Fossil reserves are limited and finite and dependence on fossil-based energy should be minimized. In addition, recent climate change and global warming trends are believed to be due to the anthropogenic burning of fossil fuels. The target is to limit global warming to 2 °C above pre-industrial levels by 2050 and this requires keeping the concentration of CO₂ below 400 parts per million. To achieve this an 80% reduction in emissions is needed [1]. According to the UK government’s Department of Energy and Climate Change (DECC), in 2009, the residential sector accounted for 27% of final-user emissions. In order to meet the national target of 80% CO₂ reductions by 2050, and to react to higher energy prices, it is critical that energy conservation in dwellings is a major performance objective in the design or rehabilitation of buildings. One step towards this objective in the UK is the requirement that from 2016 all new dwellings will have to be “zero carbon”. Due to the very low numbers of new dwellings that are built each year the impact of this requirement will be incremental rather than immediately significant. Since 70% of all current homes in the UK will still exist in 2050, the drive to reduce carbon emissions from housing has shifted from new-build to retrofit. As a result, eco-refurbishment is critical if reduction goals are to be met [2].

This study considers the eco-refurbishment of a 19th century solid wall end-terraced house. The aim of the eco-refurbishment was to go beyond current UK thermal building regulations criteria and to try and achieve the more exacting German Passivhaus standard. This standard sets strict limits on the overall U-value of the external envelope (≤ 0.15 W/m²K); airtightness@50Pa (≤ 0.6 air changes per hour@50Pa and the use of mechanical ventilation with heat recovery); the elimination of cold bridges and the overall energy use for space heating (≤ 15 kWh/m² of floor area per year). Compliance with the Passivhaus standard is checked by assessing the building using the Passivhaus Planning Package (PHPP) [3]. It is has been estimated that applying the Passivhaus standard to a typical UK building would reduce overall CO₂ emissions by approximately 80% [4].

In the context of carbon reduction, eco-refurbishment is about making buildings more thermally efficient and sustainable. The emphasis is on energy services, rather than energy purchase. It is about shifting the debate on to demand reduction rather than energy supply. As with the Passivhaus standard, eco-refurbishment also concentrates on three areas: improving the air-tightness of the house, reducing heat bridges, and, to a lesser extent, considering heat gain from the sun. In addition, eco-refurbishment means having energy services that use fuels with low carbon intensity and incorporating building integrated renewable energy systems. Essentially, a typical eco-refurbished house is air-tight and super-insulated, with mechanical ventilation and heat recovery (MVHR) to solve the issue of internal air quality and with a small heating element within the ventilation system. In addition, generally, windows are triple glazed [5]. The overall aim of eco-refurbishment is to find methods by which the UK government’s commitment to an 80% reduction in CO₂ emissions by 2050 (compared to 1990 levels) might be met within the existing housing stock.

The dwelling used for the eco-refurbishment study was a solid wall, 19th century terraced house in Liverpool. In March 2009 the UK’s Technology Strategy Board (TSB) launched a competition, Retrofit for the Future, to encourage and enable building and renovation companies to retrofit and refurbish existing housing to make deep cuts in carbon emissions [6]. The Retrofit for the Future targets were linked to the Climate Change Act and, for the purpose of the competition, the following targets were set:

Maximum CO₂ emissions—17 kg/m²yr;

The primary energy consumption figure includes all the energy used in the house, including that for appliances (white goods) and consumer electronics [6]. The Retrofit for the Future competition was looking for creativity in developing 100 “exemplar” retrofit projects across the UK. The TSB offered 100% funding to each proposal and eventually 87 schemes were approved. Six were in the North West of England, including the terraced house, which was chosen as the case study for this research project [7]. Figure 1, Figure 2 show the selected case study which was refurbished by the Plus Dane Group with the aim of achieving Passivhaus standards as far as possible by applying 21st century solutions to a 19th century end terraced house. Passivhaus and/or sustainability principles were introduced in the demonstration project. The retrofit of the building included: very high levels of insulation, timber framed triple glazed windows, very high levels of airtightness@50Pa (1.0 ACH@50Pa), mechanical ventilation with heat recovery, solar gain via a new conservatory, LED lighting in the kitchen and bathroom, and A-rated low energy appliances [7].

