SeaBAM Publications

Readme

		Date: Mon, 24 Mar 97 8:20:20 PST

SeaBAMers,

Our preliminary report is in.
  
We have included 44 more stations from CalCOFI cruises 9608 and 9610.

As we got more data points in the 5-20 mg/m3 Chl range we have
updated our CalCOFI models for estimating Chl-a and Chl-a + Phaeo.

In short, the CalCOFI 3- and 4-band linear regression models do not seem
to add predictive power to single band ratio models. They also
tend to be very specific to the particular dataset. Instead, a better fit
to the data can be achieved by adding quadratic (and cubic) terms to the
linear (in the log-log space) Rrs490/Rrs555 ratio. The higher terms follow
well the curvature in the high Chl region.  In the very low Chl range
(0.05 mg/m3 and lower) - CalCOFI does NOT have any values below 0.05 mg/m3 -
these higher terms bend the curve in the wrong direction and therefore
the linear fit works better there.

The current models are given below. The models are
(all for both Chl-a and Chl-a + Phaeo):

Rrs490/Rrs555 linear 
Rrs490/Rrs555 linear, low Chl (derived from Rrs490/Rrs555 > 2.0)
Rrs490/Rrs555 quadratic
Rrs490/Rrs555 cubic
3-band linear (Rrs490/Rrs555, Rrs510/Rrs555)
4-band linear (Rrs443/Rrs555, Rrs412/Rrs510)

The best fit is achieved with the Rrs490/Rrs555 cubic (the quadratic is
practically the same).

Regards,

Mati

----------------------------------------------------------------------
Date: Thu, 22 May 97 

Files updated to 'version 2'.

seabam_report_2.ps  - the text and tables in postscript format
ALL_FIGS.eps        - figures 1 to 13 in EPS format
figs_eps.tar.Z      - figures 1 to 13 in EPS format, tarred, compressed

Other formats of same docs (MS WORD 97, Mac WORD 5.1) available at SPG web site:
ftp://spg.ucsd.edu/pub/seabam/report

-Brian
----------------------------------------------------------------------
Files

SeaBASS

Readme

Files
