Forum: General Topics

Forums / General Topics / countours in high tatras

 

Subject:countours in high tatras 

jose1711

22:19
Sunday
25-Mar-2012

Location:
slovakia

Phone Model:
bb, e51

hello stephen,

could you please check countours near gerlachovsky stit (use mapquest search api)? i don't know if this is a limitation of osm data processing or maybe there's something wrong with my phone (settings).

thank you.

jose
 
 

Stephen

9:51
Monday
26-Mar-2012

Location:
Surrey, UK

Phone Model:
BB 9800 Torch
BlackBerry 8900
SE W910i
Nokia 5800

Hi Jose,

(reply relates to both threads)

I suspect the problem is not related to either OSM or your phone, but rather to the NASA SRTM data that I'm using for the contours. The data comes from space shuttle radar which scanned the planet surface to record altitude. The process was not perfect so there are numerous small gaps in the data that are marked accordingly. The heights are supplied as 16-bit integers with the value -32768 being used to represent a 'null' or non-available value (hence the weird altitude in the geocode lookup!).

I've used the GDAL Tools to convert the SRTM height maps into contours, but where there is no data then the contours will just stop, as you have noticed in that mountain area.

Incidentally, if you look at that area on the OpenCycleMap it shows a similar noisiness of contours, although I think they have used a separate algorithm to guess the unknown heights by interpolating the surrounding areas, so the problem is less pronounced.

I did look into this smoothing algorithm myself (it was probably a couple of years ago that I added the contours), but I seem to remember I had trouble getting it to work reliably on the tile server, so decided to just leave the contours blank.

I will have a look at the geocoding altitude lookup to at least make it return a zero altitude rather than the crazy -32768 figure!!

Cheers,
Stephen
 

(You must be logged in to post a reply to this thread)