Decarbonizing Local Mobility and Greenhouse Agriculture through Residential Building Energy Upgrades: A Case Study for Qu é bec

: Electriﬁcation is an efﬁcient way to decarbonize by replacing fossil fuels with low-emission power. In addition, energy efﬁciency measures can reduce consumption, making it easier to shift to a zero-carbon society. In Qu é bec, upgrades to aging buildings that employ electric resistance heating offer a unique opportunity to free up large amounts of hydroelectricity that can serve to decarbonize heating in other buildings. However, another source of energy would be needed to electrify mobility because efﬁciency measures free up small amounts of electricity in summer compared to winter. This study reveals how building efﬁciency measures combined with solar electricity generation provide an energy proﬁle that matches the requirements for decarbonizing both mobility and heating. The TRNSYS software was used to simulate the annual energy performance of an existing house and retroﬁtted/rebuilt low-energy houses equipped with a photovoltaic (PV) roof in Montreal, Qu é bec, Canada (45.5 ◦ N). The electricity that is made available by upgrading the houses is mainly considered for powering battery and fuel cell electric vehicles (BEVs and FCEVs) and electrifying heating in greenhouses. The results indicate that retroﬁtting 16% or rebuilding 12% of single-detached homes in Qu é bec can provide enough electricity to decarbonize heating energy use in existing greenhouses and to operate the new greenhouses required for growing all fresh vegetables locally. If all the single-detached houses that employ electric resistance heating are upgraded, 33.4 and 21.8 TWh year − 1 of electricity would be available for decarbonization, equivalent to a 19% and 12% increase of the province’s electricity supply for the retroﬁtted or rebuilt houses, respectively. This is enough energy to convert 83–100% of personal vehicles to BEVs or 35–56% to FCEVs. Decarbonization using the electricity that is made available by upgrading to low-energy solar houses could reduce the province’s greenhouse gas (GHG) emissions by approximately 32% (26.5 MtCO 2eq ). The time required for the initial embodied GHG emissions to surpass the emissions avoided by electriﬁcation ranges from 3.4 to 11.2 years. Building energy efﬁciency retroﬁts/rebuilds combined with photovoltaics is a promising approach for Qu é bec to maximize the decarbonization potential of its existing energy resources while providing local energy and food security.


Introduction
As humans strive to reduce environmental degradation, fossil fuels will increasingly be substituted with low-emission energy sources such as renewable electricity. The sectors that require decarbonization the most are transportation, which represents 45% of Québec's greenhouse gas (GHG) emissions, and the energy used to heat buildings, greenhouses, and industrial processes [1]. The transportation sector is particularly in need of transformation heating decarbonization include electrification, solar energy [9][10][11], and the use of biomass boilers/gasifiers for heating [12] or combined heat and power for lighting and heating [13], including using crop waste as an energy source [14]. A challenge with electrifying heating in greenhouses is their large thermal energy consumption and power demand during peak periods, often making the use of heat pumps cost-prohibitive. If greenhouses would switch from natural gas to electric resistance heating, the impact on grid peak power demand can be considerable, particularly for Québec, where peak demand occurs in winter due to the widespread use of electric heating. Hydro-Québec now offers incentives for demand-side management and estimates that a new power supply during peak periods will be needed by 2025 [15,16]. Therefore, redirecting the electricity that is made available from building energy upgrades to secure low-emission greenhouse operation is prioritized in this work.
For high-latitude locations, building energy efficiency frees up significantly less electricity in the summer (and shoulder seasons) because the potential energy reductions during these periods (e.g., achieved mainly by more efficient appliances, lighting, and water heating) are small compared to the potential energy savings from space heating. Since electrifying mobility requires a near-constant annual supply of low-emission electricity as vehicles travel approximately the same distance throughout the year, the potential decarbonization that can be achieved from building upgrades is limited by the relatively low amount of electricity that can be freed up by energy efficiency measures in summer. Therefore, to address the priority objective of decarbonizing the mobility sector, another source of energy (concentrated in the summer) would be needed to complement the energy that is freed up due to efficiency measures (concentrated in the winter).
This study investigates the combination of energy efficiency and solar electricity generation as a promising way to increase the potential for decarbonizing mobility in the summer. During this period, the higher solar energy generation complements the low amount of electricity that is freed up by efficiency measures, and the opposite occurs in the winter. The combo of energy that is made available by implementing energy efficiency measures and solar energy generation might provide a greater potential for electrifying constant loads (such as required for electric vehicles (EVs) and domestic hot water (DHW) heating) than would be possible if they were implemented separately. Various studies have been conducted to quantify the potential energy savings from efficiency measures [17,18], energy generation from on-site solar energy conversion technologies [19,20], and their associated cost and emissions analysis [21][22][23]. However, from an extensive literature review, there are not any studies that evaluate how the energy supplied from building energy upgrades can serve for decarbonization purposes. More specifically, how can upgrading aging residential buildings to low-energy solar buildings increase the potential for decarbonizing mobility and heating?
In particular, this study analyzes existing single-detached residential houses that are retrofitted/rebuilt and equipped with photovoltaic (PV) roofs in Montreal, Québec, Canada (45.5 • N, mid-latitude, 4457 heating degree-days). Figure 1 illustrates how upgrading existing buildings can make electricity available to the building itself and/or the utility that redirects it for decarbonization purposes. The existing building can be retrofitted (i.e., the structure is kept and energy savings achieved by improving the envelope design and heating, ventilation, and air conditioning system, etc.) or demolished and rebuilt to a low-energy solar building. The energy generated from PV may be partly consumed by the building (e.g., self-consumed for its operation, to power the occupant's EV(s), or to operate a greenhouse located on the land lot) while any excess is exported to the utility who redirects it for decarbonization applications elsewhere. The sum of the produced solar power and the savings due to energy efficiency provides the utility with a low-emission power supply that can help to achieve society's decarbonization targets. In particular, this study considers the potential use of the electricity that is made available from the building upgrade for powering battery and fuel cell personal EVs and for electrifying the heating energy used in buildings, including for greenhouse crop production. produced solar power and the savings due to energy efficiency provides the utility with a low-emission power supply that can help to achieve society's decarbonization targets. In particular, this study considers the potential use of the electricity that is made available from the building upgrade for powering battery and fuel cell personal EVs and for electrifying the heating energy used in buildings, including for greenhouse crop production.

Energy and Emissions Analysis
The energy consumption of a typical house is compared to two upgrade scenarios: one where the existing single-detached house is retrofitted (efficiency measures applied while keeping the main structure) and the other where it is demolished and rebuilt. The upgraded houses generate electricity on-site with a building-applied PV roof for the retrofit house and a more resource-efficient building-integrated PV roof for the new construction. The first step of the analysis is to quantify the total amount of electricity that can be made available by upgrading a single-detached house or "upgrade energy", which includes contributions from solar electricity generation plus the electricity that is freed up by implementing energy efficiency measures. Then, the amount of mobility and heating (including for greenhouses) that can be electrified using the upgrade energy will be determined based on various scenarios that are described in detail in Section 2.4. This analysis requires energy models for each house and the greenhouse. Annual energy simulations will be performed to quantify their energy performance and the associated GHG emissions.

House Characteristics
Two main scenarios are compared for upgrading an existing house. The first is to retrofit the existing house and the second is to rebuild it. For the rebuild option, the house geometry changes, whereby two houses are built on the same land lot to provide results that reflect the growing trend of densification (reduces inhabitant land and energy requirements). The existing and retrofitted single-story houses have a basement and a main living floor above it, with a slab-on-grade garage connected to it (Figure 2a). The new houses are designed to have the same living area (typically excluding basement and garage) as their existing counterparts (140 m 2 ). Although the footprint of residential buildings in Canada has increased by approximately 20% in the last 30 years [1], future trends may be different as compact cities and environmentally friendly lifestyles are favored. For a balanced prediction given these uncertainties, the newly built houses were specified to have the same living area as its existing counterpart which also allows a fairer comparison of their energy performance. To build two houses on the same land lot (∼850 m 2 ) while maximizing available outdoor space, the new homes will have two stories (Figure 2b). Similarly, the new houses will have an attached garage and a basement.

Energy and Emissions Analysis
The energy consumption of a typical house is compared to two upgrade scenarios: one where the existing single-detached house is retrofitted (efficiency measures applied while keeping the main structure) and the other where it is demolished and rebuilt. The upgraded houses generate electricity on-site with a building-applied PV roof for the retrofit house and a more resource-efficient building-integrated PV roof for the new construction. The first step of the analysis is to quantify the total amount of electricity that can be made available by upgrading a single-detached house or "upgrade energy", which includes contributions from solar electricity generation plus the electricity that is freed up by implementing energy efficiency measures. Then, the amount of mobility and heating (including for greenhouses) that can be electrified using the upgrade energy will be determined based on various scenarios that are described in detail in Section 2.4. This analysis requires energy models for each house and the greenhouse. Annual energy simulations will be performed to quantify their energy performance and the associated GHG emissions.

