Advances in environmental engineering and green technologies book series,
Год журнала:
2023,
Номер
unknown, С. 115 - 129
Опубликована: Дек. 30, 2023
The
goal
of
the
chapter
on
earthquake
multi-magnificence
detection
and
use
synthetic
intelligence
is
to
discover
exhibit
usage
device
learning
AI
techniques
for
appropriately
efficiently
detecting
different
lessons
earthquakes.
seeks
offer
a
complete
understanding
strategies
spotlight
capability
in
advancing
this
subject.
present
detailed
analysis
existing
recommend
novel
AI-primarily
based
that
could
enhance
category
accuracy
timeliness.
Conventional
seismology
commonly
focus
earthquakes
an
unmarried
seismic
event.
However,
algorithms
can
investigate
significant
quantity
information,
which
includes
ancient
facts,
geological
capabilities,
actual-time
signals,
become
aware
patterns
classify
into
multiple
instructions.
Geological Journal,
Год журнала:
2023,
Номер
58(9), С. 3515 - 3543
Опубликована: Март 16, 2023
Among
several
devastating
natural
hazards,
flooding
is
a
common
and
serious
threat
to
society
causing
huge
loss
of
lives,
properties,
infrastructure
throughout
the
world.
The
intensity
frequency
this
extreme
weather
event
are
expected
increase
due
significant
changes
in
present‐day
climate
land
use
cover
(LULC)
pattern.
India
has
very
systematic
organized
structural
program
policies
but
lacks
proper
implementations,
adverse
effect
change
goes
on
society.
This
paper
an
analysis
floods
hazards
LULC
patterns.
Three
models,
namely
“Eco‐biogeography‐based
optimization
(EBO),
Random
forest
(RF),
Support
vector
machine
(SVM)”
were
used
obtain
final
output
prepare
“Flood
susceptibility
map”.
result
was
validated
through
“Receiver
operating
characteristics
(ROC)”
with
“Area
under
curve
(AUC)”
values.
future
rainfall
scenario
been
estimated
by
considering
“General
circulation
models”
different
“shared
socioeconomic
pathways
(SSPs)”.
values
AUC
0.915
0.887
0.869
(SVM),
respectively.
After
consideration
SSPs,
shows
that
there
increasing
tendency
flood
projected
period.
all
employed
modelling
approaches,
EBO
model
notable
potential
delineating
possible
flood‐prone
regions
for
effective
planning
management.
Decision‐makers
can
benefit
from
country‐specific
information
regional
planner
implement
sustainable
long‐term
measures
overcome
type
hazardous
situation.
Land,
Год журнала:
2023,
Номер
12(6), С. 1151 - 1151
Опубликована: Май 30, 2023
Landslides
along
the
main
roads
in
mountains
cause
fatalities,
ecosystem
damage,
and
land
degradation.
This
study
mapped
susceptibility
to
landslides
Saqqez-Marivan
road
located
Kurdistan
province,
Iran,
comparing
an
ensemble
fuzzy
logic
with
analytic
network
process
(fuzzy
logic-ANP;
FLANP)
TOPSIS
logic-TOPSIS;
FLTOPSIS)
terms
of
their
prediction
capacity.
First,
100
identified
through
field
surveys
were
randomly
allocated
a
70%
dataset
30%
dataset,
respectively,
for
training
validating
methods.
Eleven
landslide
conditioning
factors,
including
slope,
aspect,
elevation,
lithology,
use,
distance
fault,
river,
road,
soil
type,
curvature,
precipitation
considered.
The
performance
methods
was
evaluated
by
inspecting
areas
under
receiver
operating
curve
(AUCROC).
accuracies
0.983
0.938,
FLTOPSIS
FLANP
Our
findings
demonstrate
that
although
both
models
are
known
be
promising,
method
had
better
capacity
predicting
area.
Therefore,
map
developed
is
suitable
inform
management
planning
prone
allocation
development
purposes,
especially
mountainous
areas.
Landslides
around
the
main
roads
in
mountains
not
only
cause
fatal
events
but
also
ecosystem
damage,
including
land
degradation.
This
study
aims
to
map
susceptibility
of
landslides
Saqqez-Marivan
rod
Kurdistan
province,
Iran,
using
ensemble
Fuzzy
logic
with
Analytic
Network
Process
(Fuzzy
Logic-ANP;
FLANP),
and
TOPSIS
Logic-TOPSIS;
FLTOPSIS).
A
total
100
were
first
recognized
by
field
surveys
then
they
randomly
divided
into
a
70%
dataset
(70
locations)
30%
(30
locations),
respectively,
for
training
validating
methods.
Eleven
landslide
conditioning
factors,
slope,
aspect,
elevation,
lithology,
use,
distance
fault,
river,
road,
soil
type,
curvature,
precipitation
used.
The
performance
methods
was
checked
areas
under
receiver
operating
curve
(AUCROC).
Results
concluded
that
prediction
accuracy
based
on
datasets
were,
0.882
0.918
FLANP
FLTOPSIS
Our
findings
demonstrated
although
both
models
known
as
promising
techniques,
method
had
better
capacity
predicting
studied
area.
Therefore,
developed
can
be
used
proper
management
high
potential
managers
planners
during
implementation
allocation
development
projects,
especially
mountainous
areas.
