correct Ecopuck bbp calculation

[messiem]

Ecopuck bbp (particulate backscatter in m-1) in variable ecopuck_bbp700 may in reality be beta (m-1 s-1).

[MBARIMike]

Information about the Ecopuck instrument installed on Dorado in September of 2020 starting with 2020.245.00 is sparse. The output is RAW values:

> head -10 /mbari/AUVCTD/missionlogs/2020/2020245/2020.245.00/FLBBCD2K.log
# binary FLBBCD2K
# timeTag time %8.8e ,time ,UNKNOWN
# integer NU_Chl %d ,NU_Chl ,RAW
# integer Chl_Sig %d ,Chl_Sig ,RAW
# integer NU_bb %d ,NU_bb ,RAW
# integer BB_Sig %d ,BB_Sig ,RAW
# integer NU_CDOM %d ,NU_CDOM ,RAW
# integer CDOM_Sig %d ,CDOM_Sig ,RAW
# integer Thermistor %d ,Thermistor ,RAW
# begin

There is a corresponding "device" file for each mission that appears to have some calibration parameters:

> cat /mbari/AUVCTD/missionlogs/2020/2020245/2020.245.00/FLBBCD2K-3695.dev
ECO 	FLBBCD2K-3695
Created on: 	10/29/2014

COLUMNS=9
N/U=1
N/U=2
N/U=3
CHL=4		0.0073		45
N/U=5
Lambda=6	1.633E-06	46	700	700
N/U=7
CDOM=8		0.0909		45
N/U=9

and there is this Calibration sheet for the backscatter output (the BB_Sig and Lambda rows from above):

The code that writes ecopuck_bbp700 is incorrect as the $\beta$ value appears to be Volume Scattering, not a particulate backscattering coefficient.

[MBARIMike]

We can test the code fix in #9 using the Notebook 6.1-mpm-cross-calibrate-dorado-pontus-backscatter.ipynb. It's possible that we can also use this same _compute_backscatter() function to properly compute bbp for the HS2 that's being tracked in #3.

[MBARIMike]

The Dorado mission used in 6.1-mpm-cross-calibrate-dorado-pontus-backscatter.ipynb was reprocessed with the corrections made in #10 and loaded into a temporary test database: stoqs_lrauv_jul2022_t. The 6.1-mpm-cross-calibrate-dorado-pontus-backscatter.ipynb Notebook was then updated to use the test database and the corrected units for bbp. The correlation patterns and R-squared values are nearly identical to the data from before these corrections were made. A rendered version of this Notebook will be shared in the internal plankton-proxies Slack channel.

[MBARIMike]

Comparing the new ecopuck bbp (particulate backscatter in m-1) calculation to the new corrected hs2 bbp computation can be done in STOQS using the test database:

Previous to these corrections (from the original non-test database), the slope was much further away from 1.0: