Slope
stability
evaluation
is
a
complex
and
uncertain
system
problem,
carrying
out
slope
prediction
the
prerequisite
foundation
for
disaster
prevention.
In
order
to
achieve
fast
accurate
of
stability,
this
paper
considers
height,
total
angle,
unit
weight,
cohesion,
internal
friction
pore
water
pressure
ratio
as
input
features
proposes
an
intelligent
method
based
on
grid
search
optimization
ensemble
learning
model
by
soft
voting
(GSOEM-SV).
First,
390
sets
on-site
data
were
collected
form
dataset,
analyses
including
correlation
coefficients,
density
estimates,
box
lines
carried
out.
Then,
algorithm
used
optimize
hyperparameters
five
algorithms—Gradient
Boosting
Decision
Trees,
Light
Gradient
Machine,
Categorical
Boosting,
Support
Vector
Random
Fores,
integrates
them
through
voting.
Furthermore,
optimizes
above
algorithms
search,
particle
swarm
simulated
annealing
algorithms,
builds
15
improved
models
2
conducts
comparison.
The
results
reveal
that
GSOEM-SV
has
highest
accuracy,
up
91%,
area
under
curve
(AUC)
0.950,
its
F1
score
0.917,
which
are
better
than
integrated
models.
addition,
set
APP
uni-app
developed
in
paper.
It
provides
technical
open
shareable
information
service
platform
hazard
geotechnical
engineering.
Applied Sciences,
Год журнала:
2025,
Номер
15(6), С. 3139 - 3139
Опубликована: Март 13, 2025
Sinkholes,
naturally
occurring
formations
in
karst
regions,
represent
a
significant
environmental
hazard,
threatening
infrastructure,
agricultural
lands,
and
human
safety.
In
recent
years,
machine
learning
(ML)
techniques
have
been
extensively
employed
for
sinkhole
susceptibility
mapping
(SSM).
However,
the
lack
of
explainability
inherent
these
methods
remains
critical
issue
decision-makers.
this
study,
Konya
Closed
Basin
was
mapped
using
an
interpretable
model
based
on
SHapley
Additive
exPlanations
(SHAP).
The
Random
Forest
(RF),
eXtreme
Gradient
Boosting
(XGBoost),
Light
Machine
(LightGBM)
algorithms
were
employed,
interpretability
results
enhanced
through
SHAP
analysis.
Among
compared
models,
RF
demonstrated
highest
performance,
achieving
accuracy
95.5%
AUC
score
98.8%,
consequently
selected
development
final
map.
analyses
revealed
that
factors
such
as
proximity
to
fault
lines,
mean
annual
precipitation,
bicarbonate
concentration
difference
are
most
variables
influencing
formation.
Additionally,
specific
threshold
values
quantified,
effects
contributing
analyzed
detail.
This
study
underscores
importance
employing
eXplainable
Artificial
Intelligence
(XAI)
natural
hazard
modeling,
SSM
example,
thereby
providing
decision-makers
with
more
reliable
comparable
risk
assessment.
PLoS ONE,
Год журнала:
2025,
Номер
20(3), С. e0318835 - e0318835
Опубликована: Март 28, 2025
Accurate
forecasting
of
blowing
snow
events
is
vital
for
improving
numerical
models
processes,
yet
traditional
predictive
methods
often
lack
interpretability.
This
study
leverages
eXtreme
Gradient
Boosting
(XGBoost)
to
detect
using
meteorological
and
flux
monitoring
data
from
three
weather
stations
in
the
Alps.
Through
5-fold
cross-validation,
model
achieved
impressive
performance
metrics,
with
precision
rates
exceeding
0.94
non-blowing
0.77-0.80
events.
The
SHAP
framework
was
employed
analyze
relative
importance
factors,
revealing
that
maximum
wind
speed
(WS-MAX),
average
(WS-AVG),
air
temperature
(AT),
humidity
(AH)
are
most
influential
factors.
Additionally,
Partial
dependence
plots
(PDP)
demonstrated
a
linear
correlation
between
increased
WS-MAX
probability
snow,
while
WS-AVG
showed
diminishing
returns
beyond
10
m/s.
Notably,
AT
below
-3°C
strongly
correlates
occurrence,
whereas
above
exhibits
negative
relationship.
Relative
plays
significant
role,
values
60%
stabilizing
peaking
near
100%.
research
contributes
drifting
event
dynamics
by
integrating
explainable
artificial
intelligence
techniques
(XAI),
thereby
interpretability
supporting
data-driven
decision-making
applications.
