Draft climate change plan: draft third report on policies and proposals 2017-2032

15.1 Annex A: The TIMES model

Introduction

15.1.1 The development of the draft Climate Change Plan ( CCP) and Energy Strategy were informed by the use of a TIMES (The Integrated Markal EFOM System) for Scotland. The TIMES modelling approach was developed by the International Energy Agency - Energy Technology System Analysis Programme ( IEA- ETSAP) and is a Whole System Energy Model ( WSEM). Such models aim to capture the main characteristics of an energy system and the interlinkages within it. They are particularly useful for understanding the strategic choices that are required to decarbonise an economy. While this is the first time a model like this has been available for Scotland, they are widely used internationally in modelling climate and energy policy choices.

Figure 25: Countries which are members of the ETSAP modelling community or have used TIMES models

15.1.2 The Scottish TIMES model is a high level strategic model, covering the entire Scottish energy system, and containing thousands of variables capturing existing and future technologies and processes.

15.1.3 The model combines two different, and complementary, approaches to modelling energy: a technical engineering approach and an economic approach. The model uses this information to identify the effectiveness of carbon reduction measures in order to provide a consistent comparison of the costs of action across all sectors of the economy. The Scottish model was built by an international consortium of experts from E4TECH, E4SMA, KANORS, SYSTRA and Imperial College London and we are grateful for advice and assistance provided by the analytical team in the UK Department for Business, Energy and Industrial Strategy.

15.1.4 The aim of the model is to capture the main characteristics which effect the deployment of technologies, their costs and associated greenhouse gas emissions for Scotland as a whole given a range of policy and other constraints. This allows consideration of the strategic choices which Scotland faces as it seeks to decarbonise its energy system.

15.1.5 In building Scotland's first TIMES model we have involved both academic, non-governmental organisations and other independent experts in the project and we will continue to develop the model in partnership with ClimateXChange ( CXC) and the Scottish academic community.

15.1.6 This technical annex sets out our objectives in commissioning the model, an overview on how it functions and how it was used to inform the development of the CCP.

Objectives

15.1.7 The two previous Reports on Proposals and Policies ( RPP1 and RPP2) have necessarily been limited by our inability to consider in detail the system wide impacts of changes in one sector on other sectors. By using a whole system energy model, we are able to address this. The model limits us to the consideration of pathways that are internally consistent so a demand for say electricity in one sector needs to be met by increased generation and any fuel required for that additional electricity generation needs to be sourced from finite supplies.

15.1.8 For example biomass can be used to fulfil a wide variety of different energy needs depending upon how it is processed. In the TIMES model the availability of raw biomass is limited. As a result of this limitation the model will divert the available supply of biomass into its most economically valuable energy use and will ensure that total biomass use across all areas of the economy does not exceed the available supply. This ensures that decisions made in one part of the system have impacts which ripple out across the system and ensures that the resulting pathway proposed by the model is internally consistent.

15.1.9 The insights that this has provided on energy supply are explored in more detail in the complementary Energy Strategy.

15.1.10 As a result of taking this system wide view we have been able to:

• provide a clear roadmap to achieve required reductions in CO ₂ at minimum cost to government, industry and householders, whilst taking account of other policy priorities such as maintaining energy security
• consider a range of additional policy options, and to maximise the cost effectiveness of these, as required
• in effect, replicate a significant portion of the IPCC analysis for Scotland

15.1.11 This new approach results in a fundamental change in how we calculate sector emissions and consider cross-sectoral interactions compared to the two previous Reports on Proposals and Policies. The key differences are set out in the box below.

Text box 15-1: Difference between RPP and CCP

Difference in analytical approach between RPP and CCP

Previous approach: For the first two RPPs we did not have access to a whole-system energy model. As a result we estimated a top down 'business as usual' ( BAU) set of emissions for each year of the RPP, for each sector, and then netted off the estimated impact of policies and proposals to ensure our net emissions in each year were in line with the targets.

The key limitation with this approach was that, while sectors could consider where they individually felt they could best save carbon, they could not see how the costs of their efforts compared on a consistent basis with other sectors and what the wider system impacts might be of the package of policies and proposals as a whole.

