Extreme
urban
temperatures
have
emerged
as
crucial
threats
to
ecosystems
and
sustainable
development.
Against
this
background,
we
developed
a
Random
Forest
(RF)
model
by
means
of
eXplainable
Artificial
Intelligence
(XAI)
examine
the
contributions
various
impact
features
in
regulating
land
surface
temperature
(LST)
study
area
Beijing,
China.
Multiple
data
sources
were
investigated,
including
LST,
Normalized
Difference
Vegetation
Index
(NDVI),
cover,
elevation,
tree
height,
building
height.
A
grid
(inner
outer
cities),
composed
3416
boxes,
3x3
km,
was
used
extract
mean
values
features.
RF
outperformed
Linear
Regression
(R2
0.89
vs
0.83)
predicting
demonstrating
complex
non-linear
relationships
between
LST
By
applying
method,
our
results
suggest
that
major
Beijing
elevation
(44.19%),
compactness
impervious
(17.27%),
NDVI
(11.12%),
proportion
(8.04%),
height
(3.83%).
The
relationship
highlights
need
for
systematic
planning
landscapes.
This
provides
state-of-the-art
technology
gain
novel
insights
into
managing
green
spaces,
development
mitigate
hot
environments.
Land Degradation and Development,
Год журнала:
2024,
Номер
35(14), С. 4330 - 4342
Опубликована: Июль 4, 2024
Abstract
This
research
provides
a
brief
insight
into
the
spatial
nature
of
urban
sprawl
in
functional
areas
(FUAs)
European
capitals,
as
opposed
to
most
investigations
growth
that
focus
on
case
studies.
Its
purpose
is
identify
capitals
grow
and
characterize
aspect
dynamics
ensuing
land‐cover
change
between
2006,
2012,
2018.
We
employed
open
data
from
Urban
Atlas,
which
we
processed
with
analyses
NUASI
(Normalized
Atlas
Sprawl
Indicator)
quantify
scale
investigated
areas.
The
results
demonstrate
dynamic
for
Central
Eastern
Iberian
Peninsula
while
it
seems
be
slowing
down
other
Western
Balkan
countries.
Moreover,
found
out
FUAs
differ
terms
pace
uncontrolled
sprawl.
our
comparative
are
relevant
development
because
they
can
exhibiting
various
scales
paces
take
targeted
actions
depending
needs
defined
by
decision‐makers
strategies.
Sustainability,
Год журнала:
2024,
Номер
16(16), С. 7030 - 7030
Опубликована: Авг. 16, 2024
Understanding
the
relationship
between
urban
form
and
shrinkage
is
crucial
for
developing
sustainable
policies,
particularly
in
medium-sized
cities
facing
demographic
economic
challenges.
This
study
investigates
complex
Polish
(population
of
20,000
to
100,000),
highlighting
implications
sustainability.
Utilising
a
comprehensive
multi-factor
approach,
it
analyses
growth
trends
over
15
years
(2006–2021)
by
establishing
shrinkage/growth
score
based
on
social,
demographic,
factors
each
city.
It
examines
spatial
aspects,
compactness
population
density,
using
Corine
Land
Cover
(CLC)
data,
making
methodology
applicable
areas
across
Europe.
The
findings
reveal
no
significant
overall
correlation
all
cities.
However,
positive
exists
within
“urban
municipalities”,
indicating
that
less
compact
tend
experience
more
shrinkage.
Additionally,
temporary
negative
density
was
observed
from
2006
2016,
which
shifted
trend
municipalities”
2016
2021.
These
results
highlight
shrinkage’s
dynamic
nature
its
potential
ties
form.
concludes
with
recommendations
policymakers
planners
regarding
dense
strategies
mitigate
adverse
effects
enhance
resilience
While
change,
highlights
need
further
analysis
these
relationships.
Extreme
urban
temperatures
have
emerged
as
crucial
threats
to
ecosystems
and
sustainable
development.
Against
this
background,
we
developed
a
Random
Forest
(RF)
model
by
means
of
eXplainable
Artificial
Intelligence
(XAI)
examine
the
contributions
various
impact
features
in
regulating
land
surface
temperature
(LST)
study
area
Beijing,
China.
Multiple
data
sources
were
investigated,
including
LST,
Normalized
Difference
Vegetation
Index
(NDVI),
cover,
elevation,
tree
height,
building
height.
A
grid
(inner
outer
cities),
composed
3416
boxes,
3x3
km,
was
used
extract
mean
values
features.
RF
outperformed
Linear
Regression
(R2
0.89
vs
0.83)
predicting
demonstrating
complex
non-linear
relationships
between
LST
By
applying
method,
our
results
suggest
that
major
Beijing
elevation
(44.19%),
compactness
impervious
(17.27%),
NDVI
(11.12%),
proportion
(8.04%),
height
(3.83%).
The
relationship
highlights
need
for
systematic
planning
landscapes.
This
provides
state-of-the-art
technology
gain
novel
insights
into
managing
green
spaces,
development
mitigate
hot
environments.
