4 ms·
This is an extraordinary effort. The NY Times took LIDAR data showing the shape of every building in NYC and then calculated the shadows on every square meter o
by dankohn1 10y ago
This is an extraordinary effort. The NY Times took LIDAR data showing the shape of every building in NYC and then calculated the shadows on every square meter over every minute of every day of the year, then integrated the totals to show the % in shadow during summer, spring/fall, and now (winter).
I asked my 10yo to find our apartment building and I was interested to see that he had trouble because he uses the World Financial Center ferry dock as a reference (who knew). The ferry dock is missing from the picture because it presumably is not in the building LIDAR database.
Anyway, a crazy amount of computation and a cool example of data journalism.
- dagw 10y agoAnyway, a crazy amount of computation It really isn't. I do this sort of stuff all the time at work (and over much bigger areas) and the tools available these days make it almost trivial. Edit: Not to say that what they did isn't cool, it is. But anybody with access to the right data and the right software could do the same thing on a halfway decent desktop computer.
- dankohn1 10y agoIs any of the software open source? If not, how much does it cost?
- dagw 10y agoPersonally I use FME, which costs ~$2500, for processing the point cloud data. But if I put in some effort I could probably duplicate everything I do in FME entirely open source using GDAL,PDAL,GEOS and a bunch of python scripts. The actual shadow casting code is some pretty simple C code I wrote, but both SAGA and GRASS (again both open source) have similar functions you you can either use or be inspired by. Finally for the statistics and post-processing it's just some simple python/numpy code and then GDAL again to output the pretty pictures.