26:44,669 --> 00:26:46,905
on the ground every 30 minutes,

550
00:26:47,905 --> 00:26:49,607
and you can see the kind
of bumps and wiggles

551
00:26:49,640 --> 00:26:50,675
over the course of the year.

552
00:26:50,708 --> 00:26:55,213
Landsat comes over only
about every 16 days,

553
00:26:55,246 --> 00:26:56,381
and if there's
clouds in the way

554
00:26:56,414 --> 00:26:59,717
then you know every 32
days or multiples thereof.

555
00:26:59,750 --> 00:27:02,053
So even though it's got
the spatial resolution,

556
00:27:02,086 --> 00:27:05,023
it doesn't quite have
that temporal resolution.

557
00:27:05,056 --> 00:27:09,027
ECOSTRESS is gonna
come in and measure,

558
00:27:09,060 --> 00:27:12,163
or fly over us every
about three to five days

559
00:27:12,196 --> 00:27:14,599
really picking up
that seasonal cycle.

560
00:27:17,501 --> 00:27:20,571
Another interesting aspect
about plants is that

561
00:27:20,604 --> 00:27:22,974
there's a diurnal cycle.

562
00:27:23,007 --> 00:27:25,443
Some plants when
there's water stress

563
00:27:25,476 --> 00:27:27,445
will close those
stomata in the afternoon

564
00:27:27,478 --> 00:27:30,381
when it's really hot so they
don't lose a lot of water,

565
00:27:30,414 --> 00:27:32,784
and then they'll open
back up in early evening

566
00:27:32,817 --> 00:27:36,888
before they lose sunlight to do

567
00:27:36,921 --> 00:27:38,489
a little bit more
photosynthesis.

568
00:27:38,522 --> 00:27:41,826
Most of our satellites pass
over us at the same time

569
00:27:41,859 --> 00:27:46,264
every day at the polar
orbiters, 10:30 every time,

570
00:27:47,131 --> 00:27:48,800
so they miss this
diurnal cycle,

571
00:27:48,833 --> 00:27:52,270
they miss this daily
functioning of plants.

572
00:27:52,303 --> 00:27:54,405
There are some satellites
that hover over us,

573
00:27:54,438 --> 00:27:57,842
geostationary all the time,
but because of the orbit

574
00:27:57,875 --> 00:27:59,644
the pixels are very coarse

575
00:27:59,677 --> 00:28:02,580
so they would lump together
plants that are shutting down

576
00:28:02,613 --> 00:28:05,349
and plants that aren't
shutting down into one,

577
00:28:05,382 --> 00:28:07,352
so you wouldn't be able
to distinguish this.

578
00:28:09,954 --> 00:28:12,857
At the end of the day,
we want to take a look

579
00:28:12,890 --> 00:28:16,260
at this landscape and
apply some color to it.

580
00:28:16,293 --> 00:28:19,263
We want to figure out
if there's droughts

581
00:28:19,296 --> 00:28:22,633
which trees, which species,
are gonna die first?

582
00:28:22,666 --> 00:28:24,635
Because some will
more than others.

583
00:28:24,668 --> 00:28:26,437
Some need more
water than others.

584
00:28:26,470 --> 00:28:28,973
Some are less efficient
with water than others.

585
00:28:30,174 --> 00:28:32,043
So that's a little
bit about ECOSTRESS,

586
00:28:32,076 --> 00:28:36,981
and we're really looking
forward to the launch in June.

587
00:28:37,014 --> 00:28:40,251
Hopefully I'll come back
and give another talk later

588
00:28:40,284 --> 00:28:43,321
on some of those results
when we get them.

589
00:28:44,588 --> 00:28:46,924
I'll just briefly mention
a couple of other missions

590
00:28:46,957 --> 00:28:48,493
going to the space station.

591
00:28:48,526 --> 00:28:50,161
I'm not necessarily
involved in,

592
00:28:50,194 --> 00:28:52,330
but they're my
colleagues and partners,

593
00:28:52,363 --> 00:28:54,732
and there's
synergies among them.

594
00:28:54,765 --> 00:28:58,203
HISUI is coming out of
Japan, and this will measure,

595
00:28:59,370 --> 00:29:02,140
it's called hyperspectral
or spectroscopic signatures,

596
00:29:02,173 --> 00:29:04,108
like the unique
fingerprints of plants.

597
00:29:04,141 --> 00:29:05,910
So even though ECOSTRESS
will be able to tell you

598
00:29:05,943 --> 00:29:08,546
which plants need
more or less water,

599
00:29:08,579 --> 00:29:12,450
this one will tell you kind
of what those plants are.

600
00:29:12,483 --> 00:29:15,386
So that's extremely
useful because we'll know

601
00:29:15,419 --> 00:29:18,923
what the plants are and how
much water they're using.

602
00:29:20,357 --> 00:29:25,096
GEDI is using LIDAR, these
lasers from the space station,

603
00:29:25,129 --> 00:29:28,533
to map out how big trees
are, and that will tell you

604
00:29:28,566 --> 00:29:33,171
how much carbon is
stored in ecosystems.

605
00:29:33,204 --> 00:29:36,741
Then finally, OCO-3,
another one let out of JPL,

606
00:29:36,774 --> 00:29:38,676
will be measuring that
fluorescence again,

607
00:29:38,709 --> 00:29:40,912
as well as CO2 in
the atmosphere.

608
00:29:40,945 --> 00:29:42,647
All on the space station.

609
00:29:42,680 --> 00:29:45,516
All with that diurnal
cycle sampling.

610
00:29:45,549 --> 00:29:48,253
So very exciting to
have these all up.

611
00:29:49,420 --> 00:29:53,724
So essentially we go up on a
SpaceX rocket, cargo resupply.

612
00:29:53,757 --> 00:29:56,427
I tell people that we're
going up with he pizzas,

613
00:29:56,460 --> 00:29:58,062
the pizzas for the astronauts.

614
00:29:58,095 --> 00:30:01,499
Then the Dragon capsule docks.

615
00:30:01,532 --> 00:30:02,800
What's really interesting about

616
00:30:02,833 --> 00:30:04,168
the International
Space Station is that

617
00:30:04,201 --> 00:30:06,704
we've gone up on
an American rocket,

618
00:30:06,737 --> 00:30:10,174
that's the Canadian robotic
arm removing the instrument,

619
00:30:10,207 --> 00:30:13,544
and we're mounting to
the Japanese module.

620
00:30:13,577 --> 00:30:17,615
So we're very much
internationally coming together

621
00:30:17,648 --> 00:30:22,520
both engineering and
scientifically to tackle

622
00:30:22,553 --> 00:30:24,322
global scale questions.

623
00:30:25,289 --> 00:30:26,691
ECOSTRESS is going
up first so we get

624
00:30:26,724 --> 00:30:28,259
the best real estate on here.

625
00:30:28,292 --> 00:30:29,560
It's kind of like
the cul-de-sac

626
00:30:29,593 --> 00:30:31,028
of the space station,

627
00:30:31,061 --> 00:30:34,833
whereas OCO-3 kinda gets the
side street a little bit.

628
00:30:38,969 --> 00:30:41,839
What's interesting
about these instruments

629
00:30:41,872 --> 00:30:45,843
is that they measure the
structure, GEDI for example,

630
00:30:45,876 --> 00:30:48,646
the composition,
which was HISUI,

631
00:30:48,679 --> 00:30:52,750
the evapotranspiration, the
water use which was ECOSTRESS,

632
00:30:52,783 --> 00:30:55,119
and the fluorescence,
which is OCO-3.

633
00:30:55,152 --> 00:30:57,255
These are essentially
lenses of ecosystems.

634
00:30:57,288 --> 00:30:59,423
These are what
comprises ecosystems.

635
00:30:59,456 --> 00:31:00,925
Ecosystems are
comprised of structure,

636
00:31:00,958 --> 00:31:05,496
composition, and function,
and there's this CO2 aspect.

637
00:31:05,529 --> 00:31:09,567
We can also get all of these
using airborne spacecraft.

638
00:31:09,600 --> 00:31:12,703
Not spacecraft,
airborne platforms,

639
00:31:12,736 --> 00:31:15,239
which is also very useful
for being able to target

640
00:31:15,272 --> 00:31:18,376
certain areas that we can't
see very well like the tropics,

641
00:31:18,409 --> 00:31:20,077
for example, where
it's very cloudy.

642
00:31:20,110 --> 00:31:22,680
So even if we have
the satellite sensors,

643
00:31:22,713 --> 00:31:25,450
we might not be able to see
through the clouds as well.

644
00:31:26,383 --> 00:31:29,053
And we need very high spatial
resolution of the tropics,

645
00:31:29,086 --> 00:31:32,323
because all the plants are very
different from one another,

646
00:31:32,356 --> 00:31:34,225
so we need to
really drill down.

647
00:31:35,125 --> 00:31:37,461
Another important
aspect about the tropics

648
00:31:37,494 --> 00:31:40,231
is that not only are they
very sensitive to climate,

649
00:31:40,264 --> 00:31:43,367
which I've shown, but they're
also very sensitive to CO2.

650
00:31:43,400 --> 00:31:45,069
If you remember that the plot

651
00:31:45,102 --> 00:31:50,141
of the ecosystem crashing,
that was due to the droughts.

652
00:31:50,174 --> 00:31:52,476
But some of them that
were doing really well,

653
00:31:52,509 --> 00:31:54,245
and that was
because those models

654
00:31:54,278 --> 00:31:56,080
were very responsive to CO2.

