2.2 Seed data and assumptions

2.2.1 The need for seed data

1. An hourly model must do its best to represent a realistic scenario hour by hour. It will need to either accept hourly seed data or generate its own.
2. It is not realistic to generate the data. Key factors like weather and demand are neither random nor regular. They are not unrelated but not determinant (e.g. temperature and wind affect demand, but so do other human factors). The output of any model based on artificially-generated data would primarily reflect the assumptions used to generate that data.
3. The seed data must be based on historic data to provide a realistic pattern of irregular variability amongst several key components, which creates the challenge to balance supply and demand.
4. That historic data must be used raw and not aggregated to produce “typical” figures for each period. The aggregate would not be typical. It would be an average that radically dampened the inter- and intra-temporal variability that represent the key challenge in real-life operation.
5. The model must encompass all energy, not just electricity as many previous models have done. Net Zero is about recognising that decarbonisation has to cover a lot more than the electricity sector.
6. Many of the favoured solutions (e.g. electrification, hydrogen and bioenergy) create large overlaps between what were previously regarded as largely discrete sectors: electricity, transport, heat and non-energy carbon sources. The model and the seed data need to encompass these overlaps, to allow for the allocation of scarce means between alternative uses.[1]
7. Which factors are primarily exogenous and require seed data?