ArticlePDF Available

Abstract and Figures

The research presented here examined at which spatial and temporal resolution urban metabolism should be analysed to generate results that are useful for implementation of urban planning and design interventions aiming at optimization of resource flows. Moreover, it was researched whether a lack of data currently hampers analysing resource flows at this desired level of detail. To facilitate a stakeholder based research approach, the SIRUP tool – “Space-time Information analysis for Resource-conscious Urban Planning” – was developed. The tool was applied in a case study of Amsterdam, focused on the investigation of energy and water flows. Results show that most urban planning and design interventions envisioned in Amsterdam require information on a higher spatiotemporal resolution than the resolution of current urban metabolism analyses, i.e., more detailed than the city level and at time steps smaller than a year. Energy-related interventions generally require information on a higher resolution than water-related interventions. Moreover, for the majority of interventions information is needed on a higher resolution than currently available. For energy, the temporal resolution of existing data proved inadequate, for water, data with both a higher spatial and temporal resolution is required. Modelling and monitoring techniques are advancing for both water and energy and these advancements are likely to contribute to closing these data gaps in the future. These advancements can also prove useful in developing new sorts of urban metabolism analyses that can provide a systemic understanding of urban resource flows and that are tailored to urban planning and design.
Content may be subject to copyright.
Resources,
Conservation
and
Recycling
128
(2018)
516–525
Contents
lists
available
at
ScienceDirect
Resources,
Conservation
and
Recycling
jo
u
r
n
al
homep
age:
www.elsevier.com/locate/resconrec
Full
length
article
Space-time
information
analysis
for
resource-conscious
urban
planning
and
design:
A
stakeholder
based
identification
of
urban
metabolism
data
gaps
Ilse
M.
Voskampa,b,,
Marc
Spillera,
Sven
Stremkeb,c,
Arnold
K.
Bregtc,d,
Corné
Vreugdenhilc,d,
Huub
H.M.
Rijnaartsa
aSub-Department
of
Environmental
Technology,
Wageningen
University
&
Research,
P.O.
Box
17,
6700
AA
Wageningen,
The
Netherlands
bLandscape
Architecture
Group,
Wageningen
University
&
Research,
P.O.
Box
47,
6700
AA
Wageningen,
The
Netherlands
cAMS,
Amsterdam
Institute
for
Advanced
Metropolitan
Solutions,
Mauritskade
62,
1092
AD
Amsterdam,
The
Netherlands
dLaboratory
of
Geo-Information
Science
and
Remote
Sensing,
Wageningen
University
&
Research,
P.O.
Box
47,
6700
AA
Wageningen,
The
Netherlands
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
31
March
2016
Received
in
revised
form
2
August
2016
Accepted
25
August
2016
Available
online
12
September
2016
Keywords:
Urban
metabolism
Sustainable
resource
management
Urban
infrastructure
Urban
planning
Spatiotemporal
analysis
a
b
s
t
r
a
c
t
The
research
presented
here
examined
at
which
spatial
and
temporal
resolution
urban
metabolism
should
be
analysed
to
generate
results
that
are
useful
for
implementation
of
urban
planning
and
design
inter-
ventions
aiming
at
optimization
of
resource
flows.
Moreover,
it
was
researched
whether
a
lack
of
data
currently
hampers
analysing
resource
flows
at
this
desired
level
of
detail.
To
facilitate
a
stakeholder
based
research
approach,
the
SIRUP
tool
“Space-time
Information
analysis
for
Resource-conscious
Urban
Plan-
ning”
was
developed.
The
tool
was
applied
in
a
case
study
of
Amsterdam,
focused
on
the
investigation
of
energy
and
water
flows.
Results
show
that
most
urban
planning
and
design
interventions
envisioned
in
Amsterdam
require
information
on
a
higher
spatiotemporal
resolution
than
the
resolution
of
current
urban
metabolism
analyses,
i.e.,
more
detailed
than
the
city
level
and
at
time
steps
smaller
than
a
year.
Energy-related
interventions
generally
require
information
on
a
higher
resolution
than
water-related
interventions.
Moreover,
for
the
majority
of
interventions
information
is
needed
on
a
higher
resolution
than
currently
available.
For
energy,
the
temporal
resolution
of
existing
data
proved
inadequate,
for
water,
data
with
both
a
higher
spatial
and
temporal
resolution
is
required.
Modelling
and
monitoring
techniques
are
advancing
for
both
water
and
energy
and
these
advancements
are
likely
to
contribute
to
closing
these
data
gaps
in
the
future.
These
advancements
can
also
prove
useful
in
developing
new
sorts
of
urban
metabolism
analyses
that
can
provide
a
systemic
understanding
of
urban
resource
flows
and
that
are
tailored
to
urban
planning
and
design.
