International Energy Agency

Technology Collaboration Programme on District Heating

and Cooling including Combined Heat and Power

REDUCING GREENHOUSE GAS EMISSIONS

A N D E N E R G Y CO N S U M P T I O N B Y
OPTIMIZING URBAN FORM FOR DISTRICT
ENERGY

A nnex X I Fi n al Repo r t

PREPARED BY
 IEA DHC Annex XI:
COPlan4DE
N T E N T S Final Report

Plan4DE Final Report

This project has been independently funded by the International Energy Agency Energy Technology
Initiative on District Heating and Cooling including Combined Heat and Power (IEA DHC).

Any views expressed in this publication are not necessarily those of IEA DHC.

IEA DHC can take no responsibility for the use of the information within this publication, nor for any
errors or omissions it may contain.

ANNEX XI FINAL REPORT, SEPTEMBER 2016

 IEA DHC Annex XI:COPlan4DE
N T E N T SFinal Report

Plan4DE Final Report

ABOUT THE INTERNATIONAL ENERGY AGENCY

The International Energy Agency (IEA) is an intergovernmental organisation that serves as energy
policy advisor to 28 member countries in their effort to ensure reliable, affordable and clean energy for
their citizens. Founded during the oil crisis of 1973-1974, the IEA was initially established to coordinate
measures in times of oil supply emergencies.

As energy markets have changed, so has the IEA. Its mandate has broadened to incorporate the Three
Es of balanced energy policy making: energy security, economic development and environmental
protection. Current work focuses on climate change policies, market reform, energy technology
collaboration and outreach to the rest of the world, especially major consumers and producers of energy
like China, India, Russia and the OPEC countries.

With a staff of nearly 200 who are mainly energy experts and statisticians from its 28 member countries,
the IEA conducts a broad program of energy research, data compilation, publications and public
dissemination of the latest energy policy analysis and recommendations on good practices.

ABOUT IEA DHC

The Energy Technology Initiative on District Heating and Cooling including Combined Heat and
Power was founded in 1983. It organizes and funds international research which deals with the design,
performance, operation and deployment of district heating and cooling systems. The initiative is
dedicated to helping to make district heating and cooling and combined heat and power effective tools for
energy conservation and the reduction of environmental impacts caused by supplying heating and cooling.

P L A N 4 D E : R E D U C I N G G R E E N H O U S E G A S E M I S S I O N S A N D E N E R G Y CO N S U M P T I O N B Y O P T I M I Z I N G U R B A N F O R M F O R D I S T R I C T E N E R G Y | 3
 CO N T E N T S

AC K N O W L E DG E M E N T S ADDITIONAL SUPPORT FROM

Plan4DE was developed through a collaborative design Ken Church and Raymond Boulter, Natural Resources
Canada
process by the following organizations:

Sustainability Solutions Group

GGLO

EA Energy Analyses

Government of British Columbia

International District Energy Association

Farallon Consultants Limited

Mary Ellen Richardson Consultants

ANNEX XI FINAL REPORT, SEPTEMBER 2016

 CO N T E N T S

Acknowledgements 6.5.4 DEEM

Executive Summary 6.5.5 Balmorel
1 Introduction 6.5.6 District Energy Concept Advisor
1.1 Objectives and goals 6.6 Comparison of the models
2 Methods 6.7 Open source
3 Determinants of district energy 6.7.1 Benef its and challenges
3.1 District energy def ined 6.7.2 Examples
3.2 The inf luence of policy 6.8 Findings for Plan4DE
3.3 Social determinants 7 Plan4DE
3.4 Economic determinants 7.1 Model characteristics
3.5 Technical determinants 7.2 Analytical approach
3.6 The built environment 7.3 Model design
3.7 Findings for Plan4DE 7.3.1 Technologies
4 The built environment and energy density 7.3.2 Revenue generation
4.1 Energy and climate 7.3.3 Network length
4.2 Energy use and urban structure 7.3.4 DE system lifetime
4.3 Building morphology 7.3.5 Individual heating
4.3.1 Building envelope 7.3.6 Waste heat and industrial heat demand
4.3.2 Building compactness 7.3.7 Currency
4.3.3 Glazing ratio and orientation 7.3.8 Versions
4.3.4 Trade-offs at the building scale 7.3.9 Inputs
4.4 Building systems 7.3.10 Outputs
4.5 Occupant behaviour 7.4 Data
4.6 The block and neighbourhood scale 7.5 Current Limitations
4.6.1 Urban heat island effect and vegetation 8 Validation
4.6.2 Renewable energy potential 8.1 Applying Plan4DE to a neighbourhood
4.7 Future buildings archetype
4.8 Findings for Plan4DE 8.1.1 A neighbourhood archetype
5 The discipline of planning 8.1.2 Observations
5.1 Planning and energy consumption 8.2 Applying Plan4DE to the City of
5.2 Planning and district energy Ottawa
5.3 Findings for Plan4DE 8.2.1 Comparison of methods for calculating heat
6 Modelling the built environment density
6.1 Classif ication of model approaches 8.2.2 Evaluation of neighbourhoods in
6.2 Urban energy models Ottawa
6.3 District energy models 8.2.3 Observations
6.3.1 Assumptions and theory behind the 9 Dissemination
modelling approaches 9.1 Website
6.3.2 Spatial and non-spatial dimensions in the 9.2 Webinars
models 9.3 Presentations
6.3.3 Level of decision-making that the models are 10 D iscussion
designed to inf luence 10.1 Future steps
6.3.4 Methods used for modelling 10.2 Conclusion
6.3.5 Graphical user interface Glossary
6.4 Existing models Abbreviations
6.5 Review of models Figures
6.5.1 Rambll model Tables
6.5.2 COWI Varmeatlas Bibliography
6.5.3 RETScreen
P l a n 4 DE : R e d u c i n g g r e e n h o u s e g a s e m i s s i o n s a n d e n e r g y co n s u m p t i o n b y o p t i m i z i n g u r b a n f o r m f o r d i s t r i c t e n e r g y | 01
 EXECUTIVE SUMMARY
E XE C U TIVE SU MMARY
The role of cities in reducing GHG emissions
is gaining increasing prominence. It has been
highlighted by the work of the C40, a global
network of large cities committed to helping
address climate change (Arup, 2011), the f irst
Intergovernmental Panel on Climate Change
(IPCC) chapter on human settlements (Seto &
Dhakal, 2014) and, a new reporting mechanism,
the Non-State Actor Zone for Climate Action 1
launched in December under the United
Nations Framework Convention on Climate
Change (UNFCCC). As the IPCC reports,
Addressing climate change has become part
of the policy landscape in many cities, and
municipal authorities have begun to implement
policies to reduce GHG emissions within their
administrative boundaries (2014, p. 970).
These efforts ref lect a rapidly urbanizing
world in which 54% of the population lives in
urban areas, a total that increases weekly and is
projected to climb to more than 60% by 2050
an additional 2.5 billion people (United Nations,
2014). Urban areas account for between 71%
and 76% of CO2 emissions (Seto & Dhakal,
2014). This continual process of urbanization is
requiring and will continue to require massive
investments in infrastructure, which, depending
on the approaches selected, will either impede
or enhance societys ability to reduce GHG
emissions.
In the absence of consideration of heat density
by planners, proponents of district energy are
restricted to those areas of cities that through
historical accident have achieved a suff icient heat
density, or where there is a signif icant anchor load
1 See http://climateaction.unfccc.int/
ANNEX XI FINAL REPORT, SEPTEMBER 2016
in close proximity to other buildings, to merit a
district energy system.
In taking decisions on land use, local governments
unconsciously either rule in or rule out district
energy by enabling development that is either less
than or greater than energy density thresholds,
generally the former. These decisions have long-
term implications as a result of the lock-in that
accompanies the capital intensive and durable
built environment. Energy density thresholds
are dependent on many different variables that
ref lect the characteristics and use of the built
environment.
While there is a range of considerations that
contribute to the feasibility of a district energy
sy stem, the character of the built environment
is a foundational aspect. This project developed
a new model, Plan4DE, that enables planners to
quickly and easily evaluate the impact of different
conf igurations of the built environment on the
viability of district energy relative to individual
heating systems.
Plan4DE was developed following an extensive
literature review using an integrated design
process that combined the expertise of planners,
engineers and architects. The Excel-based model
has three different interfaces involving different
levels of complexity for different use cases or
practitioners with different levels of experience.
The model uses inputs on characteristics of the
built environment to determine the capital and
operating costs of district energy versus individual
energy systems over variable lifetimes, as well as
environmental impacts in the form of greenhouse
 EXECUTIVE SUMMARY

gas emissions and criteria air pollutants. Planners

are then able to make informed decisions to enable
a built environment that supports district energy.

Unique characteristics of the model include an

algorithm for calculating the cost of a district
energy network and an optimization tool to select
district energy technologies.

Plan4DE is open source and while it is populated

with data for North American climate zones,
additional or different data sets can be readily
added.

The model was validated f irstly by developing a

neighbourhood archetype and applying a planning
process to this neighbourhood informed by the
use of Plan4DE. Secondly, Plan4DE was validated
using observed data for a Canadian city and then
used to evaluate district energy viability for a
range of neighbourhoods.

Plan4DE is open source so that users can adjust

assumptions or methods as required for particular
circumstances, climates or countries. The
model and a guidebook can be downloaded at
plan4de.ssg.coop.

P l a n 4 DE : R e d u c i n g g r e e n h o u s e g a s e m i s s i o n s a n d e n e r g y co n s u m p t i o n b y o p t i m i z i n g u r b a n f o r m f o r d i s t r i c t e n e r g y | 03
 INTRODUCTION
1 . I NT RO DU CTION
1 .1 O B J E C T I V E S & G OA L S
Legislators in California, Oregon, British
Columbia and most recently Ontario have
passed laws requiring municipal plans to include
greenhouse gas emissions targets (Government of
Ontario, 2016; Oregon Legislative Assembly, 2010;
Rutherford, S., 2009; Steinberg, D., 2008). These
requirements signal a growing understanding
of the relationship between the design of
community settlements, energy consumption and
greenhouse gas emissions, which also is ref lected
in work by international agencies. Urban form
has a determining inf luence on many sources of
greenhouse gas emissions as density and land-
use patterns affect distances travelled, modes of
travel, building typology, possibilities for district
energy systems, and carbon release and uptake by
biomass. One aspect of this inf luence is known
as path dependence; where current decisions can
limit possibilities for future decision-making
(Erickson & Tempest, 2015). The combination of
an expensive transition and the fear of forfeited
investments compel one to maintain the path of
least resistance despite ongoing disadvantages
(Liebowitz & Margolis, 2009).
Of all the energy-related decisions made
at a community level, it is likely that land-
use decisions have implications on the
communitys consumption of energy for the
longest time. Physical infrastructure and the
spatial organization of society can be seen as an
important form of carbon lock-in (Erickson
& Tempest, 2015), the process by which society
becomes further entrenched in a dependent
relationship with fossil fuels. Land-use decisions
therefore need to be prioritized in efforts to
reduce energy consumption and greenhouse gas
Plan4DE
emissions, above transportation management, site
and building design, and energy supply (Jaccard,
Failing, & Berry, 1997).
There are a growing number of tools available to
assess the impact of land-use planning decisions on
GHG emissions and energy consumption. Four
recent reports, one for the State of Washington
(Fehr & Peers, 2009), two by the Lincoln Institute
(P. Condon, Cavens, & Miller, 2009; Holway et
al., 2012), one prepared for California Transport
(Caltrans) (DKS Associates, 2007) and another
by researchers at Imperial College London (James
Keirstead, Jennings, & Sivakumar, 2012) provide
detailed reviews of existing models. Both Fehr
and Peers, and DKS Associates focus primarily
on the impacts of land-use on transportation and
categorize the available models by sophistication.
The Lincoln Institute report addresses available
modelling tools and evaluates the relative climate
change benef its of alternative development
scenarios at different scales. However, none of
these tools specif ically consider the potential of
district energy systems to reduce greenhouse gas
emissions from a land-use planning perspective.
There are also a number of tools and models
that can be used to evaluate the feasibility
of district energy systems. From a land-use
perspective, they are responsive models as they
are designed primarily to enable engineers to
design a district energy system given a pre-
existing built environment. Few assessment
approaches stress the importance of identifying
the general conditions, with regards to community
frameworks, required within a community for
advancing district energy (Brent Gilmour &
 O B J E C T I V E S & G OA L S

Warren, 2007). Responsive models are therefore

limited by looking only at the feasibility of district
energy relative to a pre-existing built environment
condition. An approach that ensures consideration
of energy implications during the planning of the
built environment and development approvals
increases the scope for creating opportunities for
district energy.

In order to maximize the potential for district

energy to contribute to reductions in greenhouse
gas emissions, planners need to incorporate
district energy as a consideration in the
development of municipal plans, upstream from
the actual development of the built environment
(King & Bradford, 2013). Consideration of district
energy at this early stage in the development
process will build literacy on district energy in
communities, result in enabling land-use patterns
and create new district energy opportunities.

PLAN4DE | 05
 methods
2. METHODS
The intent of this project was to use novel research
methods to design and implement a model that
meaningfully represents the relationship between
characteristics of the built environment (an
independent variable) and the viability of district
energy (dependent variable). There is extensive
research on this relationship, which was identif ied
in a literature review described below; however,
the novel aspect of this project was to facilitate the
evaluation of the impact of the built environment
on the viability of district energy from the
perspective of a city planner and to do so using
a tool suitable to different contexts around the
world.
The construction of a model that is applicable
for the discipline of urban planning while
incorporating aspects of the f ields of systems
modelling, architecture, urban design, civil
engineering, energy systems engineering and
economics is by def inition interdisciplinary.
Building on past experience, the team elected to
employ a technique developed in the design of
high performance green buildings, the integrated
design process (IDP) (Keeler & Vaidya, 2016).
An IDP focuses on broad involvement of the
different disciplines of the team in the initial
design stages during which key decisions are taken
that determine the outcome of the project. To this
end, the team was convened for a three-day design
charette to set the stage for the interdisciplinary
collaboration that characterizes the IDP
approach. The charette began at a high level with
decisions on the goals and objectives, and became
increasingly specif ic, as model parameters were
identif ied and ultimately initial efforts at model
design and implementation were undertaken.
Plan4DE
Following the IDP, a literature review was
completed, again using a collaborative drafting
process on the web-based platform, Google Docs.
This approach enabled team members from both
Europe and North America to contribute in real
time to the same document. The literature review
took stock of the literature on district energy,
focusing on existing models and the inf luence of
planners. Aspects included a description of the
mandate of planners and the f ield of planning,
consideration of the relationship between the
built environment and energy intensity, and
energy intensity and the feasibility of district
energy. An exploration of the tools available
to planners in considering the impacts of their
work on district energy was undertaken, looking
specif ically at the state of the modelling f ield
with respect to district energy, and at methods to
generically characterize the built environment.
Finally, the co-benef its, or the ancillary benef its
that may arise if planners decide to factor district
energy as a consideration into their planning
decisions were considered.
With goals and objectives identif ied, the
completion of an initial foray into the design
and implementation of the model, and the
completion of a detailed literature review, the
team then divided into four sub-groups. A
model development team developed iterations of
the model, with the rest of the team providing
feedback on each new version. A data team
compiled relevant assumptions and inputs to
support the model for specif ic jurisdictions. A
planning and design team began conceptualizing
how the model would be applied in the context
of urban planning and design processes. Finally,
 METHODS
a dissemination team shared the proposed
concepts and approach more broadly to ensure
the relevance of the approach, in the form of
conference presentations and webinars.
The full team congregated to address specif ic
design challenges or questions that emerged
from the sub-groups. For example, one question
related to approaches to represent the length
and size of pipes in a distribution system in the
absence of detailed spatial data. A team member
compiled characteristics of a sample of existing
district energy systems and used an analysis of
these characteristics to derive an algorithm that
calculated pipe length and size for different forms
of development.
A process of data collection was undertaken to
provide key assumptions for the model. Key
requirements for data included:
Individual and district energy heating
and cooling technologies including total
eff iciency, electric eff iciency, thermal
eff iciency, lifetime, fuel source and capacity
factor,
Financial assumptions including capital
costs, f ixed and variable operating and
maintenance costs, fuel cost, carbon costs,
system integration cost, network costs,
balancing costs and power revenue costs.
Emissions factors for each technology
including CO2, CH4, N20, NOX.
Building energy characteristics including
energy intensities, peak energy demand
and total energy consumption for f ive
residential and f ifteen non-residential
building archetypes for more than twenty
different climate zones.
Data was collected for the jurisdictions of
Denmark, Canada and the US drawing on the
expertise of the team members; the user can input
PLAN4DE | 07
additional data specif ic to their jurisdiction if
available.
A process of data collection was undertaken
to provide key assumptions for the model
including specif ications for heating and cooling
energy technologies, energy consumption and
peak energy demand for building archetypes in
different climate zones, and capital and operating
costs for different district energy technologies.
Once a functioning model was developed, an
extensive process of validation was undertaken
using sensitivity testing and the application of
three case studies (neighbourhoods in the Cities
of Copenhagen and Ottawa and an abstraction
of a neighbourhood in Seattle). In the case of
Copenhagen, the model results were compared
with data from an existing district energy system
in a particular neighbourhood in order to test
the predictive ability of the model. In the case
of Ottawa, a comparison of the results of three
different approaches to characterizing the energy
prof ile of the built environment was undertaken.
Finally, the process in Seattle was f irstly a test of
usability from the perspective of the planning
profession and secondly of the f indings of
the analysis in a relatively mild climate. This
validation process resulted in various iterations of
the model to increase accuracy and usability.
The model was then evaluated in the process
of developing a neighbourhood (derived from
Seattle) as would be experienced by a planner and
subsequently a further iteration of the model was
developed as a result of insights in this process.
A user interface was developed and a guidebook
on the use of the model was prepared prior to a
public launch of the model.
 DETERMINANTS OF DISTRICT ENERGY

Transformations in energy systems are long-term

change processes (decadal or longer) in technology, the
economy, institutions, ecology, culture, behavior, and
belief systems (Patwardhan et al., 2012, p. 1176).

