search

dankelley / oceglider

R package for processing ocean glider data
https://dankelley.github.io/oceglider/
3 stars 1 forks source link

odd near-surface values #32

Closed dankelley closed 5 years ago

dankelley commented 5 years ago

At an f2f meeting today, @richardsc asked that I make a sandbox for something I was showing on my computer screen. So, I've done that. It can be seen in sandbox/dk/01_odd_surface (in the "develop" branch). That directory has an .Rmd file and the .pdf it creates. (I do not normally include .pdf files in repositories, but since the package is under active development and readers might not have the datafiles, I am doing so here, because the file might be hard to reproduce.)

I have assigned myself this issue, although I am not asking myself the questions in the report; rather, I'm asking my BIO glider pals.

dankelley commented 5 years ago

Perhaps the data in these seaexplorer files have been run through seabird processing steps that involve averaging previous measurements. If that averaging was done, then the pre-descent values would get blurred into the deep-water (start of ascent) values.

I suppose that must be the explanation.

This means that I was not understanding conversations properly. I had thought that this sort of boxcar averaging had not been done prior to creating the raw files, but only later during the processing that led to the .nc files that had been in my mind, before I started using our new code to read the raw seaexplorer files.

richardsc commented 5 years ago

Sorry, I'm only just getting around to looking at the pdf. I'm having a hard time figuring out what is going on ...

To my knowledge, there is no onboard averaging of the data -- it should be just raw data coming out of the CTD. But, as we've already seen with issue #8 because the data are first collected by the glider computer before being written to the "raw" csv files there are some strange things that can happen.

I'm starting to wonder if it might be related to the way that the glider is sampling with the CTD, and that the CTD is being turned off and on. Perhaps when it is first turned on (after being off), it still has water in the plumbing from the last time the pump was running and it takes some time to clear that all out?

dankelley commented 5 years ago

This issue was addressed a long time ago, in various f2f discussions. I won't bother summarizing all the ins and outs, but instead will simply close the issue.