nature redacted - Massachusetts Institute of Technology

Energy & the Built Environment:Assessing Renewable Energy Planning in Burlington, Vermont

by

Ethan Lay-Sleeper

B ArchSyracuse University, 2008

ARCHNvESMASSACHUSETTS INSTITUTE

OF TECHNOLOLGY

JUN 29 2015

LIBRARIES

Submitted to the Department of Urban Studies and Planningin partial fulfillment of the requirements for the degree of

Master in City Planning

at the

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

June 2015

@2015 MIT. All rights reserved.

The author hereby grants to MIT permission to reproduce and to distribute publicly paper andelectronic copies of this thesis document in whole or in part in any medium now known orhereafter created.

Sig nature redactedDepartment of Urban Studies and Planning

May 21, 2015

ature redactedProfessor Alan Berger

Department of Urban Studies and Planning

3cted Thesis Supervisor

Professor Dennis FrenchmanChair, MCP Committee

Department of Urban Studies and Planning

Signature of Author:

Sig

Certified by: Sigr

SignaAccepted by:

i

ture red

1

Energy & the Built Environment:Assessing Renewable Energy Planning in Burlington, Vermont

by

Ethan Lay-Sleeper

Submitted to the Department of Urban Studies and Planningon May 21, 2015 in Partial Fulfillment of the Requirements

for the Degree of Master in City Planning

ABSTRACT

Society's dependence on fossil fuels, spawned during the industrial revolution of the 1 9thcentury, increased the physical isolation between the sites of energy consumption, and sites ofenergy production. Rapid population growth and urbanization following this period gave rise, inthe 2 0 th century, to concerns around the impact of humans on the environment. These concernsprecipitated an increased focus on renewable energy, and sustainable development modelspresent in contemporary urban planning discourse.

Despite the increased focus on urban sustainability, the rapid expansion of renewableenergy capacity and supporting policies, municipal governments in the United States continue tostruggle with incorporating renewable energy systems into the built environment. The primarychallenges concerning this integration rest in the capacity of municipal government to reinterpretthe built environment as a framework for renewable energy, to conduct spatial analysis of thepotential capacity in the built environment, and to synthesize that analysis with municipalpolicies in order to develop more robust and specific targets for renewable energy development.

In response to these challenges, I assess opportunities and barriers for renewableenergy development in the built environment, and synthesize established methods of spatialanalysis, renewable energy policy, and project development models, to inform the role ofmunicipal government in future planning efforts around renewable energy.

To investigate the potential practical applications of this research, I focus on the city ofBurlington, Vermont, which in 2014, earned the status as the first city in the United States tosource 100% of its electricity from renewable sources. I question the replicability of the meansby which Burlington attained this status, whether further opportunities exist for Burlington toexpand its support for renewable energy, and what role the municipal government mightassume in this expansion.

I find the means by which Burlington sources its renewable energy only partiallyreplicable, but I also find significant opportunities for Burlington to expand support for renewableenergy within its municipal boundaries. I conclude my research by providing my findings to thecity, in hopes that they will strengthen the role of municipal government in renewable energyplanning.

Thesis Supervisor: Alan BergerTitle: Professor of Landscape Architecture and Urban Design

Introduction..............................................................................................4Energy & the Built EnvironmentMunicipal Planning for Renewable EnergyImplications of Approach

Renewable Energy in Context......................................................................9Global RenewablesRenewable Energy in the United StatesRenewable Energy Policies and PlanningThe Cost of SolarSolar PV CapacityState Support for Renewable EnergyEnergy Planning in BurlingtonThe Burlington Electric Department

Renewability & Replicability......................................................................22Supply Resources and Capacity ObligationClaiming RenewabilityThe Geography of BED's Renewable ResourcesLocal and Regional Implications of Burlington's Path to Renewables

Renewable Energy in the Built Environment............................................... 28Impacts of Energy FormsGenerating Energy in the Built EnvironmentBarriers in the Built EnvironmentOperational and Technical ConcernsCollective Action & Project Scale

Precedents............................................................................................. 35Community Solar Project ModelsSolarize CampaignsSolar Zoning Models

Proiecting Cavacity................................................................................. 47Spatial Analysis of Solar CapacityAn Alternative, Targeted-ModelSpatial Description of BurlingtonSpatial Analysis of StructuresSpatial Analysis of Open Space

Applications for Municipal Planning............................................................58The Role of the Municipal GovernmentIdentification of StakeholdersStrengthening Policies for SolarDeveloping a Project ModelLeveraging RECs for New Capacity Planning

Conclusion.............................................................................................64

Reference Maps & Diagrams......................................................................69Methodoloav........................................................................................... 79N otes...................................................................................................... 83

Introduction

Energy & the Built Environment

Urban areas are composed of cities, towns, and suburban development; they

function as an organizing framework for the human population. Urban areas house

more than 50% of the current global population, up from only 13% in 1900.1 Urban

areas account for between 71 % and 76% of global C02 emissions, and consume

between 67% and 76% of total global energy.2 Buildings within these urban areas

account for 32% of global energy use, and 19% of greenhouse gas emissions. Urban

areas can be understood as a layering of systems, including buildings, transportation

and energy infrastructure, industry, and the natural environment. These systems

represent massive amounts of energy inputs. These inputs include the embodied

energy required to construct the systems, the operational energy required for heating,

cooling, transmission of energy, and maintenance, as well as the transportation energy

required for moving people and goods within these systems.

The type of energy used and rate of consumption varies for urban areas

depending on multiple geographic, cultural and economic variables. However, tracing

the history of energy types and rates of consumption from pre-industrial societies

through contemporary urbanization, establishes a clear evolution toward the dominance

of fossil fuel energy, higher rates of energy consumption, and increased environmental

impacts associated with this energy use.4

After almost two centuries of evolving understandings and approaches to

industrialization and energy in the built and natural environments, there is a clear

trajectory toward design and planning processes that seek to reduce the negative

4

impacts of humans on the earth. However, there are still questions about how urban

areas can participate in this effort, and the role of planning and design in facilitating this

participation. One common trend is the growing interest in sustainability5 and its

relationship to energy conservation and renewable energy.

Sustainability oriented planning agendas now exist in major cities around the

world: Chicago, New York City, Boston, Portland, Austin, Boulder, San Francisco, Santa

Monica, Zurich, Geneva, Lyon, Paris, Berlin, Munich, Hamburg, Dusseldorf, Koln,

Lisbon, Oeiras, Oporto, Barcelona, Madrid, Singapore, and the list goes on.6 These

cities have invested large quantities of time and money developing their various

environmental action plans, green planning initiatives, climate action plans, sustainable

development principles, and other planning agendas with similarly constructed verbiage.

Even with agendas of sustainability, these cities continue to rely on a global energy

system dominated by fossil fuels, with wind, solar, geothermal and modern biomass

only comprising 1.2% of final global energy consumption.7

Addressing this issue will require global participation across a multitude of

political, economic, cultural and geographic factors, and is beyond the scope of this

thesis. What this thesis focuses on, is the potential role of municipal government in the

incorporation of renewable energy production into the built environment.

Municipal Planning for Renewable Energy

Despite the increased focus on urban sustainability and the rapid expansion of

renewable energy capacity and supporting policies in recent years; municipal

governments in the United States continue to struggle with incorporating renewable

energy systems into the built environment. The primary challenges concerning this

incorporation rest in the capacity of municipal government to reinterpret the built

environment as a framework for renewable energy, to conduct spatial analysis of the

potential capacity in the built environment, and to synthesize that analysis with

municipal policies and project development models.

In response to these challenges, this thesis assesses opportunities and barriers

for renewable energy development in the built environment, and synthesizes

established methods of spatial analysis, renewable energy policy, and project

development models, to inform the role of municipal government in future planning

efforts around renewable energy. To investigate the potential practical applications of

this research, this thesis focuses on the city of Burlington, Vermont, which in 2014,

earned the status as the first city in the United States to source 100% of its electricity

from renewable sources. Investigating the means by which Burlington attained this

status, provides insight into the relationship between municipal planning and regional

energy markets. It also reveals further opportunities for Burlington to expand its support

for renewable energy within the built environment, through the synthesis of spatial

analysis and policy at a local level, and leveraging the value of renewable energy in a

regional context.

The growth and increasing value of solar photovoltaic (PV) technology and its

suitability for integration into the built environment represents a major opportunity for

renewable energy in Burlington. To integrate this technology into the built environment,

Burlington needs to redefine its existing structures as a framework for renewable

energy. This requires that roofs be valued not only as surfaces for protection from the

elements at the individual level, but as surfaces for collection of solar energy at the

scale of the city. To support and articulate this new understanding of the built

environment, spatial analysis using GIS software provides Burlington with the ability to

quantify the total surface area suitable for solar PV. This quantification can serve as the

base material for economic and spatial planning initiatives in support of targeted

renewable energy development at the scale of the city. By synthesizing spatial analysis

and regulatory policies within the built environment, Burlington can support renewable

energy development in more calculated and spatially deterministic ways that exploit

economies of scale and collective action.

This synthesis of quantitative spatial analysis, regulatory policy, and scale of

action focused around the solar functionality of urban surfaces, increases the possibility

for decoupling individual sites of production from their necessary sources of investment.

In this way, the total solar surface area of Burlington can be optimized for maximum

production, and investors can benefit from the renewable energy regardless of the

geography or ownership of the surface on which it is installed.

Implications of Approach

The synthetic approach to renewable energy planning outlined in this thesis

argues for a shift in the dynamic between cities and energy. It seeks to add value to

urban areas as producers of energy, in order to mitigate their role as the world's primary

consumers of energy. Addressing this issue in the existing built environment confronts

the social, cultural, economic and political centers of civilization as drivers of change,

and as examples for future urbanization.

The research and investigation undertaken in this thesis provide a preliminary

outline for how one city might begin to address these issues at the scale of its own built

environment. This thesis does not claim to provide a comprehensive plan for renewable

energy, but rather a series of precedents in solar policy and project development, that

might be adapted for Burlington, and made more spatially deterministic through the

integration of GIS analysis. By adopting regulatory policy based on quantitative spatial

analysis, Burlington can prepare itself for future development of renewable energy, and

potentially play a role in instigating this development.

Before investigating the means by which Burlington transitioned to 100%

renewable sources for its electricity, and before assessing opportunities for Burlington to

increase its support of renewable energy, the following section establishes the broader

context of renewable energy. It synthesizes data from various institutions and agencies

to show the expansion of renewable energy technology and policy that occurred over

the last 40 years. It establishes wind and solar PV generation as the dominant forms of

new renewable capacity, and shows that the United States is lagging in this

development on a per-capita basis. This section discusses the costs of a potential

transition scenario for the United States, where solar PV and wind generation replace

fossil fuels. It also positions the state of Vermont in the broader context of solar PV

development.

Renewable Energy in Context

Global Renewables

The cultural understanding around renewables is evolving from the singular

viewpoint of renewables as just another energy source, to include their impact on

energy security, health and environmental benefits over nuclear and fossil fuels, their

potential for mitigating greenhouse gas emissions, and also the opportunity they

represent for new jobs and education.8 In early 2014, 144 countries had renewable

energy targets, with 138 of those also having implemented renewable energy support

policies.9 Thousands of cities and towns across the world also have renewable energy

targets, plans and policies, which are often more-ambitious than national policy, and

point to the importance of municipal leadership for the future of renewable energy

development.10

Despite these efforts, renewable energy still only accounts for approximately

20% of total global energy consumption. China, the United States and Germany are

leaders in total renewable energy capacity, although other countries like Denmark have

higher per-capita capacities." The following charts illustrate total global energy

consumption and capacity relative to various sources of renewable energy:

Total Energy Consumption All Renewables Modern RenewablesBlofuels

Re Ai a les lFuels M erRrealsTraditonal Reeal SBinaWS Bloas/got

47% Power hoeNuclear 12% 42%

3%

Hydro38%

Figure 3: 2013 Total Global Energy Consumption by SourceSource: REN21. 2014 Renewables Global Status Report (Paris: REN21 Secretariat).

9

The charts below show the role renewable sources play in electricity production; the

primary differences between the global consumption charts and the electricity

production charts are the relationships between biomass and hydro power, with hydro

playing a much larger role in electricity production.

Total Electricity Production Renewable ElectricitySolar PV Geothermal,

Fossil Fuels & 3% CSP, OceanAll FoslFes&Sl Power 2%Renewable. Nuclear 418%

Figure 4: 2013 Total % of Global Electricity Production by SourceSource: REN21. 2014 Renewables Global Status Report (Paris: REN21 Secretariat).

In 2013, the world produced a total of 560 Gigawatts (GW) of electricity from renewable

sources not including hydro. The relationship between global capacity, the European

Union, BRICS (Brazil, Russia, India, China and South Africa) and individual countries is

illustrated below:

GW Capacity GW Capacity560

118

3-----93-----------------------------78

100 GW -- --- - - --

World EU-28 BRICS China U.S.A Germany Spain Italy India

Figure 5: 2013 Non-Hydro Renewable CapacitySource: Adapted by author from: REN21. 2014 Renewables Global Status Report (Paris: REN21 Secretariat).

The top 6 countries account for approximately 70% of total renewable capacity globally.

By energy source, the leading contributors to global renewable capacity are solar PV

and wind energy, which comprise close to 80% of this capacity.12 Between 2000 and

10

2013, global wind capacity increased from 17GW to 318GW; between 2004 and 2013,

global PV capacity increased from 3.7GW to 139GW.13

Renewable Energy in the United States

The rapid expansion of wind and PV capacity reflect increasing cost parity

between renewables and fossil fuels, which has led to significant increases in total

dollars invested in renewables in the United States over the last decade:

U.S.A Investment inRenewables

53.4

33.6 35.9 39.7 35.828.223.5

117

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013Figure 6: Total Annual Investment in Renewables, in Billion USD

Source: REN21. 2014 Renewables Global Status Report (Paris: REN21 Secretariat).