3. D E TE R M INA N TS OF DISTRICT EN ERG Y

This section reviews the concept of district energy,
its opportunities and challenges, and the role of
the built environment amongst the factors that
inf luence the deployment of district energy.
3.1 D I S T R I C T E N E R G Y
DEFINED
At its most basic level, district energy is def ined
as an approach to energy provision in which
heat is produced centrally and hot water is piped
to the buildings (Davies & Woods, 2009).
District energy can provide either heating for
space heating and hot water, or chilled water
for air conditioning. District energy systems
generally consist of three sub-systems including
the collection and/or generation of thermal
energy, the distribution of the thermal energy
from plant sites to a network of energy consumers,
and the transfer of the thermal energy to the
energy consumer (B. Gilmour & Warren, 2008;
Rodger, 2010). A district can consist of a small
neighbourhood or a large part of a community
(Rodger, 2010).
There are many existing and future energy
demands for low temperature heat that
are currently being satisf ied by employing
combustion of fossil fuels, which provides high
temperature heat a mismatch. Similarly, the
use of electricity for these heat loads means using
a high-quality energy source for a low-quality
heat demand. It is therefore preferable to use the
difference between the high temperature heat
and the low temperature demand to generate
electricity. However, this is not generally feasible
without a centralized system like district energy,
which facilitates the capture of additional and
signif icant eff iciencies (Rogner, 1993). The
improved eff iciencies translate into economic
benef its and reduced pollution. Further,
district heating helps overcome three barriers
to the wider dissemination of renewable energy,
namely: periodic f luctuations associated with the
generation; the low energy density of renewable
energy carriers relative to fossil fuels; and,
the limited utilization of renewable energy by
consumers, including provision, conversion,
storage, transport and use (Stremke & Koh, 2011).
The United Nations Environment Program
(UNEP) identif ied the following policy goals that
cities hoped to achieve in applying district energy
(United Nations Environment Program, 2014,
p.5):

Plan4DE
 DETERMINANTS OF DISTRICT ENERGY
Reduce greenhouse gas emissions: Achieve
rapid, deep and cost-effective emissions
reductions, with reduction of primary energy
consumption by 3045 per cent.
Improve air quality: Address indoor and
outdoor air pollution and their associated
health impacts, by reducing coal and oil
consumption.
Improve energy eff iciency: Achieve operational
eff iciency gains of up to 90 per cent, by using
district energy infrastructure to link the heat
and electricity sectors.
Local and renewable resources: Harness local
energy sources, including from waste streams,
reject heat, natural water bodies and renewable
energy. Pilot new technologies, such as thermal
storage, to integrate renewables.
Resilience and energy access: Reduce import
dependency and fossil fuel price volatility.
Manage electricity demand and reduce the risk
of brownouts.
Green economy: Achieve savings from avoided
or deferred investment in generation
infrastructure and peak power capacity.
Create wealth through reduced fossil fuel
bills and generate local tax revenue. Create
jobs for design, construction, operation and
maintenance.
In a perfectly rational society with scarce
resources, district energy is installed when the
cost of constructing and operating district energy
is less than the cost of other technologies or
strategies to provide the same service. The cost of
district energy is determined by a myriad of factors
including the scale of the installation, the density
of heat demand, the availability of expertise, the
cost of energy generation, the cost of carbon, local
geology and geography, the availability and type
of fuel sources, the design of existing buildings,
the future planning regime, the legal context
PLAN4DE | 09
and others. There is an equally large range of
variables that determine the cost of a building-
level heating or cooling system. In an assessment
of the potential for district energy expansion
in Europe, the key variables that were assessed
included heat demands, settlement structures,
land-use priorities, excess heat activities, local
heat resources, and general geographic properties
of any given location (Connolly et al., 2013).
However, in addition to the economic factors,
there are also externalities that inf luence the
uptake of district energy.
A helpful distinction in understanding the
convergence of these and other factors, is to
identify two specif ic contexts for district energy,
where:
District energy is mature. These countries have
relatively higher-priced fossil fuels, policies
which support district energy, mature and
extensive systems, extensive experience with
the technical and f inancing aspects of district
energy, and a history of public acceptance
of local energy infrastructure. This group
includes Scandinavian countries, Finland,
France, and Germany.
District energy is emerging. These countries
have relatively lower-priced fossil fuels,
policies which tend to support energy systems
based on centralized utilities, few district
energy systems, modest district energy
experience, and limited public acceptance
of new energy infrastructure entering a
community. This group includes Canada, the
United States, the United Kingdom, and most
developing countries.
Evaluating the feasibility of district energy
for countries in the f irst group is relatively
straightforward, based on well-established urban
planning policies, a high degree of certainty
regarding the accuracy of capital and operating
cost estimates, and conf idence in public
acceptance of new or expanded systems. These
 DETERMINANTS OF DISTRICT ENERGY
factors tend to reduce the risk of developing or
expanding a district energy system.
For a proposed district energy system in a country
belonging to the second group, the economic
and social advantages must be strongly positive
to overcome policy barriers and f inancial
uncertainty. Examples of such favourable
conditions would include new, high-density
developments in northern communities, where the
cost of conventional energy sources is high.
3.2 T H E I N F L U E N C E
OF POLICY
Policies that support district energy include
regulations requiring local governments to
understand local energy needs and to plan to meet
those needs with renewable sources. Examples of
such regulations that either indirectly or directly
support district energy include:
The Merton Rule adopted by in the Merton
Council in the United Kingdom. The rule
requires new commercial buildings with an
area greater than 1,000 square metres to meet
more than 10% of their energy needs with
renewable sources (Pitts, 2008).
The Heat Supply Law in Denmark, which
required local governments to identify their
energy needs and to recognize alternatives for
meeting those needs (Sovacool, 2013).
Regulations requiring public sector
organizations to achieve carbon neutral
operations tend to encourage schools,
universities and hospitals to consider low
carbon forms of district energy. British
Columbias Carbon Neutral Government
Regulation is an example of this kind of
regulation (Government of BC, 2008).
The cost of carbon taxes and carbon offsets
tends to support low carbon forms of district
Plan4DE
energy by increasing the cost of the business-
as-usual (e.g. fossil fuel) option (Johansson,
2000).
Streamlining the approval process for projects
that f it a local governments criteria
for sustainable development (including
developments associated with low carbon
forms of district energy) tends to support
district energy by reducing the developers
costs (R. Bush, Bale, Taylor, & Gale, 2014).
These policies are designed to account for
and or address a broad range of externalities.
Municipalities are instrumental in the deployment
of district energy because of their inf luence
over land-use and infrastructure. Municipal
governments are, however, required to deliver a
variety of public services and respond to direction
and mandates from higher order government.
As a result, many competing budgetary interests
exist. Enshrining thermal energy priorities as
they relate to broader environmental, economic
and social objectives is necessary to validate
and support local governments investment and
commitment to district energy development. This
also helps garner public support. Additionally,
having a long-term vision for community energy
infrastructure gives conf idence to developers and
investors (R. Bush et al., 2014).
In many municipalities, urban planning is
fragmented among individual local governments
and the regional government to which they
belong. In addition, plans for land use,
energy infrastructure, water infrastructure,
transportation, sustainability and climate change
are commonly developed by separate departments.
This fragmentation presents a structural challenge
to implementing district energy systems, the
planning of which requires engagement by
multiple disciplines (Dale, 2001).
Local governments, however, occupy a unique
position from which they can serve as one of, or
a combination of, the following: planners and
 DETERMINANTS OF DISTRICT ENERGY
regulators, brokers and facilitators, generators,
distributors, consumers and project coordinators.
While federal, provincial or state programs may
supply targets or provide strategic direction, local
governments often are charged with implementing
district energy solutions. Even in the case of
private sector district energy development, local
governments typically provide support through
policy frameworks, franchise agreements,
easements and zoning, taxation relief, some level
of investment or capacity building with the
public.
In North America, regulatory frameworks for
energy are designed to address issues related to
centralized utilities, which can present barriers to
smaller, locally owned energy systems. Apart from
the regulations themselves, the cost, complexity
and duration of regulatory energy processes
present a barrier to district energy systems.
In North America, utilities may have mixed
responses to district energy and distributed
energy, based on concerns over loss of business,
increased burden of the cost of common
distribution infrastructure (e.g. natural gas
pipelines and electricity distribution lines) on
remaining customers, higher upfront costs, and
the necessity of complex partnerships (Cooper &
Rajkovich, 2012).
Planning for district energy is inhibited by an
inadequate understanding on the part of urban
planners, consultants and building owners
regarding energy in general and in particular
the technological possibilities, the full business-
as-usual costs of energy systems (including
environmental and social costs of fossil fuel
exploration and production), and the full cost of
providing energy (i.e. owning and operating the
buildings boilers, and not just the cost of fuel)
(Hawkey, 2009).
P L A N 4 D E | 11
3 . 3 S OC I A L
DETERMINANTS
Social challenges to district energy include a
general lack of understanding, and the necessity of
engaging a large number of disparate stakeholders
that typically includes individuals, as well as
private and public institutions (M. R. Bush,
2013). District energy is simpler to implement in
communities that are aware of their air quality
issues (for example when pollution is caused by
open burning of wood waste). District energy
planning and implementation is simpler when
the local culture places a relatively higher value
on the greater good of the community and the
environment, versus individual interests.
Public concern regarding new energy sources
entering the community can inhibit systems based
on biomass. If inadequate time and planning
is applied to identifying and addressing public
interests during the earliest possible stages of a
district energy project, the public consultation
process itself can fail to overcome public
opposition to integrating low carbon forms of
district energy into a community (Comeault, C.,
2011). Also, interest on the part some building
owners and community members regarding
climate change can be limited.
Since design professionals tend to be
commissioned to design individual buildings
on a single site, their focus is often limited
to that site. Design professionals tend not to
look for opportunities to share energy and
water with adjacent buildings or sites, or to
work cooperatively with municipal planners or
municipal engineers to f ind energy schemes that
can help both the building site and the community
(Ronn, 2011).
 DETERMINANTS OF DISTRICT ENERGY
3.4 E CO N O M I C
DETERMINANTS
District energy requires a major initial investment
as well as ongoing operations and maintenance
costs. The economics of district energy are
improved by the relatively higher energy
consumption of buildings in colder climates,
relatively higher costs of conventional sources of
energy, larger scale projects that reduce the per
unit energy costs of systems, and the availability
of district energy equipment (e.g. heat pumps,
biomass boilers and distribution piping)
manufactured domestically, which reduces the cost
and complexity of importing materials.
Exploration subsidies for fossil fuels ($34 billion/
year in Canada and $600 billion/year worldwide)
interfere with market pricing by using public
funding to make fossil fuels less expensive
(Clements, B. et al., 2013). Since district energy is
capital intensive, the cost of capital (i.e. interest
rate) is critical. Economies of scale limit the
viability of smaller systems, which have higher
costs per unit of delivered energy. The current
shale gas boom has increased the availability of oil
and natural gas, and decreased the price of fossil
fuels.
In countries where district energy is mature, the
economics of district energy are well understood,
and the criteria for viability are well established.
The higher costs of business-as-usual energy
sources, and the lower cost of widely available
district energy infrastructure in these countries
combine to make the threshold for district energy
viability relatively lower.
In countries where district energy is emerging,
the economics of district energy can be marginal,
and since the advantages of district energy may
be less obvious to stakeholders, those involved in
planning systems must invest more effort and be
more creative to make systems viable. Lowering
the threshold for viability in these countries can
Plan4DE
involve more opportunism, for example:
Reducing the cost of the district energy
fuel source by relying on sources of waste
industrial heat.
Reducing the cost of distribution piping by
choosing routes through parks, through streets
that must be renewed for other reasons, and
by sharing the cost of trenching with other
utilities such as potable water, wastewater,
electrical utilities or communications utilities.
Finding owners of existing buildings with a
greater interest in district energy because their
boilers need replacing, and f inding developers
of new buildings who recognize the value
of re-programming utility space as revenue-
generating space.
The cost of energy from conventional sources
has a signif icant inf luence on the viability of
district energy. The cost of providing energy to a
building includes the cost of conventional fuels,
including taxes, carbon taxes, carbon offset costs
and the cost of converting fuel to energy, which
is inf luenced by the eff iciency of the boiler, the
capitalization cost of the boiler (initial cost,
f inancing cost, and life expectancy), the cost
of utility operations, maintenance, insurance,
licensing, and the opportunity cost associated
with the utility space in the building.
The operating temperature also impacts the cost
of the system. A low-temperature district heating
system reduces thermal stress on the pipes, results
in reduced heat loss in the network and storage,
increases the eff iciency of combined heat and
power, and increases the ease of renewable energy
integration (Rosa, 2012).
 DETERMINANTS OF DISTRICT ENERGY
3.5 T E C H N I C A L
DETERMINANTS
District energy systems are, like all technologies,
evolving rapidly. The f irst generation used steam
as a carrier of heat and was introduced in the late
nineteenth century in the US. A second generation
switched to pressurized hot water, to address the
heat loss, high operating and maintenance costs,
and safety issues associated with steam. A third
generation used pressurized water with lower
supply temperatures, twin pipe systems and plastic
pipes. The fourth generation is characterized as
low energy district energy and will supply new
low-energy buildings as well as buildings which
have undergone deep energy retrof its (Rosa, 2012).
The eff iciency of the district energy system
depends on the eff iciency of conversion of fuels
to useful energy and energy losses throughout
the system. The integration of energy storage
and electricity generation seamlessly with the
provision of thermal energy is also inf luencing
the eff iciency of the system (DFIC - Dr. Fromme
International Consulting, 2016).
Load diversif ication tends to reduce the peak
capacity of a district energy system, and therefore
its cost. Engineering modelling capability and
capacity to optimize district energy networks
is now well established and widely available
(e.g. Natural Resources Canadas RETScreen
International program).
New building technologies and new energy
technologies can make buildings energy self-
suff icient, which tends to reduce the need for
district energy. In the short term, the higher-
temperature design of energy systems in existing
buildings requires higher-temperature energy
sources (e.g. biomass combustion rather than
heat pumps) and higher-temperature distribution
piping (e.g. steel vs. polymer), which increases
both the capital and operating costs of district
energy (Dalla Rosa & Svendsen, 2011).
P L A N 4 D E | 13
3.6 THE BUILT
ENVIRONMENT
Keeping the above variables in mind, it is clear
that heat density has a signif icant inf luence
on the viability of district energy (The Energy
Saving Trust, 2008), with a positive non-linear
correlation between the increasing proximity of
heat demand and the cost of the system (Woods,
2012). Consideration of the extremes illustrates
the nature of the relationship: when two heat
loads are in close proximity, it is relatively easy
to connect them and heat loss is not a critical
consideration, but when two heat loads are distant
from each other, the infrastructure cost becomes
prohibitive and the necessity of minimizing heat
loss mandates costly insulation. As a consequence,
there is, in any particular location, a threshold
of heat density at which district energy is cost
effective, but this threshold varies according to a
wide range of variables. For example, in the UK
that threshold has been measured at 200 MWh/
ha (Woods, 2012) and 3,000 kWh/km2 (Davies
& Woods, 2009). In terms of linear heat density,
areas with 0.20 MWh/m,year are considered to
be feasible in Denmark (Rosa, 2012) and 0.30
MWh/m.year elsewhere (Zinko et al., 2008).
Linear heat density is inf luenced not only by
energy density but also by the conf iguration of
the buildings within that area, as this inf luences
the length and size of the pipes required (Zinko
et al., 2008). Plan4DE will provide a more precise
mechanism to develop heat densities thresholds
for specif ic contexts, according to the selection
of technologies, the climatic zone, fuel costs and
building mix.
Of the three levels of government, it is local
governments that have legal jurisdiction over
land-use planning, which ultimately determines
the proximity of heat loads, or the heat density.
As Floater and Rode write, spatial planning
provides a way to engage with the f lows of the city
in pursuing the optimal sequencing, coordination
and integration of infrastructure investments
 determinants of district energy
(2014, p. 23). In order for district energy to be
broadly successful, local governments need to
incorporate a consideration of district energy, and
energy demand in general, into land-use decisions.
In an analysis of community energy planning,
Jaccard et al (1997) propose that of the various
powers available to local governments, land use is
the most critical as it has the longest durability,
beyond policies on transportation management,
site and building design, and energy supply, for
example.
Land-use decisions result in lock-in, or path
dependence. Lock-in happens when an activity
or process continues even though it may not be
benef icial but is tied to previous investments
or commitments (Liebowitz & Margolis, 2009).
Signif icant investments in bricks and mortar
are a consequence of land-use decisions and
it is diff icult to abandon those investments
even if they are impeding other societal goals
such as reducing GHG emissions. The pattern
of development which cities plan and permit
therefore determines the energy consumption
and GHG emissions into the future (Creutzig,
Baiocchi, Bierkandt, Pichler, & Seto, 2014; World
Resources Institute, 2014).
In taking decisions on land use, local governments
unconsciously either rule in or rule out district
energy by enabling development that is either less
than or greater than energy density thresholds.
These decisions have long-term implications as a
result of the lock-in that accompanies the capital
intensive and durable built environment. Energy
density thresholds are, however, dependent
on many different variables that ref lect the
characteristics and use of the built environment.
In the absence of consideration of energy density
by planners, proponents of district energy are
restricted to those areas of cities that through
historical accident have achieved a suff icient heat
density, or where there is a signif icant anchor
load in close proximity to other buildings, to
merit a district energy system. This approach,
Plan4DE
characterized by the use of heat maps or atlases
(Mller & Nielsen, 2014), can be characterized
as reactive and inevitably results in signif icant
portions of the building stock that cannot
be economically serviced by district energy.
However, this type of energy planning is critical
to developing an understanding of energy
density and district energy feasibility. An
upstream approach would include an assessment
of energy density as integral to the creation of
off icial city plans and approving developments
in the same manner that other variables such as
transportation, servicing and topography are
considered.
Energy density is typically calculated at the
building level on the basis of intensity, for
example as GJ/m2/year or kWh/m2/year. At
the community level, energy density is typically
calculated as GJ/Ha/year or kWh/Ha/year. Energy
density is inf luenced by the climate; the energy
eff iciency of existing or planned buildings in
terms of kWh/m2/year, which is a function of
the design and use of the building; the density
of f loor space in relation to the footprint of the
building, i.e. expressed as Floor Area Ratio (FAR)
or Floor Space Ratio (FSR); and the use of the
building space (see Section 4 for more detail on
these variables). For example, hospitals and other
health care facilities consume a signif icant portion
of a communitys total energy for domestic hot
water, the consumption of which is independent
of the building envelope design. In addition,
domestic hot water is used throughout both
summer and winter seasons, which improves the
system utilization and therefore the economics of
district energy.
A lower energy density can be achieved with an
anchor load, consisting of a single large energy
demand. First, economies of scale mean that
the unit cost of providing energy to the anchor
load are likely to be the lowest in the system,
which provides an early indication of the likely
viability of the district energy system, as well as
a cost benchmark for other connections. Perhaps
 DETERMINANTS OF DISTRICT ENERGY
more importantly, the energy managers of larger
buildings (e.g. hospitals) are more likely to
understand their business-as-usual costs, as well
as the benef its of district energy. As a result,
the effort required to persuade such managers
to commit to district energy is relatively lower.
In addition, the location of the anchor load
can have a positive inf luence over the decision
regarding the location of the energy centre for
the district energy system. Locating the energy
centre close to the largest loads will reduce the
capital and operating cost of distribution piping.
Finally, the commitment by an anchor load to a
new district energy system can reassure owners
of other, smaller buildings who may have less
understanding of their business-as-usual costs and
the benef its of district energy (Davies & Woods,
2009; Rodger, 2010).
Higher urban density improves the viability
of district energy as economies of scale mean
that the district energy service providers cost
of connecting larger buildings is lower per unit
of energy delivered. As the cost of distribution
piping is one of the most signif icant components
of a district energy system, shorter distances
between high-density buildings signif icantly
reduces this cost.
In addition to higher density, developments that
include mixed uses also support district energy in
interesting ways. District energy systems must be
designed to meet the peak energy demand of the
system as a whole, and the ratio of peak demand
to average demand strongly affects the economics
of the system. If all buildings served by a district
energy system are of the same type (for example
multi-family urban buildings), then the timing
of peak energy demand for every building will
be similar (e.g. early morning and early evening).
If, on the other hand, the connected buildings
include a mix of homes and different types of
business, then the timing of peak demand will be
diverse, which tends to reduce the ratio of peak
demand to average demand (Hawkey, 2009).
P L A N 4 D E | 15
This effect of reducing the magnitude of peak
energy demand has a signif icant inf luence on the
capital and operating costs of a district energy
system. Higher peak demand requires larger
capacities for energy generation, distribution
piping, and energy transfer stations capacity
which by def inition will be under-utilized
majority of the time. In addition, because of the
relatively high capital cost of low-carbon sources
of heat (for example biomass boilers or heat
pumps), it is not practical to size this equipment
to meet peak demand. In low-carbon systems, peak
demand is typically met with conventional fuels,
which increases the operating cost and greenhouse
gas emissions of the system.
In addition to diversif ication in the timing of
demand, mixed use can result in diversif ication in
the form of energy consumed, such as temperature
and end use. For example, hospitals in Canada are
designed to use 100% makeup air, which increases
energy consumption for space heating, but also
presents an opportunity to recover energy from
exhaust air. Further, in health care facilities
(e.g. hospitals, retirement homes) as much as
40% of the total energy consumed can be used
for heating domestic hot water, which presents
further opportunities for energy recovery and
energy exchanges with other buildings. Recovered
energy which cannot be used by one building
can potentially be used by another in an energy
exchange arrangement (Chu, Richardson, &
Rogowska, 2014).
3.7 FINDINGS FOR
PLAN4DE
District energy can support societal objectives
related to greenhouse gas emissions, air pollution,
resilience and local economic benef it. There are
various factors that inf luence a district energy
project, beginning with the characteristics of the
built environment, but also including the policy
context, economic factors, culture and geography.
While Plan4DE is explicitly intended to address
 THE BUILT

EN V I R ONMENT &

ENE R GY DEN s i TY

The built environment is defined as man-

made surroundings that provide the setting for
human activity, ranging from the large-scale civic
surroundings to the personal places (Moffatt & Kohler, 2008, p.
249).

