nmfs-swfsc-ast / ast-tasks

0 stars 0 forks source link

Compute biomass estimates #60

Closed kstierhoff closed 2 months ago

kstierhoff commented 8 months ago
jrenfree commented 8 months ago

Will focus on Transect 118, sampled by SD-1060:

2307SD1060_118_CPS-38 Remove Passive Pings

We conducted a pre- and post-survey calibration, which resulted in the following for 38 kHz

Parameter PRE POST
Gain 19.04 19.10
BW_along 17.8 17.9
BW_athwt 17.8 17.9
Offset_along 0.2 0.2
Offset_athwt -0.6 -0.9
EBA -12.6 -12.6
Sa_corr -0.04 0.03
RMS 0.16 0.13

So both were very similar, and we applied the Pre-survey results for the data processing. The template ECS file appears to have the correct environment and calibration parameters entered:

image

The CTD cast used to process this transect was cast 158 from the Shimada hake survey, which appeared to have a harmonic sound speed of 1495.67 m/s, which matches with what's in the ECS file used to process this transect:

image

The calibration parameters in the ECS value are from the template file but compensated by change in sound speeds. With a calibration (template) sound speed of 1501.37 and transect sound speed of 1495.67, my calculations are:

Parameter Template Transect
Gain 19.04 19.073
EBA -12.6 -12.633
BW_along 17.8 17.7324
BW_athwt 17.8 17.7324

This matches what's in the transect ECS file, so looks like it's being compensated correctly.

So unless both calibrations were way off for this echosounder, I think this checks out just fine. Suppose the question then becomes if the scatterers near the seabed are indeed CPS.

kstierhoff commented 8 months ago

Compelling and thorough analysis. I think we can consider that item completed and debunked.

kstierhoff commented 2 months ago

Final biomass estimates computed for all species/stocks, after accounting for deep backscatter from CSNA in the nearshore region of the SCB. Results incorporated into biomass report.