of 9
iScience, Volume
26
Supplemental information
Conspiracy spillovers and geoengineering
Ramit Debnath, David M. Reiner, Benjamin K. Sovacool, Finn Müller-Hansen, Tim
Repke, R. Michael Alvarez, and Shaun D. Fitzgerald
1
Supplementary Information
(SI)
:
Conspiracy spillover
s
and geoengineering
Ramit Debnath, David M. Reiner, Benjamin K. Sovacool, Finn Muller
-
Hansen, Tim Repke, R.
Michael Alvarez, and Shaun Fitzgerald
The descriptive characteristics of the #geoengineering
t
weets is illustrated in SI Table1
.
SI Table 1: Descriptive characteristics of the Tweet corpus (n = 814, 924)
,
r
elated to STAR Methods
User_metrics
Minimum
Median
Mean
Maximum
Retweet_count
0
1
12.70
2029
Like_count
0
0
0.45
2631
User_t
weet
_count
1
44479
184712
3980074
User_
Followers
0
1895
4539
19370201
User_
Followings
0
1694
3016
925922
SI
Figure 1:
Spearman correlation between negative emotion and daily tweet volume of #geoengineering (n = 814, 924). The adjusted R
-
squared value is 0.1002,
s
tandard
-
error is 0.01920 significant at 0.001 level.
[Related to Figure 1]
SI
Figure
2
show
s
that
severe
toxicity
is distinctively right
-
skewed with
a single
peak, with a 13
-
year
mean score of 0.12 and
a
median score of 0.07. The
toxicity
attribute has multiple peaks with
a
right
ward
skew, with a 13
-
year mean score of 0.17 and
a
median score of 0.11. Furtherm
ore, our
approach
classified
~
60% of
t
weets as
toxic
relative to
severe toxic
between
February 2009
and
August 2021 (see
SI Figure 2b
).
The share of
severe toxic
tweets
further fell by 62.5% from August 2021, which coincides with the IPCC AR6
announcements
on the assessment of
solar geoengineering
.
The descriptive statistics details of the toxicity
attributes
are
presented in SI Table 2.
SI Table
2
: Descriptive characteristics of the
toxicity attributes in the Tweet corpus
, related to Figure 3
Attributes
Minimum
Median
Mean
Maximum
TOXICITY
0
0.13
0.17
0.99
SEVERE_TOXICITY
0
0.08
0.12
0.99
R
=
0.12
,
p
<
0.001
0
1
2
3
0
2
4
6
T
w
eet v
olume/da
y(log10)
E
s
t
i
m
a
t
e
d
n
e
g
a
t
i
v
e
e
m
o
t
i
o
n
s
c
o
r
e
s
(
l
o
g
1
0
)
2
SI Figure 2:
(a) Density characteristics of TOXICITY and SEVERE_TOXICITY estimates; (b) Temporal shifts in share of
the
TOXICITY and
SEVERE_TOXICITY score
s
over the 13
-
year period; (c)
Spearman correlation between
t
weet
s
per day, TOXICITY (adjusted R
-
square = 0.0812,
std. error = 0.0013) and SEVERE_TOXICITY (adjusted R
-
square = 0.0719, std. error = 0.0014), significant at 0.01 level.
[Related to Figure 3 and
Table 1]
0
2
4
6
8
0.00
0.25
0.50
0.75
1.00
v
alue
d
e
n
s
i
t
y
v
ar
iab
le
T
O
XICITY
SEVERE_T
O
XICITY
a
)
b
)
f
)
R
=
0.38
,
p
<
0.001
0.00
0.25
0.50
0.75
1.00
0
2
4
6
T
w
eets/da
y (log10)
T
o
x
i
c
i
t
y
s
c
o
r
e
R
=
0.36
,
p
<
0.001
0.00
0.25
0.50
0.75
1.00
0
2
4
6
T
w
eets/da
y (log10)
S
e
v
e
r
e
T
o
x
i
c
i
t
y
s
c
o
r
e
c
)
3
We
show in SI Figure 3
that the UK and the USA
labelled tweets
have similar toxicity
distributions
, with similar
right skewness. The distribution
differs
drastically for Sweden and India, with India demonstrating localised
peaks in high
TOXICITY (0.85) and SEVERE_TOXICITY (0.73) scores.
SI Figure 3:
Density estimate characteristics of TOXICITY and SEVERE_TOXICITY
by country label
[Relate to Table 2]
.
