How How far out can a city model be before it changes a planning decision?

When people talk about 3D city models, they usually talk about how they look. 

Crisp facades, realistic trees, a nice sunset over the skyline. We like all of that too. But the question we get asked most by planners and architects is a less glamorous one: can I trust it?

It is a fair question. A city model is increasingly where a scheme gets judged. Officers look at it, committee members look at it, and residents look at it at consultation events. If the model is wrong, everyone is judging the wrong building.

So how wrong does a model have to be before it actually changes a decision? Our honest answer is: not very. A few metres is often enough. Here is why.

Shadows stretch small mistakes

Shadows are where small errors get big, fast. The lower the sun, the longer the shadow, and the more every extra metre of height matters.

Take a winter's day in London. At midday in late December, the sun sits only about 15 degrees above the horizon. At that angle, every metre of building height throws roughly three and a half to four metres of shadow. So if a model shows a neighbouring building just two metres taller than it really is, its shadow lands around seven metres further across the ground. That can be the difference between a shadow stopping at a fence and covering half a garden.

Most sunlight assessments in the UK follow BRE guidance, which looks at how much of a garden or open space still gets sun on a set date in spring. If the existing buildings in the model are the wrong height or in the wrong place, the "before" picture is wrong. And if the before picture is wrong, the comparison with the new scheme does not mean much.

Protected views leave very little room

In London, some views are protected. The best known run to St Paul's Cathedral, including one from King Henry's Mound in Richmond Park, around ten miles away. New buildings along those sightlines are tested to make sure they do not break into the view or crowd the landmark.

The margins here are tight. A tall scheme sitting miles from the viewpoint might pass or fail by a few metres of height. Now picture the model the test is run on. If the viewpoint, the ground level or the buildings in between are slightly out, a roofline that looks safely below the line could in reality sit above it. Or the other way round, and a perfectly acceptable scheme gets pushed down for no good reason.

Either way, the model has quietly made the decision.

What "accurate" really means

Accuracy sounds like one thing, but in a city model it is really four questions:

• Is it in the right place? Buildings, streets and viewpoints sitting where they are on the ground.
• Is it the right height? Including ground levels, which are easy to overlook on a sloping site.
• Is it the right shape? A pitched roof and a flat roof at the same height cast very different shadows.
• Is it up to date? A model can be perfectly accurate for a city that no longer exists. Schemes get built, and some get demolished.

A model can look beautiful and still fail on any one of these. That is why we think it is worth asking how a model was made, not only how it looks.

The short version

A few metres does not sound like much. But in a city model, a few metres can move a shadow onto a garden, lift a roofline into a protected view, or make a good scheme look worse than it is. If a decision rests on the model, the model has to be right.

Want to see how accurate a model of your site could be? Click below to book a demo.