Measurement of the abundance and optical significance of sub-micron sized particles in the ocean

Visible-band satellite images have been a valuable tool in marine science for 30 years. Through them we have learned a great deal about the distribution and seasonal variation of phytoplankton in the ocean. It has been possible to estimate primary production and the role the oceans play in taking up carbon dioxide from the atmosphere. In shelf seas, visible band images have enabled us to map out the concentration of small inorganic sediment particles stirred up from the sea bed. In both of these applications the particles are visible because they absorb and scatter sunlight. An important question, to which the answer is not at all clear at present, is exactly how large the particles are that are mainly responsible for scattering the light that is 'seen' by the satellites. It is often assumed that the size distribution of particles in the ocean follows a 'Junge' distribution, in which the number of particles increases rapidly as the size of the particles decrease. With this assumption, and using an optical theory which assumes spherical particles, it has been shown that most of the light scattering is performed by particles smaller than 1 micron in diameter. If this were true, it would mean that the particles seen in satellite imagery are mostly very small with slow settling speeds and long residence times in the surface of the ocean. This would have important implications for those who interpret these images and who use them to verify numerical models of particles in the ocean. However, no-one has ever directly observed in the sea the numbers of such small particles predicted by the Junge distribution. Photographs of undisturbed samples of seawater show that particles tend to gather together in 'flocs'. The measurements of particle size distribution which support the Junge distribution use a disruptive technique which potentially breaks up flocs and hence possibly over-estimates the number of small particles. Current instruments designed to measure the size of particles in situ and without disturbance are limited to particles greater than a few microns in size and hence greater than the critical particle size thought to be important in remote sensing. Holgraphic cameras enable focused images of small particles suspended in water to be made. The camera images the diffraction pattern of the particle and the particle is then reconstructed mathematically from this pattern. In the case of small particles, the diffraction pattern is much larger than the particle itself and so the holographic technique can reconstruct very small particles indeed, smaller than the wavelength of light, which cannot be measured in any other way. We have demonstrated this technique in the laboratory and imaged particles down to about 0.5 micron. With further magnification and improved optics it will be possible to image particles down to 0.2 micron. In this proposal we will package this technology for field work. By using different magnifications and commercially available in situ particle sizing instruments, we can make a package of instruments for measuring the undisturbed particle size distributions from 0.2 micron to 1 mm. This package will first be used in a turbulence tank to 'film' the flocculation process. The insight this gives will be used to construct new theoretical models of the particle size distribution. Field work will be carried out at one coastal site over a seasonal cycle and at sea through a variety of water types before and after the spring bloom. We will also make improved measurements of absorption and scattering by particles. Because the camera also measures the shape of the particles, differences between observed and calculated optical properties can be compared, for the first time, to particle shape. Finally, we will put together the complete data set to determine what size particles, under what conditions, are primarily responsible for the signals seen in visible band satellite images of the oceans.