There were several reasons for choosing this 19th century terraced house as a case study of the feasibility of eco-refurbishment to a Passivhaus standard. Firstly, although over 20,000 Passivhaus accredited properties exist, the UK has only three of them. Secondly, Passivhaus refurbishment has not been tried before in the UK on an old terraced house. Thirdly, virtually all old terraced housing has a single skin, uninsulated solid wall structure, making them hard and expensive to treat and therefore not many have been subject to major refurbishment [8]. Finally, with more than 6 million 19th century terrace houses in the UK, they represent an important part of the country’s housing resource and architectural heritage. Consequently, research on the experience of Passivhaus and low energy retrofit of such buildings can examine the challenges and opportunities of significant retrofits in conservation areas.

The primary objective of this paper was to evaluate the benefits of sustainable refurbishment on the energy performance of 19th century type end-terrace houses under current and future climates. To do so a regional approach was adopted to assess innovations in the eco-refurbishment agenda. Locally and regionally, UK housing has quite different characteristics. There is also a geographical diversity, which causes different energy demands in different parts of the UK. Heating demand is typically higher in Scottish cities than in, say, Manchester, whilst the changing climate may lead to a significant increase in demand for summer cooling in London and the south east of the UK, but not necessarily elsewhere. A second objective of this study was to consider how to apply the knowledge and technologies of eco-refurbishment in the context of future climates and to make well-founded decisions based on these considerations. A final objective was to see how savings (which are increased by reusing and refurbishing an existing building) might relate to the operational costs during the building’s life cycle.

The 19th century type end-terraced house was modeled (before and after refurbishment) using the advanced thermal simulation package DesignBuilder. As there was not enough information about the previous condition of the house the characteristics of the typical terraced houses in the neighborhood were studied to create a pre-refurbishment model of the existing house in the chosen software (DesignBuilder). Table 1, Table 2 compare the pre and post refurbishment thermal features of the house. The occupancy densities and occupied period profiles were selected based on the real house schedules obtained from monitoring results for a family of four (two adults and two children). In addition, internal heat gain allowances were selected according to the activity of each zone in the building. Then, simulations were carried out with all treated zones assumed to be kept to a nominal set point temperature of 21 °C and heated to set points of between 19 °C and 21 °C (depending on the nature of usage). Current weather data for Liverpool were used to evaluate the effect of eco-refurbishment on energy consumption and CO₂ emissions.

In addition, the energy and CO₂ performances of the same pre and post refurbishment dwellings were examined for current and future climates at three other cities across the UK.

The Intergovernmental Panel on Climate Change (IPCC) stated in its Technical Summary for their Fourth Assessment Report that an increase in global average surface temperature would occur towards the end of the 21 century [11]. For UK specific climate projections the UK Climate Impacts Programme (UKCIP), a UK based agency, in conjunction with the UK’s Meteorological Office’s Hadley Centre, published its first climate change projections in 1998, which were substantially updated in 2002 and again in 2009. For the purpose of this research climate change scenarios and future hourly weather data released by the UK’s Charted Institution of Building Services Engineers (CIBSE), which incorporated the UKCIP02 projections for three timelines (2020s, 2050s, 2080s) and four carbon emissions scenarios (low, medium-low, medium-high and high), have been used. These data are in TRY (Test Reference Year) format but need to be generated in the EPW (EnergyPlus Weather) format to be readable in DesignBuilder. No TRY files existed for Liverpool at the time of this study and only current weather data in the EPW format were available. However, current TRY weather data were available for three UK locations (Edinburgh, Manchester and London). These TRY files were morphed in to future weather data using the software CCWeatherGen [12] (time slices: 2020s, 2050s, 2080s; carbon emission scenarios: low, medium and high). Therefore, the same dwelling was assumed to be located in these three cities and simulated using these weather data to model the effect of climate change and geographical diversity on CO₂ emissions, heating and cooling energy usage. In this paper only the results from 2020 to 2050 high emission scenarios will be presented.