House Characteristics
Two main scenarios are compared for upgrading an existing house. The first is to retrofit the existing house and the second is to rebuild it. For the rebuild option, the house geometry changes, whereby two houses are built on the same land lot to provide results that reflect the growing trend of densification (reduces inhabitant land and energy requirements). The existing and retrofitted single-story houses have a basement and a main living floor above it, with a slab-on-grade garage connected to it (Figure 2a). The new houses are designed to have the same living area (typically excluding basement and garage) as their existing counterparts (140 m 2 ). Although the footprint of residential buildings in Canada has increased by approximately 20% in the last 30 years [1], future trends may be different as compact cities and environmentally friendly lifestyles are favored. For a balanced prediction given these uncertainties, the newly built houses were specified to have the same living area as its existing counterpart which also allows a fairer comparison of their energy performance. To build two houses on the same land lot (∼850 m 2 ) while maximizing available outdoor space, the new homes will have two stories (Figure 2b). Similarly, the new houses will have an attached garage and a basement. The houses are equipped with heating and cooling and the upgraded houses also have energy recovery ventilation.
The more compact design of the rebuilt houses causes them to have a smaller area for PV production compared to the retrofitted house. Both houses are upgraded to the same level of energy efficiency. The applied energy efficiency measures are related to the space and DHW heating, the building envelope (insulation, windows and movable shades), and ventilation. The south-facing roof of the upgraded houses is equipped with a PV system consisting of monocrystalline PV modules and an inverter. The roof surface area covered Energies 2021, 14, 6820 5 of 31 by PV is 97.7 m 2 and 46.1 m 2 , producing 20.7 kW and 9.8 kW of peak electricity for the retrofitted and rebuilt houses, respectively. Additional details for the modeled houses can be found in a previous study by Bambara, Athienitis and Eicker [7].
The houses are equipped with heating and cooling and the upgraded houses also have energy recovery ventilation.
(a) (b) Figure 2. Schematic showing the two modeled houses [7]: (a) single-story existing house; (b) two two-story solar houses built on the same land lot (they can be semi-detached or spaced with a small distance).
The more compact design of the rebuilt houses causes them to have a smaller area for PV production compared to the retrofitted house. Both houses are upgraded to the same level of energy efficiency. The applied energy efficiency measures are related to the space and DHW heating, the building envelope (insulation, windows and movable shades), and ventilation. The south-facing roof of the upgraded houses is equipped with a PV system consisting of monocrystalline PV modules and an inverter. The roof surface area covered by PV is 97.7 m 2 and 46.1 m 2 , producing 20.7 kW and 9.8 kW of peak electricity for the retrofitted and rebuilt houses, respectively. Additional details for the modeled houses can be found in a previous study by Bambara, Athienitis and Eicker [7].

Greenhouse Characteristics
The modeled greenhouse is a typical single pane glass greenhouse with artificial lighting used for the consistent production of crops year-round. Heating and ventilation are used to control inside humidity and temperature. Thermal shading screens reduce night heat loss and are also used to control sunlight levels. The modeled greenhouse has an equal length and width of 30.5 m, a height of 3.7 m, and a total area of 929 m 2 (10,000 sqft). More of the details for the greenhouse can be found in a previous study by Bambara and Athienitis [24].

Energy Modeling and Simulation
The TRNSYS 17.2 software (Thermal Energy System Specialists, LLC, Madison, WI, USA) was selected for the transient simulation of the house and greenhouse climate [25]. A majority of the population and thus house upgrades would be located between Montreal and Québec City. Since these two locations have similar weather, the simulations will be performed for Montreal.
House energy performance Table 1 provides the monthly energy consumption and production for the existing, retrofitted, and rebuilt houses. These monthly values were obtained by summing hourly simulation output data. The annual energy consumption of the existing house is sufficiently close to the value reported by [26] for single-detached houses.  [7]: (a) single-story existing house; (b) two two-story solar houses built on the same land lot (they can be semi-detached or spaced with a small distance).

Greenhouse Characteristics
The modeled greenhouse is a typical single pane glass greenhouse with artificial lighting used for the consistent production of crops year-round. Heating and ventilation are used to control inside humidity and temperature. Thermal shading screens reduce night heat loss and are also used to control sunlight levels. The modeled greenhouse has an equal length and width of 30.5 m, a height of 3.7 m, and a total area of 929 m 2 (10,000 sqft). More of the details for the greenhouse can be found in a previous study by Bambara and Athienitis [24].

Energy Modeling and Simulation
The TRNSYS 17.2 software (Thermal Energy System Specialists, LLC, Madison, WI, USA) was selected for the transient simulation of the house and greenhouse climate [25]. A majority of the population and thus house upgrades would be located between Montreal and Québec City. Since these two locations have similar weather, the simulations will be performed for Montreal.
House energy performance Table 1 provides the monthly energy consumption and production for the existing, retrofitted, and rebuilt houses. These monthly values were obtained by summing hourly simulation output data. The annual energy consumption of the existing house is sufficiently close to the value reported by [26] for single-detached houses.
Greenhouse energy consumption The greenhouse energy model developed by Bambara and Athienitis [24] was used for this study. The simulations were performed for Ottawa and assumed to be adequate for representing conditions in Montreal, which is nearby and has sufficiently similar weather. The energy consumed for heating and lighting is in good agreement with those reported for Québec greenhouses with a thermal screen [27,28]. Since approximately 98% of Canadian greenhouse production is dedicated to growing fruiting crops such as tomatoes [29], the daily lighting integral was increased to 21 mol m −2 d −1 to reflect their growing needs [28]. The heating energy is provided by burning natural gas inside the greenhouse. Other energy such as pumping, air circulation fans, etc. is taken as 34 kWh m −2 and ventilation fan power is calculated using 4.7 W m −3 h −1 [30]. Table 2 provides the peak winter demand and monthly energy consumption that was obtained by summing hourly simulation output data.

Energy Analysis
The energy that is made available by upgrading an existing building to a low-energy solar building is saved primarily by the utility and a portion may be self-consumed on-site ( Figure 1). For the scenarios considered in this study, it is assumed that all of the electricity that is made available from building upgrades (sum of contributions from solar and energy efficiency) is saved by the utility that redirects it for mobility and heating decarbonization applications.
The present analysis considers single-detached houses but the methodology could be expanded to cover other residential buildings such as attached houses and apartments. In addition, only houses that employ electric resistance heating were considered because they are the most commonly used heating systems in Québec and their low efficiency provides significant potential for freeing up electricity from the applied energy efficiency measures.
As a first step, the energy that is made available (or upgrade energy) by retrofitting and rebuilding an existing single-detached house to a low-energy solar house is quantified through annual energy simulations. Once the annual profile of upgrade energy is known, Energies 2021, 14, 6820 7 of 31 the maximum amount of electricity that can serve to decarbonize mobility is determined and the remainder is used for heating electrification. Figure 3 illustrates the two main scenarios considered in this study. The first scenario uses the results for a single-detached house to estimate how many houses would need to be upgraded in Québec to provide all the energy required to electrify greenhouse operations. This includes the energy needed to convert all of the existing greenhouses from natural gas to electric heating and the electricity required to operate all of the planned new greenhouses (to achieve fresh vegetable production autonomy in Québec). The second scenario expands upon the findings of scenario 1 by considering the impact of upgrading all of the singledetached houses in Québec that employ electric resistance heating (42% of single-detached houses). Since all of the greenhouse operations have already been decarbonized in scenario 1, the additional upgrade energy would serve to electrify the heating in buildings that use natural gas for heating (e.g., commercial and institutional buildings). In all cases, the use of upgrade energy is first prioritized for decarbonizing mobility because it carries the highest GHG emissions and contributes to urban pollution the most.
ments. In addition, only houses that employ electric resistance heating were considered because they are the most commonly used heating systems in Québec and their low efficiency provides significant potential for freeing up electricity from the applied energy efficiency measures.
As a first step, the energy that is made available (or upgrade energy) by retrofitting and rebuilding an existing single-detached house to a low-energy solar house is quantified through annual energy simulations. Once the annual profile of upgrade energy is known, the maximum amount of electricity that can serve to decarbonize mobility is determined and the remainder is used for heating electrification. Figure 3 illustrates the two main scenarios considered in this study. The first scenario uses the results for a single-detached house to estimate how many houses would need to be upgraded in Québec to provide all the energy required to electrify greenhouse operations. This includes the energy needed to convert all of the existing greenhouses from natural gas to electric heating and the electricity required to operate all of the planned new greenhouses (to achieve fresh vegetable production autonomy in Québec). The second scenario expands upon the findings of scenario 1 by considering the impact of upgrading all of the single-detached houses in Québec that employ electric resistance heating (42% of single-detached houses). Since all of the greenhouse operations have already been decarbonized in scenario 1, the additional upgrade energy would serve to electrify the heating in buildings that use natural gas for heating (e.g., commercial and institutional buildings). In all cases, the use of upgrade energy is first prioritized for decarbonizing mobility because it carries the highest GHG emissions and contributes to urban pollution the most. Although the present analysis considers that all of the upgrade energy is saved by the utility, some the solar power generated on-site would likely be self-consumed by the building itself, used for the occupants' EV, or to operate a greenhouse located the land lot ( Figure 1). Increasing self-consumption may even be incentivized in certain low voltage grids to avoid power grid congestion. For the case where the house is fully operated using electricity (current situation), shifting energy use from the utility to consumption on-site would not change the net amount of upgrade energy and associated decarbonization potential presented in this investigation.
To quantify the decarbonization potential for the two main scenarios, the energy analysis is separated into different sections and the key assumptions for each can be found below. A house upgrade implies either a retrofit (efficiency measures applied while keeping the main structure) or a rebuild and the results for both are presented and compared. In addition, mobility decarbonization is achieved by switching to battery or fuel cell EVs and the results for both are provided.