SI Table
3
: Descriptive characteristics of the
toxicity attributes at the
country scale
, related to Table 2
Attributes
Minimum
Median
Mean
Maximum
TOXICITY (USA)
0.001
0.150
0.202
0.990
SEVERE_TOXICITY (USA)
0.000
0.094
0.139
0.941
TOXICITY (UK)
0.000
0.120
0.160
0.980
SEVERE_TOXICITY (UK)
0.000
0.008
0.110
0.920
TOXICITY
(Sweden)
0.010
0.160
0.190
0.950
SEVERE_TOXICITY
(Sweden)
0.010
0.150
0.160
0.890
TOXICITY (India)
0.000
0.150
0.210
0.960
SEVERE_TOXICITY (India)
0.000
0.100
0.150
0.870
0
2
4
6
8
0.00
0.25
0.50
0.75
1.00
v
alue
d
e
n
s
i
t
y
v
ar
iab
le
T
O
XICITY
SEVERE_T
O
XICITY
U
K
0
2
4
6
0.00
0.25
0.50
0.75
1.00
v
alue
d
e
n
s
i
t
y
v
ar
iab
le
T
O
XICITY
SEVERE_T
O
XICITY
U
S
A
0
2
4
0.00
0.25
0.50
0.75
v
alue
d
e
n
s
i
t
y
v
ar
iab
le
T
O
XICITY
SEVERE_T
O
XICITY
S
w
e
d
e
n
0
2
4
6
0.00
0.25
0.50
0.75
1.00
v
alue
d
e
n
s
i
t
y
I
n
d
i
a
v
ar
iab
le
T
O
XICITY
SEVERE_T
O
XICITY
4
SI Figure 4:
A 6
-
months moving
-
average representation of Twitter emotions associated with major SG projects and governance events
for
country
hashtag
s
between
2009
-
2021. The emotion scores are estimated using a lexicon
-
based approach (NRC
-
lexicon) with the following
embedded emotions in each category: positive emotion
s
(optimism, joy and trust), negative emotion
s
(disgust, fear, anger and sadness) and
neutral e
motion
s
(anticipation and surprise)
. The country hashtags are:
(a) United Kingdom (UK) (n = 41,386), (b) United States of America
(USA) (n = 81,310), (c) Sweden (n = 3,691), (d) India (n = 2,689)
.
The red
-
dot shows the
mean,
and the red
-
line shows max and
min values.
[Related to Figure 5, Figure 6 and Table 2]
T
weets with
UK
hashtags have
the highest share of negative and neutral emotions of the countries
examined. For example, moving average estimation shows that neutral emotions such as surprise and
an
ticipation increased (by more than 50%) in
UK hashta
g
ged
communications when SG projects (like SPICE,
GRGP and GRIP) were launched (
SI
Figure 4a).
Unlike the UK
hashtagged tweets
, the USA
hashtagged
have
a consistently higher share of positive
emotions a
cross the timeline (
SI
Figure 4b). However, positive emotions (almost 40%) were more distinct
for governance
-
related events like the launch of significant reports (
SI
Figure 4b). On the other hand, SG
funding announcements and project launches were associa
ted with a 25% rise in negative emotions
(disgust, fear, anger and negative) (
see SI
Figure 4b).
0
%
1
0
%
2
0
%
3
0
%
4
0
%
5
0
%
6
0
%
7
0
%
8
0
%
9
0
%
1
0
0
%
N
O
A
A
p
r
o
j
e
c
t
s
:
S
t
r
a
t
o
s
p
h
e
r
i
c
A
e
r
o
s
o
l
I
n
j
e
c
t
i
o
n
(
S
A
I
)
a
n
d
M
a
r
i
n
e
C
l
o
u
d
B
r
i
g
h
t
e
n
i
n
g
(
M
C
B
)
b
)
U
S
A
C
I
A
p
a
r
t
n
e
r
s
w
i
t
h
N
a
t
i
o
n
a
l
A
c
a
d
e
m
y
o
f
S
c
i
e
n
c
e
s
(
U
S
)
t
o
f
u
n
d
p
r
o
j
e
c
t
s
S
e
a
F
u
e
l
o
f
A
c
t
o
f
2
0
1
9
:
D
i
r
e
c
t
A
i
r
C
a
p
t
u
r
e
a
n
d
B
l
u
e
C
a
r
b
o
n
R
e
m
o
v
a
l
T
e
c
h
n
o
l
o
g
y
P
r
o
g
r
a
m
(
2
0
2
0
-
2
0
2
3
)
C
o
n
s
p
i
r
a
c
y
t
h
e
o
r
i
s
t
c
l
o
s
e
s
H
A
A
R
P
S
C
o
P
E
x
a
n
d
I
c
e
9
1
1
l
a
u
n
c
h
e
d
N
a
t
i
o
n
a
l
A
c
a
d
e
m
y
o
f
S
c
i
e
n
c
e
s
,
E
n
g
i
n
e
e
r
i
n
g
a
n
d
M
e
d
i
c
i
n
e
(
U
S
)
c
o
n
s
e
n
s
u
s
r
e
p
o
r
t
U
S
$
1
1
0
m
i
l
l
i
o
n
U
.
