of 13
Patterns, Volume
3
Supplemental information
Anthroponumbers.org: A quantitative database
of human impacts on Planet Earth
Grif
fi
n Chure, Rachel A. Banks, Avi I. Flamholz, Nicholas S. Sarai, Mason Kamb, Ignacio
Lopez-Gomez, Yinon Bar-On, Ron Milo, and Rob Phillips
Supplemental Note 1: Selection, Validation, and Curation of Values
The
breadth
of
measurements
of
human
impacts
on
the
planet
is
enormous,
covering
a
wide
array
of
disciplines
and
methods.
While
this
is
a
boon
for
science,
this
imposes
a
very
important
burden
any
value
we
care
to
enter
into
the
Human
Impacts
database
must
be
carefully
examined
and
deemed
credible
and
appropriate
for
the
database.
While
we
certainly
acknowledge
we
are
not
domain
experts
in
all
of
these
fields,
the
members
of
the
administrative
team
span
a
broad
range
of
backgrounds,
and
are
all
quantitative
scientists
who
both
deeply
value
the
utility
of
quantitative
measurements
and
have
the
domain
expertise
to
assess
whether
the
reported
values
make
sense
and
are
determined
with
trustworthy
methods.
In
this
section,
we
briefly
outline
the
general
procedure
undertaken
before
a
value
is
entered
into
the
Human
Impacts Database.
Identifying a Potential Entry
Our
first
action
is
identifying
a
value
or
set
of
values
and
determining
whether
they
are
pertinent
to
Human
Impacts.
We
take
a
broad
definition
of
“Human
Impacts”,
but
enforce
that
any
value
must
be
either
(i)
a
direct
result
of
anthropogenic
action,
(ii)
contributed
to
by
anthropogenic
activities,
or
(iii)
is
directly
relevant
to
human
consumption
and/or
production.
Most
importantly,
any
candidate
entry
must
reflect
an
impact
on
some
natural
process.
For
example,
a
value
quantifying
the
standing
population
of
all
livestock
on
Earth
would
fall
under
criterion
(i)
making
it
an
appropriate
candidate
entry.
As
a
counter
example,
the
fraction
of
a
country’s
GDP
resulting
from
fossil
fuel
export
would
not
be
considered
as
a
candidate
value
as
it
describes
an
economic
impact
rather
than
an
impact
on
a
natural
process.
Of
course,
rigid
lines
cannot
always
be
drawn
and
inclusion
of
a
value
is
ultimately
at
the
discretion
of
the
administration
team.
Vetting a Potential Entry
Next,
we
determine
if
the
quantity
is
scientifically
valid
and
appropriate.
This
not
only
includes
the
precise
value
of
the
quantity,
but
the
reliability
of
the
source
and
the
methods
of
measurement.
In
general,
we
consider
data
from
large,
international
efforts
such
as
the
Food
and
Agriculture
Organization
of
the
UN
(FAO)
or
the
Intergovernmental
Panel
on
Climate
Change
(IPCC)
to
be
highly
reliable
sources
of
information.
We
take
these
sources
to
be
reliable
as
they
clearly
report
the
methods
of
their
measurements
or
meta-analyses,
emphasizing
where
assumptions
and
approximations
have
been
made.
Furthermore,
given
the
internationality
of
its
contributors
and
the
deep
well
of
scientists
they
consult
and
employ,
we
find
that
the
FAO
and
IPCC
are
largely
free
of
bias
as
they
have
little
stake
in
reporting
overly-rosy
or
negative
results.
For
this
reason,
we
are
less
likely
to
include
values
from
industry
reports,
which
have
potential
conflicts
of
interest.
Whenever
industry
reports
are
used,
we
try
to
find
multiple
sources
for
that
particular
value
to
place
it
in
context.
For
example,
we
extensively
use
the
BP
Annual
Statistical
Report
on
Energy
in
the
human
impacts
database.
As
BP
is
a
private
company
with
financial
interests
in
reporting
global
energy
use,
we
compare
these
values
with
those
from
the
US
Energy
Information
Administration
(EIA)
and
the
International
Energy
Agency
(IEA)
to
judge
their
consistency.
