1  Introduction

1.1 Summary

Objectives Develop modelling system to quantify features of land use in urban environment
Output

Prediction of the quality of life indicators following modelled scenarios of development

Our case study is Tyne and Wear county, comprising the following local authorities: Gateshead, Newcastle upon Tyne, North Tyneside, South Tyneside, and Sunderland.
How Creating key indicators for assessing a baseline scenario and future scenarios
Duration September 2022 – present

1.2 Project aims

The project aims to provide insight into the impact of land use policies in cities across the UK, piloting on the case of Tyne and Wear. It:

  1. Derives indicators of quality of life.
  2. Develops machine learning models which can predict the impact of land use changes on such indicators.
  3. Inversely, develops a neural network able to predict the required land use change to reach target levels of quality of life indicators.

All these technological components are presented in an interactive tool allowing quick and easy exploration of impacts aimed at policymakers.

1.3 Quality of life indicators

The project defines four indicators related to the quality of life, which are intended to capture selected dimensions of the environment, society and economy:

  1. Air pollution
  2. House prices
  3. Job accessibility
  4. Greenspace accessibility

Air pollution and house prices are directly derived from observed values. For these two indicators, we developed learning-based, predictive models which assess the impact of of land use changes. These models are based on land use variables derived from the Urban Grammar project.

On the other hand, both accessibility metrics (job and greenspace) were generated for this project. These were calculated for four modes of transport (walking, bicycle, vehicles, and public transit), between a relevant set of origins and destinations at the UK 2011 Census Output Area level.

More details about the indicators, as well as the explanatory variables used in the model, can be found in the Methodology chapter.

1.4 Development scenarios

The project is done in collaboration with the Geospatial Commission and Newcastle City Council (NCC). It defines development scenarios of the Tyne and Wear county, for which it reports predicted changes of selected indicators, allowing assessment of a proposed land use change based on machine learning.

The indicators are first modelled for a baseline, reflecting the current land use in the area under study. Once the baseline is created, the project defines a number of scenarios of future development (e.g. densification of the city centre, or land release in the green belt area) and assesses the effect of those scenarios on the quality of life indicators.

See the chapter on scenarios for more details.

1.5 Interactive web app

As part of the project, an interactive web app has also been created. The web app allows users to project both the proposed land use, as well as the predicted values of the quality-of-life indicators, onto a map of Newcastle. Policymakers and professionals will thus be able to explore the outputs of this project by visualising and comparing different scenarios.