©
2016
The
Authors.
Published
by
Elsevier
B.V.
This
is
an
open
access
article
under
the
CC
BY
license
(http://creativecommons.org/licenses/by/4.0/).
1.
Introduction
The
notion
of
urban
metabolism
(UM)
has
inspired
new
ideas
about
how
cities
can
be
made
sustainable
and
it
has
fostered
quantitative
approaches
to
the
analysis
of
urban
resource
flows
(Agudelo-Vera
et
al.,
2012;
Castán
Broto
et
al.,
2012;
Zhang,
2013).
UM
refers
to
the
processes
whereby
cities
transform
raw
materials,
energy,
and
water
into
the
built
environment,
human
biomass,
and
waste
(Decker
et
al.,
2000).
UM
can
be
traced
back
to
Marx
in
1883,
who
used
the
term
metabolism
to
describe
the
exchange
of
materi-
Corresponding
author
at:
Sub-Department
of
Environmental
Technology,
Wageningen
University
&
Research,
P.O.
Box
17,
6700
AA
Wageningen,
The
Netherlands.
E-mail
address:
ilse.voskamp@wur.nl
(I.M.
Voskamp).
als
and
energy
between
society
and
its
natural
environment
(Pincetl
et
al.,
2012;
Zhang,
2013).
In
1965
Wolman
re-launched
the
term
as
he
presented
the
city
as
an
ecosystem,
and
later
others
also
used
the
term
UM
in
representing
a
city
as
an
organism
(Barles,
2010
Castán
Broto
et
al.,
2012;
Pincetl
et
al.,
2012;
Zhang,
2013).
Since
Wolman’s
early
study
of
urban
metabolic
processes,
two
distinct
quantitative
UM
approaches
have
developed
that
aim
to
describe
and
anal-
yse
the
material
and
energy
flows
within
cities.
One
describes
the
UM
in
terms
of
solar
energy
equivalents
(‘emergy’).
Related
school
of
scholars
emphasizes
the
earth’s
dependence
on
the
sun
as
an
energy
source
and
the
qualitative
difference
of
mass
or
energy
flows.
The
second
and
most
widely
used
approach,
is
associated
with
the
fields
of
industrial
ecology
and
engineering
(Barles,
2010;
Castán
Broto
et
al.,
2012;
Pincetl
et
al.,
2012).
Related
research
largely
consist
of
empirical
studies
that
account
for
the
energy
and
http://dx.doi.org/10.1016/j.resconrec.2016.08.026
0921-3449/©
2016
The
Authors.
Published
by
Elsevier
B.V.
This
is
an
open
access
article
under
the
CC
BY
license
(http://creativecommons.org/licenses/by/4.0/).
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
517
material/mass
flows
of
a
city,
using
methods
such
as
material
flow
analysis
(MFA),
mass
balancing,
life
cycle
analysis
(LCA)
and
eco-
logical
footprint
analysis
(Castán
Broto
et
al.,
2012;
Kennedy
et
al.,
2011;
Pincetl
et
al.,
2012;
Zhang,
2013).
Multiple
scholars
have
argued
that
the
latter
type
of
UM
anal-
yses,
the
flow
quantifications
associated
with
the
mainstream
UM
approach,
are
useful
for
urban
planning
and
design
(Castán
Broto
et
al.,
2012;
Chrysoulakis
et
al.,
2013;
Kennedy
et
al.,
2011;
Moffatt
and
Kohler,
2008;
Pincetl
et
al.,
2012).
However,
these
authors
also
argue
that
major
efforts
are
still
needed
to
make
UM
analyses
useful
for
informing
urban
planning
and
design
aiming
at
optimiza-
tion
of
urban
resource
flows
(Kennedy
et
al.,
2011).
Indeed,
only
three
examples
of
application
of
UM
for
designing
more
sustain-
able
urban
infrastructures
are
referred
to
in
literature
(Codoban
and
Kennedy,
2008;
Oswald
and
Baccini,
2003;
Quinn,
2008),
of
which
just
one
is
a
peer-reviewed
article.1The
only
recent
scientific
contributions
on
this
topic
all
discuss
the
planning
support
system
developed
in
the
BRIDGE
project
(Blecic
et
al.,
2014;
Chrysoulakis
et
al.,
2013;
Mitraka
et
al.,
2014).
In
professional
literature,
some
other
recent
examples
can
be
found.
In
the
Netherlands,
research
on
the
resource
flows
of
Rotterdam
was
conducted
and
used
as
a
basis
for
urban
design
strategies,
in
the
context
of
the
International
Archi-
tectural
Biennale
2014
Urban
by
Nature
(Tillie
et
al.,
2014).