S
.
F
e
d
e
r
a
l
S
u
p
p
o
r
t
P
r
o
g
r
a
m
M
a
r
i
n
e
C
l
o
u
d
B
r
i
g
h
t
e
n
i
n
g
e
x
p
e
r
i
m
e
n
t
s
f
u
n
d
e
d
b
y
B
i
l
l
G
a
t
e
s
G
e
o
e
n
g
i
n
e
e
r
i
n
g
g
o
v
e
r
n
a
n
c
e
a
n
d
t
e
c
h
n
o
l
o
g
y
p
o
l
i
c
y
r
e
p
o
r
t
b
y
U
S
C
o
n
g
r
e
s
s
i
o
n
a
l
R
e
s
e
a
r
c
h
S
e
r
v
i
c
e
U
S
G
o
v
e
r
n
m
e
n
t
A
c
c
o
u
n
t
a
b
i
l
i
t
y
R
e
p
o
r
t
s
e
t
t
i
n
g
f
e
d
e
r
a
l
r
e
s
e
a
r
c
h
s
t
r
a
t
e
g
y
o
n
g
e
o
e
n
g
i
n
e
e
r
i
n
g
C
a
r
n
e
g
i
e
G
o
v
e
r
n
a
n
c
e
P
r
o
j
e
c
t
L
a
u
n
c
h
0
%
1
0
%
2
0
%
3
0
%
4
0
%
5
0
%
6
0
%
7
0
%
8
0
%
9
0
%
1
0
0
%
0
0
0
0
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
5
5
5
5
6
6
6
6
7
7
7
7
8
8
8
8
9
9
9
9
0
0
0
0
1
1
1
1
R
o
y
a
l
S
o
c
i
e
t
y
r
e
l
e
a
s
e
s
r
e
p
o
r
t
o
n
G
e
o
e
n
g
i
n
e
e
r
i
n
g
g
o
v
e
r
n
a
n
c
e
S
P
I
C
E
p
r
o
j
e
c
t
c
a
n
c
e
l
l
e
d
c
i
t
i
n
g
l
a
c
k
o
f
r
e
g
u
l
a
t
i
o
n
a
n
d
g
o
v
e
r
n
a
n
c
e
o
f
S
R
M
r
e
s
e
a
r
c
h
I
n
t
e
g
r
a
t
e
d
A
s
s
e
s
s
m
e
n
t
o
f
G
e
o
e
n
g
i
n
e
e
r
i
n
g
P
r
o
p
o
s
a
l
(
I
A
G
P
)
(
O
c
t
o
b
e
r
2
0
1
0
t
o
F
e
b
r
u
a
r
y
2
0
1
5
)
S
P
I
C
E
p
r
o
j
e
c
t
l
a
u
n
c
h
e
d
i
n
t
h
e
U
K
G
R
I
P
-
G
r
e
e
n
h
o
u
s
e
G
a
s
R
e
m
o
v
a
l
I
n
s
t
r
u
m
e
n
t
s
&
P
o
l
i
c
i
e
s
p
r
o
j
e
c
t
(
J
u
l
y
2
0
1
6
t
o
M
a
y
2
0
1
9
)
a
)
U
K
£
8
0
0
m
i
l
l
i
o
n
C
C
S
I
n
f
r
a
s
t
r
u
c
t
u
r
e
F
u
n
d
C
G
G
P
r
o
j
e
c
t
l
a
u
n
c
h
G
e
o
e
n
g
i
n
e
e
r
i
n
g
R
e
s
e
a
r
c
h
G
o
v
e
r
n
a
n
c
e
P
r
o
j
e
c
t
(
G
R
G
P
)
l
a
u
n
c
h
(
2
0
1
6
t
o
p
r
e
s
e
n
t
)
G
G
R
E
W
-
G
r
e
e
n
h
o
u
s
e
G
a
s
R
e
m
o
v
a
l
b
y
E
n
h
a
n
c
e
d
W
e
a
t
h
e
r
i
n
g
p
r
o
j
e
c
t
(
J
u