We
draw
a
large
number
of
the
values
in
the
Human
Impacts
Database
from
peer-reviewed
scientific
reports.
For
these
data
sources
we
thoroughly
examine
the
reported
methods
used
to
determine
the
value.
If
details
regarding
the
method
are
not
clearly
reported
(e.g.
the
value
“was
fitted”
without
explaining
the
fitting
procedure),
we
are
strongly
inclined
to
not
trust
that
particular
source.
Furthermore,
if
the
method
is
not
stated
or
the
code/data
are
available
under
only
“reasonable
request”,
the
value
is
not
considered
as
appropriate.
When
possible,
we
also
compare
the
reported
value
to
other
measurements
and
check
if
the
source
explains
any
discrepancy
between
their
measurement
and
others.
In
many
cases,
however,
there
are
not
multiple
reported
values
for
a
given
quantity.
In
these
cases,
we
assess
the
trustworthiness
of
the
reported
value
and
reach
out
to
domain
experts
as
needed.
With
rare
exceptions,
we
do
not
factor the publishing journal in assessing the veracity of a value.
Once
a
value
is
entered
into
the
database,
we
label
it
with
a
primary
and
secondary
category.
Human
impacts
are
inherently
connected
by
webs
of
interactions
and
often
affect
multiple
subsystems
within
the
Earth
system.
Meanwhile,
most
human
impacts
can
be
categorized
according
to
the
systems
with
which
they
interact
most
strongly.
While
incomplete,
these
category
labels
are
meant
to
give
users
an
impression
of
the
subsystems
that
are
most
strongly
influenced
by
or
related
to
the
value.
Users
are
able
to
filter
the
database
by
these
categories
and subcategories.
Considering Uncertainty
While
the
numeric
value
of
a
candidate
quantity
is
an
important
factor
we
consider,
so
too
is
the
reported
uncertainty.
Many
scientific
reports
will
give
an
assessment
of
uncertainty,
either
at
the
statistical,
measurement,
or
systematic
level.
The
clarity
of
the
presented
uncertainty
analysis
is
critical
in
our
determination
of
whether
a
candidate
value
should
be
entered
in
the
database.
While
scientific
reports
often
address
the
uncertainty,
this
is
rarely
reported
in
governmental
and
industry
reports.
Many
numbers
from
governmental
or
intergovernmental
bodies
come
from
surveys
and
are
thus
self-reported
by
countries,
adding
some
uncertainty
to
the
data
and
requiring
some
level
of
interpolation
from
the
reporting
agency.
These
numbers
are
still
considered,
though
we
are
cognizant
of
the
number
of
significant
digits
that
are
reported.
Often,
we
report
these
numbers
as
approximate,
representing
the
uncertainty
with
the
data.
In
all
cases,
we
state
a
concise
yet
sufficiently
detailed
description
of
the
method
and
quantification
of
uncertainty in the “method” field of an HuID entry.
Considering Data Use Protections
As
we
do
not
directly
generate
the
data
presented
in
this
work,
we
are
very
careful
to
ensure
that
the
data
we
add
to
the
database
follows
all
legal
requirements.
All
data
presented
in
the
database
must
be
explicitly
stated
to
be
under
a
generally
permissive
license
such
as
a
Creative
Commons
Attribution
license
(CC-BY).
Data
sources
which
reserve
all
rights
to
their
data
are
not
included
in
the
database
in
any
form.
While
we
ensure
that
we
have
the
legal
right
to
share
these
data,
we
strongly
implore
the
users
of
the
Human
Impacts
Database
to
directly
cite
the
original data source alongside the database if a value or entry is used in a later publication.
Continued Curation and Maintenance of the Database
Unlike
similar
databases
in
chemistry
and
biology
(such
as
BioNumbers
or
the
CRC
Handbook
of
Chemistry
and
Physics),
the
Human
Impacts
Database
faces
a
unique
maintenance
challenge
as
the
values
it
houses
will
undoubtedly
change
with
time
as
will
our
understanding
of
the
facets
of
the
Earth
system
that
are
impacted
by
human
activities.