1.
The first part covers the energy analysis of a single-detached house to answer the following questions: What is the maximum amount of mobility electrification that is achieved using upgrade energy? 1.3.
How much energy is available for heating decarbonization and what is the maximum amount of greenhouse area that can be decarbonized/operated by using it? 2.
The second part consists of evaluating the impact of upgrading many houses to answer the following questions: 2.1. Scenario 1: How many houses should be upgraded to provide enough energy for decarbonizing/operating all of the greenhouses needed to achieve fresh vegetable production autonomy in Québec? 2.2.
Scenario 1: How many houses should be upgraded to provide enough energy for decarbonizing/operating all of the greenhouses needed to achieve fresh vegetable production autonomy in Québec?
The details, parameter values, and key assumptions for the energy modeling are presented below and given in Table 3. The values for the modeled house are given in [7,24] for the greenhouse. Section 1.1: For each hour of the year, the quantity of electricity that is freed up by implementing energy efficiency measures is calculated by subtracting the electricity consumed by the upgraded house from that of the existing house. The electricity that is made available for decarbonization (E decarb in kWh year −1 ) is found by summing the electricity that is freed up by implementing energy efficiency measures (E eff in kWh year −1 ) and the solar electricity generation (E PV in kWh year −1 ). The sum of these hourly values over one year provides the annual quantity of energy for the case where the house is retrofitted or rebuilt. Monthly totals are used to visualize and calculate some of the results. Section 1.2: This study considers using a portion of the electricity that is made available from the house upgrade (or upgrade energy) to power battery or fuel cell electric vehicles (BEVs and FCEVs). Personal vehicles are assumed to be used for traveling approximately the same distance each day and modeled as a constant load throughout the year. The maximum amount of energy that can serve to electrify personal vehicles is taken as the lowest value of the monthly electricity (E EV in kWh month −1 ) that is made available from the house energy upgrade. A value of one month was selected for convenience and because the hydropower grid with large reservoirs allows it to provide energy flexibly. Therefore, each month, an amount of energy equal to E EV is reserved for EVs. When the monthly value of upgrade energy exceeds E EV , the surplus serves to electrify heating. The annual quantity of electricity that is available for EVs is E EV multiplied by twelve.
The distance a BEV or FCEV can travel (D EV in km year −1 ) using the electricity that is provided by upgrading a single-detached house is given by: where E EV is the monthly energy reserved for EVs (kWh month −1 ) EC EV is the BEV or FCEV energy consumption (kWh km −1 ) the factor 12 is the number of months per year (month year −1 ).
The energy consumption (grid-to-wheel) for BEV (EC BEV ) was taken as the weighted value between cars (0.19 kWh km −1 for 63% of personal vehicles) and light trucks (0.25 kWh km −1 for 37% of personal vehicles) [31]. The energy consumption of a FCEV is estimated by comparing the efficiencies of BEV and FCEV for converting grid electricity into power at the wheels. The grid-to-wheel efficiency of a BEV is estimated to be 83% by multiplying the following efficiencies: 97% for the AC/DC inverter, 95% for the batteries, 97% for the DC/AC inverter, and 93% for the electric motor [32]. The grid-to-wheel efficiency of an FCEV is taken as 35% by multiplying the following efficiencies and assuming hydrogen is produced directly at the refuel station: 97% for the AC/DC inverter, 73% for water electrolysis, 92% for hydrogen compression, 60% for the fuel cell, 97% for the DC/AC inverter, and 93% for the electric motor [32,33]. The ratio of these efficiencies (R BEV:FCEV ) equals 2.37 (83%/35%), meaning that a FCEV requires that many times more energy than the BEV for a similar range. The energy consumption (EC FCEV ) of a personal FCEV is found by multiplying the energy consumption of a BEV (EC BEV ) by this ratio of efficiencies (R BEV:FCEV ). If the energy that is made available by upgrading a given number of houses exceeds the need for electrifying all the personal vehicles, then the surplus energy can be used to decarbonize other vehicles such as fuel cell electric trucks (FCET) for regional or long-haul transport. The energy consumption of a FCET (EC FCET ) is estimated the same way as for the FCEV using the energy consumption for battery electric trucks (BET) provided by Ref. [34].
Then, the number of BEVs or FCEVs that can be powered using the energy from a single house upgrade (N EV ) is calculated by: where D EV_yr is the personal EV annual travel distance (km year −1 ). Section 1.3: The electricity that is made available by upgrading to a low energy solar house (or upgrade energy) is divided between mobility and heating decarbonization purposes. To satisfy the energy needs of the newly electrified heating loads, it is necessary to ensure that there is sufficient energy over the desired period and that enough power is available during peak demand. Therefore, the amount of heating decarbonization will be limited either by energy or power availability. The monthly amount of electricity that is available for heating decarbonization is found by subtracting the amount of electricity dedicated to EVs (E EV in kWh month −1 ) from the monthly amount of upgrade energy (E decarb_mo in kWh month −1 ). The amount of power available during periods of peak demand is found by subtracting the EV power demand from the maximum hourly value of power that is freed up by implementing energy efficiency measures. The contribution of solar energy is ignored as it is not guaranteed to be available during peak periods. The power demand from EV is assumed to be constant with time and calculated by dividing the monthly amount that is reserved for EV (E EV in kWh month −1 ) by the number of hours in one month.
The portion of the upgrade energy that is not used for EVs will be prioritized for use in greenhouses and any surpluses will then serve to electrify heating in other buildings. It is desired to determine the greenhouse area that can be decarbonized (for existing greenhouses) and operated (for new greenhouses) using the electricity that is made available by upgrading a single-detached house. Two cases are considered to address the different needs for decarbonizing the heating in existing greenhouses and providing low-emission power for the operation of new greenhouses (including heating, lighting, and other electricity needs).
The annual electricity that is available for greenhouses (E GH in kWh year −1 ) is given by: where E decarb is the electricity available from upgrading one house (kWh year −1 ) the factor 12 is the number of months per year (month year −1 ).
When the potential greenhouse area is designed based on energy availability, it is necessary to find the area (A GH_per_house in m 2 ) which minimizes the annual difference between the house upgrade energy and the actual energy consumed by a given greenhouse area. The greenhouse area based on power availability can be found by dividing the power that is dedicated to greenhouses (P max in kW, obtained by subtracting the EV power demand (P EV in kW) from the maximum power freed up by implementing efficiency measures (P eff in kW)) by the greenhouse's peak power demand per unit area (P GH in kW m −2 ). The calculated greenhouse areas (A GH_per_house calculated for existing and new greenhouse) based on energy and power are compared and the lowest value obtained is selected as the final design. The annual thermal energy consumed by the existing greenhouses (E GH_exist_per_house in kWh year −1 ) that are heated using the electricity that is made available from upgrading a house is determined from: where A GH_exist_per_house is the greenhouse area that can be decarbonized from one house upgrade (m 2 ) E GH_heat is the electricity consumed for heating per unit area of a greenhouse (kWh m −2 year −1 ).
Similarly, the annual energy consumed by the new greenhouses (E GH_new_per_house in kWh year −1 ) that are operated using the upgrade energy may be written as: where A GH_new_per_house is the greenhouse area that can be operated from one house upgrade (m 2 ) E GH is the electricity consumed per unit area of a greenhouse (kWh m −2 year −1 ). Section 2.1: For scenario 1, it is desired to know how many houses would need to be upgraded to decarbonize/operate greenhouses in Québec. For the first case, it is desired to know how many houses would need to be upgraded to provide enough energy for electrifying the heating energy used in existing greenhouses, which are nearly all heated using natural gas today. For the second case, it is desired to know how many houses would need to be upgraded to provide enough energy to operate all of the new greenhouses (including heating, lighting, and other electricity needs) that will be built to achieve local fresh vegetable production autonomy.
Currently, a greenhouse area of approximately 1,280,000 m 2 is used to grow 41,000 t yr −1 of vegetables in Québec [29]. The government has plans and established financial incentives to achieve fresh vegetable production autonomy by doubling the greenhouse production area by 2025 [8,35]. Therefore, approximately 1,280,000 m 2 of additional greenhouse area would be needed to produce a total of 82,000 tonnes of vegetables per year.
The previous analysis determined the maximum greenhouse area that could be decarbonized/operated using the electricity that is made available by upgrading a house. The number of houses (N house_GH_exist ) needed to provide enough power for electrifying the heating in all existing greenhouses is estimated using: where A GH_tot_exist is the total existing greenhouse area (m 2 ). Similarly, the number of houses (N house_GH_new ) needed to provide enough electricity for operating all the new greenhouses is determined from: where A GH_tot_new is the total new greenhouse area (m 2 ). The total number of houses that would need to be upgraded to provide enough electricity to decarbonize and operate all greenhouses in Québec (N house_GH_tot ) is defined as: The fraction of total single-detached houses that would need to be upgraded (F upgraded_GH in %) is derived from: (9) where N house_tot is the total number of single-detached homes in Québec the factor 100 serves to express the result in percentage.
The total electricity that is available for decarbonization (E decarb_tot in TWh year −1 ) is calculated as: where E decarb is the electricity that is available for decarbonization (kWh year −1 ) the factor 10 9 serves to convert kWh to TWh (month year −1 ).
The fraction of total electricity consumption (F decarb in %) is determined by dividing total electricity that is available for decarbonization (E decarb_tot ) by the total electricity consumption in Québec (E QC in TWh year −1 ) and multiplying this by 100 to express the result in percentage.
The total electricity consumed for decarbonizing the existing greenhouses and operating the new greenhouses (E GH_tot in TWh year −1 ) is computed as: where the factor 10 9 serves to convert kWh to TWh (month year −1 ). Later, the electricity (E GH_heat_tot in TWh year −1 ) that is used solely for heating the existing and new greenhouses will be needed to determine the avoided GHG emissions and is described by: where the factor 10 9 serves to convert kWh to TWh (month year −1 ). The total electricity for EV (E EV_tot in TWh year −1 ) is calculated by: where the factors 12 is the number of months per year (month year −1 ) and 10 9 serves to convert kWh to TWh (month year −1 ). The leftover electricity (E leftover in TWh year −1 ) and available for other decarbonization purposes such as electrifying heating in buildings that employ natural gas is expressed as: The number of BEVs or FCEVs that can be powered using a portion of the electricity that is made available by upgrading these houses (N EV_tot ) is given by: The fraction of total personal vehicles that could be converted to EVs (F EV_tot in %) is estimated using: where N PV_tot is the total number of personal vehicles the factor 100 serves to express the result in percentage.
Energies 2021, 14, 6820 12 of 31 Section 2.2: For scenario 2, it is desired to quantify the potential decarbonization that can be achieved by converting all of the single-detached houses that employ electric resistance heating to low energy solar houses. This scenario builds upon the results of scenario 1, where a certain number of houses are upgraded to provide energy for greenhouses. In addition to this, scenario 2 includes the decarbonization potential of an additional number of houses (N house_added ) equal to: (17) where N house_EH_tot is the number of houses that use electric resistance heating. The analysis for these additional houses follows a similar procedure as for scenario 1 and uses the same equations, Equations (10), (13) and (14) (where E GH_tot is ignored), (15), and (16), but using N house_added instead of N house_GH_tot . The total values for scenario 2 are the sum of the contributions from N house_GH_tot and N house_added .