Even with the significant increases in investment and development of renewable

energy, there are still challenges, not the least of which is that fossil fuels continue to

receive massive subsidization, far outweighing that of renewables. In 2011, petroleum

products, natural gas and coal received $480 billion in subsidies globally before taxes;

after taxes, factoring in negative externalities the global value of these subsidies is

calculated to be closer to $1.9 trillion.14 Energy subsidy reform is a complex issue that

could create significant positive change, but also lacks public support, and raises

governmental concerns around inflation.1 The countries of Denmark, Germany, and

Spain have taken unprecedented national leadership in promoting and developing

renewable energy through policy reform and direct investment.16 The fact that Germany,

11

which is the size of Montana, has developed 78GW of renewable capacity, while the

entirety of the United States has only developed 93GW of renewable capacity, implies

that United States has an impressive amount of work left to do. In addition to lagging

behind other nations on a per-capita basis of renewable energy development, the

United States is a global leader in per-capita consumption of non-renewable resources

like petroleum, natural gas and coal.17

To address the current situation around energy in the United States, in 2010, the

Civil Society Institute released an evaluation of a strategy for the U.S. electric industry

to reduce carbon emissions and improve public health, through a large-scale transition

to energy efficiency, and renewable energy. The evaluation draws on growing concerns

over health and environmental risks around fracking, water consumption for cooling in

power plants, and new EPA regulations for greenhouse gas emissions.18 The study

compares a business as usual (BAU) scenario to a transition scenario, and assesses

the potential financial implications using a cost benefit model and sensitivity analysis. In

the transition scenario, all coal fired power plants and 25% of nuclear plants are retired

by 2050, natural gas consumption is lower than in 2011, and energy efficiency and

renewable energy make up the difference.19 The net annual cost impacts of the

transition scenario are evaluated in by decade, and range from savings of $18 billion in

2050, to costs of $9 billion in 2040, with the overall conclusion that the transition

scenario will likely be less expensive than BAU.20 This is especially significant in

consideration of the fact that the BAU scenario doesn't include carbon costs or carbon

reductions, which would further increase savings of the transition scenario.2 1 To achieve

this transition at the projected costs, the model assumes that solar PV grows from 2-

14GW under BAU to 384GW, and that onshore and offshore wind grows from 38-64GW

under BAU to 188GW. To accommodate for the intermittency of these resources, the

scenario also relies on increased storage capacity. The largest increase in renewable

capacity identified in the study is in distributed generation from solar PV. While solar PV

is by no means the only solution, it is one that can be readily adopted by municipal

governments, and has more direct implications for the built environment and the

discipline of city planning.

Renewable Energy Policies and Planning

For municipal government to participate effectively in this planning process, it is

important to position cities in the context of broader federal and state policies that

support renewable energy. Policy plays a significant role in making renewable energy

economically viable, especially when it comes to PV. The main components for solar PV

policy focus on financing the installation of projects, ensuring that the electricity

produced is sold at certain prices, and generating renewable energy credits. A major

contributor to project financing is the federal Investment Tax Credit (ITC) for solar. This

credit has been in effect since 2008 and is set to expire at the end to 2016. It provides a

30% tax credit on the installed cost of solar projects for both commercial and residential

properties. One hurdle for certain municipal projects is that cities don't have tax

appetites, so to take advantage of the tax credits cities need to work with the

commercial sector to negotiate contracts where both parties benefit.

Net metering is a policy tool that incentivizes distributed energy generation by

requiring utilities to purchase excess power generated by customers at a rate slightly

higher than retail. The system is monitored by utility companies through meters that

keep track of the energy consumed and produced by a net metered customer; net

metering also bypasses the need for customers to install expensive battery banks for

storage, as excess electricity is fed back into the grid. Currently, 44 states have

mandatory statewide net metering regulations.23

A Renewable Portfolio Standard (RPS) is a state-level regulatory mandate that

requires a certain percentage of a state's electricity portfolio to come from renewable

sources such as solar, wind and biomass. These requirements typically increase

gradually to allow for new renewable capacity to be developed. As of February of 2015,

31 U.S. states had mandatory renewable portfolio standards, 8 states had voluntary

renewable energy standards or targets, and only 13 states had no standard or target.

Renewable Portfolio Standards also increase the demand for Renewable Energy

Certificates (RECs), which are another financial mechanism designed to further

strengthen the market for renewable energy. For every megawatt hour (MWh) of

electricity produced by renewable sources, one REC is generated. RECs can be sold

with the actual electricity or separately. Purchasing RECs is one way for states to meet

the requirements of Renewable Portfolio Standards. There are generally two classes of

RECs, Class I and Class 11. Class I RECs typically have higher values and are

comprised of new generation such as solar PV, wind and biomass, while Class II RECs

typically have lower values and are comprised of older hydroelectric facilities. RECs are

tracked by a series of regional systems throughout the U.S. For instance, the New

England Power Pool Generation Information System (NEPOOLGIS) provides an online

system that tracks production and sales of RECs in the 6 New England States. In

14

addition to RECs there are Solar Renewable Energy Certificates or SRECs, which focus

solely on PV, and tend to have much higher values than RECs. After purchasing

RECs or SRECs, the owner may choose to retire these in order to claim renewability

and meet required or voluntary renewable portfolio standards.

Another major driver of new PV development in the United States has been the

steady and significant decline in the installed price of solar over the last 15 years. This

is primarily attributed to improved technology and lower module costs, but there are also

other factors that influence installed cost.

The Cost of Solar

In 2014, the Lawrence Berkley Laboratory, in collaboration with the U.S.

Department of Energy (DOE), released a historical summary of the installed costs of

solar PV from 1998-2013. The study makes the following conclusions about efficiencies

of scale, trends in installed costs, as well as estimates for future costs. Nationally, PV

installations for new residential construction have shown consistent savings relative to

retrofit projects, in 2013 the difference was $0.9/watt. When comparing roof mounted

systems to ground-mounted systems, size is a key factor; typically roof mounted

systems are more economical in systems less than 10 kW ($0.8/watt), but this margin

decreases as system size increases, and in 2013 ground-mounted projects over 1,000

kW saved $.07/watt over roof-mounted systems. The report attributes declining costs in

installation over the last decade to a steady decrease in PV module prices (despite

diminishing ITCs), but notes that in 2013 and 2014 installed costs continued to drop

despite steady or increasing module costs. This is attributed to new efficiencies in labor

and installation procedures. Prices in modules are not expected to decrease, so further

price reductions are expected to come from increased innovation in non-module

components and installation methods. These reductions are seen as attainable, based

on comparisons between installed costs in Europe relative to the United States. In 2013

the installed cost before taxes was $2.1/watt in Germany, $2.7/watt in the United

Kingdom, $2.9/watt in Italy, $4/watt in France, and $4.4/watt in the United States. To

reduce the soft costs associated with installation, the report suggests the development

policy that fosters competition within the delivery infrastructure, and specific targeting of

permitting, interconnection and research.26

Even though growth in solar has increased due to technological improvements

and declining costs, it is still more expensive than most conventional energy sources.

Solar also benefits from a variety of fiscal and regulatory incentives, and in order to

achieve large-scale deployment, solar still needs to overcome a variety of technical,

financial, regulatory and institutional barriers, all of which necessitates increased policy

support for solar.2 7

Solar PV Capacity

As a result of declining costs and support on behalf of state and federal

government, electricity generation from solar in the U.S. increased by a factor of 5

between 2008 and 2012.28 In terms of overall PV capacity, California has long been a

leader, followed by New Jersey and Arizona. The top 10 states by total PV capacity are

shown in figure 7 below.

16

PV Capacity (MW)

3,500,000 -- ------------ -----------------------------------------

3,000,000 ---- - - - -- -- ------- ------ ----

2,500,000 - -

2,000,000 - - -- -- --- - ------ --

1,500,000 -- ---- --- --- --

100,000- --

5ooooK - -- - -Um - -- -

CA NJ AZ MA NV TX NY PA NM CO

Figure 7: Total PV Capacity, in MWSource: National Renewable Energy Laboratory, Open PV Project.

The vast disparity between geographic region and total PV capacity help to support the

argument that state policy plays a major role in promoting solar development. Similarly

to total renewable energy at a global scale, these ranking shift if they are considered on

a per capita basis. Figure 8 shows Arizona and New Jersey as leaders in per capita PV

capacity, followed by California, Nevada and New Mexico similarly to the graph of total

PV capacity. What is interesting about the per capita graph is that smaller states like

Delaware and Vermont rank very closely to larger and sunnier states.

PV MW Per Capita0.180 ------ --

0.160 - - -

0.140 -- - - - -

0.120 -- - -- - -

0.100 - - -

0.080 - - -- - - ----- ------- -

0.060 - ---

0.040 - - - - -

0.020-- -

AZ NJ CA NV NM DE MA VT CT COFigure 8: Total PV Capacity Per Capita, MW

Source: National Renewable Energy Laboratory, Open PV Project; United States Census, 2010.

The combination of Vermont's relatively high rate of PV capacity per capita, combined

with fact that Burlington, Vermont is the first city in the United States to have

17

transitioned to 100% renewable sources for its electricity, suggests that there are

opportunities to learn from Vermont's state and local policies.

State Support for Renewable Energy

Vermont is not a major producer or consumer of electricity at a national scale; it

also lacks fossil fuel resources, with hydro and other renewables accounting for 99.5%

of in-state electricity generation.29 In 2012, Vermont also became the first state to ban

fracking. These factors combined, make in-state renewable energy development one of

the Vermont's only options for reducing its reliance on imported energy. This geographic

and resource-related reality, combined with policy has no doubt contributed to the rise

of renewable generation in the state.

Stastically, Vermont adheres to national trends with respect to efficiencies of

scale, with system sizes from 10-100 kW coming in at consistently lower cost per watt

than systems smaller than 10 kW. In Vermont, the median price for systems under 10

kW in 2013 was $4.7/watt, and $4.2/watt for systems between 10-100 kW. The installed

costs in Vermont have dropped from over $10/watt in 2004.30 For a typical 4 kW

residential installation in Vermont, the expected payback period is between 9 and 10

years.31 The Vermont ITC parallels the timeline of the federal ITC, offering an additional

7.2% tax credit to commercial, industrial and agricultural sectors. Vermont also offers

sales and property tax exemptions, and several low-interest loan programs that assist in

financing solar PV installations.

Vermont allows group net metering, which distributes the benefits of net metering

between groups of customers within a utility's service area. Vermont's net metering law

caps the size of projects at 500 kW, and also caps the total capacity of net metering

within individual utility service areas at 15% of their peak load. Utilities are required to

purchase excess generation from net metered systems at $.20/kWh for systems 15 kW

or less, and $.19/kWh for systems greater than 15 kW. Customers in Burlington typically

pay $.14/kWh, so the higher purchase price provides additional incentive for customers

to install solar PV. In summary, Vermont supports renewable energy and solar PV in a

variety of ways that parallel efforts in other states, including the implementation of an

RPS. In addition to participating in best practices for renewable energy policy, Vermont

also has developed ambitious goals for statewide renewable energy development.

Vermont aspires to obtain 90% of its total energy from renewable sources by

2050 in order to reduce its greenhouse gas (GHG) emissions. In 2011, the state

released a 336 page Comprehensive Energy Plan (CEP) outlining a multitude of

strategies for achieving this goal, while protecting Vermont's natural resources and

working landscape. Statewide electricity consumption accounts for .5 million megatons

(MMt) of C02 emissions annually, while residential, commercial and industrial fuel use

accounts for 2 MMt, and transportation accounts for 3.5 MMt.32 In 2009, electricity

comprised 37.5% of total energy consumption, transportation fuels comprised 34.4%,

and heating fuels comprised 28.1%.33 As such, the CEP advocates for increasing

electricity generation and transmission capacities to produce lower-emissions

alternatives to fossil fuels, and to provide the necessary infrastructural capacity for

electric vehicles. After an evaluation of the economics, public sentiment and land use

considerations, the CEP proposes that new electricity generation be met primarily

through increased capacity provided by solar, wind, and combined heat and power

(CHP) facilities powered by natural gas and biomass.34

These state-level factors in policy and planning help to explain Vermont's high

per capita rankings in energy efficiency and renewable energy, and these certainly

influenced and helped the city of Burlington in its transition to renewable electricity, but

municipal policy and planning also played a large role.

Energy Planning in Burlington

The city of Burlington has a history of supporting as well as developing

renewable energy resources. Renewable energy and efficiency have been incorporated

into a number of municipal planning efforts, including:

* Burlington's Municipal Development Plan" Burlington's Climate Action Plan" Burlington's District Energy Study" Burlington's Comprehensive Development Ordinance" The Burlington Electric Department's Integrated Resource Plan

Burlington's energy plan (a component of the Municipal Development Plan) sets policy

goals of educating citizens on the benefits of public utility ownership, and renewable

electricity generation. It also identifies local energy self-sufficiency as an important

component of Burlington's future sustainability.35 The climate action plan provides a

framework for measuring and reducing greenhouse gas emissions, and establishes

goals for compact development, transportation, urban forestry, energy efficiency in

buildings, renewable energy development, waste reduction and recycling.36 Burlington's

district energy study provides analysis and recommendations for utilizing existing

energy resources to develop a Community Energy System (CES). The study focuses on

the potential for improved energy efficiency through interconnecting the McNeil

generating station with the campuses of the University of Vermont Campuses and

Fletcher Allen Health Care.37 Burlington's comprehensive development ordinance (or

zoning ordinance) clearly establishes the encouragement, conservation, utilization and

development of renewable energy resources in the intents and purposes section of its

general provisions. 38 The Burlington Electric Department's Integrated Resource Plan

aims to meet the public's need for energy services at the lowest cost, including both

environmental and economic costs. The plan analyzes potential resource decisions,

proposes strategies for resource acquisition, and must be approved by the city and

Public Services Board. 39 The Burlington Electric Department is the third-largest of

Vermont's 18 electric utility companies, and plays a critical role in shaping and

implementing energy planning in Burlington.

