Abstract.
Seagrass
meadows
are
a
highly
productive
and
economically
important
shallow
coastal
habitat.
Their
sensitivity
to
natural
anthropogenic
disturbances,
combined
with
their
importance
for
local
biodiversity,
carbon
stocks
sediment
dynamics,
motivate
frequent
monitoring
of
distribution.
However,
generating
time-series
seagrass
cover
from
field
observations
is
costly,
mapping
methods
based
on
remote
sensing
require
restrictive
conditions
seabed
visibility,
limiting
the
frequency
observations.
In
this
contribution,
we
examine
effect
accounting
environmental
factors
such
as
bathymetry
median
grain
size
(D50)
substrate,
well
coordinates
known
patches,
performance
Random
Forest
(RF)
classifier
used
determine
cover.
Using
148
Landsat
images
Venice
Lagoon
(Italy)
between
1999
2020,
trained
RF
only
spectral
features
surveys,
respectively
2002
2017.
Then,
by
adding
above
applying
time-based
correction
predictions,
created
multiple
models
different
feature
combinations.
We
tested
quality
resulting
predictions
each
model
against
showing
that
bathymetry,
D50
patches
exert
an
influence
dependant
training
image
survey
chosen.
2017,
where
using
causes
overestimate
surface
area,
no
significant
change
in
was
observed.
Conversely,
2002,
addition
out-of-image
particularly
vegetated
greatly
improves
predictive
capacity
model,
while
still
allowing
detection
beds
absent
reference
survey.
Applying
eliminates
small
temporal
variations
improving
performed
before
correction.
conclude
together
correction,
has
most
potential
produce
reliable
While
case
study
alone
insufficient
explain
how
geographic
location
information
influences
classification
process,
suggest
it
linked
inherent
spatial
auto-correlation
meadow
interest
develop
our
map
vegetation
across
time,
identify
phenomenon
warranting
further
research.
Biology Open,
Journal Year:
2022,
Volume and Issue:
11(8)
Published: July 25, 2022
Plants
endure
environmental
stressors
via
adaptation
and
phenotypic
plasticity.
Studying
these
mechanisms
in
seagrasses
is
extremely
relevant
as
they
are
important
primary
producers
functionally
significant
carbon
sinks.
These
not
well
understood
at
the
tissue
level
seagrasses.
Using
RNA-seq,
we
generated
transcriptome
sequences
from
of
leaf,
basal
leaf
meristem
root
organs
Posidonia
australis,
establishing
baseline
situ
transcriptomic
profiles
for
tissues
across
a
salinity
gradient.
Samples
were
collected
four
P.
australis
meadows
growing
Shark
Bay,
Western
Australia.
Analysis
gene
expression
showed
differences
between
types,
with
more
variation
among
leaves
than
or
roots.
Gene
ontology
enrichment
analysis
largely
due
to
role
photosynthesis,
plant
growth
nutrient
absorption
organs,
respectively.
Differential
upregulation
regulation
processes
higher
meadows.
Our
study
highlights
importance
considering
when
evaluating
whole-plant
responses
change.
This
article
has
an
associated
First
Person
interview
first
author
paper.
Ecological Modelling,
Journal Year:
2024,
Volume and Issue:
495, P. 110802 - 110802
Published: July 19, 2024
In
a
context
of
worldwide
decline
and
given
the
critical
ecological
role
marine
seagrasses
to
coastal
ecosystem
structure
functioning,
regional
conservation
initiatives
have
emerged
over
past
thirty
years
protect
these
important
habitat-forming
species.Yet,
effective
interventions
need
account
for
site-specific
processes
stressors.Thus,
our
ability
accurately
predict
seagrass
dynamics
is
pivotal
support
management
interventions.To
date,
determinist
process-based
modelling
has
provided
insights
on
drivers
dynamics.Here,
we
developed
an
original
model
framework
that
combines
hydrodynamics
ocean
with
local
data-driven
models
rely
Boosted
Regression
Trees
seasonal
patch-level
plant-level
features
as
function
environmental
conditions.Based
only
12-month
monitoring
across
nine
sites,
traits
successfully
reproduce
overall
based
mostly
inferred
relationships
monthly
light
temperature,
lesser
extent,
exposure
physical
stressors
(i.e.,
currents
waves).While
fail
finely
capture
spatial
discrepancies
all
sites
(especially
where
demonstrates
higher
growth
potential),
spatially-explicit
simulations
highlight
how
seagrass-hydrodynamics
feedback
whole
bay
can
dampen
potential
due
shear
stress.However,
this
offers
simulate
long-term
changes
in
extent
status
meadows
Arcachon
Bay,
explicit
resolving
hydro-sediment
effects
appears
priority
better
range
between
conditions.
