vlahm / rsfme

flux_estimation_methods_evaluation

TODO:

• make single USGS and HBEF annual flux calcs loop, which runs flux calcs by site-year, and also runs calcs for each site with all available years
• run this loop and share / first look at results
• look at HBEF Watershed 3, run full annual flux estimates for all site years, each site-year seperately, and ocmpare to 'true flux' from sensor data for NO3_N
• look at HBEF Watershed 3, run full annual flux estimates for all site years, each site-year seperately, and ocmpare to 'true flux' from sensor data for Ca AND HBEF offficial published estimate for Ca flux
• WRTDS results (currently) make sense as-is, need to rigorously document UNITS and ensure that it all maths out to kg/day
• look into model paramter 'tuning' does it have built in for settting a Max Q or Max Flux? If not, post processing?
• look at fdaily time series of years where extreme outlier flux overestimates are occuring- what kind of days/flows are these overestimates occuring at?
• diagnostic plots such as showing daily flux with the days of sampling indicated, and printed statistic about the 'balancedness' of the sampels across the year.
• compare plot Q:C by WY vs Q:C over all site data
• average annual hydrograph against hydrograph of any particular year
• improve USGS flux comparison plots, currently impossible to tell outliers or diff in good ones even

purpose

The purpose of this repo is to develop and maintain tools to evaluate how to best calculate riverine solute fluxes.

structure

main script

The master script is written to coordinate between all subscripts and data sources.

data

Most data in this analysis will be pulled from USGS servers during each analysis.

All data that is not pulled from in-script web requests will reside in the data folder. The data folder is organized following the same procedure as Macrosheds (https://github.com/MacroSHEDS/macrosheds), nested by domain and site.

source

Each method of flux estimation is replicated in it's own subscript. These scripts and any associated helper functions are stored in the source folder.

out

The output from selected methods and visualizations specificed in the master script are output to the out folder.