The Burlington Electric Department

The Burlington Electric Department (BED) is a municipally owned utility company

that manages the power purchase agreements and distribution of electricity for the city

of Burlington. As a public power entity BED makes its decisions through an open

process and public input. As such, it has been and continues to be a goal for BED to

provide power that is as clean and as locally produced as possible, while maintaining

affordability. BED helped to pave the way for renewable energy through the construction

of the McNeil wood-powered electric generating facility, which was constructed in 1984.

Between 1989 and 2011 BED invested over $38 million in energy efficiency programs;

Burlington's electricity use in 2011 was 4.7% lower than in 1989, which equates to

approximately $10 million in savings per year.40 BED is currently Vermont's largest

municipally owned electric utility, serving over 19,000 customers in a 16 square mile

area.

In 2010, BED received $69 million in federal grant funds through the Smart Grid

Investment Grant. This enabled increases in efficiency through dynamic billing, better

service analytics and easier integration of renewables. In recent years BED made

significant steps in transitioning to renewable sources, the most recent of these steps

includes the acquisition of the Winooski One hydroelectric facility in September of 2014.

BED provides net metering as a way to encourage private solar PV development within

the municipality; and also contracts for PV generation with several housing projects and

city schools. BED negotiated a lease on the roof of the parking structure at Burlington

International Airport, and constructed a 576 kW facility that BED owns and operates as

of January, 2015. BED continues to pursue contracts for renewable energy from wind

and solar resources within Burlington and the state of Vermont.

Renewability & Replicability

The following section discusses the means by which Burlington currently claims

renewability for its electricity, and assesses the extent to which these means are

replicable by other municipalities.

Supply Resources and Capacity Obligation

To understand Burlington's transition to renewable electricity, it is important to

examine BED's responsibility to its local customers as well as to the regional electric

grid. BED needs to provide electricity not only for the residents of Burlington, but also

needs to contribute capacity to ISO-NE to ensure reliability of the regional electric grid.

This is referred to as BED's capacity obligation, and is based on the load ratio share of

the total ISO-NE capacity requirement. The requirement is a percentage of BED's load

at the time of the annual peak load. The peak load usually occurs in July, and even

though BED peaks at around 65 MW, its capacity obligation is to ISO-NE is 85 MW.

BED meets its municipal energy needs through a series of energy purchase contracts

with approximately 15 generation facilities throughout the Northeast.41 However, not all

of these 15 facilities count toward the capacity obligation due to ISO-NE requirements

and the intermittency of renewable energy. To meet obligations to ISO-NE, BED

purchases capacity through the ISO-NE Forward Capacity Market (FCM). In 2015 BED

expects to purchase 41.5% of its capacity requirements in this way, and anticipates that

unless new capacity is developed, this number will remain relatively constant. While

BED's capacity obligation doesn't contribute to its ability to claim renewability for supply

resources, it does represent a significant cost. Cost of service is a major component of

BED's declared responsibility to its customers, and has played an important role in

directing the utility's support of and transition to renewable energy.

The interplay between flows of energy and money at the local and regional

scales has played an important role in Burlington's energy transition. BED achieved this

transition through a series of long-term initiatives in energy efficiency, expanding utility-

ownership of renewable energy generation, investing in new renewable energy

generation, purchasing power from renewable sources, and finally, through the sale and

purchase of Renewable Energy Credits (RECs) in a regional marketplace.

Claiming Renewability

BED has steadily increased its ability to claim renewability by contracting with

more renewable energy sources, while at the same time stabilizing rates through the

sale of RECs. BED is awarded Class I RECs for contracts it maintains with renewable

generation sources. BED sells these RECs to other entities and then purchases lower-

cost Class 11 RECs which it then retires to regain its claim to renewable energy.

In 2011, BED purchased 46% of its electricity from renewable sources, and sold

the associated Class I RECs to stabilize rates thereby reducing its claim to renewability

to 12.6%. BED then purchased less-expensive Class 11 RECs to reclaim its original 46%

renewability. BY 2013, BED increased its purchases of renewable supply resources so

that before selling its Class I RECs, it could claim that renewable sources provided 95%

of its electricity. In 2013, selling these RECs provided BED with revenues close to 20%

of its cost of service, but lowered the percentage of renewability it could claim to 39%.

With the revenue from selling the Class 1 RECs, BED was able to both control rate

increases and purchase enough Class 2 RECs to reclaim 100% renewable electricity.

The transactions of supply resources and RECs for 2013 are shown below in figure 9.

Renewability Prior to REC Transactions Renewability after REC Sales Renewability after REC Purchasescomincklicu9a% al3-% 1 0

24% 40IJ OA7%ASG6w

6.040

OHM

320% 2.67%

co67m6.73%

Figure 9: Renewability and REC Transactions for Burlington Electric DepartmentSource: Adapted by author from Burlington Electric Department, where we get our power, 2013.

24

As BED acquired more renewable supply resources, the price of Class I RECs

increased significantly, along with BED's revenue from REC sales. Figure 10 shows the

increasing prices BED received for RECs from the McNeil plant, wind and other sources

between 2013 and 2015, as well as the total sales from those sources between 2011

and 2015.

REC Prices (BED) REC Sales (BED)m McNeil a Wind Other $12,000,000

$10,000,000

$8,000,000 Other$8,000,000 Wind

$4,000,000

McNeil$2,000,0

2013 2014 2015 2011 2012 2013 2014 2015

Figure 10: REC Prices and REC sales by Burlington Electric DepartmentSource: Burlington Electric Department, Annual Budget Projections 2013-2015.

The demand created for RECs by the Renewable Portfolio Standards of other New

England states has enabled BED to stabilize its own electricity rates, while still claiming

renewability. Currently, the state of Vermont has a voluntary RPS which will become

mandatory in 2017 and require retail utilities to provide 50% of their electricity from

renewable sources. Vermont's RPS will increase by 4% annually until it reaches 75% in

2032.4 In this regard, Burlington is already well ahead of schedule, in part because of

its early investment in the McNeil generating station in the 1980s. However, as

Vermont's RPS become more stringent, demand for both Class I and Class I RECs will

increase, and eventually more renewable capacity will need to be developed. Where

this capacity is developed will depend on the availability of natural resources and

geography in the case of biomass, wind and hydro, but for solar PV it will depend more

on land use planning and financing models. This new capacity will require significant

25

financial resources in the short-term, but has the potential to produce long-term

economic benefits for the communities in which it is located. In order to assess potential

sites for new renewable energy capacity, it is important to first understand the

geography and extent of existing resources.

The Geography of BED's Renewable Resources

For supply resources purchased through the ISO-NE Exchange, utilities typically

assign a ratio of nuclear, coal and gas based on the average regional mix, thereby

giving a general geographic extent to those resources. However, for thel5 energy

resources for which BED has direct contracts, it is possible to more accurately identify

the sites of production.

In accordance with Rule 5.200 Notification of Power Supply, BED is obligated to

provide a description of its power sources to the Vermont Public Service Board and the

Department of Public Service. The 2015-2019 edition of this report was submitted on

January 3 0th, 2015, and identifies BED's power resources by geography and MWh. In

keeping with BED's 2012 Integrated Resources Plan (IRP), BED's power sources show

a commitment not only to renewable energy, but to local generation facilities, with 11 of

the 15 sources located in Vermont, and 13 qualifying as renewable resources. Together

these 13 renewable resources are projected to provide between 300,000 MWh and

375,000 MWh of annual supply resources between 2015 and 2020.4 These resources

will meet or exceed the forecasted system load starting halfway through 2015. The

largest single contributor to BED's supply resources is the McNeil biomass facility

located in Burlington. The remaining supply resources are composed primarily of a mix

of wind and hydro power, with a much smaller proportion coming from solar. The

following diagrams indicate the geographic extent of these renewable resources and

their respective capacities. Several caveats are that while Burlington only draws 8% of

its electricity from Hydro Quebec, it doesn't identify a specific facility, and thus more

Hydro Quebec facilities are shown than BED actually draws resources from. This is also

the case for BED's contracts through Vermont's Standard offer program which provides

an even smaller percentage of BED's resources from a number of distributed sites, but

is only represented by one site in the following supply resources diagram. All other

sources can be traced to specific singular facilities. M See d-agrasn & 7

Local & Regional Implications of Burlington's Path to Renewables

A better understanding of the geography and quantity of Burlington's current

supply resources and capacity obligation helps to put future decisions about developing

new resources in perspective at both local and regional scales. According to the

Vermont Department of Public Service, in 2011, Vermont received approximately 20%

of its total energy from renewable sources, and imported 40-45% of its total electricity

(including renewable sources) through a combination of direct contracts and short-term

purchases from ISO-NE.45 In the context of the statewide goal of acquiring 90% of its

energy from renewable sources by 2050, there is still a large gap to be filled through the

development of new renewable resources in the next 35 years.

There is no doubt that Vermont will continue to rely on and contribute to the

regional network of renewable energy providers in the Northeast, but there are also

opportunities to take steps toward developing new renewable energy generation

capacity in-state. It is important to understand the efforts in Burlington as a dynamic

feedback between local and regional forces that depend on and affect one another. By

purchasing renewable energy sources locally and regionally, BED is helping to support

renewable energy development at multiple scales. By participating in the regional REC

market BED is simultaneously providing supply and demand, while producing local

economic benefits and regional environmental benefits. The development of new

renewable capacity should therefore be both a regional and municipal priority. Since

municipal governments can't directly take advantage of state or federal ITCs 46 , they

need to develop new approaches for promoting renewable energy.

Different municipalities will have different options in terms of the ways they can

act, based on population, geography, density, and form of the built environment. As

Vermont's largest municipally owned utility, and as a national leader in the transition to

renewable energy, BED is uniquely positioned to pioneer new approaches in municipal

planning for renewable energy. A first step is to identify what types of renewable energy

are appropriate for the built environment, and how the municipal government can better

prepare itself to promote and incorporate new capacity.

Renewable Energy in the Built Environment

Impacts of Energy Forms

The 100 largest metros of the United States comprise only 12% of the nation's

land area.47 In this regard it is unrealistic to expect renewable energy development to

remain confined to previously developed areas. Renewable energy projects need to be

sited in both developed and non-developed areas in order to provide enough renewable

energy for the future. However, in the context of the built environment, there are

demonstrated advantages of installing distributed generation that reduces peak demand

and increases energy security.

There are a number of renewable energy options that any community can

consider, however some have more impact on the organization of the built environment

than others. Biomass generation facilities for instance, exist as singular structures that

can be sited anywhere that permits industrial functions. Wind turbines obviously need to

be sited in windy areas, which immediately rules out certain places. They also carry

spatial and political implications for the built environment; the first of which is that tall

buildings can obstruct wind patterns, the second of which is that even in remote areas,

wind turbines are met with political opposition grounded in concerns over aesthetics,

real estate values, environmental degradation and public health.

Siting renewable energy projects in developed areas decreases the likelihood

that they will disrupt habitat, reduce availability of prime farmland, or contribute to

fragmented development patterns. These concerns are especially paramount in the

State of Vermont, where a high cultural value is placed on the state's natural beauty.

Negative cultural perceptions of renewable energy plagued installations of wind projects

in Vermont in recent years, and a dramatic increase of multi-acre solar arrays on open

lands is also spurring public opposition over aesthetic concerns. A study from 2014

shows that perceived risks of siting renewable energy in Vermont outweigh perceived

benefits by a factor of 2.67.48 Incorporating PV into the built environment presents less

of a perceived risk for Vermonters, and has the potential to increase public awareness

and acceptance of renewable energy projects in general, and to enhance the state's

capacity to transition toward its goal of 90% renewable energy by 2050.

Solar thermal and solar PV can be integrated directly into the built environment,

and their installed efficiency depends heavily on spatial factors such as land use

considerations for open space, building height, orientation and roof integrity, and

shading caused by other buildings, trees and topography. Solar installations on

buildings are generally innocuous and tend not to attract the political opposition of more-

visible sources of generation. Paradoxically, solar in the built environment-especially

building integrated PV-presents less of a political risk, but is more-dependent on the

spatial configuration of buildings for its efficient operation. With respect to these

assertions about renewable energy generation, not only is solar PV more likely to be

accepted in developed areas, but its implementation carries greater implications for the

spatial reorganization and adaptation of the built environment.

Generating Energy in the Built Environment

At the municipal level, there are certain advantages to installing generation

capacity behind the meter, within the boundaries of the municipality. If a resource is

installed behind the meter (BTM) it contributes directly to reducing the load within a

distribution system, it is not technically counted as capacity and doesn't enter the

regional transmission grid. The primary impact that BTM generation can have for utilities

like BED is in reducing the peak load. By reducing the peak load, the capacity obligation

to the regional grid (ISO-NE) can be reduced, thereby reducing the amount of capacity

a utility needs to purchase from the FCM. 49 In this sense, utility-owned and operated

BTM installations can provide revenue through sales of electricity and potentially RECs,

while also reducing capacity obligation expenses. The installation of solar PV as BTM

capacity can have an added advantage as the peak-energy produced by solar PV often

aligns with the peaking demand of municipalities in warm summer months (when there

is more AC use, but also more sun). With that said, officials at BED have acknowledged

peak consumption may be shifting to later in the day, in which case the role of solar PV

in reducing peak demand would need to be integrated with storage solutions. Other

arguments for BTM PV include the generation of local jobs, energy security, and long-

term economic benefits for the host community and less strain on transmission grids

during peak hours.

