search

PNNL-PREMIS / bc-incubation2018

Aditi's 2018 Beaver Creek incubation
MIT License
0 stars 0 forks source link

flux analyses #6

Open AditiGit opened 4 years ago

AditiGit commented 4 years ago

Evaluate the following:

  1. Average/median and summary statistics of flux per site temporally or at the end of the experiment.
  2. Immediate, 24 hr, 48 hr, 1 week response after each inundation event per site.

Plots for 1 and 2 as well as a .csv version that can be used as variables for statistical analyses downstream

bpbond commented 4 years ago

@AditiGit Is this for me, or you?

AditiGit commented 4 years ago

For you @bpbond.

bpbond commented 4 years ago

@AditiGit

fluxes_co2

Average/median and summary statistics of flux per site temporally or at the end of the experiment.

What does "per site" mean? Are BC13, BC14, etc. different sites, so "BC3-2" is core 2 and BC3?

Immediate, 24 hr, 48 hr, 1 week response after each inundation event per site.

When does the inundation event occur? I.e. if not at time zero, how do I know?

Sorry, fuzzy on the experimental details, please remind me.

AditiGit commented 4 years ago

Sent inundation event date/time and experimental details over email. Let me know if you need additional information/phone call to discuss in detail.

bpbond commented 4 years ago

The BCx-y were pre-inundation cores (these were a set of 5 replicates from each site that were held for a week in the chamber after being brought from the field, and then destructively harvested. Apart from these 35 cores, the rest of the cores were split into 5 control and 5 inundation cores from each site. The inundation cores were treated with seawater. Both inundated and non-inundated controls were then harvested at the end for two depths (because we suddenly decided to change plans and sample at both depths...which we hadn’t done for the pre-inundated cores). The sites are BC-2,3,4,12,14,and 15.

As a reminder, each site (BC2, BC3, BC4, BC12, BC13, BC14, and BC15) had 15 cores (5 replicates for pre-inundation control, 5 replicates for control and 5 replicates for inundation). The preinundation cores were harvested before the experiment began. The control and inundation cores were monitored throughout the experiment. The inundated cores received seawater thrice. I added seawater in a staggered manner, as follows: On 7/24/2018, inundation cores from BC2, 3, 4, and 13 received seawater On 7/31/2018, inundation cores BC12, BC14, and BC15 received seawater On 8/14/2018, inundation cores from BC2, 3, 4, and 15 received seawater On 8/21/2018, inundation cores from BC12, BC13, and BC14 received seawater On 9/4/2018, inundation cores from BC2, 3, 4, and 15 received seawater On 9/11/2018, inundation cores from BC12, BC13, and BC14 received seawater

bpbond commented 4 years ago

@AditiGit

fluxes

Average/median and summary statistics of flux per site temporally or at the end of the experiment.

Can you be more specific please? You want over the entire experiment mean and median per site...what about treatment? Also see email about cores missing from inundation data file. Thanks!

AditiGit commented 4 years ago

Sent the edited core id to include BC15-6 and BC3-5.

  1. Need treatment specific mean and median at each site. So envisioning three panels per site (Pre-inundation, control-cores, inundation cores).
  2. For sites BC2, BC3,BC4, and BC15 or BC12, BC13, BC14, and BC15 (imagining a transect from creek to upland), how does flux change or not. Can be done again by looking at treatment specific response per site and also by evaluating post-inundation 24-hr response between inundated and control cores. Trying to understand if proximity to creek (and therefore vegetation and salinity gradient) impacts the flux. It is not a straightforward salinity gradient (BC3 is sometimes has higher salinity in field since it is low in elevation and tidal water sits in the area for quite a while).
bpbond commented 4 years ago

Inundation events for each core.

inundations_qc

bpbond commented 4 years ago

Need treatment specific mean and median at each site. So envisioning three panels per site (Pre-inundation, control-cores, inundation cores).

Like this?

inundations_meanmedian

AditiGit commented 4 years ago

Thanks. Yes. Please drop the Pre-inundatedControl so as to visualize only control v inundated during the incubation time period.

bpbond commented 4 years ago

This is over the whole incubation period. inundations_meanmedian

AditiGit commented 4 years ago

Thanks. These work. Couple of things:

  1. Are there plots for CH4?
  2. Can you send the data associated with this figure (and CH4 if available)? I can run some statistics.
bpbond commented 4 years ago

OK. Other variants/color possible. I will generate CH4 ones as well and send via email.