GHG Emissions Analysis
This section aims to quantify the reduction in GHG emissions that is achieved by decarbonizing mobility and heating using the energy that is made available by upgrading the number of houses that were determined in scenarios 1 and 2. The analysis also includes estimating the embodied emissions associated with the house upgrades and electrified loads and determining the emissions payback time.

GHG emissions avoided by decarbonization
The reduction in emissions from mobility electrification (GHG EV_per_house in tCO 2eq ) that can be achieved using a portion of the upgrade energy is determined from: where FE PV is the average fuel efficiency of personal vehicles (L km −1 ) EF PV is the emissions factor of gasoline-powered personal vehicles (kgCO 2eq L −1 ) the factor 1000 serves to convert kg to tonne (t).
If the portion of upgrade energy that can be dedicated to EVs exceeds the energy needs of all of the personal vehicles in Québec, the surplus is assumed to be used for decarbonizing long-haul trucks by switching them to fuel cell electric trucks (FCETs). The associated reduction in emissions (GHG FCET in MtCO 2eq year −1 ) is estimated using: where E FCET is the surplus electricity available for FCETs (kWh year −1 ) EC FCET is the FCET energy consumption (kWh km −1 ) FE T is the average fuel efficiency of a diesel-powered truck (L km −1 ) EF T is the emissions factor of diesel-powered truck (kgCO 2eq L −1 ) the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The reduction in emissions that can be achieved by electrifying mobility (GHG EV in MtCO 2eq year −1 ) in scenario 1 (N house_GH_tot ) or scenario 2 (N house_EH_tot ) is given by: where the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt). The reduction in emissions (GHG GH in MtCO 2eq year −1 ) from electrifying the heating in existing greenhouses and from the emissions that are avoided by employing electric heating in the new greenhouses (assuming they would otherwise be heated using natural gas burned inside the greenhouse) is calculated by: where EF NG is the emissions factor of natural gas (kgCO 2eq m −3 ) EV is the energy value of natural gas (MJ m −3 ) the factor 10 9 serves to convert TWh to kWh the factor 3.6 serves to convert MJ to kWh the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The analysis considers that existing heating systems that burn natural gas directly inside the greenhouse are replaced with electric resistance heating. The use of heat pumps (e.g., geothermal or thermal storage using surplus air thermal energy) and dual fuel systems could be considered in future studies.
The reduction in emissions that is achieved using the leftover electricity (GHG leftover in MtCO 2eq year −1 ) to convert building heating from natural gas boilers to electric heat pumps is approximated by: where η is the natural gas boiler efficiency (dimensionless) COP is the average coefficient of performance of an electric heat pump (MJ m −3 ) the factor 10 9 serves to convert TWh to kWh the factor 3.6 serves to convert MJ to kWh the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The total reduction in emissions (GHG tot in MtCO 2eq year −1 ) is equal to: The reduction in Québec's emissions (R GHG in %) is expressed as: where GHG QC is Québec's total annual GHG emissions (MtCO 2eq year −1 ) the factor 100 serves to express the result in percentage.

Embodied GHG emissions
This study considers the emissions that are required for the house upgrade and the production of EVs. The embodied energy of the new greenhouses will not be considered because they will be built regardless of the decarbonization scenario. They will be treated like the existing greenhouses, whereby the embodied emissions produced by converting the heating system from natural gas to electric resistance are relatively small and can be neglected. The embodied emissions produced by retrofitting or rebuilding the houses (EE houses in MtCO 2eq ) in scenario 1 (N house_GH_tot ) or scenario 2 (N house_EH_tot ) is calculated by: where EE house is the embodied emissions per unit area for a retrofitted or rebuilt house (kgCO 2eq m −2 ) A is the living area of the house (m 2 ) the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The operational (e.g., O&M) and downstream processes (e.g., removal/recycling) for roof-mounted PV systems are small compared to the upstream emissions (e.g., productionrelated) and ignored in this study [36]. The embodied emissions produced by the PV system (EE PV in MtCO 2eq ) in scenario 1 (N house_GH_tot ) or scenario 2 (N house_EH_tot ) is estimated using: where EE PV_m2 is the embodied emissions per unit area of the PV system (kgCO 2eq m −2 ) A PV is PV area (m 2 ) the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The embodied emissions related to the production of FCETs (EE FCETs in MtCO 2eq ) is determined from: where D ET_yr is the FCET annual travel distance (km year −1 ) EE FCET is the embodied emissions per truck (kgCO 2eq ) the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The embodied emissions related to the production of EVs (EE EVs in MtCO 2eq ) in scenario 1 (N house_GH_tot ) or scenario 2 (N house_EH_tot ) is computed as: where the factor 1000 serves to convert kg to tonne (t) the factor 10 6 serves to convert tonnes (t) to megatonnes (Mt).
The total embodied emissions (EE tot in MtCO 2eq ) are approximated by: Emissions payback time The emissions payback time (PT GHG in yr) is equal to:

