This session on in‐water algorithms will trace the history of the development of algorithmic methods applied to Ocean Colour data starting with the very first algorithms applied to global datasets obtained from the CZCS sensor. We will outline the evolution of algorithm development as our knowledge (and data!) on the optical properties of both open‐ocean and coastal waters have improved over the last three decades. Additionally we will examine the mathematical and statistical approaches (neural networks, non‐linear optimisation, spectral un‐mixing, principal component analysis etc.) that have been explored to make best use in using the radiometric quantities measured by the sensors in retrieving the relevant geophysical quantities of interest. Specific attention will be placed on emphasising the complexities of applying such methods in coastal regions, and considerations will be made on the limitations and uncertainties that need to be understood. Furthermore we will analyse the parallel progress of both the empirical and semianalytical method, and consider the merits and deficiencies of each of these, providing a clear understanding of the difference between these methods and their practical application in the operational processing of data (see flowchart below). Complimentary to this we will specifically consider the results from an intensive round robin intercomparison of different semi‐analytical methods (performed by the IOCCG). In considering all of these various aspects of different available algorithms we will underline which algorithms have been considered for routine processing by the major space agencies (and why). Continuing we shall address the relative benefit of using standard global coverage products compared to regional algorithms and vice versa, and explore various alternatives for the implementation of regional algorithms. Here we will investigate the “minimum requirements” for the implementation of such regional algorithms (i.e. required datasets, “Level” of satellite data required, computing requirements). Finally we will make some considerations on the future direction for research on these topics. And deal with any real world examples/questions that participants may have and want to address. Bibliography In preparation for the course, it is suggested that participants consult the following, freely available, IOCCG reports (a more detailed bibliography, on specific topics, will be provided during the course): Working group on Ocean Colour in Case 2 Waters (Chaired by Shubha Sathyendranath): IOCCG Report 3 (2000). Remote Sensing of Ocean Colour in Coastal, and Other Optically‐Complex, Waters (http://www.ioccg.org/reports/report3.pdf) Working group on Ocean Colour Algorithms (Chaired by ZhongPing Lee): IOCCG Report 5 (2006). Remote Sensing of Inherent Optical Properties: Fundamentals, Tests of Algorithms, and Applications. (http://www.ioccg.org/reports/report5.pdf)
Objectives This lecture is designed to provide an overview of the fundamentals of Inherent Optical Properties (IOPs), its relationship with AOPs, algorithms to invert IOPs from AOPs, as well as applications of IOPs. Topics Lecture 1: Fundamentals of IOPs and IOP‐AOP relationships Approach I will give focused lectures along with hands‐on practices. Advanced reading materials will be handed out to broaden knowledge.
This session on in‐water algorithms will trace the history of the development of algorithmic methods applied to Ocean Colour data starting with the very first algorithms applied to global datasets obtained from the CZCS sensor. We will outline the evolution of algorithm development as our knowledge (and data!) on the optical properties of both open‐ocean and coastal waters have improved over the last three decades. Additionally we will examine the mathematical and statistical approaches (neural networks, non‐linear optimisation, spectral un‐mixing, principal component analysis etc.) that have been explored to make best use in using the radiometric quantities measured by the sensors in retrieving the relevant geophysical quantities of interest. Specific attention will be placed on emphasising the complexities of applying such methods in coastal regions, and considerations will be made on the limitations and uncertainties that need to be understood. Furthermore we will analyse the parallel progress of both the empirical and semianalytical method, and consider the merits and deficiencies of each of these, providing a clear understanding of the difference between these methods and their practical application in the operational processing of data (see flowchart below). Complimentary to this we will specifically consider the results from an intensive round robin intercomparison of different semi‐analytical methods (performed by the IOCCG). In considering all of these various aspects of different available algorithms we will underline which algorithms have been considered for routine processing by the major space agencies (and why). Continuing we shall address the relative benefit of using standard global coverage products compared to regional algorithms and vice versa, and explore various alternatives for the implementation of regional algorithms. Here we will investigate the “minimum requirements” for the implementation of such regional algorithms (i.e. required datasets, “Level” of satellite data required, computing requirements). Finally we will make some considerations on the future direction for research on these topics. And deal with any real world examples/questions that participants may have and want to address. Bibliography In preparation for the course, it is suggested that participants consult the following, freely available, IOCCG reports (a more detailed bibliography, on specific topics, will be provided during the course): Working group on Ocean Colour in Case 2 Waters (Chaired by Shubha Sathyendranath): IOCCG Report 3 (2000). Remote Sensing of Ocean Colour in Coastal, and Other Optically‐Complex, Waters (http://www.ioccg.org/reports/report3.pdf) Working group on Ocean Colour Algorithms (Chaired by ZhongPing Lee): IOCCG Report 5 (2006). Remote Sensing of Inherent Optical Properties: Fundamentals, Tests of Algorithms, and Applications. (http://www.ioccg.org/reports/report5.pdf)
Objectives This lecture is designed to provide an overview of the fundamentals of Inherent Optical Properties (IOPs), its relationship with AOPs, algorithms to invert IOPs from AOPs, as well as applications of IOPs. Topics Lecture 2: Algorithms to invert IOPs Approach I will give focused lectures along with hands‐on practices. Advanced reading materials will be handed out to broaden knowledge.
