Reinterpreting turbidity: new methodologies for suspended-sediment research

Existing instruments for turbidity measurement vary considerably in terms of the principles of
operation, the physical design, and the cost to the researcher. The operational methodologies of
late twentieth century turbidity instruments have led to the development of new turbidity
measurement standards, and the invention of new turbidity measurement units. These
measurement units are invalid and do not have a sound footing with regard to the underlying
physics of the scattering and absorption of light by suspended particles. A review of the turbidity
literature has shown that the proliferation of these incommensurate units of measurement
throughout the physical sciences has caused extensive misinterpretation of turbidity data,
particularly concerning its use as a surrogate for suspended sediment concentration (SSC).
Turbidity is a complex phenomenon, and its measurement reported in terms of a single numeric
quantity in some physically indeterminate units of measurement. It is not necessarily useful to
reduce complex data to a single value, since this approach does not permit the researcher any
a posteriori opportunity to reinterpret existing data in light of innovations in analysis
methodology. This thesis proposes a new way to present turbidity data that will facilitate the
cross-comparison of turbidity measurements made by different instruments on any type of
suspended sediment. The creation of a new turbidity research instrument that illustrates the
application of the new method for reporting turbidity data as a ratio of light attenuation values in
decibels, promotes a positive change in direction away from the traditional measurement units.
The design process focusses on the instrument calibration procedure. With a simple
reinterpretation of the phase-function description of light scattering from suspended particles,
measurements of light attenuation made at multiple angles with respect to the axis of the incident
light beam, compare easily with the same measurements made using different wavelengths of
incident light. This work goes on to introduce new nomenclature that requires the citing of
measurement angle and wavelength of light to be an integral part of any recorded turbidity
measurement.
A modelling approach is used in the evaluation of the new turbidity instrument. This modelling is
important for three reasons. First, it identifies which instrument parameters affect the result of a
turbidity measurement – the key parameter being the mathematical function that describes the
spatial divergence of the incident light beam, which is important to measurement systems that
employ incoherent light sources such as LEDs, rather than to laser-based systems. Secondly, the
modelling reveals two fundamental theories of light scattering due to suspended particles, both
of which are required to describe adequately the turbidity of sediment-laden water. These two
theories are Mie scattering and geometric optics. Mie scattering is well accounted for by the
developed model - geometric optics, less so. The extent to which the model predictions diverge
from the empirical data is characterised by a metric related to the backscatter fraction, and the
consistency and linearity of the model is established. Ideas for the improvement of the geometric
optics feature of the model are discussed, as is the third reason for the importance of numerical
modelling. This third reason relates to the use of multi-parameter turbidity measurements as a
means to characterise the properties of a suspended sediment. By simulating precisely the
measurement response of the turbidity instrument, then it is notionally possible to infer the
properties of an unknown sediment by tuning the model parameters to match the empirical
response of the unknown sediment. This tuning process could reveal information pertaining to
particle size and shape.
Finally, potential applications for the new research instrument focus on improvements to the
instrument itself and the methodology, and the further development of the turbidity data
reporting nomenclature. Prototype methodologies that relate turbidity to suspended sediment
concentration are suggested, which also consider ways in which the optical measurements can
potentially classify the physical properties of a sediment.