4 . THE BUILT EN V I R ONMENT & ENE R GY

DENSITY
the conf iguration of the built environment, it can
also provide insight in f inances and policies and
act as an educational tool.
The built environment is def ined as man-
made surroundings that provide the setting for
human activity, ranging from the large-scale civic
surroundings to the personal places (Moffatt
& Kohler, 2008, p. 249). While there are many
human activities that occur within and around
the built environment, this project explores the
implications of the built environment with respect
to the requirement for and provision of thermal
energy, using the strategy of district energy. To
better inform these considerations, planners need
to consider a number of the factors that inf luence
the intensity of energy consumption (kWh/m2)
and energy demand (kW/m2) at the land-use scale.
As buildings become more eff icient, through
the adjustment of the factors considered below,
energy density decreases, and the viability of
district energy therefore also decreases. Planners,
when assessing future built environments,
face a complex trade-off in that they need to
contribute to and plan for decreased energy
density at the building level, while increasing
energy density at the land-use level to increase
the feasibility of district energy. To better inform
these considerations, planners need to consider
a number of the factors that inf luence energy
intensity at the land-use scale.
Heating and cooling energy density are a
function of the characteristics and use of the
built environment. While these characteristics
represent design and planning decisions, they in
turn are driven by economic activity, fuel prices,
climate and other variables (Creutzig et al., 2014).
The growing focus on the relationship between
the built environment and energy demand is
highlighted in Chapter 8 of the IPCCs Fifth
Assessment Report and explicitly recognizes of
the extent of urbanization anticipated, and the
idea that urbanization can itself be a greenhouse
gas emissions mitigation strategy that integrates
multiple sectors (Seto & Dhakal, 2014). The
IPCC chapter specif ically references the urban
form characteristics of density, land-use mix,
connectivity, and accessibility as drivers of energy
and GHG emissions.
Data on energy consumption in buildings is
currently measured by the utilities that provide
energy to these buildings, and is ref lected in
a utility bill. This data provides a reliable and
accurate basis from which to understand the
energy demands of the existing building stock
and to model the energy demand of the future
building stock. This data however is often hard to
come by, due to privacy constraints.
Energy demand for buildings is expressed using
energy intensity factors, such as GJ/m/yr or
eKWh/m2/yr. Higher building energy intensity
factors combined with higher built form densities
(i.e. more building f loor area) result in higher
energy demands, increasing the potential benef it

Plan4DE
 THE BUILT
ENVIRONMENT &
ENERGY DENSITY
and feasibility of a district energy system. Figure
1, adapted from Ratti et al. (2005), Salat (2009),
and Miller (2013), illustrates the factors that affect
energy consumption in buildings; these are factors
that inf luence the heating and cooling energy
density.
Urban form or land-use planning plays a role in
building energy consumption in both building
morphology and urban structure, and in two
key ways: heat transfer and solar access (Miller,
2013). At the individual building scale, building
morphology refers to characteristics such as:
size and shape of a building, surface to volume
ratio (S:V), plan depth, building height, and
faade design characteristics (glazing ratios and
distribution, building orientation, and envelope
performance). At the block or neighbourhood
scale, urban structure refers to the arrangement
and spacing of buildings, streets, and open space,
which include measures such as density (f loor area
ratio or FAR), land coverage, and spacing between
buildings. The neighbourhood scale also captures
the cumulative impact of the relationships
between multiple buildings (building massing,
heights, arrangements and spacing), and their
impact on energy consumption. The impacts
of urban form on energy at the neighbourhood
scale are less frequently considered and have less
developed methods available for modelling in
comparison with those available at the individual
building scale (Miller, 2013)
and spacing of buildings, streets, and open space,
which include measures such as density (f loor area
ratio or FAR), land coverage, and spacing between
buildings. The neighbourhood scale also captures
the cumulative impact of the relationships
between multiple buildings (building massing,
heights, arrangements and spacing), and their
impact on energy consumption. The impacts
of urban form on energy at the neighbourhood
URB AN B UILDING
CLIMATE ST RUC T URE/ MO RPH OLOG Y
CO NT EXT
Figure 1: Factors influencing energy intensity in buildings
P L A N 4 D E | 17
scale are less frequently considered and have less
developed methods available for modelling in
comparison with those available at the individual
building scale (Miller, 2013).
4 . 1 E N E R G Y & C L I M AT E
In general, energy demand is climate and context
specif ic. For heating demand, the warmer the
climate the lower the operational energy due to
the reduction in heating energy required (Newton
et al., 2000). However, this does not necessarily
imply that buildings in warmer climates use less
energy overall, as warmer climates may result
in increases in energy demand for mechanical
cooling. The most prevalent working f luids in
mechanical cooling systems on the market today
are hydrof luorocarbons (HFCs), which contribute
to climate change.
Urban form or land-use planning plays a role in
building energy consumption in both building
morphology and urban structure, and in two
key ways: heat transfer and solar access (Miller,
2013). At the individual building scale, building
morphology refers to characteristics such as:
size and shape of a building, surface to volume
ratio (S:V), plan depth, building height, and
faade design characteristics (glazing ratios and
distribution, building orientation, and envelope
performance). At the block or neighbourhood
scale, urban structure refers to the arrangement
The mechanical cooling load is also dependent
upon building design and whether non-mechanical
cooling methods, such as operable windows, are
implemented. Climate has a signif icant impact on
heating and cooling energy consumption, and the
extent of this inf luence is determined by the way
in which the built environment responds to the
climate, through design, or the lack thereof.
B U ILD ING OCCU PA NT E N ER G Y
SY ST E MS B E H AVIOU R CO N S U M P TI O N
 THE BUILT
ENVIRONMENT &
ENERGY DENSITY
4.2 E N E R G Y U S E &
URBAN STRUCTURE
Density is a measure of the concentration of
development (residential, commercial, industrial,
etc.) per unit of land. Density is measured in a
number of ways including people per hectare/
acre, dwelling units per hectare/acre (for
residential), and f loor space ratio (FSR) or FAR.
FSR and FAR are a measurement of the ratio of
a buildings f loor area to the area of the lot on
which the building is built (Senbel et al., 2010). As
density increases, total energy demand increases
(i.e. the energy intensity factor is applied across
a greater f loor area). It is important, however,
to understand how energy-intensity factors,
particularly for residential and commercial uses,
change as building density changes.
As Doherty et al. (2009) explains, the concept of
density is plagued by several def initions that vary
depending on the purpose, and confusion arises
partly because density can refer to either dwellings
or people per unit area, as noted above. Where
calculations of people per unit area are used,
these are based on an assumed average number of
people per dwelling/unit. Most def initions are
linked as they attempt to identify low-, medium-
and high-density categories in a relative sense for
planning purposes, but the precise meaning of
such categories varies signif icantly both within
and between countries.
A variety of researchers note that as density
increases, buildings typically consume less
energy on a per capita or per area basis (Salat,
2009; Steemers, 2003). That is, energy intensity
factors decrease as building density increases.
Shared walls and f loors/ceilings and shared
building mechanical systems contribute to lower
heating and cooling loads (Senbel, M, et al,
2010). Newton et al. (2000) show that annual
heating and cooling energy per square metre
for apartments are very similar to those for
detached houses, despite the total f loor area of
Plan4DE
an apartment being less than half of a typical
detached house. However, when comparing
energy per person (by taking into account people
per unit), the energy usage for apartments is
signif icantly lower than for detached houses,
as the occupancy per square meter is higher for
apartments than for detached houses.
In contrast, models examining increased density in
commercial buildings have shown potential overall
increases in energy consumption per area from
increased cooling demands, due to high levels of
internal heat gains present in these buildings.
4.3 BUILDING
M O R P H O L OG Y
4. 3. 1 B U I L D I N G E N V E L O P E
The building envelope (i.e. the physical
separator between a conditioned indoor space
and unconditioned outdoor space) and its
performance have a signif icant impact on building
energy consumption. As the majority of energy
consumed in buildings is for space heating and
cooling, heat transfer between interior and
exterior space determines the level of energy
required in a building to maintain comfortable
levels of heating and cooling. Improving the
thermal performance level of a dwelling can have a
dramatic effect on its heating and cooling energy
consumption. In general, energy use decreases
as thermal performance increases (Newton et
al., 2000). However, results also show that as
envelope performance increases in certain cases,
heating demand decreases while cooling demand
increases (Miller, N., 2013). This is in part due
to other factors such as glazing ratio, building
compactness, and local shading.
4. 3. 2 B U I L D I N G CO M PAC T N E S S
A buildings compactness refers to the surface-to-
volume ratio of a building, or the ratio between
total building surface area and total enclosed
building volume. In general, as building density
 THE BUILT
ENVIRONMENT &
ENERGY DENSITY
increases (and building height increases), more
building volume is enclosed by building surface,
resulting in a lower surface-to-volume ratio
(S:V), making a building more compact. The
predominant effect of compactness is reducing
heat transfer, as more compact building shapes
enclose more building volume with less surface
area through which heat can escape (Miller, 2013).
4.3 .3 G L A Z I N G R AT I O & O R I E N TAT I O N
Glazing ratio refers to the proportional amount
of glazing (including windows) on a building,
that is, the ratio between total glazing area and
total building area. Heat loss through glazed
surfaces is higher than typical wall assemblies, and
as such is a signif icant factor in overall envelope
performance and building energy demand. In
general, higher glazing ratios result in higher
heating and cooling demand, counteracting much
of the benef it of eff icient surface-to-volume
ratios (Miller, 2013). During cold weather, heating
demand increases with increased glazing due to
loss of heat through the less thermally resistive
glazing components of the building envelope.
However, increased glazing allows for passive solar
heating opportunities, where heating demand
is decreased as passive zones (areas adjacent to
glazing) experience heat gain from sunlight. The
loss of heat, however, generally outweighs passive
heating effects during cold weather. In contrast,
cooling demand primarily increases with increased
glazing due to the increase in solar gains during
warm weather (Miller, 2013). Buildings with
excessive glazing ratios and untreated faades
(symptomatic of high-rise buildings) make them
particularly vulnerable to overheating during
the summer and to heat losses during the winter
(Salat, 2009).
A buildings orientation, and the distribution of
glazing, relative to the suns position, also impacts
a buildings opportunities for passive heating. For
space heating, glazing oriented to the east and
south (for buildings in the northern hemisphere)
provides the most useful gains. In comparison,
western orientations provide solar gains late in
P L A N 4 D E | 19
the day when temperatures are at a maximum and
buildings are already heated (Baker & Steemers,
2000).
4. 3. 4 T R A D E O F F S AT T H E B U I L D I N G S C A L E
While increasing compactness and building
density generally reduces heat loss, increasing
these characteristics does not guarantee overall
building energy reductions. The trend for
increased glazing as building density increases
counteracts some of the thermal performance
benef its of more eff icient surface-to-volume
ratios and shared walls and f loors, increasing
energy demand (Miller, 2013). Minimizing heat
losses during the winter requires minimization
of the surface-to-volume ratio; but this implies
a reduction of the building envelope exposed
to the outside environment, thus reducing the
availability of daylight and sunlight and increasing
energy consumption for artif icial lighting and
natural ventilation (Ratti, Raydan, & Steemers,
2003).
The relationships between compactness,
building density and building energy are further
complicated by differences between residential
and commercial uses (Miller, 2013). For residential,
increasing building density from detached housing
to apartments can reduce heating energy demand.
However, such density increases typically require
increased building depth, increased building
height or reduced building spacing that limit
access to passive heating and daylighting. In
certain contexts, low-density residential design
with high surface-to-volume ratios maximized
for passive solar heating may provide greater
opportunities than density for reduced energy
demand (Steemers, 2003). For commercial
buildings, increasing the building depth of off ices
reduces the availability of natural ventilation
and daylight, resulting in an anticipated increase
in mechanical ventilation and artif icial lighting.
However, heat losses are likely to decrease as the
surface-to-volume ratio decreases with increasing
plan depths (Steemers, 2003). In contrast,
however, commercial buildings with high cooling
 THE BUILT

ENVIRONMENT &

ENERGY DENSITY

demands may benef it from higher surface-to-
volume ratios, which can assist in reducing cooling
demand (Steemers, 2003).
Notwithstanding the complexity discussed above,
Miller (2013) notes that the impacts of building
compactness, density and glazing ratios on heating
demand are reduced as envelope performance
improves. This suggests that at the building
scale, there should be a greater focus on envelope
performance than on other factors. Steemers
(2003) notes that the arguments for and against
density for building design are f inely balanced
and dependent on local contexts.
4.4 B U I L D I N G
SYSTEMS
The operational energy consumed in buildings is
predominantly used in space heating and cooling,
and in running appliances, including heating hot
water and lighting (Doherty et al., 2009). Energy
consumption changes for different building uses
or activities. In general, residential dwellings are
dominated by space heating, while commercial
buildings have signif icantly larger requirements
for space cooling and lighting.
Commercial buildings also consume more energy
per square metre than both residential and
industrial land uses (Steemers, 2003). Differences
in space heating and cooling requirements vary
climatically, particularly latitudinally (Doherty
et al., 2009). As a general rule, the warmer the
climate, the lower the operational energy, because
of the reduction in heating energy required
(Newton et al., 2000). Heating is the major driver
of energy consumption, ahead of cooling and
refrigeration (Doherty et al., 2009).
Different building-use activities require
different types of heating and cooling systems
and equipment, lighting and hot water systems,
with large variations in capacity and eff iciency.
In general, as systems and equipment eff iciency
decreases, energy demand increases. Also,
buildings containing different activities do not
use energy at the same rate and at the same time
of day. In general, commercial and industrial uses
consume more energy during working hours,
whereas residential energy consumption peaks in
the morning and evening, before and after regular
working hours, respectively.

Figure 2: End-use of energy in buildings (OECD/IEA, 2013)

Plan4DE
 THE BUILT

ENVIRONMENT &

ENERGY DENSITY

their respective consumption are more inclined to

4.5 OCC U PA N T save energy, especially where energy is expensive,
B E H AV I O U R compared to occupants who consume energy from
collective systems.
Occupant behaviour affects building energy use
directly and indirectly; opening/closing windows, In a study completed by Hong & Lin (2013),
turning on/off or dimming lights, turning on/ results showed that occupants of private off ices
off equipment and electronics, turning on/off who are proactive in saving energy consume up to
heating, ventilation and air-conditioning (HVAC) 50% less energy, while occupants who are wasteful,
systems, and setting indoor thermal, acoustic consuming energy at will and lacking motivation
and visual comfort criteria (Hong, 2014). Grey to reduce energy use, consume up to 90% more
and Gleeson (2007) discuss occupant behaviour energy.
factors that tend to increase energy demand. These
include: increasing aff luence or household income
(related to larger residential homes and increasing
4 . 6 T H E B L OC K &
area per person), uptake of more electronic and N E I G H B O U R H OO D
consumer goods due to declining prices, and,
where applicable, low energy prices. SCALE
As discussed earlier, arguments for and against
Hong and Lin (2013) note that occupant
density, balancing the need to reduce heat loss
behaviour is one of the most signif icant sources
with the need for solar access (resulting in passive
of uncertainty in the prediction of building
heating) are complex and affected not only by
energy use by simulation programs due to the
factors of building design and use, but also local
complexity and inherent uncertainty of occupant
context.
behaviour. Ratti et al. (2005) argue that occupant
behaviour and its effect on energy performance
At the block or neighbourhood scale, building
can vary by a factor of 2 between buildings, even
energy consumption is inf luenced primarily by the
among buildings with comparatively similar
shading of adjacent buildings, which limits solar
functions. They theorize that occupants could be
access. This substantially reduces opportunities
more likely to adopt energy eff icient behaviour
for passive solar heating and natural daylight,
if they live in an energy eff icient house and are
increasing overall heating and lighting demand
aware of its eff iciency measures. Similarly, Baker
(Miller, 2013; Ratti et al., 2005; Steemers, 2003).
and Steemers (2000) argue that good urban form
These same shading effects, however, also reduce
design may result in better occupant behaviour.
overall cooling demand (Miller, 2013).
In contrast, Salat (2009) suggests that if occupants
are aware that the building they occupy is very
Solar access is inf luenced primarily by building
well insulated, they might feel less incentivized to
spacing and arrangements, which are dependent
monitor their heating consumption.
upon a number of urban pattern elements such
as street widths, parcel size, building setbacks,
Research conducted by Salat (2009) shows that
building heights, and the distribution of open
for buildings heated by electricity, occupants
spaces. A measure of building spacing and
consume only half their theoretical heating needs,
arrangement (used in research by Steemers, 2003;
whereas occupants who receive heat through
Ratti et al., 2005; and Miller, 2013) is urban
district energy generally consume 30% more than
horizon angle (UHA). The UHA is the average
their theoretical needs. Salat (2009) concludes that
angle of elevation of surrounding buildings from
occupants who receive individual pricing based on
the centre of a given faade, and is affected by

PLAN4DE | 21
 THE BUILT
EN V I R ONMENT &
ENE R GY DEN s i TY
the height of and distance between structures.
The UHA accounts for building shading and
signif icantly affects solar access, particularly in the
winter when the sun is lower in the sky (Baker and
Steemers, 2000).
The predominant (yet limited) literature reviewed
in this area is by Steemers (2003) and Miller
(2013) who conducted research in the UK and
Vancouver, BC, respectively. It is important to
note that their f indings are context and climate
specif ic. Steemers (2003) shows that the UHA can
affect heating energy demand by as much as 30%,
cooling demand by 20%, and lighting demand
by 150%, with the greatest impacts occurring
for south-facing faades. Miller (2013) found
that increased UHA results in increased heating
demands and decreased cooling demands, higher
density urban patterns tend to require less heating
energy and more cooling energy per capita and per
unit of building area, and residential buildings
will experience larger increases in heating energy
demand and smaller improvements in cooling
demand with high UHA conditions.
Millers overarching f inding is that, for the urban
patterns and development context chosen for
analysis, there is relatively little variation in energy
performance among patterns of similar density
but distinctly different urban forms, suggesting
that urban form may not have the potential
for the factor of 2 impact on building energy
suggested by Ratti et al. (2005). The type and mix
of building activities included in buildings, for
otherwise similar urban forms and patterns, play a
larger role in heating and cooling demand.
Millers research supports the general assertion
of urban-form building-energy literature, that
higher-density development provides the benef it
of reduced heating demand through more compact
building shapes and more eff icient use of f loor
area (e.g. smaller residential units). However,
this research also identif ies several key urban
form factors tied to increased density that work
against heating demand benef its, such as high
Plan4DE
glazing ratios, high-rise buildings with lower
envelope performance requirements, and increased
shading effects due to high UHAs. Particularly
when it comes to local shading, it is important
to understand that there will be trade-offs;
implementing high-density development may
decrease heating demand (per area or per capita)
for the new building, in comparison with a lower-
density development. But it may also increase
shading on adjacent buildings, causing increased
heating demand. At the block or neighbourhood
scale, positive impacts for one building may have
negative impacts for another.
4. 6. 1 Ur b a n h e at i s l a n d e f f e c t a n d
v e g e tat i o n
Urban areas in many parts of the world are
subject to the urban heat island (UHI) effect.
Increased densities are generally accompanied by
increases in built materials and paved areas, and
decreases in natural land cover (e.g. vegetation
and bare soil). These conditions exacerbate UHI,
resulting in higher temperatures in denser city
centres, which lead to increased cooling demand
(Doherty et al., 2009). Miller (2013) argues that
UHI increases cooling demand, but decreases
heating demand, resulting in a net reduction in
energy consumption. Doherty et al. (2009) discuss
further how vegetation can play a signif icant role
in regulating the urban microclimate and can
inf luence operational energy demand through
solar absorption and the cooling effects provided
by shade and evapotranspiration.
4. 6. 2 R e n e wa b l e e n e r g y p ot e n t i a l
Research by Sarralde et al. (2011) explores renewable
energy potential (REP), that is the potential to install
several renewable energy technologies (RET) in existing
neighbourhoods and cities. The RETs reviewed include
Diffuse RET those technologies that once installed
will not block any urban space that could be used
for other purposes (e.g. buildings or infrastructure)
and that will not require any further material f lows
for their functioning, including photovoltaic panels
(PV) installed on roofs or faades, or meso-scale
 THE BUILT
ENVIRONMENT &
ENERGY DENSITY
wind turbines installed on rooftops; and Concrete
RET those options that use extra urban space and/
or generate material f lows, such as biomass-fuelled
combined heat and power (CHP) systems or macro-
scale wind turbines, which might require a non-built
safety area around them.
Sarralde et al. (2011) concluded that lower
building density is benef icial for REP in existing
neighbourhoods and cities. Higher building density
means that fewer portions of land are freely available
for installing RET; however, this may not necessarily
mean that the f inal renewable energy output will be
lower, since it will depend on what types of RET can
be installed. While high density can be detrimental for
some types of RET, it can be benef icial for others. For
example, a high-density area might not be optimal for
installing wind turbines or domestic shallow ground
source heat pumps (GSHPs), but it can be very good
for the installation of building-mounted PV, due to a
larger proportion of envelope area.
In subsequent research, Sarralde, Quinn, Wiesmann,
& Steemers (2015) show that while there is more room
for improvement in increasing the amount of solar
radiation received on faades (by modifying variables
such as plot ratio, distance between buildings and
building heights), this is still a small portion of what
can be achieved on roofs. Miller (2013) explores the role
that roof shape plays in the solar potential of urban
patterns (i.e. local energy generation potential), and
concludes that f lat roofs provide the greatest benef it
for solar potential due to their reduced dependence on
building and roof orientation. Where building height
varies (generally associated with higher densities),
reduced solar potential occurs for shorter buildings
within the urban structure, while taller buildings
maintain maximum solar potential due to limited
shading on roof surfaces. This effect can, however,
be minimized if building heights are kept relatively
consistent.
PLAN4DE | 23
4.7 FUTURE
BUILDINGS
The introduction of new policy measures to
increase the energy eff iciency of buildings as well
as the adoption of voluntary standards such as
PassivHaus will likely result in signif icantly lower
energy density, and indeed major eff iciency gains
will be required to limit climate change.
As of 2020 in Europe, nearly zero-energy
buildings are assumed to be the norm due to
the Energy Performance of Buildings Directive
(Vandevyvere & Stremke, 2012a). An IEA report
documents low carbon or zero emissions building
policies in ten countries, with targets between
2020 and 2050 (OECD/IEA, 2013).
In response to a saturation of district energy in
countries where district energy already connects
the buildings with higher heat density, a major
study (Zinko et al., 2008) identif ied a wide range
of strategies to reduce the cost of district heating
to increase the feasibility of low heat density areas.
These strategies sought to reduce investment
and heat losses, for example, by reducing the
pipe dimensions, increasing insulation with new
piping systems, decreasing the size of the trench,
integrating the install with other infrastructure
projects, identifying new energy services such
as absorption chillers, washing machines,
dishwashers and dryers, the notion of branching
out beyond the provision of energy for heating
(Spth, 2005). There is also an opportunity for
individual buildings to become energy generators
and traders as part of a district energy network
(Nystedt, Shemeikka, & Klobut, 2006).
District energy systems can create perverse
incentives that result in the avoidance of
successful energy savings strategies, as well as
legal requirements that can inhibit more eff icient
dwellings or buildings another form of path
dependence. One option is to design a tariff
structure that ref lects ecological impact, analogous
 THE BUILT

ENVIRONMENT &

ENERGY DENSITY

to a carbon tax so that low-energy dwellings are

not penalized (Spth, 2005). The development
of highly eff icient buildings increases the focus
on urban planning to maximize the viability of
district energy.

