Abstract.
This
study
provides
the
first
inter-comparison
of
different
state-of-the-art
approaches
and
frameworks
that
share
a
commonality
in
their
utilization
satellite
remote
sensing
data
to
quantify
irrigation
at
regional
scale.
The
compared
vary
reliance
on
either
soil
moisture
or
evapotranspiration
data,
joint
both.
two
combine
rainfed
hydrological
models
baseline
framework
use
water
balance
modeling
moisture-based
inversion
framework.
is
conducted
over
lower
Ebro
catchment
Spain
where
observed
amounts
are
available
for
benchmarking.
Our
results
showed
within
framework,
approach
using
both
ET
only
differed
by
+17
mm
from
benchmark
(922
mm)
during
main
season
years,
+41
-228
relying
solely
ET,
respectively.
A
comparison
advantage
more
complex
was
consistency
between
components
model,
which
made
it
unlikely
one
ended
up
representing
all
use.
However,
simplicity
coupled
with
its
direct
conversion
changes
into
actual
volumes,
effectively
addresses
key
challenges
inherent
associated
uncertainties
related
an
unknown
observation
depth
static
layers
conceptual
model.
performance
came
closest
able
account
precipitation
input,
resulted
plausible
temporal
distributions
than
what
expected
observations.
Agricultural and Forest Meteorology,
Год журнала:
2024,
Номер
346, С. 109882 - 109882
Опубликована: Янв. 2, 2024
Recently,
data
assimilation
(DA)
has
garnered
significant
attention.
Integration
of
DA
approaches
and
crop
models
could
diminish
model
uncertainties
improve
the
precision
simulations.
While
previous
research
extensively
focused
on
assimilating
leaf
area
index
(LAI)
or
soil
moisture
(SM),
feasibility
effectiveness
evapotranspiration
(ET)
have
been
rarely
explored.
In
this
study,
we
proposed
a
novel
framework
ET
assimilation.
Then,
together
with
commonly
assimilated
LAI
SM,
evaluated
performance
new
method
in
simulating
key
indicators
(i.e.,
daily
interannual
scales,
yield)
based
long-term
eddy
covariance
observations
well-calibrated
model.
strategies
utilized
to
evaluate
consist
two
Ensemble
Kalman
filter
(EnKF)
EnKF
simultaneous
state-parameter
estimation
(EnKF-SSPE))
combinations
three
LAI,
ET).
Our
results
demonstrate
that
joint
EnKF-SSPE
performs
best
for
wheat
while
SM
is
maize.
For
single
observation,
play
dominant
role
maize,
respectively.
This
because
variability
growth
primarily
influenced
by
agricultural
management
(e.g.,
cultivar
change)
can
be
represented
LAI.
maize
which
mostly
rainfed,
water
stress
usually
occurs.
Therefore,
ET,
its
ability
reflect
status,
proves
effective.
outperforms
EnKF,
exhibiting
potential
revealing
parameter
evolution
during
modeling,
especially
when
cultivars
are
regularly
renewed.
study
evaluates
different
methods
through
newly
sequential
framework,
might
illuminating
future
applications
DA.
Hydrology and earth system sciences,
Год журнала:
2024,
Номер
28(3), С. 441 - 457
Опубликована: Фев. 6, 2024
Abstract.
This
study
provides
the
first
inter-comparison
of
different
state-of-the-art
approaches
and
frameworks
that
share
a
commonality
in
their
utilization
satellite
remote-sensing
data
to
quantify
irrigation
at
regional
scale.
The
compared
vary
reliance
on
either
soil
moisture
or
evapotranspiration
joint
both.
two
extract
information
from
residuals
between
observations
rainfed
hydrological
models
baseline
framework
use
water
balance
modeling
soil-moisture-based
inversion
framework.
is
conducted
over
lower
Ebro
catchment
Spain
where
observed
amounts
are
available
for
benchmarking.
Our
results
showed
within
framework,
approach
using
both
(ET)
only
differed
by
+37
mm
benchmark
(922
mm)
during
main
season
2
years
+47
−208
relying
solely
ET,
respectively.
A
comparison
advantage
more
complex
was
consistency
ET
components
model,
which
made
it
unlikely
one
ended
up
representing
all
use.
However,
simplicity
coupled
with
its
direct
conversion
changes
into
actual
volumes,
effectively
addresses
key
challenges
inherent
associated
uncertainties
related
an
unknown
observation
depth
static
layers
conceptual
model.
performance
came
closest
able
account
precipitation
input,
resulted
plausible
temporal
distributions
than
what
expected
observations.
Agricultural Water Management,
Год журнала:
2024,
Номер
293, С. 108704 - 108704
Опубликована: Фев. 2, 2024
Irrigation
is
the
most
water
consuming
activity
in
world.
Knowing
timing
and
amount
of
irrigation
that
actually
applied
therefore
fundamental
for
managers.
However,
this
information
rarely
available
at
all
scales
subject
to
large
uncertainties
due
wide
variety
existing
agricultural
practices
associated
regimes
(full
irrigation,
deficit
or
over-irrigation).
To
fill
gap,
we
propose
a
two-step
approach
based
on
15
m
resolution
Sentinel-1
(S1)
surface
soil
moisture
(SSM)
data
retrieve
actual
weekly
scale
over
an
entire
district.
In
first
step,
S1-derived
SSM
assimilated
into
FAO-56-based
crop
balance
model
(SAMIR)
each
type
both
(Idose)
threshold
(SMthreshold)
which
triggered.
do
this,
particle
filter
method
implemented,
with
particles
reset
month
provide
time-varying
SMthreshold
Idose.
second
retrieved
Idose
values
are
used
as
input
SAMIR
estimate
its
uncertainty.
