Ecological Chemistry and Engineering S,
Journal Year:
2023,
Volume and Issue:
30(2), P. 167 - 173
Published: June 1, 2023
Abstract
Dynamic
features
from
remote
sensing
photos
may
be
successfully
extracted
using
deep
learning
and
symmetric
network
structure,
which
can
then
used
to
direct
them
carry
out
accurate
classification.
The
DBN
model
more
effectively
extract
since
it
uses
unsupervised
learning.
It
reduced
the
many
Restricted
Boltmann
Machines
(RBM)
training
problem.
In
this
paper,
a
soil
rocky
desertification
(RD)
assessment
based
on
belief
(DBN)
is
created
in
light
of
complicated
influencing
aspects
Karst
RD
risk
encompassing
several
geographical
elements.
builds
upon
conventional
RBM
framework
incorporates
influence
layer
related
elements
as
an
auxiliary
requirement
for
retrieving
Geographic
Information
System
(GIS)
score
data.
Then,
order
forecast
level
desertification,
learns
experimental
results
show
that
proposed
paper
has
better
prediction
performance
faster
convergence
speed,
its
classification
different
degrees
are
consistent
with
actual
results.
Sustainability,
Journal Year:
2025,
Volume and Issue:
17(7), P. 3094 - 3094
Published: March 31, 2025
The
Wenchuan
earthquake,
an
unexpected
magnitude
8.0
mega-earthquake
that
struck
on
12
May
2008,
significantly
changed
land
cover
(LC),
particularly
affecting
vegetation
and
rock
cover.
However,
the
long-term
effects
of
LC
changes
ecosystem
services
(ESs)
remain
unclear
in
earthquake-affected
regions,
especially
across
different
spatial
scales.
This
study,
focusing
County,
employs
a
multi-model
framework
integrates
fractional
coverage
(FVC),
exposure
rate
(FR),
(ESs),
combining
correlation
analysis,
geographically
weighted
regression
(GWR),
Self-organizing
map
(SOM)
clustering,
XGBoost-SHAP
model,
to
analyze
spatiotemporal
dynamics,
interrelationships,
driving
mechanisms
(LC)
ESs
before
after
earthquake.
Results
show
that:
(1)
From
2000
2020,
FVC
FR
fluctuated
markedly
under
earthquake
influence,
with
slight
declines
habitat
quality
(HQ)
carbon
storage
(CS)
notable
improvements
soil
conservation
(SC)
water
yield
(WY).
(2)
With
increasing
elevation,
FVC–CS–SC
group
exhibited
downward
trend
synergy,
while
FR–HQ–WY
increased
also
showed
synergy;
trade-offs
synergies
became
more
pronounced
at
larger
scales,
displaying
strong
heterogeneity.
(3)
Elevation
(explaining
10–60%
variance)
was
main
driver
for
ESs,
use,
slope,
human
activities,
climate,
geological
conditions
impacting
individual
indicators.
At
same
time,
existing
hazard
points
are
mainly
concentrated
along
both
sides
river
valleys,
which
may
be
associated
intensified
human–land
conflicts.
These
findings
offer
valuable
insights
into
ecological
restoration
sustainable
development
regions.
Geography and sustainability,
Journal Year:
2024,
Volume and Issue:
5(2), P. 289 - 301
Published: Jan. 26, 2024
China's
Grain
to
Green
Program
(GTGP),
which
is
one
of
the
largest
payments
for
ecosystem
services
(PES)
in
world,
has
made
significant
ecological
improvements
environment.
However,
current
understanding
its
outcomes
on
social-ecological
system
(SES)
remains
limited.
Therefore,
taking
South
China
Karst
as
an
example,
a
SES
resilience
evaluation
index
was
constructed
followed
by
exploratory
spatial
analysis,
root
mean
square
error,
and
Self-Organizing
Feature
Map
clarify
spatiotemporal
changes
relationship
resilience,
achieve
zoning
provide
restoration
measures.
The
results
showed
upward
trend
social
from
2000
2020,
especially
subsystem
development.
Regional
stable,
owing
slightly
declined
increased
landscape
pattern.
Spatially,
nearly
half
counties
exhibited
distribution
mismatch
resilience.
There
obvious
inverted
U-shaped
indicating
clear
threshold
effect,
constraint
eased
over
time,
demonstrating
effectiveness
program.
GTGP
played
positive
role
reducing
regional
trade-off,
but
this
effect
limited,
reflecting
limitations
overemphasizing
conversion
farmland
forest
grassland.
can
be
divided
into
four
clusters,
were
key
optimization
zone
system,
safety
zone,
system.
Adaptive
adjustments
these
zones
should
taken
maximum
benefits
future.
