Retrieval of the spectral diffuse attenuation coefficientKd(λ) in open and coastal ocean waters using a neural network inversion

Jamet, Cédric; Loisel, Hubert; Dessailly, David

doi:10.1029/2012jc008076

Cited by 87 publications

(64 citation statements)

References 76 publications

(129 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been proven in the last years that NNs produce reasonable approximations of ocean color products from optically complex (Case-2) waters. NNs have been applied to different satellite sensors in order to derive concentrations of water constituents, inherent and apparent optical properties (IOPs and AOPs), and photosynthetically available radiation (PAR), or to discriminate algae species (Gross , 1999;Schiller and Doerffer, 1999;D'Alimonte and Zibordi, 2003;Zhang et al, 2003;Tanaka et al, 2004;Schiller, 2006;Bricaud et al, 2007;Schroeder et al, 2007;Ioannou et al, 2011;Jamet et al, 2012;Chen et al, 2014;Hieronymi et al, 2015;D'Alimonte et al, 2016). Due to their speed, NN-based ocean color algorithms are deployed for operational and near-real time satellite observations, e.g., the MERIS Case-2 water algorithm (Doerffer and Schiller, 2007) and C2RCC .…”

Section: Introductionmentioning

confidence: 99%

The OLCI Neural Network Swarm (ONNS): A Bio-Geo-Optical Algorithm for Open Ocean and Coastal Waters

2017

View full text Add to dashboard Cite

The processing scheme of a novel in-water algorithm for the retrieval of ocean color products from Sentinel-3 OLCI is introduced. The algorithm consists of several blended neural networks that are specialized for 13 different optical water classes. These comprise clearest natural waters but also waters reaching the frontiers of marine optical remote sensing, namely extreme absorbing, or scattering waters. Considered chlorophyll concentrations reach up to 200 mg m −3 , non-algae particle concentrations up to 1,500 g m −3 , and the absorption coefficient of colored dissolved organic matter at 440 nm is up to 20 m −1 . The algorithm generates different concentrations of water constituents, inherent and apparent optical properties, and a color index. In addition, all products are delivered with an uncertainty estimate. A baseline validation of the products is provided for various water types. We conclude that the algorithm is suitable for the remote sensing estimation of water properties and constituents of most natural waters.

show abstract

Section: Introductionmentioning

confidence: 99%

The OLCI Neural Network Swarm (ONNS): A Bio-Geo-Optical Algorithm for Open Ocean and Coastal Waters

2017

View full text Add to dashboard Cite

show abstract

“…This kind of empirical approach has been adopted by various ocean color sensors (OCM, SeaWiFS, MODIS, MERIS, OCTS, CZCS) for operational use with good results over clear open ocean waters. Problems associated with previous studies may be overcome by a neural network algorithm 6 , still its applicability in a wide range of waters needs to be assessed.…”

Section: Datamentioning

confidence: 99%

“…K d mainly depends on the inherent optical properties (IOPs) and the angular distribution of the surface light field [4][5][6] . K d can be considered as an indicator of the turbidity in the water column 7 , thus, the greater the attenuation of light, the lower the clarity.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of downwelling diffuse attenuation coefficient algorithms in the Red Sea

2016

View full text Add to dashboard Cite

Despite the importance of the optical properties such as the downwelling diffuse attenuation coefficient for characterizing the upper water column, until recently no in situ optical measurements were published for the Red Sea. Kirby et al.1 used observations from the Coastal Zone Color Scanner to characterize the spatial and temporal variability of the diffuse attenuation coefficient (K d (490)) in the Red Sea. To better understand optical variability and its utility in the Red Sea, it is imperative to comprehend the diffuse attenuation coefficient and its relationship with in situ properties. Two apparent optical properties, spectral remote sensing reflectance (R rs ) and the downwelling diffuse attenuation coefficient (K d ), are calculated from vertical profile measurements of downwelling irradiance (E d ) and upwelling radiance (L u ). K d characterizes light penetration into water column that is important for understanding both the physical and biogeochemical environment, including water quality and the health of ocean environment. Our study tests the performance of the existing K d (490) algorithms in the Red Sea and compares them against direct in situ measurements within various subdivisions of the Red Sea. Most standard algorithms either overestimated or underestimated with the measured in situ values of K d . Consequently, these algorithms provided poor retrieval of K d (490) for the Red Sea. Random errors were high for all algorithms and the correlation coefficients (r 2 ) with in situ measurements were quite low. Hence, these algorithms may not be suitable for the Red Sea. Overall, statistical analyses of the various algorithms indicated that the existing algorithms are inadequate for the Red Sea. The present study suggests that reparameterizing existing algorithms or developing new regional algorithms is required to improve retrieval of K d (490) for the Red Sea.

