Ancillary Data Uncertainties within the SeaDAS Uncertainty Budget for Ocean Colour Retrievals

Atmospheric corrections introduce uncertainties in bottom-of-atmosphere Ocean Colour (OC) products. In this paper, we analyse the uncertainty budget of the SeaDAS atmospheric correction algorithm. A metrological approach is followed, where each of the error sources are identified in an uncertainty t...

Full description

Saved in:
Bibliographic Details
Published inRemote sensing (Basel, Switzerland) Vol. 14; no. 3; p. 497
Main Authors De Vis, Pieter, Mélin, Frédéric, Hunt, Samuel E., Morrone, Rosalinda, Sinclair, Morven, Bell, Bill
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.01.2022
Subjects
Aerosols
Algorithms
ancillary parameters
Atmosphere
Atmospheric correction
Budgets
Color
Data assimilation
Datasets
Ocean color
ocean colour
Relative humidity
Remote sensing
Sensors
Uncertainty
Variables
Weather forecasting
Wind speed
Online AccessGet full text

Cover

More Information
Summary:Atmospheric corrections introduce uncertainties in bottom-of-atmosphere Ocean Colour (OC) products. In this paper, we analyse the uncertainty budget of the SeaDAS atmospheric correction algorithm. A metrological approach is followed, where each of the error sources are identified in an uncertainty tree diagram and briefly discussed. Atmospheric correction algorithms depend on ancillary variables (such as meteorological properties and column densities of gases), yet the uncertainties in these variables were not studied previously in detail. To analyse these uncertainties for the first time, the spread in the ERA5 ensemble is used as an estimate for the uncertainty in the ancillary data, which is then propagated to uncertainties in remote sensing reflectances using a Monte Carlo approach and the SeaDAS atmospheric correction algorithm. In an example data set, wind speed and relative humidity are found to be the main contributors (among the ancillary parameters) to the remote sensing reflectance uncertainties.
ISSN:2072-4292
2072-4292
DOI:10.3390/rs14030497