The influence of schlieren on in situ optical measurements used for particle characterization

Abstract

In pycnoclines, the density differences can cause light scattering -schlieren -even though only few particulate scatterers may be present. This may pose problems for the interpretation of results obtained with instruments relying on light scattering and transmission, for example the LISST (Laser In Situ Scattering and Transmissometry) particle sizer, and various cameras. Here, the influence of schlieren on in situ forward light scattering, beam attenuation and image analysis is evaluated using a LISST-100 and a digital floc camera. Automated image analysis routines detect schlieren as particles, causing an apparent increase in particle size and volume. Re-analysis omitting schlieren-affected parts of the images reveals no increase. LISST beam attenuation and Volume Scattering Function (VSF) measurements indicate that schlieren can cause increases in beam attenuation due to a marked increase in the VSF at angles smaller than ~1.5°-2°, and falsely indicate accumulation of suspended particles in the pycnocline. Light scattering caused by density differences can also cause multiple scattering, which produces an apparent decrease in particle size derived from the LISST. Schlieren is visible in images when the buoyancy frequency exceeds ~0.12 s -1

How schlieren affects beam transmissometers and LISST-Deep: an example from the stratified Danube River delta, NW Black Sea

“…The discrepancy observed here between our results and those of Mikkelsen et al (2008) could be related to the profiling package itself; in a stratified environment there are three plausible factors that will influence the degree to which schlieren will affect the measurements: (i) Richardson gradient (i.e., velocity differences across the pycnoline, which by itself would cause turbulence); (ii) profiling speed -fast profiling would cause more turbulence than slow profiling; and (iii) profiling package design -a bulky package would cause more turbulence than a sleek and slim package. It could be that the singlemost important factor in creating schlieren in stratified environments is the profiling speed and/or the design of the profiling package, i.e.…”

“…how much it disturbs the water (thus causing turbulence and schlieren) as it is descending. It is evident that Mikkelsen et al (2008) did not observe schlieren at buoyancy frequencies below 0.025 s . Conceivably, one could minimize the effect of schlie- …”

“…very high particle volume concentrations and large particle diameters, were observed at all stations where a pycnocline was developed, attributed to artificial particles due to schlieren. Mikkelsen et al (2008) found maximum values of N in the range 0.22-0.24 s -1 in the strong pycnocline of the Hudson River that coincided with large variability of both temperature and salinity. The same authors, found N to be an order of magnitude lower (~0.03 s -1 ) in Cardigan Bay in the Irish Sea, UK, and concluded that buoyancy frequencies exceeding 0.025 s -1 could be considered as the threshold for schlieren influence.…”

“…6). Superimposing D 50 values on the plot, we observe that values of N> 0.025 s -1 (Mikkelsen et al, 2008) correspond to large particle median diameters in the range 100-250 μm, which we atμm, which we atm, which we attribute to schlieren. Those high values correspond to almost 20% of the particle-size data collected during the cruise.…”

“…Small changes of the refractive index n should exist in parallel with turbulence at that depth to develop buoyancy frequency variations and consequently to trigger schlieren. According to Mikkelsen et al (2008), the schlieren effect is a combination of the density gradient, the shear in the pycnocline, and the profiling speed of the instruments. Station SGT2 shows high c 670 (21.9 m -1 ) when compared to Stn.…”