Volumetric estimation of thermal fields inferred from glider-like and remote-sensing measurements in undersampled coastal regions

Abstract

A method to estimate volumetric oceanographic conditions in a given coastal area merging information from remote sensing and gliders is investigated. The method relies on the application of a linear model that makes use of a degenerated 3-D covariance model inferred from in situ and remote-sensing measurements and takes into account the functional nature of glider data. The optimum weighting function results from a Fredholm integral equation of the first kind. A centroid collocation method with piecewise constant bases and a specific class of stabilizing functional are used to numerically solve the problem. The investigated methodology was validated during the oceanographic cruise TSS08 conducted by the NATO Undersea Research Center in September 2008 in a region of the Black Sea, offshore the Bosphorus Strait. Data sampled with a towed vehicle emulating a glider were merged with remote-sensing information to estimate the 3-D temperature field at unobserved locations. A set of the observations was only considered for validation. Estimations from the functional model are confronted with those obtained from a discrete approach. Results indicate a better performance of the functional model and raise the question of whether functional techniques are more appropriate than algebraic ones to analyze high-resolution data obtained from new robotic platforms. Copyright 2010 by the American Geophysical Union.