Lecture 1. Hyperspectral remote sensing of optically shallow waters. I'll give an of overview of the problem, what information people need in shallow waters, and what differences there are with deep water.
Lecture 2. Atmospheric correction for shallow waters. I'll talk about why atmosphere correction for deep case 1 water doesn't work for shallow water, and what techniques are used for shallow waters.
Objectives This lecture is designed to provide an overview of the fundamentals of Inherent Optical Properties (IOPs), its relationship with AOPs, algorithms to invert IOPs from AOPs, as well as applications of IOPs. Topics Lecture 3: Applications of IOPs Approach I will give focused lectures along with hands‐on practices. Advanced reading materials will be handed out to broaden knowledge.
Lecture 3. Techniques for inverting spectra: I'll talk about semi‐analytic and spectrum matching techniques for retrieving bathymetry, bottom type, and water IOPs, with the emphasis on bathymetry and error analysis.
Lecture 4. Ecosystem modeling. I'll talk about improvements to ocean ecosystem models when more accurate light calculations are used.
Lecture 5. More Ecosystem modeling. I guess I'll finish up Lecture 4. HydroLight training. 5 lectures and labs. All sorts of things from an overview of the software, to demonstration runs, to students running H on their own computers
Lecture 1: Above‐ et In‐Water Radiometry (Methods et Calibration Requirements) In situ optical radiometric measurements have direct application in the development and assessment of: theoretical models describing extinction of light in seawater; and empirical algorithms linking the seawater apparent optical properties to the optically significant constituents expressed through their inherent optical properties or concentrations. In addition, in situ radiometric data are essential for the vicarious calibration of space sensors and the validation of remote sensing products. The most accurate input data is always the most desirable for any bio‐optical modeling and calibration or validation activity. However, accuracy requirements impact methodological and instrumental investment which should be weighed against the specific need for each application. The lecture, after a brief introduction to radiometric concepts and terminology, addresses the fundamentals of above‐ and in‐water field radiometry. This includes a review of instruments, measurement methods and data analysis. Additional elements addressed in the lecture include an overview of calibration requirements and methods for ocean color field radiometry. These latter include an introduction to absolute radiometric calibration of radiance and irradiance sensors, determination of the geometric response of cosine collectors, characterization of immersion factors for in‐water sensors. Finally, inter‐comparisons of radiometric products derived from different measuring systems and methods are used to discuss individual performances. Bibliography G. Zibordi and K.J. Voss. Field Radiometry and Ocean Color Remote Sensing. In Oceanography from Space, revisited. V. Barale, J.F.R. Gower and L. Alberotanza Eds., Springer, Dordrecht, pp. 365‐398, 2010.
