search

Riverscapes / pyBRAT

pyBRAT - Beaver Restoration Assessment Tool (Python)
http://brat.riverscapes.xyz
GNU General Public License v3.0
10 stars 10 forks source link

Idaho BRAT Project Status Info (Internal) #263

Closed bangen closed 5 years ago

bangen commented 5 years ago

@mhallerud @wally-mac

I took a quick look at the tracking sheet that Maggie put together. Thanks for getting this rolling!

I have a few questions/comments (see below). This is a little long-winded but we need to hash out the details so our batch runs are smooth.

Project Area

I noticed there are several basins is Maggie's spreadsheet that aren't in the original scope (Big Hole, Bitteroot, GreysHobock, LowerSnackTucannon, Madison, MiddleClarkFork, RedRock, Yaak). If I remember correctly these were basins that barely (or do not) cross into Idaho so we removed them from the project area. @wally-mac does that sound familiar? Or were these basins added back in based on feedback from the project partners.

I added a 'ProjectBoundary' folder to the '00_Projectwide' data folder that has the original scoping shapefiles.

VBET Valley Width Parameters

I added the VBET valley width parameters to the 'Basin Specific Parameters' tab. These values had been populated by (I think) Karen and Mic. Widths haven't been gathered for all the basins. I'm not sure why they stopped but there was this issue (#125) that went dead. @wally-mac should @mhallerud just use default values for all the basins or should we task someone with finishing the remaining basins.

Updating Major River DA Values

I added the a 'Major River Upstream DA Values' tab. These are the values that Karen had calculated. There are a lot of major rivers crossing HUC8 boundaries in Idaho! We need to make sure that anyone running BRAT for the Idaho project is checking that tab and updating the 'iGeo_DA' values accordingly. @mhallerud I can walk you through this piece.

DEMs

I added a 'NED' folder to the '00_Projectwide' folder on Box. I'm currently uploading the project area DEM and the original tiles that were used to create it (under the 'tiles' folder).

There are 5 basins (Middle Kootenai, Moyie, Lower Kootenai, Pend Oreille, Priest) that extend partially into Canada that do not have full DEM coverage. In other words, we only have NED DEM data for the US portion of the HUC8. @wally-mac this is something we chatted about really early on in the project but I'm pretty sure it was overlooked and is something we still need to task a tech with.

I previously clipped out all the individual basin DEMs into a 'DEM' subfolder. This was early on before we'd really standardized all of our naming conventions and created the batch scripts. Long story short, the individual HUC DEM names include the HUC8 ID (e.g., 'NED_DEM_10m_17040206') rather than just 'NED_DEM_10m'. From a batch processing perspective it may just be easiest to run the batch DEM clip code on the project-wide DEM rather than re-write the code. Thoughts?

wally-mac commented 5 years ago

Thanks Sara,

Here is my feedback: Maggie just use default values for the watersheds that we do not have V-BET values for yet. In regards to areas in Canada, I say we clip all data to the international boundary.

bangen commented 5 years ago

The main issue with clipping to the US boundary and not having DEM data for Canada is that our drainage area values will be off.

mhallerud commented 5 years ago

Thanks for taking a look @bangen and @wally-mac. @bangen it may be quicker to pull the DEMs over to the folders. I already ran something similar for the GYE NHD data (was not sure where you got the projectwide data at that point) so it should be easy enough to adapt for the DEMs.

wally-mac commented 5 years ago

Back to the Canada issue... is there a way to capture the drainage area without having a DEM? Similar to what we do when a large river runs through a watershed. Thoughts? I want to avoid if possible using DEM data from Canada... it could be a rabbit hole.

bangen commented 5 years ago

@wally-mac

Capturing DA without DEM

I'm not really sure how we would go about this and end up with reasonable estimates. With major rivers it's much easier and defensible since we grab the US drainage area of the contributing basin(s).

I checked and there are 1 arc second (~30 m) DEMs for Canada available for download from the USGS TNM. Merging those with the US 1/3 arc second tiles for the 5 basins would probably be easier than coming up with a non-DEM based DA estimation workflow.

@mhallerud - Until we make a decision on how to handle these 5 basins just hold off on running them through BRAT and VBET.

wally-mac commented 5 years ago

Thanks Sara! I like the idea of using the 30m data. I suggest we do it that way. Can you please line Maggie out on this task?

On Sat, Feb 16, 2019 at 3:45 PM bangen notifications@github.com wrote:

@wally-mac https://github.com/wally-mac

Capturing DA without DEM

I'm not really sure how we would go about and end up with reasonable estimates. With major rivers it's much easier and defensible since we grab the US drainage area of the contributing basin(s).

I checked and there are 1 arc second (~30 m) DEMs available for download from the USGS TNM. Merging those with the tiles for the 5 basins would probably be easier than coming up with a non-DEM based DA estimation workflow.

@mhallerud https://github.com/mhallerud - Until we make a decision on how to handle these 5 basins just hold off on running them through BRAT and VBET.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Riverscapes/pyBRAT/issues/263#issuecomment-464395456, or mute the thread https://github.com/notifications/unsubscribe-auth/AU-QUsdkqu9KZ9kAYH-IoE4PGzrRUKvoks5vOIn9gaJpZM4a6K0y .

-- Wally Macfarlane 435.512.1839