On 15/06/2023 20:47, Jez Nicholson wrote:
Whilst it is a great idea to capture local knowledge about flooding,
especially where it is currently not available, I am concerned that
this doesn't have on-the-ground verification.
I don't think that anyone has suggested that - at least not in this thread?
The original email said "The location and extent of these hazard areas
is often well known by local communities with knowledge of past events."
When I talked about "what is currently flooded based on current measured
level and previous observations"I meant exactly that - recording that
when a river level at a known point reads X, land at Y (in the vicinity
of X) will also be flooded.
Flood risk areas are predictions generated via modelling software and
it depends on which software you use, and the quality of the input data.
Indeed - the Environment Agency in the UK (and other agencies elsewhere)
make extensive use of this sort of model, but I suspect that mapping
this sort of thing goes bit beyond what can usefully done within OSM,
though of course it can be combined with OSM data by a data consumer to
create "risk maps".
The current hazardous areas get away with it by mapping areas marked
out by signage. Sure, the signs may have been placed following
predictions, but they are physically there to be seen.
I suspect that this isn't true for all "boundary=hazard" in OSM at the
moment (picking one at random, the signage for
https://www.openstreetmap.org/way/500428513 and the wider
https://www.openstreetmap.org/relation/15680620 doesn't look especially
extensive - see
https://www.abc.net.au/radio/programs/australia-wide/australia-wide/13490888#:~:text=Wittenoom%20is%20the%20largest%20contaminated,site%20in%20Western%20Australia%27s%20Pilbara.
)
Best Regards,
Andy
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