Lecture 2: In Situ Radiometric Products (uncertainty analysis and applications) Data products from in–water radiometric measurements generally include spectral values of: irradiance reflectance, remote sensing reflectance, normalized water–leaving radiance, diffuse attenuation coefficient and the so called Q‐factor. Data products from above–water radiometric measurements are generally restricted to the normalized water–leaving radiance and the remote sensing reflectance. By restricting the analysis to the normalized water‐leaving radiance, the lecture addresses the various sources of uncertainties affecting in situ radiometric measurements (e.g., accuracy of absolute calibration, superstructure perturbations, changes in illumination conditions, wave effects and selfshading for in‐water methods only). Emphasis is placed in the evaluation of methods allowing for the minimization of the various perturbing effects and additionally in the quantification of contributions of these latter to uncertainty budgets. Further element considered in the lecture is the application of in situ radiometric data to the assessment of satellite primary products (i.e., the normalized water leaving radiance determined from top‐of‐atmosphere radiance corrected for the atmospheric perturbations). Focus is placed on the use of in situ data to evaluate differences in cross‐mission products (i.e., normalized water leaving radiance from SeaWiFS, MODIS‐A, MODIS‐T and MERIS), variations in space system performance with time and intra‐annual changes in accuracy. Bibliography G. Zibordi and K.J. Voss. Field Radiometry and Ocean Color Remote Sensing. In Oceanography from Space, revisited. V. Barale, J.F.R. Gower and L. Alberotanza Eds., Springer, Dordrecht, pp. 365‐398, 2010.
Lecture 1. In situ Measurements (1). Saturday, 0900‐1030 h. Starting from a base in the radiative transfer equation, the fundamental measurement of the underwater radiance field will be introduced. Historical observations will be traced through two recent developments. These measurements will be used to introduce (or reintroduce) various integrated radiometric quantities: the planar and scalar irradiances, the average cosines, and reflectances. Variations in the radiance distribution along a path (say in the vertical) brings together consideration of the fundamental inherent optical properties, the absorption coefficient and volume scattering function, as well as the beam and diffuse attenuation coefficients and scattering coefficients over various solid angles. Modern methods for the independent measurement of these quantities will be presented and discussed. Attempts at optical closure field experiments will be presented. Bibliography Ocean Optics Web Book. http://www.oceanopticsbook.info/view/introduction/overview
One of the main questions you will be asked as a remote sensing expert is: how reliable and good is information, which we derive from remotely sensed ocean colour data? Can we trust them? What is the error or uncertainty range of these data? In this section of the IOCCG training course, which consists of 3 lectures and exercises, we will look into this problem. Lectures The first lecture will be dedicated to the sources of uncertainties. We have to consider that our observations are the reflectivity in a number of spectral bands, which are measured at the top of atmosphere (TOA) or, in case of an aircraft platform, in a certain height above the water. We try nothing less than to isolate, retrieve and quantify a small effect on these spectra, which is caused by absorption and scattering of e.g. of phytoplankton, from a large number of other effects, of which in particular the atmosphere dominates the TOA spectrum. Problems of this kind may induce large uncertainties. In some cases it might be even impossible to retrieve reliable information of the ocean from remotely sensed reflectance spectra. Thus, one important area of ocean colour research is to analyze sources of uncertainties, to develop methods to quantify uncertainties and finally to find way to reduce uncertainties. In this lecture we will consider Natural factors, which determine uncertainties, and their variability Uncertainties, which are induced by reducing the manifold of factors to a few dominant wavelength (nm) Radiance (Wm‐2 sr‐1 μm‐1) air molecules different aerosols thin clouds Sky reflectance Sun glint foam floating material chlorophyll Suspended particles different phytoplankton species dissolved organic matter Vertical distribution Bottom reflection contrails Factors, which determine top of atmosphere reflection spectra, from which try to retrieve e.g. the chlorophyll concentration Errors caused by spaceborne or airborne instruments: calibration, ageing, noise Errors caused by in situ measurements, sampling and procedures Problem of comparing in situ with space borne In the second lecture we look into procedures, how to determine uncertainties: How to quantify uncertainties: scatter, bias, robustness, stability Validation procedures and strategies Testing of algorithms Round robin exercises Sensitivity studies Determination of uncertainties on a pixel by pixel bases flagging The third lecture will finally discuss the results of our exercises and will be dedicated to the question, how to reduce uncertainties. This is a wide field, where a lot of research is still needed, and it offers themes for your future work. Detection of spectra / pixels, which are out of scope of the algorithm Masking of clouds and cloud shadows Use of additional information Pre‐classification of water types and use of dedicated algorithms How to produce maps from satellite data, which include information about uncertainties.
