VBET Sausage Links - Metrics wishlist brainstorm

[joewheaton]

9-6-21 Making tasks for transfer over to metrics sheet in #395

Global Metrics

These are reported for every sausage link and then summary statistics are reported in dashboard and report

Basic

• Area (hectares, sq km, acres, sk mi.)

• Landscape Proportion VB (e.g. for a HUC there is 200 km2 of area, and 5% is valley bottom)

  • Gross
  • By network/riverscape category (e.g. perennial, intermittent, i.e. NHD FCODE)
  • By Ownership
  • By Ecoregion

• Length of transportation network intersecting valley bottom (or specific - road, road type, railroad)

RS Context stuff we have

Not sure if this belongs in RCAT or here because it might not require run-specific parametrization

• Land Use Intensity
• LANDFIRE Veg Class proportions within - producing a categorical bar plot distribution or pie chart

Geometry

• Total Riverscape Length
• Integrated Riverscape Width
• Bankfull Width (from Input)
• Ratio of Bankfull Width to Valley Bottom Width
• Total Channel Length
• Primary Channel Length
• Valley Bottom Sinuosity = RS Length / Straight-line Riverscape Length
• Sinuosity Flavors (GNAT)?
• Effective Flow Length
• Riverscape Density (Riverscape Length / Total Watershed Area)

Interactive Tools

• Playing with probabilistic threshold as first-cut estimate of Active Channel vs. Active Floodplain vs. Inactive Floodplain
• Playing with detrended elevation threshold as first-cut estimate of Stage 0 surfaces, incised surfaces, Inactive surfaces (almost TErX within VB)
• Once symbolized and segregated, give areas (and horizontal bar plots) by above categories -
  • Proportion Active
  • Proportion Inactive
  • Proportion Active Channel
  • Proportion Active FP
• For those polygons, calculate average height above or below Stage 0 gradeline
• Maybe more RCAT... but Recovery Potential can be estimated initially and conservatively by:
  • A user-defined LAND USE INTENSITY THRESHOLD
  • Floodplain Inaccessibility

The above could be produced as horizontal bar plots for each of the thresholds

Mask Metrics

Using the concept we've described in GCD of AOI vs. Class Masks, to produce same global metrics, but for particular areas of interest, or of polygons of same type (e.g. USFS vs. BLM vs. Private) summaries.

Local (Sausage Link) Metrics

Is actually all above...

[joewheaton]

@MattReimer this is where we laid out the Masking ideas.

@lauren-herbine and @shelbysawyer, please have a read and start asking questions if anything isn't clear here.

[wally-mac]

@joewheaton thanks for flushing this out. @Cashe93 and I have been brainstorming about similar metrics for the on-screen digitizing that we have been doing for the BLM and BoR in Utah so it's great to see that we are coming up with similar and the same metrics.

I understand the Metrics you have come up and think that they will add a ton of value but I'm not sure what you are asking for. Is it an attribute table in a shapefile that we populate and pass on to NAR, or what?

Please let me know and then we'll work on the task.

[joewheaton]

Just a reminder to folks, please provide some critical feedback here. @lauren-herbine and @shelbysawyer I will ask you to provide yours in some vignettes you can link to from here.

[wally-mac]

@joewheaton , @lauren-herbine , @shelbysawyer @Cashe93 Here is some feedback on the Global Metrics:

Geometry

• In order to have the lengths calculated we will need to run a centerline of the valley bottom and the centerline will house the global metric attributes.

Interactive Tools

Active/inactive floodplain comparisons

• As we know, currently our tools do not generate Active Channel, Active Floodplain, and Inactive Floodplain. So, currently this can only be done where these features have been on-screen digitized.
  Note: As of 12/1/20, Karen and Tyler are working on extending RIM to include these comparisons

Detrended elevation thresholds

• This seems super useful but we need to discuss this more as a team and get all the specifics on how to do this lined out.

Recovery potential

• To make sure we are on the same page, an initial and conservative Recovery potential polygon could be derived from the accessible portion of the valley bottom, which could be further refined by land use intensity values. Is that what you are thinking?

