Basemap’s Director Simon Court explores the issues around “Crow Flies” calculations for travel time analysis in this latest blog.
Recently, I have been doing a lot of research into travel time analysis software in areas such as North America and Australia and was quite interested to see that a lot of software is still reliant on a crow fly (straight line) walking in a total travel time analysis.
It’s quite obvious to see how this can impact on walking only based calculations, as this example from TRACC below shows.
We can see totally different contours and almost twice the amount of people within the 1km catchment thus around a 50% level of inaccuracy with crow fly walking.
But how about a public transport total travel time where the walking element of the calculation is also impacted? The walking time from origin to stop, stop to stop (interchange) and stop to destination can have an impact on total travel time, especially when the maximum walking time cut off is applied.
What better place to look at this than St Louis… with the majestic Mississippi River carving a barrier between Missouri and Illinois. TRACC can provide PT travel time maps using a crow fly approach or by utilising a digitised road and path layer such as OSM, HERE, TomTom etc. Taking the Botanical Gardens as a destination, the following maps were produced….
Visually, the differences are quite striking over a relatively small area. The crow fly map, as expected, extends out over the Mississippi bridging the gaps between the two states whilst the road based one cuts tight to the river boundary. Importing some demographic data into TRACC allowed the demographic reports to be automatically created for these 80 minute catchments and the differences could then be quantified. Using crow flies, allows 926,392 people to access the gardens yet with a road based this is reduced down to 726,013, so a difference of just over 200,000 people.
Over the bridge to the East of St Louis has a lot more coverage with the crow fly approach, yet the population here is actually quite small in comparison with other areas to the West. It seems that there is a much higher population to the West of the St Louis boundary which would be included and this really bumps up the figures. Therefore, this shows a 21% over estimation with the crow fly analysis.
With crow flies, this just isn’t an issue with large natural hazards like rivers or lakes, the issue can be much more local. Does your region have a detailed path data that shows where you can walk? What about motorways and dual carriageways that are impossible for pedestrians to cross? When looking at accessibility you need to be certain that it’s safe to access the public transport network.
Do you get involved in new service design? When using crow flies approach this gives a simplistic view on how people will access the stop. Do you need to model new pedestrian and cycling routes or do you need to look at putting a new entrance to bus station to improve access? All of these are impossible when only looking at crow flies methodology.
Hopefully, not too many important decisions are being made off the back of inaccurate crow fly data!
To find out more about how TRACC can help with your travel time analysis, please contact Simon at firstname.lastname@example.org