This
means
that
a
concerted
effort
to
keep
the
values
in
the
database
up
to
date,
within
reason,
is
needed.
In
this
section,
we
outline
steps
we
have
taken
to
ensure
that
the
database
can
be
properly
maintained
and be useful for many years to come.
Composition of the Administrative Team
The
primary
authors
on
this
work
(GC,
RAB,
AIF,
ILG,
NSS,
and
MK)
are
the
primary
members
of
the
administrative
team
of
the
Human
Impacts
Database.
All
of
these
authors
are
practicing
research
scientists
working
at
the
interface
of
biology,
chemistry,
physics,
and
earth
science.
As
a
result,
this
database
will
be
an
invaluable
resource
for
our
specific
research
objectives,
imposing
a
self-interest
in
keeping
the
entries
up
to
date.
All
members
of
the
administrative
team
frequently
read
primary
scientific
literature
covering
these
topics,
meaning
that
critical
new
values
or
updates
to
extant
entries
can
be
reliably
found.
Furthermore,
the
majority
of
the
administrative
team
intend
to
enter
into
leadership
positions
in
academic
and
industry
contexts,
allowing
us
to
mentor
and
train
more
administrators
with
different
domain
expertises.
As
this
database
is
primarily
a
scientific
tool,
we
believe
our
specific
yet
diverse
training
well
prepares
us
as
careful
curators
of
the
database.
Furthermore,
all
authors
are
well-versed
in
computational
methods
with
some
administrators
having
expertise
in
web
development
technologies.
This
added
expertise
helps
ensure
that
the
database
will
reliably
operate
at
both
the
front
and
backend
levels.
In
addition,
the
two
PIs
who
have
led
this
work,
RP
and
RM,
have
support
from
the
Resnick
Sustainability
Center
at
Caltech
and
the
Weizmann
Institute
to
continue work on this project.
Many
of
the
sources
behind
the
HuID
entries
are
updated
on
a
regular
basis,
but
updates
may
not
be
immediately
updated
on
the
database
itself.
For
example,
the
FAO
routinely
updates
their
data
as
new
data
arrive
or
corrections/improvements
to
previously
reported
data
are
released.
The
frequent
nature
of
these
releases
precludes
a
mirror
reflection
of
these
values
in
the
Human
Impacts
Database.
For
continually
updating
sources,
we
update
these
values
at
an
annual
basis
within
the
third
quarter
of
the
calendar
year.
Other
sources,
such
as
the
BP
statistical
report
on
energy
and
IPCC
reports,
also
typically
release
updates
around
this
time.
For
values
that
are
more
frequently
updated
(such
as
the
atmospheric
CO
2
concentration,
which
is
updated
on
a
near-daily
basis),
we
update
these
values
semiannually
coinciding
with
the
spring and fall of the calendar year.
While
the
administration
team
is
diverse
in
their
scientific
interests
and
expertise,
it
is
unreasonable
to
believe
that
our
collective
knowledge
is
all-encompassing
of
Human
Impacts.
There
will
invariably
be
important
values
that
we
are
unaware
of
that
should
be
included
in
the
database.
To
this
end,
we
have
developed
a
community-feedback
system
into
the
database
(
https://anthroponumbers.org/catalog/contact
)
where
the
general
public
can
submit
recommendations
for
new
values
or
updates
and/or
corrections
to
extant
values
in
the
database.
Whenever
feedback
is
submitted,
the
administrative
team
is
notified,
preventing
important
feedback
from
being
cast
into
the
void.
Furthermore,
contact
information
is
provided
for
each
administrative
member
(
https://anthroponumbers.org/catalog/about
)
if
a
user
wishes
to
contact us individually.
As the curation procedures enumerated in the preceding sections are laborious and require
a level of comfort in digesting scientific methods and data, we have opted to not open core
maintenance privileges to the general public. However, all values housed within the database
are also housed within a public GitHub repository
(
https://github.com/rpgroup-pboc/human_impacts
) where
we enthusiastically encourage forking
of the repository and submission of new issues and pull requests. The issues and pull requests
are also monitored by the administrative team.
