back
64 comments
Related: map of noise pollution in Berlin https://interaktiv.morgenpost.de/laermkarte-berlin/
This is so cool! Would love to have this kind of information before I rent or buy an apartment.
If you live in the EU, there's an European directive making noise maps mandatory: Member States shall adopt the measures necessary to ensure that no later than 30 June 2012, and thereafter every five years, strategic noise maps showing the situation in the preceding calendar year have been made and, where relevant, approved by the competent authorities for all agglomerations and for all major roads and major railways within their territories. END Directive (2002/49/EC)

There's a basic online viewer here: http://noise.eea.europa.eu/ but the GIS data is available here: http://cdr.eionet.europa.eu/

Unfortunately, legislation in some countries (at least 1, Austria) prohibits use of this data for commercial projects.
Is it me or is Austria most paranoid country in Europe in regards to privacy?

IIRC, dash/helmet cams are also not allowed here due to privacy, which is crazy. A friend of mine was hit by a car while riding his bike and now has to fight the driver in court to get justice since there were no witnesses. A cam would have made a night/day difference.

Its a really weird rule, but Austria is full of weird rules.

First - its not true that you cannot use a dash cam. You just can't ever publish any media from that camera that might be used to identify a person - even in a court of law, when attempting to prove your own innocence.

You can use a dash cam to record your journey - but if you ever show the video to anyone, you must render anyone in that video un-identifiable, and that goes for license plates as well.

Also, a common misconception among Austrians is that you need permission to film someone in public. This is not true - no such permission is required, and you can record anyone in any public space, any time you want.

You're just not allowed to use that recording to identify someone, nor for commercial purposes - without express permission.

I do find that a bit weird myself, but I'm happy to see that there are countries that still value privacy. Towards the other side of the spectrum, you get countries like the UK where you are being filmed pretty much everywhere you go, yet the police mostly shrugs when you report a crime that was caught on video.
Ah, that's unfortunate indeed. In my city that data is CC-BY 4.0.
The University of Minnesota, Center for Transportation Studies, Accessibility Observatory has been working on such tools to provide a connection between studies and analysis and applicable geospatial visualizations (i.e. noise pollution, access to jobs within 30 minutes, access to food, etc.).

I'm surprised there isn't more of this already published and well known, such that you can go on Zillow and see such things.

Pure speculation, but perhaps this would only detract from more sales and revenue-generating turnover?

If site doesn't respond for you also here's a mirror

https://web.archive.org/web/20190414020642/https://interakti...

Here's one with noise by day and night as well as a bunch of other metrics: https://berlin.placeilive.com/daynoise
How do they get reliable estimated provided that noise is changing somewhat based on rush hours?
Hah! I live inside one of these bubbles in Mitte.

The truth is the grid of Altbau (pre-war) buildings is great to live in, especially if you don't share the flat and can utilise the "back" room as a bedroom. But pretty much most side streets (at least in the West) are super quiet -- thanks to speed bumps, narrow lanes, treeline and wide sidewalks.

EDIT: Also many points are in heavy-industry areas (like Westhafen) so being far away from the street doesn't change much.

Central London (of all places!) also has really quiet and walkable side streets.
What neighbourhoods do you have in mind?
I used to live in Clerkenwell and usually walked a bit north and East. Eg towards the Earl of Essex, or even just the side streets in the City itself.
Nice!

If I re-engineered your algorithm correctly, it tries to find the single largest circle within a grid of streets.

This, unfortunately doesn't work well for a river that isn't frequently broken up by bridges.

E.g. the whole area of Treptower Park, Insel der Jugend, Stralau and Rummelsburger Bucht doesn't have a Rueckzugsort next to the water (there is only a big one in Plaenterwald for that whole part of the Spree).

Nice work, though. I'd be interested in a version of this that maximizes walking or biking time without crossing streets!

Working with circles seems weird. I would have computed a Voronoi tessellation with Fortune's algorithm (also n log n) and colored the result with OpenGL triangles. Even better would be to estimate traffic density of the roads based on road type and some measure of centricity/connectedness and also weight the resulting values with that. A small central spot sourrounded by small streets may be calmer than a large peripheral one right next to a high way.
It's a cheap and hand-wavy, yet titled impressively piece made by an artist so don't expect anything too sophisticated ;)
I can think of computationally intensive probabilistic algorithms to find these circles (e.g. hill-climbing by randomly mutating circles), but is there a way to guarantee optimality? I.e. find the absolute largest circle inside a given polygon.