Lecture 2. In situ Measurements (2). Monday, 1100‐1230 A brief review of the instrument concepts introduced in the first lecture will be followed by a discussion of some new and novel platforms for the measurement of underwater optics. Emphasis will be on autonomous measurement systems: moored platforms, autonomous underwater and surface vehicles, gliders, surface drifters, profiling floats, and animal tags. The lecture will finish with a discussion of two ancient optical measurement devices, and the surprising utility of these for solving modern problems in biological oceanography. Bibliography Ocean Optics Web Book. http://www.oceanopticsbook.info/view/introduction/overview
One of the main questions you will be asked as a remote sensing expert is: how reliable and good is information, which we derive from remotely sensed ocean colour data? Can we trust them? What is the error or uncertainty range of these data? In this section of the IOCCG training course, which consists of 3 lectures and exercises, we will look into this problem. Lectures The first lecture will be dedicated to the sources of uncertainties. We have to consider that our observations are the reflectivity in a number of spectral bands, which are measured at the top of atmosphere (TOA) or, in case of an aircraft platform, in a certain height above the water. We try nothing less than to isolate, retrieve and quantify a small effect on these spectra, which is caused by absorption and scattering of e.g. of phytoplankton, from a large number of other effects, of which in particular the atmosphere dominates the TOA spectrum. Problems of this kind may induce large uncertainties. In some cases it might be even impossible to retrieve reliable information of the ocean from remotely sensed reflectance spectra. Thus, one important area of ocean colour research is to analyze sources of uncertainties, to develop methods to quantify uncertainties and finally to find way to reduce uncertainties. In this lecture we will consider Natural factors, which determine uncertainties, and their variability Uncertainties, which are induced by reducing the manifold of factors to a few dominant wavelength (nm) Radiance (Wm‐2 sr‐1 μm‐1) air molecules different aerosols thin clouds Sky reflectance Sun glint foam floating material chlorophyll Suspended particles different phytoplankton species dissolved organic matter Vertical distribution Bottom reflection contrails Factors, which determine top of atmosphere reflection spectra, from which try to retrieve e.g. the chlorophyll concentration Errors caused by spaceborne or airborne instruments: calibration, ageing, noise Errors caused by in situ measurements, sampling and procedures Problem of comparing in situ with space borne In the second lecture we look into procedures, how to determine uncertainties: How to quantify uncertainties: scatter, bias, robustness, stability Validation procedures and strategies Testing of algorithms Round robin exercises Sensitivity studies Determination of uncertainties on a pixel by pixel bases flagging The third lecture will finally discuss the results of our exercises and will be dedicated to the question, how to reduce uncertainties. This is a wide field, where a lot of research is still needed, and it offers themes for your future work. Detection of spectra / pixels, which are out of scope of the algorithm Masking of clouds and cloud shadows Use of additional information Pre‐classification of water types and use of dedicated algorithms How to produce maps from satellite data, which include information about uncertainties.
Part 1 covers the nature of hyperspectral imaging and the history of the development including airborne systems AVIRIS, PHILLS and the new PRISM, and the spaceborne HICO instrument on the International Space Station (http://hico.coas.oregonstate.edu). Calibration and characterization of the sensors and on‐orbit calibration is also covered. Bibliography Corson, M. and C. O. Davis, 2011, “The Hyperspectral Imager for the Coastal Ocean (HICO) provides a new view of the Coastal Ocean from the International Space Station,” AGU EOS, V. 92(19): 161‐162. Davis, C. O., J. Bowles, R. A. leathers, D. Korwan, T. V. Downes, W. A. Snyder, W. J. Rhea, W. Chen, J. Fisher, W. P. Bissett and R. A. Reisse, 2002, Ocean PHILLS hyperspectral imager: design, characterization, and calibration, Optics Express, 10(4): 210‐221. Davis, C. O., K. L. Carder, B‐C Gao, Z. P Lee and W. P. Bissett, 2006, The Development of Imaging Spectrometry of the Coastal Ocean, IEEE Proceedings of the International Geoscience and Remote Sensing Symposium, V. 4: 1982‐1985. Gao, B‐C, M. J. Montes, C. O. Davis, and A. F.H. Goetz, 2009, Atmospheric correction algorithms for hyperspectral remote sensing data of land and ocean, Remote Sensing of Environment, doi:10.1016/j.rse.2007.12.015 Lee, Z‐P, B. Casey, R. Arnone, A. Weidemann1, R. Parsons, M. J. Montes, Bo‐Cai Gao, W. Goode, C. O. Davis, J. Dye, 2007, Water and bottom properties of a coastal environment derived from Hyperion data measured from the EO‐1 spacecraft platform, J. Appl. Remote Sensing, V. 1 (011502): 1‐16. R. L. Lucke, M. Corson, N. R. McGlothlin, S. D. Butcher, D. L. Wood, D. R. Korwan, R.‐R. Li, W. A. Snyder, C. O. Davis, and D. T. Chen, 2011, “The Hyperspectral Imager for the Coastal Ocean (HICO): Instrument Description and First Images,” Applied Optics, V. 50 (11): 1501‐1516 doi:10.1364/AO.50.001501 Mobley, C. D., L. K. Sundman, C. O. Davis, T. V. Downes, R. A. Leathers, M. J. Montes and J. H. Bowles, W. P. Bissett, D. D. R. Kohler, R. P. Reid, E. M. Louchard and A. Gleason, 2005, Interpretation of hyperspectral remote‐sensing imagery via spectrum matching and look‐up tables, Applied Optics, 44(17): 3576‐3592.