Energy Available by Upgrading a House
As expected, energy efficiency upgrades significantly reduce a house's energy consumption. The retrofitted and rebuilt houses consume similar amounts of energy, equal to about 65% less than an existing house. As shown in Table 4, implementing efficiency measures in single-detached houses results in around 14,000 kWh year −1 of electricity being freed up and available for decarbonization elsewhere. In addition, the retrofitted and rebuilt houses generate approximately 14,800 kWh year −1 and 31,300 kWh year −1 of solar electricity on-site, respectively. This large difference is caused by the retrofitted house having a PV area more than double the size of the rebuilt house. The simulated solar energy production for the rebuilt house (14,800 kWh year −1 ) is in good agreement with a simple calculation that yields 15,500 kWh year −1 based on the annual global radiation on a south-facing surface at a tilt angle equal to the latitude for Montreal (4.35 kWh m −2 year −1 ) provided by Ref. [48], a PV module efficiency of 21.16%, and a PV area of 46.1 m 2 .
The total electricity that is made available by upgrading one house (or upgrade energy) equals the sum of the energy freed up by implementing energy efficiency measures and the solar energy generation. This amounts to approximately 45,500 kWh year −1 per retrofitted house and 29,500 kWh year −1 per rebuilt house. In other words, the retrofitted house could provide energy for decarbonization that is equal to twice the electricity consumption of the existing house and 1.3 times more for the rebuilt house. Solar energy provides 50% and nearly 70% of the total upgrade energy for the retrofitted and rebuilt houses, respectively, with the remainder provided by the efficiency measures. Meanwhile, the amount of energy that is freed up due to efficiency is highest during periods of peak demand in winter and is found to be 7.1 and 7.3 kW per retrofitted and rebuilt house, respectively. When the electricity that is made available by upgrading to a low-energy solar house is used to decarbonize elsewhere, the amount of energy and peak power that is required for the electrified load must not exceed these values.  Figure 4 provides the monthly quantity of electricity that is freed up by implementing energy efficiency measures, from solar electricity generation, and the total electricity that is made available by retrofitting ( Figure 4a) and rebuilding (Figure 4b) a house. As expected, the energy that is freed up by implementing efficiency measures is greatest in the winter months because heating is the largest contributor to total energy use and is therefore reduced most. The opposite trend occurs with solar energy, whereby energy production in summer is about 2-3 times more than in winter for sloped south-facing roofs. The annual profile of the total upgrade energy is highest in winter due to the relatively large contribution from the energy efficiency measures during that period. The impact of a larger PV system can be observed by comparing the total electricity profile (green line) for the retrofitted house ( Figure 4a) and the rebuilt house (Figure 4b). The profile's U-shape is better defined for the rebuilt house because of a lower amount of solar electricity is generated compared to the electricity that is freed up from efficiency measures.  As illustrated in Figure 5, the process of upgrading residential buildings to low-energy solar buildings provides an energy profile that is suitable for both mobility and heating decarbonization. The relative proportion of energy that is freed up by implementing various efficiency measures and solar energy generation is what determines how much energy is available for the desired decarbonization application(s).  The use of upgrade energy is prioritized for electrifying mobility because transportation is the largest single contributor to health-damaging air pollution in cities. Since EVs typically require a near-constant supply of energy year-round, the minimum annual value of the monthly upgrade energy is what dictates the amount that can be dedicated to EVs. Energy efficiency measures alone provide little energy for EV mobility due to the low availability in summer. For solar energy, the opposite effect occurs, where minimum energy availability occurs in winter, and this would dictate the portion that can be available for EVs. The combination of the electricity that is freed up by implementing energy efficiency measures and the solar electricity generation is a promising way to obtain a larger amount of energy for EVs than would be provided by each of the energy supply streams separately. This symbiosis occurs because the minimum value of monthly energy for one is close to the maximum of the other. For example, for the retrofitted house, energy efficiency (blue line in Figure 4a) provides a minimum value of 2 kWh m −2 month −1 (in summer) and this is what limits the electricity that can be reserved year-round for EVs. For solar electricity generation (red line in Figure 4a), the minimum value is 8 kWh m −2 month −1 in November. However once combined (green line in Figure 4a), the minimum value becomes 19 kWh m −2 month −1 (in November), which is 90% more than the sum of the minimum values for efficiency and solar individually (total of 10 kWh m −2 month −1 ). Therefore, the efficiency and solar combo can achieve the desired goal of attributing a greater share of the total upgrade energy for EV, and the remainder is concentrated in winter and is ideal for heating decarbonization ( Figure 5). It would be possible to use an alternative source of energy such as biomass to assist solar electricity production in winter (neglects efficiency) or to complement efficiency in summer (neglects solar) to achieve a constant annual energy profile. However, this study focuses on how to employ energy from building upgrades as the sole means for providing energy needed for decarbonization.  As illustrated in Figure 5, the process of upgrading residential buildings to low-energy solar buildings provides an energy profile that is suitable for both mobility and heating decarbonization. The relative proportion of energy that is freed up by implementing various efficiency measures and solar energy generation is what determines how much energy is available for the desired decarbonization application(s). It should be noted that the energy consumption of an existing house is representative of a recently built house. More energy would be freed up by implementing energy efficiency measures if older, less efficient houses are selected for the analysis. Therefore, this paper provides conservative estimates that are more representative of what could be expected in the future, when recently built houses will be renovated or rebuilt. Additionally, a portion of solar energy generation would usually be consumed on-site (self-consumption) to minimize energy transactions with the grid and peak demand. Although it is assumed they would be implemented, these grid-interactive flexibility and self-consumption strategies were not considered in the analysis as they would not significantly impact the quantity of house upgrade energy. In addition, separating them   As illustrated in Figure 5, the process of upgrading residential buildings to low-energy solar buildings provides an energy profile that is suitable for both mobility and heating decarbonization. The relative proportion of energy that is freed up by implementing various efficiency measures and solar energy generation is what determines how much energy is available for the desired decarbonization application(s).
It should be noted that the energy consumption of an existing house is representative of a recently built house. More energy would be freed up by implementing energy efficiency measures if older, less efficient houses are selected for the analysis. Therefore, this paper provides conservative estimates that are more representative of what could be expected in the future, when recently built houses will be renovated or rebuilt. Additionally, a portion of solar energy generation would usually be consumed on-site (self-consumption) to minimize energy transactions with the grid and peak demand. Although it is assumed they would be implemented, these grid-interactive flexibility and self-consumption strategies were not considered in the analysis as they would not significantly impact the quantity of house upgrade energy. In addition, separating them enables a better understanding of how each energy supply stream contributes to the profile of the house upgrade energy.

Heating Decarbonization
The electricity that is made available by upgrading to a low-energy solar building will be used to electrify mobility and heating. Figure 6 illustrates the energy profiles that are made available for mobility and heating decarbonization by retrofitting ( Figure 6a) and rebuilding (Figure 6b) an existing single-detached house. A greater amount of electricity is provided by the retrofitted house because it has a larger PV system that generates over two times more solar electricity than the rebuilt house. enables a better understanding of how each energy supply stream contributes to the profile of the house upgrade energy.

Heating Decarbonization
The electricity that is made available by upgrading to a low-energy solar building will be used to electrify mobility and heating. Figure 6 illustrates the energy profiles that are made available for mobility and heating decarbonization by retrofitting ( Figure 6a) and rebuilding (Figure 6b) an existing single-detached house. A greater amount of electricity is provided by the retrofitted house because it has a larger PV system that generates over two times more solar electricity than the rebuilt house.
(a) (b) Figure 6. Energy profile that is made available for mobility and heating decarbonization by upgrading to: (a) a retrofitted house and (b) a rebuilt house.
As a first step (scenario 1), this study investigates the case where the electricity would be used to decarbonize the heating energy used in existing greenhouses and provide the electricity required to operate the new greenhouses needed to grow all fresh vegetables locally. Table 5 provides the greenhouse area that can be decarbonized/operated using the electricity that is made available by upgrading a single-detached house. The greenhouse area is selected as the minimum value that is calculated based on available energy and peak power demand. It was found that the energy provided by a house retrofit could provide enough electricity to heat 12.5 m 2 of the existing greenhouse and operate 7 m 2 of the new greenhouse. Both of these areas were selected based on the availability of peak power. Rebuilding a house can provide enough electricity to heat 14.1 m 2 of existing greenhouse (selected based on energy availability) and to operate 10.1 m 2 of new greenhouse (selected based on peak power availability). Table 5. Greenhouse (GH) area that can be decarbonized/operated using the energy that is made available from retrofitting/rebuilding a single-detached house. As a first step (scenario 1), this study investigates the case where the electricity would be used to decarbonize the heating energy used in existing greenhouses and provide the electricity required to operate the new greenhouses needed to grow all fresh vegetables locally. Table 5 provides the greenhouse area that can be decarbonized/operated using the electricity that is made available by upgrading a single-detached house. The greenhouse area is selected as the minimum value that is calculated based on available energy and peak power demand. It was found that the energy provided by a house retrofit could provide enough electricity to heat 12.5 m 2 of the existing greenhouse and operate 7 m 2 of the new greenhouse. Both of these areas were selected based on the availability of peak power. Rebuilding a house can provide enough electricity to heat 14.1 m 2 of existing greenhouse (selected based on energy availability) and to operate 10.1 m 2 of new greenhouse (selected based on peak power availability).