655
00:31:56,113 --> 00:31:57,648
CO2 being good for plants.

656
00:31:57,681 --> 00:32:01,319
Droughts of course related
to CO2 being bad for plants.

657
00:32:02,419 --> 00:32:04,822
So we need to understand
how these plants,

658
00:32:04,855 --> 00:32:07,792
how these ecosystems, are
going to respond in the future

659
00:32:07,825 --> 00:32:10,027
also to rising CO2.

660
00:32:10,060 --> 00:32:11,429
So what do we do?

661
00:32:11,462 --> 00:32:16,334
We conduct experiments where
we pump CO2 onto ecosystems

662
00:32:16,367 --> 00:32:19,237
and see how they respond.

663
00:32:20,838 --> 00:32:23,341
This is very difficult to do.

664
00:32:23,374 --> 00:32:27,678
As you can imagine it requires
a lot of import of CO2,

665
00:32:27,711 --> 00:32:31,248
expensive infrastructure,
and it costs a lot of money,

666
00:32:31,281 --> 00:32:32,249
tens of millions of dollars,

667
00:32:32,282 --> 00:32:34,552
and you don't actually
get a lot of trees.

668
00:32:38,288 --> 00:32:42,393
More importantly what you
also don't get is this long,

669
00:32:42,426 --> 00:32:44,095
we're interested in
the next century,

670
00:32:44,128 --> 00:32:47,832
over the next 10 years,
20 years, 50 years,

671
00:32:47,865 --> 00:32:50,434
and we can't run these
experiments for that long.

672
00:32:50,467 --> 00:32:53,971
So we don't know how the
ecosystems might be shifting,

673
00:32:54,004 --> 00:32:57,608
adapting, to this
longterm change in CO2.

674
00:32:59,443 --> 00:33:02,513
So these CO2
enrichment experiments

675
00:33:02,546 --> 00:33:04,615
have been incredibly value.

676
00:33:04,648 --> 00:33:07,985
In fact, kind of one of our
only tools to assess this.

677
00:33:08,018 --> 00:33:11,022
It's still limited,
especially in the tropics

678
00:33:11,055 --> 00:33:13,691
because there's not a
single one in the tropics

679
00:33:13,724 --> 00:33:16,027
and this longterm aspect.

680
00:33:16,060 --> 00:33:17,695
So we've been
really stuck there.

681
00:33:18,695 --> 00:33:21,366
With kind of no solution,
we've been just stuck.

682
00:33:22,533 --> 00:33:24,201
Something new that
I've been exploring

683
00:33:24,234 --> 00:33:26,470
in the past couple of
years and spinning up

684
00:33:26,503 --> 00:33:27,872
is an entirely different field.

685
00:33:27,905 --> 00:33:30,141
You thought you were coming
for an ecosystems talk,

686
00:33:30,174 --> 00:33:32,010
I'm gonna talk about
volcanology now.

687
00:33:36,280 --> 00:33:37,715
We tend to think of volcanoes

688
00:33:37,748 --> 00:33:40,918
as that thing in the background

689
00:33:40,951 --> 00:33:44,255
spewing out lava,
or whatever it does,

690
00:33:44,288 --> 00:33:48,626
but the volcano complex
actually extends well in

691
00:33:48,659 --> 00:33:51,562
to the surrounding forest.

692
00:33:53,530 --> 00:33:55,166
So if you think about
a landscape of trees,

693
00:33:55,199 --> 00:33:58,836
when a volcano forms there's
these cracks and fissures

694
00:33:58,869 --> 00:34:02,873
in the Earth's surface,
and the number one dry gas

695
00:34:02,906 --> 00:34:05,543
that comes out of
volcanoes is CO2,

696
00:34:05,576 --> 00:34:07,478
and it comes out in
very high amounts.

697
00:34:09,113 --> 00:34:12,550
Then it diffuses
into the ecosystems

698
00:34:12,583 --> 00:34:15,619
until it reaches
background conditions.

699
00:34:15,652 --> 00:34:19,290
These are exactly the
concentrations that we

700
00:34:19,323 --> 00:34:24,295
expect to see globally
over the next 50-100 years.

701
00:34:24,862 --> 00:34:29,100
And this has been occurring
since geologic age,

702
00:34:29,133 --> 00:34:30,668
hundreds of years.

703
00:34:30,701 --> 00:34:33,704
So we actually kind
of now have a natural

704
00:34:33,737 --> 00:34:37,274
longterm CO2
experiment given to us

705
00:34:37,307 --> 00:34:39,243
by Mother Nature
through the volcanoes.

706
00:34:41,545 --> 00:34:44,381
So I kind of stole this image

707
00:34:44,414 --> 00:34:46,550
from the gravitational
wave guys.

708
00:34:46,583 --> 00:34:48,519
[laughing]

709
00:34:48,552 --> 00:34:50,387
We now are taking
two communities

710
00:34:50,420 --> 00:34:51,655
that normally actually

711
00:34:51,688 --> 00:34:53,691
don't ever interact,
the ecology community

712
00:34:53,724 --> 00:34:56,527
and volcanology community, and
we're putting them together,

713
00:34:56,560 --> 00:34:59,864
and we think this can
really radically change

714
00:34:59,897 --> 00:35:03,134
our knowledge of ecosystems,

715
00:35:03,167 --> 00:35:05,302
and actually
volcanology as well.

716
00:35:05,335 --> 00:35:06,871
I think I have a slide on that.

717
00:35:07,938 --> 00:35:08,906
One of the first things we did,

718
00:35:08,939 --> 00:35:11,008
we went to our
backyard, Mammoth.

719
00:35:11,041 --> 00:35:14,011
A lot of people don't realize
that it's an active volcano.

720
00:35:14,945 --> 00:35:16,147
I never realized that.

721
00:35:16,180 --> 00:35:18,482
I realized it was a good
place to go snowboarding

722
00:35:18,515 --> 00:35:20,684
if I wanted to spend
a lot of money.

723
00:35:20,717 --> 00:35:23,187
[laughing]

724
00:35:23,220 --> 00:35:25,556
So what we did is
we ran our aircraft

725
00:35:25,589 --> 00:35:29,093
over it to look at the
biomass and the chemistry

726
00:35:29,126 --> 00:35:31,328
and the greenness
and the evaporation,

727
00:35:31,361 --> 00:35:33,964
and we saw these
really tight signals

728
00:35:33,997 --> 00:35:38,569
across the landscape to when
there was increasing CO2

729
00:35:38,602 --> 00:35:42,473
we saw direct changes in
the ecosystem structure.

730
00:35:42,506 --> 00:35:46,710
We got the CO2 because the
USGS had mapped on the ground

731
00:35:46,743 --> 00:35:47,745
all the CO2 in the area,

732
00:35:47,778 --> 00:35:49,947
you know, it's in
their backyard as well.

733
00:35:51,615 --> 00:35:53,851
So that's great, and
that was really exciting.

734
00:35:53,884 --> 00:35:55,653
But what we really need
are the tropics, right?

735
00:35:55,686 --> 00:35:57,621
That's the breathing
of the biosphere,

736
00:35:57,654 --> 00:36:02,093
and we don't have the USGS
in the tropics necessarily.

737
00:36:03,694 --> 00:36:07,097
What we've discovered is
there's this chain of volcanoes

738
00:36:07,130 --> 00:36:11,235
in Costa Rica that
has been emitting CO2

739
00:36:11,268 --> 00:36:14,572
into the rainforest
for hundreds of years

740
00:36:14,605 --> 00:36:16,373
at different amounts,

741
00:36:16,406 --> 00:36:18,809
kind of giving us a
nice experimental setup.

742
00:36:20,010 --> 00:36:25,016
We went into these volcanic
jungles a few years ago

743
00:36:25,415 --> 00:36:29,320
to examine and figure out on
the ground what was happening.

744
00:36:29,353 --> 00:36:32,590
So we really wanted to
figure out was there

745
00:36:32,623 --> 00:36:35,593
a window into the
future of the Earth

746
00:36:35,626 --> 00:36:39,663
hidden in the jungles of
Costa Rica's volcanoes.

747
00:36:39,696 --> 00:36:42,833
Here we are taking a
lot of measurements.

748
00:36:42,866 --> 00:36:45,336
So this is a picture to
show we came out alive.

749
00:36:48,372 --> 00:36:51,742
What we found were, again,
clear and direct signals

750
00:36:51,775 --> 00:36:54,912
that when there was
greater CO2 exposure

751
00:36:54,945 --> 00:36:56,814
the plants were changing.

752
00:36:59,283 --> 00:37:01,885
What's also particularly
interesting, at least to me,

753
00:37:01,918 --> 00:37:03,787
is that the CO2 coming
out of volcanoes

754
00:37:03,820 --> 00:37:06,090
is very different
than the CO2 in air.

755
00:37:06,123 --> 00:37:08,959
It has a different what we
call isotopic signature.

756
00:37:08,992 --> 00:37:12,129
It's a different kind of
chemical signature if you will.

757
00:37:12,162 --> 00:37:14,531
So trees when they breathe
in the volcanic CO2,

758
00:37:14,564 --> 00:37:17,501
it becomes part of
their chemistry.

759
00:37:17,534 --> 00:37:21,405
So trees in these systems
are actually made up

760
00:37:21,438 --> 00:37:25,175
of volcanic CO2 when
they're more exposed.

761
00:37:25,208 --> 00:37:29,079
So they actually keep longterm
records of volcanic CO2

762
00:37:29,112 --> 00:37:30,648
in their wood.

