|
When you look at a map of the U.S., New Jersey and Georgia stand out has having a lot more public land rendered green. It doesn't seem possible that those two states have so much more public land than all the rest... does it? |
|
New Jersey and Georgia just have more landuse mapped at the moment. This data seems to come from a Land Use and Land Cover (LULC) import, for example the data for New Jersey has source=NJ2002LULC (see way 250286927) and seems to be made available by the New Jersey Department of Environmental Protection (NJDEP). Likewise, some of the data for Georgia has source=USGS-LULC (seeway 31388469) and seems to originate from the USGS Land Cover Institute (LCI). If you want to improve the current situation then just start mapping :) 3
I took a look at a bit of the NJ areas mapped as wood and IMO the import wasn't that accurate. I think such imports are a mixed blessing. They add detail to the map but there's a downside too. It's harder to correct poor data than to begin from nothing in many cases. Maine is a state that is also heavily wooded but has not had the benefit of such imports; it's appears almost totally white on the OSM map.
(08 Apr '15, 10:48)
AlaskaDave
3
Yes, that's true for most imports. They also make it harder to create a local community.
(08 Apr '15, 11:04)
scai ♦
2
Even careful hand traced landuse creation from Bing make the future edits much more difficult. It's not just a problem of landcover imports.
(08 Apr '15, 11:39)
Mike N
But Mike, without Bing we would be severely limited. I have never heard of any mapper walking an area with a GPS to determine land cover. The best and IMO most beautiful topographical maps available in the U.S. are those from the US Geological Survey and even they relied on aerial imagery.
(08 Apr '15, 11:59)
AlaskaDave
1
The use of both local survey and Bing together is not exactly unheard of. Here's an area near me that has changed somewhat since the imagery was taken - the natural heathland is being allowed to reclaim more of the area: http://www.openstreetmap.org/way/322977120#map=17/53.13686/-1.13203 However, it's absolutely correct that more detailed mapping (imports, tracing from imagery or from any other source) makes further mapping more difficult, but there are ways of trying to make the lives of future mappers easier - I've tried to "de-multipolygonise" the linked area above, for example.
(08 Apr '15, 12:45)
SomeoneElse ♦
|