Climate Change Plan approach: In the CCP we are able to take a different, more integrated, approach as a result of our TIMES whole-system energy model. Instead of focusing on future emissions on a sector by sector basis, and then netting off abatement, we can identify the most efficient parts of the system to remove carbon and allocate sectoral carbon envelopes accordingly. This bottom up approach allows us to ensure that cross sectoral effort in each year is in line both with the targets and with the effort in other sectors. This ensures that we take the most economically efficient decarbonisation pathway.

Modelling approach

15.1.12 The former Department for Energy and Climate Change ( DECC) built a TIMES model of the UK energy system. This UK TIMES model was used as a starting point for modelling the Scottish energy system. This will enable future linking between the two national models, and provided the core data sets required by the Scottish model. Using UK TIMES as a starting point was justified by the fact that technology costs and performance were found to be common in the majority of cases.

15.1.13 The modelling process was undertaken via the following steps:

1. Model Scope and Architecture

The architecture of the UK TIMES model was reviewed in order to provide a starting point for development of the Scottish TIMES model. The result of this review, alongside a review of available Scottish data was then considered by a Steering Group of officials and reviewed by sector experts.

2. Data Collection

Clarification of model architecture in (1) above led to the description of data required. This data collection step involved an iterative process of identifying the data required, appropriate data sources, data collection and checking. Iteration across this process was required where data was found to be either unavailable or of insufficient quality, leading to minor alterations to model architecture in outlying cases. The data collection phase culminated in a stakeholder data validation workshop. Data collection and improvement is an on-going process and we will continue to adjust the underpinning data as new information becomes available.

3. Sectoral Models Structural Analysis

The Scottish Government has historically employed a range of sector-specific models. This TIMES model development phase considered the structure of these models to ensure that cross-calibration between the whole-system and sector specific models is possible. This activity further refined the architecture of the Scottish TIMES model. TIMES drew on the following Scottish Government Models:

• the Scottish Electricity Dispatch Model ( SEDM)
• the Scottish Heat Map
• the National Housing Model ( NHM)
• the Transport Model for Scotland ( TMfS)
• the Scottish Government Heat Model

In addition analysts at Transport Scotland commissioned detailed work from Element Energy which informed the transport pathway.

4. TIMES Model Implementation and Calibration

The Scottish TIMES model was then built using a standard TIMES model development process that has been used to construct the models used in other countries. This entailed creation of a depiction of the 'base year' which represents a specific historical year to give model runs a starting point, and characterisation of the set of future technologies available to the model including their technical performance and economic parameters.

Scottish TIMES model architecture

15.1.14 The figure below sets out a simplified overview of the Scottish TIMES model.

Figure 26: Schematic of TIMES inputs and outputs; source (Remme et al., 2001)

15.1.15 The above diagram is a simplified representation of the energy system showing the following main parts:

• Resources (Red) - includes current and potential availability of 'traditional' energy sources such as oil and gas, but also includes sources such as wind, biomass, wave/tidal and solar. Trade is also included in this section.
• Conversion (Green) - includes the range of technologies and industrial processes used to convert resources into usable energy - from the production of petrol to the generators atop wind farms.
• Consumption (Purple) - details the range of potential uses (energy-services) for the energy being converted and distributed by the system, along with the technologies used to convert into a usable form, e.g. light, heat, computing and refrigeration.
• Demands (Light blue) - details a range of information shown to influence the amount of energy-services required. These are generally specified via other models, e.g. future population growth will influence many of these demands.

15.1.16 The model will always satisfy the demands we place on it subject to the availability of sufficient energy resources.

Exploring Scotland's future energy system

15.1.17 A key use of the model is to explore at a strategic level how the energy/climate system in Scotland might change as we move through the period of the Climate Change Plan and the companion Energy Strategy.

15.1.18 As TIMES looks into the future, existing technologies and processes will require replacement, possibly with more efficient alternatives, as they reach the end of their natural or economic lifetimes and new processes and technologies enter the market.

15.1.19 In responding to these changing conditions and user constraints, the model will make a series of investment decisions to ensure that demands for energy continue to be met. In effect, the model identifies the investments to be made at any given point in time which represent the lowest overall cost solution to meet demand subject to the constraint of delivering this within the emissions permitted by Scotland's climate change targets and other policy considerations.