763
00:37:33,317 --> 00:37:35,686
I mentioned that we didn't
have the USGS in there.

764
00:37:35,719 --> 00:37:39,857
What we've been exploring is
we've been developing drones,

765
00:37:39,890 --> 00:37:43,560
a partnership with a
company called Black Swift,

766
00:37:43,593 --> 00:37:45,596
and they have these
drones that can fly

767
00:37:45,629 --> 00:37:49,667
close to the tree tops
and sniff out the CO2

768
00:37:49,700 --> 00:37:51,702
where it's leaking
out of the landscape,

769
00:37:52,636 --> 00:37:55,072
thereby mapping out the CO2

770
00:37:55,105 --> 00:37:58,142
of the larger landscape.

771
00:38:02,879 --> 00:38:05,983
This is also really valuable
not just to me as an ecologist,

772
00:38:06,016 --> 00:38:08,352
but to the volcanologist.

773
00:38:08,385 --> 00:38:11,722
Volcanic CO2 is
an early indicator

774
00:38:11,755 --> 00:38:13,724
of volcanic activity
or eruption.

775
00:38:15,158 --> 00:38:20,097
But they cannot monitor
CO2 on every volcano.

776
00:38:21,098 --> 00:38:23,367
Even in the tropics
they have CO2 monitors

777
00:38:23,400 --> 00:38:24,735
and they get easily damaged.

778
00:38:24,768 --> 00:38:27,504
But if the trees can
act as those sensors,

779
00:38:27,537 --> 00:38:28,872
and not just a
couple of sensors

780
00:38:28,905 --> 00:38:30,574
but thousands of sensors

781
00:38:30,607 --> 00:38:32,309
telling us what the
volcanoes are doing,

782
00:38:32,342 --> 00:38:35,079
this is potentially
a major breakthrough

783
00:38:35,112 --> 00:38:36,980
for the volcanology community.

784
00:38:37,013 --> 00:38:38,782
And I'm not just saying
this as an outsider,

785
00:38:38,815 --> 00:38:41,819
this is words coming from
my volcanology colleagues.

786
00:38:44,888 --> 00:38:48,592
All right, so I've talked
a lot about the tropics,

787
00:38:48,625 --> 00:38:51,228
and that's where a lot
of my interests lie,

788
00:38:51,261 --> 00:38:53,564
but I would remiss not
to talk about the Arctic,

789
00:38:53,597 --> 00:38:55,032
two poles of the carbon cycle.

790
00:38:55,065 --> 00:38:57,801
I'm not gonna talk too much
about it for the sake of time,

791
00:38:57,834 --> 00:39:01,071
but these same
type of techniques,

792
00:39:01,104 --> 00:39:03,741
you know, we've got a
lot of carbon locked up,

793
00:39:03,774 --> 00:39:06,777
methane locked up
that's being released.

794
00:39:06,810 --> 00:39:09,113
The models of course
are all over the place.

795
00:39:09,146 --> 00:39:11,515
This was a paper I
put out on Alaska,

796
00:39:11,548 --> 00:39:12,583
and you can see
different colors

797
00:39:12,616 --> 00:39:15,486
which is basically
each of these Alaskas

798
00:39:15,519 --> 00:39:16,620
are different models.

799
00:39:16,653 --> 00:39:18,222
The point of this
is that almost every

800
00:39:18,255 --> 00:39:20,190
color combination is shown.

801
00:39:20,223 --> 00:39:21,759
It's like a giant
game of Twister

802
00:39:21,792 --> 00:39:23,193
where the models
completely disagree

803
00:39:23,226 --> 00:39:25,896
as to what's going
on in the Arctic.

804
00:39:25,929 --> 00:39:29,933
So NASA's launched this
giant almost 10 year campaign

805
00:39:29,966 --> 00:39:33,137
to really tackle the Arctic,
and I'm a part of that.

806
00:39:33,170 --> 00:39:37,141
I helped write the
study and I'm a PI

807
00:39:37,174 --> 00:39:38,743
on one of the
projects for that.

808
00:39:40,010 --> 00:39:42,112
These airborne capabilities
that I talked about,

809
00:39:42,145 --> 00:39:45,816
we've just flown for the
last two summers in Canada

810
00:39:45,849 --> 00:39:49,620
and Alaska really trying to
tackle the ecosystem responses

811
00:39:49,653 --> 00:39:54,658
to warming and other climate
impacts, but in the Arctic.

812
00:39:57,928 --> 00:40:01,098
Back to this CHANGE diagram.

813
00:40:01,131 --> 00:40:04,334
I mentioned a lot about
the CO2 fertilization.

814
00:40:04,367 --> 00:40:05,803
Plants love CO2.

815
00:40:05,836 --> 00:40:08,872
They'll take up more CO2
if you give them more CO2,

816
00:40:08,905 --> 00:40:13,444
but they also need water
and light and nutrients.

817
00:40:13,477 --> 00:40:15,979
So nutrients is
kind of one of those

818
00:40:16,012 --> 00:40:17,281
things we forget
about with plants.

819
00:40:17,314 --> 00:40:18,882
We always think
to water plants,

820
00:40:18,915 --> 00:40:20,851
and of course they need
light, but nutrients.

821
00:40:20,884 --> 00:40:22,686
And that's the same
with climate models,

822
00:40:22,719 --> 00:40:25,622
it's one of those last things
we've developed in models;

823
00:40:25,655 --> 00:40:27,858
there's a lot to be developed.

824
00:40:27,891 --> 00:40:29,993
So we look at that
CO2 rise again.

825
00:40:30,026 --> 00:40:31,295
What does this mean in terms of

826
00:40:31,328 --> 00:40:33,931
the equation for
photosynthesis?

827
00:40:33,964 --> 00:40:36,667
CO2 being right there
with water and energy

828
00:40:36,700 --> 00:40:41,172
creating oxygen and the CH2O
is our sugars, or our wood.

829
00:40:42,138 --> 00:40:44,575
So as CO2 goes up, does
that mean our trees

830
00:40:44,608 --> 00:40:46,343
are continuing to get
bigger and bigger?

831
00:40:46,376 --> 00:40:51,081
So big that they basically
take over our houses?

832
00:40:51,114 --> 00:40:54,485
I mean there's gotta be a limit
to how big these trees get.

833
00:40:55,452 --> 00:40:58,055
The climate models
had originally not had

834
00:40:58,088 --> 00:40:59,923
nutrient limitations in them.

835
00:40:59,956 --> 00:41:03,827
So they were projecting too
much CO2 being taken up.

836
00:41:03,860 --> 00:41:07,464
Plants were too happy
in some of the models.

837
00:41:07,497 --> 00:41:09,032
So we started looking at this.

838
00:41:09,065 --> 00:41:13,237
When we put this nitrogen
being one of the most

839
00:41:13,270 --> 00:41:15,539
limiting nutrients for
plants in the models,

840
00:41:15,572 --> 00:41:17,741
it left a fundamentally
altered behavior.

841
00:41:17,774 --> 00:41:20,244
I surveyed my colleagues
around the world as to

842
00:41:20,277 --> 00:41:23,080
what processes they were
gonna put in their models

843
00:41:23,113 --> 00:41:24,915
over the next five years.

844
00:41:24,948 --> 00:41:29,887
I sized their response by the
frequency of the response.

845
00:41:29,920 --> 00:41:34,492
The most common response
across the world was nitrogen.

846
00:41:36,826 --> 00:41:39,563
I've been developing a lot of
the nitrogen modeling myself.

847
00:41:39,596 --> 00:41:43,500
There's a lot of
mathematics here.

848
00:41:43,533 --> 00:41:45,802
The most important
thing with this model

849
00:41:45,835 --> 00:41:47,504
that you should take
from this is that

850
00:41:47,537 --> 00:41:50,240
I gave it a really cool
name which was FUN,

851
00:41:50,273 --> 00:41:51,909
which is Fixation and
Uptake of Nitrogen.

852
00:41:51,942 --> 00:41:54,178
Some I'm literally
putting fun into modeling.

853
00:41:56,513 --> 00:41:58,081
Now, again we need these

854
00:41:58,114 --> 00:42:00,717
observational constraints
to the models.

855
00:42:00,750 --> 00:42:03,253
One thing that we've
discovered is that

856
00:42:03,286 --> 00:42:06,423
there's this underground
economy associated with plants,

857
00:42:06,456 --> 00:42:08,959
associated with fungi,

858
00:42:08,992 --> 00:42:10,227
mycorrhizal fungi
they're called,

859
00:42:10,260 --> 00:42:13,030
and there's two
dominant types of fungi

860
00:42:13,063 --> 00:42:14,532
that associate with plants,

861
00:42:15,532 --> 00:42:17,067
and they're scattered
throughout the landscape.

862
00:42:17,100 --> 00:42:19,836
Some tree species are
associated with one,

863
00:42:19,869 --> 00:42:22,773
AM, arbuscular, and the
others are associated

864
00:42:22,806 --> 00:42:26,176
with the other, ECM,
ectomycorrhizal.

865
00:42:27,277 --> 00:42:31,181
These fungi go out and get
nutrients for the plants.

866
00:42:31,214 --> 00:42:35,018
In exchange, the plants
pay them in sugar.

867
00:42:35,051 --> 00:42:37,621
It's a carbon economy
for the nutrients.

868
00:42:38,588 --> 00:42:40,891
Now as you can imagine,

869
00:42:40,924 --> 00:42:43,460
some fungi charge
more than the others,

870
00:42:43,493 --> 00:42:46,163
and some plants pay
more than the others.

