Dredging to create and maintain navigable shipping channels and allow safe ship access is a necessary component of most ports and coastal infrastructure developments. Dredging activities generate suspended sediment that could impact upon nearby marine communities i.e. it is a potential hazard. Well recognised cause-effect pathways include suspended sediment interfering with filtering and feeding mechanisms, increased turbidity (water cloudiness) changing light quantity and quality and increased sediment deposition causing smothering. How the hazard translates into risk was investigated in this study for dredging in inshore coastal areas near reefs in Cleveland Bay (inshore central, Great Barrier Reef), where turbidity regimes and light levels are very different from what is considered a ‘typical’ reef setting. Data analysed include (1) a time series of benthic light and turbidity levels at five sites supplied by the Port of Townsville Limited, (2) a time series of multispectral irradiance and turbidity on a fringing reef and (3) a study involving 90 vertical turbidity and multispectral light depth profiles collected mostly behind a working trailing suction hopper dredge. This information was used to re-create environmentally realistic exposure scenarios in an advanced aquarium system at the Australian Institute of Marine Science and the physiological responses of corals and sponges examined over an extended, 28 d period. From the time series data, the 10-minute turbidity and light readings were reduced to daily mean values (producing daily light integrals (DLI) in units of mol quanta m2 for light) and percentile (P) values from P0-P100 calculated for running mean periods from 1–42 d. This largely encompasses the length of a typical maintenance dredging program. A characteristic feature of the data was that it was highly skewed, indicating water quality was very good for most of the year (hence supporting reefs), but sometimes subjected to multiple short-term periods of poor water quality resulting in a divergence of mean and median values. Overall upper percentiles of turbidity (P95 etc) and lower percentiles of light (P5) were the best descriptors of the data, showing a very clear gradient across the Bay. The water quality time series included six dredging campaigns and at some of the bays dredging may increase the turbidity by 0.6-0.7 times the mean expected values, but these are between two and five times lower than the effects of natural events caused by wind or waves. The deployment of the multispectral light logger occurred over a brief (7 d) natural turbidity event and also periods of low turbidity, both of which were interspersed with cloudy and cloud- free days. During cloudy and low turbidity days, the benthic light levels were reduced without changing the underwater light spectrum. During turbid days benthic light levels were reduced and the spectrum changed, with relatively greater loss of more photosynthetically usable blue light. A simple ratio of blue (455 nm) to green (555 nm) wavelengths could identify these different periods of light reduction (i.e. cloud versus turbidity). The vertical turbidity profiling identified complex 3-dimensional profiles of the plumes showing surface, mid-water and bottom maxima, as well as well mixed homogenous SSC profiles. Overall, the most common profile was as an increase in SSC with depth, with measurements at the surface 3.5  less than the seabed behind the dredge and 10  lower at the dredge material placement area. Vertical light profiles in low turbidity ‘blue water’ outside the Bay showed the well-known exponential decrease in light quantity with depth, the rapid attenuation Risk assessing dredging activities 1 Jones, et al of red light in the first few metres, and furthest penetration of green and blue light. In low turbidity water inside the Bay, and beside a fringing reef, there was a similar loss of red light but also much more pronounced attenuation of the blue wavelengths with depth, shifting the spectral profile to green. This pattern is consistent with the attenuation caused by chromophoric (coloured) dissolved organic matter (CDOM). Under elevated SSCs behind the dredge, both the quantity of light and the spectral profile shifted strongly to the green-yellow (550–600 nm) range, with a maximum peak at 575 nm. The spectral shift could be due to absorption by the suspended sediment particles, but it is also likely to be due to the increased scattering of light by the suspended sediments which increases the probability of being absorbed by CDOM. The spectral shift is significant as it means a loss of the quality as well as quantity of light. This spectral shift needed to be replicated as close as practicable in laboratory experiments in order to properly evaluate pressure-response relationships. Using the wavelength specific light attenuation coefficients for downwelling light under different turbidity levels and using a turbidity to SSC conversion factor from samples collected behind the dredge, an empirical spectral solar irradiance model was constructed for Cleveland Bay. This model was used to produce nomograms for light quantity and colour spectrum at different depths, in different SSCs and in its full form the model could also accommodate different solar zenith angles from the solar declination cycle. The model outputs were used in conjunction with the analysis of the long-term water quality to define likely pressure-fields generated by dredging. The conditions (SSC, light quality and quantity) were then replicated in the AIMS SeaSimulator in a fully automated, computer-controlled dosing system with custom made light emitting diode (LED) lights that could replicate the spectral shifts. Sublethal responses of 3 adult corals (Acropora millepora, Pocillopora verrucose, Montipora aequituberculata) juvenile corals (A. millepora) and an encrusting sponge species, Cliona orientalis were then examined over a one month exposure period to 5 treatments levels: SSCs ranging from 2.3 to 15.7 mg L-1 of Cleveland Bay sediment (with a modal size of 25 μm) and predominantly green-yellow light (peaking at 550 nm) of 5.7–0.06 mol quanta m-2 d-1. All corals and sponges survived the exposures with no whole colony or partial mortality, but clear physiological responses were measured including changes in pigmentation, lipid concentrations, the ratio of structural to storage lipids and density of symbiotic dinoflagellates. All coral species showed changes in lipid ratios at 2.2 mol quanta m-2 d-1 (and 0.85 mol quanta m-2 d-1 for the sponge) consistent with mobilizing lipid reserves under sub-optimal light. The branching pocilloporid Pocillopora verrucosa was the most sensitive, showing bleaching (the dissociation of symbiosis) at <1 mol quanta m-2 d-1(for 30 d), which is a more consequential physiological response. When these physiological responses were mapped back onto the light data around Cleveland Bay the analyses showed corals and sponges occasionally naturally experience light limitation even in shallow (<5 m) depths. This is consistent with known depth distributions and zonation patterns in turbid zone reef communities i.e. that they are shallow- water mesophotic reef systems. The study included the first empirical measurements of elevated sediment accumulation rates caused by maintenance dredging using newly re-designed deposition sensors. Accumulation rates were highly elevated at a distance of a few hundred metres from a working Trailing Suction Hopper Dredge (TSHD) but there was a strong gradient of decreasing accumulation rates with increasing distance and no effects detectable after a few hundred metres. Given the novelty of the instrumentation, the results are preliminary, but provide evidence to support the 2 idea that (1) high SSCs produced by dredging in a low energy water column is conducive to rapid settling and enhanced deposition (2) the effects are quite localized. The light-based monitoring conducted here with the multispectral PAR sensors, supported by the hyperspectral vertical profiling offers many more advantages for inshore water quality monitoring than turbidity measurements. The report concludes with management implications section and suggestions for how to use the results in monitoring and in risk-response reactive management cascades to guide capital and maintenance dredging in inshore coral reef communities.

Depth range
3- 21 m

Mesophotic “mentions”
8 x (total of 30340 words)

Classification
* Presents original data
* Focused on 'mesophotic' depth range

Fields
Ecology
Oceanography
Physiology

Focusgroups
Porifera (Sponges)
Scleractinia (Hard Corals)

Locations
Australia - Great Barrier Reef

Platforms
Aquarium-based
In-situ instrumentation