Supplemental Note 2: References and Explanations For Values Reported in Figure 1
In
this
section,
we
report
our
extensive
and
detailed
referencing
for
each
and
every
quantity
reported
in
the
subpanels
of
Figure
1
of
the
main
text.
As
described
in
the
Materials
&
Methods,
each
value
comes
from
the
manual
curation
of
a
piece
of
scientific,
industrial,
governmental,
or
non-governmental
organization
reports,
articles,
or
databases.
Each
value
listed
here
contains
information
about
the
original
source,
the
method
used
to
obtain
the
value,
as
well
as
accession
identification
numbers
for
the
Human
Impacts
Database
(
https://anthroponumbers.org
),
listed
as
HuIDs.
For
each
value,
we
attempt
to
provide
an
assessment
of
the
uncertainty.
For
some
values,
this
corresponds
to
the
uncertainty
in
the
measurement
or
inference
as
stated
in
the
source
material.
In
cases
where
a
direct
assessment
of
the
uncertainty
was
not
clearly
presented,
we
sought
other
reported
values
for
the
same
quantity
from
different
data
sources
to
present
a
range
of
the
values.
For
others,
this
uncertainty
represents
the
upper-
and
lower-bounds
of
the
measurement
or
estimation.
Each
value
reported
here
is
prefixed
with
a
symbol
representing
our
confidence
in
the
value.
A
symbol
of
equality
(=)
represents
that
either
i)
the
value
is
known
within
a
measurable
uncertainty
or
b)
multiple
sources
confirm
this
value.
A
symbol
of
approximation
(≈)
represents
that
we
are
confident
in
the
reported
value
within
a
factor
of
a
multiplicative
factor
less
than
10.
In
some
cases,
an
approximation
symbol
(≈)
represents
a
range
where
the
values
from
different
sources
differ
within
three
significant
digits
and
the
range
is
then
reported.
Some
values
in
the
database
are
only
known with a lower-bound limit. In these cases, the value is reported with an inequality symbol (>).
A.
Surface Warming
Surface temperature change from the 1850-1900 average ≈ 1.0 - 1.3
(HuID:
7
9
5
9
8
,
7
6
5
3
9
,
1
2
1
4
7
)
°C
Data
Source(s):
HadCRUT.4.6
(Morice
et
al.,
2012,
DOI:
10.1029/2011JD017187
),
GISTEMP
v4
(
GISTEMP
Team,
2020:
GISS
Surface
Temperature
Analysis
(GISTEMP),
version
4
.
NASA
Goddard
Institute
for
Space
Studies.
Dataset
accessed
2020-12-17
at
https://data.giss.nasa.gov/gistemp/
&
Lenssen
et
al.,
2019,
DOI:
10.1029/2018JD029522
)
and
NOAAGlobalTemp
v5
(Huang
et
al,
2020,
DOI:
10.1175/JCLI-D-19-0395.1
)
datasets.
Notes:
The
global
mean
surface
temperature
captures
near-surface
air
temperature
over
the
planet’s
land
and
ocean
surface.
The
value
reported
represents
the
spread
of
three
estimates
and
their
95%
confidence
intervals
for
the
year
2019.
Since
data
for
the
period
1850-1880
are
missing
in
GISTEMP
v4
and
NOAAGlobalTemp
v5,
data
are
centered
by
setting
the
1880-1900
mean
of
all
datasets
to
the
HadCRUT.4.6
mean over the same period.
B.
Annual Ice Melt
Glaciers = (3.0 ± 1.2) × 10
11
m
3
/ yr (HuID:
32459
)
Data
Sources:
Intergovernmental
Panel
on
Climate
Change
(IPCC)
2019
Special
Report
on
the
Ocean
and
Cryosphere in a Changing Climate. Table 2.A.1 on pp. 199-202.
