Inputs
Four datasets are required. Two establish the historical relationship, one projects it forward, and one explains it.
1. Observed traffic speed and volume
| Source | National Highways WebTRIS |
| Period | 2022 to 2024 |
| Resolution | Hourly average speed and traffic volume |
| Coverage | Strategic Road Network sensor sites |
| Used for | Estimating the rainfall-speed relationship on each link |
This is the dependent side of the relationship: what the network actually did.
The same source as the gravity model
WebTRIS also supplies the base-year traffic volumes for the gravity model, over the same 2022 to 2024 window.
Using one observational source for both keeps the demand projection and the resilience estimation consistent with each other.
2. Observed rainfall
| Source | Met Office MIDAS-Open |
| Period | 2022 to 2024 |
| Resolution | Hourly rainfall intensity |
| Coverage | Gauges across the study area |
| Used for | Matching rainfall to each link |
This is the explanatory side: what the weather was doing at the same hours.
Matching gauges to links is a modelling choice
Rainfall is measured at gauges; speed is measured at sensor sites. Neither is where the other is.
The handbook states that gauge data are "matched to each link" without specifying the matching method. Since the estimated sensitivity depends on which rainfall series is paired with which speed series, the approach used is worth documenting.
MIDAS is also the source of secondary weather variables for the weather generator, so it appears on both the observational and the projection sides of the IMP.
3. Transport demand projection
| Source | Gravity model |
| Provides | Projected 2045 traffic volume for each site |
| Used for | Representing future demand in the forward-looking scenarios |
Documented in full at transport demand: inputs and outputs.
A gap between the demand and climate horizons
The overview states that the model is applied to extreme climate rainfall scenarios for 2030, 2045 and 2050. The demand projection listed here supplies 2045 only.
How demand is represented for the 2030 and 2050 scenarios is not stated. Either the gravity model produces additional horizons that are not listed, or 2045 demand is held constant across all three, which would be a material assumption worth recording.
This should be clarified in the source.
4. Road characteristics
| Source | National Highways open data, via the Travel Time Reporting Tool |
| Provides | Link-level attributes |
| Used in | The "Explain" stage of the framework |
These attributes account for why the sensitivity of speed to rainfall varies between links. The handbook's worked example: links with more lanes and higher speed limits are found to be more resilient.
See GLM Resilience Model for what that stage contributes.
Summary
| Input | Source | Role |
|---|---|---|
| Observed traffic speed and volume | WebTRIS | Estimate the relationship |
| Observed rainfall | MIDAS-Open | Estimate the relationship |
| Transport demand projection | Gravity model | Project it forward |
| Road characteristics | National Highways open data | Explain the variation |
All four are open data
Every input is publicly available: WebTRIS, MIDAS-Open and National Highways open data are all openly published.
That matters for the transferability claim made in the introduction. This component, at least, can be reproduced by anyone.