Finally, in addition to the decreased heating and

cooling demand due to design eff iciencies, there
is also the impact of climate change on heating
degree days and cooling degree days, both of
which drive energy consumption (Aebischer,
Catenazzi, & Jakob, 2007; Christenson, Manz, &
Gyalistras, 2006).

4.8 F I N D I N G S F O R
PLAN4DE
The viability of district energy systems relies
in part on the total energy demand for the area
to be serviced the general approach being to
maximize the total energy demand. This total is
calculated by multiplying energy intensity factors
for different types of building use (expressed as
GJ/m/yr) by the total f loor area of that use (a
function of building or development density). It
follows therefore that maximizing total energy use
involves maximizing the energy intensity factor,
the density, or both.

In general, as density increases (and building

characteristics improve), energy intensity factors
improve (i.e. decrease), which counteracts the
intent to establish a maximized total energy
demand. There are also a wide range of factors
that inf luence energy requirements, including
shape, size, occupant behavior, location, building
mix and others. Plan4DE needs to provide insight
on these factors to facilitate the optimal built
environment given broader societal parameters
such as existing built environment conditions,
demographics and economic context.

Plan4DE
 THE DISCIPLINE OF PLANNING

Urban design, including the clustering of buildings

and mixing of different building types within a given
area greatly affect the opportunities for and cost of
district heating and cooling systems (Levine & rge-Vorsatz,
2007).

5 . T H E DIS CIPL IN E OF PL AN N IN G
A key task in designing a model for city planners as politics and economics may have a strong
is thoroughly understanding the role of urban inf luence on zoning control and patterns, and
planning as it relates to district energy. Planning therefore the resultant urban form.
is empowered by a legal and regulatory framework
and is implemented through an administrative
system, both of which vary considerably from
5. 1 P L A N N I N G
country to country (Newman & Thornley, & ENERGY
1996). In general, planners have inf luence on
urban form through: the direct administration CO N S U M P T I O N
or implementation of zoning control in
A buildings energy performance is inf luenced by,
existing areas; the changing of zoning control
and under the control of, different professional
(through amendments) in existing areas; and the
actors in the urban context, at building and
development of new zoning controls alongside
neighbourhood scale, namely: planners with
urban design and street pattern, for new urban
respect to urban structure, architects/designers
developments.
with respect to urban and building morphology,
systems engineers with respect to building
In these settings, the planners role is either as
systems, and occupants/building owners with
administrator of an existing control framework, or
respect to occupant behaviour. The division
developer and/or proponent of new or different
between the private and public sector in the
control framework. Generally speaking, major
actual construction of the built environment
inf luence or change to urban form comes through
varies by jurisdiction, but local governments
the development of a new set of principles or
have a signif icant inf luence, as is illustrated in
control that is then followed by development.
the case studies and policy catalogue developed
In this instance, planners tend to play either
in IEAs Annex 51 (Strasser et al., 2011). Local
an expert role (making recommendations on
governments can carry their policies relating to the
appropriate changes) or a facilitation role
urban structure through to building morphology,
(consolidating widespread recommendations,
building systems and even to occupant behaviour.
achieved through engagement, on what should
be changed). These recommendations and expert
As discussed above, planners have the most
opinions are often grounded in the contextual
inf luence and control over urban structure, form
planning approach. It is very seldom, however,
and density. However, through zoning control,
that planners have ultimate decision-making
some aspects of planning can also inf luence
authority, as this is typically retained at the
building morphology, such as height (Wende,
political level. In the political realm such factors

PLAN4DE | 25
 THE DISCIPLINE OF PLANNING
Huelsmann, Marty, Penn-Bressel, & Bobylev,
2010). At the building scale, architects and
engineers are bound to not only follow zoning
and bylaw requirements, but also building-code
requirements. Building code can (and often
does) have an inf luence on energy consumption,
particularly as it relates to envelope performance.
Another way to conceptualize the different
spheres of planning operations refers to
geographic scale (Vandevyvere & Stremke, 2012b).
Interventions at the building level are the micro
scale, interventions in the larger scale energy
infrastructure occur at the macro scale, whereas
the discipline of planning focuses primarily, but
not exclusively, on the built environment, which
could be considered the meso scale.
When considering energy consumption in
the context of urban form, zoning control,
urban pattern and building-code regulations
need to work together (van den Dobbelsteen,
2007). Dwelling density can increase as a result
of deliberate policies under a compact-city
philosophy. However, if building code and energy
performance standards are not improved, overall
energy eff iciency may not be achieved.
Urban form refers to the spatial and physical
pattern or structure of developed areas, and the
distribution and relationship of uses within
this structure. Urban patterns are made up of
a number of elements (buildings, parcels, open
spaces and streets) that are linked by a consistent,
repeated set of systematic relationships (Miller,
2013). Variations to these elements and their
functions or uses generally occur within an
organized set of principles or control, which
ref lect the development approach (Table 1).
Urban forms, and approaches to urban form,
vary depending on context, across different
jurisdictions, countries, and climates, due
to a number of elements. These include
differences in architectural style and building
materials, planning approaches, policy, codes
Plan4DE
and regulations, local and/or federal politics,
economics, and approaches to environmental
sustainability.
5. 2 P L A N N I N G &
DISTRICT ENERGY
Planners and other local authorities have a number
of tools to encourage district energy development
and connection, using bylaws and local building
codes to ensure building design and systems are
compatible with district energy.
Green building or neighbourhood standards
can be set to establish various targets for energy
eff iciency. For example, a standard that sets an
electricity conservation target will naturally
discourage the use of electric heaters, and a
renewable energy procurement target will
naturally discourage the use of fossil-based fuels
(King & Bradford, 2013).
The ability of local government to encourage
or, in some cases, require district energy ready
(DE-R) buildings varies considerably across
jurisdictions. In parts of Europe and Asia (e.g.
Denmark, South Korea), local governments can
mandate energy utility services across districts.
The Canadian context provides both a legal
framework and environmental values that help
support the development of district energy. The
City of North Vancouver in British Columbia
established a Hydronic Heat Energy Service
Bylaw to create a district heating service area for
the Lower Lonsdale neighbourhood. The bylaw
requires all new or retrof itted buildings greater
than 1,000 square meters be connected to and use
the district energy system.
 THE DISCIPLINE OF PLANNING

Table 1: Description of common planning strategies that influence urban form

P L A N N I N G S T R AT E G Y DESCRIPTION

Zoning Control What gets built on a parcel of land is generally determined by the zoning control and other
regulatory requirements that apply to that land.

Land Use Residential, commercial, industrial, institutional, recreational, open space, agricultural, right
of ways (roads, rail, utilities).

Density Number of residents, households, employees, units, or building area per total parcels (e.g.
f loor space ratio (FSR), dwelling units per area).

Setbacks (which can also be Inf luences the distance, and therefore spacing pattern, between buildings and between
a function of density) buildings and streets.

Building typology (which E.g. detached houses, semi-detached houses, apartment blocks.
can also be a function of
density)

Building geometry Vertical and horizontal distribution height, volume, perimeter distance.

Building character Architectural look and feel in certain contexts, this can be prescribed through control,
in other circumstances, designers are given more artistic freedom to create individualistic
designs.

Urban geometry How parcels of land f it together within a developed area is determined by street pattern and
block sizing.

Street patterns Street patterns make up the spatial structure of urban areas. This includes the street width
and geometric layout.

Block sizing Block sizes determine the length of a block (or collective parcels) of land between intersecting
streets, and therefore, a corresponding length in street. No one street pattern is associated
with a particular block size block sizes can vary within the same street pattern. For example,
a traditional grid street pattern may have 100m x 50m blocks, or 100m x 100m.

5.3 F I N D I N G S F O R
PLAN4DE
Although planners do not develop district energy,
they have a wide range of tools that can either
enable or inhibit district energy through their
inf luence on the built environment. Planners
increasingly have a mandate to incorporate
strategies to reduce energy and emissions of which
district energy is one.

PLAN4DE | 27
 MODELLING THE

BUILT ENVIRONMENT

A model is not a piece of absolute and universal

science but the expression of a point of view that will
take place in a social context where it will more or less
be relevant (Bouleau, 1999, p. 340).

6. MODELLING THE BUILT ENVIRONMENT

In order to understand relevant modelling
approaches and strategies, a scan of the existing
modelling ecosystem was undertaken, focusing on
the built environment, urban energy systems and
district energy.
checking that the ex-ante reasoning makes sense,
testing theory and method, and exploring design
alternatives and design decisions (Devisme, Bosse,
Dumont, & Ouvrard, 2011).

What is a model? In this project, a model

6 . 1 C L A S S I F I C AT I O N
is def ined as a conceptual abstraction of an OF MODEL
existing or proposed real system that captures
characteristics of interest, consisting at its essence A P P R OAC H E S
of inputs, calculations and outputs. There are two
Modelling of urban systems has been evolving
streams of modelling of interest to this project:
rapidly, driven by urban theory and computing
urban energy modelling and district energy
abilities, but the history of their use is relatively
modelling. There is overlap across these two
short, spanning f ifty to sixty years (Oryani
streams as those interested in district energy seek
& Harris, 1997). The earliest land-use models
to characterize the heating and/or cooling demand
attracted signif icant interest and effort as a
of the built environment to investigate the
strategy to understand and manage the rapid
feasibility of district energy, and those interested
growth in automobile use. Douglass Lee (1973)
in energy f lows within the built environment also
published a watershed paper in 1973 titled
seek to understand heating within buildings.
Requiem for large-scale models in which he
identif ied major weaknesses. Following this paper,
A model, if designed to address the capabilities
interest waned and the focus shifted to short-term
and context of a user group, in this case city
problem solving for more than a decade (Pinto
planners, can facilitate an enhanced understanding
& Antunes, 2007). The emergence of GIS in the
of cause and effect and therefore improved policy
1980s increased computer power, and increased
outcomes that support energy eff iciency and
data availability reinvigorated the relevance
greenhouse gas emissions. A workshop under
of models as signif icant tools for urban policy
the auspices of the Organisation for Economic
decision-making (L. Li, Sato, & Zhu, 2003).
Co-operation and Development (OECD) that
explored the development and state of urban
Of the urban systems that can be modelled,
models globally, identif ied motivations for using
transportation has attracted the most interest,
models as: exploring and discussing policy scenario
evolving from the classic four-stage model which
development and alternative futures, informing
used a sequential analysis of trip generation, trip
policy decisions, convincing professionals,

Plan4DE
 MODELLING THE
BUILT ENVIRONMENT
distribution, modal split and trip assignment to
generate traff ic f lows to cellular automata, which
simulate individual or household behaviour
(Kane & Behrens, 2002) and employs complexity
theory to generate unpredictable outcomes. In
addition to these resource-intensive models, a
new generation of sketch-planning-type models
has been developed which range from full-scale
simulation to diagnostic-indicator-like approaches
(Batty, 2004). The overall result of the various
streams of urban systems model development over
the past forty years is a very diverse ecosystem
of different approaches that are diff icult to
categorize or standardize.
The most critical question in terms of model
structure is the question of what is generated
within the model (endogenous) and what the
inputs from other sources (exogenous) are.
Any agent or process whose attributes and/or
behaviour are determined within the model is
considered endogenous to the model, and factors
which affect system performance, but whose
values are provided as inputs, are called exogenous
factors. The boundary of the model is one factor
that determines which variables are inputs and
which variables are the subject of evaluation in
the model. Keirstead et al. (2012, p. 5) noted three
different boundaries of urban energy systems, each
with signif icant implications for the type of model
employed:
Pure geographic, i.e. the urban energy system
consists only of those technologies that lie
within a citys administrative boundaries;
Geographic-plus, i.e. everything within the
administrative boundaries plus easily traceable
upstream f lows, like electricity consumption;
Pure consumption, i.e. the energy system
encompasses all energy activities of a citys
occupants wherever they occur. For example,
attributing a resorts energy consumption and
emissions to the home cities of the visiting
tourists.
PLAN4DE | 29
One framework evaluated nineteen models
against three axes representing temporal,
spatial and human decision-making complexity
(Agarwal, Green, & Grove, 2002). The authors
indicate that the relationship between human
and environmental systems contains many
nonlinearities and spatial and temporal lags which
are not captured by statistical techniques, creating
a signif icant challenge for modelling. Nonlinear
dynamic models respond to this challenge, using
stock variables (for example, capital stock in an
economy), f low variables (annual investment in
capital) and parameters to create a model that
is not dependent on historical data and that
generates emergent, unpredictable outcomes. It
is, however, still diff icult to tune these models to
local contexts (Batty, 2004). In another approach
to classif ication, Table 2 identif ies nine different
families of characteristics of energy models (Beeck,
1999).
Finally, a review of models for analyzing the
GHG impacts of land-use decisions classif ied
eleven models by scope, methodology, scale
and support for policy making (P. M. Condon,
Cavens, & Miller, n.d.). Methodological categories
included spatial/nonspatial, top-down/bottom-
up, simulation/end-state and observation-
based/process-based. Of these options for GHG
modelling, the authors indicated that a spatial
dimension is advantageous as it interprets the
physical organization of a city and more effectively
communicates the implications of decisions.
Simulation models generate outcomes based on
a series of rules, whereas the end-state method
describes a future condition and work backwards.
In terms of this classif ication schema, Plan4DE is
intended to be spatial in that considers the spatial
conf iguration of the built environment, bottom-
up in that it derives results based on individual
unit, or specif ic buildings and end-state oriented
in that it doesnt seeks to describe the evolution
of the built environment, but rather a specif ic
outcome.
 MODELLING THE

BUILT ENVIRONMENT

Table 2: Characteristics of energy models

A P P R OAC H C AT E G O R I E S

1. Purposes of energy models General: forecasting, exploring, backcasting

Specif ic: energy demand, energy supply, impacts, appraisal, integrated

approach, modular build-up

2. The model structure: internal assumptions Degree of endogenization, description of non-energy sectors, description
& external assumptions of end-uses, description of supply technologies (how many variables are
generated within the model vs. captured from other sources)

3. The analytical approach Top-down (i.e. deconstructs the economy to understand households)
or bottom-up (i.e. constructs a city transportation system based on
household behaviour)

4. The underlying methodology Econometric, macro-economic, economic equilibrium, optimization,

simulation, spreadsheet/toolbox, backcasting, multi-criteria.

5. The mathematical approach Linear programming, mixed-integer programming, non-linear

programming

6. Geographical coverage Global, regional, national, local, or project

7. Sectoral coverage Energy sectors or overall economy

8. The time horizon Short, medium, long term

9. Data requirements Qualitative, quantitative, monetary, aggregated, disaggregated

assess urban heating demand, employs LiDAR

6 .2 U R B A N E N E R G Y
MODELS
Urban energy models seek to track the spatial and
temporal variation in urban energy demand. In
general these models are developed for a particular
location or set of circumstances, although there
are more generic approaches emerging such as
SynCity (J. Keirstead, Samsatli, & Shah, 2010).
A sub-stream of this broader category seeks to
quantify GHG emissions associated with urban
areas in order to guide local governments in their
mitigation efforts (Devisme et al., 2011). The
techniques and strategies employed to quantify
energy vary widely. One approach, seeking to
(airborne laser scanning) to identify three-
dimensional prof iles of buildings. A transmission
heat-loss method is then used to identify heat
demand (Strzalka, Bogdahn, & Eicker, 2010).
Another approach divides a city into typologies
of relatively homogeneous built environment
characteristics, assigns resource consumption to
each typology and calculates energy and other
resources used a top-down method (Juan Jos
Sarralde, Quinn, & Wiesmann, 2011). Similarly,
but more precisely, both the Energy and
Environmental Protection Model (Cf, 2001) and
the Urban Archetypes Project (Webster, 2008)
classify buildings into one of twenty types and
attach them to a particular spatial location to

Plan4DE
 MODELLING THE
BUILT ENVIRONMENT
enable calculation of building and transportation
energy consumption. Common to all of these
approaches is the use of GIS, essentially a
manipulatable database of spatial information.
But the rise of big data gives rise to a whole new
range of as yet untapped sources of data (Martino
et al., 2010.). For example, Googles NEST, a
learning thermostat, not only tracks energy
consumption online but also the behaviour of
residents over time (Barker, Mishra, Irwin, &
Cecchet, 2012).
One notable stream of effort within the urban
energy modelling world is that of urban
metabolism. The urban metabolism method seeks
to track the f lux of energy or carbon, and/or other
resources through the urban system including
changes in storage (Sahely, Dudding, & Kennedy,
2003), a complex but systematic approach that
includes embodied as well as operational energy.
In their literature review on the subject, Holmes
and Pincetl (2012), note that urban metabolism
provides a complete picture of actual and
embedded energy use over short- and long-term
time scales.
New strategies for validating models include
actual measurements of carbon f luxes, although
there are challenges extrapolating the accounting
framework against actual location-based emissions
(Christen et al., 1990).
6 .3 D I S T R I C T E N E R G Y
MODELS
Typically, all the studied models used for analyzing
the feasibility of district energy projects include
the following elements for the analyzed area:
Calculation of consumption of heating or
cooling for all customers;
Technical and economic calculations for
individual cooling and heating;
Technical and economic calculations for district
P L A N 4 D E | 31
energy cooling and heating, including network
costs;
Comparison of base case (individual heating or
cooling) with district energy.
This approach is shown in Figure 3 on the
following page.
6. 3. 1 A S S U M P T I O N S A N D T H E O R Y B E H I N D
T H E M O D E L L I N G A P P R OAC H E S
The most common methodology used for
comparing different individual solutions for
heating and cooling with district energy systems
is to make a calculation of the net present value
of the costs for the different scenarios over a
planning period (typically twenty years) and
compare the individual solution with district
energy (F. Li, 2013).
An overview of typical assumptions is shown in
Figure 4 on the following page. .
6. 3. 2 S PAT I A L A N D N O N - S PAT I A L
DIMENSIONS IN THE MODELS
None of the models studied has a focus on the
spatial dimensions of the area in the analysis. The
user typically specif ies the number of buildings,
the area of the buildings, the length of the needed
pipes in the district energy network, as well
as other related infrastructure. This approach
requires that the user have detailed knowledge of
district energy systems and the related technology
and cost data. However, in two of the models
discussed below (the COWI Heat Atlas and the
study using Balmorel) a GIS database has been
used as input for the district energy model to
calculate energy density and the rough layout of
the district heating network. A further model,
the District Energy Concept Advisor provides
guidance to urban planners and developers at
an early stage of the development of the built
environment, but it is not grounded in a specif ic
community (Erhorn-kluttig, Erhorn, Weber,
Wssner, & Budde, 2013).
 MODELLING THE

BUILT EN V I R ONMENT

technical
& e co n o m i c
c a l c uat i o n s
f o r i n d i v i d ua l
h e at i n g & coo l i n g

technical
definition of & e co n o m i c
analyzed area: co m pa r i s o n o f b a s e
h e at i n g & coo l i n g c a s e ( i n d i v i d ua l
n e e d s , g e og r a p h y , coo l i n g & h e at i n g )
energy density with a district
energy system
technical
& e co n o m i c
c a l c uat i o n s
for district
energy system:
production &
N e t w o rk co s t s
c a l c u l at i o n s
Figure 3: Modelling of District Energy

specifics inputs for data f o r

district energy area
Archetypes for building areas:
inputs on general
Number, size and types of
f r a m e w o rk
buildings -> energy density,
installation costs
Fuel prices, power prices,
taxes and subsidies Consumption prof ile (seasonal,
diurnal load variations
Lifetime and interest rate
for investments
Network costs & losses
Local sources for cooling or
heating: sea water, ground
water, geothermal energy,
industrial surplus heat, etc.
r e l e va n t e n e r g y
production units
- i n d i v i d ua l &
district energy
solutions
Size
Technical & economic
daat: eff iciencies,
variable & f ixed O&M
costs, investments, etc.
Emissions data
Technical constraints
Political constraints

Figure 4: Assumptions in district energy models.