In
the
Circular
Buiksloterham
project
in
Amsterdam
an
‘Urban
Metabolism
Scan’
was
performed
and
used
as
foundation
for
a
vision
for
the
Buiksloterham
area,
including
site-specific
technical
interventions
and
a
design
concept
(Gladek
et
al.,
2015).
So,
although
the
theo-
retical
potential
of
UM
analysis
for
urban
planning
and
design
is
increasingly
addressed
in
the
scientific
literature,
scientific
reports
that
illustrate
how
this
potential
can
be
realised
with
practical
implementation
remain
limited
thus
far.
Possibly,
UM
analyses
are
still
of
limited
use
for
urban
planning
and
design
because
they
are
performed
on
a
scale
level
that
does
not
match
urban
planning
and
design
practice
(Moffatt
and
Kohler,
2008;
Pincetl
et
al.,
2012;
Spiller
and
Agudelo-Vera,
2011).
The
UM
is
usually
analysed
for
a
period
of
a
year
on
city
or
regional
scale
(Kennedy
et
al.,
2011;
Niza
et
al.,
2009);
analyses
on
a
more
detailed
level
are
said
to
be
hampered
by
lack
of
data
(Codoban
and
Kennedy,
2008;
Pincetl
et
al.,
2012;
Shahrokni
et
al.,
2015).
Such
large-scale
analyses,
however,
do
not
reveal
which
metabolic
processes
and
functions
are
operating
at
various
spatial
and
temporal
scales.
Yet,
planners
and
designers
need
such
information
to
decide
upon
the
appropriate
interventions
to
realize
a
resource-conscious
strategy.
In
other
words,
they
need
this
information
to
inform
their
plan-
ning
and
design
decision-making
regarding
interventions
aimed
at
urban
climate
adaptation,
climate
mitigation
and/or
resource
effi-
ciency.
To
be
useful
for
urban
planners
and
designers,
UM
analyses
should
thus
provide
detailed
and
spatial
and
temporal
explicit
data
on
the
scale
at
which
these
practitioners
work
(Chrysoulakis
et
al.,
2013;
Golubiewski,
2012;
Moffatt
and
Kohler,
2008;
Pincetl
et
al.,
2012;
Vandevyvere
and
Stremke,
2012).
Therefore,
the
study
presented
here
aims
to
answer
the
follow-
ing
questions:
(a)
“at
which
spatial
and
temporal
resolution
should
resource
flows
be
analysed
to
generate
results
that
are
useful
for
implementation
of
urban
planning
and
design
interventions?”
and
(b)
“is
UM
analysis
at
this
desired
level
of
detail
currently
hampered
by
a
lack
of
data?”.
To
answer
these
questions
the
“Space-time
Infor-
mation
analysis
for
Resource-conscious
Urban
Planning”
(SIRUP)
tool
was
developed
and
applied
in
a
case
study
of
the
city
of
Amsterdam,
the
Netherlands.
The
SIRUP
tool
enables
an
analysis
on
two
levels:
I)
assessing
on
which
level
of
detail
in
space
and
time
stakeholders
1Although
Codoban
and
Kennedy
(2008)
were
the
first
to
refer
to
Oswald
and
Baccini
(2003)
in
this
light,
Kennedy
et
al.
(2011)
were
the
first
to
mention
all
three
examples.
need
information
on
resource
flows
to
inform
urban
planning
and
design
decision-making
aimed
at
developing
resource-conscious
strategies,
and
II)
evaluating
whether
existing
data
can
provide
the
information
needed
or
that
there
is
a
data
gap.
The
qualitative
tool
facilitates
information
and
knowledge
sharing
and
discussion
between
stakeholders.
Stakeholder
involvement
in
UM
research
is
essential
to
leverage
availability
of
and
access
to
urban
resource
data
and
it
allows
identifying
the
information
needs
of
urban
plan-
ning
and
design
practitioners
(Voskamp
et
al.,
2016;
Zhang
et
al.,
2015).
2.
Methods
and
materials2
2.1.
Development
of
the
SIRUP
tool
The
“Space-time
Information
analysis
for
Resource-conscious
Urban
Planning”
(SIRUP)
tool
is
based
on
the
work
of
Vervoort
et
al.
(2014),
who
developed
the
tool
Scale
Perspectives
to
elicit
societal
perspectives
and
generate
dialogue
on
governance
issues.
Their
tool
consists
of
a
frame
with
pre-defined
spatial
and
temporal
scales
in
which
stakeholders
can
outline
the
relevant
scales
for
a
partic-
ular
governance
issue.
For
the
SIRUP
tool,
this
frame
is
adapted
for
the
purpose
of
identifying
on
which
spatiotemporal
resolution
stakeholders
need
information
on
resource
flows
and
for
assessing
whether
existing
data
can
provide
this
information
on
the
reso-
lution
needed
(Supplementary
material,
Fig.
