[talk-au] seaward admin boundaries

cleary osm at 97k.com
Tue Jan 19 08:40:47 UTC 2021



My knowledge of legislation etc. is limited and primarily about NSW legislation. I know nothing about other jurisdictions.  In NSW, it seems that definitions by word in the Government Gazette have been replaced by maps.  Some years ago, boundaries used to be "gazetted" in word form but now it seems that graphical representation in maps are the defining form of documentation. If anyone can direct me to an authoritative verbal description of LGA or suburb boundaries in NSW, I would happily accept whatever it stated.

However, the only authoritative source I can now locate is the NSW Spatial Services office. Their website refers to their maps of various administrative boundaries and asserts that they are joint custodian of these layers which are maintained by them.  They go on to state that most of these layers are described by legislation, notifications described by the NSW Government Gazette or more recently by changes described in the register of public surveys and that Spatial Services tracks temporal and spatial changes to these layers through an authority reference layer.

If there is an authoritative source providing verbal descriptions of boundaries passing along the centre of a river or the midline of a particular road (or whatever) then we could accurately align administrative boundaries with those features. However, unless such sources can be referenced, it seems to me that the official administrative boundary maps from NSW Spatial Services  (and replicated by PSMA) are the accurate boundaries. 

Incidentally, in quickly looking again this evening, I did find that section 205 of the NSW Local Government Act (1993) as amended specifies that the coastal boundaries of local government areas extend to the low-water mark. I could find no other verbal descriptions of administrative boundaries.







On Tue, 19 Jan 2021, at 7:02 PM, Warin wrote:
> On 19/1/21 2:50 pm, Andrew Harvey wrote:
> > 
> > On Tue, 19 Jan 2021 at 11:48, cleary <osm at 97k.com> wrote:
> >> 
> >> As with other boundaries, I'd prefer to keep administrative boundaries separate from natural features even where they approximate and may once have precisely aligned.  I'd like to see administrative boundaries consistent with the authoritative government source/s while natural features such as rivers, coastline etc. are mapped from satellite imagery.  Even where coastline erodes or changes in other ways, I think the administrative boundaries in OSM should remain unchanged until the relevant government authority redraws them.
> >> 
> >> In regard to high-water and low-water marks, I defer to others with better knowledge.
> > 
> > In my opinion, where the boundary is defined by the natural feature (coastline, river, road centerline etc) then the boundary way should be snapped (share nodes) with the natural feature. This provides the most accurate representation and encodes the "defined by the natural boundary" information which would otherwise be lost. As the coastline/river etc changes then the boundaries are kept up to date because they have shared nodes, this is a feature, not a bug.
> > 
> > This is in contrast to your preference cleary. The PSMA and other government datasets aren't the exact boundary definition, only a digital representation of it, if we have a better coastline or river data we should use ours.
> 
> 
> OSM data is also not exact. So I don't think claims of OSM being more 
> exact than other data is an argument I would make ... considering the 
> number of 'inexact' OSM data I come across. And yes I correct it where 
> I can. 
> 
> > 
> > Of course this is all based on that assumption about what defines the legal boundary, but I doubt it is the GIS files the government and it's 3rd parties (PSMA) produce. Phil's comments seem to backup this claim too.
> 
> 
> Looks like the area of NSW that does not conform to the PSMA/DCS Base 
> map (I think those are both the same) is simply around Sydney and 
> Newcastle  ... so I think I'll change those and be done. All the rest 
> looks to use the PSMA data.... which looks to be the low water mark and 
> that is not mapped in OSM (yet)... would be difficult to map and given 
> the number of contributors willing to map it ... it is never going to 
> be done. So you my way of thinking it is the PSMA data that is the way 
> forward. 
> 
> _______________________________________________
> Talk-au mailing list
> Talk-au at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-au
>



More information about the Talk-au mailing list