Brian King

Drag force due to vegetation in mangrove swamps

Field studies of tidal flows in largely pristine mangrove swamps suggestthat the momentum equation simplifies to a balance between the water surfaceslope and the drag force. The controlling parameter is the vegetation lengthscale LE, which is a function of the projected area ofmangrove vegetation and the volume of the vegetation. The value ofLE varies greatly with mangrove species and water depth. It isfound that the drag coefficient is related to the Reynolds number Re definedusing LE. The drag coefficient decreases with increasingvalues of Re from a maximum value of 10 at low value of Re (<104), and converges towards 0.4 for Re < 5 ×104.

DYNAMICS OF THE TURBIDITY MAXIMUM IN THE FLY RIVER ESTUARY, PAPUA-NEW-GUINEA

Field studies in the Fly River estuary, Papua New Guinea, show that the turbidity maximum exists only at spring tides. The wind is important in wave-driven fluidization of the bed. The erosion rate varies with the sixth power of the water velocity. The suspended sediment settling velocity varies nonlinearly with the concentration. At least three-quarters of the river sediment inflow appears to be trapped in the estuary. A numerical hydrodynamics-sediment transport model is able to reproduce a number of the key features of the turbidity maximu, and suggests that the turbidity maximum is due to the simultaneous influence of the baroclinic circulation and the tidal pumping.

Trapping and dispersion of coral eggs around Bowden Reef, Great Barrier Reef, following mass coral spawning

Bowden Reef is a 5 km long kidney-shaped coral reef with a lagoon, located on the mid-shelf of the central region of the Great Barrier Reef. Field studies were carried out, in November 1986, at the time of mass coral spawning, of the water circulation around Bowden Reef and in the surrounding inter-reefal waters. The near-reef water circulation was strongly three-dimensional although the stratification was weak. In calm weather, coral eggs were aggregated in slicks along topographically controlled fronts. In the absence of a longshore current, water and coral eggs were trapped in the lagoon and in a boundary layer around Bowden Reef, by tidally driven recirculating motions. In the presence of a longshore current, some trapping occurred in the lagoon, but the bulk of the coral eggs was advected away from Bowden Reef and reached downstream reefs in a few days. This implies a likelihood of both self-seeding of reefs, and connectivity between reefs.

DYNAMICS, FLUSHING AND TRAPPING IN HINCHINBROOK CHANNEL, A GIANT MANGROVE SWAMP, AUSTRALIA

Hinchinbrook Channel is a 44-km long tidal channel connected at its northern and southern ends to the continental shelf of the Great Barrier Reef. The channel drains 164 km 2 of mangrove swamps. It also drains the Herbert River. The dynamics within the channel and the flushing of the fringing mangrove swamps were studied both in the field and using numerical models. The dominant currents are tidal semi-diurnal and are due to two nearly identical tidal waves entering the channel at its norhtern and southern openings and meeting in the middle. The mangrove swamps contribute significantly to the tidal dynamics. When the Herbert River discharge is small, net residual currents are negligible even in the presence of wind and the residence time of water is about 2 months. Flushing of the fringing mangrove swamps is largely controlled by the trapping effect in the mangrove swamps, and is slow with an e -fold residence time of about 54 days. On the northern end, the mangrove-fringed channel is wide and shallow, and a barotropic, coastal boundary layer is formed, due to shallow water effects and trapping in mangrove swamps. This results in a long-term (2 weeks) trapping of water along the mangrove-fringed coast in calm weather. Thus, coastal waters are continuously exchanged and mixed with mangrove swamp waters, but mixing between coastal and offshore waters is much slower.

Water circulation in the Gulf of Papua

Abstract The Gulf of Papua has the shape of a half-moon of radius of about 200 km and mean depth 1 m s−1. The dominant M2 tide propagates from the Coral Sea through the Gulf to enter both Torres Strait and the large estuaries of Papua New Guinea. The low-frequency currents have, in costal waters, little vertical shear associated with the salinity stratification, but, at the shelf break, a strong vertical shear in the well-mixed layer typically 100 m thick. A dominant forcing of the circulation in the Gulf is the eastward-flowing Coral Sea Coastal Current in the Northwest Coral Sea. This current appears to generate a counter-clockwise rotating eddy in the Gulf. The wind fluctuations result in the brackish water leaving the Gulf alternatively at its western and eastern sides. The residence time of river runoff in the Gulf, estimated using a three-dimensional hydrodynamic model, is about 2 months and this estimate agrees with that from freshwater budget estimates. Brackish water intrudes in the Torres Strait where tidal mixing maintains vertical homogeneity. The tidal mixing front is located near the northern tip of the Warrior Reefs and the intrusion is strongest in the monsoon season.

