Atmospheric chemistry and physics,
Год журнала:
2023,
Номер
23(17), С. 9837 - 9852
Опубликована: Сен. 5, 2023
Abstract.
Given
the
increasing
complexity
of
chemical
composition
PM2.5,
identifying
and
quantitatively
assessing
contributions
pollution
sources
has
played
an
important
role
in
formulating
policies
to
control
particle
pollution.
This
study
provides
a
comprehensive
assessment
between
PM2.5
characteristics,
sources,
health
risks
based
on
sampling
data
conducted
over
1
year
(March
2018
February
2019)
Nanjing.
Results
show
that
exhibits
distinct
variation
across
different
seasons,
which
is
primarily
driven
by
emissions,
meteorological
conditions,
conversion
gaseous
pollutants.
First,
mass
reconstruction
shows
secondary
inorganic
aerosols
(62.5
%)
carbonaceous
(21.3
contributed
most
mass.
The
oxidation
rates
SO2
NO2
from
summer
winter
indicate
transformation
pollutants
strongly
positively
correlated
with
relative
humidity.
Second,
positive
matrix
factorization
(PMF)
method
identified
include
aerosol
source
(SIS,
42.5
%),
coal
combustion
(CC,
22.4
industry
(IS,
17.3
vehicle
emission
(VE,
10.7
fugitive
dust
(FD,
5.8
other
(1.3
%).
Hybrid
Single-Particle
Lagrangian
Integrated
Trajectory
(HYSPLIT)
model
concentration-weighted
trajectory
(CWT)
analysis
are
used
further
explore
spatial
distributions
regional
transport
sources.
concentrations
(10–11
µg
m−3)
SIS
CC
distribute
Nanjing
central
China
winter.
(8–10
IS
VE
potentially
located
north
Jiangsu,
Anhui,
Jiangxi.
Finally,
risk
indicates
carcinogenic
non-carcinogenic
toxic
elements
(Cr,
As,
Ni,
Mn,
V,
Pb)
mainly
come
IS,
VE,
CC,
within
tolerance
or
acceptable
level.
Although
main
at
present,
we
should
pay
more
attention
burden
combustion,
industrial
processes.
Environmental Science & Technology Letters,
Год журнала:
2022,
Номер
9(2), С. 112 - 119
Опубликована: Янв. 4, 2022
Many
places
on
earth
still
suffer
from
a
high
level
of
atmospheric
fine
particulate
matter
(PM2.5)
pollution.
Formation
pollution
event
or
haze
episode
(HE)
involves
many
factors,
including
meteorology,
emissions,
and
chemistry.
Understanding
the
direct
causes
key
drivers
behind
HE
is
thus
essential.
Traditionally,
this
done
via
chemical
transport
models.
However,
substantial
uncertainties
are
introduced
into
model
estimation
when
there
significant
changes
in
emissions
inventory
due
to
interventions
(e.g.,
COVID-19
lockdown).
Here
we
applied
Random
Forest
coupled
with
Shapley
additive
explanation
algorithm,
post
hoc
technique,
investigate
roles
major
meteorological
primary
chemistry
five
severe
HEs
that
occurred
before
during
lockdown
China.
We
discovered
that,
addition
PM2.5
these
episodes
was
largely
driven
by
effects
(with
average
contributions
30–65
μg
m–3
for
HEs),
followed
(∼15–30
m–3).
Photochemistry
likely
pathway
formation
nitrate,
while
air
humidity
predominant
factor
forming
sulfate.
Our
results
highlight
machine
learning
data
has
potential
be
complementary
tool
predicting
interpreting
Atmosphere,
Год журнала:
2021,
Номер
12(12), С. 1625 - 1625
Опубликована: Дек. 6, 2021
In
recent
years,
haze
pollution
is
frequent,
which
seriously
affects
daily
life
and
production
process.
The
main
factors
to
measure
the
degree
of
smoke
are
concentrations
PM2.5
PM10.
Therefore,
it
great
significance
study
prediction
PM2.5/PM10
concentration.
Since
PM10
concentration
data
time
series,
their
characteristics
should
be
considered
in
prediction.
However,
traditional
neural
network
limited
by
its
own
structure
has
some
weakness
processing
related
data.
Recurrent
a
kind
specially
used
for
sequence
modeling,
that
is,
current
output
correlated
with
historical
output.
this
paper,
model
established
based
on
deep
recurrent
network.
We
obtained
air
Chengdu
from
China
Air
Quality
Online
Monitoring
Analysis
Platform,
conducted
experiments
these
results
show
new
method
can
predict
smog
more
effectively
accurately,
social
economic
purposes.
Geophysical Research Letters,
Год журнала:
2021,
Номер
48(11)
Опубликована: Июнь 2, 2021
Abstract
Responding
to
the
2020
COVID‐19
outbreak,
China
imposed
an
unprecedented
lockdown
producing
reductions
in
air
pollutant
emissions.
However,
driven
pollution
changes
have
not
been
fully
quantified.
We
applied
machine
learning
quantify
effects
of
meteorology
on
surface
quality
data
31
major
Chinese
cities.
The
meteorologically
normalized
NO
2
,
O
3
and
PM
2.5
concentrations
changed
by
−29.5%,
+31.2%,
−7.0%,
respectively,
after
began.
part
this
effect
was
also
associated
with
emission
due
Spring
Festival,
which
led
∼14.1%
decrease
∼6.6%
increase
a
mixed
studied
cities
that
largely
resulted
from
festival
fireworks.
After
decoupling
weather
Festival
effects,
attributable
were
much
smaller:
−15.4%,
+24.6%,
−9.7%
for
respectively.
Environmental Research,
Год журнала:
2022,
Номер
216, С. 114386 - 114386
Опубликована: Сен. 24, 2022
Volatile
organic
compound
(VOC)
emissions
have
attracted
wide
attention
due
to
their
impacts
on
atmospheric
quality
and
public
health.
However,
most
studies
reviewed
certain
aspects
of
natural
VOCs
(NVOCs)
or
anthropogenic
(AVOCs)
rather
than
comprehensively
quantifying
the
hotspots
evolution
trends
AVOCs
NVOCs.
We
combined
bibliometric
method
with
tree
Markov
chain
identify
research
focus
uncover
in
VOC
emission
sources.
This
study
found
that
mainly
focused
characteristics,
effects
air
health,
under
climate
change.
More
concerned
NVOCs,
AVOC
shifted
a
decreasing
proportion
transport
an
increasing
share
solvent
utilization
countries
high
publications
(China
USA).
Research
is
imperative
develop
efficient
economical
abatement
techniques
specific
sources
BTEX
species
mitigate
detrimental
effects.
NVOCs
originating
from
human
risen
application
medicine,
while
sensitive
change
grew
slowly,
including
plants,
biomass
burning,
microbes,
soil
oceans.
long-term
responses
derived
various
warming
warranted
explore
feedback
global
climate.
It
worthwhile
establish
inventory
all
kinds
sources,
accurate
estimation,
spatial
temporal
resolution
capture
synergy
industrialization
as
well
simulate
quality.
review
both
health
point
out
directions
for
comprehensive
control
mitigation
O3
pollution.
