Article ID Journal Published Year Pages File Type
4460386 Remote Sensing of Environment 2008 19 Pages PDF
Abstract

Accurate quantification of photo-induced marine processes requires a realistic description of UV radiation (UVR) penetration in the surface ocean. Remote sensing of ocean color could supply global views of UVR attenuation if reliable algorithms were made available. Although a few models for the retrieval of UV diffuse attenuation coefficients, Kd(λ), have been proposed in the past, robust and properly validated algorithms are still needed. We have developed and validated two algorithms for the retrieval of Kd(λ) at λ = 320, 340, 380, 412, 443, 490 nm from multispectral remote-sensing reflectances, Rrs(λ) (SeaWiFS wavebands). These algorithms, SeaUV and SeaUVC, were developed from a large “training dataset” comprising simultaneous in situ measurements of Rrs(λ) and Kd(λ) made in a variety of water types ranging from blue oligotrophic to turbid estuarine systems. Four orthogonal dimensions of ocean color defined from the original Rrs(λ) spectra are used as predictors in multiple linear regressions of Kd(λ) to develop the algorithms. In the case of SeaUVC, an optical classification permits the development of water-type specific parameterizations. Validation against an independent synthetic Hydrolight® data set and an in situ data set showed that the algorithms can be used with confidence over the range of water types for which they have been developed. SeaUV and SeaUVC will find applicability in the fields of marine photochemistry and photobiology where their performance and inherent simplicity make them suitable for batch implementation on large data sets and at global scales.

Related Topics
Physical Sciences and Engineering Earth and Planetary Sciences Computers in Earth Sciences
Authors
, , ,