[Tagging] Tagging water bodies areas in which water level
Bert -Araali- Van Opstal
bert.araali.afritastic at gmail.com
Sat May 1 16:07:11 UTC 2021
The approach as described on
is a good appraoch, which works well for over years. However it doesn't
work well for the ever changing water levels in mostly lakes.
A good example are in my area of interest is Lake Victoria and Lake
Lake Wamala has an average depth of only around 5m, Lake Victoria has
many areas with very shallow shores and many small islands.
In the recent year, we have noticed significant variance in the water
levels. This variance is mainly caused by climate change and
restrictions in the outflow (in the case of Lake Victoria), being
hydropower dams on the main outlet rivers of Lake Victoria, not the
seasonal variations we have due to dry and wet seasons. These changes
can be consistent for many years.
Due to the shallowness of these water bodies or their shorelines, on
satellite imagery you can notice whole islands disappeared, the
shoreline varying up to 100m from the existing ones. It also has abig
influence on human behaviour, like we have seen that f.i. on Lake
Victoria, many have established buildings along shorelines since they
were not flooded the past 10 years, however this change rapidly since
the second half of 2019. There is no strict legislation in place or
when it does, enforcement to prevent this landuse along the shorelines,
resulting in humanitarian crisis situations during long time rising
We have no consensus yet locally how to handle this, what we do want to
prevent is people all the time remapping the shores and deleting other
landuse, just because we have satellite imagery showing a different
situation, which might just persist during a short time (shorter then a
As all mappers, we do face the same problems, namely not having enough
imagery available or a longer period of time, on the ground truth
doesn't give you an indication of the actual lake shores, especially
during sustained high waters.
The proposed feature for floodplains could be a good basis, and we
tested it however doesn't work nicely for large bodies. Currently I am
experimenting with high and low water lines defined as hazard
boundaries, the hazard being flood_prone or flooding.... The low and
high water lines can be changed, whenever imagery comes available,
perfectly OK to adjust them as extremes might change. The actual lake
shore is mapped as approximately in the middle. I know this does not
reflect the actual 10 year median water level, but at least gives us an
approach which is feasible for most users. It has the advantage that
the whole lake is also available as one are with ever low time water
level and all time high water level.
On 01/05/2021 17:43, Christoph Hormann wrote:
>> Saeed Hubaishan <hubaishan at outlook.sa> hat am 01.05.2021 16:10 geschrieben:
>> Water areas have many levels. So where to draw the area in low level of water or in high level.
>> The wiki is clear for coastline is the highest line the water reaches in normal circumstances. Is that same for water areas.
> No, the coastal water levels mainly vary with the tides on a daily basis, other waterbodies primarily have seasonal variations.
> You can map that explicitly using intermittent=yes/seasonal=yes on those parts of the waterbody that are intermittently dry. Otherwise see:
> Christoph Hormann
> Tagging mailing list
> Tagging at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Tagging