[wally-mac]

@Cashe93 I want you to do as you described BUT the task at hand is something entirely different. Please re-read what Philip wrote above regarding VBET. @tyler1218hatch can you please help Cashe with understanding the task?

[Cashe93]

@wally-mac gotcha. I've just met with Tyler and have a better grasp of what needs to be done. I'll keep you updated on progress.

[lauren-herbine]

@Cashe93 @tyler1218hatch would you all mind briefly outlining what you discussed? It would be helpful for @shelbysawyer and I. Thanks!

[Cashe93]

@lauren-herbine and @shelbysawyer Sure thing! For my benefit, we just went over what specifically needs to be done for this task and why. In a nutshell, we will be taking an existing VBET output and:

1. manually calculating the metrics that Joe had laid out at the beginning of this discussion thread
2. for each metric, documenting how it was calculated and what inputs/datasets were used to calculate the metric.

This will help us determine the feasibility of calculating each metric as well as help us narrow down which final set of metrics should be reported with each VBET output. My confusion spawned from some of the discussions that were had between the initial post and now and I just needed a push in the right direction. @tyler1218hatch feel free to add anything I may have missed.

[Cashe93]

@lauren-herbine and @shelbysawyer, Tyler and I will be meeting tomorrow morning at 10:30 am to go over how we should be calculating some of these metrics. If you would like to join here is the zoom link.

[shelbysawyer]

We've talked about users being able to specify their specific project area. When users do this, it seems very unlikely that their custom area would perfectly line up with our pre-determined "sausage" segments. This makes me wonder about the dashboard and its capabilities.. If the idea is to have a dashboard provide a set of metrics for each valley bottom segment (ex. proportion valley bottom by land ownership), are we only able to provide these metrics for the existing segments? What is our ability to do some of these metrics on the fly for custom areas which have either been digitized in the app by the user or uploaded via shapefile, etc?

In terms of the metrics that VBET will provide.. my thought it that the report will mirror what can be accessed in the app. @lauren-herbine and I discussed the possibility of offering customized reports, meaning allowing users to specify the metrics they care about before downloading. Is it better just to offer a standardized report? What is our capability to do customized?

I have additional questions, but I wanted to get some discussion rolling on these now. I can break these out into separate discussions if needed.

[lauren-herbine]

@shelbysawyer here is some info from a discussion @joewheaton, @philipbaileynar, and @KellyMWhitehead and I were having regarding calculating gradient and slope a couple of months ago. This might be relevant to your above comment/our conversation from yesterday. Linked above to original discussion, quoted below:

Thinking about @philipbaileynar :

@philipbaileynar let's start segmenting at 500m. I think level IV ecoregions would be too long in some cases and could "pinch" a reach in others:

and

The length over which we measure sinuosity affects the result. How do we pick the right length if our reaches are so fragmented that they are nearly all straight?

So another approach is with @KellyMWhitehead and Carol stole from the Fluvial Corridor Tool, in which some metrics can be calculated at length scales other than their segmentation with a moving window (also facilitates calculating at multiple scales). In other words, it doesn't matter what your segmentation length scale is, you choose a "metric length" calculation scale appropriate for the calculation (e.g. a fixed length window of VB for example). This is the approach laid out in Notebaert & Piegay (2013) of using DGOs and AGOs...

Notebaert B, Piegay H. 2013. Multi-scale factors controlling the pattern of floodplain width at a network scale: The case of the Rhone basin, France. Geomorphology 200 : 155–171. DOI: 10.1016/j.geomorph.2013.03.014

[shelbysawyer]

Thanks @lauren-herbine! This is an interesting approach... @joewheaton do you think this is likely what we're leaning towards for VBET?

[joewheaton]

Thanks @lauren-herbine! This is an interesting approach... @joewheaton do you think this is likely what we're leaning towards for VBET?

Absolutely. This is what we will be doing in QRiS as well. More to follow in #326 and other places. stay tuned.