Plan4DE
 MODELLING THE

BUILT ENVIRONMENT

6 .3 .3 L E V E L O F D E C I S I O N - M A K I N G T H AT T H E need to be optimized at the same time, taking

M O D E L S A R E D E S I G N E D TO I N F L U E N C E
The studied models are all designed to inf luence
decision-making at the level of politicians or
planners. However, most of the models require the
insight of technical staff to set up the necessary
inputs and interpret the calculations and results.
6 .3 .4 M E T H O D S U S E D F O R M O D E L L I N G
Four of the f ive studied models use spreadsheets
as the model platform. The mathematical model
structure is simple and employs basic theory for
energy balances and f inancial calculations. One
model is based on linear programming, using the
model language GAMS, but this model is only
necessary if a number of district energy areas
into account the dynamics of the regional power
system.
6. 3. 5 G R A P H I C A L U S E R I N T E R FAC E
The user interface of the models is typically made
up of tables and graphs in the spreadsheet, an
approach that is easily accessible for those familiar
with spreadsheets and f inancial analysis.
6.4 EXISTING MODELS
There are a large number of quantitative models
used for different aspects of the district energy
lifecycle. The models can be divided into different
types that serve different purposes, as illustrated
in Table 3.

MODEL TYPE DESCRIPTION EXAMPLES

Hydraulic models Models of district heating networks, pressure losses, f lows and pumping Systemrrnet, Termis

Energy system models
Models of load dispatch
in single district heating
systems
Costing models
Models that explore the role of district heating in the energy system. Balmorel, EnergyPlan,
Simple representation of district heating networks (demand, production, Times, SUNTool
fuel use).
Models that calculate hour by hour dispatch of production units and heat EnergyPRO
storage.
Models for calculating costs of district energy systems, dimensioning and RETScreen
layout of pipes. May include GIS applications.

Decision models for
district energy projects
Urban planning mode
Models that evaluate the feasibility of district energy taking into account Rambll model
investments in district heating network, heat/cooling production price developed for the
and energy density. Some models use GIS tools. Danish Energy Agency,
COWI Heat Atlas
(combination of
models), RETScreen,
DEEM, Balmorel.
Decision models for district energy also including planning of urban None identif ied. This
structure to f it with district energy systems is the target use case for
Plan4DE.

Table 3: Model characteristics

P L A N 4 D E | 33
 MODELLING THE
BUILT ENVIRONMENT
6 .5 R E V I E W O F
MODELS
Models used to evaluate the feasibility of district
energy are used by most consultants and some
industrial companies working in the f ield of
district energy. Many of these are proprietary
models, which means that developing an
exhaustive list is diff icult. The models studied
include the following:
Rambll
COWI Heat Atlas
RETScreen
DEEM
Balmorel
District Energy Concept Advisor
One model that was not studied but is notable is
the SUNtool, a highly technical simulation model
that includes urban planning dimensions and
district energy (Robinson et al., 2007). However,
it does not appear to be currently available.
6. 5 . 1 R A M B L L M O D E L
In recent years many municipalities in Denmark
have started working more systematically with
cross-cutting strategic energy planning including
energy eff iciency measures, heat planning,
transportation and other variables. To support
this work the Danish Energy Agency has developed
guidelines and tools to be used in planning. One
of these tools is an Excel model for evaluating the
feasibility of district heating in new developments
in towns and cities. The model includes an input
sheet, an output sheet and a number of calculation
sheets. The inputs from the user include: number
and area of different types of buildings (i.e. single
family, multi-storey, off ices), energy standard
Plan4DE
of buildings, distance to nearest district heating
network, costs and environmental impact of
nearest district heating network (choice between
3 levels of costs), and specif ication of individual
heating alternatives.
Based on these inputs the model calculates
the costs of the individual and district heating
alternatives and makes a comparison of the two.
The model includes both a calculation of the
consequences for the users (including taxes and
subsidies) and a socio-economic calculation.
6 .5 .1 COW I VA R M E AT L A S
In 20132014, the Danish consultancy companies
COWI and Ea Energy Analyses carried out an
analysis of the future for district heating in
Denmark for the Danish Energy Agency. This task
included an analysis of the optimal split between
district heating and individual heating in Danish
buildings. Today district heating has a share of
about 50%, but the analysis showed that it is
economically feasible to increase this to about 60
%. As part of the project an Excel-based model was
developed to analyze the competition between the
individual and district heating alternatives. The
model includes data for all buildings in Denmark
based on the Danish National Building and
Houses Database (BBR) and COWIs existing GIS
tools. The buildings were grouped in about 4,000
town areas with data for energy consumption of
the buildings, energy density and distance to the
nearest existing district heating network.
Based on assumptions for fuel prices, taxes and
subsidies, CO2 prices, power prices and technical
and economic data for heating technologies, the
costs for individual heating of the buildings were
calculated. Costs for the heat production in the
district heating networks were computed with
the energy market model Balmorel and fed into
the Heat Atlas model. It was also possible to
calculate the district heating costs for alternative
land uses and use these results as inputs into the
model. Finally, the model included inputs and
calculations of the costs for connecting buildings
 MODELLING THE
BUILT ENVIRONMENT
to a district heating network dependent on the
energy density and the distance of the buildings to
existing district heating networks.
The results of the model include the costs
for individual heating systems and district
heating for all existing buildings in Denmark
and a comparison between the costs of the two
alternatives. Based on these results the optimal
split between individual and district heating for a
particular jurisdiction can be evaluated depending
on the input assumptions.
6 .5 .3 R E T S C R E E N
RETScreen is a software system used to evaluate
energy eff iciency, renewable energy and
cogeneration project feasibility, and ongoing
energy performance. The RETScreen software is a
unique decision support tool developed with the
contribution of numerous international experts
from government, industry and academia, and
is provided free of charge for worldwide use.
RETScreen evaluates energy production and
savings, costs, emission reductions, f inancial
viability, and risk for various types of renewable-
energy and energy-eff icient technologies (RETs).
The software (available in multiple languages)
also includes product, project, hydrology and
climate databases, a detailed user manual, and a
case studybased college/university-level training
course, including an engineering e-textbook.
The software has quite a broad scope and the
analysis of district energy systems is one of a
number of applications of the model. It is a
spreadsheet model where the user can choose the
type of project analyzed, i.e. energy eff iciency,
power production, heating, combined heat and
power, and others. The possible project types
include district energy systems and the user
needs to specify inputs including the number,
area and energy consumption of buildings,
the required temperature, the pipe size and
length of the district heating network, as well as
detailed technical and economic data for the heat
production technologies in the base case and in
P L A N 4 D E | 35
the district energy case. The outputs from the
model include a cost analysis, a f inancial analysis
and an emissions analysis in which the district
energy system is compared to the base case.
6. 5 . 4 D E E M
The District Energy Economic Model (DEEM)
is a spreadsheet tool designed specif ically
to help developers, engineers and municipal
planners forecast how the benef its of a district
energy system translate into revenue, savings
and jobs for a specif ic community. The tool has
been developed by Natural Resources Canadas
CanmetENERGY and the focus of the tool is
district energy in Canada.
The model includes f ive input sheets where the
user specif ies data for a business as usual case
(individual heating or cooling) and the district
energy case. The required inputs are detailed
and include data for the number and energy
consumption of buildings, and economic and
technical data for the two cases. The model then
calculates the costs of the two cases and completes
a comparison. The model includes a number
of alternative perspectives and calculations to
show the range of economic benef its of district
energy including wages and salary earnings, gross
domestic product, taxes, jobs and environmental
impacts.
6.5.5 Balmorel
The Balmorel model simulates investment and
operations of a combined electricity and CHP
system from an international perspective. It
permits integrated analyses of demand and supply
of electricity and heat while balancing operations
against investment, local electricity generation
against import/export of electricity, and price
elastic demand against generation and investment.
It represents demand of electricity and heat in
a number of geographical locations, each with
individual time variations over the year.
Key inputs are existing electricity and heat
 MODELLING THE

BUILT ENVIRONMENT

generation capacities, electricity transmission possibilities for building service systems (space
capacities and demand. Additional key heating, hot water generation, ventilation,
assumptions relate to fuel prices, CO2 costs, cooling) and the integration of renewable energy
taxes and support schemes. The main outputs sources (Erhorn-kluttig et al., 2013).
are consumption, production, electricity
transmission, emissions and other characteristics
of the energy system.
6 . 6 CO M PA R I S O N O F
THE MODELS
The model is formulated as a deterministic linear
optimization model in the General Algebraic Table 4 provides an overview of a comparison of
Modeling System (GAMS) modelling language. the models.
This permits changes and adaptations of the
model to specif ic applications that are not
presently covered by the standard version of the
model. The models GAMS code is available under
open-source conditions at www.Balmorel.com.

In the study The role of district heating in the

future Danish energy system (Mnster et al.,
2012) the Balmorel model was adapted in order to
analyze the competition between district heating
and individual heating of buildings. In this version
of Balmorel, approximately 400 Danish district
heating areas were grouped into 25 aggregated
areas. For each of these areas the potential for
expanding the district heating networks into areas
with individual heating and the related costs of
the network expansions was calculated using a
heat atlas from Aalborg University. Based on this
input the Balmorel model was able to optimize the
expansion of district heating based on the costs
for individual heating, and the costs of district
heating production and network expansion.

6 .5 .6 D I S T R I C T E N E R G Y CO N C E P T A D V I S O R

The District Energy Concept Advisor was

developed as part of International Energy Agency
Annex 51 Energy Eff icient Communities. The
program is designed for urban planners, housing
companies, developers and local political decision
makers in the very early planning stages of the
energy supply of neighbourhoods. The tool is
populated with national building archetypes with
a f ixed geometry, a usage prof ile and U-values of
the building envelope (related to the age of the
building). The user can explore a wide range of

Plan4DE
 MODELLING THE

BUILT ENVIRONMENT

Table 4: Comparison of models reviewed

MODEL C H A R AC T E R I S T I C S CO M P L E X I T Y OPEN
SOURCE OR
P R O P R I E TA R Y
Rambll Spreadsheet model. Simple input structure. Low Open source

COWI Varmeatlas Spreadsheet model. Not very user-friendly. Medium Proprietary

The strength of the model is that it is able to
analyze a number of district energy systems at
the same time. It is not very detailed on the
level of each individual DH system.
RETScreen Spreadsheet model. The model includes the Medium Open source
possibility to analyze a number of different
renewable-energy and energy-eff icient
technologies (RETs) and district energy is
just one. The many details needed in the
input structure set high demands for the
technical insight of the user.
DEEM Spreadsheet model. Many details needed Low Open source
in input. Includes a broader view on the
benef its of district energy and includes
economic benef its such as GDP, taxes, jobs,
etc.
Balmorel GAMS model. Complex input and model High Open source
structure. Not suitable for analyzing
district energy in a single area but useful for
analyzing the role of district energy in the
energy system of a country or region.
District Energy Users can choose from databases of building Medium Proprietary
Concept Advisor types and customize the buildings, select
technologies and evaluate different options
using an interface.

P L A N 4 D E | 37
 MODELLING THE
BUILT ENVIRONMENT
6 .7 O P E N S O U R C E
A number of the models reviewed are open source.
Open source referred initially to a way of writing
and distributing software in which it was given
away for free, in direct contrast to the traditional
approach in which software is developed as a trade
secret and sold (Healy & Schussman, 2003). The
formal def inition requires that the source code
for the program must be available at little or no
charge, redistribution of the program, in source
code or other form, must be allowed without fee,
distributions of modif ied software must be allowed
without discrimination, and the distributions of
those modif ications must be permitted on the same
terms as the original program (Lerner & Tirole,
2005).
6 .7.1 B E N E F I T S A N D C H A L L E N G E S
Successful open-source projects rely on a
community of users who are willing to innovate
and share their innovations so that the software is
enhanced based on the users direct experiences.
Participants in an open-source process may
accrue social benef its in contrast to atomized
agents relying on price signals to make decisions,
community participants should be well-connected
(pervasively networked) and information should
f low well between them, as well as the practical
benef its of having software that works properly,
the increase in reputation that comes from being
associated with a successful project, and the
potential [for] further commercial opportunities
(Healy & Schussman, 2003, pp. 67). In their
analysis, Healy & Schussman conclude that the
more successful a project, the more professional its
core contributors will be, that a strong hierarchy is
necessary and that successful projects will have core
participants mobilized in a very similar manner to
core participants in a social movement.
6 .7.2 E X A M P L E S
Many urban energy models use open-source licenses
including UrbanSim (Waddell, 2010), LEAP
(Heaps, Clark, & Howells, n.d.), SPRAC (York,
2012), Urban Footprint (Calthrope Associates, 2011)
and others.
Plan4DE
6.8 FINDINGS FOR
PLAN4DE
There are many examples of different modelling
strategies to support decision making related
to district energy. No models, however, were
identif ied that support planners in optimizing
the built environment for district energy. The
models vary in focus, complexity and accessibility
according to their positioning relative to their
stage in lifecycle of the development of district
energy. Plan4DE is uniquely positioned upstream
as an intervention in the development of the
built environment, prior to the articulation
of a particular district energy project. The
implication of this positioning is that Plan4DE
must have a bottom-up orientation, deriving
energy consumption from the characteristics or
properties of specif ic buildings by type; it must
be spatial in order to understand the effects of
different conf igurations of buildings on the costs
of providing district energy networks; it must be
able to compare different scenarios, specif ically a
district energy scenario and an individual heating
scenario; it must be able to identify or specify
an optimal district energy technology mix for a
particular location given a particular combination
of energy demands. Plan4DE must therefore
integrate considerations of the built environment,
energy demand, consumption and supply and
f inancial implications. Because it is future-
oriented, it is not as concerned with precision of
the assumption, but rather the relative impact
of different technology choices for for different
contexts on the viability of district energy for a
specif ic built environment. Plan4DE also needs to
enable easy adjust of built environment parameters,
in order that planners can easily test possible
futures.
A number of models are open source, a mechanism
to increase the accessibility to the model, the
transparency of its assumptions and to create
a pathway for ongoing evolution. Plan4DE
will benef it from these attributes, which are
conditional on the development of an active user
base.
 PLAN 4 DISTRICT ENERGY
7. P L A N4DE
The preceding sections have identif ied the role of
energy density amongst other factors that enable
or hinder district energy (Section 3), the aspects
of the built environment that inf luence energy
density (Section 4), the role of planners in shaping
the built environment (Section 5), and existing
tools or models that are used to explore different
aspects of district energy (Section 6). This section
describes the results that emerged from this
analysis and the integrated design process; the
objectives, structure and methods that constitute
Plan4DE.
Objectives were established for Plan4DE that
are broader than the technical task of developing
equations that illustrate the relationship between
the built environment and district energy; they
also address social determinants. Plan4DE was
designed to be:
Persuasive: Plan4DE clearly and simply
communicates to planners, engineers and
elected off icials the benef its of urban form
that support district energy, both in terms of
the benef its resulting from district energy and
co-benef its.
Cross-disciplinary: Plan4DE builds linkages
between disciplines including planners and
engineers, and planners and politicians,
by enabling and encouraging bridging
discussions. Because district energy is not the
traditional remit of planners, they will likely
consult or engage with engineers responsible
P L A N 4 D E | 39
for city infrastructure. Because planning
typically involves participatory processes and
engagement of elected off icials, a tool which
integrates district energy into planning will
likely prompt discussions between planners,
politicians and engineers.
Practical: Plan4DE enables land-use planners,
through iteration to create a built form that
is condusive district energy, using language,
concepts and practices that are directly
relevant and applicable.
Enabling: Plan4DE serves as a decision gate,
a binary tool that describes whether or
not district energy is viable for a specif ic
combination and conf iguration of buildings
and then enabling the use to explores the
variables that enable or disable that option.
Educational: Plan4DE enables professionals
who dont normally think about or consider
energy to incorporate energy into their plans.
It helps unveil the energy implications of land-
use decisions to those making the decisions.
An unanticipated insight emerged from the design
process: that the process of applying Plan4DE was
as important as the results, in that it fostered the
questions and discussion that give rise to more
eff icient communities.
 PLAN 4 DISTRICT ENERGY

7.1 M O D E L C H A R AC T E R I S T I C S
Key characteristics of Plan4DE were determined during the design process. The model was designed for
specif ic countries, with the potential to be expanded to other locations as a result of accessibility to data
and the teams knowledge of those countries. In keeping with the imperative of inf luencing the design
and development of the built environment, the model was designed to accommodate geographies that are
common in the world of city planning; both the city and the neighbourhood. Consideration was given to
both greenf ield sites in which buildings are constructed on a blank slate, and existing built environments
which are confronting change in the form of new developments or intensif ication. The audience for
Plan4DE was identif ied as city planners and those who have a signif icant inf luence over the built
environment.

Table 5: Characteristics of Plan4DE

PA R A M E T E R MODEL DESCRIPTION
Geographic The tool will be applicable for all IEA countries. Initial data catalogues will be
provided for Denmark, Canada and the US.
Application Applicable for community plans, secondary plans and/or neighbourhood
plans.

Greenf ield or intensif ication projects- can address contexts with new
construction and/or existing buildings.
Audience The primary target audience with be land-use planners and educators in
universities.

Secondary audience includes politicians, bridging to developers and engineers

as well as consultants supporting the planning community.

There were many implications of these decisions for Plan4DE. The requirement to address broadly
different spatial contexts (neighbourhood to city) in combination with the need for an accessible platform
resulted in the decision to use Excel as the modelling platform and to undertake spatial analysis separately
in a GIS platform. Given that most planners have limited modelling experience, there was an imperative to
ensure that Plan4DE was simple to use, with a concern that the richness of the model would be lost in the
simplicity. As a result, three different interfaces were developed to enable its application with different
levels of technical ability.

On a district energy feasibility trajectory, Plan4DE was def ined as pre-pre-feasibility; pre-feasibility
was def ined as an initial scan to identify whether or not a feasibility study (a detailed assessment of
the economic parameters of a project) is merited. As a pre-pre-feasibility assessment, Plan4DE is not
concerned with the details of a particular project, but rather whether or not the existing or proposed built
environment could be optimized to enhance the feasibility of a generic district energy system.

Plan4DE
 PLAN 4 DISTRICT ENERGY

Pre-Pre-
Pre-Feasibility Feasibility Design Tender
Feasibility

Scope of Options

Engineering Detailed
Plan4DE RETScreen
Models Costing

Before Construction 5-30 Years 3-5 Years 2-3 Years 1-2 Years 0-1 Year

Cost Estimates +/- 20-50% +/- 20-50% +/- 15-30% +/- 5-20% +/- 3-10%

Figure 5: Plan4DE relative to other tools

P L A N 4 D E | 41
 PLAN 4 DISTRICT ENERGY
7.2 A N A L Y T I C A L A P P R OAC H
DEFINE URBAN
DISTRICT
E S TA B L I S H
THE
FEASIBILITY
OF DISTRICT
ENERGY
CO M PA R E
REFERENCE
& DISTRICT
ENERGY
SOLUTIONS
Figure 6: Key tasks in applying Plan4DE
The f irst task in applying Plan4DE is to def ine
the district including its size, whether it is a
greenf ield or redevelopment project, how the
district is conf igured, the number of buildings,
the type of buildings and the density of buildings.
These calculations can be performed in GIS or
estimated according to the use case of the analysis.
Subsequently, the reference case for thermal
energy is def ined; what technologies are used to
heat or cool the buildings and with what fuels?
The next step is to def ine tahe district energy
system including whether technologies will be
optimized according to cost, or whether specif ic
technologies will be selected based on available
fuels or policies. The economics of the reference
case and the district energy solution are then
compared; based on the results, the user can go
back into the model to ref ine the characteristics of
Plan4DE
DEFINE &
C A L C U L AT E
REFERENCE
ENERGY
SOLUTION
DEFINE &
C A L C U L AT E
DISTRICT
ENERGY
SOLUTION
the district in order to improve the conditions for
district energy or to test other variables.
7. 3 M O D E L D E S I G N
Plan4DE was implemented in Microsoft Excel
to enable broad access and easy adaptability.
Figure 7 illustrates the model structure. A general
framework includes catalogues of technologies and
assumptions on prices and emissions. A district
is def ined in terms of buildings and a network.
Constraints of various forms are applied to the
technologies, costs, buildings and other variables
based on the specif ic circumstances of a particular
project. With the above considerations, the model
calculates heat generation costs, network costs
and the total heat supply costs. The outputs
 PLAN 4 DISTRICT ENERGY

are a comparison of district energy costs versus
individual heating costs and environmental
impacts in terms of air pollutants and greenhouse
gas emissions.
At the heart of the model is a cost curve for district
energy, which is calculated for each district based
on the optimization of technologies or user-selected
technologies. The district energy cost curve is
generated specif ically for each conf iguration of the
built environment, varying according to the optimal
combination of district energy technologies required to
supply heat for a particular combination of buildings.
A second linear cost curve represents the cost of
individual heating systems and this cost is constant.
The desirable outcome is to identify a heat density at
which point the cost of district energy is lower than the
cost of the individual heat supply, indicated as You
win in Figure 8.

