THIS ARTICLE IS AN AUTOMATIC TRANSLATION, IT MAY INCLUDE MIS-TRANSLATIONS
In recent years, attention for walking and/or running has increased. In both the city and the landscape, the environment should be more inviting to go out on foot. In nature reserves there is also the challenge of maintaining or bringing recreation and nature into balance.
To upgrade the running infrastructure and running environment to current and future running needs, it is desirable to know where different types of runners, many and few, run. Only then can you organize routes in targeted locations attractively according to the wishes of walking people. And, balance it with other interests.
But there has never been a large-scale picture of use in the Netherlands. Until now.
The largest activity tracking app in the Netherlands, Strava, has recently started sharing - under conditions and with partner parties - the aggregated data that their users generate. Which means; street maps showing the collective/anonymous walking use of paths and roads. In the Netherlands, there were more than 800,000 different runners in 2022, who together performed ~20 million running activities. These activities spread fairly evenly across the city, the landscape and nature reserves.
On behalf of the Ministry of Infrastructure and the Environment, and partly initiated by Wandelnet, Track-landscapes mapped the route use of these 20 million Strava running activities throughout the Netherlands, in all forms of analysis that we consider useful. Then the data part is done and we can focus on what we really find important: translating data into insights, policy, and concrete solutions.
What analyzes can be made with this?
Use of all paths and roads in one overview
The most important form of analysis of Strava Metro is mapping path use, so-called 'heat maps' show how often all roads and paths were passed by someone who recorded his/her activity and made it available to Strava Metro. Maps shown below provide examples of this. The best used paths in the Netherlands are given the color black here. For example, the Vondelpark gets this color. The other colors are expressed as a percentage compared to the black value, for example yellow is 18%-27% of this. In this public document we may only express on the basis of percentages; After connecting to Strava Metro of a municipality/province, maps/reports can be created internally with the absolute values. However, absolute Strava values are also only relative; not all runners use Strava. The value lies mainly in comparing walking use or in different areas.
Runners and walkers separated in the picture
Based on the average walking speed on a path, an estimate can be made of the share of runners and the share of runners on the path. This way we can reasonably separate these uses (unfortunately Strava Metro does not offer this distinction by default). And that is necessary: runners and walkers use routes significantly differently. Runners run more on continuous cycle paths and roads in clear and logical laps, while walkers use small paths more with a more dispersed, fine-grained usage pattern. Runners also start relatively more often from home, while walkers start relatively more often from parking lots/entry points in forests and recreational areas. The images below show the differences between Amersfoort center and Den Treek.
Moreover, relatively speaking, many more runners use Strava than walkers. Mixing those two uses in skewed proportions produces a skewed/diffuse image that is easily misinterpreted. Although separating the groups is successful to a reasonable extent; it would be even better if Strava offered that separation itself in Strava Metro. This is where the greatest potential lies for improving the data structures in Strava Metro.
Walking in the dark
Another interesting analysis of route use at different times, the difference in walking use of paths, between 'summer evenings' and 'winter evenings'. Blue means that the path was used a lot on winter evenings , orange means that this path was used relatively often on summer evenings . The purpose of the analysis: to investigate which routes people can still walk (comfortably) in the dark, and which routes they cannot. It can provide insight into places where good/better routes and lighting are certainly desirable. The map below shows the city of Utrecht and the surrounding area.
Walking from A to B
The route use of 'commute walking', runs (both running and walking activities) that ran from A to B (start and end points are far apart) can also be mapped separately. That is only ~10% of all running data. One very clear line of 501 kilometers crosses the Netherlands from North Groningen to the Sint-Pietersberg in Maastricht: The Pieterpad. But other long-distance stage walks are also visible, such as the Marskramerpad and the Trekvogelpad. In addition to the stage walk, there is another specific type of walk: NS walks.
For which running tasks can we use this?
Data - such as from Strava Metro - is only of value if it leads to valuable insights. This requires interpretation and translation of the data into implications. That is not simple or obvious. We identify three types of areas or tasks in which this Strava data can be valuable:
(1) In the city the recreational walking data can be used as an additional reason to make streets, neighborhoods, parks and centers more walking-friendly. In new urban developments such as housing, new structures can connect to movement lines that Strava shows walking data. Last year, for example, we used this data in Dordrecht to outline opportunities for sports in the city. The insight into frequently and rarely used routes and areas (in combination with a survey among residents) gave rise to very concrete design proposals that make sports in and around the city more attractive.
(2) There are many application possibilities in the connection between city and country . Walking data can show where barriers are located between city and country, consisting of highways or railways, but also agricultural areas that are not finely meshed. Interests for cyclists can be taken into account, because runners also visibly use new cycling connections between city and country. In the Amersfoort region we used Strava running data (and also cycling data) to identify opportunities in city-country connections. The success factors of the frequently used routes also provide opportunities for improvement of connections that are not yet widely used.
(3) Much walking use was visible in the nature and recreational areas ; where it can provide valuable input into area challenges. Recreation and nature interests must be in balance; the walking area offers an opportunity for refinement of zoning plans and opportunities for development/connection of nature and recreation. Walking use can also provide insight into where friction between recreational users - walkers, cyclists and horse riders - can occur.
For the NTFU and NOCNSF we applied the same method with the Strava sporty cycling data, with which we mapped the change in use of nature reserves by mountain bikers after the construction of a mountain bike route. We are working with OAK (ecology) consultants to develop an analysis tool in which Strava recreational data, together with ecological area values, can provide direction for the development and zoning of nature and recreation.
Is that representative?
The simple answer: no. Most runners don't use Strava; and Strava is not expected to provide a perfect cross-section of 'walkers' or 'runners'. However; This does not mean that you cannot use this data usefully. It shows real (collective) movements of real people: so it is by definition representative of the type of runners and runs that are often undertaken with it. The question about representativeness is mainly: which ones are they? What we know, for example, is that Strava is much more often used for brisk runs than for 'just a short walk around the block' or 'lunch walk'. For this reason, the large nature/recreational areas, as well as long-distance walks, are clearly highlighted in route use. And also the city-country connections. And so you should mainly use it in problems where typing walking is relevant, and keep the science of typing users in mind when interpreting.
Moreover, the value of data is not only in having a factual/hard 'truth'. In our view, in addition to informing, data should mainly be used as a means of inspiration . Think of the use of paths and roads by runners as 'speaking with the feet'; about which routes are 'attractive' (people were literally attracted to them). Our experience in area developments is that the questions that the data raise due to the empathy with the user, and the conversations that follow from it, are at least as important as the 'factual' data (measuring is knowing) that it produces. It leads to conversations and reasoning based on the interests of the user. That is as important as it is self-evident.
It often also works the other way around; Municipalities or provinces already have certain wishes or ambitions in mind for new/better walking connections or other types of walking facilities. Usage data provides the opportunity to demonstrate their importance, which increases the chance of implementation.
Knowing more?
You can find the full report at ( https://ruimtevoorlopen.nl/kennisitem/kansen-met-strava-wandel-en-hardloopdata/ ) or https://www.track-landscapes.com/spelen-strava-loopdata . Also curious about what insight into recreational walking use can mean for your municipality/province?: feel free to message me via LinkedIn or info@track-landscapes.com .
Comentários