Part 2 covers the algorithms and processing of the data to produce ocean products including atmospheric correction, algorithms for both optically shallow (e.g. coral reefs) and optically deep (e.g. river plume) coastal environments. Example applications using airborne hyperspectral and HICO data are presented. Bibliography Corson, M. and C. O. Davis, 2011, “The Hyperspectral Imager for the Coastal Ocean (HICO) provides a new view of the Coastal Ocean from the International Space Station,” AGU EOS, V. 92(19): 161‐162. Davis, C. O., J. Bowles, R. A. leathers, D. Korwan, T. V. Downes, W. A. Snyder, W. J. Rhea, W. Chen, J. Fisher, W. P. Bissett and R. A. Reisse, 2002, Ocean PHILLS hyperspectral imager: design, characterization, and calibration, Optics Express, 10(4): 210‐221. Davis, C. O., K. L. Carder, B‐C Gao, Z. P Lee and W. P. Bissett, 2006, The Development of Imaging Spectrometry of the Coastal Ocean, IEEE Proceedings of the International Geoscience and Remote Sensing Symposium, V. 4: 1982‐1985. Gao, B‐C, M. J. Montes, C. O. Davis, and A. F.H. Goetz, 2009, Atmospheric correction algorithms for hyperspectral remote sensing data of land and ocean, Remote Sensing of Environment, doi:10.1016/j.rse.2007.12.015 Lee, Z‐P, B. Casey, R. Arnone, A. Weidemann1, R. Parsons, M. J. Montes, Bo‐Cai Gao, W. Goode, C. O. Davis, J. Dye, 2007, Water and bottom properties of a coastal environment derived from Hyperion data measured from the EO‐1 spacecraft platform, J. Appl. Remote Sensing, V. 1 (011502): 1‐16. R. L. Lucke, M. Corson, N. R. McGlothlin, S. D. Butcher, D. L. Wood, D. R. Korwan, R.‐R. Li, W. A. Snyder, C. O. Davis, and D. T. Chen, 2011, “The Hyperspectral Imager for the Coastal Ocean (HICO): Instrument Description and First Images,” Applied Optics, V. 50 (11): 1501‐1516 doi:10.1364/AO.50.001501 Mobley, C. D., L. K. Sundman, C. O. Davis, T. V. Downes, R. A. Leathers, M. J. Montes and J. H. Bowles, W. P. Bissett, D. D. R. Kohler, R. P. Reid, E. M. Louchard and A. Gleason, 2005, Interpretation of hyperspectral remote‐sensing imagery via spectrum matching and look‐up tables, Applied Optics, 44(17): 3576‐3592.