l
y
2
0
1
7
t
o
J
u
n
e
2
0
2
1
)
C
o
m
p
a
r
a
t
i
v
e
a
s
s
e
s
s
m
e
n
t
a
n
d
r
e
g
i
o
n
-
s
p
e
c
i
f
i
c
o
p
t
i
m
i
s
a
t
i
o
n
o
f
G
G
R
p
r
o
j
e
c
t
(
J
u
l
y
2
0
1
7
t
o
D
e
c
e
m
b
e
r
2
0
2
1
)
d
)
I
n
d
i
a
I
n
d
o
-
G
e
r
m
a
n
p
r
o
j
e
c
t
L
O
H
A
F
E
X
h
a
l
t
e
d
D
e
p
a
r
t
m
e
n
t
o
f
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
(
D
S
T
)
l
a
u
n
c
h
e
s
s
c
o
p
i
n
g
f
o
r
g
e
o
e
n
g
i
n
e
e
r
i
n
g
0
%
1
0
%
2
0
%
3
0
%
4
0
%
5
0
%
6
0
%
7
0
%
8
0
%
9
0
%
1
0
0
%
U
n
i
t
e
d
N
a
t
i
o
n
s
B
i
o
d
i
v
e
r
s
i
t
y
M
e
e
t
i
n
g
o
b
s
t
r
u
c
t
s
g
e
o
e
n
g
i
n
e
e
r
i
n
g
C
E
E
W
c
o
n
f
e
r
e
n
c
e
o
n
g
o
v
e
r
n
a
n
c
e
o
f
s
o
l
a
r
g
e
o
e
n
g
i
n
e
e
r
i
n
g
r
e
s
e
a
r
c
h
a
n
d
t
e
c
h
n
o
l
o
g
i
e
s
C
E
E
W
c
o
n
f
e
r
e
n
c
e
o
n
g
o
v
e
r
n
a
n
c
e
o
f
s
o
l
a
r
g
e
o
e
n
g
i
n
e
e
r
i
n
g
C
C
R
C
p
l
a
n
s
o
c
e
a
n
i
o
n
f
e
r
t
i
l
i
z
a
t
i
o
n
w
i
t
h
N
I
O
,
G
o
a
(
2
0
1
9
-
2
0
2
3
)
C
l
o
u
d
s
e
e
d
i
n
g
p
r
o
j
e
c
t
b
y
M
a
h
a
r
a
s
h
t
a
R
a
i
n
f
a
l
l
E
n
h
a
n
c
e
m
e
n
t
(
l
a
s
t
f
l
i
g
h
t
)
G
o
I
a
p
p
r
o
v
e
s
U
S
$
1
0
m
i
l
l
i
o
n
o
f
c
l
o
u
d
s
e
e
d
i
n
g
t
o
f
i
g
h
t
d
r
o
u
g
h
t
i
n
M
a
h
a
r
a
s
h
t
r
a
a
n
d
K
a
r
n
a
t
a
k
a
A
t
m
o
s
p
h
e
r
i
c
S
u
l
f
a
t
e
I
n
j
e
c
t
i
o
n
s
p
a
r
k
s
c
o
n
t
r
o
v
e
r
s
y
C
l
o
u
d
s
e
e
d
i
n
g
c
o
n
t
r
o
v
e
r
s
y
c
)
S
w
e
d
e
n
0
%
1
0
%
2
0
%
3
0
%
4
0
%
5
0
%
6
0
%
7
0
%
8
0
%
9
0
%
1
0
0
%
S
C
o
P
E
x
f
l
i
g
h
t
c
a
n
c
e
l
l
e
d
S
w
e
d
i
s
h
S
o
c
i
e
t
y
f
o
r
N
a
t
u
r
e
C
o
n
s
e
r
v
a
t
i
o
n
R
e
p
o
r
t
o
p
p
o
s
i
n
g
g
e
o
e
n
g
i
n
e
e
r
i
n
g
H
a
n
d
s
O
f
f
M
o
t
h
e
r
E
a
r
t
h
!