Barriers in the Built Environment

Despite the various arguments for increasing renewable energy generation within

the built environment, numerous barriers exist in terms of cost, operational concerns,

policy, and collective action. Even with dropping installation costs, and the availability of

financial incentives, installing solar PV still requires a significant amount of planning and

access to capital. Solar PV is typically a long-term investment, the economic returns on

a system depend on a multitude of factors, including electricity consumption, cost/kWh,

local incentives, and the size of the system. The ways in which ownership and benefits

from a project are structured have implications for investors, utility companies and

customers. There can also be tension between state-level net-metering regulations and

the actual market value of PV for utility companies.

For instance, if BED only values PV at 12-14 cents per KWh, but has to pay net

metered customers 19-20 cents per KWh, it would be more beneficial from the

perspective of the utility to pursue municipally owned solar development rather than

community solar which would take advantage of net metering incentives.50 Under

Vermont's current net metering laws, customers maintain the right to retain RECs or

turn them over to the BED. Customers typically do not turn the RECs over, which raises

questions about how RECs should be managed and where they can provide greatest

benefit.51 Future policy decisions affecting RECs, net metering and tax incentives will

play a major role in directing where new resources are sited, how they are operated,

and who benefits from their installation.

Operational & Technical Concerns

Solar PV projects on roofs of existing buildings present a variety of operational

concerns for building owners. A major concern voiced by facilities managers of buildings

owned by the City of Burlington has been roof maintenance. Once a solar system is

installed on a roof, maintenance of the roof becomes more difficult, and the logistics of

replacing a roof become more complicated. An ideal scenario for retrofit applications is

installing PV over a new or well-maintained roof that will not need major overhauls

during the lifecycle of the PV installation which is around 30 years.

A secondary concern is with the method of attachment. There are two primary

ways of installing roof mounted PV, the first involves mechanically fastening the system

to a building's roof structure by penetrating the roof membrane. Despite successful

installations with leak-free track records, this method adds risk from the perspective of

facility managers. At the scale of an individual PV installation, it is easy for these

concerns to overshadow the margin of return on the investment. A second method of

securing PV systems on roofs is by using ballasted racking, but this is only a viable

solution on flat to low slope roofs with sufficient structural capacity for the added weight.

Both methods of installation bring up issues of liability for roof maintenance and PV

system maintenance. These types of agreements can be complex and act as deterrents

for building owners as well as installation companies. Addressing these concerns will be

an important component of community outreach, in order to help overcome hesitation

about installing solar on a large number of roofs.

Collective Action & Project Scale

In 2009, Burlington voters approved a tax-exempt bond for improvements of the

city electrical system, including $4 million for renewable energy development. In 2013,

the city released an RFP for solar on approximately 30 municipal structures. Of these

potential sites, only the eight largest received proposals, and only the airport garage

installation was executed. According to city comments, only sites with the highest

potential ROI received bids, and of those that came close to development, operational,

aesthetic and financial concerns won out over renewable energy. These concerns either

eliminated sites entirely, or reduced the site area to a point where the economics only

broke-even, according to the city. The RFP for solar on municipal buildings, while

largely unsuccessful from an installations perspective, should be seen as an opportunity

to rethink how and where municipalities promote solar PV. This brings up barriers to

collective action. Although Burlington's municipal solar RFP was a form of collective

action, the economics were still focused on the scale of individual buildings. At the

building scale, the size of the array needs to be a certain proportion to the total square

footage of the building in order to pencil out economically. In the built environment, this

ratio makes solar unattractive for many multi-story buildings, where the ratio of roof area

to building area is relatively low. Another barrier to collective action in cities is the added

complication of rental properties-where building owners who don't pay electricity bills

see little incentive in installing rooftop solar or improving efficiency. Over 40% of BED's

residential customers live in rental properties, and 65% of their commercial customers

lease their space.

Despite these challenges, even if the success rate of installing solar on all of

Burlington's 10,000+ buildings was the same as the city's municipal solar effort (1 in

30), that would equate to over 333 installations city-wide. A city-wide campaign presents

numerous hurdles in terms communication, governance, economics and physical

planning. On the other hand, casting a wider net should reveal more opportunities,

engage more political and economic capacity, have the potential to make a bigger

impact and gain more public support. To be successful, such a campaign needs to

prepare innovations in existing operational, ownership and planning models; it needs to

engage a broad base of community members, and it needs to develop an actionable

plan that addresses the spatial, political, and economic potential of a city-wide initiative.

Building on the barriers and opportunities for solar in the built environment, the

following section investigates precedents in operational models for community solar,

public outreach and education campaigns focused on solar, as well as a variety of

municipal policy approaches for regulating and promoting solar development. This base

of information can be adapted in order to address specific concerns and opportunities in

Burlington.

Precedents

Community Solar Project Models

In 2011, the National Renewable Energy Lab (NREL) published a guide to

project development for community solar, and defines community solar as: a solar-

electric system that, through a voluntary program, provides power and/or financial

benefit to, or is owned by, multiple community members. NREL cites an internal study

finding that only 22-27% of residential rooftop area is suitable for solar in the United

States, and acknowledges that community solar only makes up a portion of the total

solar market.53 The guide focuses on projects designed to increase access and reduce

up-front costs, achieve economies of scale, optimal project siting, increased public

understanding of PV, generation of local jobs, and opportunities for new models of

marketing, financing and service delivery. Before addressing community solar project

models, two other common mechanisms for deploying solar are discussed, Power

Purchase Agreements (PPAs) and Solar Services Agreements (SSAs). A PPA is an

agreement between a wholesale energy producer and a utility, where the utility

purchases power at an agreed upon rate for a specified time period. A SSA is an

agreement between the owner of a solar system and the owner of the site where the

system is located; the system owner designs, installs and maintains the system under a

contract with the system host. The following models for community solar build on the

basic components of PPAs and SSAs.

NREL point to three primary models for community solar: the utility sponsored

model, where the utility owns or operates a project that is open to voluntary ratepayer

participation, the special purpose entity (SPE) model where individual investors create a

business and develop a community solar project, and finally the non-profit model, where

donors contribute to a community solar project owned by a non-profit corporation. Each

model serves a different set of financial and needs, and can be applied in a variety of

contexts, not necessarily exclusive of one another.

In a utility sponsored model, participating ratepayers are awarded production

incentives, the value of energy and in some cases RECs, in exchange for either an

upfront or ongoing monthly payment, or credit on their electric bill which is proportional

to their contribution and the energy production of the project. The customer has no

ownership stake in the project, but rather buys rights to the benefits of the system.

While public utilities are unable to take advantage of tax credits, the ease of

participation in these models can sometimes offset those benefits through higher

enrollment and managerial efficiency.54

SPE models are designed take advantage of the significant benefits provided by

the federal investment tax credits, but in doing so bring a higher level of contractual

complexity to community solar. Most participants in community solar projects don't have

passive income, and therefore can't take advantage of the tax credits. Developers of

SPE models need to find key investors with passive income and large tax appetites,

such as building owners with large amounts of rental income. NREL describes three

financing structures for SPE models that attempt to maximize the benefits of the federal

tax incentives: self-financing, flip structures, and sale / leaseback. A self-financed model

is the simplest structure and grants ownership to the community of investors, however

its success hinges on securing an investor within the community with a large tax

appetite. In a flip structure, the community SPE would partner with a tax-motivated

investor to form a new SPE that would own and operate the project. This new SPE

would provide most of the initial equity, and would also receive up to 99% of the project

benefits for the first five years. After this investor recouped tax benefits and achieved an

agreed upon rate of return, the community SPE would then have the right to purchase

the project at fair market value. The sale / leaseback approach requires the community

SPE to host and or develop the project then sell it to a tax investor and lease it back

while continuing to operate and maintain the project. This option also includes the

option for the community SPE to purchase the project from the tax investor, after the tax

benefits have been monetized. Each of these models relies extensively on complex

legal, financial and tax regulations, and as such demand additional capital to pay for

consultants.

A non-profit model isn't able to provide electricity directly to donors who write off

their contributions as tax-deductible. However, there are precedents of non-profits

teaming up with third party for-profit entities that can install and own the system to take

advantage of investment tax credits. Although non-profits are not eligible for tax

deductions, they can qualify for other sources of funding from foundations that are

inaccessible to businesses. It is also important to note that contributions to a non-profit

are only tax deductible at whatever percentage the donor's tax bracket specifies, so the

potential benefit is much less than the investment tax credit.56

Officials in Burlington's department of public works and BED have indicated that

there is some interest in community solar, but that there hasn't been a concerted effort

to analyze or implement it yet. In Burlington's RFP for municipal solar, several proposals

for community solar were submitted. The proposals were for the developers to finance

and install systems on space leased from the city, and to then sell the benefits of

individual panels to BED ratepayers. For the city, it was unclear what additional

advantage this model provided over other proposals for PPAs which distributed benefits

equally to all ratepayers. The city also expressed issues about what would happen if the

proposals were under-subscribed. Based on the NREL analysis of community solar

models, appears that for community solar to be effective, it needs to be implemented at

a certain scale, and engage investors with large tax appetites. Based on the limited

scale of the RFP for municipal solar in Burlington, it is unlikely that either of these

parameters for success were fully addressed.

In an effort to mitigate the complicating factors of community solar, the residents

of Portland, Oregon started a campaign with a local neighborhood organization called

Solarize Portland. The first solarize campaign in 2009 was wildly successful, and now

cities around the United States have implemented similar campaigns.

Solarize Campaigns

The first solarize campaign was started in southeast Portland, Oregon, and

eventually spread to the entire city. As the program grew, it drew the support of the U.S.

Department of Energy (DOE) Solar America Communities Program, the City of Portland

Bureau of Planning and Sustainability, Solar Oregon, and the Energy Trust of Oregon.57

Solarize published a guidebook in 2011 that outlines the basic components of a solarize

campaign, and provides statistics about the success of Solarize Portland. The basic

model is designed to overcome cost, complexity and customer inertia. To address cost

concerns, the campaign provided a summary of all state and federal incentives and

rebates, and used this as marketing material to generate project leads which in turn

drove contractor marketing costs down by as much as 35%. To reduce complexity a

neighborhood committee pre-selected a contractor through competitive bidding and

eliminated the number of choices for individual consumers. The typical sales cycle in

Portland at the time of the first campaign was 2 years from first inquiry to installation; by

introducing a limited time offer for group action the campaign made consumers feel

more comfortable collectively and reduced the cycle to 3-6 months.58 The first campaign

signed up 300 customers, installing solar on 130 homes to provide 130 kW of PV

capacity in the first 6 months, which created 18 full-time professional wage local jobs.

Between 2009 and 2011, Solarize Portland added over 1.7 MW of PV capacity on over

500 homes throughout Portland, creating 50 jobs and helping to spur market growth in

independent PV installations.59

The solarize guidebook emphasizes the importance of competitive contractor

selection, community-led outreach and education, and a limited-time offer as key factors

for the success of any solarize campaign. The guidebook outlines a process to guide a

neighborhood or city through a collective purchase program, and offers

recommendations for campaign partners. This begins with cultivating awareness

through grassroots communication efforts, and then education programs are offered

throughout potential neighborhoods to provide opportunities for community members to

ask questions and voice concerns. Following community outreach efforts, signup is

facilitated through an online system, and individual sites are assessed by the installation

contractor who provides estimates to each enrolled member. After the site assessment

potential customers decide whether or not to accept the contractors offer, and

contractors can offer added incentive by providing a discount if volume targets are met.

To engage the community in the initial phases of the process, the guidebook

recommends partnering with an established non-profit organization with a history of

providing benefits in the concerned areas. The guidebook also identifies the need for a

technical advisor to ensure quality and walk participants through technical hurdles; in

Portland the Oregon Energy Trust filled this role. A project organizer is also essential for

coordinating outreach, overseeing the project timeline and streamlining permitting, for

Solarize Portland the city government filled this role.

In 2010, the Vermont Public Research Interest Group (VPIRG) created a VPIRG

Energy subsidiary to start a similar group purchasing program. VPIRG used its existing

network of members to initiate education and outreach, running 10 campaigns in 5

Vermont communities which eventually lead to the installation of 60 solar PV projects

totaling 300 kW of capacity.60 VPIRG used a $0.25/watt lead-generation fee to recoup

its operational costs. In 2012 VPIRG Energy finished their campaigns and started a for-

profit venture Suncommon, which provide homeowners with a solar lease for zero

money down. The fixed-rate lease payment is around the same as the homeowner's

current electricity bill, and operates under the premise that as electricity prices rise,

customers will see savings in their electric bills. Since 2011, 18 states have run 175

Solarize Campaigns, with Connecticut, Massachusetts, Oregon and Vermont leading in

total campaigns.

Solarize Campaigns38

29

241

13 15 1

6 8

1 1 1 1 1 2 4

CO MD PA RI VA UT NY TX WI NH NC CA WA DC VT OR MA CT

Figure 11: Solarize Campaigns by StateSource: Solar Outreach, sponsored by: United States Department of Energy, SunShot Program, 2014.

