Environmental Research Letters,
Год журнала:
2024,
Номер
19(7), С. 073003 - 073003
Опубликована: Июнь 3, 2024
Abstract
Flood
risk
in
urban
areas
will
increase
massively
under
future
urbanization
and
climate
change.
Urban
flood
models
have
been
increasingly
applied
to
assess
impacts
of
on
risk.
For
this
purpose,
different
methodological
approaches
developed
order
reflect
the
complexity
dynamics
growth.
To
state-of-the
art
application
scenarios,
we
conducted
a
structured
literature
review
systematically
analyzed
93
publications
with
141
case
studies.
Our
shows
that
hydrological
hydrodynamic
are
most
commonly
used
simulate
Future
is
mostly
considered
as
sprawl
through
adjustment
land
use
maps
roughness
parameters.
A
low
number
additionally
consider
transitions
structures
densification
processes
their
scenarios.
High-resolution
physically
based
advanced
well
suited
for
describing
quantifiable
data-rich
contexts.
In
regions
limited
data,
argue
reducing
level
detail
increasing
patterns
should
be
improve
quality
projections
urbanization.
also
call
development
integrative
model
such
causal
network
greater
explanatory
power
enable
processing
qualitative
data.
Water Resources Research,
Год журнала:
2023,
Номер
59(3)
Опубликована: Фев. 21, 2023
Abstract
Estimating
uncertainty
in
flood
model
predictions
is
important
for
many
applications,
including
risk
assessment
and
forecasting.
We
focus
on
physics‐based
urban
flooding
models.
consider
the
effects
of
model's
complexity
key
input
parameters.
The
effect
rainfall
intensity
water
depth
also
studied.
As
a
test
study,
we
choose
Interconnected
Channel
Pond
Routing
(ICPR)
part
city
Minneapolis.
ICPR
floodwater
quantified
terms
ensemble
variance
using
multilevel
Monte
Carlo
(MC)
simulation
method.
Our
results
show
that
uncertainties
studied
domain
are
highly
localized.
Model
simplifications,
such
as
disregarding
groundwater
flow,
lead
to
overly
confident
predictions,
is,
both
less
accurate
uncertain
than
those
more
complex
model.
find
same
number
parameters,
increasing
resolution
reduces
(and
increases
MC
method's
computational
cost).
employ
method
reduce
cost
estimating
high‐resolution
Finally,
use
estimates
mean
covariance
real‐time
forecasting
physics‐informed
Gaussian
process
regression
even
with
few
measurements,
proposed
framework
forecast
provided
by
prediction
Environmental Research Letters,
Год журнала:
2024,
Номер
19(7), С. 073003 - 073003
Опубликована: Июнь 3, 2024
Abstract
Flood
risk
in
urban
areas
will
increase
massively
under
future
urbanization
and
climate
change.
Urban
flood
models
have
been
increasingly
applied
to
assess
impacts
of
on
risk.
For
this
purpose,
different
methodological
approaches
developed
order
reflect
the
complexity
dynamics
growth.
To
state-of-the
art
application
scenarios,
we
conducted
a
structured
literature
review
systematically
analyzed
93
publications
with
141
case
studies.
Our
shows
that
hydrological
hydrodynamic
are
most
commonly
used
simulate
Future
is
mostly
considered
as
sprawl
through
adjustment
land
use
maps
roughness
parameters.
A
low
number
additionally
consider
transitions
structures
densification
processes
their
scenarios.
High-resolution
physically
based
advanced
well
suited
for
describing
quantifiable
data-rich
contexts.
In
regions
limited
data,
argue
reducing
level
detail
increasing
patterns
should
be
improve
quality
projections
urbanization.
also
call
development
integrative
model
such
causal
network
greater
explanatory
power
enable
processing
qualitative
data.