Sector disaggregation

15.1.20 The Scottish TIMES model is designed following the same structure as UK TIMES, shown in the figure below.

Figure 27: Structure of the Scottish TIMES model

Resources and imports

15.1.21 The resources and imports sector includes all activities related to the trading of primary and secondary fuels into and out of Scotland. The model divides the sector into four subsectors:

• upstream oil and gas transport
• domestic resources
• exports
• imports

15.1.22 The energy resources are available to the model directly, through processing technologies or through infrastructure networks. The model aims to find the cost optimal mix of import, export and domestic creation of primary and secondary energy forms to meet the energy service demands. In future years new processes are available to the model which when implemented increase the import/export capacity of the model, and in the case of hydrogen infrastructure enable the distribution capacity of new energy carriers.

Upstream processing

15.1.23 The upstream sector includes the technologies that convert primary fuels into secondary fuels, excluding electricity and heat generation. The technologies available to the model can convert primary into secondary fuels or transform secondary fuels into other secondary fuels. Scottish TIMES splits the available process technologies into four categories:

• traditional processes such as oil-refining
• bioenergy and waste processes like anaerobic digestion of sewage sludge
• hydrogen production and infrastructure
• carbon capture and storage

15.1.24 The base year includes the traditional process such as oil-refining, and established biomass and waste processes such as landfill gas capture and biomass pelletisation. In the future years, multiple processes are available to the model, which include new feedstocks for fossil fuel production, new bioenergy technologies, and hydrogen technologies. The model also considers the transportation and sequestration potential required for carbon capture and storage from both electricity generation and industrial processes.

Electricity generation

15.1.25 The model treats electricity differently to other energy carriers due to the physical requirements for system balancing and the relative difficulty of electricity storage. Electricity demand is endogenous because of electricity's competition with other fuels to meet the energy services demands. As a result, the load profile for electricity demand will vary with each model run and the electricity sector must meet the peak demand at every period and time-slice, as well as a reserve capacity. A large suite of generation technologies is available to the model to meet the electricity demand each with a set of parameters and constraints. These parameters and constraints are grouped as technical, economic and environmental. The model finds the cost optimal mix of suitable generation technologies to satisfy the electricity demand for each run of the model.

15.1.26 The base year capacity of the generation technologies, broadly categorised as fossil fuel, nuclear and renewable, are calibrated to the actual installed capacity in Scotland in 2012. The future installed capacity of these technologies is estimated according to the current planned or under construction generation projects and constraints on deployment potential.

15.1.27 As the TIMES model is strategic in nature it looks at time periods in blocks rather than chronologically, which allows for manageable model run times while still providing valuable insights. It is the nature of the electricity system however that supply and demand have to match in real time. In order to check that the generation mix being proposed by the selected TIMES scenario fulfilled this criteria it was stress tested using the Scottish Government Electricity Dispatch Model ( SEDM). This confirmed that the proposed capacities are technically feasible, alongside use of interconnection with the integrated GB electricity network, relying on existing and under development network projects.

Transport

15.1.28 The transport sector contains the processes related to the movement of goods and people and the required infrastructure and fuel. The sector is composed of 9 subsectors:

• Car
• Bus
• Light goods vehicles
• Heavy goods vehicle
• Motor cycles
• Passenger rail
• Freight rail
• Aviation
• Shipping

15.1.29 The model defines the demands for each of these sectors including variations in car trip length and domestic and international demand for aviation and shipping. The model optimises the cost of the transport infrastructure, the energy carriers and the transport technologies to meet the transport demand whilst adhering to the environmental parameters. Parameters and constraints are applied to infrastructure, energy carrier and technologies.

15.1.30 During the course of the project it became clear that Transport Scotland's commissioned research undertaken by Element Energy offered additional detail over the core TIMES transport module and as such it was used to develop the transport envelopes. Element Energy's work included the construction of a detailed Scottish vehicle database, including number, age and turnover of the Scottish vehicle parc, a peer review of their database of technology costs across transport modes (used to refine their cost models for alternative fuels and technologies) and finally used their consumer demand model to develop realistic scenarios for the take up of alternatively fuelled vehicles, particularly cars. These emissions envelopes resulting from the analysis were then coded into the TIMES run so that their wider impact on the energy system would be costed and considered.