Notes:
Value
corresponds
to
the
trend
of
annual
glacial
ice
volume
loss
(reported
as
ice
mass
loss)
from
major
glacierized
regions
(2006-2015)
based
on
aggregation
of
observation
methods
(original
data
source:
Zemp
et
al.
2019,
DOI:10.1038/s41586-019-1071-0)
with
satellite
gravimetric
observations
(original
data
source:
Wouters
et
al.
2019,
DOI:10.3389/feart.2019.00096).
Ice
volume
loss
was
calculated
from
ice
mass
loss
assuming
a
standard
pure
ice
density
of
920
kg
/
m
3
.
Uncertainty
represents
a
95%
confidence
interval
calculated
from
standard
error
propagation
of
the
95%
confidence
intervals
reported
in
the
original
sources
assuming them to be independent.
Ice sheets = (4.6 ± 0.4) × 10
11
m
3
/ yr (HuIDs:
95798
;
93137
)
Data
Source(s):
D.
N.
Wiese
et
al.
2019
JPL
GRACE
and
GRACE-FO
Mascon
Ocean,
Ice,
and
Hydrology
Equivalent
HDR
Water
Height
RL06M
CRI
Filtered
Version
2.0,
Ver.
2.0,
PO.DAAC,
CA,
USA.
Dataset
accessed [2022-Feb-09]. DOI: 10.5067/TEM- SC-3MJ62
Notes:
Value
corresponds
to
the
trends
of
combined
annual
ice
volume
loss
(reported
as
ice
mass
loss)
from
the
Greenland
and
Antarctic
Ice
Sheets
(2002-2021)
measured
by
satellite
gravimetry.
Ice
volume
loss
was
calculated
from
ice
mass
loss
assuming
a
standard
pure
ice
density
of
920
kg
/
m
3
.
Uncertainty
represents
one standard deviation and considers only propagation of monthly uncertainties in measurement.
Arctic sea ice = (3.0 ± 1.0) × 10
11
m
3
/ yr (HuID:
89520
)
Data
Source(s):
PIOMAS
Arctic
Sea
Ice
Volume
Reanalysis,
Figure
1
of
webpage
as
of
January
31,
2022.
Original method source: Schweiger et al. 2011, DOI:10.1029/2011JC007084
Notes:
Value
reported
corresponds
to
the
trend
of
annual
volume
loss
from
Arctic
sea
ice
(1979-2022).
The
uncertainty
in
the
trend
represents
the
range
in
trends
calculated
from
three
ice
volume
determination
methods.
C.
Sea Ice Area
Extent of loss at yearly maximum cover (September) ≈ 4.8 × 10
10
m
2
/ yr (HuID:
6
6
2
7
7
)
Extent loss at yearly minimum cover (March) ≈ 0.4 × 10
10
m
2
/ yr (HuID:
6
6
2
7
7
)
Average annual extent loss = 2.5 × 10
10
m
2
/ yr (HuID:
6
6
2
7
7
)
Data
Source(s):
Fetterer
et
al.
2017,
updated
daily.
Sea
Ice
Index,
Version
3,
Boulder,
Colorado
USA.
NSIDC: National Snow and Ice Data Center, DOI:10.7265/N5K072F8, [Accessed 2022-Feb-16].
Notes:
Sea
ice
area
is
calculated
by
multiplying
the
percentage
of
sea
ice
in
each
pixel
by
pixel
area
and
taking
the
integral
sum
of
these
products.
Annual
value
corresponds
to
the
linear
trend
of
annual
extent
loss
calculated
by
averaging
over
every
month
in
a
given
year
(2.45
×
10
10
m
2
/
yr
HuID:
66277
).
The
minimum
cover
area
loss
corresponds
to
the
linear
trend
of
Arctic
sea
ice
area
in
September
from
1979-2021
and
the
maximum
cover
area
loss
corresponds
to
the
linear
trend
of
sea
ice
area
in
March
from
1979-2021.
The
Antarctic
sea
ice
area
trend
is
not
shown
because
a
significant
long-term
trend
over
the
satellite
observation
period is not observed and short-term trends are not yet identifiable.
D.