Given the demonstrated commitment to renewable energy and solar PV on behalf of the

city of Burlington, and the potential benefits of community solar projects, it makes sense

for Burlington to investigate the potential of a large-scale solar PV campaign. For the

city to assess the impacts a campaign, spatial analysis of the built environment can

provide basic parameters to guide targeted policy and community outreach as part of a

comprehensive planning vision.

Solar Zoning Models

Another way municipal governments have encouraged or accommodated solar

projects is through zoning models. Arguments for incorporating solar rights into zoning

emerged in the United States as early as the 1970s.6 1 These arguments draw on the

incorporation of the doctrine of ancient lights into common law. Under this doctrine, if an

individual had light or air coming into their window for a period of 20 years, a neighbor

could not block that light or air.62 In the 1970s the United States had no legal provisions

guaranteeing property owners access to the sun; this became more of a concern in the

41

context of passive and active solar technology, and the energy crisis of the early 1970s.

A proposal for a model solar zoning ordinance developed in 1978, includes definitions

for solar collectors, and draws on the concepts of beneficial uses from water law to

protect but not guarantee solar rights. The ordinance also allows local governing bodies

to determine the amount of acceptable shading of neighboring lots. The ordinance

incorporates a position for a city forester to assess and enforce vegetative shading

violations, but it also includes variances when literal interpretations of the ordinance

would result in unnecessary hardship. Another key component of the proposed

ordinance is that it allows for the transferability of solar rights by either sale or gift.

Since the 1970s, solar ordinances have become more common, and cities

around the world have developed ordinances to suit their particular geography and

structure. As these ordinances become more common and accessible, cities borrow

and adapt existing ordinances to suit their own needs. Israel was an early adopter of

solar hot water technology (SHW), and started mandating these systems in the early

1980s.

In the 1990s, Berlin developed a solar mandate model which was unsuccessful in

Germany, but was eventually adopted by the city of Barcelona, Spain in 1999.63 The

Barcelona ordinance targeted homes, hotels and gyms, requiring that 60% of their hot

water be supplied by SHW. By 2004, 11 Spanish cities including Madrid had adopted

the ordinance, and in 2006, Spain made it a national law.64

China has also adopted national laws which require the adoption of SHW

systems. The laws require local government to develop policies geared toward

integrating SHW; they require real estate developers to make provisions for solar

energy, and they require residents of existing buildings to install SHW if deemed

feasible.65 In areas with high levels of insolation, China is requiring buildings that use

intensive amount of hot water, like schools and hospitals, to gradually incorporate SHW,

and for new construction to at least provide adequate space for future installations.6

In the United States, the American Planning Association (APA) provides

extensive information on planning and zoning for solar energy. This information is

compiled in an essential information packet, and available for download on the APA

website; the packet contains over 100 municipal solar policies from municipalities

across the U.S. 67 The frequency and focus of these municipal solar policies shows

growth in solar policy over the last 15 years, as well as a shift toward more proactive

and even regulatory solar policy. The packet includes information on solar in

comprehensive plans, subarea plans, functional plans, permitting guides, solar

ordinances, easements, access/protection, orientation and siting, as well as mandatory

solar installation. The dates and focus areas of the sample policies indicate a major

increase in solar policy starting in 2008. The focus of these policies is primarily on the

definition and inclusion of systems in zoning ordinances, but in recent years these

policies have expanded to include definitions for multiple scale systems, solar ready

homes, and more-specific policies for streamlining permitting, optimizing siting and even

requiring solar systems. These policies range significantly in scope, but between 2012

and 2014 there is an increase in model solar ordinances that attempt to incorporate

multiple features of these policies. Figure 12 shows the frequency and type of policies

sampled by the APA from 1998 to 2014.

U.S. Solar Zoning Policies30

mandatory systems (1)

25 subarea (3)a easement / access permits (5)

20 a permitting guides (6)* model ordinances (6)1 multiple scale systems (6)w solar ready homes (7)a orientation and siting (8)z functional (9)a access protection (10)a comprehensive (12)

0 27 12accessory systems (12)1998 1999 2000 2001 2002 2003 2D04 2005 2006 2M MM 2M 2MO 201 2012 2013 2014

Figure 12: U.S. Solar Policies by Type and DateSource: Sample from American Planning Association, Planning and Zoning for Solar Energy Packet, 2014.

As solar policy is becoming more progressive and common in the United States, there

are now ample policy-precedents for cities like Burlington to draw on, and to further

develop its own municipal ordinances in direct support of solar energy. While solar is

more frequently mentioned in comprehensive master plans, these plans typically only

state that renewable energy is a goal, and that solar is one way to reach that goal, there

is generally a lack of specificity as to the specifics of how exactly solar systems will be

incorporated in the comprehensive plans. Functional plans start to get more specific,

focusing on particular buildings and setting general capacity targets. Permitting policies

tend not to focus on the larger picture of municipal solar, but rather on facilitating

individual installations by providing definitions, and general guidelines on calculating

capacity, finding local installers and submitting the appropriate paperwork and fees.

Solar access and easements provide a model for property owners to develop solar

envelopes without causing hardship to their neighbors, this starts to address the social

and political complications that can arise with adjacent ownership and vegetative

44

MOMMMMMEMMMMMOMr- - -w- IIIIIII.Ix- - ---

shading. Solar access protections take these easements to the next level by developing

setbacks and regulations for the physical properties of structures as well as vegetation.

Policies focusing on orientation and siting focus primarily on new development,

and require that its layout maximizes solar exposure at the neighborhood scale as well

as the scale of individual buildings. It also regulates planting choices and even

vegetation management. Policies for solar ready homes address both new and existing

structures, by requiring all construction and significant renovation to at least provide a

roof layout for solar PV or SHW, and to provide plumbing and conduit for eventual

installation. This type of policy has been developed in Arizona, California, Florida,

Nevada and Virginia. Building on solar preparedness, in 2014, the city of Lancaster,

California, introduced a zoning ordinance requiring a minimum of 1.5 kW of solar PV for

all new single family homes. The ordinance allows this provision to be met through

either installing a system on the home itself, or through the documentation of a solar

purchase agreement. 8

Model ordinances attempt to synthesize the best practices of these various

plans, and provide general language that municipal government can adapt to their own

contexts and specific goals for solar. These ordinances tend to focus on the definitions

and regulation of projects that will be installed, and are generally more reactive than

proactive from a planning perspective. Out of the various solar policy types, solar ready

homes, mandatory systems, and site planning take on a more active role, but still focus

primarily on new development or at the scale of the individual building. All of these

efforts are useful and important, but there are opportunities for municipal government to

strengthen these efforts by incorporating spatial analysis of the built environment in

order to target these efforts at a larger scale.

The 2014 edition of Burlington's Comprehensive Development Ordinance (CDO)

is the closest the city comes to providing specific guidelines for solar development.

The CDO starts to address the spatial planning implications for renewable energy and

solar PV. It encourages site plans to maximize solar access with building positioning

and open space; it requires new structures and additions to minimize shading on

existing buildings or public spaces to maximize solar opportunities; and it encourages

rooftop solar. The CDO also provides variances for renewable energy resource

structures if they can't be built to regulations, and won't be detrimental to the public

welfare.

Given the context of more explicit and aggressive solar ordinances in other cities,

there is room to strengthen the specificity and support for solar in Burlington at the scale

of the building, the block and the entire city. At the scale of the building and block,

Burlington can draw on precedents from other municipalities, but at the scale of the built

environment, Burlington's municipal government will need to explore new territory in

renewable energy planning. To strengthen and focus solar policies and project models,

cities need to understand the potential capacity of their open spaces and built

environments. There are numerous methodologies used to project solar capacity, and

cities can choose among these depending on their available resources and expertise.

The key point however, is that by projecting the potential capacity through spatial

analysis, cities will be able to make more informed decisions about renewable energy

development and policy.

Projecting Capacity

Spatial Analysis of Solar Capacity

There are multiple scales of analysis for estimating the potential solar energy of

any particular site. The primary factor is the solar insolation which is determined by a

site's latitude and longitude. It is relatively easy to find NASA-produced maps for every

country in the world showing the relative insolation of particular regions, provinces or

states. For the United States, NREL provides 10km resolution maps showing insolation

for each state. Insolation is typically measured in kWh per square meter per day. The

average insulation for Anchorage, Alaska is around 3, while in Phoenix, Arizona the

average insolation is above 6.5. Vermont's average insolation is 4-4.5. Estimating

insolation within the built environment however, entails a more complex process which

must take into account building orientation, building height, roof pitch and shading. To

negotiate this complexity, researchers have developed a variety of methodologies to

accurately calculate the potential insolation for the built environment.

In 2013, NREL released a publication reviewing the various methods for

estimating rooftop suitability for PV, including a methodology developed by NREL. The

primary methods for estimating the technical potential of PV use constant values,

manual selection or GIS-based analysis. Each method has advantages and

disadvantages: the constant value method is quick and easily computes rooftop area,

but the generalized results don't take into account local rooftop characteristics and are

difficult to validate; manual selection methods are more detailed and allow assumptions

to be made based on regional building knowledge, but these methods are time-intensive

and difficult to replicate in other regions; GIS-based analysis is also detail-specific, it

can be easily replicated or even automated, but GIS analysis is very intensive in terms

of time and computer processing capacity.69 After reviewing the literature for each

method, the NREL study concludes that there is a wide degree of variation in

recommended slope, orientation, shading and size characteristics of roofs. Based on

these conclusions, the proposed NREL method makes the following assumptions: the

slope of the roof should be no greater than 60 degrees, the orientation should range

from east-south-west, the rooftop should be in sunlight for the minimum number of

hours to achieve 80% generation based on its region, and the available roof size should

be a minimum of 10 contiguous square meters. The specific methodology of the NREL

method employs LiDAR data and building footprints as inputs for a GIS model, and uses

hill shade, slope, and aspect calculation tools to estimate the unshaded roof space and

orientation available for PV. NREL used over 200 existing PV installations in New

Jersey, Colorado and California to validate the accuracy of their model and achieved

satisfactory results. NREL is currently working to validate and improve the accuracy of

their model using 120 cities across the United States. Their ultimate goal is to use

statistical methods to generalize the results of their model to other areas in the United

States and to provide a national database of total rooftop space for PV.

Pending the release of a comprehensive national database, organizations around

the country have undertaken efforts using similar methodologies, including Mapdwell,

and the Renewable Energy Atlas of Vermont. Both of these projects use an online

mapping platform to project the potential solar capacity of rooftops in a given city.

Mapdwell uses digital elevation data and methods analogous to NREL's approach, and

currently provides data for the cities of Boston, Cambridge and Wellfleet

Massachusetts, as well as Washington DC, Washington County, Oregon, and Vitacura,

Chile. The Renewable Energy Atlas of Vermont takes more of a hybrid approach that

incorporates constant-value regional data with GIS technology, and provides data for

300,000 sites in Vermont's E-9-1-1 database. Both online systems are equipped with

slider-bars that allow users to change the amount of PV installed to estimate cost

differentials, incentives, and payback periods. The systems also provide links to local

solar installers as well as state and municipal level resources for parties who wish to

initiate installations.

An Alternative, Targeted-Model

While these models succeed as educational tools for the general public and

motivated building owners, they don't explicitly target any particular portion of the built

environment, such as clusters of high-potential parcels. Municipalities don't employ

these tools as part of specific planning initiatives. For the most part, these broad tools

allow individuals to assess isolated instances of PV capacity, while ignoring the

potential for collective action afforded by community solar models and group purchasing

campaigns like Solarize. In light of this gap between technological capacity and the

development of specific planning visions, the following spatial analysis is geared toward

identifying specific zones within Burlington's built environment, in support of a targeted

plan for community solar.

This analysis synthesizes vital elements of existing methodologies for the rapid

targeting of high-capacity areas. The straightforward and efficient model for assessing

municipal solar capacity and targeting outreach efforts can be developed with basic GIS

software, land use data and public outreach efforts. The model incorporates flexibility for

increasing the spatial resolution of its data, and it can also be applied to other

municipalities. The primary function for the model is to identify areas with high solar

capacity and their associated stakeholders, and to organize this data in order to initiate

a public planning process.

Spatial Description of Burlington

Burlington is Vermont's most populous city, and the surrounding metropolitan

statistical area houses approximately one third of Vermont's total population.70 The city

of Burlington's population is approximately 42,000 people, or 6.7% of Vermont's total

population. The municipal border of Burlington contains approximately 15.3 square

miles, of which 10.6 square miles is land area, and 4.7 square miles is the portion of

Lake Champlain designated as Burlington Bay. The land area of Burlington is a

crescent shape that opens to the southwest around Burlington Bay. The crescent shape

is approximately 6 miles long by 2 miles wide, and slopes gradually upwards to the east,

rising to 200 feet above the lake at the city's eastern border. The land area of Burlington

can be categorized by six primary functions as a percentage of total land area: open

space 31%, residential 30%, transportation and utility 17%, institutional 11%,

commercial 6%, and public 4%71 mp n

These six functions can be distilled into two primary categories: open space at

31% total land area (including various open space functions), and developed space at

69% total land area (including residential, transportation and utility, institutional,

commercial and public use parcels). The majority of open space is located just above

the midpoint of the crescent, and separates the developed space into two zones. The

northern zone is comprised primarily of lower-density residential parcels of .25 acres

and larger. The southern zone includes the majority of Burlington's commercial,

institutional and public structures, and has a higher residential density with smaller lot

sizes. This zone includes the downtown and university areas; it also houses 76% of

Burlington's total population, and contains 67% of the total buildings.72 Another

distinguishing characteristic is that the southern zone is built on a street grid that runs in

cardinal directions, while the street grid in the northern zone is rotated at a 450 angle.