Dispersal of coral larvae from a lagoonal reef—II. Comparisons between model predictions and observed concentrations

Abstract A comparison was made between observed larval concentrations around a coral reef and predicted concentrations derived from hydrodynamic and dispersion models. There was virtually no correlation between the two data sets. The most obvious failure was the inability of the model to generate the extreme patchiness which was observed in larval distributions. This patchiness is caused by the fine-scale three-dimensional flows which occur around topographically complex structures such as coral reefs. Although the model did predict the development of larval patches, they were not as concentrated and showed a different spatial and temporal pattern compared to the field data. Time-averaged model predictions were also compared to coral recruitment patterns, but these too showed no correlation with the observed patterns. This study suggests that fully three-dimensional models with a small grid size are required in order to predict the complex flow patterns and patchy particle distributions around reefs. It is stressed that any models which are to be used to make predictions about the behaviour of reef ecosystems must be validated with detailed quantitative field data. Such data should be collected at temporal and spatial scales that are relevant to the predictions being made.

Inorganic sediment budget in the mangrove-fringed Fly River Delta, Papua New Guinea

Six oceanographic moorings were maintained for 8 weeks across the mouth of the mangrove-fringed Fly River estuary from April to June 1995 in the southeast trade wind season. A further 4 moorings were deployed for 8 weeks along the estuary channel in 1992, also in the southeast trade wind season. These data were used to estimate net exchange of suspended sediment between the estuary and the Gulf of Papua. A net inflow of fine sediment into the estuary from the coastal ocean was found to be considerable, about 40 tonnes s-1 or about 10 times the riverine inflow rate, resulting in a calculated, spatially averaged vertical accretion rate of 2 mm year-1. Mangroves may account for trapping 6% of the riverine sediment inflow or about 1/4 of the riverine clay inflow. If this sediment was distributed only over the observed accumulation zones near islands the local accumulation rates in these zones would reach 4 cm year-1. Estimates of soft sediment mass accumulation rates (1–10 kg m-2 year-1) in the channel from Pb-210 and C-14 measurements from cores of deltaic mangrove mud cannot account for this accumulation rate on a 100–1000 year time scale. The fate of the remaining sediment is unknown, it may be exported from the estuary in the monsoon season.

Flushing of bowden reef lagoon, great barrier reef

Field and numerical studies were undertaken in 1986 and 1987 of the water circulation around and over Bowden Reef, a 5-km long kidney-shaped coral reef lagoon system in the Great Barrier Reef. In windy conditions, the flushing of the lagoon was primarily due to the intrusion into the lagoon of topographically induced tidal eddies generated offshore. In calm weather, such eddies did not prevail and lagoon flushing was much slower. The observed currents at sites a few kilometres apart in inter-reefal waters, have a significant horizontal shear apparently due to the complex circulation in the reef matrix. Under such conditions, sensitivity tests demonstrate the importance of including this shear in the specification of open boundary conditions of numerical models of the hydrodynamics around reefs. Contrary to established practice, the water circulation around a coral reef should not be modelled by assuming reefs are hydrodynamically isolated from surrounding ones. Little improvement appears likely in the reliability of reef-scale numerical models until the inter-reefal shear can be reliably incorporated in such models.

Dispersion and fate of produced formation water constituents in an Australian Northwest Shelf shallow water ecosystem

This was a study of produced formation water (PFW) discharged into a shallow tropical marine ecosystem on the Northwest Shelf of Australia. A combination of oceanographic techniques, geochemical tracer studies, chemical and biological assessment methods, and dispersion modelling was used to describe the distribution and fate of petroleum hydrocarbons and added nutrients discharged from an offshore production platform. Using fine scale volatile hydrocarbon data, the horizontal and vertical diffusion parameters for a three dimensional dispersion model were calibrated under local conditions. Trace hydrocarbon chemistry studies and integration of the data into a mass balance model, facilitated a comprehensive description of dispersion and degradation pathways and rates. Bio-accumulation into bivalves and water column microbial growth inhibition studies confirmed the chemistry and model predictions that the area of potential biological impact extended to 0.5 nautical miles (∼900 m) from the discharge with additional skewing in the direction of the predominant tidal flows. Impact would be expected to be concentrated in transient surface slicks and near surface seawater. Dispersion and degradation processes were fast enough to prevent any long-term build-up of contamination within the system. Trace levels of oil in the near field sandy sediments were directly related to the magnitude of the daily discharge. The study is a benchmark to help predict the effects of further oil industry expansion in this pristine coastal region.

Tidal current variability in the Central Great Barrier Reef

Abstract With tidal data from the literature and the field, a two-dimensional (depth-averaged) numerical model was formulated to simulate the dominant semi-diurnal tidal hydrodynamics of the Central Great Barrier Reef continental shelf. Importantly, the individual mesh dimensions of the numerical scheme were set at approximately 2 × 2 km which was sufficient resolution to incorporate the topography of each reef within the matrix. The model provided a new detailed understanding of the influence of the reef matrix on the tidal currents of the outer shelf. In particular, the model demonstrated that the spatial variability in the tidal currents speed and direction exists down to the scale of the 2 km grid size. The model also demonstrated the significant tidally-induced residual currents that result from the interaction with the complex topography of the reef matrix. The advective effect of these tidal currents would be significant as the tidally-induced residual currents are of similar magnitude to the non-tidal currents of the region. Further, the spatial variability in the modelled and observed tidal currents suggests highly spatially variable advective processes operate within the reef waters.