Logic trees are a key foundation in the development of a 2050 Calculator. What is a logic tree, and why they are so important to develop calculators?
Written by Charles Vander Linden and John Watterson
Logic trees are a key foundation in the development of a 2050 Calculator. We’ll look at an example of a logic tree later in this article, but let’s start by explaining what a logic tree is, and why they are so important in the development of country-specific Calculators. We’ll also show how logic trees connect to other key parts of the Calculator – essentially the levers which can be used to control supply and demand.
If you’ve read the ‘creating a Calculator’ guidance, you may have seen the term ‘driver tree’. Driver trees are another way of displaying the data and calculations you’ll need in the Calculator model. Logic trees are equivalent to driver trees.
Developing logic trees will simplify the development of your 2050 Calculator. They allow a visualisation of the major logic elements of the calculations needed in the model. The visualisation will help you think through the calculations necessary, while the diagram can also be used to increase stakeholder insight and understanding. If you get the logic trees right, building your model will be a lot easier.
A logic tree diagram will help stakeholders to understand the inner workings of sectoral elements of the Calculator, and how sector policies to control emissions could act. This insight and understanding will help stakeholder engagement.
Levers
It’s important to understand how levers and logic/driver trees interact. Once you have a logic tree for each of your sectors, you can quickly see how changes in supply and demand affect your outputs. Next, you need to choose which factors you think are the major drivers of change in the sector. This dictates which levers you’ll use in the model. Each lever allows users to make a choice about what happens in the future.
How should logic trees develop? Who should be involved?
The development of logic trees is the first step in building a Calculator. Modellers decide what the historical data input will be; what levers will be used; and how the historical data and levers will influence energy consumption and related emissions. Depending on the objective and type of Calculator being built, it’s important to choose appropriate historical data input; for example, if the model needs to be used for different countries, input data needs to be easily available for all countries and can’t be too specific.
Modellers also define the main assumptions to estimate the emissions of the different sectors as efficiently as possible at this stage. Consultation with strategic stakeholders is key to validate the different decisions and assumptions.
How long do logic trees take?
There are no rules on the time it takes to develop a logic tree; rather it depends on the type of Calculator being built – are all sectors of the economy being modelled? Is it country, regional or continental scale? It’s an important step, however, and should not be neglected for the reasons mentioned earlier.
Logic tree example for a transport system
The diagram below shows what a logic tree could look like for a transport system:
Understanding the logic of the calculations used allows identification of the input data a country will need to provide (orange cells in the figure) as well as the lever values it will need to define (red cells in the figure).
To start, you’ll need to compile a list of the engine and technology types that are likely to be used in your country, both now and in the future. Don’t be afraid to use expert judgement – you can always consult with peers or stakeholders to refine your list. In the transport example above, the specific list of inputs that need to be complied are on the right side of the diagram:
- Share of engine type according to mode (%)
- Engine intensity of use per mode, per engine type and per technology (TWh/km)
- Biofuel share of liquid fuel by mode in the base year (% of total fuel used)
- Change in biofuel share over time.
For non-transport sectors, the same general principle applies. Consider what the current and future technologies or options are, and what data you’ll need to have access to, and make a list.
Next, start to build the logic tree diagram – normally from the top, downwards. You don’t have to construct the diagram in this way, but it’s important to provide a clear narrative, so the reader can understand how the logic tree was developed. This will ensure that your communications are transparent.
Turning back to the transport example above, the last box at the bottom of the diagram shows the end point of the calculations – the greenhouse gas emissions per mode, engine type and technology in units of tonnes CO2 equivalent. To reach this calculation, the logic tree begins on the first line of the diagram with two groups of calculations: 1) the product of vehicle distance and share of engine type per mode; and 2) biofuel share for the base year and change in biofuel mode.
For the transport sector, it’s important to consider biofuel separately (note biofuel may be a blend of fossil fuel and biocarbon fuel) as the CO2 emissions from the use of biocarbon can be estimated and reported, but must not be included in national greenhouse gas total emissions. The non-CO2 emissions from biofuel use should be included in national total emissions.
The second line of the logic tree shows the next step in the calculation. Here, the calculation is the product of three things: 1) vehicle distance travelled per mode and per engine type; 2) energy intensity per mode, per engine type and per technology; and 3) biofuel share of liquid fuel per mode.
The third line of the calculations is the product of two terms: 1) energy balance per mode, per engine and per technology; and 2) emission factors applicable for the technology. If you don’t have access to the country-specific emission factors, you can use default Intergovernmental Panel on Climate Change (IPCC) factors. For a key category in your greenhouse gas inventory, you may wish to consider developing country-specific emission factors in the future.
The product of the two terms leads to the greenhouse gas emissions per mode, engine type and technology in units of tonnes CO2 equivalent. This is what we need to know. It provides the information needed to start to understand the key sources of emissions in the transport sector and allows planning of mitigation to tackle the sources. The lever values applicable to mitigation that could be built into the Calculator in this example are:
- Share of engine type according to mode (%)
- Change in biofuel share
- Engine intensity of use per mode, per engine type and per technology (TWh/km).
Understanding these calculations in logic trees is one of the first steps required to integrate the UK’s MacKay Calculator into a country.
Do logic trees work?
Yes! This is a testimonial from one of the team members during a recent training session of the Malaysia Climate Action Simulator (MCAS): “Logic trees simplify the complicated structure of modelling which helps the developer of the carbon Calculator to easily understand the calculations involved across the sectors and sub-sectors and the data needed for the model.”
Can you get help with developing logic trees?
Yes. The UK-funded 2050 Calculator programme not only supports the creation of country Calculators; it also provides ongoing support to existing Calculator teams, including for logic trees.
Further support
beis2050calculator@mottmac.com
Written by Charles Vander Linden from Climact and John Watterson from Ricardo.
Article text (excluding photos or graphics) © Imperial College London.
Photos and graphics subject to third party copyright used with permission or © Imperial College London.