show abstract

“…Fichot et al [35] proposed a method based on principal component analysis to derive K d at 320, 340, 380, 412, 443, and 490 nm using data from the multispectral bands of SeaWiFS. Similarly, Jamet et al [36] used an ANN-based algorithm to compute K d (490) in coastal waters using multispectral remote sensing reflectance. The use of hyperspectral data would certainly improve the retrieval of the diffuse attenuation coefficient for statistically based approaches, provided that the training dataset is large enough and encompasses a variety of optical environments such as those encountered in coastal waters.…”

Section: Backscatteringmentioning

confidence: 99%

“…The higher number of wavebands available with HyspIRI would give more opportunities to test different combinations of wavelengths and undoubtedly improve the accuracy with which K d (490) could be retrieved. Moreover, a simple linear regression or a more complex formulation [34] of band ratios against K d (490), or algorithms based on more than two wavelengths [35,36], could be used to improve the retrieval of K d (490) in coastal waters. Given the number of wavebands available on HyspIRI, the diffuse attenuation coefficients could also be derived at other wavelengths besides 490 nm using simple mathematical formulations.…”

Section: Backscatteringmentioning

confidence: 99%

Future Retrievals of Water Column Bio-Optical Properties using the Hyperspectral Infrared Imager (HyspIRI)

et al. 2013

View full text Add to dashboard Cite

Interpretation of remote sensing reflectance from coastal waters at different wavelengths of light yields valuable information about water column constituents, which in turn, gives information on a variety of processes occurring in coastal waters, such as primary production, biogeochemical cycles, sediment transport, coastal erosion, and harmful algal blooms. The Hyperspectral Infrared Imager (HyspIRI) is well suited to produce global, seasonal maps and specialized observations of coastal ecosystems and to improve our understanding of how phytoplankton communities are spatially distributed and structured, and how they function in coastal and inland waters. This paper draws from previously published studies on high-resolution, hyperspectral remote sensing of coastal OPEN ACCESSRemote Sens. 2013, 5 6813 and inland waters and provides an overview of how the HyspIRI mission could enable the retrieval of new aquatic biophysical products or improve the retrieval accuracy of existing satellite-derived products (e.g., inherent optical properties, phytoplankton functional types, pigment composition, chlorophyll-a concentration, etc.). The intent of this paper is to introduce the development of the HyspIRI mission to the coastal and inland remote sensing community and to provide information regarding several potential data products that were not originally part of the HyspIRI mission objectives but could be applicable to research related to coastal and inland waters. Further work toward quantitatively determining the extent and quality of these products, given the instrument and mission characteristics, is recommended.

show abstract

Retrieval of the spectral diffuse attenuation coefficientK_d(λ) in open and coastal ocean waters using a neural network inversion

Cited by 87 publications

References 76 publications

The OLCI Neural Network Swarm (ONNS): A Bio-Geo-Optical Algorithm for Open Ocean and Coastal Waters

The OLCI Neural Network Swarm (ONNS): A Bio-Geo-Optical Algorithm for Open Ocean and Coastal Waters

Evaluation of downwelling diffuse attenuation coefficient algorithms in the Red Sea

Future Retrievals of Water Column Bio-Optical Properties using the Hyperspectral Infrared Imager (HyspIRI)

Contact Info

Product

Resources

About