Annual Material Production
Concrete production ≈ (2 - 3) × 10
13
kg / yr (HuID:
25488
;
81346
;
16995
)
Data
Source(s):
United
States
Geological
Survey
(USGS)
National
Minerals
Information
Center,
Commodity
Statistics
and
Information,
Cement
Statistics
and
Information
.
Miller
et
al.
2016,
Table
1,
DOI:10.1088/1748-9326/11/7/074029.
Monteiro
et
al.
2017,
DOI:10.1038/nmat4930.
Krausmann
et
al.
2017,
DOI:
10.1073/pnas.1613773114
Notes:
Concrete
is
formed
when
aggregate
material
is
bonded
together
by
hydrated
cement.
The
USGS
reports
the
mass
of
cement
produced
in
2019
as
4.1
×
10
12
kg.
As
most
cement
is
used
to
form
concrete,
cement
production
can
be
used
to
estimate
concrete
mass
using
a
multiplicative
conversion
factor
of
7
(Monteiro
et
al.).
Miller
et
al.
report
that
the
cement,
aggregate
and
water
used
in
concrete
in
2012
sum
to
2.3
×
10
13
kg.
Krausmann
et
al.
report
an
estimated
value
from
2010
based
on
a
material
input,
stocks,
and
outputs
model.
The
value
is
net
annual
addition
to
concrete
stocks
plus
annual
waste
and
recycling
to
estimate gross production of concrete.
Steel production ≈ 1.9 × 10
12
kg / yr (HuID:
51453
;
44894
;
85981
)
Data
Source(s):
United
States
Geological
Survey
(USGS)
National
Minerals
Information
Center,
Commodity
Statistics
and
Information,
Iron
and
Steel
Statistics
and
Information
.
World
Steel
Association,
World
Steel
in
Figures 2021, p. 7. Krausmann et al. 2017, DOI:
10.1073/pnas.1613773114
Notes:
Crude
steel
includes
stainless
steels,
carbon
steels,
and
other
alloys.
The
USGS
reports
the
mass
of
crude
steel
produced
in
2019
as
1.860
×
10
12
kg.
The
World
Steel
Association
reports
a
production
value
of
1.874
×
10
12
kg
in
2019.
Krausmann
et
al.
report
an
estimated
value
from
2010
based
on
a
material
input,
stocks,
and
outputs
model.
The
value
is
net
annual
addition
to
steel
stocks
plus
annual
waste
and
recycling
to
estimate gross production of steel.
Plastic production ≈ 4 × 10
11
kg / yr (HuID:
97241
;
25437
)
Data
Source(s):
Geyer
et
al.
2017,
Table
S1,
DOI:10.1126/sciadv.1700782.
Krausmann
et
al.
2017,
DOI:
10.1073/pnas.1613773114
Notes:
Value
represents
the
approximate
sum
total
global
production
of
plastic
fibers
and
plastic
resin
during
the
calendar
year
of
2015.
Comprehensive
data
about
global
plastic
production
is
sorely
lacking.
Geyer
et
al.
draw
data
from
various
industry
groups
to
estimate
total
production
of
different
polymers
and
additives.
Some
of
the
underlying
data
is
not
publicly
available,
and
data
from
financially-interested
parties
is
inherently
suspect.
Krausmann
et
al.
report
an
estimated
value
from
2010
based
on
a
material
input,
stocks,
and
outputs
model.
The
value
is
net
annual
addition
to
stocks
plus
annual
waste
and
end-of-life
recycling
to
estimate
gross production of plastics.
E.
Livestock Population
Chicken standing population ≈ 3.5 × 10
10
(HuID:
94934
)
Cattle standing population ≈ 1.5 × 10
9
(HuID:
92006
)
Swine standing population ≈ 1.5 × 10
9
(HuID:
21368
)
All livestock standing population ≈ 4.6 × 10
10
(HuID:
43599
)
Data
Source(s):
Food
and
Agriculture
Organization
(FAO)
of
the
United
Nations
Statistical
Database
(2022)
— Live Animals.
Notes:
Counts
correspond
to
the
estimated
standing
populations
in
2019.