871
00:42:46,196 --> 00:42:49,499
So we need to understand
this below ground economy

872
00:42:49,532 --> 00:42:51,935
to really understand
the nutrient constraints

873
00:42:51,968 --> 00:42:54,104
and functioning
of the ecosystems.

874
00:42:54,137 --> 00:42:57,341
So what we've learned is that

875
00:42:57,374 --> 00:43:00,944
these fungi actually
kind of pulse the trees

876
00:43:00,977 --> 00:43:03,313
in different ways
that are visible

877
00:43:03,346 --> 00:43:07,117
to some of the instruments
that we can see from space.

878
00:43:07,150 --> 00:43:10,621
So if we were able to see each
tree species individually,

879
00:43:10,654 --> 00:43:12,589
we'd know which fungi
was associated with it,

880
00:43:12,622 --> 00:43:16,026
but we can't, we can't
see each tree like that.

881
00:43:16,926 --> 00:43:18,261
What we've
found is that

882
00:43:18,294 --> 00:43:19,963
instead of looking
at each tree,

883
00:43:19,996 --> 00:43:24,568
we look at groups of trees
that respond more similarly

884
00:43:24,601 --> 00:43:27,471
to each other than
other groups of trees.

885
00:43:27,504 --> 00:43:31,742
The fungi act like hands to
the trees like they're puppets.

886
00:43:34,244 --> 00:43:36,546
And of course we've gone out
and done a lot of field work.

887
00:43:36,579 --> 00:43:39,516
This is at the top
of the Andes in Peru

888
00:43:39,549 --> 00:43:42,486
looking out into the Amazon
basin in the cloud forest.

889
00:43:42,519 --> 00:43:44,955
I spent four years
in there conducting

890
00:43:44,988 --> 00:43:46,890
a fertilization experiment,

891
00:43:46,923 --> 00:43:48,926
putting nitrogen
and phosphorus down,

892
00:43:50,160 --> 00:43:52,562
and collecting soils and leaves

893
00:43:52,595 --> 00:43:55,966
to really test a lot of
these models and hypotheses.

894
00:43:57,333 --> 00:44:02,072
We're now able to really use
a lot of the remote sensing,

895
00:44:02,105 --> 00:44:06,810
airborne, and ground data to
constrain the nutrient aspects.

896
00:44:06,843 --> 00:44:08,745
We've put these into
the climate models,

897
00:44:08,778 --> 00:44:12,182
so now we have global models
that are now constrained

898
00:44:12,215 --> 00:44:14,951
by nutrients and fungi.

899
00:44:14,984 --> 00:44:18,689
You wouldn't think that
we could see and even care

900
00:44:18,722 --> 00:44:22,192
about fungi when we're
talking about ecosystems

901
00:44:22,225 --> 00:44:25,896
or the global carbon cycle.

902
00:44:30,133 --> 00:44:32,369
Finally, the last
part of the triangle

903
00:44:32,402 --> 00:44:35,806
I'll talk a little
bit about is water.

904
00:44:35,839 --> 00:44:40,343
So we've been interested
in climate and CO2

905
00:44:40,376 --> 00:44:41,779
and ecosystem response,

906
00:44:43,012 --> 00:44:46,983
but at the end of the day
we're also concerned about us,

907
00:44:47,016 --> 00:44:50,220
people, and what
we need is water,

908
00:44:50,253 --> 00:44:51,955
especially in California,

909
00:44:51,988 --> 00:44:54,091
and we need certainty
in our water,

910
00:44:54,124 --> 00:44:56,460
especially as the
water cycle's changing.

911
00:44:56,493 --> 00:44:58,428
This is all related to climate,

912
00:44:58,461 --> 00:45:00,997
but we need to just
know how much water

913
00:45:01,030 --> 00:45:03,700
we're gonna have and
can we grow our crops,

914
00:45:03,733 --> 00:45:05,936
and when can we thrive
as civilization.

915
00:45:06,803 --> 00:45:11,274
We have this incredible
uncertainty in our water cycle.

916
00:45:11,307 --> 00:45:14,211
The water cycle being
made up of rain and snow

917
00:45:14,244 --> 00:45:15,813
and evapotranspiration.

918
00:45:16,980 --> 00:45:19,282
The surface soil
moisture leading to

919
00:45:19,315 --> 00:45:22,686
the deep ground water storage,
that's that triangle S,

920
00:45:22,719 --> 00:45:24,555
delta S, and the runoff.

921
00:45:25,555 --> 00:45:28,258
So our uncertainty in
our water availability

922
00:45:28,291 --> 00:45:31,528
is tied to each of these
components in the water cycle.

923
00:45:33,029 --> 00:45:36,299
We are now at an age in NASA

924
00:45:36,332 --> 00:45:38,902
and the space community
internationally

925
00:45:38,935 --> 00:45:40,170
where we can observe

926
00:45:40,203 --> 00:45:45,209
every single component of
the water cycle from space,

927
00:45:45,508 --> 00:45:46,810
or about to be from space.

928
00:45:47,911 --> 00:45:49,446
I mentioned that TRMM,

929
00:45:49,479 --> 00:45:51,114
the Tropical Rainfall
Measuring Mission,

930
00:45:51,147 --> 00:45:52,482
that has since died.

931
00:45:52,515 --> 00:45:53,516
We have now launched

932
00:45:53,550 --> 00:45:56,686
a Global Precipitation
Measurement Constellation

933
00:45:56,719 --> 00:46:00,557
in collaboration with space
agencies around the world.

934
00:46:03,259 --> 00:46:06,463
We have some ability
to get snow from space.

935
00:46:06,496 --> 00:46:07,798
We're still working on that.

936
00:46:07,831 --> 00:46:11,802
But JPL has been running an
Airborne Snow Observatory, ASO,

937
00:46:13,136 --> 00:46:15,772
which uses LIDAR, those lasers,

938
00:46:15,805 --> 00:46:18,909
to see how thick
the snowpack is

939
00:46:18,942 --> 00:46:21,144
and also how dark
the snowpack is

940
00:46:21,177 --> 00:46:23,079
because that determines
the melt rate;

941
00:46:23,112 --> 00:46:25,849
darker snow melting
faster than brighter snow.

942
00:46:27,016 --> 00:46:29,319
I've talked a lot about
evapotranspiration.

943
00:46:29,352 --> 00:46:31,421
I won't talk too
much more about that,

944
00:46:31,454 --> 00:46:36,059
but ECOSTRESS is gonna
provide a major breakthrough

945
00:46:36,092 --> 00:46:39,663
in our ability to transcend
scales across the globe.

946
00:46:42,131 --> 00:46:44,835
JPL also launched the SMAP,

947
00:46:44,868 --> 00:46:47,070
the Soil Moisture
Active Passive Mission.

948
00:46:47,103 --> 00:46:48,338
Have you guys heard
of that at all?

949
00:46:48,371 --> 00:46:50,340
It's been out for a
couple of years now.

950
00:46:53,509 --> 00:46:56,146
I used to tell kids
at the open house that

951
00:46:56,179 --> 00:46:58,882
with the active part,
which doesn't work anymore,

952
00:46:58,915 --> 00:47:01,484
it basically is like a
giant finger from the sky

953
00:47:01,517 --> 00:47:03,720
that goes and sticks
its finger in the mud

954
00:47:03,753 --> 00:47:05,989
and comes back and
says, it's this wet.

955
00:47:07,156 --> 00:47:09,292
Now that part doesn't work,

956
00:47:09,325 --> 00:47:10,927
but we have this
passive measurement

957
00:47:10,960 --> 00:47:12,395
that uses light
reflectants to tell us

958
00:47:12,428 --> 00:47:14,264
how much moisture
is in the soil.

959
00:47:17,467 --> 00:47:19,970
There's another
satellite called GRACE.

960
00:47:21,004 --> 00:47:23,607
My colleague Felix Landerer
gave a von Karman talk

961
00:47:23,640 --> 00:47:26,176
I think a few months
ago or some weeks ago

962
00:47:26,209 --> 00:47:29,746
on GRACE Follow-On, which is
continuing this amazing record.

963
00:47:29,779 --> 00:47:33,516
GRACE uses a gravitational
anomaly where

964
00:47:33,549 --> 00:47:36,620
these satellites kind of
get pulled towards the Earth

965
00:47:36,653 --> 00:47:37,921
when there's more mass, right,

966
00:47:37,954 --> 00:47:40,457
'cause more mass
equals more gravity.

967
00:47:40,490 --> 00:47:43,059
So when you have a
lot of groundwater,

968
00:47:43,092 --> 00:47:44,861
it'll get pulled more
towards the Earth.

969
00:47:44,894 --> 00:47:46,763
And if you've sucked
out all the groundwater

970
00:47:46,796 --> 00:47:48,865
'cause you pumped it
out or there's droughts,

971
00:47:48,898 --> 00:47:51,568
next time they fly over
they get pulled less.

972
00:47:51,601 --> 00:47:53,670
So we can actually use
this gravitational anomaly

973
00:47:53,703 --> 00:47:56,006
to figure out how much water
there is in the ground.

974
00:47:56,039 --> 00:48:00,777
They also use this to look at
ice caps and sea level rise.

975
00:48:03,279 --> 00:48:04,981
Then the last component
is river discharge,

976
00:48:05,014 --> 00:48:06,383
how much water's
coming off the rivers.

977
00:48:06,416 --> 00:48:08,218
We actually don't have
this from space yet,

978
00:48:08,251 --> 00:48:09,819
but it will be soon.

