Environmental Engineering Reference
In-Depth Information
1.2.2.3 Summary: laser diffraction as a
suspended-sediment surrogate technology
samplers. The cost for a complete unit without envi-
ronmental packaging is similar to that for a fully
equipped turbidimeter. The instrument-measurement
realm of a digital-optic measurement is a point. Like
the LISST technology, routine instrument calibra-
tions are unnecessary.
The principal components of the system are up
to three charged-coupled-device progressive scan
cameras (each with a selected lens) and a multi-port
fl ow-through cell. Each lens is affi xed to the fl ow-
through cell using extension tubes, keeping a precise
optical alignment between the cameras, lenses, tar-
geted area, and backlighting (Fig. 1.12a). All com-
ponents other than the fl ow-through cell, for which
a patent is pending, and extension tubes are com-
mercially available.
The key component of the system, and the only
part developed explicitly for this application, is the
multi-port fl ow-through cell (Fig. 1.12b). The fl ow-
through cell serves two purposes: to separate parti-
cles into fractions smaller and larger than 75
A major advantage of the LISST technology is real-
time measurement of PSD in 32 ¼-phi-diameter size
classes, a capability shared by no other currently
available sediment-surrogate monitoring instrument.
LISST instruments do not require instrument calibra-
tion when used for PSD or volume SSC.
Nevertheless, the technology has some limitations.
The measurement is a point sample. In addition, SSC
measurements are in volume units, thus requiring
estimates or measurements of sediment density to
convert to mass SSC units. When deployed
in situ
,
the LISST is susceptible to biofouling unless anti-
fouling shutters are used. Reductions in data accu-
racy due to the presence of non-spherical particles
and loss of data from signal saturation can occur.
Finally, the cost of a LISST instrument is two to six
times that of a fully equipped
in situ
turbidimeter.
However, for applications that require long-term
repetitions of at-a-point or spatially dense measure-
ments, especially if PSD data are required, the LISST
suite of instruments may represent the most cost-
effective approach for suspended-sediment data
acquisition.
m,
thus enabling a relatively unobstructed analysis of
the smaller particles; and to disjoin and isolate par-
ticles to create a more robust digital image of each
particle. If imaged particles are separated, or can be
digitally separated, they easily can be identifi ed,
measured, and counted by the software.
Computing SSC is based on four attributes derived
from the images: particle population, particle shape,
grayscale relation to turbidity, and the amount of
light passing through the entire image. The amount
of light (average image brightness) and average
image grayscale are measured over a sequence of
several images from the fl ow-through cell taken
within 2-6 seconds. The net changes for brightness
and grayscale are relative to a reference image using
clear water contrast against the sample images.
Particle volumes are estimated by calculating a “
z
”
axis length (the third unmeasured axis in the two-
dimensional image) based on the particle shape,
texture, chord length, and the particle center of
gravity from the two-dimensional image.
A multi-camera confi guration measures PSDs in
the range of 4-4000
μ
1.2.3 Digital Optical Imaging
Daniel J. Gooding
1.2.3.1 Background and theory
A digital optic-image analysis and pattern recogni-
tion system that does not require routine calibration
has been developed and is being adapted to quantify-
ing SSCs and selected size and shape characteristics
of suspended sediment in water samples. The tech-
nology, commercially promoted by the medical
industry in the 1990s to quantify cells in a blood
sample, computes size statistics based on automated
measurements of individual particles. Volumetric
SSC is inferred from the size statistics.
The technology, in development and testing at the
USGS Cascades Volcano Observatory, Vancouver,
Washington, USA, was conceptualized for applica-
tion in the laboratory. However, a fi eld version is
planned for testing as part of a stream-side pumping
system. The technology may eventually be adapted
for use in manually deployed isokinetic sediment
m. This three-order-of-
magnitude range cannot be accomplished using a
single magnifi cation, hence the use of multiple
cameras and lenses is required. The software is
designed to integrate images from up to three cameras
depending on the particle-size range required by the
μ
