Perfiles,
Год журнала:
2024,
Номер
1(32), С. 63 - 72
Опубликована: Дек. 13, 2024
This
study
evaluates
the
water
quality
of
main
tributary
and
effluent
Santo
Domingo
Lagoon.
The
objectives
were
to
analyze
physical-chemical
biological
characteristics
determine
its
influence
on
quality,
using
ICA-NSF,
ICA-Dinius,
ABI
IBF
indices.
For
this,
samplings
carried
out
during
months
April
June,
evaluating
a
total
15
parameters
identifying
benthic
macroinvertebrates
present
in
bodies.
Physicochemical
indicators
show
slightly
higher
(maximum
value
42.2%)
compared
June
35.5%).
Meanwhile,
evaluation
showed
poor
at
sampling
points
with
values
from
14
21
according
index,
while
index
indicated
excellent
points,
suggesting
discrepancy
between
physicochemical
quality.
suggests
presence
contaminants
that
could
affect
biodiversity
long
term,
so
it
is
important
carry
constant
more
in-depth
monitoring
establish
health
aquatic
ecosystem.
Water Science & Technology Water Supply,
Год журнала:
2025,
Номер
unknown
Опубликована: Янв. 15, 2025
ABSTRACT
The
prediction
of
saltwater
intrusion
in
estuaries
plays
an
important
role,
supporting
decision-makers
or
farmers
building
strategies,
and
also,
policies
made
for
agricultural
development
water
resource
management.
objective
this
study
is
to
develop
machine
learning
models,
namely
gated
recurrent
unit
(GRU),
GRU-GWO
(grey
wolf
optimiser),
GRU-SFO
(sailfish
optimiser
algorithm)
predict
1,
7,
15,
30
days
ahead
the
Mekong
estuary
Vietnam.
Several
statistical
indices,
root
mean
square
error
(RMSE),
absolute
(MAE),
coefficient
determination
(R2),
were
applied
evaluate
accuracy
model.
results
showed
that
GWO
SFO
optimisation
algorithms
successfully
improved
GRU
model
intrusion.
For
a
one
day
forecast,
R2
value
proposed
models
ranged
from
0.89
0.91,
seven
it
0.81
0.85,
15-day
0.67
0.76,
30-day
0.52
0.55.
indicated
ability