Analysis for the Energy Needs of New Greenhouses
The upgrade energy that is not used for decarbonizing mobility and greenhouses would be available for other decarbonization purposes. Since this surplus is predominantly available during the heating months, it can serve to electrify heating in other buildings, for instance converting a natural gas boiler to an electric heat pump. Figure 7 illustrates the differences that occur between the energy that is available for heating decarbonization and the energy consumed for the greenhouse when it is designed based on peak power ( Figure 7a) and based on energy availability (Figure 7b). The two lines show how a better overlapping of energy supply and demand results in more energy being used for the greenhouses (i.e., less surplus available for electrifying heating elsewhere). When the greenhouse design is based on energy availability, there is very little surplus energy for decarbonization elsewhere. The opposite typically occurs for greenhouse areas designed based on peak power, where up to 16% of the upgrade energy is available for other decarbonization purposes (Table 5 and Figure 7a). A possible solution for using more of the upgrade energy for greenhouse purposes (i.e., in the case where the design is based on peak availability) is to employ a dual heating system, which would switch to an alternative fuel such as biomass to provide heat below a predetermined exterior air temperature. As shown in Figure 7b, there will be some months where the greenhouse energy consumption is greater than the energy that is available from the house upgrade. These periods would be supplied from the surplus that is available during the other parts of the year (e.g., surplus energy in winter covers the deficit in fall).  Figure 7. Monthly electricity available to decarbonize heating compared to the electricity consumption of the greenhouse (based on the area determined in Table 5) for: (a) a retrofitted house; (b) a rebuilt house.
The electricity needed to electrify heating in existing greenhouses and to operate the new greenhouses would require 2.1 TWh yr −1 of electricity and would add nearly 1000 MW of peak demand to the existing grid. This represents a 2.6% increase compared to today's peak demand of approximately 39,000 MW [5], or half of the new peak demand that is predicted to be required in the province by 2029 [15]. As shown in Table 6, all the energy needed to have an entirely local and decarbonized supply of fresh vegetables can be achieved by retrofitting 16% or rebuilding 12% of single-detached homes in Québec, respectively. Table 6. Energy that is provided by upgrading enough houses to achieve greenhouse (GH) crop production autonomy (scenario 1) and all houses heated by electric resistance (scenario 2). The second scenario considers the impact of upgrading all of the single-detached houses that currently use electric resistance heating in Québec (42% of the total or 734,000 houses). This would provide 7.9 and 5.1 TWh yr −1 of surplus electricity for decarbonizing heating in other buildings for the retrofitted and rebuilt houses, respectively. The total energy that is made available by retrofitting and rebuilding these houses equals 33.4 and 21.8 TWh yr −1 or 19% and 12% of the total electricity consumption of the province, respectively. Monthly energy (kWh)

Month
Electricity available to decarbonize heating Electricity consumption for greenhouse heating Figure 7. Monthly electricity available to decarbonize heating compared to the electricity consumption of the greenhouse (based on the area determined in Table 5) for: (a) a retrofitted house; (b) a rebuilt house.
It should be noted that this is possible owing to the flexible nature of Québec's hydropower system, which has large reservoirs that provide significant energy storage capacity. Hydroelectricity production can be reduced to accommodate distributed energy generation (e.g., from PV) and this saved energy (in the form of water stored in a reservoir) can be used at a later time (e.g., when the greenhouse energy consumption exceeds the amount that is available from building energy upgrades). The exceptionally large water reservoirs in Québec could enable seasonal grid energy balancing without the need for other forms of energy storage. Nuclear power plants would operate in a similar manner, where unused uranium can be stored for later use.
The electricity needed to electrify heating in existing greenhouses and to operate the new greenhouses would require 2.1 TWh year −1 of electricity and would add nearly 1000 MW of peak demand to the existing grid. This represents a 2.6% increase compared to today's peak demand of approximately 39,000 MW [5], or half of the new peak demand that is predicted to be required in the province by 2029 [15]. As shown in Table 6, all the energy needed to have an entirely local and decarbonized supply of fresh vegetables can be achieved by retrofitting 16% or rebuilding 12% of single-detached homes in Québec, respectively. Table 6. Energy that is provided by upgrading enough houses to achieve greenhouse (GH) crop production autonomy (scenario 1) and all houses heated by electric resistance (scenario 2). The second scenario considers the impact of upgrading all of the single-detached houses that currently use electric resistance heating in Québec (42% of the total or 734,000 houses). This would provide 7.9 and 5.1 TWh year −1 of surplus electricity for decarbonizing heating in other buildings for the retrofitted and rebuilt houses, respectively. The total energy that is made available by retrofitting and rebuilding these houses equals 33.4 and 21.8 TWh year −1 or 19% and 12% of the total electricity consumption of the province, respectively.

Item
The analysis considers the case where PV is only applied to house roof surfaces. For multi-story residential buildings such as apartments, the available solar energy collection area may not be enough for providing the desired energy profile of the electrified load. In such cases, solar energy can be generated from other surfaces within a district, for instance solar carports or semi-transparent PV cladding for greenhouses. Using this cladding causes the greenhouse lighting energy to increase in winter. However, on an annual basis, it reduces the overall demand for electricity by potentially covering all lighting electricity needs [11].

Mobility Decarbonization
The energy that is reserved for electrifying mobility (which requires a constant supply of energy year-round) is taken as the minimum monthly value of the house upgrade energy (see Section 3.1 for additional details). Table 7 provides the quantity of energy that can be dedicated to electrifying mobility and the number of EVs that can be operated using the energy that is made available from retrofitting/rebuilding a single-detached house. The maximum amount of energy that can be dedicated to EVs is found to be 2660 kWh month −1 for the retrofitted house and 1652 kWh month −1 for the rebuilt house. The rebuilt house provides 38% less energy for EVs than the retrofit house because less solar electricity is generated by its smaller roof area. Approximately 70% of the house upgrade energy can serve to electrify mobility. This energy is sufficient to power 9.8 and 6.1 personal BEVs for the retrofitted and rebuilt houses, respectively. For FCEVs, this energy is sufficient to power 4.1 and 2.6 personal vehicles for the retrofitted and rebuilt houses, respectively. The lower values for FCEVs are due to their lower well-to-wheel efficiency compared to BEV (35% versus 82%). Table 7. Number of personal EVs that can be operated using the energy that is made available from retrofitting/rebuilding a single-detached house.

Item Retrofit Rebuild
Monthly energy reserved for EVs (E EV kWh month −1 ) 2660 1652 Annual energy for EVs ( Table 8 provides the amount of energy that can be used to power EVs for scenarios 1 and 2. If enough houses would be upgraded to provide greenhouse fresh vegetable production autonomy (scenario 1), then 9.1 and 4.3 TWh year −1 of electricity would be available to decarbonize mobility for the retrofitted and rebuilt house, respectively. This would provide enough energy to convert 52% and 24% of personal vehicles to BEVs and 22% and 10% to FCEVs for the retrofitted and rebuilt house, respectively. If all houses that employ electric resistance heating would be upgraded (scenario 2), then 23.4 and 14.6 TWh year −1 of electricity would be available to decarbonize mobility for the retrofitted and rebuilt house, respectively. This would provide enough energy to convert 133% and 83% of personal vehicles to BEVs and 56% and 35% to FCEVs for the retrofitted and rebuilt house, respectively. For the case where BEVs are powered using energy that is made available by retrofitting houses, there is more than enough energy to cover all the energy required for personal vehicles. This surplus energy can be used for other mobility electrification purposes such as producing hydrogen for long-haul fuel cell electric trucks. Figure 8 shows the profile of power that is made available by rebuilding a house, on a sunny summer and winter day. In winter (blue line), the energy efficiency measures free up electricity all day plus there is a contribution from solar electricity generation. In summer (orange line), upgrade energy is provided almost entirely by solar electricity generation (energy efficiency lowers electricity use mainly by employing a heat pump for DHW heating but this is often outweighed by the higher cooling energy requirements of the upgraded houses when there is no sun). heavy-duty FCEVs, could be a more viable method for supplying power to the grid when needed (i.e., the EV serves for both transport and backup power generation purposes). Another issue is low voltage power grid congestion when a large amount of solar energy is being generated within a residential neighborhood. One way to avoid the need for upgrading power distribution networks is to store the solar electricity closer to where it is being produced. Various self-consumption strategies for the buildings and EVs could be employed for this purpose. For instance, during the sunniest period of the day, solar electricity could be converted to heat and stored in tanks (e.g., for space and DHW heating at night), used to charge BEVs or to produce hydrogen that can be stored at home. In addition, it may be more viable to store excess solar power at a larger scale such as per district block, in batteries or as hydrogen fuel, at distributed recharge/refuel stations. With more people working from home, the solar panels could be charging/refueling the EVs during the daytime and feeding it to the grid at night, providing substantial flexibility to the grid.

