Heliyon,
Journal Year:
2024,
Volume and Issue:
10(16), P. e35987 - e35987
Published: Aug. 1, 2024
Rivers
worldwide
are
warming
due
to
the
impact
of
climate
change
and
human
interventions.
This
study
investigated
river
heatwaves
in
Vistula
River
Basin,
one
largest
systems
Europe
using
long-term
observed
daily
water
temperatures
from
past
30
years
(1991-2020).
The
results
showed
that
increased
frequency
intensity
Basin.
total
number
clear
increasing
trend
with
an
average
rate
1.400
times/decade,
duration
at
14.506
days/decade,
cumulative
53.169
°C/decade.
Mann-Kendall
(MK)
test
was
also
employed,
showing
statistically
significant
trends
number,
duration,
for
all
rivers,
including
main
watercourse
its
tributaries,
few
exceptions.
Air
temperature
is
major
controller
each
hydrological
station,
increase
air
temperatures,
will
intensity.
Another
impacting
factor
flow,
tend
decrease
suggested
mitigation
measures
shall
be
taken
reduce
effect
on
systems.
Frontiers in Environmental Science,
Journal Year:
2024,
Volume and Issue:
12
Published: July 9, 2024
Assessing
water
quality
becomes
imperative
to
facilitate
informed
decision-making
concerning
the
availability
and
accessibility
of
resources
in
Korattur
Lake,
Chennai,
India,
which
has
been
adversely
affected
by
human
actions.
Although
numerous
state-of-the-art
studies
have
made
significant
advancements
classification,
conventional
methods
for
training
machine
learning
model
parameters
still
require
substantial
material
resources.
Hence,
this
study
employs
stochastic
gradient
descent
(SGD),
adaptive
boosting
(AdaBoosting),
Perceptron,
artificial
neural
network
algorithms
classify
categories
as
these
well-established
methods,
combined
with
Bayesian
optimization
hyperparameter
tuning,
provide
a
robust
framework
demonstrate
performance
enhancements
classification.
The
input
features
from
2010
2019
comprise
such
pH,
phosphate,
total
dissolved
solids
(TDS),
turbidity,
nitrate,
iron,
chlorides,
sodium,
chemical
oxygen
demand
(COD).
is
employed
dynamically
tune
hyperparameters
different
select
optimal
best
performance.
Comparing
algorithms,
AdaBoosting
exhibits
highest
level
indicated
its
superior
accuracy
(100%),
precision
recall
F1
score
(100%).
top
four
important
factors
classification
are
COD
(0.684),
phosphate
(0.119),
iron
(0.112),
TDS
(0.084).
Additionally,
variations
or
changes
levels
likely
coincide
similar
levels.
Journal of Hydraulic Research,
Journal Year:
2024,
Volume and Issue:
62(4), P. 370 - 382
Published: July 3, 2024
Deep
learning
(DL)
is
a
powerful
tool
that
has
proven
highly
effective
in
many
applications,
but
creating
new
deep
models
becoming
increasingly
challenging.
However,
some
fields,
such
as
fluid
dynamics,
theoretical
can
help
design
DL
models.
Based
on
the
existing
air2water
(A2W)
model,
this
paper
proposes
hybrid
neural
network
DL-A2W,
which
combines
long
short-term
memory
(LSTM)
with
A2W
model
to
predict
lake
water
temperature.
The
DL-A2W
was
established
using
datasets
of
UK
Centre
for
Ecology
&
Hydrology,
and
performance
evaluated
through
three
experiments.
Compared
other
models,
lowest
mean
absolute
error
(MAE),
percent
(MAPE),
root
squared
(RMSE),
highest
Nash-Sutcliffe
efficiency
coefficient
(NSC)
at
any
given
prediction
step.
values
MAE,
MAPE,
RMSE
NSC
test
set
were
0.223–0.388,
1.946–3.296%,
0.375–0.647
0.985–0.995,
respectively.
results
show
good
generalization
ability
portability,
accurately
perform
multi-step
ahead