S1).
The
SIRUP
tool
is
applied
in
four
steps
(Fig.
1).
These
steps
aim
to
(I)
generate
an
inventory
of
UM
interventions,
(II)
determine
the
information
needed
for
implementing
each
of
these
interventions,
(III)
describe
the
spatiotemporal
resolution
of
existing
data
relevant
for
the
inter-
vention
and
(IV)
identify
whether
the
resolution
of
identified
data
can
satisfy
the
stakeholders’
intervention
information
needs.
2.2.
Application
of
the
SIRUP
tool
The
SIRUP
tool
was
applied
in
a
case
study
of
Amsterdam.
As
part
of
this
case
study,
stakeholders
were
involved
that
are
engaged
with
urban
planning
and
design
decision-making
aimed
at
devel-
oping
resource-conscious
strategies
for
the
city
of
Amsterdam.
The
stakeholders
comprised
researchers,
environmental
managers
from
utilities,
landscape
architects
and
urban
planning
&
design
practitioners.
Eleven
of
these
stakeholders
were
interviewed,
using
semi-structured
interviews,
and
thirteen
stakeholders
participated
in
a
workshop.
Step
I
and
II
of
the
SIRUP
tool
were
used
to
identify
on
which
spa-
tiotemporal
resolution
stakeholders
need
information
on
resource
flows.
In
step
I,
the
stakeholders
were
asked
to
describe
a
resource-
conscious
intervention
that
they
envision
to
be
implemented
in
Amsterdam.
Participants
were
also
asked
to
specify
in
the
SIRUP
frame
(Supplementary
material,
Fig.
S1)
at
which
spatial
scale
level
the
intervention
would
take
places
and
which
time
frame
they
envisioned
for
implementation.
In
step
II,
participants
were
asked
to
specify
the
information
needed
for
implementing
the
interven-
tion
mentioned
and
to
indicate
the
required
spatial
and
temporal
resolution
of
this
information
in
the
SIRUP
frame
(Fig.
1).
The
interviewer
or
workshop
facilitator
had
to
ensure
that
participants
described
the
information
on
resource
flows
that
is
necessary
to
enable
the
intervention.
Pen-and-paper
format
was
used
because
this
allows
for
greater
flexibility
than
a
digital
setting
(Vervoort
et
al.,
2014).
After
the
workshop,
all
contributions
were
digitalized
and
labelled
to
enable
the
selection
of
interventions
that
are
within
the
scope
of
the
research.
The
interventions
were
labelled
according
2A
more
elaborate
description
of
the
method
is
provided
as
Supplementary
mate-
rial
(S1.
Elaborate
description
of
methodology).
518
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
III. Inventory of existing data
The researcher makes an
inventory of existing data
that are relevant for the
intervention and place the
data in the SIRUP frame
according to their spatio-
temporal resolution.
I. Inventory of interventions
Stakeholders describe a resource
-conscious
intervention they envision implementing and they
specify
the envisioned
spatial
scale
level
and time-
frame of implementation.
IV. Data gap assessment
The researcher compares
the spatiotemporal resolution
of identified datasets with
the resolution of
the
stakeholders’
intervention
information needs, and
identifies gaps and overlaps.
II. Inventory of information needs
Stakeholders specify the
information needed for
implementing an intervention
,
mentioned in step I, and they
indicate the required
spatio-
temporal resolution of
this
.
Fig.
1.
The
four
steps
of
the
SIRUP
tool.
Note:
The
figure
includes
as
subfigures
a
schematic
representation
of
the
potential
outcome
of
the
different
steps.
The
green,
red
and
blue
circles/ellipses,
rectangles
and
arrows
represent
respectively
three
different
information
needs,
existing
data
sets
and
data
gaps
for
a
single
intervention
on
the
SIRUP
frame
(in
grey).
See
also
Supplementary
material
S1:
Elaborate
description
of
methodology.
to
the
type
of
intervention
and
the
resource
flow(s)
for
which
infor-
mation
is
needed.
We
limited
the
research
to
spatial
and
technical
interventions
aimed
at
urban
climate
adaptation,
climate
mitiga-
tion
and/or
resource
efficiency,
focussing
on
energy
and
water
flows
because
these
are
strongly
related
to
such
interventions
(Mitraka
et
al.,
2014;
Pincetl
et
al.,
2012)
(see
also
Supplementary
material
S1.
Elaborate
description
of
methodology).
In
step
III
a
desk
study
was
conducted
to
identify
which
data
exist
on
Amsterdam’s
energy
and
water
flows
and
to
compose
an
overview
of
these
data,
including
a
description
of
the
spatiotem-
poral
resolution
of
these
data.