979
00:48:09,852 --> 00:48:13,256
The Surface Water Ocean
Topography Mission, SWOT,

980
00:48:13,289 --> 00:48:15,292
again uses those
lasers, the LIDAR,

981
00:48:15,325 --> 00:48:18,361
to measure how tall
those rivers are

982
00:48:18,394 --> 00:48:20,797
and that tells you how
much water's flowing off.

983
00:48:21,898 --> 00:48:23,800
As we start to put
these together,

984
00:48:25,034 --> 00:48:28,805
our uncertainty in the water
cycle starts to come down.

985
00:48:30,840 --> 00:48:32,208
We would never have
zero uncertainty,

986
00:48:32,241 --> 00:48:34,912
but since it's like a cartoon,
I made it come to zero,

987
00:48:36,045 --> 00:48:37,915
but you can get
where this is going.

988
00:48:39,015 --> 00:48:41,685
What's next now is
for us to integrate.

989
00:48:41,718 --> 00:48:45,655
It's about integration and
talking across the missions.

990
00:48:45,688 --> 00:48:48,825
Just like in the International
Space Station those missions,

991
00:48:48,858 --> 00:48:52,262
same with our other missions
that have synergies together

992
00:48:52,295 --> 00:48:54,764
to tell us something about
the Earth's system as a whole

993
00:48:54,797 --> 00:48:56,733
more than the sum of its parts.

994
00:48:57,800 --> 00:49:02,105
So there's been a lot of
droughts worldwide in the US,

995
00:49:02,138 --> 00:49:06,109
in California, affecting
crops, agriculture,

996
00:49:06,142 --> 00:49:08,778
and we were able to pick up
on one of the biggest droughts

997
00:49:08,811 --> 00:49:12,182
in US history
since the Dust Bowl

998
00:49:12,215 --> 00:49:14,284
which happened a few years ago

999
00:49:14,317 --> 00:49:16,286
creating this kind
of bullseye right on

1000
00:49:17,587 --> 00:49:21,725
the Midwest where
almost 80% of our GDP

1001
00:49:21,758 --> 00:49:23,794
was impacted by the drought.

1002
00:49:28,131 --> 00:49:32,502
NASA has stood up a Western
Water Applications Office

1003
00:49:32,535 --> 00:49:37,173
which has a number of
personnel from JPL housed here,

1004
00:49:37,206 --> 00:49:41,111
and we're reaching out to
water resource managers,

1005
00:49:41,144 --> 00:49:46,116
policy makers, farmers,
and trying to make our data

1006
00:49:46,549 --> 00:49:49,953
useful for them in their
decision making process.

1007
00:49:49,986 --> 00:49:52,923
So we're developing
web applications

1008
00:49:55,458 --> 00:49:57,060
and phone apps,

1009
00:49:57,093 --> 00:50:00,263
there's a lot of citizen
science at NASA as well,

1010
00:50:00,296 --> 00:50:04,100
to really enable
society to respond

1011
00:50:04,133 --> 00:50:05,902
to a changing
environment,

1012
00:50:05,935 --> 00:50:07,404
especially with
regards to water.

1013
00:50:11,474 --> 00:50:15,879
So that's a little bit
of what I do here at JPL

1014
00:50:15,912 --> 00:50:20,383
and what my colleagues do here
at JPL and across the world.

1015
00:50:20,416 --> 00:50:25,389
Again, we end with this
fairy tale or this story of

1016
00:50:25,588 --> 00:50:27,257
the princess and the pea.

1017
00:50:27,290 --> 00:50:30,260
We are trying to establish
the true identity,

1018
00:50:30,293 --> 00:50:32,729
the true future of the Earth.

1019
00:50:32,762 --> 00:50:37,333
How are ecosystems
responding in reality

1020
00:50:37,366 --> 00:50:39,002
now and into the future?

1021
00:50:41,671 --> 00:50:43,940
These are some of the take-home
points that I hope you

1022
00:50:43,973 --> 00:50:46,476
got from the talk.

1023
00:50:46,509 --> 00:50:49,245
Terrestrial ecosystems
exert this dominant force

1024
00:50:49,278 --> 00:50:52,415
in Earth's climate, and
they're very complex,

1025
00:50:52,448 --> 00:50:54,517
and that's why we have
a lot of uncertainties

1026
00:50:54,550 --> 00:50:57,020
in projections of
future responses.

1027
00:50:58,087 --> 00:51:02,158
The projected climate change,
particularly in droughts,

1028
00:51:02,191 --> 00:51:04,394
is being borne out
in front of our eyes.

1029
00:51:05,928 --> 00:51:09,666
CO2 fertilization sensitivity
is a major uncertainty

1030
00:51:09,699 --> 00:51:11,968
in our understanding of
ecosystem future responses,

1031
00:51:12,001 --> 00:51:15,004
and we've been developing
innovative technological

1032
00:51:15,037 --> 00:51:17,140
and interdisciplinary
ways to tackle

1033
00:51:17,173 --> 00:51:19,109
this part of the
equation as well.

1034
00:51:20,343 --> 00:51:24,147
NASA, and satellite and
airborne remote sensing

1035
00:51:24,180 --> 00:51:26,683
in general, provides a
deeper understanding

1036
00:51:26,716 --> 00:51:29,953
of ecosystem responses
across the Earth,

1037
00:51:29,986 --> 00:51:32,655
and it enables this reduction
of model uncertainties,

1038
00:51:32,688 --> 00:51:37,594
and can help us improve
societal responses to change.

1039
00:51:37,627 --> 00:51:39,629
Thank you.

1040
00:51:39,662 --> 00:51:43,100
[applauding]

1041
00:51:47,937 --> 00:51:50,707
And I'm happy to
take any questions.

1042
00:51:50,740 --> 00:51:52,041
I think we have
time for questions.

1043
00:51:52,074 --> 00:51:56,112
So if you have any questions,
there's a microphone there,

1044
00:51:56,145 --> 00:51:58,148
'cause they want, they told
me to tell you guys to use

1045
00:51:58,181 --> 00:52:00,049
the microphone.
- Okay.

1046
00:52:00,082 --> 00:52:03,887
- So go line up at the
microphone and pepper away.

1047
00:52:08,724 --> 00:52:10,193
- First, thank you
so much for this.

1048
00:52:10,226 --> 00:52:12,962
I trained with the
former Vice President,

1049
00:52:12,995 --> 00:52:14,531
the honorable Al Gore.

1050
00:52:14,564 --> 00:52:16,332
I'm a climate
reality volunteer.

1051
00:52:16,365 --> 00:52:17,534
- [Josh] Yeah, me too.

1052
00:52:17,567 --> 00:52:20,203
- Fantastic, where's your pin?

1053
00:52:20,236 --> 00:52:22,872
But question is when you
mentioned the lasers,

1054
00:52:22,905 --> 00:52:27,177
can they now measure carbon
emissions in the ice?

1055
00:52:27,210 --> 00:52:28,545
- In the ice?
- Yes.

1056
00:52:31,447 --> 00:52:34,317
- So as carbon
gets released from.

1057
00:52:34,350 --> 00:52:38,354
So, yeah, so it's not using
the lasers or the LIDAR,

1058
00:52:38,387 --> 00:52:42,258
we have other airborne
capabilities that can take,

1059
00:52:42,291 --> 00:52:44,561
that we fly over and
can take air samples,

1060
00:52:44,594 --> 00:52:48,665
and then we measure the
CO2 from those platforms.

1061
00:52:48,698 --> 00:52:52,368
OCO-2 and three and
related satellites

1062
00:52:52,401 --> 00:52:56,372
also can measure the
CO2 in the atmosphere.

1063
00:52:56,405 --> 00:52:59,042
Instead of using lasers,
they use the sun essentially.

1064
00:52:59,075 --> 00:53:01,878
So the sun will bounce
down, hit the Earth,

1065
00:53:01,911 --> 00:53:03,646
and bounce back
through the atmosphere.

1066
00:53:03,679 --> 00:53:06,883
But CO2 absorbs
some of that light,

1067
00:53:06,916 --> 00:53:08,551
so when there's more CO2,

1068
00:53:08,584 --> 00:53:11,621
there's less of that light
hitting the satellite.

1069
00:53:11,654 --> 00:53:13,423
So that's how we get the CO2,

1070
00:53:13,456 --> 00:53:15,792
not just in the
Arctic, but globally.

1071
00:53:15,825 --> 00:53:17,060
- All right, thank
you very much.

1072
00:53:17,093 --> 00:53:18,361
- Did that answer you
question a little bit?

1073
00:53:18,394 --> 00:53:19,729
- Appreciate it, thank you.

1074
00:53:21,831 --> 00:53:26,669
- Hello, are there any
areas in particular

1075
00:53:26,702 --> 00:53:29,939
that are abiotic
that are of interest

1076
00:53:29,972 --> 00:53:34,110
that are not
concerned with biomes?

1077
00:53:36,512 --> 00:53:38,515
- Yeah, so I'm very
much a plant person,

1078
00:53:39,448 --> 00:53:41,417
but I definitely
work with geologist,

1079
00:53:41,450 --> 00:53:44,887
so for instance the
volcanologists in that aspect.

1080
00:53:44,920 --> 00:53:48,324
So my interest in the
abiotic land component

1081
00:53:48,357 --> 00:53:49,459
has been on that.

1082
00:53:49,492 --> 00:53:51,194
I'm also interested in soils,

1083
00:53:51,227 --> 00:53:54,430
but there's always somewhat
of an interaction with plants

1084
00:53:54,463 --> 00:53:56,633
when we're talking about soils.

