npj Climate and Atmospheric Science,
Journal Year:
2024,
Volume and Issue:
7(1)
Published: Jan. 6, 2024
Abstract
Accurately
estimating
the
concentration
of
carbon
monoxide
(CO)
with
high
spatiotemporal
resolution
is
crucial
for
assessing
its
meteorological-environmental-health
impacts.
Although
machine
learning
models
have
predictive
ability
in
environmental
research,
there
are
relatively
few
explanations
model
outputs.
Utilizing
top-of-atmosphere
radiation
data
China’s
new
generation
geostationary
satellites
(FY-4A
and
FY-4B)
interpretable
models,
24-hour
near-surface
CO
concentrations
China
was
conducted
(resolution:
1
hour,
0.04°).
The
improved
by
6.6%
when
using
all-sky
dataset
(cloud-contained
model,
R
2
=
0.759)
compared
to
clear-sky
(cloud-removed
model).
interpretability
analysis
estimation
used
two
methods,
namely
ante-hoc
(model
feature
importance)
post-hoc
(SHapley
Additive
exPlanations).
importance
daytime
meteorological
factors
increased
51%
nighttime.
Combining
partial
dependency
plots,
impact
key
on
elucidated
gain
a
deeper
understanding
variations
CO.
Journal of Geophysical Research Atmospheres,
Journal Year:
2022,
Volume and Issue:
127(9)
Published: April 18, 2022
Abstract
The
rapid
urbanization
in
China
and
the
long‐range
transport
dust
(LRTD)
from
arid
semi‐arid
areas
has
resulted
an
increase
of
PM
10
concentration.
In
this
study,
interpretable
deep
learning
model
[deep
forest
(DF)]
with
FY‐4A
top‐of‐the‐atmosphere
reflectance
(TOAR)
data
were
used
to
obtain
hourly
China.
optimal
average
R
2
10‐fold
cross
validation
can
achieve
0.85
(13:00
Beijing
time);
(RMSE,
μg/m³)
daily,
monthly,
annual
averages
0.82
(24.16),
0.97
(6.53),
0.99
(2.30),
respectively.
Using
TOAR
data,
DF
performed
better
than
other
machine
models.
feature
importance
TOAR‐PM
showed
that
meteorological
elements
both
contributed
significantly
model.
spring,
northern
was
greater
southern
China,
which
may
be
related
LRTD.
Excluding
weather
periods,
high
values
mainly
cities
their
suburbs,
where
correlated
human
activities.
During
a
process,
LRTD
increased
by
80.4%.
mixture
haze
130.2%
led
73.7%.
sources
(from
Taklimakan
Desert
China)
transmission
paths
these
two
processes
similar.
contribution
intensity
conditions.
results
local
pollution
important
periods.
Atmospheric Research,
Journal Year:
2022,
Volume and Issue:
274, P. 106199 - 106199
Published: April 15, 2022
Due
to
urbanization
and
industrialization,
PM2.5
(particulate
matter
with
a
diameter
less
than
2.5
μm)
pollution
has
become
serious
environmental
problem.
The
low
spatial
resolution
insufficient
coverage
of
observation
stations
affect
research
on
causes
human
health
risks.
With
the
launch
FY-4A,
new
generation
Chinese
geostationary
weather
satellites,
it
is
possible
obtain
high
temporal
covering
all
China.
In
this
study,
FY-4A
top-of-the-atmosphere
reflectance
data,
meteorological
factors,
geographic
information
were
input
into
deep
forest
(DF)
model
hourly
in
samples
based
10-fold
cross
validation
DF
an
R2
0.83–0.88,
root
mean
square
error
8.81–14.7
μg/m3,
while
result
sites
was
0.77.
monthly
(R2
=
0.98)
seasonal
0.99)
estimated
results
showed
consistency
observations.
Feature
importance
that
contribution
features
varies
regions
seasons.
Estimation
indicated
substantial
spatiotemporal
differences
PM2.5,
highest
between
09:00–10:00
then
gradually
decreased.
Regions
China
mainly
distributed
Tarim
Basin
Central
assessment
that:
1)
more
80%
winter
days
higher
World
Health
Organization
interim
target
3
(37.5
μg/m3);
2)
bimodal
distribution
there
are
obvious
cities
suburbs;
3)
autumn
winter,
where
population-weighted
IT-3
Beijing-Tianjin-Hebei,
China,
Guanzhong
Plain,
Sichuan
Basin,
Yangtze
River
Delta.
Our
advantages
thus
shows
great
potential
for
estimating
pollutants.
Journal of Geophysical Research Atmospheres,
Journal Year:
2023,
Volume and Issue:
128(14)
Published: July 1, 2023
Abstract
Long‐range
transport
and
vertical
distribution
of
aerosols
are
important
factors
for
assessing
the
uncertainty
in
aerosol
radiative
forcing.
This
paper
reveals
trends
optical
properties
China
using
15
years
Cloud‐Aerosol
Lidar
with
Orthogonal
Polarization
(CALIOP)
data.
The
Hybrid
Single‐Particle
Lagrangian
integrated
trajectory
model
was
used
to
analyze
layer
highest
occurrence
frequencies
dust,
polluted
continental
elevated
smoke
aerosols.
results
indicated
that
(a)
there
were
significant
regional
seasonal
differences
depth
(AOD)
trend
a
given
region
depends
on
changes
type
frequency
corresponding
largest
AOD
profile.
dust
Beijing‐Tianjin‐Hebei
(BTH)
Central
China.
considerable
decrease
0–2
km
led
column
AOD.
(b)
main
types
also
affected
by
sources
long‐range
pathways.
In
BTH,
originated
from
Mongolian
Plateau,
accounting
57.88%
total
trajectories.
Pearl
River
Delta
dominated
aerosols,
trajectories
mainly
originating
Myanmar
Vietnam,
27.38%
29.59%,
respectively.
15‐year
backward
Tibetan
Plateau
India
is
increasing.
