Future improvement for BankfullChannel

[jenna-walsh]

Some minor issues with the BankfullChannel tool:

• Output network doesn't retain input segments but is instead segmented by the thiessen valley bottom. Any attributes of the input network are lost.
• In watersheds with inflow from upstream watershed(s), the user must manually edit drainage area values and recalculate bankfull width. Automatic some or all of this task could make the tool more efficient and user-friendly.
• Adding a drainage area check would improve outputs. Drainage area raster doesn't always align with the valley bottom, leaving reaches with DA = 0.
• The bankfull polygon output is created by buffering the network by the calculated bankfull width, thus the polygon width is twice the calculated bankfull width. This is useful for the tool's application in determining confinement, but is confusing. May want to clarify in the documentation, and/or rename fields for clarity.

[jenna-walsh]

The tool also still leaves some segments with PRECIP = 0 which results in a minimum bankfull width.

[mhallerud]

Another major change needed is allowing a user to enter stream names and drainage area values to correct for upstream drainage area values. I started to implement these changes in the confinement_dev branch, specifically the lines relating to these changes are:

• Bankfull channel: lines 40-49, 86-92, and 292-332
• RCAT.pyt: lines 1054-1071

Corrections that still need to be completed:

1. Joining the network with the input NHD network so that the 'StreamName' field is present before feeding into the correct_upstream_da() function
2. Making the list parameter (param11) input fully functional in the toolbox plugin

[jenna-walsh]

There are watersheds with upstream drainage areas that enter partway up a stream. The North Fork John Day and Lower Salmon are both examples of this. Is the tool able to handle that kind of scenario? When I've correct drainage areas in those watersheds I've manually removed segments from the selection. Maybe there is a way to have the user enter the ID of the segment at the confluence and have the tool figure out which segments are downstream of that?

[mhallerud]

@jenna-walsh Unfortunately not, in the past we've just corrected these situations manually. I can't think of a quick and easy way to code this in. We could enter the FID of the segment but determining what's "downstream" of that segment is a challenge in itself.