These observations about Burlington's built environment help to set up a basic

framework for a more-detailed analysis of the potential area suitable for solar PV. The

next level of analysis addresses the specific geometries of the individual buildings

based on structure-type and solar orientation.

Spatial Analysis of Structures

In 2013, collaboration between Open Street Maps contributors and the City of

Burlington produced a public dataset that provides digital vector files for 10,066 building

footprints within Burlington. By overlaying aerial photography with the building footprints

and land use designations in GIS, it is possible to isolate individual buildings by land

use, and to determine the basic geometry of the building rooftops. Derived from this

data are the following characteristics about rooftops: Structures within commercial,

public, institutional, transit and utility land use designations have primarily flat roofs,

while structures within residential designations have primarily pitched roofs. The

designation of a flat versus pitched roof allows the building footprints to be divided into

two distinct groups.

Buildings with flat roofs are less-dependent on their solar orientation when it

comes to suitability for accepting solar panels. At Burlington's latitude, buildings with

pitched roofs need to be oriented as close to south as possible in order to achieve

maximum insolation. As one of the leading solar installers in Vermont, Suncommon

states that an orientation up to 150 east or west of true south has minimal impact on

efficiency, and merely favors morning or afternoon insolation.73 This is corroborated by

the United States Green Building Council's LEED rating system, which encourages

individual buildings as well as city blocks to be oriented within 150 of south, to optimize

passive and active solar opportunities.74 The data derived for roof pitch, building

orientation and land-use type is summarized in table 1 below:

Type No. Buildings GSF (Footprint)

ResIdentiall Total 3,089 4,066,62,5Semi-Optimal: SE 30*-15* 305 401,568Semi-Optimal: SW 30*-15 320 437,446

Optimal: SE 15*-00 1,896 2,450,776Optimal: SW, 150-O0 568 776,835,Commercial Total 472 3,778,804PubAic Total 76 396,711Institutional Total 349 2,811,778TransitIUtility Total 131 897,127Suitable Buildings Total 4,117 11,951,045Buildings, Total 101066 20,274,235% Suitable 41% 59%

Table 1

The table includes all primarily flat-roofed structures (commercial, public,

institutional, transit and utility), but only includes residential structures whose orientation

is within 300 of south. The residential structures are further subdivided into optimal and

semi-optimal categories based on orientations between 00-150 and 15*-30* respectively.

Residential structures with orientations greater than 30* are eliminated. This yields

52

4,177 potential structures for solar, which equates to 41% of Burlington's total buildings,

and 59% of the total area of building footprints.

Roof slope is not typically a delimiting factor for installing solar panels. For

instance, Suncommon can install their mounting systems on roofs with slopes from 100-

450, which encompasses the vast majority of vernacular roofs found in Burlington. For

flat roofs, a different mounting system is used that tilts panels at around 350, which

provides optimal insolation for both summer and winter at Burlington's latitude.75 What

roof slope does impact, is the total surface area of the roof, which increases as steeper

slopes lengthen the hypotenuse. For flat roofs, the roof area is typically a 1:1 ratio

relative to the building footprint, but for pitched roofs this ratio ranges from 1:1.11 for

6:12 pitches, to 1:1.30 for 10:12 pitches. A survey of the classical architectural styles

after which the majority of Burlington homes are modeled, suggests that slopes in

Burlington range from 6:12 to 12:12.76 Assuming an average roof slope of 8:12, allows

for the calculation of total roof area at a ratio of 1:1.20 relative to the average size of

residential building footprints. Residential roof areas must also be divided by half since

only one plane of the roof can face south. Table 2 shows the estimated total roof area

that is optimal for installing solar PV based on slope and orientation:

Type GSF (Footprint) GSF (Roof Area)

Residential Total (Optimal) 3,227,61 1.,936,567Commercial Total 3,778,804 3,778,804Public Total 396,711 396,711Institutional Total 2,811,778 2,811,778Trafl ltIUtity Total 897 127 897,127Total: 11,951,045 10,324,395

Table 2

0 See maps on pages 73-75

These roof areas must be further reduced to account for obstructions from

building elements such as chimneys, dormers, and mechanical equipment. An analysis

of ten city blocks in Burlington, containing 346 residential structures showed that an

average of 33% of structures had physical obstructions reducing available roof area by

an average of 31%. The percentage of structures with obstructions is closer to 50% in

the southern zone of the city, where the building stock is older, and has a higher

frequency of dormers and more-complicated roof lines. The following table and keyed-

map show the location and results for each of the 10 city blocks:

Block Total Buildings w/ % Buildings w/ % RoofBuildings Obstructions Obstructions Obstructed

1 16 4 25% 44%2 49 10 20% 48%3 23 4 17% 29%4 51 26 51% 29%5 48 27 56% 32%6 33 18 55% 32%7 27 15 56% 28%8 23 4 17% 18%9 59 8 14% 2W%

10 17 4 24% 24%T ot: 346 120 3 3%0, 31%

Table 3

S)ee map on page 76

By using the more conservative average percentiles for the amount of roof area

obstructed (from blocks 4-7), the total residential roof area suitable for solar PV is

reduced from approximately 1.9 million sf to 1.5 million sf.

For the remaining structures, categorized as commercial, institutional, public,

utility or transportation, the primary concern for obstruction is mechanical equipment.

Given the structural similarities between these building types, this study uses a

composite constant value number derived by four separate studies of commercial

54

structures in the United States, designating 60%-65% of total roof area as suitable for

solar PV. 77 The following table summarizes the total available roof area after accounting

for obstructions from structural elements and mechanical equipment:

Type GSF (Roof Area) After ObstructionsResidential Total (Optimal) 1,936,567 1,500,000Commercial Total 3,778,804 2,456,223Public Tot l 396,711 257,82Institutional Total 2,811,778 1,827,656Transit/tit Total 897,127 583,133Total: 9,820,987 6,624,873

Table 4

There are two additional factors for consideration of roof shading: vegetation and

neighboring buildings. Roof shading from neighboring buildings is not a big concern in

Burlington, as the building stock is relatively homogenous, and buildings of similar

heights are clustered together. The majority of non-residential structures in Burlington

are multi-story and their roofs exceed the height of nearby vegetation. They are also

located in zones with a low Normalized Difference Vegetation Index (NDVI), and fewer

established trees. In Burlington, shading from vegetation is more of an issue in for

residential structures.

The residential zones to the north and south of the downtown area have a higher

NDVI and more established trees. Trees can be trimmed or removed to eliminate

shading, but the decision to do so depends on the subjective valuation of personal

property. Vegetative shading is further complicated by the fact that it can be produced

by trees on neighboring parcels, and that over the 30-year life span of a solar

installation, the extent of shading can change significantly. Rather than assign

vegetative shading a constant reduction factor, this model provides data identifying

zones with high NDVIs, so that stakeholders can determine the appropriate course of

action, most-likely avoiding heavily shaded properties.

In summary, the total non-residential potential roof area, after accounting for

obstructions, is around 5.1 million sf. The total potential roof area for residential

structures, after accounting for orientation and slope is around 1.5 million sf, but some

of this is contingent on stakeholders' willingness to alter existing vegetation. The

primary implications of this data are that the greatest opportunity for solar PV based on

available roof area in Burlington, is in non-residential structures, with 64% of that square

footage accounted for by commercial and institutional structures. With the spatial extent

of potential area for solar PV identified, the next steps are to identify the various

stakeholders associated with these parcels, and to formulate a general planning

approach. 0 See map np-age .

Spatial Analysis of Open Space

Despite the relatively large quantity of open space in Burlington, there are not

many instances where this space is likely to be suitable for ground mounted PV. The

primary factors limiting the suitability of Burlington's open space have to do with current

land uses that provide cultural and environmental value. The majority of Burlington's

large parcels of non-forested open space are either designated as wetland area, located

in floodplains, or used as agriculture. The remaining smaller parcels consist primarily of

wooded areas, parks, lake front property, cemeteries, and lawns, all of which provide

either cultural or physical obstacles for solar.

After excluding remaining open area based on size and an evaluation of their

cultural and environmental value to the city, only 10 potential sites remain for ground

mounted solar. These 10 sites comprise approximately 100 acres, or 4 million sf of

space, a significant amount as compared to the 6.6 million sf of potential roof area.

Despite the opportunity in terms of potential area, these 10 sites are likely to be more

difficult to develop than rooftops. This is due to the reality that ground mounted systems

are more visible than rooftop systems, more subject to vegetative shading, and they

also prevent sites from other potential uses during the 30-year lifespan typical of solar

installations.M See n page 7

One of the largest potential sites is the retired landfill owned by the city. This site

seemed like a prime candidate for a ground mounted system, and was included in the

city's efforts to develop solar on city-owned property in 2013, but encountered

community resistance due to its high visibility.78 There are also risks with mounting

systems for landfills; pole mounted systems can penetrate the landfill membrane, and

ballasted systems can cause uneven settling which can also jeopardize the membrane.

Ground mounted systems can take a long time to go through permitting, as is evidenced

by the proposed South Forty solar project, which has been under development since

BED released a memorandum in October of 2013, stating it had reached a conceptual

power purchase agreement with the developer, and that the project would be beneficial

to BED customers. Given the demonstrated and potential barriers that are unique to

installing ground mounted PV in Burlington; it may make sense for the city to pursue

these systems separately from roof mounted systems.

Applications for Municipal Planning

The Role of the Municipal Government

Municipal government can play a significant leadership role in directing

development and organizing stakeholders, but those efforts need to be coordinated

around quantifiable resources. Open or developable land is an obvious resource for

many municipalities, and the focus of various municipal zoning and development

policies is directed at regulating how these resources are used. Existing buildings and

their rooftops are less commonly viewed as potential municipal resources, but in the

context of renewable energy development, municipalities need to expand the extent of

how they understand and define their resources. This is especially the case in land-poor

municipalities like Burlington, where building rooftops represent the greatest opportunity

for solar development. The mapping exercises undertaken in this project provide

Burlington with the necessary data to incorporate the renewable energy potential of the

built environment into a new definition of municipal resources. With this new definition,

Burlington's municipal government can function as a leader to identify and convene

stakeholders, develop supportive regulatory policies for solar in the built environment,

and adopt an appropriate project development model.

Identification of Stakeholders

Burlington's municipal leadership has the capacity and the responsibility to

convene stakeholders around common interests. By synthesizing this project's spatial

analysis with municipal property records, the leadership can engage in more precise

and informed outreach. The 2014 edition of the Burlington Assessor's Office Grand List

includes 11,147 unique parcel IDs, and their associated owners, addresses, areas, and

other data. By spatially joining the address fields of the Grand List to the state's E-9-1 -1

data points, all of the current owners of high-potential structures can be organized into a

targeted stakeholder database. Using this database, the municipal government can

initiate planning and outreach efforts to engage key stakeholders. The flexibility of the

spatial model allows for adjustment of parameters to expand or decrease the scope of

the stakeholder database.

While this database has the potential to be a powerful tool for increasing PV

generation in Burlington, its functionality is contingent on the capacity of the municipal

government to use it as part of a broader solar development plan. To orchestrate this

outreach and planning effort, key city departments already engaged in energy planning

can play important roles. These departments include the Department of Public Works,

the Planning Department, and the Burlington Electric Department. Together, these

departments can work to strengthen existing energy policies around specific regions of

the city, while incorporating community input from those who would be most affected.

Strengthening Policies for Solar

The primary tools by which municipalities govern the built environment are

through regulatory policies such as zoning ordinances, building codes, and incentives or

disincentives created through taxes and development rights. These tools have been

used in various combinations to address land use issues ranging from sprawl reduction

to urban land management. 79 The question for the city of Burlington is how these policy

tools might be applied to increase solar PV on existing structures, or at least prepare

these structures for future installations. There is no question of Burlington's commitment

to supporting local renewable energy projects, including solar PV. These goals are

stated multiple times in BED's Integrated Resource Plan, and in Burlington's Municipal

Development Plan, Climate Action Plan, and Comprehensive Development Ordinance

(CDO or Zoning Ordinance).

Of these plans, the CDO is the one that most explicitly engages design and

planning in the built environment, and it provides an existing regulatory framework for

the city to strengthen its spatial policies around solar. By drawing on best practices from

other solar ordinances, Burlington can incorporate approaches that incentivize and

prepare new development for solar. To address existing development, identification of

solar-optimal properties is a necessary first step. The identification of these properties

also lays the groundwork for a broader vision of solar development at the scale of the

city. An expanded section on solar development in Burlington's CDO could provide

better preparedness regulation and incentives at the scale of individual buildings, but

also aggregate these policies within a citywide database of solar-optimal properties.

A first step would be to use the database of solar-optimal properties to delineate

subgroups of properties based on building type and location within the city, and to

develop more focused regulation and incentives for these groupings. Even the act of

officially acknowledging these property groups in the CDO would be a success, and

help initiate targeted planning for solar development. By engaging stakeholders in these

target areas, the city can develop regulations and incentives to maximize solar potential

while addressing community concerns. Upon establishing regulatory parameters for

these property groupings or districts, the city can develop a customized project model

for ownership and financing that addresses concerns at multiple scales of development.

Developing a Project Model

Various models for developing large-scale solar projects involve different

ownership and benefit arrangements between private developers, public utilities, utility

customers and nonprofit organizations. In developing a model that works for Burlington,

the advantages and disadvantages of each model must be considered, and synthesized

with the specific opportunities and barriers for large-scale distributed solar in Burlington.