(
H
O
M
E
)
C
a
m
p
a
i
g
n
l
a
u
n
c
h
e
d
B
E
C
C
S
r
e
s
e
a
r
c
h
s
t
u
d
y
f
u
n
d
e
d
b
y
S
w
e
d
i
s
h
E
n
e
r
g
y
A
g
e
n
c
y
c
a
u
s
e
s
p
u
b
l
i
c
o
u
t
c
r
y
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
N
e
g
a
t
i
v
e
C
O
2
E
m
i
s
s
i
o
n
R
e
d
u
c
t
i
o
n
c
a
u
s
e
s
c
o
n
t
r
o
v
e
r
s
y
S
C
o
P
E
x
a
n
n
o
u
n
c
e
m
e
n
t
S
C
o
P
E
x
l
a
u
n
c
h
e
d
N
e
g
a
t
i
v
e
N
e
u
t
r
a
l
P
o
s
i
t
i
v
e
e
)
T
O
X
I
C
I
T
Y
S
E
V
E
R
E
T
O
X
I
C
I
T
Y
5
The share of neutral emotions (anticipation and surprise) rose fourfold at the launch of the Carnegie
Governance Project in May 2016 (see
SI
Figure 4b). By c
ontrast, sharp peaks in negative (and neutral)
emotions were observed at the launch of the gates
-
funded Marine Cloud Brightening experiments (August
2018). As previously noted, this association has
attracted conspiracy
theorists.
SG
-
related interactions o
n Twitter for Sweden
-
hashtagged tweets
have a distinct cyclical pattern. Figure 4c
shows that the share of negative emotions rises following various SG opposition and governance events. A
prominent anti
-
solar geoengineering campaign called 'Hands Off Mothe
r Earth! (HOME)'
[1]
(December
2020) created a nationwide public appeal to stop the SCoPEx launch. An open letter signed by authors in
45 countries contributed to abandoning the experiment. During this time, the share of negative emotions
increased strongl
y, demonstrating the ability of social media to reflect and potentially shape public
outcomes on SG events.
The Indian case is also interesting since the share of positive emotions
in Indian hashtagged tweets
gradually decreased over the 13 years (
SI
Fi
gure 4d). Abrupt changes in emotion were tied to prominent
national SG events. Notably, the cloud seeding controversy by non
-
state actors led to negative emotions
rising more than threefold in August 2013 (
SI
Figure 4d). However, when state
-
based agencies
sanctioned
the same cloud seeding projects as drought management measures in May 2015, positive emotions
increased sharply by almost 35% (see
SI
Figure 4d). A distinct feature of the Indian case is that many SG
programs (especially cloud seeding) are opera
ted by the federal government's Department of Science and
Technology, which sets the overall governance agenda and is in favour of scientific research on SG
[2].
SI
Figure 4e show the characteristics of
toxicity
and
severe toxicity
in the
t
weet corpus across the four
countries. Mean
severe toxicity
scores for India (0.15) and Sweden (0.16) are slightly higher than for the
USA (0.14) and UK (0.11). For the
toxicity
scores India (0.21) remains highest, followed by the USA (0.20),
Sweden (0.19)
, and the UK (0.16) (see
SI
Figure 4e and SI Table 3). Tweets with high toxicity attributes
across the countries are illustrated in Table 2.
SI Table
4
: Descriptive characteristics of
the
normalised
emotion
scores
in the geospecific hashtags
, related to
Figure 5
and Figure 6
Emotions
Minimum
Median
Mean
Maximum
#geoengineering
Positive
0
0.061
0.091
1
Negative
0
0.344
0.062
1
Neutral
0
0.023
0.051
1
#
UK
Positive
0
0.004
0.011
1
Negative
0
0.008
0.015
1
Neutral
0
0.004
0.003
1
#
USA
Positive
0
0.019
0.034
1
Negative
0
0.014
0.028
1
Neutral
0
0.006
0.015
1
#
INDIA
Positive
0
0.014
0.019
1
Negative
0
0.007
0.017
1
Neutral
0
0.018
0.027
1
#
SWEDEN
Positive
0
0.011
0.032
1
6
Negative
0
0.011
0.038
1
Neutral
0
0.010
0.031
1
SI Table
5
:
Network characteristics of the
geospecific hashtag network
, related to Figure 6
Country
Nodes
Edges
Mean
weighted
degree
Mean
modularity
class
Mean
clustering
coefficient
Mean
eigenvector
centrality score
UK
4581
24843
12.23
171.99
0.77
0.021
USA
7496
40359
9.80
337.31
0.41
0.013
India
418
1793
10.00
25.88
0.051
0.82
Sweden
668
3043
8.74
29.57
0.83
0.089
7
SI Figure
5
:
Dimension reduction embeddings
of ‘chemtrails’ in the
t
weet corpus across the three
-
time scale
, related Figure 3
and Figure 4
.