GENERAL FRAMEWORK L OC A L D I S T R I C T A R E A
Technology def initinons Building typology
Fuel prices, emission costs
Network typology
Detailed information (possible
to modify)

D E F I N I T I O N O F R E L E VA N T
G E N E R AT I O N U N I T S

Model guidance on technology

Technical constraints
Political constraints
Options for individual supply

MODEL ENGINE OUTPUT

Calculation of heat generation
costs Comparison District
Energy vs. Individual
Option for optimization
Cost
Calculation of network costs
Calculation of total heat Environmental indicators
supply costs

Figure 7: Model structure

P L A N 4 D E | 43
 PLAN 4 DISTRICT ENERGY

DO N T B OT H E R
CO S T

T R Y AG A I N
I N D I V I D UA L H E AT S U P P L Y

YO U W I N

H E AT D E N S I T Y
Figure 8: Illustration of cost curve concept

If the cost of individual heating is lower than the cost of district energy, in the try again circle, the user
can adjust variables in the built environment to increase the heat density.

Table 6: Costs included in Plan4DE

D I S T R C T E N E R G Y CO S T S I N D I V I D UA L S Y S T E M CO S T S

Generation capacity (capital costs) Generation capacity (capital costs)

Operations and maintenance Operation costs and maintenance
Fuel costs Fuel costs
Environmental costs Environmental costs
Network costs
Electricity generation (sales)

Plan4DE
 PLAN 4 DISTRICT ENERGY

Cost for the district energy system includes capital, f ixed and variable operation and maintenance (O&M),
fuel, carbon & other emissions costs, power revenue (in the case of combined heat and power (CHP)), and
network/infrastructure costs. Similarly, the cost for individual heating (reference case) includes capital,
f ixed and variable O&M, fuel, carbon & other emissions, but excludes power revenue and network costs.

The total cost for both district energy and individual heating are annualized based on a user-chosen
system lifetime in years. The annual cost to supply the total annual heat demand through an optimized
district energy system is compared with the cost to supply that same heat demand through individual
heating.

District energy is considered viable when the annual cost of district energy is more costeffective than
individual heating.

Table 7: Technologies addressed in Plan4DE

D I S T R I C T H E AT I N G I N D I V I D UA L H E AT I N G
Large CHP-coal Boiler- natural gas, single family

Large CHP- natural gas Boiler- biomass, single family

Large CHP- wood pellets Boiler-oil, single family

Medium CHP- natural gas Air to air heat pump, single family

Medium CHP- wood chips Ground source heat pump, single family

Medium CHP- straw Air to water heat pump, single family

DH boiler- natural gas Solar heating, single family

DH boiler- wood chips Electric heating, single family

Figure 9: Cost data for various technologies

P L A N 4 D E | 45
 PLAN 4 DISTRICT ENERGY

Through cost optimization, the model selects the most cost-effective combination of district energy
technology to supply the total heat demand. However, the user can turn off optimization and select the
district energy technology they wish to explore. System costs are generated for the district energy system.

Figure 10: Sample optimisation duration curve for a district energy system

7. 3 .2 R E V E N U E G E N E R AT I O N

If the user is interested in considering revenue generation (from electricity generated through CHP),
the model is able to take this preference into consideration when optimizing technology and calculating
overall system cost. Annual revenue generated is deducted from the annual system cost.

7. 3 .3 N E T W O R K L E N G T H

The model includes the functionality to estimate the network length to supply the district area if network
length is unknown; alternatively, network length can be calculated in GIS and entered into Plan4DE.
If network length is not available, an algorithm was developed to give an estimate of the cost of district
energy distribution piping and energy transfer stations, using inputs including number of buildings and
total area.

Table 8: Inputs and outputs of Plan4DE

INPUTS OUTPUTS
Total plan area Trunk piping from energy centre
Number of buildings in the plan area Trunk piping to primary lines
Number of clusters in the plan area Primary distribution piping
Total energy consumption Secondary distribution piping
Excess capacity (for future growth) Connection piping
Ration of peak to average power demand Energy transfer station costs
Temperature of DES
Location of energy centre
Price modif ier

Plan4DE
 PLAN 4 DISTRICT ENERGY

The algorithm was derived by analyzing the relationship between plan area, number of buildings
connected and total network length for twenty-three different district energy systems. The total length of
pipe was found to be proportional to the square root of the number of buildings and to the square root of
the system area; the results of this analysis are indicated in Figure 11.

Figure 11: Actual vs modelled network length

7. 3 .4 D E S Y S T E M L I F E T I M E

The annualized cost of the DE system is driven by the expected lifetime of the DE system. The user can
select 30yrs, 50 yrs, or 100yrs.7.3.3 Network length

7. 3 .5 I N D I V I D UA L H E AT I N G

The model allows for the user to adjust the assumptions of the business-as-usual case, that is, specify the
energy supply mix for how buildings would otherwise be heated where district energy is not present.

7. 3.6 WA S T E H E AT A N D I N D U S T R I A L H E AT D E M A N D

The model is able to incorporate waste heat recovery from industry (if known), as well as

heat demand for industrial facilities (if known). However, the user is required to input this information as
there is no data or underlying assumption included in the model

7. 3 .7 C U R R E N C Y

The model allows for results to be shown in Canadian dollars, US dollars, or Danish Krone. The default
setting in the Basic and Intermediate versions is US dollars. Currency selection and assumptions on
exchange rate can be found and modif ied in the Advanced version.

P L A N 4 D E | 47
 PLAN 4 DISTRICT ENERGY

7. 3 .7 V E R S I O N S

The annualized cost of the DE system is driven by the expected lifetime of the DE system. The user can
select 30yrs, 50 yrs, or 100yrs.7.3.3 Network length
Table 9: Three different interfaces

BASIC I N T E R M E D I AT E A DVA N C E D

Simple interface (few inputs More complex interface (more No interface

required by user) inputs required by user)
Fully open and unlocked
Model calculations locked and Model calculations locked and
hidden hidden All data and assumptions can
be changed and modif ied
Most data assumptions are Some data assumptions are
predetermined and set predetermined and set

There are three versions of the model available: Basic; Intermediate; and Advanced. All three versions use
the same Excel Plan4DE model, the only difference is that the Basic and Intermediate versions allow for a
limited number of inputs made available to the user through an interface, some assumptions are pre-set,
and majority of the model is locked/hidden.

In the Advanced version, there is no interface, all the inputs variables are open for input, there are no
preset assumptions, and all the calculations and data are open and transparent.

Plan4DE
 PLAN 4 DISTRICT ENERGY

About Inputs Results Simulate

Enter your data in the input boxes below Click Simulate Review results on the Results Tab

District area km
2

?
Climate Zone Climate Zones
?
Lifetime of District Heating Network years

Revenue Consideration Yes No

Network Length Calculation

?
Calculate Network Length Known Network Length

?
Location of Energy Centre Centre

?
Number and Area of Buildings Number of Buildings Area of Building Types
?
Energy for Heating Override Yes No

Heating Energy Use

Number of Total Area of Building
2
Intensity (EUI)
Buildings Types (m )
(GJ/m2/year)
Residential Buildings
Single-Family Detached

Single-Family Attached

?
Low Rise Apartment

?
High Rise Apartment

Mobile Home
Commercial Buildings
?
Small Office
?
Medium Office

?
Large Office

Warehouse

Strip Mall

Stand-Alone Retail

Primary School

Secondary School

Supermarket

Quick Service Restaurant

Full Service Restaurant

Hospital

Outpatient Health Care

?
Small Hotel

?
Large Hotel

Mixed Use Buildings

?
Mixed Use Buildings

Figure 12: Intermediate interface

P L A N 4 D E | 49
 PLAN 4 DISTRICT ENERGY

7. 3.8 I N P U T S

The core inputs are designed to be accessible to planners, and at the most basic level, relatively easy to
access.
Table 10: Summary of inputs

I N P U T C AT E G O R Y DESCRIPTION/SOURCE
District area (km^2) Calculated in GIS

Piping system length Calculated in GIS or in the Network Algorithm

Lifetime of the system Select 30, 50 or 100 years

Excess capacity Additional capacity that is not used expressed in a %

Climate zone Climate zone selection drives which set of energy use intensities (EUI)
the model uses, as EUIs vary by location. In general, EUIs are higher
in colder regions, as more energy is required to heat buildings in these
areas on an annual basis.

Number of buildings The total number of buildings and f loorspace that are to be modelled
must be within the identif ied district area. If there are buildings
outside the boundary that need to be included, a new boundary should
be identif ied, and a new district area calculated.

Building types There are 21 different building use types: 5 residential; 15 commercial;
and 1 for mixed-use. If data is available, f loorspace can also be entered
for each building type.

7. 3.9 O U T P U T S

There are a number of outputs/results that the model generates, the most relevant of which are listed
below and included in the Basic and Intermediate interfaces, under the Results tab.

Cost comparison: A table and bar chart showing the annual costs for the optimized (or pre-selected) DE
system compared with individual heating.

Figure 13: Comparison of individual vs district heating costs

Plan4DE
 PLAN 4 DISTRICT ENERGY

Density threshold cost curve: A cost curve showing the specif ic cost ($/Mwh) relative to heat density
(kWh/m2). Heat density in this case refers to the average district heat density - the total heat demand (of
all buildings in a district) divided by the area of the district. The graph indicates the current level of heat
density for the input building mix.

Figure 14: Comparison of individual vs district heating costs

Emissions: A bar chart showing carbon dioxide (CO2), sulphur dioxide (SO2) and nitrous oxide (NOx)
emissions for both the optimized (or pre-selected) DE system and individual heating.

Figure 15: GHG emissions and air pollutants

PLAN4DE | 51
 PLAN 4 DISTRICT ENERGY

7.4 DATA
All data used within Plan4DE is accessible and can be revised for a specif ic context (in the advanced
version of the model). In order to help guide the user in customizing assumptions, colour coding is
employed to indicate which cells can be modif ied (orange), without impacting the function of the model.
7.4.1 CO R E DATA S E T S

The model is prepopulated with a number of key data sets. 1

Table 11. Core data sets and data scenarios.

C AT E G O R Y D E S C R I P T I O N SOURCE TA B
Heating/ Energy for heating or cooling- total consumption US Department of Heating
Cooling and peak load for 24 climate zones in US and Energy; Natural
Canada. Resources Canada;
Building sizes. Danish Energy
District energy network data, including costs. Agency.
Fuel prices Includes fuel cost projections to 2075. The user can IEA World Energy FuelPrices
select from multiple scenarios including: Outlook 2015,
ENS 2012 National Energy
ENS 2014-base Board 2015 (Canada).
ENS 2014-450 ppm
ENS 2014- Current prices
Fuel Canada 2
Fuel USA
Custom Fuel
Emissions A range of emissions price scenarios are included, IEA World Energy Emissions
prices based on IEA projections: Outlook 2015. Prices
ENS 2012
ENS 2014-base
ENS 2014-450 ppm
ENS 2014- Current prices
Fuel Canada
Fuel USA
Custom Fuel
Technology The model includes detailed technology data for a Danish Energy TechData
data range of scenarios: Agency Technology
Current data Catalogue.
Custom tech
Tech base
Tech low
Tech high
Tech Canada
Tech USA
Emissions Emissions factors for each fuel source by year are Danish Energy Emissions
factors (CO2, included. Agency Technology Factors
SO2, NOx, Catalogue.
N2O, CH4)

1 Currently the Canada and US price scenarios have not been populated, but will be f illed in on subsequent releases.
Plan4DE
 PLAN 4 DISTRICT ENERGY

The model includes extensive pre-populated data on buildings for 24 different climate zones in Canada
and the United States, as well as detailed cost and performance information for each of the technology
types (Figure 16). In the event that the model is used in other jurisdictions or for other technologies, it is
relatively easy to add new datasets as required, either using the custom data entry option or replacing an
existing dataset.

Figure 16: Heating intensity by building type by climate

PLAN4DE | 53
 PLAN 4 DISTRICT ENERGY

7.5 C U R R E N T L I M I TAT I O N S
In the process of validating the model various limitations were identif ied. In some cases these limitations are a
function of the requirement to make the model accessible to planners. The limitations are as follows:

Thermal storage: the model does not account for thermal storage capacity.

Duration curve: the model uses a static predetermined load duration curve, that is, while the curve
shifts, the shape of load duration curve does not adjust as building use mix changes.

Feasibility: Plan4DE is not a feasibility tool, and hence does not determine whether a DE system is
economically feasible. It functions as a cost comparison tool, enabling planners to understand and
compare the costs of a DE system with that of individual heating over the lifetime of the system for a
predetermined building mix. However, Plan4DE could play a valuable role in determining feasibility
of DE by contributing full cost accounting information to inform a business case.

Plan4DE
 VA L I DAT I O N
8. VA L IDATIO N
8.1 A P P L Y I N G
P L A N 4 D E TO A
N E I G H B O U R H OO D
ARCHETYPE
In addition to using archetypes at the building
scale to increase the accessibility of the model
for planners, an archetype neighbourhood was
developed and Plan4DE was used to evaluate
and inform development scenarios. Once
the neighbourhood was conceptualized, a
typical planning process was undertaken that
incorporated Plan4DE.
An archetype is def ined as a very typical example
of a certain person or thing, almost, but not
quite, one of the repeating components contained
in a pattern, or a pattern itself (Eisenack, Luedeke,
& Kropp, 2006). An archetype can equally be
spatial, referring to a tangible, physical space
or intangible as in circumstances, reactions or
relations. It may or may not refer to the idealized
generalization of a particularity, but it certainly
captures dimensions which can be generally
encountered in a similar context (Cicchetti,
2006). Archetypes are used in a variety of f ields
ranging from literature to design, from psychology
to ecology, as a way of simplifying complexity
without forfeiting the ability to provide
meaningful insights.
PLAN4DE | 55
Archetypes in architecture and planning are used
to categorize typical types of development and
help guide best practices and design decisions.
As described in Section 1, urban planning is the
domain of what are termed wicked problems
challenges to which there are no easy solutions.
Archetypes are thus a useful tool to identify the
essential characteristics of a particular aspect
or pattern of the built environment and then
to add, take away or substitute this pattern in
other places. The attributes of an archetype can
describe quantitative dimensions such as energy
use, impermeable space, building types and others.
When these archetypes are employed to revise
the landscape, conclusions can then be drawn
by analyzing the attributes associated with each
archetype.
Archetypes are commonly used to describe
energy characteristics of the built environment.
Van der Laan (2011, p. 9) identif ies nine recent
studies in the academic literature that have used
archetypes to describe energy consumption. In the
Urban Archetypes Project, a multi-year initiative
funded by Natural Resources Canada, Canadian
researchers led a project (20042010) studying
energy consumption of typical households in
different neighbourhoods across Canada, linking
energy consumption with neighbourhood urban
form and demographics (Webster, 2007).
Archetypes are frequently used in energy
modelling, in particular to construct bottom-
up models in which each archetype represents
 VALIDATION

a dwelling, generalized to its basic shape with
standard spatial and energy characteristics
(Marique & Reiter, 2014; Medina, 2011;
Natarajan, Padget, & Elliott, 2011). As illustrated
above, archetypes can also be def ined at the
neighbourhood level, in one case as a tool to
explore retrof itting suburbia (Barton, Grant,
Rice, & Horswell, 2011). The District Energy
Concept Advisor relies on archetypes derived from
real cases to illustrate the potential for district
energy in different conf igurations (Erhorn-kluttig
et al., 2013).
Very few studies were identif ied that employed
archetypes to help inform land-use that would
support district energy or that would inform
the development of district energy systems. One
Swedish study developed a series of building
archetypes to model how solar energy and load
shifting could be employed to reduce the peak
load for district energy systems (Molander &
Olofsson, 2012), and another evaluated the heating
and cooling load of seventeen neighbourhood
typologies (Mazia et al., 2009).
and suburbia that are expected to experience
incremental development.
Neighbourhoods in Seattle, Burnaby, Vancouver,
Boston, Lille and Copenhagen were identif ied and
considered as demonstrating the characteristics
desired for this archetype. The prototypical
neighbourhood was based on Lake City in
Seattle but was reconf igured to represent aspects
that neighbourhoods in many cities possess,
for example, a cross cutting boulevard which is
primarily auto-dominated; moderately dense
blocks populated with single family and multi-
family housing, low rise professional off ice
clusters, parks, schools and civic buildings,
including a library. A three dimensional virtual
model of the neighbourhood was developed in
order to test the application of Plan4DE (Figure
17).

In this case, an archetypical neighbourhood was

developed, but the planning process was itself also
an archetype of a common planning process.

8. 1.1 A n e i g h b o u r h oo d a r c h e t y p e

In developing an archetype neighbourhood for

Plan4DE, consideration was given to the most
likely urban forms wh ich would support district
energy. Cities in North America and Europe are
frequently characterized by an urban and dense
core or centre with large swathes of ex-urban
or suburban areas on the edges or perimeters.
There are few greenf ield locations remaining in
which a new development or neighbourhood is
constructed from the ground-up; instead most
new development occurs in the midst of existing
structures, either as retrof its or new construction
in locations where demand is perceived by the
private market. Given these considerations, the
focus of the archetype was not on the urban code
of a city but on the areas in-between the core

Plan4DE
Visualizing District Energy VA L I DAT I O N
Planning Archetype

A typical urban neighborhood

approximately 10 min. walk in diameter

Planning and District Energy GGLO

DESIGN
Figure 17: Archetype neigbourhood

Planners will typically initiate a district energy screening process by establishing a boundary within
which they will perform the analysis. This boundary will likely correspond to existing or future land
use boundaries such as neighborhood boundaries, zoning boundaries or other boundaries unique to
the planning area. The archetypical neighbourhood was designed so that it could be crossed on foot
in 10 minutes, an average distance that people will walk to a high frequency transit hub or a central
business district, and a distance frequently used to delimitate a neighhourhood. The total area of the
neighbourhood archetype was calculated to be 600,000 m3 (148.3 acres) with existing buildings totaling
140,000m2 (Figure 18).

It was assumed that the neighbourhood was developing with moderate market demand over a planning
horizon of 20 years.

When establishing new planning policies or land use regulations and development incentives, planners
will often attempt to focus the location of development where it can best achieve planning goals. In most
cases, this involves limiting the area for new planning policies to a boundary that the development market
can absorb within the def ined planning period. Figure 19 illustrates those areas for the neighbourhood
archetype.

PLAN4DE | 57
Visualizing District Energy VA L I DAT I O N
Planning Archetype

Planning Boundary

600,000 m2 (148.3 acres) Land Area

140,000 m2 (1.5mil gsf) Existing Building Area

Visualizing District Energy

Figure 18: Planning boundary
Planning Strategy
Planning and District Energy GGLO

DESIGN
Planning Boundary
Phase 1
600,000 m2 (148.3 acres) Land Area

140,000 m2 (1.5mil gsf) Existing Building Area

475,000 m2 (117.4 acres) Phase 1 Land Area

Figure 19: Phase 1 development boundary

Planning and District Energy GGLO
DESIGN

Plan4DE
 VA L I DAT I O N

As new development incentives and land use regulations are deployed to focus development on the
priority areas, consideration is typically given to the mix of building uses to ensure compatibility with
existing uses. As will be illustrated, the mix of building uses inf luences the heat demand of the area and
thus the viability of a future district energy system. At this stage, a potential location for the physical
district energy plan was identif ied within close proximity to these focus areas. In many jurisdictions,
district energy plants have specif ic zoning requirements (Figure 20).

In the case of the archetype, three zones were established that would be likely to contribute to the
Visualizing District Energy Planning
feasibility of district energy. A residential zone with some commercial use enables buildings Strategy
four to six
storeys, a mixed use off ice enables off ice towers of a similar height, typical of this type of context, and a
low rise zone enables townhouses and walk-ups. Zoning is replicated on both sides of the street to ensure
a consistent streetscape, a best practice in planning.

Land Use Zone to Achieve

District Energy Feasibility
MIXED USE
RESIDENTIAL
(MEDIUM INTENSITY)
ZONED FAR: 2.25 TO 5.75

Proposed MIXED USE

Plant OFFICE
(MEDIUM INTENSITY)
Location ZONED FAR: 2.25 TO 5.75

LOW RISE
ZONE
(LOW INTENSITY)
ZONED FAR: 1.0 TO 2.0

Figure 20: Phase 1 Land-use zones

Planning and District Energy GGLO
DESIG N

While newer buildings would connect to the district energy system, existing buildings would require a
subsidy or incentive to connect including the grocery store, two schools, off ice complex, amongst others
(Figure 21).

