Optimizing Air Pollution Forecasting Across Temporal Scales: A Case Study in Salamanca, Mexico
Earth,
Journal Year:
2025,
Volume and Issue:
6(1), P. 9 - 9
Published: Feb. 9, 2025
Air
pollution
forecasting
is
essential
for
understanding
environmental
patterns
and
mitigating
health
risks,
especially
in
urban
areas.
This
study
investigates
the
of
criterion
pollutants—CO,O3,SO2,NO2,PM2.5,
PM10—across
multiple
temporal
frames
(hourly,
daily,
weekly,
monthly)
Salamanca,
Mexico,
utilizing
temporal,
meteorological,
pollutant
data
from
local
monitoring
stations.
The
primary
objective
to
identify
robust
models
capable
short-
mid-term
predictions,
despite
challenges
related
inconsistencies
missing
values.
Leveraging
low-code
PyCaret
framework,
a
benchmark
analysis
was
conducted
best-performing
each
pollutant.
Statistical
evaluations,
including
ANOVA
Tukey
HSD
tests,
were
employed
compare
model
performance
across
different
time
frames.
results
reveal
significant
variations
prediction
accuracy
depending
on
both
windows,
with
stronger
predictive
observed
weekly
monthly
research
indicates
that
incorporation
variables
enhances
forecast
highlights
value
AutoML
tools,
such
as
PyCaret,
streamlining
selection
improving
overall
efficiency.
Language: Английский
Enhancing Environmental Policy Decisions in Korea and Japan Through AI-Driven Air Pollution Forecast
Yushin Kim,
No information about this author
J S Kim,
No information about this author
Sunghyun Cho
No information about this author
et al.
Sustainability,
Journal Year:
2024,
Volume and Issue:
16(23), P. 10436 - 10436
Published: Nov. 28, 2024
(1)
Background:
Although
numerous
artificial
intelligence
(AI)-based
air
pollution
prediction
models
have
been
proposed,
research
that
links
key
drivers,
such
as
regional
industrial
facilities,
to
actionable
policy
recommendations
is
required.
(2)
Methods:
This
study
employs
the
radial
basis
function
(RBF)
and
spatial
lag
features
capture
interactions
among
regions,
utilizing
a
transformer
model
for
analysis.
The
was
trained
on
quality
data
from
South
Korea
(2010–2022)
Japan
(2017–2020).
(3)
Results:
achieved
mean
squared
error
of
0.045
Korean
dataset
0.166
Japanese
dataset,
outperforming
benchmark
models,
including
Support
Vector
Regression,
neural
networks,
AutoRegressive
Integrated
Moving
Average
model.
(4)
Conclusions:
By
capturing
complex
dynamics,
proposed
provides
valuable
insights
can
assist
policymakers
in
developing
effective,
data-driven
strategies
reduction
at
national
levels,
thereby
supporting
broader
goals
sustainability
through
informed,
equitable
environmental
interventions.
Language: Английский