Is there any information about how these particular circles were generated?

There appears to be an O(n log n) algorithm: https://stackoverflow.com/questions/4279478/largest-circle-i...
Nice work! Would be nice if the tooling could be open sourced so it can be applied to any area of interest.
Anyone interested in building a "livability" map with noise, air pollution, traffic, demographic indicators, proximity to nuclear reactors, flood and other natural hazards risk etc. or anyone doing this already?
This would be cool. Would need toggles with a wide variety of options since 'livability' means very different things to different people.

Like for me, I love that now that I live in Munich, I can get around by foot, by bike, by bus, by train, etc. But a lot of Americans really do like that in the US, everything is so spread out you have to drive. Different strokes and all.

>proximity to nuclear reactors

Don't forget proximity to vaccines!

On a more serious note, a coal plant (like there are plenty in Germany) will spew 10x more radionuclides in the air than a nuclear plant for the same amount of produced electricity.

So why aren't you avoiding coal plants rather than nuclear power plants again ?

Coal plants are already covered by air pollution measurements.
The point being made is that a properly funtioning nuclear power plant emits very little radiation when compared to a coal-powered plant.

Coal power plants, even with emission regulations, are very dirty as far as radioactive emissions are concerned.

The risk of cancer for small children is doubled near a nuclear plant, see https://www.bfs.de/DE/bfs/wissenschaft-forschung/ergebnisse/...
There could be, of course, confounding factors; I can't read German, but I can say that in the US at least, cancer rates are elevated in industrial zones in general -- exactly the places that nuclear reactors are constructed.
As it’s small numbers and small distances involved, I wonder if the risk is really from parents working at the plant, or if they controlled for that.
> The point being made is that a properly funtioning nuclear power plant emits very little radiation when compared to a coal-powered plant.

Nobody is denying that or excluding the possibility to add coal plants to such a map. Excuse me for not adding a complete, prioritized list of features to my question. I find it disturbing that some people start insinuating "humorously" that someone is an "anti-vaxxer" because he's considering functionality for people who - for whatever reason - don't like living near nuclear plants. It's this kind of derailing that makes discussions on the Web unbearable sometimes...

Questioning superstition is important. What you are offended by is identical to what you would expect to experience were you to suggest that you want to avoid vaccines for 'whatever reason'. Leaving scientifically unsupported opinions unchallenged is irresponsible and I at least certainly don't feel like I've overstepped my bounds by challenging such an uninformed belief.
So, Chernobyl and Fukushima never happened on your planet, it's all superstition. Got it, let's leave it at that.
It's pretty simple and fun to compute these for your own use cases. Openstreetmap data is really useful here. One implementation:

http://notes.secretsauce.net/notes/2015/05/06_poles-of-inacc...

Nice idea.

Though looking at my area a few of the larger dots are inaccessible industrial areas. Not exactly what you're looking for. Wonder if that could be made more intelligent.

Nice idea. I'd like to see something applicable for other cities (via openstreetmap?) to explore possible partially remote areas where I could be living; along with the possibility to hook it up with more data.

Background: pretty bad case of asthma, car exhausts being one thing that makes things worse.

Very nice how the resolution increases when zooming in. This should exist for every city.
Would me interesting to compare this with a map of criminal activities statistics.
Can someone mathematically formulate the exact problem that is solved here?
There's a big bubble right over the airport Tegel (TXL) hwich I would consider no place of silence ;)
Probably should have included railroads as well.
Nice! Berliner here, living very closely to a large green bubble in Neukölln (Tempelhofer Feld)
Nice idea! This seems to ignore smaller streets though, why?
Probably because those are mapped in OpenStreetMap as service roads (which includes alleys, driveways, parking lanes, etc.). The author of this map seems to have queried all roads in Berlin from Autobahn down to residential streets, and left out service roads, and bicycle and foot paths.
Yes like Pfaueninselchaussee. A street but with very low traffic.
Great work!