Regression of in-water radiometric profile data

This study is addressed to the regression of in-water radiometric profiles with the objective of investigating solutions for enhancing the accuracy of data reduction results like subsurface radiance and irradiance values (Ed and Lu) and diffuse attenuation coefficients. Analyses have been conducted using optical profile generated through Monte Carlo simulations and field measurements. A nonlinear NL approach is presented as an alternative to the standard linear method LN. Results indicate that the LN method, relying on log-transformed data, tends to underestimate regression results with respect to NL, which operates directly on input radiometric values. The log-transformation is thus identified as the source of negative biases in data products. Observed differences between LN and NL regression results for Lu are of the order of 1-2%, that is well below the target uncertainty for data products from in situ measurements (i.e., 5%). For subsurface Ed, instead, difference can easily exceed the 5% threshold. Difference are much larger for Ed than Lu because the bias due to the log-transformation depends on the amplitude of perturbations induced by the effects of light focusing and defocusing induced by waves at the sea-surface. This work also remarks the importance of applying the multi-cast measurement scheme in view of ensuring repeatability of data products.

D'ALIMONTE Davide;  SHYBANOV Eugeny;  ZIBORDI Giuseppe;  KAJIYAMA Tamito; 

2013-11-13

OPTICAL SOC AMER

JRC81619

1094-4087,   

http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-21-23-27707,    https://publications.jrc.ec.europa.eu/repository/handle/JRC81619,   

10.1364/OE.21.027707,