In
the
study,
data
portals,
databases
and
reports
were
considered
that
contain
open
or
restricted
data
on
Amsterdam’s
energy
and
water
flows.
Expert
consultation
was
used
to
identify
relevant
datasets
and
to
obtain
access
to
restricted
datasets.
Metadata
was
described
for
all
datasets
obtained,
using
a
format
that
was
based
on
the
ISO
19115
and
the
INSPIRE
(Infrastruc-
ture
for
Spatial
Information
in
the
European
Community)
metadata
standards
to
ensure
compatibility
with
other
datasets
in
the
world.
The
mandatory
elements
of
the
metadata
standard
used
were,
amongst
others,
the
level
of
detail
of
the
data
in
the
spatial
dimen-
sion
and
in
the
time
dimension,
i.e.
the
spatial
and
temporal
resolution
of
the
data.
Moreover,
it
was
required
to
state
the
lim-
itations
and
rules
on
accessing,
use
and
publishing
of
the
existing
data
to
indicate
whether
the
data
is
open
or
restricted
(Supple-
mentary
material,
Table
S1).
Based
on
the
metadata-description
on
spatial
and
temporal
resolution,
the
datasets
were
placed
in
the
SIRUP
frame
(Fig.
1).
In
the
final
step,
step
IV,
the
data
inventory
was
used
to
analyse
whether
the
identified
data
can
satisfy
the
stakeholders’
infor-
mation
needs.
For
each
intervention
it
was
evaluated
whether
the
attributes
of
each
data
set
were
relevant
for
the
information
needed.
Then,
the
relevant
data
and
the
information
needs
for
the
intervention
were
combined
in
one
SIRUP
frame
and
arrows
were
drawn
from
the
dataset
to
the
information
needed
(Fig.
1).
When
the
spatiotemporal
resolution
of
information
needs
are
equal
to
the
resolution
of
existing
data,
these
exact
matches
were
indicated
by
a
circle
()
in
the
SIRUP
frame.
Subsequently,
the
size
and
dimension
of
the
arrows
were
analysed
to
assess
the
presence
and
severity
of
data
gaps.
An
arrow
either
indicates
a
two-dimensional
data
gap,
when
both
the
spatial
and
temporal
resolution
of
existing
data
are
insufficient,
or
it
indicates
a
one-dimensional
data
gap,
when
either
the
spatial
or
the
temporal
resolution
of
existing
data
is
lower
than
required.
A
two-dimensional
data
gap
is
indicated
by
arrows
point-
ing
towards
the
lower
left
corner
().
Arrows
pointing
downwards
or
to
the
lower
right
corner
(,)
indicate
a
one
dimensional
data
gap,
in
the
spatial
dimension
only.
The
reason
for
this
is
that
the
arrows
indicate
that
the
temporal
resolution
is
equal
to
or
higher
than
needed.
Because
aggregation
from
higher
temporal
resolu-
tion
to
lower
resolution
is
possible
without
an
information
loss,
the
required
temporal
resolution
can
be
derived
from
this
infor-
mation.
For
example,
hourly
totals
can
be
derived
from
data
on
minute
level
by
summing
all
available
minute
data
points.
Arrows
pointing
to
the
left
or
higher
left
corner
(←− ,)
represent
a
gap
in
the
temporal
dimension
only,
because
the
spatial
resolution
of
the
data
is
sufficient
or
higher
than
needed.
There
is
no
data
gap
when
existing
data
has
either
the
right
resolution
in
one
dimension
and
a
higher
resolution
in
the
other
or
a
higher
resolution
in
both
dimen-
sion,
implying
that
data
can
be
aggregated
to
get
to
the
required
resolution.
These
matches
are
indicated
by
arrows
pointing
up
(),
to
the
right
(),
or
diagonally
in
the
upper
right
direction
().
3.
Results
3.1.
Inventory
of
interventions
In
this
case
study
a
total
of
52
different
interventions
were
sug-
gested
by
the
stakeholders
during
the
interviews
and
workshop.
We
selected
fourteen
of
these
interventions
for
further
analysis,
namely
the
spatial
and
technical
interventions
for
which
infor-
mation
on
energy
and/or
water
flows
is
required.
The
selected
interventions
have
a
total
of
26
information
needs
that
relate
to
energy
and/or
water
flows.
These
information
needs
were
cate-
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
519
Table
1
The
14
interventions
and
26
related
information
needs
that
were
considered
in
this
research.