The Science of The Total Environment,
Год журнала:
2023,
Номер
896, С. 165221 - 165221
Опубликована: Июнь 29, 2023
Snow
avalanches
are
gravitational
processes
characterised
by
the
rapid
movement
of
a
snow
mass,
threatening
inhabitants
and
damaging
infrastructure
in
mountain
areas.
Such
phenomena
complex
events,
for
this
reason,
different
numerical
models
have
been
developed
to
reproduce
their
dynamics
over
given
topography.
In
study,
we
focus
on
two-dimensional
simulation
tools
RAMMS::AVALANCHE
FLO-2D,
aiming
compare
performance
predicting
deposition
area
avalanches.
We
also
aim
assess
employment
FLO-2D
model,
normally
used
water
flood
or
mud/debris
flow
simulations,
motion
For
purpose,
two
well-documented
avalanche
events
that
occurred
Province
Bolzano
(IT)
were
analyzed
(Knollgraben,
Pichler
Erschbaum
avalanches).
The
each
case
study
was
simulated
with
both
through
back-analysis
processes.
results
evaluated
primarily
comparing
observed
one
statistical
indices.
Subsequently,
maximum
depth,
velocity
depth
compared
between
results.
showed
generally
reproduced
deposits
better
simulation.
provided
suitable
wet
dry
after
meticulous
calibration
rheological
parameters,
since
they
not
those
typically
considered
rheology
studies.
can
be
propagation
could
adopted
practitioners
define
hazard
areas,
expanding
its
field
application.
Atmosphere,
Год журнала:
2024,
Номер
15(11), С. 1343 - 1343
Опубликована: Ноя. 9, 2024
Snow
avalanches,
one
of
the
most
severe
natural
hazards
in
mountainous
regions,
pose
significant
risks
to
human
lives,
infrastructure,
and
ecosystems.
As
climate
change
accelerates
shifts
snowfall
temperature
patterns,
it
is
increasingly
important
improve
our
ability
monitor
predict
avalanches.
This
review
explores
use
remote
sensing
technologies
understanding
key
geomorphological,
geobotanical,
meteorological
factors
that
contribute
avalanche
formation.
The
primary
objective
assess
how
can
enhance
risk
assessment
monitoring
systems.
A
systematic
literature
was
conducted,
focusing
on
studies
published
between
2010
2025.
analysis
involved
screening
relevant
sensing,
dynamics,
data
processing
techniques.
Key
sources
included
satellite
platforms
such
as
Sentinel-1,
Sentinel-2,
TerraSAR-X,
Landsat-8,
combined
with
machine
learning,
fusion,
detection
algorithms
process
interpret
data.
found
significantly
improves
by
providing
continuous,
large-scale
coverage
snowpack
stability
terrain
features.
Optical
radar
imagery
enable
crucial
parameters
like
snow
cover,
slope,
vegetation
influence
risks.
However,
challenges
limitations
spatial
temporal
resolution
real-time
were
identified.
Emerging
technologies,
including
microsatellites
hyperspectral
imaging,
offer
potential
solutions
these
issues.
practical
implications
findings
underscore
importance
integrating
ground-based
observations
for
more
robust
forecasting.
Enhanced
fusion
techniques
will
disaster
management,
allowing
quicker
response
times
effective
policymaking
mitigate
avalanche-prone
regions.
Water,
Год журнала:
2024,
Номер
16(22), С. 3247 - 3247
Опубликована: Ноя. 12, 2024
The
focus
of
this
study
is
to
introduce
a
hybrid
predictive
framework
encompassing
different
meta-heuristic
optimization
and
machine
learning
techniques
identify
the
regions
susceptible
snow
avalanches.
To
accomplish
aim,
present
research
sought
acquire
best-performed
model
among
nine
scenarios
three
meta-heuristics,
namely
particle
swarm
(PSO),
gravitational
search
algorithm
(GSA),
Cuckoo
Search
(CS),
ML
approaches,
i.e.,
support
vector
classification
(SVC),
stochastic
gradient
boosting
(SGB),
k-nearest
neighbors
(KNN),
pertaining
families.
According
diligent
analysis
performed
with
regard
blinded
testing
set,
PSO-SGB
illustrated
most
satisfactory
performance
an
accuracy
0.815,
while
precision
recall
were
found
be
0.824
0.821,
respectively.
F1-score
predictions
was
area
under
receiver
operating
curve
(AUC)
obtained
0.9.
Despite
attaining
similar
success
via
CS-SGB
model,
time-efficiency
underscored
PSO-SGB,
as
corresponding
process
consumed
considerably
less
computational
time
compared
its
counterpart.
SHapley
Additive
exPlanations
(SHAP)
implementation
further
informed
that
slope,
elevation,
wind
speed
are
contributing
attributes
detecting
avalanche
susceptibility
in
French
Alps.