Plants,
Год журнала:
2024,
Номер
13(21), С. 3070 - 3070
Опубликована: Окт. 31, 2024
The
aboveground
biomass
(AGB)
of
summer
maize
is
an
important
indicator
for
assessing
crop
growth
status
and
predicting
yield,
playing
a
significant
role
in
agricultural
management
decision-making.
Traditional
on-site
measurements
AGB
are
limited,
due
to
low
efficiency
lack
spatial
information.
development
unmanned
aerial
vehicle
(UAV)
technology
agriculture
offers
rapid
cost-effective
method
obtaining
information,
but
currently,
the
prediction
accuracy
based
on
UAVs
limited.
This
study
focuses
entire
period
maize.
Multispectral
images
six
key
stages
were
captured
using
DJI
Phantom
4
Pro,
color
indices
elevation
data
(DEM)
extracted
from
these
stage
images.
Combining
measured
such
as
plant
height,
which
collected
ground,
three
machine
learning
algorithms
partial
least
squares
regression
(PLSR),
random
forest
(RF),
long
short-term
memory
(LSTM),
input
feature
analysis
PH
was
carried
out,
model
constructed.
results
show
that:
(1)
spectral
(CIS)
alone
predict
has
relatively
poor
accuracy.
Among
models,
LSTM
best
simulation
effect,
with
coefficient
determination
(R
Ecological Informatics,
Год журнала:
2024,
Номер
80, С. 102485 - 102485
Опубликована: Янв. 17, 2024
Understanding
the
relationship
between
land-use
patterns
and
regional
carbon
storage,
as
well
predicting
future
changes
for
sink
emission
management,
are
of
immense
significance.
This
study
utilized
data
from
1990,
2000,
2010,
2020,
InVEST
model,
to
evaluate
spatiotemporal
evolution
storage
in
Sanjiangyuan
area
over
past
three
decades.
Furthermore,
predictions
2035
were
presented
using
PLUS
model.
The
findings
revealed
following
key
results:
(1)
land
types
mainly
low
cover
grassland,
medium
grassland
unused
land,
among
which
decreased
significantly
1990
wetland
increased,
is
main
reason
increase
storage.
(2)
Climatic-environmental
social-economic
factors
jointly
influenced
change
area.
Except
expansion
other
was
by
climatic
environmental
factors.
(3)
During
1990–2020,
source
region
showed
an
overall
upward
trend,
with
a
total
39.97
×
107
t,
had
positive
potential
impact
on
whole.
(4)
Under
natural
scenario,
both
density
increased
simulation
2035,
positive.
On
this
basis,
paper
puts
some
suggestions
forward
improve
capacity
future.
provides
valuable
scientific
insights
management
decision-making
promotes
sustainable
development
functions
region.
Ecological Informatics,
Год журнала:
2024,
Номер
82, С. 102712 - 102712
Опубликована: Июнь 30, 2024
Quantifying
above
ground
biomass
(AGB)
and
its
spatial
distribution
can
significantly
contribute
to
monitor
carbon
stocks
as
well
the
storage
dynamics
in
forests.
For
effective
forest
monitoring
management
case
of
complex
tropical
Indian
forests,
there
is
a
need
obtain
reliable
estimates
amount
sequestration
at
regional
national
levels,
but
estimation
quite
challenging.
The
main
objective
study
validate
usefulness
gridded
density
(AGBD)
(ton/ha)
spaceborne
LiDAR
Global
Ecosystem
Dynamics
Investigation
data
(GEDI
L4B,
Version
2)
across
two
heterogeneous
forests
India,
Betul
Mudumalai
Methodology
includes,
for
each
area,
linear
regression
model
which
predicts
AGB
from
Sentinel-2
MSI
was
developed
using
reference
comparing
it
with
GEDI
AGBD
values.
