A summer town, counted.
All of July we counted the phones present in Drøbak — every ten minutes, day and night — and scaled them to people. Then we put those counts next to our model of the town's everyday life: who lives here, who commutes away, who comes to visit. The measurement tells you what happened. The model tells you what it means.
Where we measured
Two zones over Drøbak — the old town and the slopes above it. The blue shading is how far you can actually walk from the town square in ten and twenty minutes, on the real streets. The dots are the town's parking.
More people asleep than live here
Our model puts about 6,113 residents inside these zones. On a quiet July night the network finds 7,423 people there — roughly 1,310 more than actually live in the town. Those are the summer people: holiday houses, boats in the guest harbour, spare rooms. A town that grows overnight, and no register would show it.
Two zones were drawn at different sizes over the same town centre and returned near-identical counts, so a zone's figure reflects the area the network resolves, not the exact polygon. Compare zones over time, not against each other by area.
In a normal week Drøbak empties in the daytime — people leave for work and come home at night. In July the town filled up instead. That gap between the dashed line and the green one? Those are the visitors.
July, day by day
Daily peak presence, measured core. Visits/day = peak minus the overnight floor — weekdays average 1,875, weekends 2,117.
Measured centre, modelled municipality
Drøbak town
Every 10 minutes, all month, 31/31 complete days. Sentrum alone: 26,301 visits, weekends +21%.
All of Frogn
The model already simulates ~35,000 trips into Frogn on an ordinary weekday — for the 113 km² the sensors don't see. It can do the same for any town in Norway, tomorrow, without installing anything.
Where visitors come from — before any sensor
We simulate every resident of Norway through an ordinary day — so before a single sensor went up, the model already knew who comes to Drøbak, how they get here, and why. Here is what it says.
How people arrive on an ordinary weekday. That 56% car share is the whole reason parking becomes the story in July.
Why they come — with residents heading home filtered out. The model knows the difference.
Who visits in July
Top foreign countries · people present at any moment, on average. Norway: 6,387 — a Norwegian summer town with a European accent.
Three questions nobody has asked yet
Anyone can hand over a visitor count. The interesting part is what happens when you put the measurement next to everything else we know about the town.
Is there parking for July?
At the July peak the visitors alone want about 1,249 spaces. The two central zones hold 684. The whole town holds 2,511 — so the spaces exist, they are just in the wrong place. (56% arrive by car, 1.8 people per car.)
Who could walk instead?
Around one in eight of Frogn's residents can reach the square on foot in ten minutes — on the actual streets, hills included. Every one of them who leaves the car at home in July hands a parking space back to a visitor.
When does the town have room?
The busiest day of the month, hour by hour. From 15:00 the town ran nearly full until midnight — but the morning had room to spare. If Drøbak wants more visitors without more pressure, the space is already there. It just starts before lunch.
Same data, other questions
Who is here at 03:00?
Preparedness planning needs the night-time population, not the census — and in a holiday town those are different numbers, in July by about 1,310 people. Measured here; the model extends it to the rest of the municipality.
Where do they come from?
The model already knows Frogn's travel patterns zone by zone — parking pressure, catchments, new-development impact. A national RAN-trace sample adds observed origin–destination on request.