Information
needs
Piped
water
Non
piped
water
Energy
demand
Energy
supply
Interventions
and
interven-
tion
ID
(#)
Drinking
water
quantity
Waste
water
quantity
Waste
water
quantity
Fluvial
flooding
risk
Groundwater
quantity
Groundwater
quality
Intensity
and
repeat
time
of
extreme
rains
Surface
water
level
and
flow
rate
Household
cooling
demand
Electricity
demand
of
public
lighting
Total
urban
electric-
ity
demand
Cooling
supply
by
drinking
water
network
Electricity
supply
in
a
regional
smart
grid
Potential
electric-
ity
supply
by
PVpanels
1
Converting
cellulose
in
waste
into
power
I
2
Dike
rein-
forcement
I
3
More
concentrated
sewage
flows
I
I
4
Park
on
a
brownfield
site
II
I
5
Parking
garage
as
battery
I
6
Phytoremedi-
ation
of
green
areas
II
II
7
PVs
on
roofs
for
public
lighting
I
I
8
Rainwater
buffering
and
infiltration
I
9
Recovery
of
protein
from
sewage
I
I
10
Regional
smart
grid
I
II
11
Small-scale
parks
I
12
Usage
of
cold
from
drinking
water
for
cooling
I
I
I
13
Water-
robust
vital
infrastruc-
ture
II
14
Water
square
I
Note:
‘–‘
means
that
no
information
need
is
expressed
in
the
corresponding
category
for
the
indicated
intervention.
“I”
means
that
one
information
need
is
expressed
in
the
corresponding
category
for
the
indicated
intervention.
‘II’
means
that
two
information
needs
are
expressed
in
the
corresponding
category
for
the
indicated
intervention.
520
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
gorized
into
four
different
clusters
to
facilitate
interpretation:
I)
piped
water,
including
waste
water
and
drinking
water;
II)
non-
piped
water,
including
groundwater,
surface
water,
storm
water
and
rainwater;
III)
energy
demand;
and
IV)
energy
supply
(Table
1).
3.2.
Piped
water
Results
show
that
out
of
the
five
information
needs
related
to
piped
water,
two
can
be
met
by
existing
open
data
and
one
by
restricted
data
(Fig.
2).
Fig.
2a
shows
that
the
information
needs
that
were
expressed
for
piped
water
are
scattered
over
the
SIRUP
frame.
However,
no
information
needs
appear
in
the
lower
left
cor-
ner
of
the
frame,
up
to
12
h
and
district,
nor
at
the
highest
scale
levels,
that
is
metropolitan
region
and
five
years
and
higher.
The
SIRUP
frame
of
existing
data,
on
the
other
hand,
shows
a
different
pattern
(Fig.
2b).
In
terms
of
temporal
resolution,
these
data
fall
in
the
range
‘minutes’
to
‘one
year’.
In
terms
of
spatial
resolution,
open
data
ranges
from
the
scale
of
a
small
neighbourhood
to
the
metropolitan
region.
Additionally,
one
restricted
access
database
provides
drinking
water
quantity
data
on
building
level.
No
piped-
water
data
has
been
identified
that
has
both
high
temporal
and
spatial
resolution.
When
the
resolution
of
information
needs
and
existing
data
are
compared,
it
appears
that
two
information
needs
can
be
met:
I)
the
quantity
and
II)
the
quality
of
waste
water
that
enters
Amsterdam’s
waste
water
treatment
plants
at
seasonal
up
to
yearly
level
(indicate
as
,
Fig.
2c).
For
the
remainder
of
the
data
gaps,
it
shows
that
the
size
and
dimension
of
the
gaps
depend
on
which
existing
data
source
is
considered.
The
drinking
water
data
gaps,
for
instance,
are
either
two
dimensional
(),
using
dataset
3,
or
with
a
spatial
dimension
only
(,),
when
using
the
open
data
from
source
4
or
6.
Yet,
with
access
to
restricted
datasets,
there
is
a
data
gap
with
a
temporal
dimension
only
(←− ).
Nevertheless,
when
using
the
restricted
data,
the
size
of
the
gap
in
the
temporal
dimen-
sion
remains
the
same
as
when
dataset
3
is
used,
from
‘one
year’
to
‘month’.
On
the
other
hand,
restricted
data
can
close
the
data
gap
regarding
waste
water
quantity
at
municipal
and
minute
level
entirely.
Regarding
waste
water
quality,
there
is
a
two
dimensional
data
gap
when
using
open
data
()
and
a
data
gap
in
the
spatial
dimension
only
when
using
restricted
data
().
The
size
of
the
gap
in
the
spatial
dimension
remains
equal
when
restricted
data
can
be
used.
3.3.
Non-piped
water
In
the
case
of
non-piped
water,
the
spatiotemporal
resolution
of
existing
open
data
meets
the
resolution
of
six
out
of
the
twelve
information
needs
(Fig.
3).