1085
00:53:58,634 --> 00:54:01,938
But in terms of areas
like deserts or so on

1086
00:54:01,971 --> 00:54:06,743
where there are no plants,
again, there are plants

1087
00:54:06,776 --> 00:54:09,246
kind of everywhere,
even in those deserts.

1088
00:54:10,646 --> 00:54:12,649
Sometimes there are
these rain events,

1089
00:54:12,682 --> 00:54:16,085
and these ecosystems are
adapted to really grow

1090
00:54:16,118 --> 00:54:19,155
and take up a lot of CO2 when
there are these rain events,

1091
00:54:19,188 --> 00:54:20,690
and we need to be
able to capture that

1092
00:54:20,723 --> 00:54:21,658
and predict that as well.

1093
00:54:21,691 --> 00:54:23,359
I don't know if that
answered your question.

1094
00:54:23,392 --> 00:54:24,827
Did that answer your question?

1095
00:54:24,860 --> 00:54:27,530
- I think you did, thank you.

1096
00:54:27,563 --> 00:54:28,965
- [Josh] Feel free to come
up and talk to me afterwards

1097
00:54:28,998 --> 00:54:30,834
if you wanna.
- Thank you very much.

1098
00:54:33,669 --> 00:54:35,004
- Excellent
presentation, thank you.

1099
00:54:35,037 --> 00:54:38,508
Question, we're all very
familiar on a daily basis

1100
00:54:38,541 --> 00:54:40,276
with weather forecasts,

1101
00:54:40,309 --> 00:54:43,680
and some of us take note
of when they're right

1102
00:54:43,713 --> 00:54:44,814
and when they're wrong,

1103
00:54:44,848 --> 00:54:47,984
but it seems that at least
going about three days out

1104
00:54:48,017 --> 00:54:49,952
we can feel pretty confident
they got a good idea

1105
00:54:49,985 --> 00:54:51,621
about what's happening,
sometimes they're off.

1106
00:54:51,654 --> 00:54:55,058
But beyond that it diverges
wildly with entropy.

1107
00:54:56,492 --> 00:54:58,294
But that means that
as these models

1108
00:54:58,327 --> 00:55:00,630
in weather forecasting improve,

1109
00:55:01,664 --> 00:55:06,670
we can tell almost immediately
if they're any good, right;

1110
00:55:07,002 --> 00:55:09,038
if they say it's gonna
rain and it doesn't rain,

1111
00:55:09,071 --> 00:55:11,974
or if it's really windy
like it was today.

1112
00:55:12,007 --> 00:55:14,711
Can you give us a
little bit of an idea,

1113
00:55:14,744 --> 00:55:17,280
talk a little bit about,
the timeframes that we use

1114
00:55:17,313 --> 00:55:20,283
both what you're trying to
do as far as forecasting,

1115
00:55:20,316 --> 00:55:22,985
is this about knowing
what it's gonna be like

1116
00:55:23,018 --> 00:55:26,456
one season from now, or just
on a 10 year, 20 year model,

1117
00:55:26,489 --> 00:55:31,494
and how have we
been doing recently

1118
00:55:31,827 --> 00:55:34,230
on how that's changed
our accuracy with that?

1119
00:55:34,263 --> 00:55:35,298
- Yeah, absolutely.

1120
00:55:35,331 --> 00:55:38,167
So I know you must
have a lot of questions

1121
00:55:38,200 --> 00:55:39,101
wrapped up in there,

1122
00:55:39,135 --> 00:55:40,536
but these are exactly
the same questions

1123
00:55:40,569 --> 00:55:42,772
we ask ourselves in the
scientific community.

1124
00:55:42,805 --> 00:55:45,875
We've been looking to the
weather forecasting community

1125
00:55:45,908 --> 00:55:48,878
as an analog, as a guide,

1126
00:55:48,911 --> 00:55:50,813
because it didn't
used to be that way

1127
00:55:50,846 --> 00:55:52,115
with weather forecasting.

1128
00:55:52,148 --> 00:55:55,118
They actually had a rigorous,
systematic development

1129
00:55:55,151 --> 00:55:59,122
of their weather models against
key observable benchmarks,

1130
00:55:59,155 --> 00:56:02,358
and structured scoring systems

1131
00:56:02,391 --> 00:56:04,527
that helped improve
the weather models.

1132
00:56:04,560 --> 00:56:07,764
So we are just coming into that

1133
00:56:09,098 --> 00:56:12,068
kind of ability in
the carbon cycle,

1134
00:56:12,101 --> 00:56:14,170
or the rest of the water cycle,

1135
00:56:14,203 --> 00:56:16,873
just talking about the
carbon cycle for now.

1136
00:56:16,906 --> 00:56:21,911
So we're setting up the
structure to be able to create

1137
00:56:22,344 --> 00:56:26,215
these forecasts that use
these NASA type benchmarks

1138
00:56:26,248 --> 00:56:29,485
that evaluate the
models against them

1139
00:56:29,518 --> 00:56:32,088
and improve their
predictability

1140
00:56:32,121 --> 00:56:34,290
in the same way that
weather models improved

1141
00:56:34,323 --> 00:56:39,295
over their history,
and we have to do that.

1142
00:56:39,428 --> 00:56:41,130
With weather, unlike carbon,

1143
00:56:41,163 --> 00:56:44,767
you do kind of need to know
every day or within the week.

1144
00:56:44,800 --> 00:56:49,806
Carbon cycle you can be fine
with annual time scales.

1145
00:56:50,339 --> 00:56:52,308
So there's this kind
of difference as well.

1146
00:56:52,341 --> 00:56:55,278
Weather models, yes, they break
down over a couple of weeks.

1147
00:56:55,311 --> 00:56:56,512
But if you think about it,

1148
00:56:56,546 --> 00:57:00,183
they're actually still pretty
good at an annual time scale

1149
00:57:00,216 --> 00:57:03,986
'cause they kinda know
if the year will be warm

1150
00:57:04,019 --> 00:57:06,222
or hot in a kind of weird way.

1151
00:57:06,255 --> 00:57:11,194
So there is this kind of
predictability based on

1152
00:57:11,227 --> 00:57:13,963
your interest of your
temporal resolution.

1153
00:57:15,898 --> 00:57:17,733
In terms of climate
projections,

1154
00:57:17,766 --> 00:57:18,901
what we're interested
in are like

1155
00:57:18,934 --> 00:57:22,605
more core scale climate
projections, not every day,

1156
00:57:22,638 --> 00:57:26,409
much more five, 10
year time scales.

1157
00:57:29,144 --> 00:57:31,113
In theory, we could drive
it down even further

1158
00:57:31,146 --> 00:57:33,616
if there was a demand for it.

1159
00:57:33,649 --> 00:57:36,119
- Thanks, follow up?
- Yeah.

1160
00:57:37,286 --> 00:57:41,557
- So naturally policy
makers need to see, hey,

1161
00:57:41,590 --> 00:57:44,694
Fritz said it would be
like this, if you will.

1162
00:57:44,727 --> 00:57:46,362
They need to see,

1163
00:57:46,395 --> 00:57:48,931
build confidence by
seeing accuracy in this.

1164
00:57:48,964 --> 00:57:51,934
So would it be accurate to
say that this is almost,

1165
00:57:51,967 --> 00:57:54,770
'cause it sounds almost
like it's in its infancy,

1166
00:57:54,803 --> 00:57:56,339
not the study of the climate,

1167
00:57:56,372 --> 00:57:58,274
but when you talk about all
these different elements

1168
00:57:58,307 --> 00:57:59,642
of finally having the whole

1169
00:58:00,609 --> 00:58:02,879
geological cycle work together.

1170
00:58:02,912 --> 00:58:05,014
Would it be accurate to say
it's almost in its infancy

1171
00:58:05,047 --> 00:58:06,349
trying to work out models?

1172
00:58:06,382 --> 00:58:08,484
Now that we have so much
more information to work with

1173
00:58:08,517 --> 00:58:10,520
that it might be a
while before we'll have

1174
00:58:10,553 --> 00:58:14,290
a timeline that will give
policy makers what they need?

1175
00:58:14,323 --> 00:58:16,626
- Yeah, I mean
you can see that.

1176
00:58:17,493 --> 00:58:20,129
You can see how much
activity there is, right?

1177
00:58:20,162 --> 00:58:22,832
I showed a lot of
activity I'm part of.

1178
00:58:22,865 --> 00:58:24,433
There's even more
as you can imagine.

1179
00:58:24,466 --> 00:58:26,435
So we're definitely developing.

1180
00:58:26,468 --> 00:58:27,770
If you look at the
history of these models,

1181
00:58:27,803 --> 00:58:29,238
they haven't been
around that long.

1182
00:58:29,271 --> 00:58:32,642
So you can, infancy to some,

1183
00:58:32,675 --> 00:58:35,811
you know a whole career
for somebody else.

1184
00:58:35,844 --> 00:58:38,180
My volcanologist works
in geologic time,

1185
00:58:38,213 --> 00:58:40,617
I work in ecological time.

1186
00:58:42,084 --> 00:58:45,488
There's absolutely, we're
not done developing them.

1187
00:58:45,521 --> 00:58:48,858
And I think that there's
this aspect of how much more

1188
00:58:48,891 --> 00:58:51,761
complexly to do we
put into these models

1189
00:58:51,794 --> 00:58:54,831
versus running more
simple models faster.

