This conservation unit has the wrong outlet in the database. This is due to the edge of the conservation unit being very close to a bend in the Thompson river. At the lowered resolutions we are using for the VIC-GL flow map, the river flows briefly outside the conservation unit, which the outlet-finding script interprets as the mouth of the conservation unit. Script selects a "typical" point in the watershed and traces downstream until it leaves the region, which it counts as the outlet.
There might be some way to make the script resilient to these kinds of mapping errors, perhaps selecting several starting points to trace downstream from and letting them "vote" on the correct outlet, but that seems like a lot of trouble for one pathological region. Probably just needs to be hand-edited in the database.
This conservation unit has the wrong outlet in the database. This is due to the edge of the conservation unit being very close to a bend in the Thompson river. At the lowered resolutions we are using for the VIC-GL flow map, the river flows briefly outside the conservation unit, which the outlet-finding script interprets as the mouth of the conservation unit. Script selects a "typical" point in the watershed and traces downstream until it leaves the region, which it counts as the outlet.
There might be some way to make the script resilient to these kinds of mapping errors, perhaps selecting several starting points to trace downstream from and letting them "vote" on the correct outlet, but that seems like a lot of trouble for one pathological region. Probably just needs to be hand-edited in the database.
Here's a map showing the problem: