search

PNNL-TES / TES-drydown

MIT License
2 stars 1 forks source link

Concentration visualizations #11

Open stephpenn1 opened 3 years ago

stephpenn1 commented 3 years ago

Comparison of the distribution of concentration data from EGM (red) and picarro (blue), shown as a density plot.

CO2 concentration on the x axis and kernel density estimate on the y, which is a smoothed version of a histogram. image

This is a potential way to show treatment differences. The dot represents the mean and the lines are the standard deviation for the Picarro data only.

I had some trouble trying to match Core and Sample_number and a corresponding treatment, hence the NA column. image

A few questions:

  1. Is there a dataframe that has a treatment label for each sample number and core?
  2. Do you want to visualize the concentration data over time?
  3. Overall thoughts?
kaizadp commented 3 years ago
  1. Is there a dataframe that has a treatment label for each sample number and core?

data/ cpcrw_valve_map.csv and sr_valve_map.csv

  1. Do you want to visualize the concentration data over time?

I don't think we need concentration data over time, we already have some time-series graphs. I think binning them by treatment is the most straightforward way for now.

  1. Overall thoughts?
    • The distribution comparisons are very interesting. It's good to see a similar general trend, although we'd have some differences for sample number, sample type, etc.
    • I like the general layout of the second plot.
    • but -- some of the patterns seem strange to me. (drought should be very low; sat should be high). Possibly because we're mixing different drought durations (30d, 90d, 150d), dry down type (FAD, CW) and sites(?).
    • because we have so many data points and so much variability, I prefer plotting the actual points instead of just the mean/sd
    • I'll check the NAs. Which site was this?
stephpenn1 commented 3 years ago

I'm not super familiar with the different levels of data.. this is actually all of it together by treatment...oops