Examining innovative unsupervised learning techniques for automated characterization of complex groundwater systems
Results in Engineering,
Journal Year:
2024,
Volume and Issue:
23, P. 102594 - 102594
Published: July 26, 2024
This
research
proposes
an
innovative
approach
utilizing
geophysical
well
logging
data
analyzed
with
multiple
machine
learning
(ML)
methods
including,
self-organizing
maps
(SOMs),
k-means
cluster
analysis
(CA),
and
most
frequent
value-assisted
(MFV-CA),
to
automatically
identify
lithological
variation
within
a
complex
groundwater
system.
The
MFV
method
is
introduced
enhance
center
identification.
In
MFV,
automatedly
weighted
Euclidean
distance
applied
in
which
closer
points
are
assigned
higher
weights
that
emphasize
their
proximity
clusters.
proved
be
efficient,
however,
ensure
robustness,
histogram-based
selection
employed
for
initial
positioning
minimize
the
risk
of
choosing
inappropriate
starting
value.
proposed
methodology
tested
on
Quaternary
aquifer
system
Debrecen
area,
Eastern
Hungary
known
its
high
heterogeneity.
results
were
evaluated
both
1D
2D
reveal
vertical
horizontal
distribution
lithofacies.
Accordingly,
MFV-CA
demonstrated
exceptional
noise
rejection
capabilities
efficient
recognition
Moreover,
continuous
estimation
hydraulic
parameters
including
conductivity
critical
flow
velocity
along
hydrostratigraphical
units
showed
close
agreement
identified
units.
Language: Английский
Integrating geospatial, hydrogeological, and geophysical data to identify groundwater recharge potential zones in the Sulaymaniyah basin, NE of Iraq
Scientific Reports,
Journal Year:
2025,
Volume and Issue:
15(1)
Published: March 22, 2025
Abstract
Groundwater
is
a
critical
resource
for
sustaining
human
activities,
particularly
in
urban
areas,
where
its
importance
exaggerated
by
growing
water
demands,
expansion,
and
industrial
activities.
Ensuring
future
security
necessitates
an
in-depth
understanding
of
groundwater
recharge
dynamics,
which
are
often
complex
influenced
rapid
urbanization.
The
alarming
decline
resources
both
rural
regions
underscore
the
urgency
advanced
management
strategies.
However,
identifying
evaluating
potential
zones
(GWPZs)
remains
challenge
due
to
dynamic
interplay
hydrogeological
development
factors.
This
study
employs
integrated
approach
combining
geographic
information
system
(GIS),
remote
sensing,
multi-criteria
decision
analysis
using
analytical
hierarchy
process
(MCDA-AHP)
delineate
GWPZs
Sulaymaniyah
Basin
(SB).
methodology
further
supported
data
validated
through
geophysical
investigation
electrical
resistivity
tomography
(ERT)
data.
For
MCDA-AHP,
six
thematic
layers
including
rainfall,
geology,
lineament
density,
slope,
drainage
land
use/land
cover
were
derived
from
satellite
imagery,
geological
surveys,
well
These
ranked
based
on
their
relative
influence
GIS-based
weighted
overlay
generate
maps.
results
identified
three
recharge:
low
(11.26%),
moderate
(45.51%),
high
(43.23%).
Validation
ERT
receiver
operating
characteristics
(ROC)
revealed
strong
agreement,
with
area
under
curve
(AUC)
accuracy
86%.
findings
demonstrate
robustness
approach,
providing
reliable
tool
minimizing
hydrogeophysical
exploration
costs
reducing
number
unsuccessful
boreholes.
Language: Английский
Groundwater exploration, management strategies and sustainability: Geophysical approaches
Geosystems and Geoenvironment,
Journal Year:
2025,
Volume and Issue:
unknown, P. 100395 - 100395
Published: April 1, 2025
Language: Английский
Isotopic and Geophysical Investigations of Groundwater in Laiyuan Basin, China
Weiqiang Wang,
No information about this author
Zhiwei Meng,
No information about this author
Chenglong Wang
No information about this author
et al.
Sensors,
Journal Year:
2024,
Volume and Issue:
24(21), P. 7001 - 7001
Published: Oct. 31, 2024
Due
to
the
complex
intersection
and
control
of
multiple
structural
systems,
hydrogeological
conditions
Laiyuan
Basin
in
China
are
complex.
The
depth
research
on
relationship
between
geological
structure
groundwater
migration
needs
be
improved.
supply
each
aquifer
is
still
uncertain.
This
paper
systematically
conducts
characteristics
hydrogen
oxygen
isotopes,
combines
magnetotelluric
impedance
tensor
decomposition
two-dimensional
fine
inversion
technology
carry
out
exploration
strata
structures
Basin,
as
well
comprehensive
replenishment.
results
indicate
following:
(i)
values
all
fall
near
local
meteoric
water
line,
indicating
that
precipitation
main
recharge
source.
(ii)
excess
deuterium
decreased
gradually
from
karst
mountain
basin,
pore
experienced
different
flow
processes.
(iii)
three
runoff
channels
described
by
MT
processing
techniques.
Finally,
it
concluded
limestone
moved
discharge
area,
mixed
with
deep
dolomite
along
fault
under
F2,
eventually
rose
surface
unconsolidated
sediment
blocked
F1
emerge
into
an
ascending
spring.
Language: Английский