Values
are
reported
directly
by
countries.
The
FAO
uses
non-governmental
statistical
sources
to
address
uncertainty
and
missing
(non-reported) data. Reported values are therefore approximations.
F.
Annual Synthetic Nitrogen Fixation
Annual mass of synthetically fixed nitrogen ≈ (1.4 - 1.5) × 10
11
kg N / yr (HuID:
60580
;
61614
)
Data
Source(s):
United
States
Geological
Survey
(USGS)
National
Minerals
Information
Center,
Commodity
Statistics
and
Information,
Nitrogen
Statistics
and
Information
.
International
Fertilizer
Association
(IFA)
Statistical
Database
(2021)
Ammonia
Production
&
Trade
Tables
by
Region.
Smith
et
al.
2020,
DOI:
10.1039/c9ee02873k.
Notes:
Ammonia
(NH
3
)
produced
globally
is
compiled
by
the
USGS
and
IFA
from
major
factories
that
report
output.
The
USGS
estimates
the
approximate
mass
of
nitrogen
in
ammonia
produced
in
2019
as
1.42
×
10
11
kg
N
and
the
International
Fertilizer
Association
reports
a
production
value
of
1.50
×
10
11
kg
N
in
2019.
Nearly
all
of
this
mass
is
produced
by
the
Haber-Bosch
process
(>96%,
Smith
et
al.
2020).
In
the
United
States
most
of
this
mass
is
used
for
fertilizer,
with
the
remainder
being
used
to
synthesize
nitrogen-containing
chemicals
including
explosives,
plastics,
and
pharmaceuticals
(
88%,
USGS
Mineral
Commodity
Summaries
2020
Nitrogen
).
G.
Ocean Acidity
Surface ocean [H+] ≈ 0.2 parts per billion (HuID:
9
0
4
7
2
)
Annual change in [H+]
= 0.36 ± 0.03% (HuID:
19394
)
Data
Source(s):
Figures
1-2
of
European
Environment
Agency
report
CLIM
043
(2020).
Original
data
source
of the report is “Global Mean Sea Water pH” from Copernicus Marine Environment Monitoring Service.
Notes:
Reported
value
is
calculated
from
the
global
average
annual
change
in
pH
over
years
1985-2018.
The
average
oceanic
surface
pH
was
8.057
in
2018
and
decreases
annually
by
0.002
units,
giving
a
change
in
[H+]
of
roughly
10
-8.055
-
10
-8.057
4x10
-11
mol/L
or
about
0.4%
of
the
global
average.
[H+]
is
calculated
as
10
-pH
10
-8
mol/L
or
0.2
parts
per
billion
(ppb),
noting
that
[H
2
O]
55
mol/L.
Uncertainty
for
annual
change
is
the
standard error of the mean.
H.
Land Use
Agriculture ≈ 5 × 10
13
m
2
(HuID:
29582
)
Data
Source(s):
Food
and
Agriculture
Organization
(FAO)
of
the
United
Nations
Statistical
Database
(2020)
— Land Use.
Notes:
Agricultural
land
is
defined
as
all
land
that
is
under
agricultural
management
including
pastures,
meadows,
permanent
crops,
temporary
crops,
land
under
fallow,
and
land
under
agricultural
structures
(such
as barns). Reported value corresponds to 2017 estimates by the FAO.
Urban ≈ (6 - 8) × 10
11
m
2
(HuID:
41339
;
39341
)
Data Source(s):
Florczyk et al. 2019 (https://tinyurl.com/yyxxgtll)
and Table 3 of Liu et al. 2018 DOI:
10.1016/j.rse.2018.02.055
Notes:
Urban land area is determined from satellite
imagery. An area is determined to be “urban” if the total
population is greater than 5,000 and has a minimum population density of 300 people per km
2
. Reported
value gives the range of recent measurements of ≈ 6.5×10
11
m
2
(2015) and ≈ (7.5 ± 1.5) ×10
11
m
2
(2010) from
Florczyk et al. 2019 and Liu et al. 2018, respectively.
I.
