Earth system science data,
Journal Year:
2021,
Volume and Issue:
13(10), P. 4881 - 4896
Published: Oct. 26, 2021
Abstract.
Forest
age
can
determine
the
capacity
of
a
forest
to
uptake
carbon
from
atmosphere.
However,
lack
global
diagnostics
that
reflect
stage
and
associated
disturbance
regimes
hampers
quantification
age-related
differences
in
dynamics.
This
study
provides
new
distribution
circa
2010,
estimated
using
machine
learning
approach
trained
with
more
than
40
000
plots
inventory,
biomass
climate
data.
First,
an
evaluation
against
plot-level
measurements
reveals
data-driven
method
has
relatively
good
predictive
classifying
old-growth
vs.
non-old-growth
(precision
=
0.81
0.99
for
non-old-growth,
respectively)
forests
estimating
corresponding
estimates
(NSE
0.6
–
Nash–Sutcliffe
efficiency
RMSE
50
years
root-mean-square
error).
there
are
systematic
biases
overestimation
young-
underestimation
old-forest
stands,
respectively.
Globally,
we
find
large
variability
tropical
regions
Amazon
Congo,
young
China,
intermediate
stands
Europe.
Furthermore,
high
rates
deforestation
or
degradation
(e.g.
arc
Amazon)
composed
mainly
younger
stands.
Assessment
space
shows
old
either
cold
dry
warm
wet
regions,
while
young–intermediate
span
climatic
gradient.
Finally,
comparing
presented
series
regional
products
rooted
different
approaches
situ
observations
global-scale
products.
Despite
showing
robustness
cross-validation
results,
additional
methodological
insights
on
further
developments
should
as
much
possible
harmonize
data
across
approaches.
The
dataset
here
into
better
understand
dynamics
water
cycles.
datasets
openly
available
at
https://doi.org/10.17871/ForestAgeBGI.2021
(Besnard
et
al.,
2021).
Forests,
Journal Year:
2018,
Volume and Issue:
9(6), P. 305 - 305
Published: June 1, 2018
Amazonia
is
home
to
more
than
half
of
the
world’s
remaining
tropical
forests,
playing
a
key
role
as
reservoirs
carbon
and
biodiversity.
However,
whether
at
slower
or
faster
pace,
continued
deforestation
causes
forest
fragmentation
in
this
region.
Thus,
understanding
relationship
between
fire
incidence
intensity
region
critical.
Here,
we
use
MODIS
Active
Fire
Product
(MCD14ML,
Collection
6)
proxy
(measured
Radiative
Power—FRP),
Brazilian
official
Land-use
Land-cover
Map
understand
among
deforestation,
fragmentation,
on
frontier
Amazonia.
Our
results
showed
that
vary
with
levels
habitat
loss
fragmentation.
About
95%
active
fires
most
intense
ones
(FRP
>
500
megawatts)
were
found
first
kilometre
from
edges
areas.
Changes
made
2012
main
law
regulating
conservation
forests
within
private
properties
reduced
obligation
recover
illegally
deforested
areas,
thus
allowing
for
maintenance
fragmented
areas
reinforce
need
guarantee
low
order
avoid
degradation
its
by
related
emissions.
Emerging Topics in Life Sciences,
Journal Year:
2019,
Volume and Issue:
3(2), P. 207 - 219
Published: April 24, 2019
Abstract
Biodiversity
continues
to
decline
under
the
effect
of
multiple
human
pressures.
We
give
a
brief
overview
main
pressures
on
biodiversity,
before
focusing
two
that
have
predominant
effect:
land-use
and
climate
change.
discuss
how
interactions
between
change
in
terrestrial
systems
are
likely
greater
impacts
than
expected
when
only
considering
these
isolation.
Understanding
biodiversity
changes
is
complicated
by
fact
such
be
uneven
among
different
geographic
regions
species.
review
evidence
for
variation
changes,
relating
differences
species
key
ecological
characteristics,
explaining
disproportionate
certain
leading
spatial
homogenisation
communities.
Finally,
we
explain
overall
losses
larger
upon
types
species,
lead
strong
negative
consequences
functioning
ecosystems,
consequently
well-being.
Earth system science data,
Journal Year:
2021,
Volume and Issue:
13(10), P. 4881 - 4896
Published: Oct. 26, 2021
Abstract.
Forest
age
can
determine
the
capacity
of
a
forest
to
uptake
carbon
from
atmosphere.
However,
lack
global
diagnostics
that
reflect
stage
and
associated
disturbance
regimes
hampers
quantification
age-related
differences
in
dynamics.
This
study
provides
new
distribution
circa
2010,
estimated
using
machine
learning
approach
trained
with
more
than
40
000
plots
inventory,
biomass
climate
data.
First,
an
evaluation
against
plot-level
measurements
reveals
data-driven
method
has
relatively
good
predictive
classifying
old-growth
vs.
non-old-growth
(precision
=
0.81
0.99
for
non-old-growth,
respectively)
forests
estimating
corresponding
estimates
(NSE
0.6
–
Nash–Sutcliffe
efficiency
RMSE
50
years
root-mean-square
error).
there
are
systematic
biases
overestimation
young-
underestimation
old-forest
stands,
respectively.
Globally,
we
find
large
variability
tropical
regions
Amazon
Congo,
young
China,
intermediate
stands
Europe.
Furthermore,
high
rates
deforestation
or
degradation
(e.g.
arc
Amazon)
composed
mainly
younger
stands.
Assessment
space
shows
old
either
cold
dry
warm
wet
regions,
while
young–intermediate
span
climatic
gradient.
Finally,
comparing
presented
series
regional
products
rooted
different
approaches
situ
observations
global-scale
products.
Despite
showing
robustness
cross-validation
results,
additional
methodological
insights
on
further
developments
should
as
much
possible
harmonize
data
across
approaches.
The
dataset
here
into
better
understand
dynamics
water
cycles.
datasets
openly
available
at
https://doi.org/10.17871/ForestAgeBGI.2021
(Besnard
et
al.,
2021).
