A spatiotemporal analysis of the impact of lockdown and coronavirus on London’s bicycle hire scheme: from response to recovery to a new normal

Figures & data

Table 1. Data fields in the santander cycles dataset.

Data Field	Description
Rental ID	Unique ID for each trip
Duration	Duration of the trip in seconds
Bike ID	The unique ID of the Santander bicycle used for the trip
Start/EndDate	The start or end date and time of the trip, to the nearest minute
Start/EndStation ID	The unique ID of the start or end docking station of the trip
Start/EndStation Name	The name of the start or end docking station of the trip

Table 2. Descriptive statistics of filtered BSS trips.

Year	Total Stations	Total Trips	Distance/meters				Duration/seconds
			Mean	S.D.	Min.	Max.	Mean	S.D.	Min.	Max.
2019	797	10,207,268	2,139.78	1,450.49	0	11,996.48	993.25	845.29	60	7,200
2020	798	10,033,213	2,239.92	1,637.61	0	11,999.09	1,216.42	1,040.74	60	7,200
2021	797	10,680,265	2,181.47	1,520.79	0	11,997.28	1,093.16	9,21.63	60	7,200

Table 3. Time periods for analysis based on London’s lockdown timeline.

Description	Start Date	End Date
First national lockdown	26th March, 2020	9th May, 2020
First Lockdown restrictions eased gradually	10th May, 2020	13th September, 2020
Reimposing restrictions	14th September, 2020	4th November, 2020
Second national lockdown	5th November, 2020	30th December, 2020
Third national lockdown	6th January, 2021	7th March, 2021
Leaving lockdown	8th March, 2021	18th July, 2021
Back to normal	19th July, 2021	8th December, 2021

Figure 1. Daily temporal trend of bike share usage from 2019 to 2021.

Figure 2. Monthly temporal trend of bike share usage from 2019 to 2021.

Figure 3. The total number of bike hire trips by day of the week in each month from 2019 to 2021.

Figure 4. BIC score of GMM model selection.

Figure 5. The scatter plot of overall 2020 trips with the pre-trained GMM model.

Figure 6. Total hourly trip counts in Cluster 1 per month of 2019, 2020, and 2021.

Figure 7. Total hourly trip counts in Cluster 2 per month of 2019, 2020, and 2021.

Figure 8. Total hourly trip counts in Cluster 3 per month of 2019, 2020, and 2021.

Figure 9. Change ratios (%) in the trips of each cluster among each period compared with 2019.

Figure 10. Spatial patterns of top 1% of OD flows in the Cluster 1 (a) C1-amp (b) C1-IP (c) C1-PMP.

Figure 10. (Continued).

Figure 11. Spatial patterns of top 1% of OD flows in the Cluster 2 (a) C2-AMP (b) C2-IP (c) C2-PMP.

Figure 11. (Continued).

Figure 12. Spatial patterns of top 5% of OD flows in the Cluster 3 (a) C3-AMP (b) C3-IP (c) C3-PMP.

Figure 12. (Continued).

Data availability statement

The cycling data that support the findings of this study are openly available in Transport for London at https://cycling.data.tfl.gov.uk/.