Regarding
the
resolution
of
these
twelve
information
needs,
a
cluster
of
six
shows
on
the
right
side
of
the
field
at
the
temporal
scales
‘quarter
of
a
year’
till
‘five
years’
(Fig.
3a).
Of
the
remaining
six
information
needs,
four
appear
in
the
lower
left
corner
of
the
SIRUP
frame,
delineated
by
week
and
district.
Two
of
these
information
needs
are
rainwater
related,
the
other
two
relate
to
groundwater
quality
and
quantity.
For
these
four
information
needs
a
data
gap
exists,
because
existing
rainfall
and
groundwater
data
have
a
lower
resolution
than
required
(Fig.
3bc).
In
the
case
of
groundwater
quantity,
this
is
an
exception
because
the
resolu-
tion
of
the
existing
data
small
neighbourhood
to
small
district
at
seasonal
level
is
sufficient
for
the
other
three
groundwater
quantity
related
information
needs.
By
contrast,
the
resolution
of
groundwater
quality
data,
which
is
the
metropolitan
region
and
one
to
four
year
resolution,
is
insufficient
for
the
two
related
infor-
mation
needs.
Likewise,
the
resolution
of
rainfall
data
in
between
municipality
and
metropolitan
region
for
a
day
to
half
a
year
is
inadequate
for
all
rainwater
related
information
needs.
The
reso-
lution
of
existing
surface
water
data
is
sufficient
to
meet
the
three
related
information
needs
(,).
Overall,
six
data
gaps
for
non-
Fig.
2.
Results
for
piped
water.
a.
Information
needs;
b.
existing
data;
c.
data
gaps
(results
of
SIRUP
step
II,
III,
IV).
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
521
Fig.
3.
Results
for
non-piped
water.
a.
Information
needs;
b.
existing
data;
c.
data
gaps
(results
of
SIRUP
step
II,
III,
IV).
piped
water
remain,
including
two
data
gaps
in
both
dimensions
(),
one
data
gap
with
a
temporal
dimension
only
()
and
three
gaps
with
a
spatial
dimension
only
(,)
(Fig.
3c).
3.4.
Energy
demand
Out
of
the
five
information
needs
on
energy
demand
detected,
one
can
be
met
by
existing
restricted
data
(Fig.
4).
The
information
need
that
can
be
met,
electricity
demand
of
a
neighbourhood
at
yearly
basis
(),
is
the
only
one
that
is
not
part
of
the
cluster
in
the
lower
left
corner
of
the
SIRUP
frame,
delineated
by
week
and
district
(Fig.
4a).
Existing
data
sources
on
Amsterdam’s
energy
demand,
on
the
contrary,
primarily
provide
data
on
yearly
totals,
within
a
spa-
tial
range
of
building
to
country
level
(Fig.
4b).
The
exception
to
this
is
a
restricted
dataset
that
provides
data
on
the
electricity
demand
of
a
streetlight
per
day.
Accordingly,
for
all
four
information
needs
in
the
high-resolution
cluster
there
is
a
data
gap
(Fig.
4c).
Although
these
data
gaps
are
similar
because
they
have
a
temporal
dimen-
sion
only
(←− ,),
they
differ
in
the
size
of
the
gap.
The
data
gap
is
smallest
for
household
cooling
demand,
from
one
year
to
month
resolution,
whereas
the
data
gap
regarding
total
urban
electricity
demand
is
more
substantial,
from
one
year
to
week
up
to
from
one
year
to
one
hour.
3.5.
Energy
supply
Results
for
energy
supply
show
that
out
of
the
four
defined
infor-
mation
needs,
one
can
be
met
by
existing
open
data
(Fig.
5).
All
four
information
needs
appear
on
the
lower
half
of
the
SIRUP
field,
that
is
a
spatial
resolution
of
district
level
or
higher
(Fig.
5a).
In
terms
of
temporal
resolution
the
information
needs
cover
a
larger
range,
namely
from
‘seconds’
to
‘one
year’.
When
the
temporal
resolution
of
existing
energy
supply
data
is
considered,
it
appears
that
data
is
primarily
available
for
yearly
totals
(Fig.
5b).
The
exceptions
to
this
are
a
restricted
database
that
provides
electricity
supply
data
on
a
monthly
temporal
resolution
and
open
data
on
drinking
water
cooling
supply
with
a
half
yearly
resolution.
In
terms
of
the
spatial
resolution
of
existing
data,
findings
show
that
open
data
with
a
res-
olution
as
high
as
the
building
level
exists.
As
a
result,
energy
supply
related
data
gaps
have
a
temporal
dimension
only
(←− ,)
(Fig.
5c).
When
only
open
data
is
considered,
the
gap
for
household
cooling
supply
is
the
smallest,
from
one
year
to
month
resolution.