Residential

15.1.31 The residential sector includes all energy service demands and emissions from the Scottish housing stock. Scottish TIMES splits the sector into three broad demand sub-sectors:

• heating (space and hot water heating), using a range of fuels
• cooking (hobs, ovens and other cooking), using a range of fuels
• other demands (lighting, refrigerators/freezers, space cooling, wet appliances, consumer electronics, computers and other appliances), which consume electricity

15.1.32 Assumptions are made for each of the above energy service demands in the base year for urban houses, rural houses and flats. The only demands for which a distinction is made between existing and new dwellings are the space and hot water heating demands. For all other energy service demands, dwellings of all ages are combined, but the distinction between urban/rural/flat is maintained.

15.1.33 The National Housing Model attempts to completely represent all individual housing types and, as such, a very large number of housing 'archetypes' can be extracted from the model. For computational simplicity in TIMES, and following a statistical review, these were aggregated into 6 different types which captured the bulk of statistical variability.

Industry

15.1.34 The industrial sector module in the Scottish TIMES model is composed of six subsectors. The industry sector in Scotland is currently dominated (in emissions terms) by chemicals, with smaller contributions from food and drink, cement, non-ferrous metals, steel works and a range of other small industries. Scottish TIMES uses a process-oriented approach to model energy-intensive industry sectors including iron and steel, cement, pulp and paper. This means that the actual production processes are represented in the model.

15.1.35 The chemicals sector is highly heterogeneous. The high value chemicals and ammonia subsectors are also modelled in a process-based manner and take into account the different process related emissions. The energy services demands are the basis for the production processes of other chemicals.

15.1.36 The remaining, less energy intensive, industrial subsectors such as other chemicals, food and drink as well as other industries and which have a highly heterogeneous production structure, are modelled using an end-use or energy service demand approach. The model optimizes the combination of process technologies to meet the specified level of energy service demand. The final energy service demands are met through the transformation of energy carriers by the process technologies.

Services

15.1.37 The service sector is broadly defined in the same way as the residential sector. The primary energy service demands are space heating, water heating, lighting, computer, cooking, refrigeration, and other electricity. The sector is divided into public and private buildings and within each of those sectors the building stock is split into high and low energy consumption buildings. The technology options available to the model are very similar to those available in UK TIMES with the base year stock adapted to reflect the current capacity in Scotland. Data is currently not available on the split of public to private buildings or the total energy demand in the Scottish service sector. They are estimated based on the current demands estimated in UK TIMES and the public/private sector building greenhouse gas emissions in Scotland from the Greenhouse Gas Inventory.

Agriculture

15.1.38 The agriculture sector includes all energy consuming processes related to agriculture, and land-use and agriculture-related non- CO ₂ non-energy greenhouse gas emissions mitigation technologies. Relative to the UK, agriculture is a more important sector in Scotland and has a significant emissions profile. Scottish TIMES considers the energy demands for heat, electricity and transport related to agriculture activities. The model includes greenhouse gas emissions mitigation measures related to agricultural practices and new technologies. The agriculture emissions profile was calculated exogenously following discussion with sector experts and mapped into Scottish TIMES.

Land Use, Land Use Change and Forestry

15.1.39 The model includes greenhouse gas emissions from land use; in particular the carbon sink effects of afforestation and peatland restoration are included. As a result of historic tree planting rates the size of the carbon sink associated with land-use in Scotland is reducing. This basic profile, provided by the Centre for Ecology and Hydrology ^[70] , is included in TIMES and additional abatement associated with increasing tree planting and peatland restoration is added as a policy option.

Data sources

15.1.40 In populating the model with data and assumptions, Scottish-specific references were used as far as possible. Given that many of the processes and technologies defined within the model align with those used in UK TIMES, most of the cost and performance characteristics were aligned to UK TIMES. Two key Scottish sources were used in calibrating the base year stock and commodity flows, in particular the energy balance from the Energy in Scotland publication ^[71] and the Scottish Greenhouse Gas Emissions publication ^[72] . These allowed for a top-down calibration of commodity flows through each supply and demand sector, as well as an estimate of the installed capacity of technologies in each sector.

15.1.41 The key sources of reference used in each sector are highlighted below:

Resources: Most of the resource flow data (i.e. imports, exports, indigenous production) for the base year is derived from the 2012 Energy Balance ^[73] . Scottish specific sources were used for fossil fuel reserve potentials, biomass potentials and waste potentials, and are summarised in the Resource sector document. Where data on future resource potentials were not available data from UK TIMES was scaled for Scotland.

