Cycling in a changed climate

摘要

The use of bicycle is substantially affected by the weather patterns, which is expected to change in the future as a result of climate change. It is therefore important to understand the resulting potential changes in bicycle flows in order to accommodate adaptation planning for cycling. We propose a framework to model the changes in bicycle flow in London by developing a negative binomial count-data model and by incorporating future projected weather data from downscaled global climate models, a first such approach in this area. High temporal resolution (hourly) of our model allows us to decipher changes not only on an annual basis, but also on a seasonal and daily basis. We find that there will be a modest 0.5% increase in the average annual hourly bicycle flows in London's network due to a changed climate. The increase is primarily driven by a higher temperature due to a changed climate, although the increase is tempered due to a higher rainfall. The annual average masks the differences of impacts between seasons though - bicycle flows are expected to increase during the summer and winter months (by 1.6%), decrease during the spring (by 2%) and remain nearly unchanged during the autumn. Leisure cycling will be more affected by a changed climate, with an increase of around 7% during the weekend and holiday cycle flows in the summer months.