The
data
gaps
regarding
total
urban
electricity
demand
range
from
one
year
to
one
hour
or
minutes.
These
gaps
reduce
in
terms
of
the
number
of
temporal
scale
levels
to
be
bridged
when
there
is
access
to
the
restricted
database
of
the
electricity
supply
of
the
waste-to-power
plant.
The
information
need
that
can
be
met,
potential
yearly
elec-
tricity
supply
by
PV
panels
on
neighbourhood
level
(),
relates
to
the
same
intervention
for
which
the
energy
demand
related
infor-
mation
need
can
be
met,
namely
“PVs
on
roofs
for
public
lighting”.
4.
Discussion
4.1.
Patterns
in
information
needs
and
existing
data
To
inform
resource-conscious
urban
planning
and
design,
infor-
mation
on
water
and
energy
is
required
on
a
higher
spatiotemporal
resolution
than
the
resolution
of
current
UM
analyses.
For
12
out
of
14
interventions,
stakeholders
require
information
on
a
higher
level
of
detail
than
the
city/region
scale
and
the
annual
time
inter-
val
at
which
UM
analyses
are
currently
performed
(Kennedy
et
al.,
2011;
Niza
et
al.,
2009).
In
detail,
three
of
the
26
expressed
informa-
tion
needs
are
on
the
city-annual
resolution.
Ten
information
needs
have
either
only
a
temporal
resolution
that
is
higher
than
annual
or
only
a
spatial
resolution
that
is
higher
than
city
level,
including
six
information
needs
on
the
neighbourhood-annual
level.
Another
522
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
Fig.
4.
Results
for
energy
demand.
a.
Information
needs;
b.
existing
data;
c.
data
gaps
(results
of
SIRUP
step
II,
III,
IV).
Fig.
5.
Results
for
energy
supply.
a.
Information
needs;
b.
existing
data;
c.
data
gaps
(results
of
SIRUP
step
II,
III,
IV).
I.M.
Voskamp
et
al.
/
Resources,
Conservation
and
Recycling
128
(2018)
516–525
523
13
information
needs
are
on
a
high
resolution
in
both
the
temporal
and
spatial
dimension.
The
temporal
resolution
of
these
informa-
tion
needs
is
within
the
range
seconds
to
week
and
the
spatial
resolution
is
between
building
and
district
level.
The
required
spa-
tiotemporal
resolution
appears
to
be
linked
to
the
resource
flow
targeted
by
an
intervention.
The
resolution
of
water
related
infor-
mation
needs
is
scattered
across
the
SIRUP
frame,
including
a
large
range
of
both
low
and
high
spatiotemporal
levels,
whereas
energy
related
information
needs
are
on
a
high
spatiotemporal
resolution.
That
water
related
information
needs
cover
a
large
range
of
scale
levels
could
be
indicative
for
current
developments
towards
total
water
cycle
management,
also
known
as
sustainable
or
inte-
grated
urban
water
management
(Van
De
Meene
and
Brown,
2009).
Such
an
integrated
management
approach
requires
an
understand-
ing
of
the
dynamics
of
urban
water
flows
and
the
processes
that
affect
these
flows
at
multiple
scales
in
time
and
space
(Leusbrock
et
al.,
2015;
Pahl-Wostl,
2007).
These
different
scales
are
required
because
of
the
complexity
of
the
urban
water
cycle,
which
includes
sewerage,
drinking
water
and
drainage
as
well
as
surface
water
runoff,
open
water
bodies
and
rainwater.
All
of
these
flows
have
different
dynamics
in
space
and
time
and
therefore
the
level
of
spa-
tiotemporal
detail
at
which
information
is
needed,
varies
with
the
water
flows
targeted
by
an
intervention.
This
can
be
illustrated
by
comparing
the
interventions
“Water
square”
and
“Dike
reinforce-
ment”.
High
spatiotemporal
resolution
rainwater
data
up
to
small
neighbourhood
hourly
level,
is
needed
for
the
storm
water
man-
agement
intervention
“Water
square”.
The
information
need
for
“Dike
reinforcement”,
on
the
other
hand,
covers
the
range
months
to
five
years,
and
large
district
up
to
provincial
level.
The
relatively
high
spatiotemporal
resolution
of
rainwater
related
information
needs
is
known
to
be
prerequisite
to
assess
and
predict
urban
runoff
behaviour
(Fletcher
et
al.,
2013).
Likewise,
for
urban
flood
manage-
ment
it
is
essential
to
understand
the
dynamics
of
surface
water
flows
at
different
spatiotemporal
scales,
including
the
scale
of
the
catchment
level
and
a
long-term
perspective
(Zevenbergen
et
al.,
2008).
In
contrast,
energy-related
interventions
require
information