Weeks of 8/29/2023 & 9/12/2023

Fall plan: #1090
Continuing/Finishing WA: #818
Variables to be added to foundational data frame; https://github.com/HARPgroup/HARParchive/issues/1081#issuecomment-1682641348
[x] Tasks involving fn_get_pd_min laid out in #818
[x] Add message within FDC for when Qbaseline < 0 causes an unreliable FDC (title or subtitle): http://deq1.bse.vt.edu/d.dh/om-model-info/7200944
[ ] #1092
- Water availability article: Estimating 7Q10 confidence limits from data: a bootstrap approach DP Ames - Journal of water resources planning and management, 2006 - ascelibrary.org --> see who’s cited it, and summarize how folks are calculating uncertainty for streamflow records
- water stats guy website
- Goal: we want to know how much uncertainty and variation our calculations will have. How reliable is our calculation?
- [x] Analysts try to run: Gage vs Model Rmd (serves as a good starting point): https://github.com/HARPgroup/vahydro/blob/master/R/examples/gage_vs_model.Rmd
  - Use vahydro to connect a riverseg to a gage id and to get the area factor
- [x] Also try to run this old HARP rmd: https://github.com/HARPgroup/HARParchive/blob/master/HARP-2018-2019/DEQ_Model_vs_USGS_Comparison/code/dashboard_creator.Rmd
  - [x] 9/12/23: decided to hold off on any dev on this Rmd for now, good resource to have on hand if we decide we want to reuse any of these analytical components later on
- [x] Task 1 (Analysts) - this can be done at this time for run 11 for any models that have gage stats from the State Plan
  - [x] Use the script Megan developed during the weekly meeting (now compare_model_gage_metrics.R)which utilizes om_vahydro_metric_grid()
    - Compare vahydro-1.0 runid_11 to usgs-1.0 runid_11
    - Following the weekly meeting RB and JK were able to rename vars on usgs-1.0 model to l1_Qout, l30_Qout, l3_Qout, l7_Qout, l90_Qout
      - This should now allow om_vahydro_metric_grid() to work seamlessly
      - Note: for now only l30_Qout and l90_Qout can be used for these comparisons since the vahydro-1.0 models lack the other metrics
- [ ] Task 2 (RB/JK) - build plumbing to generate and stash usgs model metrics
  - [ ] Use New Markdown gage comparison script, but amend it to:
    - [ ] Look on the river feature for a usgs weighted model with propcode = usgs-1.0
    - [ ] load the USGS model if it exists, or create it if it does not
    - [ ] store values calculated for metrics L90, etc. in scenario metrics for
    - [ ] store area_factor in model scenario metrics
    - [ ] store usgs_gageid in model scenario metrics
    - [ ] if user passes in 'area_factor=auto`, try to load the vahydro model and get the drainage area, and load the drainage area from the USGS gage and calculate an area_factor (model drainage area / gage drainage area)
[ ] start working to turn what we have into a full document
- [x] Add placeholders to Rmd for where narratives will be included ; they'll eventually be pulled from vahydro but just outline and set variable name for the narrative for now, see #853 for locations
- [ ] update the detailed work plan to show more of what we have, get questions that we can ask soon before new generation
- [ ] edit for using new foundation doc
[x] Megan and Ella start getting more into water availability
We will not have accepted/accurate WA values to compare our calculated values to; we will apply model error to our WA equation to determine how model error impacts our WA calculations and margin of safety
Think about: how will error be displayed/ represented in the Rmd? Numerical representation in table(s) and/or visual representation (map)?

HARPgroup / HARParchive

Weeks of 8/29/2023 & 9/12/2023 #1091

gage_vs_model.Rmd tests & outputs