1190
00:58:55,898 --> 00:58:57,867
At the same time in parallel,

1191
00:58:57,900 --> 00:59:00,736
our computing capabilities
start to come up.

1192
00:59:00,769 --> 00:59:03,406
So sometimes we might
have really great models,

1193
00:59:03,439 --> 00:59:05,341
but our computing
capabilities can't actually

1194
00:59:05,374 --> 00:59:06,943
run the models that well.

1195
00:59:06,976 --> 00:59:08,444
Other times we have
good computing,

1196
00:59:08,477 --> 00:59:10,413
and our models don't
take advantage.

1197
00:59:10,446 --> 00:59:12,615
So there's this
kind of handshake

1198
00:59:12,648 --> 00:59:16,052
between computational
power and model complexity

1199
00:59:16,085 --> 00:59:17,787
in terms of its development.

1200
00:59:17,820 --> 00:59:20,323
In terms of, and again it
comes down to your question,

1201
00:59:20,356 --> 00:59:22,992
or the policy maker's
question, or my question.

1202
00:59:23,025 --> 00:59:25,761
Am I interested in
ecological processes?

1203
00:59:25,794 --> 00:59:28,497
That's actually my main love

1204
00:59:28,530 --> 00:59:33,069
is how do ecosystems
function, right?

1205
00:59:33,102 --> 00:59:36,172
But if you're a policy
maker, you might wanna know,

1206
00:59:37,539 --> 00:59:39,375
you might wanna know
something about CO2

1207
00:59:39,408 --> 00:59:41,544
or mega storms or whatnot,

1208
00:59:41,577 --> 00:59:43,512
or you might be
interested also like me

1209
00:59:43,545 --> 00:59:44,680
in how ecosystems function

1210
00:59:44,713 --> 00:59:47,550
because you wanna know which
trees are gonna die first

1211
00:59:47,583 --> 00:59:49,485
and if your state's gonna
have massive wildfires,

1212
00:59:49,518 --> 00:59:51,887
and how to mobilize,
and things like that.

1213
00:59:51,920 --> 00:59:54,724
So there's different
questions that correspond

1214
00:59:54,757 --> 00:59:56,292
to different types of models

1215
00:59:56,325 --> 00:59:59,095
and required accuracies
as you mentioned.

1216
00:59:59,128 --> 01:00:01,764
That comes back, switching
to the water side,

1217
01:00:01,797 --> 01:00:02,965
to the last bit

1218
01:00:02,998 --> 01:00:04,767
where I mentioned the
water applications.

1219
01:00:04,800 --> 01:00:08,170
They are trying to define
the accuracies required

1220
01:00:08,203 --> 01:00:11,340
by decision makers
in the water realm,

1221
01:00:11,373 --> 01:00:13,542
and whether or not we can
meet them or exceed them

1222
01:00:13,575 --> 01:00:14,810
on the NASA side.

1223
01:00:14,843 --> 01:00:17,179
That helps us drive
our requirements

1224
01:00:17,212 --> 01:00:18,848
both in the observations
and in the models

1225
01:00:18,881 --> 01:00:21,684
to be able to meet
the societal needs.

1226
01:00:21,717 --> 01:00:22,685
- All right, thanks.

1227
01:00:24,620 --> 01:00:27,223
- Hi, thanks for your talk.

1228
01:00:27,256 --> 01:00:30,893
I just wondered, you
touched on nitrogen,

1229
01:00:30,926 --> 01:00:34,931
and my Google alert popped up

1230
01:00:36,165 --> 01:00:37,867
a piece on climate change,

1231
01:00:37,900 --> 01:00:40,569
a piece on nitrogen today
it said that there's

1232
01:00:40,602 --> 01:00:43,105
a very large component
of what we expect to see

1233
01:00:43,138 --> 01:00:47,810
in the amount of nitrogen
in the ecosystem, I guess,

1234
01:00:47,843 --> 01:00:50,579
and terrestrial system
that's not there,

1235
01:00:50,612 --> 01:00:52,948
and they just kind of figured
out where it might be.

1236
01:00:52,981 --> 01:00:54,150
I wondered if you could...

1237
01:00:54,183 --> 01:00:55,518
- [Josh] Yeah, in the
rocks or something.

1238
01:00:55,551 --> 01:00:56,852
- Yeah, do you know anything?

1239
01:00:56,885 --> 01:00:58,421
They said it would
effect models.

1240
01:00:58,454 --> 01:00:59,689
Do you have any idea?

1241
01:00:59,722 --> 01:01:03,159
I didn't quite follow,
but which way it would go?

1242
01:01:04,660 --> 01:01:07,229
- So my buddy Ben Houlton
was the lead author

1243
01:01:07,262 --> 01:01:09,332
of that study up at UC Davis,

1244
01:01:10,299 --> 01:01:12,268
and I just hung out
with him for a bit.

1245
01:01:13,902 --> 01:01:14,837
He's a smart guy.

1246
01:01:14,870 --> 01:01:15,871
I don't want to
say anything bad,

1247
01:01:15,904 --> 01:01:17,873
especially because
I'm being televised.

1248
01:01:17,906 --> 01:01:20,143
[laughing]

1249
01:01:22,811 --> 01:01:26,315
Yeah, so he's coming up
with really good discoveries

1250
01:01:26,348 --> 01:01:29,151
about the nitrogen
and nutrient cycles,

1251
01:01:29,184 --> 01:01:30,486
as well as the rest of us.

1252
01:01:32,588 --> 01:01:33,522
Where I hung out with him was

1253
01:01:33,555 --> 01:01:35,958
at a modeling
workshop at Caltech

1254
01:01:35,991 --> 01:01:37,893
where we were trying to
figure out how to put

1255
01:01:37,926 --> 01:01:40,062
our newest science into
models and whether or not,

1256
01:01:40,095 --> 01:01:42,498
like in my response
to the last question,

1257
01:01:42,531 --> 01:01:46,102
is that complexity
useful to models.

1258
01:01:46,135 --> 01:01:47,803
I mentioned that in
the earlier talk.

1259
01:01:47,836 --> 01:01:49,071
Do we want more soil layers?

1260
01:01:49,104 --> 01:01:50,172
Do we want this aspect?

1261
01:01:50,205 --> 01:01:51,040
Do we want fungi?

1262
01:01:52,007 --> 01:01:54,577
It all just adds
computational demand

1263
01:01:54,610 --> 01:01:56,479
and it slows down our models,

1264
01:01:56,512 --> 01:01:58,180
but maybe it's
really important.

1265
01:01:58,213 --> 01:02:01,450
So these are those questions
that we have to face.

1266
01:02:01,483 --> 01:02:04,653
We're not gonna model every
electron in photosynthesis.

1267
01:02:04,686 --> 01:02:06,989
Maybe we should,
but probably not,

1268
01:02:07,022 --> 01:02:09,425
so we have to make
these decisions.

1269
01:02:09,458 --> 01:02:13,262
It's not obviously
quantifiable.

1270
01:02:13,295 --> 01:02:16,298
I think it comes down to
the previous questions of

1271
01:02:16,331 --> 01:02:18,501
what are your questions,
what are your objectives,

1272
01:02:18,534 --> 01:02:19,935
and what are the
accuracies required

1273
01:02:19,968 --> 01:02:21,103
to meet those objectives,

1274
01:02:21,136 --> 01:02:22,571
and they're different
for different people.

1275
01:02:22,604 --> 01:02:25,775
So something like that
story would relate

1276
01:02:25,808 --> 01:02:27,943
to different types
of questions as well.

1277
01:02:27,976 --> 01:02:29,245
- [Woman] Thank you.

1278
01:02:33,215 --> 01:02:35,718
- All right, I'll take
one more question,

1279
01:02:35,751 --> 01:02:37,553
and then I'll take
a few questions

1280
01:02:37,586 --> 01:02:40,623
from the crazy
internet.

1281
01:02:40,656 --> 01:02:43,626
I'm nervous about this.
[laughing]

1282
01:02:43,659 --> 01:02:46,929
- My question has to do
with the longterm increase

1283
01:02:46,962 --> 01:02:49,165
of carbon dioxide
in the atmosphere.

1284
01:02:49,998 --> 01:02:54,103
A number of years ago I
saw a speaker talk about

1285
01:02:54,136 --> 01:02:56,705
paleolithic times where
he said that the amount

1286
01:02:56,738 --> 01:02:59,275
of carbon dioxide at
that time was 20-50 times

1287
01:02:59,308 --> 01:03:00,509
greater than it is today,

1288
01:03:00,542 --> 01:03:02,678
and I'm wondering if
when we're talking about

1289
01:03:02,711 --> 01:03:04,880
the historical increase
of carbon dioxide

1290
01:03:04,913 --> 01:03:08,617
whether you've included
that in your predictions

1291
01:03:08,650 --> 01:03:10,286
of what's gonna
happen in the future.

1292
01:03:10,319 --> 01:03:12,021
- Yeah, the kind
of paleo record?

1293
01:03:12,054 --> 01:03:14,256
Paleo record's really
value, absolutely.

1294
01:03:14,289 --> 01:03:17,459
I think that we do a lot
of these model evaluations

1295
01:03:17,492 --> 01:03:21,764
against the paleo record,
or the geologic record,

1296
01:03:21,797 --> 01:03:24,133
however you wanna
kinda phrase it.

1297
01:03:24,166 --> 01:03:28,070
It's in some ways easier to
do the past than the present

1298
01:03:28,103 --> 01:03:30,806
or the future, because
we weren't around,

1299
01:03:30,839 --> 01:03:33,042
and we add a lot of complexity.