Following the spatial and economic analysis discussed in this thesis, Burlington's

commercial properties represent the majority of suitable rooftop area for solar;

commercial property owners are also more likely to have larger tax appetite, and thus

the ability to take advantage of state and federal tax incentives. Based on these

findings, it can be argued that Burlington's commercial real estate represents the

greatest opportunity for solar production, in terms of both potential area and economic

advantage. As such, it may be advantageous for Burlington to focus on developing a

project model specifically for commercial properties, and if this model is successful, it

can be expanded and adapted for institutional, public and residential properties.

In developing a project model for commercial properties, the city will need to

address the primary concerns that arose during its previous attempt at developing solar

on city-owned structures. The primary concerns that came up were economic payback,

the structural suitability of roofs, and maintenance. These concerns could be mitigated

by drawing on cost savings of bulk purchasing, streamlined contractor selection, and

guaranteed pricing strategies of the solarize model, as well as distributed ownership

and financing models.

Another approach to address the issue of payback is to decouple the installation

sites from the parties that own and benefit from the systems. In this way anyone who

wants to invest in and benefit from solar can take advantage of the most suitable

locations. This can also help mitigate concerns about the ratio of roof area to energy

consumption of specific buildings. For instance, if a large commercial property could

only recoup 5% of its annual energy consumption from an installation on its roof, it may

not be interested in the investment. However, if three residential property owners could

recoup 75% of their energy consumption from that same commercial roof (but were not

able to install solar on their own roofs), suddenly the installation has more perceived

value. Depending on the property owner, roof area could be leased, or donated to

interested parties, or incorporated into municipal regulations and incentives for rooftop

solar.

Leveraging RECs for New Capacity Planning

By synthesizing spatial analysis, policy, and innovative project models, the city of

Burlington can take steps toward developing new renewable energy within the municipal

borders. This will not only count towards overall renewable energy usage, but it has the

potential to reduce capacity obligation, provide local jobs, and further establish

Burlington as a leading proponent of sustainability. To successfully integrate all of these

components into an innovative strategy for solar development, the city will need to

invest significant resources in outreach, research and project management. While there

are numerous ways for cities to finance research and planning initiatives, Burlington

may be able to leverage its existing renewable resources to develop new ones.

As discussed previously, BED currently sells its Class I RECs to stabilize rates,

and then purchases less expensive Class 11 RECs to reclaim renewability. Since

Vermont's RPS will not be mandatory until 2017, BED may be able to temporarily

channel some of the revenue it receives from RECs to fund research and planning for

new renewables. In this way the money would still be used in support of renewability,

but in a different way. Depending on the amount of RECs used for this purpose,

Burlington's claim to renewability might drop in the short-term, but would eventually

return to 100%. Using revenue from RECs sold on the regional market to help finance

local renewable energy starts to create a positive feedback loop between existing

renewable assets and new ones. Once local renewables are developed the RECs they

produce can then feed back into the market to incrementally produce economic and

environmental benefits at both local and regional scales. For Burlington, the opportunity

for leveraging RECs is time sensitive, and would provide the maximum impact if

implemented before state and federal ITCs expire at the end of 2016, and before

Vermont's RPS becomes Mandatory in 2017.

By acting soon and assuming a leadership role, the city can leverage more

financial resources toward engaging stakeholders in developing policies and project

models for high-potential solar zones. If this approach succeeds at improving renewable

energy planning in Burlington, it holds the potential to serve as a model for other

communities and utility companies interested in developing local renewable energy.

Conclusions

The preliminary research presented in this thesis establishes a trend toward

urban sustainability and renewable energy, as one of many responses to mitigating the

impact of humans on the environment. It also juxtaposes the current deficit of current

global capacity in renewable energy, with the rapid growth of new wind and solar

capacity in recent years, as an opportunity for municipal government to incorporate

renewable energy production into the built environment.

The primary research undertaken in this thesis addresses the potential for

strengthening the role of municipal government in renewable energy planning,

especially for solar PV. The thesis focuses on the city of Burlington, Vermont, as case

study for existing leadership in the transition to renewable energy. It investigates the

contributing factors to Burlington's success, and also identifies opportunities for

Burlington to improve its already impressive track record of supporting renewable

energy.

This thesis advocates for synthesizing spatial analysis and regulatory policies

within the built environment, so that municipal governments can support renewable

energy development in more calculated and spatially deterministic ways. In order for

this synthesis to occur, municipalities need to reframe the way they define resources by

incorporating the built environment into these definitions. Following the synthesis of

these factors, municipal government must take leadership in convening stakeholders to

realize the opportunities for producing renewable energy in the built environment.

Findings & Products

This thesis finds that Burlington's ability to claim renewability for the electricity it

uses, benefits from the value of RECs created by the adoption of renewable portfolio

standards in neighboring New England states. In this sense, there is reciprocity

between the notion of self-sufficiency provided by local renewable energy production,

and the interconnection of electricity transmission and the market for RECs at the

regional scale. This is significant in the potential for municipalities to leverage the

regional value of renewable energy, to develop new capacity at a local level, which in

turn provides regional environmental benefits, while also-in the case of solar-

reducing peak demand. As renewable portfolio standards become more stringent in

New England, the demand for new renewable energy capacity will increase, forcing

municipalities and utility companies to find ways of importing and producing renewable

energy.

To promote new sources of renewable energy in Burlington, this thesis provides

a GIS model incorporating environmental, structural and demographic data that

identifies and categorizes individual properties and structures based on their potential

for generating power through solar PV technology. The GIS model is paired with a

selection of precedents in solar project development structures, solarize campaigns,

and municipal solar zoning policies. These materials are presented to the city of

Burlington as a collection of base materials, intended to instigate a conversation about

how regulatory policy can incorporate quantitative spatial analysis to prepare the built

environment for future development of renewable energy.

65

Implications

The findings and products of this thesis carry theoretical, technical and

practicable implications for the future role of municipal government in renewable energy

development. Conceptually, municipal government needs to reevaluate the existing built

environment as a framework for the installation of new renewable energy capacity,

especially solar PV. This will necessitate the adaptation of existing understandings and

institutional arrangements around the use of private property for a public benefit (i.e.

roofs and renewables). Cities also need to develop strategies for balancing the costs

and benefits of local energy production within the context of regional transmission,

capacity obligation and demand for renewable energy.

From a technical perspective, the tools required for formulating more specific

spatial models of renewable capacity, for developing community solar project models,

and for legislating targeted renewable energy policies in the built environment, already

exist. The challenge is in the capacity of municipal government to synthesize these tools

and opportunities, and to assume a leadership role in their practical application.

The practicable implications of this thesis rest in the future incorporation and

continuation of this research by the city of Burlington. At a minimum, this research

provides a spatial and political framework for Burlington to identify specific zones within

the municipality as having high or low solar energy potential. At a maximum, the city of

Burlington will succeed in strengthening its support for renewable energy, and inspire

other cities to follow suit.

At a disciplinary level, the approach to renewable energy planning outlined in this

thesis argues for a shift in the dynamic between cities and energy. It seeks to add value

to urban areas by adapting them produce energy, in order to mitigate their role as the

world's primary drivers of energy consumption. Addressing this issue in the existing built

environment confronts the social, cultural, economic and political centers of civilization

as drivers of change, and as examples for future urbanization.

Limitations & Further Research

The implications around the opportunities and barriers for strengthening the role

of municipal government in renewable energy development, address a multitude of

complex economic, political and spatial issues. As such there is ample room for future

research on each of these issues. It is the intent of this this thesis, to provide the city of

Burlington with the digital files for the spatial analysis conducted herein, so they may be

used as base documentation for further research, such as:

* A deeper analysis of community solar financing and ownership models throughcase studies, and the investigation of how they can be specifically adapted toprovide the most benefit for the city of Burlington.

" Further calibration of the GIS model provided, to account for the latest changesin Burlington's built environment, and to incorporate more local knowledge on thesuitability of specific areas.

" A comprehensive economic analysis of the potential impacts of a large-scalecommunity solar project on Burlington's net-metering program, peak load, andcapacity obligation to ISO-NE.

" Developing a series of solutions to address concerns about the structuralintegrity of existing roof area, its durability and maintenance issues. Thesesolutions should be both design and finance oriented in order to prepare theexisting roof-scape to successfully receive PV.

* Engaging community members from the outset, in order to identify questions andconcerns, develop a project vision, and build political support for incorporatingrenewable energy in Burlington's built environment.

The research and potential applications outlined here are not intended to provide

a comprehensive planning approach for renewable energy in the built environment.

These approaches must be developed by their specific communities and municipal

leadership. The case studies of project development and policy examined here are not

exhaustive, and the spatial analysis performed can be refined to more accurately

identify high potential areas. With these limitations in mind, the findings presented in this

thesis provide a resource base for the city of Burlington to begin a conversation around

how it will prepare its built environment to accommodate renewable energy in the future.

Reference Maps and Diagrams: Pages 69-78

t~)

Supply Resources for 201640.2%

10.7%

9.5%

8.2%

7.6%

7.0%

3.8%

2.0%

1.6%

OMes 100 200 300 400 500 600so mm Minimum Distance---------------------- ----------------------------------- Maximum Distance

i7

BiomassWindSolarMixed Renewables

0Hydro

ISO-NECapacity Obligation85-88MW

Supply Resources+I- 380,O00MWh /year

Burlington's energyshed for electricity

69

Supply Resources for 2016

Vermont

70%266,000 MWh

McNeil BiomassVermont WindWinooski OneGMC WindBED SolarVEPPIStandard Offer

Maine18%72,000 MWh

Nextera HydroHancock Wind

Burlingtc100%

>382,000 MWh

Quebec

8%30,000 MWh

Hydro Quebec

New York

4%14,000 MWh

NYPA Hydro

ffxiu

I nio

n

70

Existing Solar PV Sites*S Roof-Mount PV Instalaton

Grund-Mount PVSl BoundaryPending Ground-Mount PV St Boundary

. Tns.. m m 3 - 6d PV ibmdW an ami g gclq dde. Aorirg b ft

Bkingom Ebm*It Dep I, Ume edumsy a 0prueft wimStw cEy.

I Mile

71

Grid Orientation & Open Space

40

I Mile

72

- , 111 MZINM=:2=-

Building Footprints

Residential

Commercial*O S..*..0**

C'

(I

Land Use

44,

Institutional /Community.i .. ?..

3,089 Potential PV Buildings1,316 Average Square Feet4,546,624 Gross Square Feet

472 Potential PV Buildings8,006 Average Square Feet3,778,804 Gross Square Feet

349 Potential PV Buildings8,057 Average Square Feet2,811,778 Gross Square Feet

73

I

Land Use

(L

I6

Building Footprints

Transit / Utility

Public /Assembly

131 Potential PV Buildings6,848 Average Square Feet897,127 Gross Square Feet

76 Potential PV Buildings5,220 Average Square Feet396,711 Gross Square Feet

74

V A

\ 1

I ---

4,117 - - - --Total PV-Sultable Structures

I Mile

75

-- I

0

/I

N

7 l~

/,

/

& ;'

9

7

Sample BlocksC3Blocks selected for analysis

of structural shading elements

Population Density / Acre0-5

- 5-1515-2525-5050-150

I Mile

76

3

2

-6

4--

t y

Suitability Zone & NVOI Shading

1i~

I1'

~ I 4 ~ ~

I Mile

77

r

--

Open Space Overlay ZonesPotential Ground-Mount PV

M wefands

Open Space TypesSurface PaddngResenoeir StructureWatersports P~fhApicultureOpen Industrialcemeteries

33GrewsForestInactive LandllVftpg GreenParke & RecreationNature Park

- Natural State / Undeveloped

2 F6

- -

1 Mile

78

0

Methodology of Spatial Analysis:

Synthetic Approach

Drawing on the NREL's 2013 publication that reviews various methods for

estimating rooftop suitability for PV, this thesis synthesized components of each method

according to requirements for expediency, practicality and data availability. To reiterate

NREL's findings, there are three primary methods for conducting a PV-capacity analysis

in urban areas: constant values, manual selection, and GIS-based analysis. Each of

method has advantages and disadvantages relative to accuracy and required inputs of

time. The goal for synthesizing these methods is to arrive at an approach that provides

accurate results in a short period of time. The primary components of the analysis

addressed the potential structures and open space available for PV.

Spatial Analysis of Structures

The analysis of structures employs GIS-based techniques and analytical

processes, manual selection of a sample, and derives context specific constant values

to determine key suitability factors. The key datasets used were building footprints, land

use, and infrared aerial photography. The primary factors for suitability include roof

slope, orientation and shading.

To determine roof slope, first individual building footprints were joined to parcels

based on land use, under the assumption that the vast majority of residential parcels in

Burlington have pitched roofs, while structures in other land use designations have flat

roofs. This assumption was then verified through visual inspection of recent aerial

photography. This broke buildings into two primary groups based on roof type, pitched

and flat.

7; q

Orientation was only a concern for pitched roofs. To determine orientation of

pitch, a constant value assumption can be made of traditional residential architecture

typical of Burlington. This assumption is that the most efficient and common way to build

pitched roofs is to minimize the span of rafters, and to orient the ridgeline parallel to the

longest dimension of the structure. Under this assumption GIS tools can calculate the

orientation of each structure, and they can be grouped according to optimal and semi-

optimal deviations from south.

As discussed in the general spatial analysis, the actual pitch of roofs is not

generally a deterrent for the installation of PV as companies can accommodate a wide

range of pitches. The most significant impact of pitch is actually on roof area. To

determine this, an average roof pitch is assumed based on manual selection methods

that draw on common architectural pitches for residential buildings found in Burlington;

this is verified through visual inspection pitches in street view. After determining the

average pitch, simple algebra (the Pythagorean Theorem, averages and area

calculations) allows this impact on area to be calculated based on average residential

building size.