CCWeatherGen transforms CIBSE/Met Office TRY/DSY weather files into future EPW weather files [12]. UKCIP02 data were used for this research because, at the time of the analysis, CCWeatherGen used the UKCIP2002 climate change scenario predictions to generate future weather files that were ready to use in building performance simulation programs (i.e., in EPW format). The PROMETHEUS project in Exeter University has recently produced a number of EPW future weather files using the UKCP09 weather generator, which can be used to “future-proof” buildings against predicted climate change. The key difference between UKCP02 and UKCP09 is the use of probabilistic information within UKCP09, which represents a random sampling of a probability distribution function and hence the probability of a particular level of climate change [13].

Simulation software is a useful tool to study the impact of climate change on building energy performance. However, it is important to validate the results from simulation to ensure confidence in the results. The Plus Dane Group used the Passive House Planning Package (PHPP) for its simulation of the terraced house (Table 3). Their predicted results were used for initial validation of the model created in DesignBuilder. This preliminary comparison showed a small percentage difference between the PHPP and DesignBuilder outcomes.

In addition, the Plus Dane Group has been monitoring many parameters in the house at five minute intervals since it was first occupied after refurbishment in November 2010. Large amounts of data have been measured such as internal carbon dioxide levels, floor, wall and ceiling surface temperatures, external air temperatures and electricity, water and gas consumption. One complete year of monitored data was made available for this study. These data were compared with the DesignBuilder simulations as part of the validation procedure. To start the validation a comparison of the weather data was necessary. DesignBuilder simulation results were based on 2002 CIBSE weather data whilst the monitoring was carried out between November 2010 and October 2011. Figure 3 shows 2002 (CIBSE) average monthly external air temperatures with 2010 (measured) weather data for Liverpool. There is generally reasonable agreement but it can be seen that the average temperatures generally increased over the period.

A comparison of monitored and predicted annual CO₂ emissions and primary energy requirements can be seen in Table 4. There are small differences between the results, which could be attributed to the small differences in the weather data.

Having validated the program made it possible to investigate how the same refurbished terraced house might perform in different current and future climate scenarios.

In the process of fulfilling an 80% reduction goal and meeting eco-refurbishment benchmarks, one of the most important questions is that of economic feasibility. The question arises due to several potential alternative investment options and the importance of finding the most economical choice in the long term. In this paper a Whole Life Cycle Costing (WLCC) technique has been used for comparative cost assessments over a 40 year period of time, taking into account the present value of initial capital costs, future operational costs and savings from this type of eco-refurbishment.

Pre and post-refurbished houses in different cities under current and future weather condition and different scenarios were simulated to see if the UK’s 80% carbon reduction target was achievable by refurbishment. Simulation results given in Figure 7 show the annual heating demand for the pre and post-refurbished houses in different cities and different scenarios. Significant heating demand reductions can be seen after refurbishment in all four cities. Figure 7 also confirms that, despite the fact that heating energy consumptions are affected by climate change, there is very little decline in future heating demand for the house with no refurbishment, while the eco-refurbished dwelling shows a sharp reduction of more than 70% in energy demand.

Changes of CO₂ emissions in London, Liverpool, Manchester and Edinburgh for pre and post-refurbished house are shown in Figure 8. It can be seen that there is a large reduction of around 83% for the house when located in Liverpool and London and 80% reduction in CO₂ emissions for the house in Manchester and Edinburgh following refurbishment. CO₂ emissions are affected by climate change and show a slight downward trend for the house with no refurbishment, but the eco-refurbished dwellings show sharp reduction in CO₂ emissions.