The
assimilation
(SSM-ASSIM)
tested
8000
hectare
Algerri-Balaguer
district
located
northeastern
Spain,
where
situ
integrating
whole
during
2019.
For
evaluation,
performance
SSM-ASSIM
compared
default
FAO-56
module
(called
FAO56-DEF),
sets
critical
value
systematically
fills
reservoir
event.
2019,
observed
annual
687
mm,
(FAO56-DEF)
shows
root
mean
square
deviation
between
6.7
(8.8)
mm
week-1,
bias
+0.3
(−1.4)
Pearson
correlation
coefficient
0.88
(0.78).
great
potential
retrieving
use
extended
areas
any
regime,
including
over-irrigation.
Water Resources Research,
Год журнала:
2024,
Номер
60(10)
Опубликована: Окт. 1, 2024
Abstract
Vegetation‐related
processes,
such
as
evapotranspiration
(ET),
irrigation
water
withdrawal,
and
groundwater
recharge,
are
influencing
surface
(SW)—groundwater
(GW)
interaction
in
districts.
Meanwhile,
conventional
numerical
models
of
SW‐GW
not
developed
based
on
satellite‐based
observations
vegetation
indices.
In
this
paper,
we
propose
a
novel
methodology
for
multivariate
assimilation
Sentinel‐based
leaf
area
index
(LAI)
well
in‐situ
records
streamflow.
Moreover,
the
GW
model
is
initially
calibrated
table
observations.
These
assimilated
into
SWAT‐MODFLOW
to
accurately
analyze
advantage
considering
high‐resolution
LAI
data
modeling.
We
develop
(DA)
framework
using
particle
filter
sampling
importance
resampling
(PF‐SIR).
Parameters
MODFLOW
parameter
estimation
(PEST)
algorithm
observation
table.
The
implemented
over
Mahabad
Irrigation
Plain,
located
Urmia
Lake
Basin
Iran.
Some
DA
scenarios
closely
examined,
including
univariate
(L‐DA),
streamflow
(S‐DA),
streamflow‐LAI
(SL‐DA).
Results
show
that
SL‐DA
scenario
results
best
estimations
streamflow,
LAI,
level,
compared
other
scenarios.
does
improve
accuracy
estimation,
while
significant
improvements
simulation,
where,
open
loop
run,
(absolute)
bias
decreases
from
75%
6%.
S‐DA,
L‐DA,
underestimates
use
demand
potential
actual
crop
yield.
Water,
Год журнала:
2025,
Номер
17(5), С. 730 - 730
Опубликована: Март 2, 2025
This
study
provides
a
comprehensive
assessment
of
the
HYDRUS-1D
model
for
predicting
root-zone
soil
moisture
(RZSM)
and
evapotranspiration
(ET).
It
evaluates
different
hydrodynamic
parameter
(SHP)
schemes—soil
type-based,
texture-based,
inverse
solution—under
varying
cropping
systems
(Zea
mays–Glycine
max
rotation
continuous
Zea
mays)
conditions
(irrigated
rainfed),
aiming
to
understand
water
transport
across
cultivation
patterns.
Using
field
measurements
from
2002,
SHPs
were
optimized
each
scheme
applied
predict
RZSM
ET
2003
2007.
The
solution
produced
nearly
unbiased
predictions
with
root
mean
square
error
(RMSE)
0.011
m3m⁻3,
compared
RMSEs
0.036
m3m⁻3
0.042
type-based
texture-based
schemes,
respectively.
For
predictions,
comparable
accuracy
was
achieved,
66.4
Wm⁻2,
69.5
68.2
Wm⁻2
three
schemes.
prediction
declined
over
time
in
mays
all
while
systematic
errors
predominated
field.
trends
mirrored
irrigated
but
diverged
rainfed
croplands
due
decoupling
under
arid
conditions.
Agricultural Water Management,
Год журнала:
2024,
Номер
303, С. 109036 - 109036
Опубликована: Сен. 2, 2024
Irrigated
agriculture
is
the
dominant
user
of
water
globally,
but
most
withdrawals
are
not
monitored
or
reported.
As
a
result,
it
largely
unknown
when,
where,
and
how
much
used
for
irrigation.
Here,
we
evaluated
ability
remotely
sensed
evapotranspiration
(ET)
data,
integrated
with
other
datasets,
to
calculate
irrigation
applications
in
an
intensively
irrigated
portion
United
States.
We
compared
calculations
based
on
ensemble
satellite-driven
ET
models
from
OpenET
reported
groundwater
hundreds
farmer
application
records
statewide
flowmeter
database
at
three
spatial
scales
(field,
right
group,
management
area).
At
field
scale,
found
that
ET-based
agreed
best
when
mean
was
aggregated
growing
season
timescale
(bias
=
1.6–4.9
%,
R2
0.53–0.74),
agreement
between
calculated
better
multi-year
averages
than
individual
years.
group
linking
pumping
wells
specific
fields
primary
source
uncertainty.
area
exhibited
similar
temporal
patterns
as
data
tended
be
positively
biased
more
interannual
variability.
Disagreement
strongly
correlated
annual
precipitation,
closely
after
statistically
adjusting
precipitation.
The
selection
model
also
important
consideration,
variability
across
larger
potential
impacts
conservation
measures
employed
region.
From
these
results,
suggest
key
practices
working
include
accurately
accounting
changes
soil
moisture,
deep
percolation,
runoff;
careful
verification
well-field
linkages;
conducting
application-specific
evaluations