PLAN4DE | 59
 VA L I DAT I O N

Figure 21: Existing buildings

In the f irst phase of development 254,800m2 was added, more than double the original building
f loorspace; a relatively high rate of development for this kind of neighbourhood. Figure 22 is an
illustration of what this level of development looks like.

Figure 22: 3d model of neighbourhood

Plan4DE
 VA L I DAT I O N

Figures 22-26 illustrate the addition of new buildings to the neighbourhood by building type as well as
indicating the share of additional heating that these new buildings represent.

Figure 23: Multi-use buildings

Figure 24: Low rise residential

P L A N 4 D E | 61
 VA L I DAT I O N

Figure 25: Office buildings

Figure 26: Civic buildings

Plan4DE
 VA L I DAT I O N

Using climatic conditions for the City of Seattle, the new development represented by the Phase
1 development was evaluated within Plan4DE. The results indicated that individual heating was
substantially more affordable than district energy (Figure 27). In other words, there were not enough
buildings that consumed enough heat in close enough proximity to justify an investment in district energy
on economic grounds.

Figure 27: Phase 1 results

Thermal energy consumption and building f loor area for Phase 1 is described in Figure 28. Mixed-
use buildings were found to use proportionately more energy per unit of f loor space than commercial
buildings, demonstrating the importance of mixed-use buildings for district energy. This illustration
helped the planners identify which types of buildings to add in order to increase the heat density for
viable district energy.

P L A N 4 D E | 63
 VA L I DAT I O N

Figure 28: Phase 1: Building mix

With the insight that the level of development in Phase 1 was insuff icient to support district, a Phase 2
land-use strategy was developed, focusing on the boulevard (Figure 29).

Figure 29: Phase 2 planning area

Plan4DE
 VA L I DAT I O N

Phase 2 adds 490,200 m2 of building area to the district energy network, as the boulevard transitions to
an urban centre and both new and existing buildings are proposed to be connected to the district energy
system (Figure 30).

Figure 30: Phase 2 buildings

The new level of development is evaluated in Plan4DE and in this case, the costs of providing district
energy are lower than the costs of individual heating for each of the buildings. District energy is therefore
considered viable for the neighbourhood.

P L A N 4 D E | 65
 VA L I DAT I O N

Figure 31: Phase 2 results

The district energy heating supply curve that is generated within Plan4DE for this neighbourhood is
illustrated in Figure 32. The yellow line illustrates the heat density of the project; the blue line represents
the cost of individual heating and the black line is the cost of district energy. In this case the position
at which the yellow line crossed the black line represents a lower cost than the blue line, signaling that
district energy is the preferred option.

Plan4DE
Phase 2 Development
VA L I DAT I O N
Heat Density

Phase 2 totals
TOTAL BUILDING AREA
745,000 m2
(8,020,000 ft2)

TOTAL HEAT LOAD

106,667 MWh

TOTAL HEAT DENSITY

178kWh/m2

HEAT SUPPLY COST

$77/MWh

Figure 32: Phase 2 Plan4DE results

the second part, Plan4DE was used to evaluate

GGLO
8. 1.2 O B S E R VAT I O N S
Planning and District Energy

An archetypical neighbourhood was developed
and a standard planning process was applied
to this neighbourhood with the additional
consideration of district energy using Plan4DE.
Plan4DE successfully provided feedback
on planning decisions to help ensure the
incorporation of district energy into the built
environment. The process demonstrated that
many different approaches could be applied; for
example focusing on enabling district energy
in Phase 1 or assessing the impact of different
locations for the district energy plant.
8.2 A P P L Y I N G
P L A N 4 D E TO T H E C I T Y
O F OT TAWA
An analysis for the City of Ottawa was completed
in two parts. In the f irst part, f ive different
approaches to calculating heating intensity were
compared using data from the City of Ottawa. In
DESIG N
areas with the highest levels of heat density.
8. 2 . 1 CO M PA R I S O N O F M E T H O D S F O R
C A L C U L AT I N G H E AT D E N S I T Y
Heat density, a crucial variable in Plan4DE,
determines the technologies and costs of
district energy systems. In order to evaluate the
impact of different methods to calculating heat
density, several data sets were accessed from
Natural Resources Canada for specif ic areas
of Ottawa including observed data for specif ic
neighbourhoods. Heat density in Plan4DE is
calculated either by identifying the number of
archetype buildings or by specifying the f loor
area associated with a building type. There are
a limited number of archetypes and building
types in Plan4DE, which necessitates a level
of generalization that enables ease of use for
planners. This analysis was intended to better
understand the difference between the head
demand predicted using these methods in
Plan4DE and other methods and observed data.

P L A N 4 D E | 67
 VA L I DAT I O N

T able 12 Comparison of methods for calculating heat density

SOURCE
Natural Resources Canada
(observed data)
Natural Resources Canada- f loor
area (building models)
Natural Resources Canada-
archetypes (building models)
Plan4DE- f loor area
METHOD GJ
Provincial average of f loor area 1,382,569
by building type multiplied by
heat intensity by building type.
Floor area by building type 1,569,709
multiplied by heat intensity by
building type.
Number of buildings by 1,062,974
archetype multiplied by
associated f loor area multiplied
by heat intensity by building
type.
Floor area by building type 1,282,122
multiplied by heat intensity by
building type.

Plan4DE- archetypes Number of buildings by 2,490,7251

archetype multiplied by
associated f loor area multiplied
by heat intensity by building
type.
Enbridge Observed natural gas 1,417,722
consumption

These initial results show a high degree of variability in results using the different assumptions methods.
While the observed data (Enbridge) is considered to be the baseline, it solely accounts for natural gas
consumption and there may be other fuels that are being used. Mapping the building types which are
available in GIS to the archetypes or appropriate heat intensity was also identif ied as a potential source
of error, as there were differences in the number of buildings as tracked by the utility (Enbridge) in
comparison with the GIS data.

The results were separated between residential and non-residential data and the non-residential results
demonstrated signif icant variation (Figure 33).

Plan4DE
 VA L I DAT I O N

Residential heating energy use for Ottawa CTs using

consumption and modelled data
Enbridge
NRCan
3 Smores/m2
Smores arch
GJ heating energy usage (x100,000)

P4DE/m2
P4DE arch

CT # 34 35 41 42 43 44 45 46 47

Non-residential heating energy use for Ottawa CTs using

consumption and modelled data
8
Enbridge
NRCan
3
Smores/m2
Smores arch
GJ heating energy usage (x100,000)

P4DE/m2
P4DE arch
2

34 35 41 42 43 44 45 46 47
Census Tract No.

Figure 33: Comparison of different methods of calculating heat demand for selected census tracts in the City of
Ottawa

P L A N 4 D E | 69
 VALIDATION

In summary, the results indicated that non- m2) and heat density (GJ/m2) were identif ied.
residential buildings categorizations and energy
use should be further scrutinized when using In addition to preliminary numerical analysis,
Plan4DE, particularly if an archetype approach is spatial analysis was also performed using heat
selected. Overall, the /m2 approach showed more maps. A higher order heat density map (Figure
consistent and accurate results from the archetype 34) was developed for all census tracts identifying
approach. hotspots for heat demand, as well as relative heat
density over the district area for each census tract.
The preliminary numerical analysis and spatial
8. 2.2 Eva l uat i o n o f n e i g h b o u r h oo d s i n
heat maps indicated specif ic census tracts that
Ot tawa
warranted analysis with Plan4DE.
Using GIS building information provided, all
buildings in the census tract (CT) areas were
categorized to the Plan4DE building categories.
Total heat demand was calculated by applying
NRCan energy use intensity (EUI) values for heat
to total f loorspace (m2) per building type.

Preliminary analysis was conducted on the

variables listed below:

Floorspace density (m2/m2): total f loorspace/

district area; an indication of average district
building density

Heat density (GJ/m2): total heat demand/

district area; an indication of the average heat
density over the district

Connection density (m2/#): total f loorspace/

unit count; an indication of average building
density, as f loorspace is divided over fewer
buildings (which results in fewer connections)

. The intent of the preliminary analysis was to

identify census tracts that show higher potential
for district energy, and to perform a more
detailed investigation on those census tracts using
Plan4DE. Each census tract was ranked according
to its performance for each variable.

In general, higher f loorspace densities are

associated with higher heat densities over the
district area. Connection density also plays a role,
as fewer connections decrease infrastructure costs.
Census tracts with higher f loorspace density (m2/

Plan4DE
 VA L I DAT I O N

Figure 34: Heat density map of census tracts in Ottawa

When analysed in Plan4DE using the climatic location for the City of Ottawa, the census tracts 34, 41, 42
and 43 returned viable results for district energy in current conditions, without additional buildings. The
results also indicated that combined heat and power with natural gas as a fuel, which generates electricity
in addition to heat, is the optimal technology from a f inancial perspective in each of the census tracts.

P L A N 4 D E | 71
 VA L I DAT I O N

Figure 35: Heat density by parcel in the City of Ottawa

8. 2.3 O B S E R VAT I O N S buildings would be or are heated. For this analysis,

it was assumed that individual heating load was
While Plan4DE is not designed to perform as a
met by a mix of natural gas (80%) and electric
feasibility tool, this analysis suggests that it could
heating (20%). By having a better understanding
nonetheless play a major role in determining
of the reference case heating mix, individual
feasibility of district energy by contributing full
heating costs could be more accurately calculated,
cost accounting information to inform a business
which may have a cost impact when determining
case. In conjunction with developing heat maps
viability. In addition, if large proportions
and spatially identifying areas with district energy
of buildings in the reference case are heated
potential, Plan4DE enables very quick calculation
electrically, there are opportunities to further
of total heat demands and heat densities for any
explore the benef its of combined heat and power
identif ied area, as well as generating district
plants to supply those buildings that would
energy system costs (for optimized or pre-selected
otherwise not be connected to a district energy
technology and fuel) in a very swift manner. This
system.
enables planners to very quickly determine the
implications of any built form for district energy
Waste heat locations: Plan4DE has the ability
potential, and understand the impact of changing
to include waste heat recovery when calculating
building or district densities on that potential.
district energy system costs. Waste heat was not
included in this analysis, but an understanding
Plan4DE calculates the reference case individual
of both location and the amount of waste heat
heating cost based on assumptions of how

Plan4DE
 VA L I DAT I O N

available could be included and perhaps play a

major role.

GIS information: Both heat mapping and

generating f loorspace areas and building counts
that are input into Plan4DE rely heavily on
GIS information. This can be, and often is,
incomplete. Improving GIS to have a clean and
clear record of building information, (particularly
in the mixed-use, multi-unit residential and
industrial uses) would help improve overall
heating demand calculations.

Industrial heat demand: When determining total

heat loads and density for industrial buildings,
a static energy use intensity is applied to all
buildings that show industrial type uses, likely
resulting in gross assumptions about industrial
heat demand. Improving both the accuracy of
the presence of different types of industrial use
buildings, and the EUIs associated with those
uses, would help improve overall heating demand
calculations.

P L A N 4 D E | 73
 D I S S E M I N AT I O N

9 . D I S S EM INATION
Dissemination was considered a critical aspect of ultimately determine the impact of the model in
the project, providing feedback and exposure that advancing district energy.
would inform the development of the model and

9.1 W E B S I T E
A website was developed that enabled people to follow the development of the model, download documents or
presentations, and follow webinars

Figure 36. Plan4DE website

Plan4DE
 D I S S E M E N AT I O N

9.2 W E B I N A R S
Three webinars were held on the relationship between the built environment and district energy, the model design,
and the relationship between planning and district energy (Figure 37).

Figure 37: Webinars

9 .3 P R E S E N TAT I O N S
Plan4DE was presented at a number of events, most notably at the International District Energy Association
conference in Boston (2015) and the United Nations Framework Convention on Climate Change in Paris (2015).

Figure 38: Presentation on Plan4DE at the UN Climate Change Summit in Paris.

P L A N 4 D E | 75
 DISCUSSION
1 0 . DISCUSSION
The intention of the project was to develop a
model that would enable planners to incorporate
consideration for district energy in decisions
with respect to the built environment. A simple
modelling tool was conceptualised that compared
the lifecycle costs of individual heating with those
of district heating.
The analysis conf irmed the importance of the city
planner both in the literature and in the f indings.
A planner can inf luence density and conf iguration
of buildings but can also provide for a location or
locations of district energy centres. The imperative
for engaging planners to consider district energy is
clear.
What emerged near the end of the project
was a sophisticated model capable of detailed
assessments of district energy for different
geographic scales. The sophistication, however,
undermined the original aim of providing a
simple, and accessible tool. As a result, three
different versions of the model were developed for
different use cases, from a very simple analysis to a
much more complex analysis.
A number of unanticipated elements emerged
from the project. One included a demonstration of
an interactive web dashboard for one climate zone
that enabled planners to test various combinations
and sizes of buildings to identify a combination
that favoured district energy. A powerful
algorithm was created that calculates pipe length
for a particular site based on the shape of the
Plan4DE
district energy system without a requirement for
any spatial analysis. An optimisation tool was
developed which identif ied the most cost effective
generation capacity for specif ic heat load. Each
of these sub-projects provided additional insight
into the design and deployment of district energy
systems.
The application of the model in an archetypical
planning project was particularly insightful. The
planning and architecture team used the model
to guide what would otherwise be a standard
planning process in an emerging centre within a
larger city. The conclusion was unequivocal that
Plan4DE helped guide their decision-making to
prioritise a built environment that would enable
district energy, with a tool that was clear and
understandable.
A more detailed analysis of the results of the
model with other data sets. including observed
data, demonstrated that the level of accuracy using
the building archetype approach is less than the
level of accuracy that accompanies the approach
in which users enter building area and building
types as anticipated. This analysis, however,
demonstrated another use case for the model
beyond planning the future built environment.
Plan4DE was used to evaluate the existing built
environment and could be used as a district energy
scanning tool by cities, companies or other levels
of government.
 DISCUSSION

Neighbourhoods were identif ied for which district energy is more cost effective than individual heating
systems, yet the dominant heating approach is individual heating systems. Even when the economics are
favourable, a number of barriers, or in economic terms, externalities remain, which impede the wide scale
development of district energy.

One of these issues was encountered in the planning of the neighbourhood archetype. While the full
build out would support district energy, the f irst phase did not incorporate adequate heat density. In this
case, the district energy developer would need to construct the district energy system taking a risk that
subsequent phases would occur but with no real assurances. If the developer waits for the development of
the second phase, building owners will have already invested in individual systems.

For this reason, the most critical entity in the development of district energy is the municipality, which
ultimately controls the provision of infrastructure and the way in which development occurs.

The interdisciplinary nature of the team was critical to the outcome; the view of the planner, the
architect and the engineer on district energy in concert was critical to developing a meaningful model
that incorporated engineering principles from district energy design, but was accessible and relevant to
planners.

10.1 F U T U R E S T E P S
In the course of the development of Plan4DE, the design team took a number of diversions to explore
promising ideas and concepts. One specif ic method that was explored was the notion of automating the
analysis of spatial data that most cities already collect and maintain. Figure 39 illustrates an example of
that data for a neighbourhood in Seattle.

Figure 39: GIS data from Queen Anne neighbourhood in Seattle

P L A N 4 D E | 77
 DISCUSSION

As an example, a system that connected GIS with the model was developed which used GIS to produce a
map that illustrates heat density for a neighbourhood (Figure 40).

Figure 40: Example of automatic generation of heat map

An online interface was then created that enabled users to see, in real-time with minimal inputs, the
likelihood of district energy resulting from the addition of buildings for a particular location (Figure 41).

Figure 41: Sample web-based interface

Plan4DE
 DISCUSSION

This process demonstrated the potential for creating a systematic and broad-based approach to evaluating
district energy at a much larger scale, laying the groundwork for a future project as data access continues
to evolve.

1 0.2 CO N C L U S I O N
Plan4DE was developed as a unique tool to support planners in considering district energy in the design
for and policies of the built environment. This process has illustrated that it is an effective tool. In order
for the model to be effectively deployed, however, further additional work is required to ensure that
planners are aware of and continue to use Plan4DE so that the model can evolve.

P L A N 4 D E | 79
 APPENDICIES

GLOSSARY
District area (km2) is the total land area within the study boundary.

District heat density (kWh/m2) is the total heat demand (converted to kWh) of all buildings in a district divided by
the area of the district.

District heating network lifetime (yrs) is the estimated lifetime of the DE system, or the time in years the system is
expected to run.

Energy use intensity (GJ/m2/yr) is a metric used to express the annual energy demand per square metre of a building
as a function of its size and other characteristics. EUI is driven by a several different factors, including building use
type and eff iciency.

Heat demand (GJ) is the annual amount of energy needed for space and water heating for a building. It is derived
from multiplying the building f loorspace with the associated energy use intensity for heat (EUI) (GJ/m2/year) for
the building type.

Network length is the total length of network piping within the DE system that supplies buildings with heat, and
includes trunk, primary, secondary and connections pipes.

Specif ic cost ($/Mwh) is the the cost in dollars per MWh of heat to supply the total heat demand.

Study area boundary is the boundary line surrounding the district area that is to be analyzed in the model.

Total building f loorspace (m2) is the total amount of f loorspace in all buildings that requires energy for heating.

Total heat demand (GJ) is the annual amount of energy needed for space and water heating for all the buildings to be
analyzed in Plan4DE.

Total number of buildings is the total number of buildings that is to be heated.

ANNEX XI FINAL REPORT, SEPTEMBER 2016

 APPENDICIES

A B B R E V I AT I O N S
(in alphabetic order)

CHP Combined Heat and Power

DHC District Heating and Cooling

IEA International Energy Agency

IDP Integrated Design Process

Plan4DE Plan for district energy

P L A N 4 D E : R E D U C I N G G R E E N H O U S E G A S E M I S S I O N S A N D E N E R G Y CO N S U M P T I O N B Y O P T I M I Z I N G U R B A N F O R M F O R D I S T R I C T E N E R G Y | 81
 APPENDICIES

Figure 37: Webinars 75

FIGURES Figure 38: Presentation on Plan4DE at the UN Climate
Change Summit in Paris. 75
Figure 1: Factors influencing energy intensity in buildings 17
Figure 39: GIS data from Queen Anne neighbourhood in
Figure 2: Energy use in buildings 20
Seattle 77
Figure 4: Assumptions in district energy models. 32
Figure 40: Example of automatic generation of heat map 78
Figure 3: Modelling of District Energy 32
Figure 41: Sample web-based interface 78
Figure 5: Plan4DE relative to other tools 41
Figure 6: Key tasks in applying Plan4DE 42
Figure 7: Model structure 43
Figure 8: Illustration of cost curve concept 44
Figure 9: Cost data for various technologies 45
Figure 10: Sample optimisation duration curve for a district
energy system 46
Figure 11: Actual vs modelled network length 47
Figure 12: Intermediate interface 49
Figure 13: Comparison individual vs district heating costs 50
Figure 14: Comparison individual vs district heating costs 51
Figure 15: GHG emissions and air pollutants 51
Figure 16: Heating intensity by building type by climate 53
Figure 17: Archetype neigbourhood 57
Figure 18: Planning boundary 58
Figure 19: Phase 1 development boundary 58
Figure 20: Phase 1 Land-use zones 59
Figure 21: Existing buildings 60
Figure 22: 3d model of neighbourhood 60
Figure 23: Multi-use buildings 61
Figure 24: Low rise residential 61
Figure 25: Office buildings 62
Figure 26: Civic buildings 62
Figure 27: Phase 1 results 63
Figure 28: Phase 1: Building mix 64
Figure 29: Phase 2 planning area 64
Figure 30: Phase 2 buildings 65
Figure 31: Phase 2 results 66
Figure 32: Phase 2 Plan4DE results 67
Figure 33: Comparison of different methods of calculating
heat demand for select census tracts in the City of Ottawa 69
Figure 34: Heat density map of census tracts in Ottawa 71
Figure 35: Heat density by parcel in the City of Ottawa 72
Figure 36: Plan4DE website 74
ANNEX XI FINAL REPORT, SEPTEMBER 2016
 APPENDICIES

TA B L E S
Table 1: Description of common planning strategies that
influence urban form 27
Table 2: Characteristics of energy models 30
Table 3: Model characteristics 33
Table 4: Comparison of models reviewed 37
Table 5: Characteristics of Plan4DE 40
Table 6: Costs included in Plan4DE 44
Table 7: Technologies addressed in Plan4DE 45
Table 8: Inputs and outputs of Plan4DE 46
Table 9: Three different interfaces 48
Table 10: Summary of inputs 50
Table 11. Core data sets and data scenarios. 52
Table 12 Comparison of methods for calculating heat density
68

P L A N 4 D E : R E D U C I N G G R E E N H O U S E G A S E M I S S I O N S A N D E N E R G Y CO N S U M P T I O N B Y O P T I M I Z I N G U R B A N F O R M F O R D I S T R I C T E N E R G Y | 83
 APPENDICIES

Bush, R., Bale, C., Taylor, P., & Gale, W. (2014).

