Turbidity and light attenuation in coastal waters of the Great Barrier Reef

Investigations were made into light attenuation and turbidity in coastal waters of the Great Barrier Reef (GBR). Turbidity and suspended sediment concentrations (SSC) cause increased light attenuation throughout the water column and can influence the location, nature and health of coral reefs and other marine biota. As a result, turbidity and light form an important part of marine monitoring and water quality plans. In the first part of this research (Chapter 2), typical inshore turbidity regimes were characterised using exceedance curves and derivatives, for 61 sites, representing the data-set with the largest spatial extent of any previously used for the GBR. Prior to this study, little published work documented 'typical' ranges of turbidity for reefs within coastal waters. The highest median turbidity (at 50% exceedance (T₅₀)) was 15.3 NTU and at 90% exceedance (T₉₀) was 4.1 NTU. The GBRMPA guideline for mean annual concentration of total suspended solids (SSC) for open coastal waters is 2.0 mg l⁻¹. However, comparisons between mean and median turbidity showed large differences (up to a factor of 3), consistent with a strongly skewed temporal turbidity distribution. Exceedance results indicated strong spatial and temporal variability in water turbidity across inter/intraregional scales. Characterisation of turbidity regimes should contribute to the understanding of turbidity and SSC in the context of environmental management of coastal reefs. The second part of this research explored the dominant (Chapter 3) and secondary (Chapter 4) drivers of turbidity. Wave-induced shear stress is a known dominant driver of inshore reef turbidity and resuspension of sediment occurs when critical bed shear-stress is reached. Wave-induced shear-stress was calculated from nephelometer-obtained pressure measurements, near the seabed in Cleveland Bay. Comparisons with concurrent turbidity measurements indicated the critical stress values required to induce sediment resuspension. A critical stress value of ~1 N/m² was found to be sufficient to produce turbidities in excess of 20 mg l⁻¹ within Cleveland Bay. This was the first time shear stress had been calculated and related back to turbidity, using such instrumentation. The result is now being applied to water quality monitoring for consultancy, within the Marine Geophysics Laboratory. Using exceedance data, potential secondary turbidity drivers: water depth, distance to shore and distance to river were also investigated. Multiple linear regression and stepwise quadratic/interaction regression analyses were implemented. No significant relationship was found between any of the potential secondary drivers and turbidity at 10, 50 and 90% exceedance levels. Results indicate that at these sites, the effect of rivers was too small to be measurable. The final part of this research investigated the greatly overlooked relationship between turbidity and light (Chapter 5). Vertical light and turbidity (T) profiles were obtained and linked for the first time, at inshore GBR locations. Attenuation coefficients (k(d)) were calculated over water-column intervals, producing linear relationships between k(d) and turbidity (R²=0.91). Site-specific, average diffuse attenuation coefficients are presented (K(d)^AVG=0.43 m⁻¹ ) and deconstructed into their clear-water (K(d)^cl=0.3 m⁻¹) and turbidity-based attenuation (α=0.076 m⁻¹NTU⁄) components. A site-specific model predicting depth-averaged turbidity (T(pred)) using light was implemented, producing a new method of measuring inshore, depth-averaged turbidity. Model results correlated well to measured turbidity (T(avg)); T(pred) = 1.0(T(avg)) and R²=0.78 (Cleveland Bay), and T(pred) = 0.77(T(avg)) and R²=0.68 (Tully coast). The euphotic depth of Cleveland Bay was found to be 10 m for a depth-averaged turbidity of 2.5 NTU. Turbidity data is generally obtained near the seabed, due to the difficulty obtaining long-term depth profiles. However, strong linear relationships between depth-averaged turbidity and seabed turbidity were discovered. Depth-averaged turbidity was between 0.3-0.4 times seabed turbidity at all sites. Importantly, this finding may be extrapolated and used to infer depth-averaged values for all other (near seabed) data in the GBR. Finally, as an extension to the PAR light attenuation study, spectral attenuation coefficients were compared for inshore-offshore and shallow/deep waters of Cleveland Bay (Chapter 6). A hyperspectral radiometer was deployed to obtain exploratory light profiles. This enabled the behaviour of light across a spectrum of 137 individual wavelengths (from 300 ~ 800nm) to be measured. Results depict the variation of light attenuation coefficients across individual wavelengths. Spectral attenuation coefficients were compared to the PAR clear water attenuation component for Cleveland Bay. In Chapter 5, this component was calculated to be K(d)^cl=0.3 m⁻¹ and this was validated across the colour spectrum at offshore sites. This work has contributed to a broader understanding of light and turbidity in waters of the coastal Great Barrier Reef