Article

Ocean Dynamics

, Volume 61, Issue 8, pp 1205-1214

Open Access This content is freely available online to anyone, anywhere at any time.

Reconstruction of MODIS total suspended matter time series maps by DINEOF and validation with autonomous platform data

  • Bouchra NechadAffiliated withManagement Unit of the North Sea Mathematical Models (MUMM), Royal Belgian Institute of Natural Sciences (RBINS) Email author 
  • , Aida Alvera-AzcaràteAffiliated withGeoHydrodynamics and Environmental Research (GHER), University of Liège, Belgium
  • , Kevin RuddickAffiliated withManagement Unit of the North Sea Mathematical Models (MUMM), Royal Belgian Institute of Natural Sciences (RBINS)
  • , Naomi GreenwoodAffiliated withCentre for Environment, Fisheries and Aquatic Sciences (Cefas)

Abstract

In situ measurements of total suspended matter (TSM) over the period 2003–2006, collected with two autonomous platforms from the Centre for Environment, Fisheries and Aquatic Sciences (Cefas) measuring the optical backscatter (OBS) in the southern North Sea, are used to assess the accuracy of TSM time series extracted from satellite data. Since there are gaps in the remote sensing (RS) data, due mainly to cloud cover, the Data Interpolating Empirical Orthogonal Functions (DINEOF) is used to fill in the TSM time series and build a continuous daily “recoloured” dataset. The RS datasets consist of TSM maps derived from MODIS imagery using the bio-optical model of Nechad et al. (Rem Sens Environ 114: 854–866, 2010). In this study, the DINEOF time series are compared to the in situ OBS measured in moderately to very turbid waters respectively in West Gabbard and Warp Anchorage, in the southern North Sea. The discrepancies between instantaneous RS, DINEOF-filled RS data and Cefas data are analysed in terms of TSM algorithm uncertainties, space–time variability and DINEOF reconstruction uncertainty.

Keywords

Total suspended matter (TSM) MODIS Data Interpolating Emprirical Orthogonal Functions (DINEOF) Cefas Optical backscatter TSM algorithm TSM time series