search

COMPASS-DOE / fluxfinder

File Parsing, Flux Computation, and Diagnostics of Greenhouse Gas Measurements
Other
1 stars 0 forks source link

QAQC: y-intercept checks #43

Open wilsonsj100 opened 7 months ago

wilsonsj100 commented 7 months ago

There has been a suggestion by Pat M. that we include some version of QAQC for the y-intercept of fluxes. In short, this is the idea that if you were taking a flux outside, your initial concentration right as you start the flux should be the same as atmospheric concentrations. This gets a bit complicated because we do not always start fluxes (i.e. start times) when we initially put the chamber down, making it potentially difficult to find this starting concentration. Another complication would be conducting laboratory experiments where the indoor concentration is higher than atmospheric or if you were measuring fluxes under an elevated CO2/CH4 scenario and your starting point will therefore be different. Another scenario I could envision would be someone measuring fluxes from the same plot or bottle overtime and not removing the chamber - ie. taking a measurement then waiting ten minutes and doing it again, but not removing the chamber. This would make it hard to assess this and they would need to look for jumps in the data as an indicator of potential ebullition.

After talking with Genevieve we think a way to solve this issue is to add some text about the importance of visually checking the y-intercept and adding a horizontal line to figures of data in the quarto for atmospheric concentrations as an example with an atmospheric starting point. This can give users some code and visual example of how they could check their y-intercepts by adding a line or band at their anticipated starting concentration in order to check this. If one of their fluxes has a starting point that is far from what might be expected, they may want to further investigate why.

So, I am thinking about adding an atmospheric concentration line to the examples in the manuscript as a way to exemplify this, then thinking about adding this also to the quarto. Let me know what you think

bpbond commented 7 months ago

We can add some y-intercept QAQC to the Rmarkdown, no problem.

adding a horizontal line to figures of data in the quarto for atmospheric concentrations as an example with an atmospheric starting point

What value are we taking as atmospheric concentration for this line?

bpbond commented 7 months ago

Add optional parameters to qaqc() for ambient atmospheric levels; these will then be added to all plots.