1300
01:03:33,075 --> 01:03:35,945
So there's very predictable
cycles of how the Earth

1301
01:03:35,978 --> 01:03:40,583
and the climate system operates

1302
01:03:40,616 --> 01:03:43,018
over the history of the Earth
that we have some records

1303
01:03:43,051 --> 01:03:44,787
through paleo record,

1304
01:03:44,820 --> 01:03:48,557
but it's this now time, when
humans came into the picture,

1305
01:03:48,590 --> 01:03:52,328
that has like jolted the signal
out of that historic record.

1306
01:03:54,096 --> 01:03:55,865
That's why a lot of our
interest has been on

1307
01:03:55,898 --> 01:03:57,533
being able to capture
that human element.

1308
01:03:57,566 --> 01:04:00,202
You can imagine how
challenging that is.

1309
01:04:00,235 --> 01:04:05,208
It's not just this is
how much CO2 we emitted

1310
01:04:05,974 --> 01:04:08,477
and how much are
we going to emit,

1311
01:04:08,510 --> 01:04:11,714
but what's gonna kill
the Amazon first,

1312
01:04:11,747 --> 01:04:15,117
droughts or people chopping
down the Amazon, right?

1313
01:04:15,150 --> 01:04:17,853
And are you gonna be able to
capture that in the models?

1314
01:04:17,886 --> 01:04:21,423
Human behavior and
economics is another beast,

1315
01:04:21,456 --> 01:04:25,060
but there's a whole community
that's focused on that as well

1316
01:04:25,093 --> 01:04:27,229
that handshakes
with my community

1317
01:04:27,262 --> 01:04:29,899
which is a little bit more
biological, physical side.

1318
01:04:31,633 --> 01:04:33,769
I guess that adds a lot
of extra information

1319
01:04:33,802 --> 01:04:35,137
that you probably
weren't asking for,

1320
01:04:35,170 --> 01:04:39,842
but that was kind of
my answer to linking

1321
01:04:39,875 --> 01:04:42,912
that historical record for
the present and the future.

1322
01:04:44,780 --> 01:04:46,682
- And what about
like exoparasites

1323
01:04:46,715 --> 01:04:47,783
like the bark beetle

1324
01:04:47,816 --> 01:04:50,519
is that being included
in the models,

1325
01:04:50,552 --> 01:04:53,455
and what is the overall
global impact of that?

1326
01:04:53,488 --> 01:04:54,757
- Great question!

1327
01:04:54,790 --> 01:04:56,726
We should definitely
do lunch sometime.

1328
01:04:58,293 --> 01:05:00,629
So I mentioned
disturbance, right?

1329
01:05:00,662 --> 01:05:03,132
And I only showed
a picture of fire.

1330
01:05:03,165 --> 01:05:05,834
But when there's a
drought, for example,

1331
01:05:05,867 --> 01:05:09,338
as you can imagine I'm very
much interested in droughts

1332
01:05:09,371 --> 01:05:11,974
having grown up in
California and in my slides,

1333
01:05:12,975 --> 01:05:16,845
it's not necessarily the water
stress that kills the trees,

1334
01:05:16,878 --> 01:05:21,583
it's the infestations,
right, it's the beetles,

1335
01:05:21,616 --> 01:05:24,586
it's the bacteria,
it's the other fungi,

1336
01:05:24,619 --> 01:05:26,322
not the good fungi,
the bad fungi.

1337
01:05:28,423 --> 01:05:33,029
So very critical, and we're
still trying to come to

1338
01:05:34,029 --> 01:05:37,933
grips with how much to include
that complexity in models.

1339
01:05:37,966 --> 01:05:39,068
A lot of the models
do have that,

1340
01:05:39,101 --> 01:05:40,035
but they're kind of more,

1341
01:05:40,068 --> 01:05:41,270
we're not like

1342
01:05:41,303 --> 01:05:43,372
modeling individual
beetles flying around

1343
01:05:43,405 --> 01:05:46,909
or whatever, it's more like
probabilistically speaking

1344
01:05:46,942 --> 01:05:49,144
if plants dry to
a certain extent

1345
01:05:49,177 --> 01:05:52,715
they are more likely
to die because of,

1346
01:05:52,748 --> 01:05:56,285
let's call it,
beetles, or something.

1347
01:05:56,318 --> 01:05:59,421
That's how we kind of
invoke some of this,

1348
01:05:59,454 --> 01:06:01,357
the other disturbances
outside of fire,

1349
01:06:01,390 --> 01:06:03,826
but yeah good question.
- Thank you.

1350
01:06:05,794 --> 01:06:09,098
- Okay, I will take a
few from the internet.

1351
01:06:12,667 --> 01:06:14,103
Ha, okay.

1352
01:06:14,136 --> 01:06:17,072
How much of an impact
could machine learning

1353
01:06:17,105 --> 01:06:21,711
and artificial intelligence
have on future climate models?

1354
01:06:23,311 --> 01:06:25,081
Section nine,
what's section nine?

1355
01:06:26,948 --> 01:06:30,319
Oh, okay, they told me
to say who asked this,

1356
01:06:30,352 --> 01:06:33,155
and I was expecting
like a Twitter handle.

1357
01:06:33,188 --> 01:06:34,723
I don't know, we
have sections at JPL.

1358
01:06:34,756 --> 01:06:36,759
It's like is this one of my
colleagues messing with me?

1359
01:06:36,792 --> 01:06:39,094
[laughing]

1360
01:06:39,127 --> 01:06:41,897
All right, so machine learning
and artificial intelligence,

1361
01:06:41,930 --> 01:06:44,066
I guess I should speak
into one of the cameras.

1362
01:06:44,099 --> 01:06:46,602
This one's not even
looking at me, geez.

1363
01:06:46,635 --> 01:06:49,371
I'm all dressed up and
it's like not even looking.

1364
01:06:49,404 --> 01:06:53,475
Okay, so we are using
machine learning.

1365
01:06:53,508 --> 01:06:56,245
I mentioned to one
of the responses

1366
01:06:56,278 --> 01:06:58,247
that we just had this
modeling workshop.

1367
01:06:58,280 --> 01:07:00,149
That was a huge part of
our modeling workshop

1368
01:07:00,182 --> 01:07:03,619
was using machine learning.

1369
01:07:03,652 --> 01:07:06,255
So a lot of the models that
I talked about actually

1370
01:07:07,422 --> 01:07:11,427
are a little bit
biased in my interests,

1371
01:07:11,460 --> 01:07:13,529
which is what we call
process modeling.

1372
01:07:13,562 --> 01:07:16,732
I want to understand how
something occurs, right?

1373
01:07:16,765 --> 01:07:20,869
It's not that, it's kind of
like a model of Newton's apple

1374
01:07:20,902 --> 01:07:24,306
hitting the floor, you
could predict it because

1375
01:07:24,339 --> 01:07:27,509
it always hits the
ground at a certain time,

1376
01:07:27,542 --> 01:07:29,978
or you could understand
that there's a process

1377
01:07:30,011 --> 01:07:32,214
related to gravity and friction

1378
01:07:32,247 --> 01:07:33,582
and mass and things like that.

1379
01:07:33,615 --> 01:07:35,884
So I'm interest in process.

1380
01:07:35,917 --> 01:07:39,188
Machine learning and
artificial intelligence

1381
01:07:39,221 --> 01:07:43,092
is more in terms of
taking that myriad of data

1382
01:07:43,125 --> 01:07:44,760
and figuring out
signals within it,

1383
01:07:44,793 --> 01:07:46,428
which is certainly helpful,

1384
01:07:46,461 --> 01:07:48,664
at least in terms
of how I use it.

1385
01:07:48,697 --> 01:07:50,065
I do use machine learning.

1386
01:07:50,098 --> 01:07:52,668
I use a lot of neural
networks and so on,

1387
01:07:52,701 --> 01:07:56,004
and data products that
use decision trees

1388
01:07:56,037 --> 01:07:57,372
and other types of
machine learning.

1389
01:07:57,405 --> 01:07:58,641
So they're very valuable.

1390
01:07:59,674 --> 01:08:03,479
The challenge, I guess, is that

1391
01:08:03,512 --> 01:08:07,182
machine learning and AI

1392
01:08:07,215 --> 01:08:10,752
they're trained to
what's available;

1393
01:08:10,785 --> 01:08:12,421
they're trained
to your data set.

1394
01:08:14,022 --> 01:08:16,458
As CO2 and climate

1395
01:08:16,491 --> 01:08:19,028
moves outside of our
existing data set,

1396
01:08:19,794 --> 01:08:24,366
it's not exactly clear how
trustworthy something trained

1397
01:08:24,399 --> 01:08:27,236
to a data set now will
behave in the future

1398
01:08:27,269 --> 01:08:31,207
if there's not something
that's traceable to a process.

1399
01:08:33,275 --> 01:08:34,710
It's definitely a compromise.

1400
01:08:34,743 --> 01:08:36,245
Definitely back and forth.

1401
01:08:36,278 --> 01:08:37,479
I have colleagues
on both sides.

1402
01:08:37,512 --> 01:08:41,650
I use both sides, so I
don't wanna downplay that,

1403
01:08:41,683 --> 01:08:45,421
but that's kind of my
response to that question.

1404
01:08:48,623 --> 01:08:49,592
Anything else?

1405
01:08:51,793 --> 01:08:54,363
All right, well thank you
for your time and attention.

1406
01:08:54,396 --> 01:09:01,303
[applauding]