HARP project: Shenandoah Drought Baseflow analysis

[rburghol]

Task Outline: Sub-projects with milestones

Note: Each 1st level bullet must have an issue page (look first to find a matching issue before creating one). Add sub-sub-tasks as needed, and add sub-pages if that sub-task requires is. Only check off the root task once the page that it references is updated such that the body of the issue has a clear and concise summary of the analysis/conclusions of the topic with examples as relevant.

• How much baseflow storage (in inches) is needed to sustain the drought of record flow
  • What is the drought of record low flow in watershed inches per day?
    • Both all-time (gage record) and during the model timespan (1984 - 2024)
  • How do we define drought? XX Days between a significant recharge event? (Use lowest L90 as DoR definition)
• Shenandoah Drought Report Drought Status Assessment Report #10
• Calculate Monthly low-flows in October, November, and December as model calibration metrics
  • What does it look like when model fit is "good", and "bad"
  • What types of visualizations help us qualitatively assess if a "good metric" is a "good model"
  • How do these metrics compare to model groundwater storage, as quantified below?
• Quantify baseflow storage in Shenandoah
  • Geologic Analysis (GIS) Hydraulic Conductivity Raster Analysis HARParchive#1427
  • Running & Analyzing hsp2: AGWS + LZS = baseflow storage
  • Calculate storage from gage flow using HSPF equations Calculating AGWS Using HSPF Methods #37
• Develop baseflow detection methods
  • Base flow even screening (initial) with any timeseries/USGS gage data analysis Timeseries Baseflow Recession Analysis #32
  • Refine Baseflow event : BF Event post-processing #33
  • Compare model to gage Baseflow Event Identification Against Modeled Stream Flow #29
  • Analyze runoff from a single landseg/landuse
  • Create analysis script to generate a baseflow event time series Drought Forecasting Shiny Module #38 (comment)
  • Analyze flow from a watershed with a single (or majority?) landseg - Mount Jackson is 80% N51165, Coote's is smaller, but less than 80% N51165; could resegment model for Passage or Smith Creek (or Linville?) to find watershed with only 1 land segment overlapping. (note: may need to up the sophistication of our resegmentation routines to subdivide only a single land segment from the parent)
• What constitutes recharge in the model? AGWI hspf Recharge & ET data harvesting & Analysis #31
  • LZS is unavailable for stream flow (but subject to ET) NO We know that water flows from LZS into AGWS, but also does LZS flow into ... interflow (IFWO)???
  • We also know that LZS can be reduced by evaporation from deeply rooted vegetation
  • FALSE Is it true that AGWS is not reduced by evaporation? (AGWET)
• Annual total percolation (i.e., the models estimate of recharge)
  • Recharge in summer events? ("warm season recharge") hspf Recharge & ET data harvesting & Analysis #31
    • Should we predefine the quantitative measures for this (i.e., give us events where 3-day total precip is > 4")
    • Structure your code so that inches and days are function inputs so that this can be repurposed effortlessly
    • i.e. changes in LZS/AGWS during hurricanes
  • (maybe. probably not.) Raster analysis of GRACE data...
  • (maybe) How does groundwater storage relate to the land use on the land segments? Can we store (ts or just summary metrics) onto each landuse property of the land unit?
• Model Sensitivity Analysis
  • Increase/Decrease LZSN max
  • Increase/Decrease AGW related params
• Analyze drought model performance in Shenandoah:
  • What years/periods have good fit?
  • What years/periods have bad fit?
  • Where do current models perform well in multi-year droughts and baseflow recharge quantification?
  • Riparian ET (BASETP) is disabled, does riparian ET explain resurgence of flow in fall/winter?
  • Do good fit winter recharge simulations = Good fit summer drought?
  • Do poor fit winter recharge simulations = Good fit summer drought?
    • Error comparisons between summer and winter of the same year (perhaps summer low vs winter mean as a starting point)
• Drought flow prediction: worst case (no rain), this year + 2002 drought (we did this in 2023), this year + 2023 drought
  • Quantify the differences between base flow recharge in these case study years
• Shiny app Drought Forecasting Shiny Module #38
  • Will integrate the full work flow if no files have been created already (or if user requests overwrite)
  • Re-apply analyses to South Fork river to demonstrate reusability of scripts.
• Function library (tracking key building blocks in this project for later modularization) Function Library #39