River Fragmentation
Global fragmented river volume ≈ 6 × 10
11
m
3
(HuID:
61661
)
Data Source(s):
Grill et al. 2019 DOI: 10.1038/s41586-019-1111-9
Notes:
Value
corresponds
to
the
water
volume
contained
in
rivers
that
fall
below
the
connectivity
threshold
required
to
classify
them
as
free-flowing.
Value
considers
only
rivers
with
upstream
catchment
areas
greater
than
10
km
2
or
discharge
volumes
greater
than
0.1
m
3
per
second.
The
ratio
of
global
river
volume
in
disrupted
rivers
to
free-flowing
rivers
is
approximately
0.9.
The
exact
value
depends
on
the
cutoff
used
to
define a “free-flowing” river. We direct the reader to the source for thorough detail.
J.
Human Population
Urban population ≈ 55% (HuID:
93995
)
Global population ≈ 7.6 × 10
9
people (HuID:
85255
)
Data
Source(s):
Food
and
Agricultural
Organization
(FAO)
of
the
United
Nations
Report
on
Annual
Population, 2019.
Notes:
Value
for
total
population
in
2018
comes
from
a
combination
of
direct
population
reports
from
country
governments
as
well
as
inferences
of
underreported
or
missing
data.
The
definition
of
“urban”
differs
between
countries
and
the
data
does
not
distinguish
between
urban
and
suburban
populations
despite
substantive
differences
between
these
land
uses
(Jones
&
Kammen
2013,
DOI:
10.1021/es4034364).
As
explained
by
the
United
Nations
population
division,
"When
the
definition
used
in
the
latest
census
was
not
the
same
as
in
previous
censuses,
the
data
were
adjusted
whenever
possible
so
as
to
maintain
consistency."
Rural
population
is
computed
from
this
fraction
along
with
the
total
human
population,
implying
that
the
total
population is composed only of “urban” and “rural” communities.
K.
Greenhouse Gas Emissions
Anthropogenic CO
2
= (4.25 ± 0.33) × 10
13
kg CO
2
/
yr (HuID:
24789
;
54608
;
98043
;
60670
)
Data
Source(s):
Table
6
of
Friedlingstein
et
al.
2019,
DOI:
10.5194/essd-11-1783-2019.
Original
data
sources
relevant
to
this
study
compiled
in
Friedlingstein
et
al.:
1)
Gilfillan
et
al.
https://energy.appstate.edu/CDIAC
2)
Average
of
two
bookkeeping
models:
Houghton
and
Nassikas
2017
DOI:
10.1002/2016GB005546;
Hansis
et
al.
2015
DOI:
10.1002/2014GB004997.
3)
Dlugokencky
and
Tans,
National
Oceanic
&
Atmospheric
Administration,
Earth
System
Research
Laboratory
(NOAA/ESRL),
https://www.esrl.noaa.gov/gmd/ccgg/trends/global.html
,
[Accessed 3-Nov-2019].
Notes:
Value
corresponds
to
total
CO
2
emissions
from
fossil
fuel
combustion,
industry
(predominantly
cement
production),
and
land-use
change
during
calendar
year
2018.
Emissions
from
land-use
change
are
due
to
the
burning
or
degradation
of
plant
biomass.
In
2018,
roughly
1.88
×
10
13
kg
CO
2
/
yr
accumulated
in
the
atmosphere,
reflecting
the
balance
of
emissions
and
CO
2
uptake
by
plants
and
oceans
(Dlugokencky
and
Tans). Uncertainty corresponds to one standard deviation.
Anthropogenic CH
4
= (3.4 - 3.9) × 10
11
kg CH
4
/ yr
(HuID:
96837
;
30725
)
Data Source(s):
Table 3 of Saunois, et al. 2020. DOI:
10.5194/essd-12-1561-2020.
Notes:
Value
corresponds
to
2008-2017
decadal
average
mass
of
CH
4
emissions
from
anthropogenic
sources.
Includes
emissions
from
agriculture
and
landfill,
fossil
fuels,
and
burning
of
biomass
and
biofuels,
but