GHG Emissions Analysis
This study investigates the particular case where house upgrade energy is used to decarbonize mobility and heating energy use in greenhouses and buildings. In scenario 1, enough houses are upgraded to provide energy for decarbonizing the heating energy used in existing greenhouses and to supply all the energy required to operate new greenhouses. The GHG emissions reductions are achieved by switching from conventional fossil fuel-powered vehicles to EVs and from the electrification of greenhouse heating (switch from natural gas to electric). Scenario 2 considers that all the single-detached homes that employ electric resistance heating are upgraded. In this case, GHG emissions arise from the electrification of personal vehicles and trucks (using any surplus electricity beyond the needs of personal vehicles) and the heating decarbonization for the greenhouses (same as scenario 1) plus any excess that could be used to electrify heating in other buildings (by switching from natural gas to electric heat pumps). EV power demand is more or less constant throughout the year but may vary significantly throughout the day depending on the BEV's charging schedule and hydrogen production times. Typically, BEVs would be charged during the day when people arrive at work or when they are at home. The power demand due to charging can increase quickly depending on the charging speed and the number of vehicles charging simultaneously. For instance, for the retrofitted house in scenario 2, if BEVs charge over one day (slow charge), over 6 h (typical charging time), and 2 h (fast charge), the power demand of Québec's grid increases by 7%, 27%, and 82%, respectively. Therefore, the daily energy consumption profile for EVs would not necessarily match what is made available from the building upgrades. Most of the time, these short-term differences between energy supplied by low-energy solar houses and energy demand for EV could be accommodated by the grid owing to the flexible nature of hydropower with large reservoirs.
However, depending on the EV market penetration and the charging/refuelling times, the additional power demand for EVs can exceed grid power supply capabilities, particularly during periods of peak demand when the grid is already under pressure. Therefore, precautionary measures would be needed to ensure that EV charging/refuelling does not cause issues as the demand for power approaches the grid's peak supply capabilities (e.g., during cold winter days). This can be achieved possibly by providing incentives to vehicle owners to slow charge over a longer period and/or charge during off-peak periods (e.g., avoid charging during the winter's morning peak of 6-9 a.m. and evening peak of 4-8 p.m.). Other methods such as Hydro-Québec's dynamic pricing incentivize consumers to reduce demand during peak periods upon receiving a signal from the utility [16]. These incentives could help to increase the economic appeal and adoption of dual heating systems, such as switching to biomass for heating buildings and greenhouses during the coldest periods of the year. Otherwise, additional grid power can be provided by building more peaking power plants but they are often costly to operate due to their low utilization factor.
The issue of power demand from the grid is easier to manage for FCEVs because hydrogen can be produced at ideal times (e.g., during off-peak periods or when variable renewable power generation is high) and stored (e.g., in tanks directly at the refuel station). It may also be possible for hydrogen refuel stations to be equipped with a fuel cell or gas turbine to assist the grid by generating electricity during peak periods and possibly to provide emergency backup power. In the future, vehicle-to-grid, particularly from heavy-duty FCEVs, could be a more viable method for supplying power to the grid when needed (i.e., the EV serves for both transport and backup power generation purposes).
Another issue is low voltage power grid congestion when a large amount of solar energy is being generated within a residential neighborhood. One way to avoid the need for upgrading power distribution networks is to store the solar electricity closer to where it is being produced. Various self-consumption strategies for the buildings and EVs could be employed for this purpose. For instance, during the sunniest period of the day, solar electricity could be converted to heat and stored in tanks (e.g., for space and DHW heating at night), used to charge BEVs or to produce hydrogen that can be stored at home. In addition, it may be more viable to store excess solar power at a larger scale such as per district block, in batteries or as hydrogen fuel, at distributed recharge/refuel stations. With more people working from home, the solar panels could be charging/refueling the EVs during the daytime and feeding it to the grid at night, providing substantial flexibility to the grid.

GHG Emissions Analysis
This study investigates the particular case where house upgrade energy is used to decarbonize mobility and heating energy use in greenhouses and buildings. In scenario 1, enough houses are upgraded to provide energy for decarbonizing the heating energy used in existing greenhouses and to supply all the energy required to operate new greenhouses. The GHG emissions reductions are achieved by switching from conventional fossil fuelpowered vehicles to EVs and from the electrification of greenhouse heating (switch from natural gas to electric). Scenario 2 considers that all the single-detached homes that employ electric resistance heating are upgraded. In this case, GHG emissions arise from the electrification of personal vehicles and trucks (using any surplus electricity beyond the needs of personal vehicles) and the heating decarbonization for the greenhouses (same as scenario 1) plus any excess that could be used to electrify heating in other buildings (by switching from natural gas to electric heat pumps). Table 9 provides the GHG emission reductions that can be achieved by upgrading the total number of houses in scenarios 1 and 2. These include emission reductions from electrifying mobility (BEVs or FCEVs plus FCET if applicable) and heating (for greenhouses plus leftovers for other buildings). Greater reductions are achieved for the retrofitted house because its larger roof area allows it to produce more solar electricity. FCEVs provide less than half the emissions reductions of BEVs because higher losses occur by producing hydrogen and generating electricity from it using fuel cells. For scenario 1, upgrading single-detached homes could reduce Québec's GHG emissions by 3-14% (2.2-11.1 MtCO 2eq year −1 ) whereas for scenario 2, emission reductions of 12-32% (10.2-26.5 MtCO 2eq year −1 ) could be achieved. Table 9. GHG emission reductions, embodied emissions, and payback time associated with upgrading enough singledetached houses to achieve greenhouse (GH) crop production autonomy (scenario 1) and all houses that employ electric resistance heating (scenario 2). Although electrification using low-emission power reduces pollution, upgrading buildings and producing new EVs requires new materials, equipment, energy, and labor that in turn generate emissions (i.e., embodied emissions). Therefore, any investment into new material/equipment (except for plant-based materials) will initially increase emissions before the desired reductions begin to occur. Retrofits generally produce fewer emissions than rebuilds because some of the building's components such as the main structure are kept. For the retrofitted house, a period of 3.4 to 3.9 years would be needed for the emission reductions to exceed the emissions that are incurred by the building upgrades plus the production of new EVs. For the rebuilt house, the embodied emissions take a longer time to recover (6.1 to 11.2 years) because of the higher material/energy/human resources needed.

Benefits of Densification
Densification of single-detached homes may consist of replacing aging homes with two new homes (and possibly more depending on the land lot size) capable of housing more occupants on the same land lot. In this study, the rebuilt house would offer similar characteristics as their aging counterparts (e.g., similar living area with a private outdoor space) while reducing the energy and land use per occupant. The decrease in energy use stems from the more compact house geometry and to a greater extent from shorter driving distances (i.e., less power needed for EVs) that result from reduced urban sprawl. In addition, significant land area is freed up by having more people live in a smaller area than before. In theory, if aging single-detached houses would implement double density rebuilds, then one block could accommodate the same number of houses (of equal living area) as two blocks. A portion or all of the freed-up land could provide valuable space for an array of activities that are in line with the sustainable transition. The land could serve to grow vegetables in greenhouses year-round close to the consumers. However, the new greenhouse area that can be operated using the energy that is made available from double density house rebuilds represents only 2-3% of the land that is freed up. Therefore, significant land would be available for other purposes such as shared spaces, playgrounds, gardens, parks, renaturalization, renewable energy production, e-mobility recharge/refuel stations, and seasonal solar energy storage systems.

Comparison to Alternative Low-Emission Energy Supply Solutions
Decarbonization will require significant amounts of new low-emission energy. This study focuses on providing this electricity through building energy upgrades. In particular, the combination of energy efficiency and solar electricity generation as a way to increase the share of energy that can be dedicated for mobility. This is a unique opportunity in Québec due to the combination of a hydroelectric-dominant grid with large reservoirs that provide energy flexibility and energy inefficient aging residential buildings that commonly employ electric resistance heating.
Alternatively, the energy required for decarbonization can be obtained from new renewable energy sources such as solar, wind, and biomass. This would be needed for most of the rest of the world where solutions cannot be geographically constrained. The most suitable source of renewable energy depends on the energy use profile for the load that is being decarbonized. For instance, at high latitude locations, electrifying heating with solar energy is unlikely (unless seasonal energy storage is implemented) because of the seasonal mismatch between generation and demand. Biomass would be a more suitable energy resource for heating provided it is available locally and sustainably sourced. For constant energy use profiles such as the one needed to electrify mobility, DHW heating, and cooking, wind power could be an ideal candidate because it provides more or less the same energy output year-round (wind produces slightly less energy in summer but an optimized mix with solar energy can supply the desired constant annual energy profile).
Since solar and wind energy sources are intermittent, backup generation capacity would be needed to cover periods of low production. For instance, if Québec were to install a new wind/solar mixed power supply to electrify mobility, a backup energy source would be needed in the case there is no wind/sun for several days. The current hydropower grid could not be relied on for backup energy because it may already be supplying peak power for its existing consumers. Today, backup generation capacity is typically provided by natural gas. However, as society strives to achieve full decarbonization, storable forms of low-emission energy would be needed for instance, renewable natural gas, hydrogen or biomass. In addition, vehicle-to-grid (particularly using large FCEVs) can supply power to the grid when backup generation is needed and to assist with peak demand. Today, the low penetration of renewables enables energy to be withdrawn from and injected into the grid (which is mainly fossil fuel power generation and offers a high degree of energy flexibility), without any practical limits imposed. However, as the share of variable energy sources increases in the grid power supply, additional integration costs associated with variability (e.g., backup generation capacity, peak demand management), balancing (e.g., frequency regulation, short-term stabilization devices), and the grid (e.g., transmission, distribution, connection) will become increasingly important components of next generation low-emission power supply.
The most effective solutions would likely provide the lowest cost to decarbonize and the shortest emissions payback time. In Québec, the low-emissions power needed for electrification will likely be provided through a combination of energy demand (e.g., energy efficiency) and supply (e.g., renewables power) solutions. For the latter, the cost to decarbonize is easier to quantify as they will be typically financed as private investments Energies 2021, 14, 6820 26 of 31 (e.g., investor-owned utilities), possibly combined with government incentives. However, when the electricity needed for decarbonization is provided by building owners, new funding mechanisms would be needed. For instance, efficiency measures such as switching DHW heating from electric resistance to heat pump may not be an attractive homeowner investment due to the low cost of electricity in Québec. However, a detailed life cycle analysis may conclude that this pathway provides an overall lower cost and/or a faster way to decarbonize compared to new larger renewable energy projects. In other words, the perceived value of house energy upgrades (e.g., in terms of the overall cost to decarbonize) may be greater for society than for the homeowner themselves. In such cases, additional financial assistance, possibly in the form of carbon or tax credits, could be made available to building owners. This could increase the appeal and expedite the adoption of energy efficiency upgrades and PV systems, with the ultimate goal of achieving national decarbonization targets in the most effective way possible. Standardized methods for comparing decarbonization options based on emissions, cost, and supply chain criticality is a challenging problem due to the many inter-dependencies and assumptions involved and deserves greater attention.