Biogeosciences,
Journal Year:
2023,
Volume and Issue:
20(22), P. 4551 - 4576
Published: Nov. 20, 2023
Abstract.
Seagrass
meadows
are
a
highly
productive
and
economically
important
shallow
coastal
habitat.
Their
sensitivity
to
natural
anthropogenic
disturbances,
combined
with
their
importance
for
local
biodiversity,
carbon
stocks,
sediment
dynamics,
motivate
frequent
monitoring
of
distribution.
However,
generating
time
series
seagrass
cover
from
field
observations
is
costly,
mapping
methods
based
on
remote
sensing
require
restrictive
conditions
seabed
visibility,
limiting
the
frequency
observations.
In
this
contribution,
we
examine
effect
accounting
environmental
factors,
such
as
bathymetry
median
grain
size
(D50)
substrate
well
coordinates
known
patches,
performance
random
forest
(RF)
classifier
used
determine
cover.
Using
148
Landsat
images
Venice
Lagoon
(Italy)
between
1999
2020,
trained
an
RF
only
spectral
features
surveys
2002
2017.
Then,
by
adding
above
applying
time-based
correction
predictions,
created
multiple
models
different
feature
combinations.
We
tested
quality
resulting
predictions
each
model
against
surveys,
showing
that
bathymetry,
D50,
patches
exert
influence
dependent
training
image
survey
chosen.
2017,
where
using
causes
overestimate
surface
area,
no
significant
change
in
was
observed.
Conversely,
2002,
addition
out-of-image
particularly
vegetated
greatly
improves
predictive
capacity
model,
while
still
allowing
detection
beds
absent
reference
survey.
Applying
eliminates
small
temporal
variations
improving
performed
before
correction.
conclude
together
correction,
has
most
potential
produce
reliable
While
case
study
alone
insufficient
explain
how
geographic
location
information
influences
classification
process,
suggest
it
linked
inherent
spatial
auto-correlation
meadow
interest
remote-sensing
develop
our
map
vegetation
across
time,
identify
phenomenon
warranting
further
research.
Botanica Marina,
Journal Year:
2022,
Volume and Issue:
65(5), P. 325 - 335
Published: Sept. 16, 2022
Abstract
During
midday
low
tides,
tropical
intertidal
seagrasses
are
challenged
by
high
irradiance
and
temperature.
This
study
assessed
photosynthetic
oxidative
stress
responses
of
Thalassia
hemprichii
Halophila
ovalis
exposed
to
150
1000
μmol
photons
m
−2
s
−1
30
40
°C
for
3
h.
High
temperature
(40
°C)
significantly
decreased
the
maximum
quantum
yield
both
this
heat-induced
photoinhibition
was
exacerbated
(1000
).
also
aggravated
effects
on
effective
T.
.
Non-photochemical
quenching
(NPQ)
induced
stressors
with
no
additive
effects.
In
contrast,
NPQ
H.
under
at
but
inhibited
°C.
Nevertheless,
antioxidant
enzyme
activity
reactive
oxygen
species
content
did
not
differ
among
treatments
in
either
seagrass.
Monitoring
chloroplast
distribution
revealed
a
partial
inhibitory
effect
avoidance
movement
irradiance.
Our
results
suggest
that
warming
events
may
cause
detrimental
impacts
shallow
water
seagrasses.
be
more
vulnerable
than
as
its
photoprotection,
i.e.
movement,
hindered