More advanced GIS techniques use LiDAR data to estimate roof pitch, which

provides a high level of accuracy, but is also very intensive in terms of processing and

time. LiDAR is more effective in identifying structural shading elements and building

height, but again involves high levels of input, and methods using LiDAR for these

purposes still recommend that the structures be inspected by a verified solar contractor

prior to installation. As an alternative to LiDAR, this analysis employed manual selection

methods and took a sample of over 300 structures in 10 residential blocks throughout

8 C'

the city, and calculated the average structural shading for the sample size, and used

that average to make a general assumption for all residential structures.

LiDAR is perhaps most effective in determining Building height, especially in

urban areas with no municipal records of height, and high levels of variation in height.

Burlington does not have building height data, but based on manual selection and

verification of aerial photography, the built environment is generally homogenous in

terms of height, with taller commercial buildings clustered away from lower residential

buildings. These observations rule out building shading from other buildings as a major

factor in Burlington, but like LiDAR methods, it is still important for individual structures

to be assessed by certified solar vendors prior to installation.

Spatial Analysis of Vegetative Shading

Vegetative shading can also be calculated using LiDAR data and GIS software to

determine the height and density of vegetation. However it is also possible to make

more general observations using color infrared aerial photography and NVDI processing

to identify areas with high concentrations of vegetation and potential shading problems.

Due to the complex nature of managing vegetation in a shared urban environment,

these issues need to examined by city arborists and addressed with land owners, both

which are beyond the purview of this analysis.

Spatial Analysis of Open Space

Open space suitability was determined by comparing land use data with aerial

photography to identify non-wooded open areas. These areas were then reduced based

on ecological and cultural factors associated with existing land use, as well as an

evaluation of potential community reactions to highly-visible ground mounted arrays. In

Burlington's case, installing PV on open areas is less of a technical planning issue and

more of a cultural one, which limits the ability of GIS software to develop accurate

suitability without community input.

Stakeholder Identification

The final and critical component of this spatial analysis involves identifying the

property owners of high-potential parcels or structures. This is done by joining available

property records from the Burlington Assessor's office to the states E-9-1-1 database of

addresses, and joining these with building footprints and parcels identified as optimal.

This synthesis takes the first step in developing a potential outreach strategy.

82

Notes

1 The Intergovernmental Panel on Climate Change, Climate Change 2014: Mitigation of Climate

Change (New York: Cambridge University Press, 2014) 929.

2 The Intergovernmental Panel on Climate Change, Climate Change 2014: Mitigation of Climate

Change (New York: Cambridge University Press, 2014) 958.

3 The Intergovernmental Panel on Climate Change, Climate Change 2014: Mitigation of Climate

Change (New York: Cambridge University Press, 2014) 675.

4 Dennis Meadows et al, The Limits to Growth (New York: Universe Books, 1972) 88-128. ; Peter

Droege ed., Urban Energy Transition from Fossil Fuels to Renewable Power (Amsterdam:

Elsevier, 2008) 1-3. ; Paulo Ferrao and John Fernandez, Sustainable Urban Metabolism (Cambridge: MIT

Press, 2013) 5-7, 115-117.

5 G.H. Brundtland, and World Commission on Environment and Development. Our Common Future:

Report of the World Commission on Environment and Development (Oxford University, 1987) 3-27.;

Patrick Geddes, Cities in Evolution (London: Williams & Norgate, 1915) 60.

6 Paulo Ferrao and John Fernandez, Sustainable Urban Metabolism (Cambridge: MIT Press, 2013) 53.

7 Renewable Energy Policy Network for the 2 1't Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 21.

8 Renewable Energy Policy Network for the 2 1 't Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 3.

9 Renewable Energy Policy Network for the 2 1st Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 14.

10 Renewable Energy Policy Network for the 2 1st Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 14.

"Renewable Energy Policy Network for the 21st Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 26.

12 Renewable Energy Policy Network for the 2 1st Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 26

13 Renewable Energy Policy Network for the 2 1't Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 49 & 59.

14 International Monetary Fund, Energy Subsidy Reform: Lessons and Implications. January 28t,

2013: 1-9.

15 International Monetary Fund, Energy Subsidy Reform: Lessons and Implications. January 28t,

2013: 5.

16 Renewable Energy Policy Network for the 2 1st Century, 2014. Renewables 2014: Global Status

Report (Paris: REN21 Secretariat, 2014). 103.

17 U.S Energy Information Administration: International Enermy Statistics 2008-2013, 1, May. 2015

<http://www.eia.gov/cfapps/ipdbproject/IEDIndex3.cfm?tid=44&pid=44&aid=2>

18 Geoff Keith et al, Toward a Sustainable Future for the U.S. Power Sector: Beyond Business as

Usual. November 16th, 2011: 6.

19 Geoff Keith et al, Toward a Sustainable Future for the U.S. Power Sector: Beyond Business as

Usual. November 16th 2011: 6.

20 Geoff Keith et al, Toward a Sustainable Future for the U.S. Power Sector: Beyond Business as

Usual. November 16 ,th 2011: 8.

21 Geoff Keith et al, Toward a Sustainable Future for the U.S. Power Sector: Beyond Business as

Usual. November 16 ,th 2011: 8.

22 Geoff Keith et al, Toward a Sustainable Future for the U.S. Power Sector: Beyond Business as

Usual. November 16 t, 2011: 22.

23 Database of State Incentives for Renewables and Efficiency: 1, May. 2015 <http://ncsolarcen-

prod.s3.amazonaws.com/wp-content/uploads/2015/04/Net-Meterinq-Policies.pdf>

24 Database of State Incentives for Renewables and Efficiency: 1, May. 2015

<http://www.dsireusa.org/>

25 United States Department of Energy, Energy Efficiency & Renewable Energy Network, Green Power

Markets: 1 May. 2015 <http://apps3.eere.ener-y.qov/qreenpower/markets/certificates.shtml?paqe=5>

26 Galen Barbose et al, Tracking the Sun VII: An Historical Summary of the Installed Price of

Photovoltaics in the United States from 1998 to 2013, September 2014. 1-66.

84

27 G.R. Timilsina et al. "Solar Energy: Markets, Economics and Policies" Renewable and Sustainable

Energy Reviews 16 2 October. 2011: 449-465

28 U.S Energy Information Administration: International Energy Statistics 2008-2013, 1, May. 2015

<http://www.eia.gov/cfapps/ipdbproject/lEDndex3.cfm?tid=44&pid=44&aid=2>

29 U.S Energy Information Administration: Vermont State Profile and Energy Estimates 2015, 1, May.

2015 <http://www.eia.gov/state/?sid=VT>

30 Galen Barbose et al, Tracking the Sun VII: An Historicial Summary of the Installed Price of

Photovoltaics in the United States from 1998 to 2013, September 2014. 1-66.

31 Solar Source Residential Payback, 2012. 1 May 2015 < http://www.solarsourcene.com/residential-

payback.html>

32 Vermont Department of Public Service, Comprehensive Energy Plan 2011: 23.

3 Vermont Department of Public Service, Comprehensive Energy Plan 2011: 162.

34 Vermont Department of Public Service, Comprehensive Energy Plan 2011: 83.

3 The Burlington City Council and Planning Commission, Burlington's Municipal Development Plan,

2014:124-132.

36 Jennifer Green and Sandrine Thibault et al, Burlington Vermont Climate Action Plan, 2011: 1-25

37 Ever-Green Energy, Burlington District Energy Study, 2014: 60.

38 City of Burlington Vermont Department of Planning and Zoning, Comprehensive Development

Ordinance, 2014: 14.

39 Burlington Electric Department, Our Integrated Resource Plan, 2012: 9.

40 Burlington Electric Department, Our Integrated Resource Plan, 2012: 81.

41 Burlington Electric Department, Rule 5.2000 Resource Report January 2015 - January 2019, 3 0 th

January, 2015: 7-10.

42 Burlington Electric Department, Our Integrated Resource Plan, 2012: 82-86.

43 State of Vermont Legislature, An act relating to the Vermont energy act of 2012, 1 May 2015

<http://www.leg.state.vt.us/docs/2012/Acts/ACT170.pdf>

"Burlington Electric Department, Rule 5.2000 Resource Report January 2015 - January 2019, 30 h

January, 2015: 5.

45 Vermont Department of Public Service, Comprehensive Energy Plan 2011: 18-27.

46Due to Municipal Government's tax exempt status, cities often need to partner with private or

commercial developers who have a large tax appetite in order to take advantage of ITCs. BED did this

with its airport solar project and development company Encore Redevelopment.

47 Bruce Katz et al, "Miracle Mets: our fifty states matter a lot less than our 100 largest metro areas"

Democracy Journal Spring. 2009: 22-35

48 Elizabeth White, Renewable EnerZ Cutting Green Tape While Improvin Ecological

Outcomes for Renewable Energy Proiects MA Thesis. University of Vermont, 2014. 22-24.

49 Phone Interview with James Gibbons, Director of Resource Planning at the Burlington Electric

Department, 24th February, 2015.

5 Phone Interview with James Gibbons, Director of Resource Planning at the Burlington Electric

Department, 24th February, 2015.

51 Phone Interview with James Gibbons, Director of Resource Planning at the Burlington Electric

Department, 24th February, 2015.

52 Burlington Electric Department, Our Integrated Resource Plan, 2012: 15.

5 National Renewable Energy Lab, J Melius et al, Estimating Rooftop Suitability for PV: A review of

Methods, Patents, and Validation Techniques, December 2013: 8.

54Jason Coughlin et al, A Guide to Community Solar: Utility, Private and Non-Profit Proiect

Development, November 2010: 7.

5, Jason Coughlin et al, A Guide to Community Solar: Utility, Private and Non-Profit Proiect

Development, November 2010: 12-19.

5 Jason Coughlin et al, A Guide to Community Solar: Utility, Private and Non-Profit Proiect

Development, November 2010: 19-21.

57 Irvine et al, The Solarize Guidebook: A community Guide to Collective Purchasing of Residential PV

Systems, February 2011: 3-4.

58 Irvine et al, The Solarize Guidebook: A community Guide to Collective Purchasing of Residential PV

Systems, February 2011: 5-6.

86

59 Irvine et al, The Solarize Guidebook: A community Guide to Collective Purchasing of Residential PV

Systems, February 2011: 5-6.

6 Irvine et al, The Solarize Guidebook: A community Guide to Collective Purchasing of Residential PV

Systems, February 2011: 15.

61 Melvin M. Eisenstadt et al, "A Proposed Solar Zoning Ordinance," Journal of Urban and

Contemporary Law Vol. 15 Urban Law Annual. 1978: 211-215.

62 Melvin M. Eisenstadt et al, "A Proposed Solar Zoning Ordinance," Journal of Urban and

Contemporary Law Vol. 15 Urban Law Annual. 1978: 211.

6 Muyiwa Adaramola, Solar Energy: Application, Economics, and Public Perception, (Boca Raton:

CRC, 2014) 200-201.

6 Muyiwa Adaramola, Solar Energy: Application, Economics, and Public Perception, (Boca Raton:

CRC, 2014) 200-201.

65 Muyiwa Adaramola, Solar Energy: Application, Economics, and Public Perception, (Boca Raton:

CRC, 2014) 200-201.

66 Muyiwa Adaramola, Solar Energy: Application, Economics, and Public Perception, (Boca Raton:

CRC, 2014) 200-201.

67 American Planning Association, Essential Information Packet: Planning and Zoning for Solar

Energy, 2014, 1 May 2015 <https://www.planning.org/pas/infopackets/open/eip30.htm>

6 Lancaster (California), City of. 2014. Code of Ordinances. Title 17, Zoning; Chapter 17.08,

Residential Zones; Article 11, Non-Urban, Urban, Medium and High Density Residential Zones; Section

17.08.060, Development Regulations by Building Types. Also, Article V, Solar, Wind, and Alternative

Energy Uses; Section 17.08.305, Implementation of Solar Energy Systems.

69 National Renewable Energy Lab, J Melius et al, Estimating Rooftop Suitability for PV: A review of

Methods, Patents, and Validation Techniques, December 2013: 14-18.

70 Annual Estimates of the Resident Population: April 1, 2010 to July 1, 2014 Source: U.S. Census

Bureau, Population Division. 1 May 2015

<http://factfinder.census.gov/faces/tableservices/sf/pages/productview.xhtml?src=bkmk>

71 Chittenden County Regional Planning Commission: November 2008, 1 May 2015

87

<http://maps.vcgi.org/gisdata/metadata/DemoCensusBLCK201 0.txt?>

7 Suncommon, Building A House with Solar in Vermont, 2 6th July. 2012, 1 May 2015

<http://suncommon.com/building-a-house-with-solar/>

4 United States Green Building Council, Solar Orientation, 1 May 2015

<http://www.usgbc.org/credits/ea51>

75 Suncommon, Building A House with Solar in Vermont, 26th July. 2012, 1 May 2015

<http://suncommon.com/building-a-house-with-solar/>

76 Urban Design Associates, Institute of Classical Architecture and Classical America, A Pattern Book

for Neighborly Houses: Architectural Styles, 2007: 1-18.

7 National Renewable Energy Lab, J Melius et al, Estimating Rooftop Suitability for PV: A review of

Methods, Patents, and Validation Techniques, December 2013: 8.

78 Phone Interview with Martha Keenan, Capital Improvement Program Manager, Department of Public

Works, City of Burlington, Vermont, February 18, 2015.

791Heather Campbell and Elizabeth Corley, Urban Environmental Policy Analysis (Armonk: M.E.

Sharpe, 2012) 208.