Approaches for characterizing uncertainties in in situ water-leaving radiance for ocean color calibration and validation

Susanne Elizabeth Craig, Dalhousie University, susanne.craig@dal.ca (Presenting)
Andrew Heath Barnard, WET Labs, Inc, andrew@wetlabs.com
Michael Twardowski, WET Labs, Inc, mtwardo@wetlabs2.com
Scott McLean, Satlantic, Inc, scott@satlantic.com
Ken Voss, University of Miami, voss@physics.miami.edu
Chuck Trees, NATO Undersea Research Centre, trees@nurc.nato.int
Marlon Lewis, Dalhousie University, marlon.lewis@dal.ca
J. Ronald Zaneveld, WET Labs, Inc, ron@wetlabs.com

To recover ocean water-leaving radiances and derive biophysical parameters from space-based ocean color sensors requires extensive validation and calibration (VC) efforts based on comparisons with in situ optical measurements. These VC efforts require detailed quantification of the in situ measurement uncertainties in order to obtain water leaving radiances that can be included in these activities. Substantial improvements can be made in the uncertainty associated with calibration/validation of ocean color sensors by combining hyperspectral apparent and inherent optical property measurements in oligotrophic and coastal areas, with state-of-the-art radiative transfer modeling. Our goal is to develop methodology to obtain high quality VC in situ radiance and inherent optical property data with well-quantified measurement uncertainty budgets. We present an overview of our methodology, which includes an empirical propagation of errors using radiative transfer modeling, and apply it to data collected at the MOBY site as part of the SORTIE project.