Conclusions
This paper demonstrates how upgrading aging residential buildings to low-energy solar buildings can provide energy with an annual profile that matches the requirements for mobility and building and greenhouse heating decarbonization. In Québec, the energy required for electrification can be supplied in this manner owing to the large fraction of buildings that employ electric resistance heating and a grid power supply that is nearly fully hydroelectric with large reservoirs that often offer energy flexibility benefits. In this simulation study, the potential electricity that can be made available by upgrading houses is quantified by comparing the energy performance of a typical single-detached house to the cases where: (1) the same house is retrofitted to a low energy solar house and; (2) two low energy solar houses (of equal living area as the existing house, i.e., double density scenario) are rebuilt on the same land lot located in the suburbs of Montreal, Québec, Canada (45.5 • N, mid-latitude, 4457 heating degree-days).
The amount of electricity that is made available for mobility and heating decarbonization is the sum of the electricity that is freed up by implementing efficiency measures and the on-site solar electricity generation. It was found that retrofitting a house would provide electricity for decarbonization equal to approximately twice the annual energy consumption of the existing house (nearly 45,500 kWh year −1 ) and 1.3 times more (about 29,500 kWh year −1 ) for the rebuilt house. This "new" electricity is characterized by an annual profile that is greater in winter, allowing it to serve for the electrification of both mobility and heating. It was found that retrofitting 12% or rebuilding 16% of single-detached houses in Québec could provide enough energy to decarbonize the heating energy used for all existing greenhouses and operate all the new greenhouses that are required to achieve local fresh vegetable production autonomy.
The second scenario assessed the potential for upgrading all single-detached houses that employ electric resistance heating in Québec (42% of total or 734,000 houses). This would provide electricity for decarbonization equal to 33.4 and 21.8 TWh year −1 which is equivalent to a 19% and 12% increase of the province's electricity supply for the retrofitted and rebuilt houses, respectively. In addition to decarbonizing the energy used by all greenhouses, 7.9 and 5.1 TWh year −1 would be available to electrify the heating energy used in other buildings, for instance by switching from a natural gas boiler to electric heat pumps, which are becoming essential because they can provide both heating and cooling which is increasingly needed with rising summer temperatures. Approximately 70% of the total house upgrade electricity or 14.6 and 23.4 TWh year −1 can be dedicated to electrifying mobility, which is enough to convert 83% and 100% of the personal vehicles to BEVs and 35% and 56% to FCEVs for the rebuilt and retrofitted houses, respectively. The available electricity that is more than the requirements for personal vehicles can serve to decarbonize other vehicles such as public transportation or delivery vehicles.
Electrification avoids significant air, water, and soil pollution by replacing the use of fossil fuels, in particular liquid fuels used for transportation. Upgrading all of the houses that employ electric resistance heating would reduce GHG emissions by 0.3 MtCO 2eq year −1 from greenhouse heating electrification, by 3.5 and 5.4 MtCO 2eq year −1 from electrifying heating in other buildings, by 19 and 26.5 MtCO 2eq year −1 from switching to BEVs, or by 10.2 and 16.1 MtCO 2eq year −1 from switching to FCEVs for the rebuilt and retrofitted houses, respectively. Using house upgrade energy for decarbonization could reduce the province's emissions by up to 32% (26.5 MtCO 2eq year −1 ). Meanwhile, the building upgrades and production of EVs generate emissions of up to 116.1 MtCO 2eq year −1 . The time required for these embodied emissions to surpass the emissions avoided by electrification ranged from 3.4 to 11.2 years (retrofits have a significantly shorter emissions payback and should be prioritized when appropriate). Therefore, emissions will initially increase before the desired reductions occur and finding ways to minimize this is a major challenge of the sustainable transition.
An average single-detached house was modeled for this study even though energy consumption changes with age, construction methods and equipment. The electricity that is freed up due to efficiency measures could thus be higher if older less efficient houses were selected for analysis. Other residential buildings such as attached houses and apartments were not included in the analysis even though they could serve for decarbonization as well. Not all energy efficiency measures were considered and their selection was not based on economic performance. This study considers installing PV on the entire south-facing roof and no analysis of the optimal coverage was conducted. Depending on the energy use profile of the loads that will be electrified, there will be an optimal balance between energy efficiency measures and renewable energy generation. Future research into how to optimize the mix based on grid interaction, cost, emissions, and material resources will be needed. Moreover, a more in-depth analysis would be required to identify strategies for managing peak power demand as the EV penetration rate increases, particularly for BEV charging.
Although upgrading buildings has significant potential to provide energy for decarbonization, it competes with other solutions including large centralized renewable power generation such as wind farms. Additional analysis would be needed to rank solutions based on standardized criteria and assess the feasibility of grid-interactive efficient solar buildings compared to larger energy projects with suitable energy storage. Key performance indices such as life cycle cost, emissions analysis, and supply chain criticality should be appropriately weighed in the decision-making process. Energy efficiency is usually the low-hanging fruit capable of providing savings at lower investment costs than new renewable energy projects. If the analysis finds that investing in low-energy solar buildings is a suitable option, then policy and incentives (e.g., carbon or tax credits, subsidies) could be developed to expedite their implementation.
Every year, some residential buildings will be upgraded due to deterioration or the owner's personal choice. The no-regret solution may be to target those buildings. This can be done in a gradual and phased manner with policies aimed at incentivizing the houses that are on a priority list, such as houses with the worst energy performance. However, the additional costs for PV systems and advanced energy efficiency measures above code requirements may not be within the economic means and/or technical competencies of the building owner. To design and market such energy upgrades, new assistance, policy tools, and business models are needed. For instance, technical assistance that reduces uncertainties and financial support that would cover a portion or all of the incremental costs could help persuade building owners to select energy upgrades that benefit society the most. In addition, promoting densified rebuilds would help to reduce travel distances (a major contributor to mobility energy use) and free up valuable land that can serve for local food production and renaturalization. Québec has the opportunity to use its abundant hydropower resources more effectively by enabling building owners to become vectors of decarbonization while contributing to local food and energy security.
Author Contributions: J.B. carried out the work and wrote the paper. U.E. and A.K.A. did proofreading and corrections. All authors have read and agreed to the published version of the manuscript.

Funding:
The work was partially supported by NSERC research grants of A.K.A. and the Canada Excellence Research Chair (CERC) on smart, sustainable and resilient cities and communities held by U.E.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: The original contributions presented in the study are included in this article and the referenced articles by the same authors, further inquiries can be directed to the corresponding author.

Acknowledgments:
The authors acknowledge the financial support of the Concordia University Horizon Postdoctoral Fellowship to J.B.

Conflicts of Interest:
The authors declare no conflict of interest. Emissions factor of diesel-powered truck (kgCO 2eq L −1 ) E GH Electricity consumed per unit area of a greenhouse (kWh m −2 year −1 ) E GH Annual electricity that is available for greenhouses (kWh year −1 ) E GH_exist_per_house Annual thermal energy consumed by the existing greenhouses (in kWh year −1 ) E GH_heat Electricity consumed for heating per unit area of a greenhouse (kWh m −2 year −1 ) E GH_heat_tot Electricity that is used solely for heating the greenhouses (TWh year −1 ) Total number of personal vehicles P eff Maximum power freed up by implementing efficiency measures (kW) P EV EV power demand (kW) P GH Greenhouse peak power demand per unit area (kW m −2 ) P max Power availability from a house (kW) PT GHG Emissions payback time (yr) R BEV:FCEV Ratio of grid-to-wheel efficiency efficiencies R GHG Reduction in Québec's emissions (%) η Natural gas boiler efficiency (dimensionless)