Section IV: Environmental Applications

The use of chromatic techniques has been reported for monitoring biodigestion at a waste-recycling unit and airborne microparticle pollution in urban areas (Jones et ah, 2008). The chromatic approach has been extended for other environmental problems. These include an extension of microparticle monitoring for long free path applications, monitoring marine water samples, evaluating the use of electric power generation by large-scale windmills and assessing the environmental suitability of various gases used for the electrical insulation of high-voltage electric power equipment. These applications use different forms of chromaticity - optical absorption/scattering for long path particle monitoring and marine water assessment, time-based analysis of windmill power production and severity of air-quality reduction potential caused by different electrical insulating gases. Details of chromatic deployment for these various environment-related aspects are considered in this section.

Optical Chromatic Monitoring of Marine Waters

J. L. Kenny

Introduction

Measuring the inherent optical properties (IOPs) of sea water from real environments (Figure I0.l) is important for understanding the propagation of light through water to determine how much light is available for primary production (photosynthesis) and for optical communications with low-energy communications between underwater autonomous vehicles. Ocean colour remote sensing models and measurements rely on a knowledge of the optical backscatter from particulate matter in the water, which can also provide information about the particulate matter characteristics (e.g., bulk refractive index, particle size distribution and composition) (Korotaev and Baratange, 2004), which is essential to understanding the optical variability of natural waters.

As instrumentation for measuring the inherent optical properties of marine waters becomes more advanced, the data produced by the emerging instruments (e.g., Wetlabs AC-S) is becoming more complex. Where simple graphical representations may have been sufficient for reductionalist data (e.g., the backscattering coefficient at a single wavelength), this is not the case for complex data. Chromatic methodology offers a means for balancing the range of relevant information with simplicity of presentation that allows for easy assimilation of information by non-specialist personnel or autonomous systems. The usefulness of the chromatic approach has been demonstrated with real marine water tests from sites l, 2 and 3 of the river estuary and a fourth sample taken from the beach marked Hilbrie Island shown in Figure 10.1.

Inherent Optical Properties of Water

Natural water has two fundamental inherent optical properties

  • 1. The absorption coefficient (ac) defined by (absorbed incident power)/(optical path length)
  • 2. The volume scattering function (sf) defined by [(light intensity 0°-180°)/(incident irradiance/ volume)]

A total scattering coefficient is defined as the integral of Sffrom 0° to 180°, whilst a backscattering coefficient is given by the integral of sf from 90° to 180° (Mobley, 1994).

Example of river estuary from which marine waters were monitored (Hilbre Island. Dee Estuary. Merseyside, UK [facing south])

FIGURE 10.1 Example of river estuary from which marine waters were monitored (Hilbre Island. Dee Estuary. Merseyside, UK [facing south]).

The amount of absorption and scattering of light in natural waters is dependent not only upon the absorption and scattering coefficients of the water but also on the optical properties of material dissolved or suspended within the water. These include Raman scattering by water molecules, fluorescence by coloured dissolved organic matter (CDOM), absorption, scattering and fluorescence by organic material (phytoplankton and algae) and absorption and scattering by non-organic material (sand, detritus and sediments).

Chromatic Processing

Polychromatic optical signals from light transmission through complex marine waters may be addressed with chromatic techniques (Chapter 1). A preferred chromatic approach has been the use of the Lab method. The algorithms of the Lab method are defined by the following equations (Schwarz et al., 1987; Ainouz et ah, 2006):

for R/Rn, G/Gn. B/Bn > 0.008856

where Rn, Gn, Bn are processor outputs when addressing data to be used as a reference source.

A reference signal is used to calculate the chromatic values; lightness (L), A and B, where L is a representative signal strength, “a” represents the chromatic space from red (+a) to green (—b) and “b” represents the chromatic space from yellow (+b) to blue (—b).

Chromatic Monitoring System

An optical system for monitoring marine water (Figure 10.2) has been developed at the National Oceanography Centre in Liverpool, UK, as part of Oceans 2025, the Natural Environment Research Council (NERC)’s proposed strategic research programme. This uses a dual halogen/deuterium light source to provide broadband white light focused onto a linear variable bandpass optical filter

Schematic diagram of the optical monitoring system

FIGURE 10.2 Schematic diagram of the optical monitoring system.

to provide a narrow optical bandwidth of approximately 40 nin. As the filter is displaced, the centre optical wavelength is shifted along the spectral range of the source. The light then passes through an optical modulator operating at 200 Hz with a 90% duty cycle, which is then divided by a glass plate, angled at 45° from the normal (Figure 10.2). Part of the light is reflected onto an avalanche photo diode (APD) to provide a reference to compensate for variance in the source output. The remaining light is focused onto the face of a 1 mm diameter optical fibre through which the light is transmitted into a water sample before being transmitted via another optical fibre to a second avalanche photo diode.

The volume scattering function at a fixed angle was calculated by multiplying the sensor response (normalised against a reference signal representing the output from the light source) by the sensitivity coefficient and the attenuation compensation function.