B I B L I OG R A P H Y To what extent is the Heat Network Delivery Unit
successful in its support of local authorities? In
Aebischer, B., Catenazzi, G., & Jakob, M. (2007).
International workshop on research f indings on
Impact of climate change on thermal comfort,
sustainable heating provisions and cities: Theory,
heating and cooling energy demand in Europe.
practice and future implications. Edinburgh, UK.
ECEEE 2007 Summer Study Saving Energy - Just
Do It!, 859870.
Calthrope Associates. (2011). Raidf ire &
Urbanfootprint. In Open source tools symposium.
Agarwal, C., Green, G., & Grove, J. (2002). A
review and assessment of land-use change models:
Christen, A., Coops, N. C., Crawford, B., Kellett,
dynamics of space, time, and human choice. US
R., Liss, K. N., Oke, T. R., & Olchovski, I. (2010).
Department of Agriculture General Technical
Comparing an urban metabolism model to long-
Reports, NE-297.
term CO2 eddy-covariance measurements. In 9th
AMS Symposium on the Urban Environment.
Arup. (2011). Climate action in megacities: C40
cities baseline and opportunities.
Christenson, M., Manz, H., & Gyalistras, D.
(2006). Climate warming impact on degree-days
Baker, N., & Steemers, K. (2000). Energy and
and building energy demand in Switzerland.
environment in architecture: a technical design
Energy Conversion and Management, 47, 671686.
guide. New York: E&FN Spon.
Chu, T., Richardson, M. E., & Rogowska, M.
Barker, S., Mishra, A., Irwin, D., & Cecchet, E.
(2014). The importance of district energy in
(2012). Smart*: An open data set and tools for
building resilient cities. In Canada: Becoming a
enabling research in sustainable homes. ACM
sustainable energy powerhouse (pp. 169186).
SustKDD 2012.
Candian Academy of Engineering.
Barton, H., Grant, M., Rice, L., & Horswell,
Cicchetti, J. (2006). Archetypes and the collective
M. (2011). Reshaping suburbs. Bristol: WHO
unconscious, Volume 9, 550.
Collaborating centre for Healthy Urban
Environments.
Clements, B., Coady, D, Fabrizio, S., Gupta,
S., Alleyne, T., & Sdralevich, C. (2013). Energy
Batty, M. (2004). Dissecting the streams of
subsidy reform: Lessons and Implications:
planning history: technology versus policy
International Monetary Fund. Retrieved from
through models. Environment and Planning B:
http://www.elibrary.imf.org/page/120
Planning and Design, 31(3), 326330.
Comeault, C. (2011). Integrated community energy
Beeck, N. Van. (1999). Classif ication of energy
system business case study: Southeast False Creek
models.
Neighbourhood Energy Utility. Sauder School
of Business. Retrieved from http://www.sauder.
Bouleau, N. (1999). Philosophies des
ubc.ca/Faculty/Research_Centres/Centre_for_
mathmatiques et de la modlisation. Du
Social_Innovation_and_Impact_Investing/
chercheur lingnieur. Paris, France:
Core_Themes/Low_Carbon_Economy/~/media/
LHarmattan.
Files/ISIS/Reports/Carbon%20Management%20
Reports/QUEST-ICES-Business-Case-Southeast-
Bush, M. R. (2013). A whole systems approach to
False-Creek-Neighbourhood-Energy-Utility.ashx
modelling low carbon urban heat provision.

ANNEX XI FINAL REPORT, SEPTEMBER 2016

 APPENDICIES

Condon, P., Cavens, D., & Miller, N. (2009). Doherty, M., Nakanishi, H., Bai, X., Meyers,
Urban Planning Tools for Climate Change J., & others. (2009). Relationships between
Mitigation. Policy. Cambridge, MA. form, morphology, density and energy in urban
environments. GEA Background Paper.
Connolly, D., Mathiesen, B. V., Ostergaard, P.
A., Moller, B., Nielsen, S., Lund, H., Trier, D. Eisenack, K., Luedeke, M., & Kropp, J. (2006).
(2013). Heat Roadmap Europe 2050. Construction of archetypes as a formal method
to analyze socialecological systems. Proceedings
Cooper, L., & Rajkovich, N. (2012). An evaluation of the Institutional Dimensions of Global
of district energy systems in North America: Environmental Change Synthesis Conference, Bali,
Lessons learned from four heating dominated Indonesia December, 17.
cities in the U.S. and Canada. In ACEEE.
Erhorn-kluttig, H., Erhorn, H., Weber, J.,
Creutzig, F., Baiocchi, G., Bierkandt, R., Pichler, Wssner, S., & Budde, E. (2013). The district
P., & Seto, K. C. (2014). Global typology of urban energy concept adviser: a software tool to support
energy use and potentials for an urbanization urban decision makers in planning district energy
mitigation wedge. Proceedings of the National supply schemes. ECEEE 2013 Summer Study:
Academy of Sciences of the United States of Rethink, Renew, Restart, 721727.
America. http://doi.org/10.1073/pnas.1315545112
Fehr & Peers. (2009). Assessment of greenhouse
Dale, A. (2001). At the edge: Sustainable gas analysis tools. Analysis. State of Washington.
development in the 21st century. Vancouver: UBC
Press. Floater, G., & Rode, P. (2014). Steering urban
growth: governance, policy and f inance (pp.
Dalla Rosa, A., & Svendsen, S. (2011). IEA- 149). Washington, DC.
ECBCS Annex 51: Energy eff icient communities.
Experience from Denmark, (May), 33643371. Gilmour, B., & Warren, J. (2007). Advancing
http://doi.org/10.3384/ecp110573364 district energy development in Canada: A
process for site selection, review and community
Davies, G., & Woods, P. (2009). The potential and participation. District Energy (p. 65). Toronto,
costs of district heating networks. ON.

Devisme, L., Bosse, A., Dumont, M., & Ouvrard, Gilmour, B., & Warren, J. (2008). The new
P. (2011). Effective modelling of urban systems district energy: Building blocks for sustainable
to address the challenges of climate change and community development. Toronto, ON.
sustainability (pp. 199). Paris, France.
Government of BC. Greenhouse Gas Reduction
DFIC - Dr. Fromme International Consulting. Targets Act: Carbon Neutral Government
(2016). Cogeneration & trigeneration- How to Regulation, Pub. L. No. 392/2008 (2008).
produce energy eff iciently. Deutsche Gesellschaft
fr Internationale Zusammenarbeit (GIZ) GmbH. Government of Ontario. (2016). Ontarios f ive
year climate change action plan-2016-2020.
DKS Associates. (2007). Assessment of local
models and tools for analyzing smart-growth Gray, R., & Gleeson, B. (2007). Energy demands
strategies. Transportation. of urban living: what role for planning. In
Proceedings of 3rd National Conference on

P l a n 4 DE : R e d u c i n g g r e e n h o u s e g a s e m i s s i o n s a n d e n e r g y co n s u m p t i o n b y o p t i m i z i n g u r b a n f o r m f o r d i s t r i c t e n e r g y | 85
 APPENDICIES
the State of Australian Cities. Retrieved from
http://soac.f be.unsw.edu.au/2007/SOAC/
energydemandsofurbanliving.pdf
Hawkey, D. J. C. (2009). Will district heating
come to town? Analysis of current opportunities
and challenges in the UK.
Healy, K., & Schussman, A. (2003). The Ecology
of Open-Source Software Development. Social
Sciences, 124.
Heaps, C., Clark, V., & Howells, M. (n.d.).
Making energy system optimization methodologies
accessible and affordable: Incorporating the Open
Source Energy Modeling System into the Long
Range Energy Alternative Planning system (pp.
14).
Holmes, T., & Pincetl, S. (2012). Center for
Sustainable Urban Systems UCLA INSTITUTE
OF THE ENVIRONMENT URBAN
METABOLISM LITERATURE REVIEW, 128.
Holway, J., Gabbe, C. J., Hebbert, F., Lally, J.,
Matthews, R., & Quay, R. (2012). Opening Access
to Scenario Planning Tools. Cambridge, MA.
Hong, T. (2014). Occupant behavior: impact on
energy use of private off ices. In ASim 2012-
1st Asia conference of International Building
Performance Simulation Association., Shanghai,
China, 11/25/12-11/27/12. Retrieved from https://
escholarship.org/uc/item/6jp5w8kn.pdf
Jaccard, M., Failing, L., & Berry, T. (1997).
From equipment to infrastructure: community
energy management and greenhouse gas emission
reduction. Energy Policy, 25(13), 10651074.
http://doi.org/10.1016/S0301-4215(97)00091-8
Johansson, B. (2000). Economic instruments in
practice 1: Carbon tax in Sweden, 112.
Jones, P. J., Lannon, S. and Williams, J. (2001).
Modelling building energy use at the urban scale.
ANNEX XI FINAL REPORT, SEPTEMBER 2016
Seventh International IBPSA Conference. Rio de
Janeiro, Brazil.
Kane, L., & Behrens, R. (2002). Transport
planning modelsan historical and critical review.
Urban Transport Group, (July), 1519.
Keeler, M., & Vaidya, P. (2016). Fundamentals of
integrated design for sustainable building. John
Wiley & Sons.
Keirstead, J., Jennings, M., & Sivakumar, A.
(2012). A review of urban energy system models:
Approaches, challenges and opportunities.
Renewable and Sustainable Energy Reviews,
16(6), 38473866. http://doi.org/10.1016/j.
rser.2012.02.047
Keirstead, J., Samsatli, N., & Shah, N. (2010).
Syncity: an integrated tool kit for urban energy
systems modelling. Energy Eff icient Cities:
Assessment Tools and Benchmarking Practices, 21.
King, M., & Bradford, B. (2013). Community
energy: Planning, development & delivery-
strategies for thermal networks.
Lee, D. (1973). Requiem for large-scale models.
Journal of the American Institute of Planners,
39(3).
Lerner, J., & Tirole, J. (2005). The scope of open
source licensing. Journal of Law, Economics, and
Organization. http://doi.org/10.1093/jleo/ewi002
Levine, M., & rge-Vorsatz, D. (2007). Climate
change 2007: Working group III: Mitigation of
climate change.
Li, F. (2013). Spatially explicit techno-economic
optimisation modelling of UK heating futures.
University College London.
Li, L., Sato, Y., & Zhu, H. (2003). Simulating
spatial urban expansion based on a physical
process. Landscape and Urban Planning, 64(12),
 APPENDICIES

6776. Mnster, M., Morthorst, P. E., Larsen, H. V.,

Bregnbk, L., Werling, J., Lindboe, H. H., &
Liebowitz, S., & Margolis, S. (2009). Path Ravn, H. (2012). The role of district heating in the
dependence, lock-in and history. Journal of Law, future Danish energy system. Energy, 48(1), 4755.
Economics and Organisation, 11(1), 205226.
Natarajan, S., Padget, J., & Elliott, L. (2011).
Mazia, M., Sze, C., Berge, S., Teller, J., Reiter, Modelling UK domestic energy and carbon
S., & Mnard, R. (2009). Energy requirements emissions: an agent-based approach, 43, 2602
of characteristic urban blocks. CISBAT 2009 - 2612. http://doi.org/10.1016/j.enbuild.2011.05.013
Renewables in a changing climate - From nano to
urban scale, 439444. Newman, P., & Thornley, A. (1996). Urban
planning in Europe: International competition,
Marique, A. F., & Reiter, S. (2014). A simplif ied national systems, and planning projects.
framework to assess the feasibility of zero- Psychology Press.
energy at the neighbourhood/community scale.
Energy and Buildings, 82, 114122. http://doi. Newton, P. W., Tucker, S. N., & Ambrose,
org/10.1016/j.enbuild.2014.07.006 M. D. (2000). Housing form, energy use and
greenhouse gas emissions. Retrieved from http://
Martino, M., Britter, R., Outram, C., Zacharias, researchbank.swinburne.edu.au/vital/access/
C., Biderman, A., & Ratti, C. (2010). Senseable manager/Repository/swin:8196
city. Boston: Senseable City Lab
Nystedt, A. , Shemeikka, J., & Klobut, K. (2006).
Medina Benejam, G., Mata, ., Sasic Kalagasidis, Case analyses of heat trading between buildings
A., & Johnsson, F. (2012). Bottom-up connected by a district heating network. Energy
characterization of the Spanish building stock conversion and management, 47, 36523658.
for energy assessment and model validation. http://doi.org/10.1016/j.enconman.2006.02.030
In Retrof it 2012 Conference, 24-26 Jan 2012,
Manchester, UK.Miller, N. (2013). Urban form OECD/IEA. (2013). Transition to sustainable
and building energy: Quantifying relationships buildings: Strategies and opportunities to 2050.
using a multi-scale approach. University of British http://doi.org/10.1787/9789264202955-en
Columbia, Vancouver.
Oregon Legislative Assembly. Senate Bill 1059
Moffatt, S., & Kohler, N. (2008). Conceptualizing (2010). Retrieved from https://olis.leg.state.
the built environment as a socialecological or.us/liz/2010S1/Downloads/MeasureDocument/
system. Building Research & Information, 36(3), SB1059/B-Engrossed
248268.
Oryani, K., & Harris, B. (1997). Review of land-
Molander, C., & Olofsson, M. (2012). Methods use models: theory and application. Sixth TRB
and potentials to reduce peaks in heating power Conference on the Application of Models, 8091.
demand. Chalmers University of Technology.
Patwardhan, A., Azevedo, I., Foran, T., Patankar,
Mller, B., & Nielsen, S. (2014). High resolution M., Rao, A., Raven, R., Verbong, G. (2012).
heat atlases for demand and supply mapping. Transitions in energy systems. Global energy
International Journal of Sustainable Energy assessment, 11731201.
Planning and Management, 1, 4158.
Pinto, N. N., & Antunes, A. P. (2007). Modeling

P l a n 4 DE : R e d u c i n g g r e e n h o u s e g a s e m i s s i o n s a n d e n e r g y co n s u m p t i o n b y o p t i m i z i n g u r b a n f o r m f o r d i s t r i c t e n e r g y | 87
 APPENDICIES
and urban studies: An introduction. ACE:
Architecture, City and Envronment, 471486.
Pitts, A. (2008). Future proof construction
Future building and systems design for energy and
fuel f lexibility. Energy Policy, 36(12), 45394543.
http://doi.org/10.1016/j.enpol.2008.09.015
Ratti, C., Raydan, D., & Steemers, K. (2003).
Building form and environmental performance:
archetypes, analysis and an arid climate. Energy
and Buildings, 35(1), 4959.
Robinson, D., Campbell, N., Gaiser, W., Kabel,
K., Le-Mouel, a., Morel, N., Stone, a. (2007).
SUNtool A new modelling paradigm for
simulating and optimising urban sustainability.
Solar Energy, 81(9), 11961211. http://doi.
org/10.1016/j.solener.2007.06.002
Rodger, S. (2010). The Missing Utility
District energy: Success factors for sustainable
implementation in Londons Lower Lea Valley.
University College London.
Rogner, H.-H. (1993). Clean energy services
without pain: District energy systems. Energy
Studies Review, 5(2).
Ronn, D. (2011). A spatial analytic method for the
preliminary design of a district energy network
utilizing waste heat in mixed-used jurisdictions.
Dalhousie University.
Rosa, A. D. (2012). The development of a new
district heating concept. Technical University of
Denmark.
Rutherford, S. (2009). Bill 27: Opportunities
and strategies for green action by BC local
governments. West Coast Environmental Law.
Retrieved from http://www.toolkit.bc.ca/sites/
default/f iles/WCEL_Bill27_Opports_and_
Strategies_for_Green_Action.pdf
Sahely, H. R., Dudding, S., & Kennedy, C. A.
ANNEX XI FINAL REPORT, SEPTEMBER 2016
(2003). Estimating the urban metabolism of
Canadian cities: Greater Toronto Area case study.
Statistics, 483, 468483. http://doi.org/10.1139/
L02-105
Salat, S. (2009). Energy loads, CO2 emissions
and building stocks: morphologies, typologies,
energy systems and behaviour. Building Research
& Information, 37(56), 598609. http://doi.
org/10.1080/09613210903162126
Sarralde, J. J., Quinn, D. J., Wiesmann, D., &
Steemers, K. (2015). Solar energy and urban
morphology: Scenarios for increasing the
renewable energy potential of neighbourhoods in
London. Renewable Energy, 73, 1017. http://doi.
org/10.1016/j.renene.2014.06.028
Sarralde, J. J., Quinn, D., & Wiesmann, D. (2011).
Urban modelling for resource performance
analysis: estimating cities renewable energy
potential. In Proceedings of Building Simulation
(pp. 1370-1377).
Senbel, M, Church, S., Bett, E., Maghsoudi, R., &
Zhang, K. (2010). The relationship between urban
form & GHG emissions: A primer for decision
makers. Urban Design Lab.
Seto, K., & Dhakal, S. (2014). Chapter 12: Human
settlements, infrastructure, and spatial planning.
Climate change 2014: Mitigation of climate
change, 9231000. Intergovernmental Panel on
Climate Change.
Sovacool, B. K. (2013). Energy policymaking in
Denmark: Implications for global energy security
and sustainability. Energy Policy, 61, 829839.
http://doi.org/10.1016/j.enpol.2013.06.106
Spth, P. (2005). District heating and passive
housesInterfering strategies towards sustainable
energy systems. ECEEE 2005Summer Study - What
Works & Who Delivers?, 339344.
Steemers, K. (2003). Energy and the city: density,
 APPENDICIES

buildings and transport. Energy and Buildings, Vandevyvere, H., & Stremke, S. (2012a).
35(1), 314. http://doi.org/10.1016/S0378- Urban planning for a renewable energy future:
7788(02)00075-0 Methodological challenges and opportunities from
a design perspective. Sustainability, 4, 13091328.
Steinberg, D. Senate Bill No. 375 (2008). Retrieved http://doi.org/10.3390/su4061309
from http://www.leginfo.ca.gov/pub/07-08/
bill/sen/sb_0351-0400/sb_375_bill_20080930_ Vandevyvere, H., & Stremke, S. (2012b).
chaptered.pdf Urban planning for a renewable energy future:
Methodological challenges and opportunities
Strasser, H., Pol, O., Church, K., Svendson, S., from a design perspective. Sustainability, 4(12),
Dalla Rossa, A., Girault, P., Koch, A. (2011). 13091328.
Annex 51: Energy eff icient communities- case
studies and strategic guidance for urban decision Waddell, P. (2010). Status Report on UrbanSim
makers. (A. Koch & J. Kersting, Eds.). Paris, and the Open Platform for Urban Simulation.
France.
Webster, J. (2007). The Canadian Urban
Stremke, S., & Koh, J. (2011). Integration of Archetypes Project: a tool facilitating the
ecological and thermodynamic concepts in the integration of energy-related information into
design of sustainable energy landscapes. Landscape urban planning decision making. 43rd ISOCARP
Journal, 30, 194213. http://doi.org/10.3368/ Congress 2007, 112.
lj.30.2.194
Webster, J. (2008). The Urban Archetypes Project.
Strzalka, A., Bogdahn, J., & Eicker, U. (2010). National Housing Research Committee.
3D city modelling for urban scale heating energy
demand forecasting, (September), 1517. Wende, W., Huelsmann, W., Marty, M., Penn-
Bressel, G., & Bobylev, N. (2010). Climate
The Energy Saving Trust. (2008). The protection and compact urban structures in spatial
applicability of district heating for new dwellings. planning and local construction plans in Germany.
London, UK. Land Use Policy, 27(6), 864868. http://doi.
org/10.1016/j.landusepol.2009.11.005
United Nations. (2014). World urbanization
prospects. New York, New York. Woods, P. (2012). ETI macro distributed energy
project: Summary report. London, UK.
United Nations Environment Program. (2014).
District energy in cities: Unlocking the full World Resources Institute. (2014). Better growth,
potential of energy eff iciency and renewable better climate: The New climate economy report.
energy. Nairobi, Kenya. (M. Davis & G. Wynn, Eds.). Washington, DC:
World Resources Institute.
van den Dobbelsteen, A., Jansen, S., Vernay, A.
L., & Gommans, L. (2007). ``Building within York, T. (2012). SPARC - Scenario Planning
an energetic context: low-exergy design based Analytical Resources Core.
on local energy potentials and excess or shortage
of energy. In PLEA Conference Proceedings Zinko, H., Bohm, B., Kirstjansson, H., Ottosson,
(National University of Singapore, Singapore). U., Rama, M., & Simila, K. (2008). District
heating distribution in areas with low heat
van der Laan, M. T. (2011). Scaling urban energy demand density.
use and greenhouse gas emissions through LiDAR.
University of British Columbia.

P l a n 4 DE : R e d u c i n g g r e e n h o u s e g a s e m i s s i o n s a n d e n e r g y co n s u m p t i o n b y o p t i m i z i n g u r b a n f o r m f o r d i s t r i c t e n e r g y | 89
 APPENDICIES

ANNEX XI FINAL REPORT, SEPTEMBER 2016