Denmark

What the record shows

Copenhagen holds one of the highest shares of documented successes of any national record in this atlas, and the two entries that failed have the same shape as each other. Everything that worked here is a physical system with a data layer on top — pipes, signals, meters, a forecast that closes a beach. Both things that failed needed somebody else to turn up.

The City Data Exchange was a marketplace for buying, selling and sharing anonymised public and private urban data, run by the city, the Capital Region and a corporate partner from May 2016. It could not attract enough data providers or enough buyers, and the partners closed it in 2018. Bycyklen put a tablet in the handlebars of every rental bike for login, payment, navigation and onward transit planning — among the first bike-shares anywhere with onboard screens — grew to about 1,860 electric bikes and lifted rides from 169,000 in 2015 to 933,000 in 2016, and then ran out of money after repeated funding cuts. Neither failed because the engineering was wrong. One needed a two-sided market to form and the other needed a subsidy to hold, and Copenhagen's successes needed neither.

The oldest entry here is also the cleanest example of a model with a consequence attached. Since 2002 a predictive bathing-water system has combined rainfall data with sewer-overflow sensor readings to estimate bacterial risk in the harbour baths after storms, and a bad forecast triggers a red flag and a closure. That is not a dashboard: the model output shuts a swimming spot. Across this atlas most measurement stops one rung lower, at publishing a number and hoping an agency reads it — which makes a twenty-four-year-old Danish forecast one of the earliest working answers to the question the rest of the corpus keeps asking.

The green wave is the entry that shows the city's habit of publishing its own trade-offs. Signals on key cycling corridors are timed to a steady 20 km/h so that a cyclist holding pace meets consecutive greens; a city evaluation found average cyclist speed rose from about 15 to 20 km/h and red-light stops fell from as many as six to zero or one. It also states what that cost: roughly 27 to 50 seconds of added delay for buses on the same corridors. Compare Zurich, where the same technique has been pointed at trams and buses since a 1977 referendum. The technology is identical and the beneficiary is a choice, and Copenhagen is unusual in naming the losing side of its own.

The heavy infrastructure is heavy on purpose. After the catastrophic cloudburst of July 2011, the city and its utility adopted a twenty-year plan built on hydraulic modelling of the city's catchments and spanning more than 300 projects — retention parks, cloudburst streets, underground tunnels. In Sankt Kjelds Kvarter, Denmark's first climate-resilient neighbourhood, about 30 per cent of local rainfall no longer reaches the sewers at all. The instrument there is a model that shaped concrete, not a sensor that watches it. EnergyLab Nordhavn ran the district-scale version from 2015 to 2019 — DC fast charging with battery buffering, ultra-low-temperature district heating, demand-response heating trialled first in 85 apartments — and closed on schedule as a research programme rather than drifting on as an underfunded service.

Where Copenhagen instruments existing plant, the reported numbers come with their provenance attached. The LED retrofit replaced about 20,000 streetlights and 8,000 poles — roughly half the city's stock — under a 2014 contract, on a wireless network that lets the city dim, schedule and monitor each lamp individually and brighten it for an approaching cyclist. Street-lighting energy fell by roughly 55 to 57 per cent, meeting the contract's minimum efficiency target, and that figure is independently corroborated by Danish trade press; the vendor's carbon saving is reported separately and is not. The district-heating demand response shows the same care about scale: a 2021/22 pilot cut morning peak heat load by up to 22 per cent across 17 municipal buildings, and the rollout that followed covers 140 buildings from 20 owners and averages a 5 per cent energy saving. The pilot number and the rollout number are different things, and the entry keeps them apart. The municipal building surveillance system catches leaks and inefficiency in real time across the city's own stock and has since been extended to major private buildings.

The data position is the one place Copenhagen chose a different answer from its neighbours. Rather than run a strong single municipal catalogue the way Zurich does, it publishes into a shared platform alongside dozens of other Danish municipalities, contributing 216 datasets of its own. That is a plausible answer to a real problem — most Danish municipalities are far too small to run a catalogue — and it means the city's holdings are legible next to its neighbours' rather than in isolation. If the entries below agree on one lesson, it is that Copenhagen is very good at systems where the city itself is the only party that has to act, and that both of its documented failures were bets on somebody else showing up.