Figure 9 compares the operative temperature (the average of the internal air and mean radiant temperatures) of the post and pre refurbished house with natural ventilation. It can be concluded from the results that during the summer months the mean peak operative temperature may exceed 25 °C after refurbishment and so there could be a risk of thermal discomfort if no adaptive measures were taken. However, for most of the year the refurbishment is greatly enhancing thermal comfort in the house. There is a suggestion from Figure 9 that some take-back could occur in the refurbished house i.e., occupants enjoying a higher internal temperature in the heating season than they might have experienced in the pre refurbished house.

Although results indicate that an 80% reduction in CO₂ emissions is achievable by refurbishment, it is important to consider the time and costs required to reach that target. Figure 10 illustrates typical heat losses from each major construction element in the terraced house, with the main sources of heat loss being the walls and loft. Consequently, improving wall and loft insulation levels can be the most effective way of improving the energy performance of buildings. Therefore, it would useful to also look at the economic performance of these components. Figure 11 indicates the cost of applying each technology.

Energy savings cost is the difference between DesignBuilder simulation results for the heating consumption of the existing house and the retrofit house. Table 7 illustrates the average internal temperature of the living room during the heating season (December, January, February) as modeled by DesignBuilder(DB) for pre and post refurbished house compared to monitoring results.

As gas is the primary energy source for heating in this case study, the energy cost is calculated by use of a mean gas price of 0.0475 £/kWh. The annual benefit (the financial saving) is calculated by multiplying energy saving by current gas price of 0.0475 £/kWh for the first year in the cash flow. The numerical results for different cities are listed in Table 8.

To calculate the future annual savings three factors must be considered: effect of climate change on annual savings, instability of future gas prices and time preference and discount rate.

Climate change affects heating consumption and accordingly the annual energy savings. Table 9 shows the reduction of heating consumption for different climate scenarios and different cities.

As any cost benefit analysis is performed over the lifetime of a house, a changing rate has to be introduced in the cost calculations to correspond with reality. Looking at historical data suggests that not only will energy prices not remain constant but that there is also a possibility of both downward and upward trends as well. Therefore, different scenarios should be investigated that take in to account different trends. To measure the amount of uncertainty and to predict gas price trends 16 years of gas price datasets were used and the change over time (standard deviation) was calculated (Table 10). The instability of the annual variation was calculated in this case to be 8.6% (see Equation 3).

(3)

Finding the ratio that controls the rising and falling values allows a binomial tree of gas price trends to be constructed for the next 16 years. To predict the gas prices for year 16–40 the standard deviation was calculated again using years 1 to 16 predicted gas prices and was found to be 23%.

Figure 12 illustrates the predicted different possible gas prices in each year. The expected gas price in each year is the sum of all possibilities weighted by their probabilities. The right vertical axis shows the predicted prices for an upward trend and the left vertical axis illustrates the projected gas prices for a downward and current trend. After each 16 years, the previous 16 years of data were used to predict the new gas prices. For a better comparison, a different scale has been chosen on the right hand side for the upward trend. Expected gas prices in the next 40 years can be seen in Table 11.

Estimating future gas prices makes it possible to calculate savings for the building’s energy expenditure. As these savings happen for different operations taking place at different times, it is essential to convert to current costs using a discounted cash flow method that incorporates the discount rate of 3.5% for the first 30 years and 3% thereafter (Table 11 shows the values for London for the period 2010–2020).

Discounted accumulated whole life cycle costs over 40 years for three gas prices and three weather scenarios are shown in Figure 13, Figure 14, Figure 15 for the terraced house located in London, Manchester and Edinburgh. A comparison of WLCC for all cities shows that this type of refurbishment is economically attractive only with the rising gas prices scenario. Clearly, adding the maintenance and replacement costs to the analysis will change these findings dramatically in favor of the refurbished house’s WLCC.

Comparing the most favorable outcomes (with upward prices) for all three cities it can be seen in Figure 16 that the total present value for Edinburgh is the most beneficial option over 40 years (£ 178,752 savings), before Manchester (£ 154,208 savings) and London (£ 138,090 savings), for the given conditions (3% discount rate, 8.6% yearly increase in energy costs for the first 16 years and 23% increase in energy costs after that).