[Imports] UN Mappers import of UNSOS waterways in Somalia

[Christoph Hormann]

On Wednesday 08 April 2020, Arne Kimmig via Imports wrote:
>
> - how can we distinguish a 'drainage shaped landscape' from a stream?

Reliably only through local knowledge or through imagery depicting 
actual waterflow.  There are occasionally good indications of recent 
waterflow even in dry situation images - like a recently newly built 
river crossing of a road next to an old and destroyed one.  Bridges or 
culverts are also a clear indicator.

> Actually, what do you mean with this term? - same question for
> 'fossil river'. Is that something like a dry valley?

As mentioned throughout subtropical Africa - both in the north and in 
the south - there are a lot of valleys evidently shaped by water that 
have not seen any more recent surface water flow.  

On images you can sometimes see that for example from visible artificial 
or natural structures of significant age that would not withstand 
significant waterflow:

https://mc.bbbike.org/mc/?lon=0.130422&lat=17.486138&zoom=15&num=2&mt0=bing-satellite&mt1=mapnik

or dunes across the riverbed:

https://mc.bbbike.org/mc/?lon=0.340901&lat=17.834483&zoom=14&num=2&mt0=bing-satellite&mt1=mapnik
https://mc.bbbike.org/mc/?lon=31.690906&lat=17.633554&zoom=16&num=2&mt0=bing-satellite&mt1=mapnik

In most cases however this is very hard to reliably see (the problem of 
proving a negative - what i mentioned in my reply to Rafael) - 
especially in relatively flat areas like much of Somalia where 
scrubland covers much of what could be useful hints.

Famous example of a fossile river is the Molopo River in South Africa 
where it is generally said that it has not carried water over its whole 
length for at least a hundred years:

https://www.openstreetmap.org/relation/7370966

I could imagine that this similarly applies today to the Lagh Dera:

https://www.openstreetmap.org/relation/5967248

-- 
Christoph Hormann
http://www.imagico.de/