Remote Sensing,
Год журнала:
2022,
Номер
14(21), С. 5461 - 5461
Опубликована: Окт. 30, 2022
The
ocean
chlorophyll-a
(Chl-a)
concentration
is
an
important
variable
in
the
marine
environment,
abnormal
distribution
of
which
closely
related
to
hazards
red
tides.
Thus,
accurate
prediction
its
East
China
Sea
(ECS)
greatly
for
preventing
water
eutrophication
and
protecting
coastal
ecological
environment.
Processed
by
two
different
pre-processing
methods,
10-year
(2011–2020)
satellite-observed
data
logarithmic
were
used
as
long
short-term
memory
(LSTM)
neural
network
training
datasets
this
study.
2021
comparison
results.
past
15
days’
predict
five
following
days.
Results
showed
that
predictions
obtained
both
methods
could
simulate
seasonal
Chl-a
ECS
effectively.
Moreover,
performance
model
driven
original
values
was
better
medium-
low-concentration
regions.
However,
high-concentration
region,
extreme
concentrations
data-driven
LSTM
models
underestimation,
considering
better.
sensitivity
experiments
accuracy
decreased
considerably
when
backward
time
step
increased.
In
study,
only
chlorophyll-a,
whose
forecasted,
effect
other
relevant
elements
on
not
considered,
current
weakness
Remote Sensing,
Год журнала:
2024,
Номер
16(4), С. 647 - 647
Опубликована: Фев. 9, 2024
In
this
study,
we
employ
in
situ,
meteorological,
and
remote
sensing
data
to
estimate
chlorophyll-a
concentration
at
different
depths
a
South
American
freshwater
ecosystem,
focusing
specifically
on
lake
southern
Chile
known
as
Lake
Maihue.
For
our
analysis,
explored
four
scenarios
using
three
deep
learning
traditional
statistical
models.
These
involved
field
(Scenario
1),
meteorological
variables
2),
satellite
(Scenarios
3.1
3.2)
predict
levels
Maihue
(0,
15,
30
m).
Our
choice
of
models
included
SARIMAX,
DGLM,
LSTM,
all
which
showed
promising
performance
predicting
concentrations
lake.
Validation
metrics
for
these
indicated
their
effectiveness
chlorophyll
levels,
serve
valuable
indicators
the
presence
algae
water
body.
The
coefficient
determination
values
ranged
from
0.30
0.98,
with
DGLM
model
showing
most
favorable
statistics
tested.
It
is
worth
noting
that
LSTM
yielded
comparatively
lower
metrics,
mainly
due
limitations
available
training
data.
employed,
use
machine
data,
have
great
potential
application
lakes
rest
world
similar
characteristics.
addition,
results
constitute
fundamental
resource
decision-makers
protection
conservation
quality.
Water,
Год журнала:
2024,
Номер
16(5), С. 707 - 707
Опубликована: Фев. 28, 2024
Dissolved
oxygen
(DO)
concentration
is
a
pivotal
determinant
of
water
quality
in
freshwater
lake
ecosystems.
However,
rapid
population
growth
and
discharge
polluted
wastewater,
urban
stormwater
runoff,
agricultural
non-point
source
pollution
runoff
have
triggered
significant
decline
DO
levels
Lake
Erie
other
lakes
located
populated
temperate
regions
the
globe.
Over
eleven
million
people
rely
on
Erie,
which
has
been
adversely
impacted
by
anthropogenic
stressors
resulting
deficient
concentrations
near
bottom
Erie’s
Central
Basin
for
extended
periods.
In
past,
hybrid
long
short-term
memory
(LSTM)
models
successfully
used
time-series
forecasting
rivers
ponds.
prediction
errors
tend
to
grow
significantly
with
period.
Therefore,
this
research
aimed
improve
accuracy
taking
advantage
real-time
(water
temperature
concentration)
monitoring
network
establish
temporal
spatial
links
between
adjacent
stations.
We
developed
LSTM
that
combine
LSTM,
convolutional
neuron
(CNN-LSTM),
CNN
gated
recurrent
unit
(CNN-GRU)
models,
(ConvLSTM)
forecast
near-bottom
Basin.
These
their
capacity
handle
complicated
datasets
variability.
can
serve
as
accurate
reliable
tools
help
environmental
protection
agencies
better
access
manage
health
these
vital
Following
analysis
21-site
dataset
2020
2021,
ConvLSTM
model
emerged
most
reliable,
boasting
an
MSE
0.51
mg/L,
MAE
0.42
R-squared
0.95
over
12
h
range.
The
foresees
future
hypoxia
Erie.
Notably,
site
713
holds
significance
indicated
outcomes
derived
from
Shapley
additive
explanations
(SHAP).
Remote Sensing,
Год журнала:
2022,
Номер
14(21), С. 5461 - 5461
Опубликована: Окт. 30, 2022
The
ocean
chlorophyll-a
(Chl-a)
concentration
is
an
important
variable
in
the
marine
environment,
abnormal
distribution
of
which
closely
related
to
hazards
red
tides.
Thus,
accurate
prediction
its
East
China
Sea
(ECS)
greatly
for
preventing
water
eutrophication
and
protecting
coastal
ecological
environment.
Processed
by
two
different
pre-processing
methods,
10-year
(2011–2020)
satellite-observed
data
logarithmic
were
used
as
long
short-term
memory
(LSTM)
neural
network
training
datasets
this
study.
2021
comparison
results.
past
15
days’
predict
five
following
days.
Results
showed
that
predictions
obtained
both
methods
could
simulate
seasonal
Chl-a
ECS
effectively.
Moreover,
performance
model
driven
original
values
was
better
medium-
low-concentration
regions.
However,
high-concentration
region,
extreme
concentrations
data-driven
LSTM
models
underestimation,
considering
better.
sensitivity
experiments
accuracy
decreased
considerably
when
backward
time
step
increased.
In
study,
only
chlorophyll-a,
whose
forecasted,
effect
other
relevant
elements
on
not
considered,
current
weakness