The sensitivity coefficient was derived by moving a plaque of spectralon (with a known reflectivity factor) away from the face of the sensor. The attenuation compensation function corrected for the attenuation of the light along the transmission and return path through the sample water. It is dependent on spectral attenuation measurements of the sample water (performed in this case by a Wetlabs AC-S).

The backscattering coefficient was then calculated using the methods described by (Maffione and Dana, 1997; Dana and Maffione, 2002).

The spectral range of the Wetlabs AC-S limited the “corrected” range of the optical system between 400 and 750 nm.

Some test results were obtained for grab samples of marine waters with natural sediment taken from four sites in the River Dee Estuary (Figure 10.1). The samples were tested with a test tower which circulated the water for testing with different systems, including the chromatic system represented in Figure 10.2.

Examples of Test Results

An important monitoring parameter was the backscattering coefficient (Section Ю.2) (Mobley, 1994; Kenny, 2015).

The corrected backscattering coefficients for water samples from the four sites (l-3 and Hilbre beach) on the Dee estuary (Figure 10.1) are shown as a function of optical wavelengths in Figure Ю.З. This shows the complex variation of the marine water between these four sites. The highest overall backscattering occurred for the sample from the deeper channel (site 1), whilst the lowest was for the shale (site 3) on the island (Hilbre). The sample from the island with red shale (site 2) had a distinctive broad peak at about 600 nm wavelength.

This figure illustrates the complexity of the test results but also some clear differences between the four samples which are not easily quantified from the backscattering characteristics. However,

Corrected backscattering versus wavelength results using Sea Krait OBS monitoring system (Kenny, 2015)

FIGURE 10.3 Corrected backscattering versus wavelength results using Sea Krait OBS monitoring system (Kenny, 2015).

chromatic Lab addressing (Section 10.3) methods may be deployed to quantify the different marine water characteristics from each of the four sites.

Triangular RGB receptors were applied to the data between 400 and 750 nm (as shown in Figure 10.3).

The reference values Rn, Gn, Bn were obtained by applying triangular RGB receptors to the backscattering coefficient for pure water (as presented by Morel. 1974) between 400 and 750 nm.

where is the backscattering coefficient of pure water and A is the wavelength of the light.

Figure 10.4 shows the Lab plot calculated from the backscattering measurements shown in Figure 10.3. The results obtained for the four sites of Figure 10.1 are shown, from which the difference in the spectral signatures of the backscattering coefficient can be quantified in reference to the backscattering coefficient of pure water. This provides a convenient way for presenting the

Chromatic Lab map of backscattering coefficient from Sea Krait OBS (Kenny. 2015)

FIGURE 10.4 Chromatic Lab map of backscattering coefficient from Sea Krait OBS (Kenny. 2015).

information for an operator to observe and interpret. For example, the Lab graph shows that the sediment from site 2 scatters more blue light than the sediment from the other sites and that the sediment from site 1 scatters more light across the measured spectrum at the concentrations used than the sediment from the other sites.

Summary and Overview

The chromatic Lab diagram provides a useful categorisation means for various marine water samples. The approach could be used by either an automated system or as an input for a machine learning system. Further w'ork could also provide additional information for marine water studies.

Acknowledgements

Special thanks to Prof. Joe Spencer (University of Liverpool) for all his advice, to Professor Alex Souza (National Oceanography Centre) and to Dr Mike Smithson (National Oceanography Centre) for all their support.

References

Ainouz. S.. Zallet, J.. de Martini. A., and Collet, C. (2006). Physical interpretation of polarisation-encoded images by clour preview. Opt. Express. Vol. 14. No. 13. pp. 5916-5927.

Dana, D.R. and Maffione, R.A. (2002). Determining the back scattering coefficient with fixed-angle backscattering sensors - revisited. Ocean Optics XVI. Santa Fe New Mexico.

Kenny, J.L. (2015). Optical properties of marine waters - A method based on chromaticity. PhD Thesis, University of Liverpool. Liverpool. UK

Korotaev, G. and Baratange, F. (2004). Particulate backscattering ratio at LEO 15 and its use to study particle composition and distribution. Journal of Geophysical Research. Vol. 109. p. C01014.

Maffione, R.A. and Dana. D.R. (1997). Instruments and methods for measuring the back-scattering coefficient of ocean waters. Applied Optics, Vol. 36, No. 24, p. 6057.

Morel, A. (1974). Optical properties of pure water and pure sea water. Optical Aspects of Oceanography. Edited by N.G. Jerlov and E. Steemann Nielsen, pp. 1-24. Academic Press, San Diego. Calif.

Mobley, C.D. (1994). Light and Water: Radiative Transfer in Natural Waters. Academic Press, San Diego, Calif.

Schwarz. M.W., Cowan, W.B., and Beatty. J.C. (1987). An experimental comparison of RGB. Y1Q. LAB, HSV, and opponent color models. ACM Transactions on Graphics, Vol. 6. No. 2. pp. 123-158.

 
Source
< Prev   CONTENTS   Source   Next >