Central
India
had
RMSE
13.9
ton/ha,
relative
=
8.7%
R2
0.88,
bias
−0.28
comparison
between
modelled
1
km
resolution
show
relatively
strong
correlation
(0.66)
no
or
little
bias.
It
also
found
that
footprint
value
underestimated
compared
according
model.
southern
an
29.1
10.8%,
0.79
−0.022.
0.84,
field
values
lies
42.2
ton/ha
238.8
75.9
353.6
ton/ha.
results
indicates
underestimates
AGB,
used
produce
product
needs
be
adjusted
provide
information
on
balance
changes
over
time
type
exists
test
areas.
Agriculture,
Год журнала:
2024,
Номер
14(3), С. 378 - 378
Опубликована: Фев. 27, 2024
Aboveground
biomass
(AGB)
is
an
important
indicator
for
characterizing
crop
growth
conditions.
A
rapid
and
accurate
estimation
of
AGB
critical
guiding
the
management
farmland
achieving
production
potential,
it
can
also
provide
vital
data
ensuring
food
security.
In
this
study,
by
applying
different
water
nitrogen
treatments,
unmanned
aerial
vehicle
(UAV)
equipped
with
a
multispectral
imaging
spectrometer
was
used
to
acquire
images
winter
wheat
during
stages.
Then,
plant
height
(Hdsm)
extracted
from
digital
surface
model
(DSM)
information
establish
improve
AGB,
using
backpropagation
(BP)
neural
network,
machine
learning
method.
The
results
show
that
(1)
R2,
root-mean-square
error
(RMSE),
relative
predictive
deviation
(RPD)
model,
constructed
directly
Hdsm,
are
0.58,
4528.23
kg/hm2,
1.25,
respectively.
estimated
mean
(16,198.27
kg/hm2)
slightly
smaller
than
measured
(16,960.23
kg/hm2).
(2)
RMSE,
RPD
improved
based
on
AGB/Hdsm,
0.88,
2291.90
2.75,
respectively,
(17,478.21
more
similar
(17,222.59
boosts
accuracy
51.72%
compared
Hdsm.
Moreover,
shows
strong
transferability
in
regard
treatments
year
scenarios,
but
there
differences
N-level
scenarios.
(3)
Differences
characteristics
key
factors
lead
model.
This
study
provides
antecedent
construction
wheat.
We
confirm
that,
when
datasets
have
histogram
characteristics,
applicable
new
Ecological Informatics,
Год журнала:
2024,
Номер
83, С. 102796 - 102796
Опубликована: Авг. 25, 2024
It
is
crucial
to
develop
a
comprehensive
method
for
estimating
the
aboveground
biomass
(AGB)
of
trees,
shrubs,
grasslands,
and
sparse
tree
areas
in
ecologically
fragile
dry,
hot
valley
regions
with
vertical
zonation.
Multi-source
remote-sensing
data
can
fulfill
this
requirement,
providing
help
monitoring
health
ecosystems
basis
regional
biodiversity
conservation
restoration.
Sentinel-2A
satellite
imagery
was
used
classify
forests,
grasslands
Yuanmou
County,
Chuxiong
Yi
Autonomous
Prefecture,
Yunnan
Province,
China.
The
Gaofen-2
(GF-2)
extract
canopy
width
calculate
valley-type
savanna
region.
These
were
combined
factors
measured
survey
data,
random
forest
(RF)
extreme
gradient
boosting
(XGBoost)
models
estimate
biomass.
Using
GF-2
images
segment
effectively
reduced
overestimation
low-resolution
images,
enabling
AGB
trees
be
accurately
estimated.
estimations
based
on
attained
coefficient
determination
(R2)
values
0.45
0.47
forest,
0.55
0.61
0.32
0.37
using
RF
XGBoost
models,
respectively,
demonstrating
variable
effectiveness
across
vegetation
types.
In
addition,
model
more
robust
than
all
three
Our
methodology
provides
scientific
support
sustainable
development
valleys
areas.