Upstream processing: The main processes included in this sector are the refinery activities at Grangemouth, biodiesel production, and various pellet production activities. Publically available estimates of Grangemouth refinery product outputs from the UKPIA ^[74] were used along with UK TIMES data on costs and future technology options. Publically available reports from the Forestry Commission and Ecofys were used to estimate domestic biofuel production capacity and potential. The assumptions about availability of imported biomass were set following discussion with BEIS.

Electricity: The majority of the data for the Electricity sector was taken from the SEDM ^[75] . Other data on generation and capacity was taken from DECC ^[76] . Technology cost and performance characteristics were mostly taken from UK TIMES.

Transport: Data on stock and performance was taken from Scottish Transport Statistics and the TMfS. Future technology performance and costs were mainly taken from UK TIMES.

Residential: Data from the Scottish Household Conditions Survey ( SHCS) ^[77] was used to characterise the base year Scottish building stock energy demands and technology stock.

Industry: Commodity production data was taken from various industry sources. A lack of data on the existing installed technology capacities in each sector meant that estimates had to be made using the emission data for each sector in the Scottish Greenhouse Gas Emissions publication ^[78] . UK TIMES technology cost and performance characteristics were used.

Services: A lack of data in the services sector meant that UK TIMES assumptions had to be scaled based on the estimated number of service sector buildings in Scotland compared to the rest of the UK.

Agriculture and land-use: The Scottish agricultural MAC curve was initially used ^[79] to estimate the mitigation potential for various measures in Scottish agriculture. Following discussions with sector experts this was replaced with an exogenous profile. The BAU land-use emissions, waste emissions and agricultural process emissions came from a mix of sources compiled and adapted by E4tech ^[80] . Agriculture sector energy demands (e.g. heat, electricity, transport) were estimated by scaling the UK TIMES assumptions for Scotland.

Methodology for projecting service demands

15.1.42 The TIMES models are driven by projections of energy service demands that are in turn based on macro-economic drivers and elasticities. The Scottish TIMES model is driven by exogenous demand specified by the list of each energy service demands (92 split by sector and service), actual values in the base year (calibration) and values for all milestone years till 2050 (projection).

15.1.43 The energy service demand projection in the model (for example amount of car road travel in passenger kilometres, residential lighting, hot water requirements of the services sector, cement production) over the horizon require two sets of parameters: base year demand value and demand drivers. Demand drivers (for example population, GDP, number of households) are usually exogenously obtained via other models or from accepted other sources. The Scottish TIMES model uses exogenous drivers based on macro-economic projections, and where not available data from the UK TIMES model. For each demand a possible driver for the data projection has been identified, based on both the available macro-economic drivers and/or the UK TIMES drivers. These drivers have been used to calculate demands driver rate by period ( DDR(t)) and then the energy service demand projection by period based on the following formulas:

DDR(t) demand driver in period t.
Demand Driver (t) is the demand driver in the actual period t.
Demand driver (t0) is the demand driver in the base year period

Demand( t) = Demand( t0)* (1+ DDR( t)) Equation 2

Demand(t) is the demand in period t.
Demand(t0) is the demand in the base year period.

How the model was used

15.1.44 While TIMES is a powerful tool for considering the implications of changes in the energy system on emissions it is simply a guide rather than a predictor of the future. As such an important part of the modelling process has been engagement with sector experts and consideration of the wider consequences of particular pathways.

15.1.45 This engagement was carried out via both analytical working groups and the Senior Suppliers Group of government officials. As a result of this engagement a number of delivery considerations were identified which resulted in additional constraints being placed on the model.

15.1.46 This process is summarised graphically below.

15.1.47 Following consideration by the Scottish Government's Cabinet Sub-Committee on Climate Change, sectors were provided with their final high level carbon envelopes. The draft Climate Change Plan explains how it is intended that each sector can meet its envelope.

15.1.48 Given the central role of energy in delivering a low carbon Scotland and the extent of the challenges faced in delivering low carbon energy, a separate Energy Strategy also informed by the TIMES model is being published as a companion document to the draft Climate Change Plan. This explores the options for delivering increasingly decarbonised electricity and heat and the technological options for doing so.