Use of ocean colour remote sensing to monitor sea surface suspended sediments

Sabatino, A.D. and Clement, R. and Heath, M.R. and McKee, D.; Side, Jonathan, ed. (2015) Use of ocean colour remote sensing to monitor sea surface suspended sediments. In: TeraWatt Position Papers. Marine Alliance for Science and Technology Scotland (MASTS), St Andrews, pp. 129-140. ISBN 978-0-9934256-0-8

[thumbnail of Sabatino-etal-TeraWatt-Position-Papers-2015-Use-of-ocean-colour-remote-sensing-to-monitor-sea-surface-suspended]
Text. Filename: Sabatino_etal_TeraWatt_Position_Papers_2015_Use_of_ocean_colour_remote_sensing_to_monitor_sea_surface_suspended.pdf
Final Published Version

Download (1MB)| Preview


Ocean colour remote sensing (OCRS) from satellite platforms has revolutionised our ability to monitor the interplay of physical and biogeochemical processes in surface waters of the ocean. Since the launch of SeaWiFS in 1996, a continuous time series of OCRS data has been accumulated from a series of satellite sensors giving near daily global coverage. These sensors measure top of atmosphere (TOA) spectral radiance which is corrected for atmospheric effects (~80% of the measured signal in the blue - Gordon 1978) to give water leaving radiances. From these putrely optical signals, it is possible to derive a wide range of higher level products such as chlorophyll concentration, diffuse attenuation coefficients, photosynthetically available radiation (PAR) and a wide range of inherent optical properties (IOPs) to name but a few. In terms of surface area and primary productivity, the global ocean is heavily dominated by deep, oceanic waters, where the optical properties are driven by phytoplankton, associated dissolved organics and water itself. It is little surprise then that early standard OCRS products were developed for optimal performance over these globally significant regions. Standard chlorophyll algorithms were developed using changes in blue-green reflectance ratios (e.g. O’Reilley et al., 1998) that can be related to the effect of changing concentrations of microscopic scale (1µm-200µm) phytoplankton (Kirk,1983) forming blooms that can stretch for thousands of km. More recently, attention has shifted to economically important coastal regions where, for example, harmful algal blooms have potential to cause significant societal and economic impact. OCRS algorithms have been developed to specifically aid in the monitoring of both toxic species e.g. Karenia brevis in the Gulf of Mexico (Stumpf et al., 2003), and also to monitor for extreme eutrophication events where excessive levels of phytoplankton cause the reduction of oxygen dissolved in the water column (hypoxia) leading to animal mortality (e.g. Mallin et al., 2006). The optically complex nature of coastal waters, more generally, presents a particular problem for OCRS applications in these regions. Shallow shelf seas and other inshore waters are subject to the influence of sediment resuspension and freshwater discharge bringing additional loads of coloured dissolved organic materials (CDOM). This results in multiple, independently varying, optically significant components, each of which influences the water leaving radiance spectrum making interpretation of spectral changes significantly more difficult. Many studies have demonstrated the breakdown in performance of standard algorithms (e.g. Chl, McKee et al. 2007) in optically complex coastal waters. In this paper we will focus on the effect of suspended sediment on optical properties of the water column. Suspended sediment has long been known to influence light penetration (Gordon and McCluney, 1975) which can limit primary production and also contribute to hypoxia (Greig et al., 2005). There is interest in monitoring sediment concentration for coastal erosion applications and various OCRS algorithms have been developed that exploit the relatively strong backscattering properties of sediment. For example, Doxaran et al. (2002) successfully presented a sediment algorithm for the highly turbid Gironde estuary. More recently a radiative transfer approach was used to refine this type of approach to incorporate the potential impact of other materials on the red reflectance values that support sediment algorithms (Neil et al., 2011). This algorithm provides estimates of maximum and minimum sediment load concentrations assuming reasonable potential ranges of Chl and CDOM for coastal waters. The aim of this paper is to determine the extent to which the Neil et al. algorithm, which was developed for Irish Sea waters, can be applied to data collected in the North Sea. The ultimate goal is to assess the potential for using OCRS data to monitor suspended sediment concentrations in coastal waters, with monitoring marine turbine arrays an obvious and potentially important application.