search

leppott / ContDataQC

Quality control checks on continuous data. Example data is from a HOBO data logger with 30 minute intervals.
https://leppott.github.io/ContDataQC/
MIT License
20 stars 10 forks source link

QC check vs. daily mean #121

Open leppott opened 3 years ago

leppott commented 3 years ago

Is your feature request related to a problem? Please describe.

Another check could be to compare value to daily mean.

Rate of Change not as useful to some users. Would prefer to compare to daily mean. RoC compares to mean and 3*stdev for the previous 24 hours.

Describe the solution you'd like Not sure of what testing.

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context daily mean comparison is a check used by another group.

leppott commented 3 years ago

Add to config file as a new QC test.

leppott commented 3 years ago

this would be a check for a given day (not a rolling time period).

E.g., max-min for April 12. then Max-min for April 13.

It'd be a simple threshold (like for the spike test...e.g., >5 C diurnal change = S, >8 C diurnal change F.

leppott commented 3 years ago

Tasks:

  1. Add thresholds to config.R (and all its variations in the package and shiny app but have a script to do that part).

  2. Do the calculation of daily max - daily min and associate with each point.

  3. Evaluate each point and assign Pass, Fail, Suspect.

  4. Add the new QC check to the overall Flag check for each site.

  5. Add the new QC check to the QC Notebook (RMD).