This lecture will provide an overview of atmospheric correction approaches for remote sensing of water properties for open oceans and coastal waters. Beginning with definitions of some basic parameters for describing ocean and atmosphere properties, the radiative transfer equation (RTE) for ocean‐atmosphere system will be introduced and discussed. Various methods for solving RTE, in particular, the successive‐order‐of‐scattering method will be described. We examine various radiance contribution terms in atmospheric correction, i.e., Rayleigh scattering radiance, aerosol radiance (including Rayleigh‐aerosol interaction), whitecap radiance, sun glint, and water‐leaving radiance. Atmospheric correction algorithms using the near‐infrared (NIR) and shortwave infrared (SWIR) bands will be described in detail, as well as some examples from MODIS‐Aqua measurements. The standard NIR atmospheric correction algorithm has been used for deriving accurate ocean color products over open oceans for various satellite ocean color sensors, e.g., OCTS, SeaWiFS, MODIS, MERIS, VIIRS, etc. Some specific issues of atmospheric correction algorithm over coastal and inland waters, e.g., highly turbid and complex waters, strongly absorbing aerosols, will also be discussed. The outline of the lectures is provided below. Outline of the Lectures Introduction Brief history Basic concept of ocean color measurements Why need atmospheric correction Radiometry and optical properties Basic radiometric quantities Apparent optical properties (AOPs) Inherent optical properties (IOPs) Optical properties of the atmosphere Molecular absorption and scattering Aerosol properties and models Non‐ and weakly absorbing aerosols Strongly absorbing aerosols (dust, smoke, etc.) Radiative Transfer Radiative Transfer Equation (RTE) Various approaches for solving RTE Successive‐order‐of‐scattering method Single‐scattering approximation Sea surface effects Atmospheric diffuse transmittance Normalized water‐leaving radiance Atmospheric Correction Define reflectance and examine the various terms Single‐scattering approximation Aerosol multiple‐scattering effects Open ocean cases: using NIR bands for atmospheric correction Coastal and inland waters Brief overviews of various approaches The SWIR‐based atmospheric correction Examples from MODIS‐Aqua measurements Addressing the strongly‐absorbing aerosol issue The issue of the strongly‐absorbing aerosols Some approaches for dealing with absorbing aerosols Examples of atmospheric correction for dust aerosols using MODIS‐Aqua and CALIPSO data Requirements for future ocean color satellite sensors Summary Bibliography Chandrasekhar, S. (1950), “Radiative Transfer,” Oxford University Press, Oxford, 393 pp. Van de Hulst, H. C. (1980), “Multiple Light Scattering,” Academic Press, New York, 739pp. Gordon, H. R. and A. Morel (1983), “Remote Assessment of Ocean Color for Interpretation of Satellite Visible Imagery: A Review,” Springer‐Verlag, New York, 114pp. Gordon, H. R. and M. Wang (1994), “Retrieval of water‐leaving radiance and aerosol optical thickness over the oceans with SeaWiFS: A preliminary algorithm,” Appl. Opt., 33, 443‐452. Gordon, H. R. (1997), “Atmospheric correction of ocean color imagery in the Earth Observing System era,” J. Geophys. Res., 102, 17081‐17106. Wang, M. (2007), “Remote sensing of the ocean contributions from ultraviolet to near‐infrared using the shortwave infrared bands: simulations,” Appl. Opt., 46, 1535‐1547. IOCCG (2010), “Atmospheric Correction for Remotely‐Sensed Ocean‐Color Products,” Wang, M. (ed.), Reports of International Ocean‐Color Coordinating Group, No. 10, IOCCG, Dartmouth, Canada. (http://www.ioccg.org/reports_ioccg.html)
Lecture 1: Model predictions of the response of ocean physical and biological processes to climate change I will summarize the main physical changes predicted by global warming models for the rest of this century and how we expect those changes to impact both lower trophic and upper trophic level processes in the ocean. I will discuss empirical as well as ecosystem model approaches for predicting the biological response and examine how the model simulations compare with estimates of chlorophyll and primary production based on ocean color observations. A major emphasis of the discussion will be on interannual as well as intra and inter‐model variability. Lecture 2: Detection of trends in ocean color data I will briefly review recent studies attempting to use ocean color products to detect climate trends then examine how variability such as that identified by the model simulations affects our ability to detect long‐term trends in the observations. Bibliography Sarmiento, J. et al. (2004), Response of ocean ecosystems to climate warming, Global Biogeochem. Cycles, 18(GB3003), doi:1029/2003GB002134. Steinacher, M., F. Joos, T. Frölicher, L. Bopp, P. Cadule, S. Doney, M. Gehlen, B. Schneider, and J. Segschneider (2010), Projected 21st century decrease in marine productivity: a multi‐model analysis, Biogeosciences, 7, 979–1005. !!!!!!!!Cours 2!!!!!!! Henson, S., J. Sarmiento, J. Dunne, L. Bopp, I. Lima, S. Doney, J. John, and C. Beaulieu (2010), Detection of anthropogenic climate change in satellite records of ocean chlorophyll and productivity, Biogeosciences, 7, 621–640.