The Australian Institute

of Marine Science

(AIMS) and the

Australian Geological

Survey Organisation

(AGSO) have combined

their expertise to conduct

the first multi-disciplinary

study of Produced

Formation Water

discharged into

Australia's shallow

coastal seas. Dr Kathryn

Burns reports on the

project.

Using the AIMS research vessel, Lady Basten, the project team conducted detailed measurements around the Harriet Alpha platform on the NW Shelf in September 1995. The Harriet A platform is a major source of PFW on the NW Shelf. A pilot study conducted the previous year provided a good basis for planning field sampling and analyses. Dr Miles Fumas and Alan Mitchell were the biological oceanographers. Two current meter arrays were set to record tidal speed and directions over the time period which covered both spring and neap tide conditions. The oceanographic properties and processes measured included water temperature, salinity, subsurface light penetration, ambient nutrient concentrations, the quantity and quality of the water column particulate matter, phyto- and zoo-plankton biomass, primary production, bacterial production, nutrient utilisation rates, and fluxes of particulate matter to the sediments. Experiments were also carried out to measure the responses of natural phytoplankton and pelagic bacterial populations to important PFW constituents.

AIMS has conducted hydrographic measurements on the NW Shelf for many years and the present study can be compared with this larger database. The waters around the area of Varanus Island and the Harriet platform were generally well mixed by tidal currents and had characteristically low dissolved nutrients and phytoplankton biomass concentrations. Despite this, primary production rates were relatively high. Doubling times for the phytoplankton were calculated to be greater than once per day, with a substantial portion of the production (19 to 79%) diverted to bacterial pools. Direct measurements of the uptake rates of limiting nutrients in the water column gave turnover times of 0.5 to 1 hour. This fast rate of growth was found both near the platform and at control sites in shallow waters far away from the platform.

Based on the records of PFW discharged and the concentrations of nutrients and organic carbon in the effluent, the team concluded that daily discharges of organic carbon and phosphates were small compared with the natural supply rates of these materials. In the case of nitrogen, the daily output was significant, but it appeared to be rapidl dispersed and utilised by the plankton. The ship-board experiments showed some inhibition of bacterial and phytoplankton production by PFW and some of its constituents.

This data will be used to estimate an area of potential impact within the vicinity of the discharge after the modelling team and chemists define the areas of the plumes, and the dilution and dispersion processes. Drs David Heggie, David Holdway and Colin Tindall miniaturised the real time gas sampling gear which AGSO normally uses for oil and gas exploration work from the much larger ship, Rig Seismic. From the Lady Basten they provided fine timeline tracers of the PFW plume using volatile hydrocarbon distributions (benzene and toluene, plus methane to hexane totals).

The ships captain, Brian McCarthy, became adept at spotting the plume and steamed the ship slowly across it during the various tidal cycles. Volatile hydrocarbons were measured every two minutes, and included both surface and depth samples. Each data point was identified by time and position. The volatile hydrocarbon analyses detected very low concentration of some hydrocarbons in the plume (close to the limits of detection of two parts per trillion) up to 10 nautical miles (nm) in the directions of tidal forcing. From this data set, it was possible to define both the horizontal and the vertical diffusion parameters and load these into a model to characterise plume mixing for the region.

Dr. Brian King running the @MUDMAP model at his computer.

This comparison revealed that both techniques provided good predictions of the near platform PFW distributions. However point source data was not useful for farfield prediction when there was spatial variability on the current field, as is usual in shallow coastal waters. The model predicted the changes in plume dispersion and concentrations due to tidal oscillations and showed that previously dispersed PFW water may return in later tidal cycles for a "second dosing" only in very calm conditions. Validating the model predictions with high resolution field data creates an effective management tool for defining potential impacts by simulating best and worst case scenarios for environmental assessment. Further, it can be used to estimate the impact of. changes in practices, such as altering pipe diameters or improving oil removal prior to PFW discharge.

The hydrocarbon chemists were Dr. Kathryn Burns and Sue Codi. They measured the dispersion of the PFW into dissolved and undissolved fractions using moored water samplers, surface screen samplers and moored sediment traps. Much of the undissolved fractions appeared to have bound to small particles. They studied bio-accumulation by using transplanted oysters. They measured dispersion into sediment with benthic grab samples. These samples were frozen and returned to AIMS for detailed analysis.

Dr. Kathryn Burns in her organic chemistry lab at AIMS.

There was a decrease in sediment oil concentrations within the 0.5 nm radius from 2.8 mg/g dry wt (average of 16 samples) in 1994 to 1.0 mg/g dry wt (average of 20 samples) in 1995. This decrease between the years probably reflects the decrease in particulate associated residues discharged within the dispersed plume by the platform, due to the addition of a centrifugal separator (hydrocyclone). Concentrations of oil from the PFW discharge associated with the production of the relatively light crude oil result in a definable pattern of trace contamination of the surrounding sediments. However, residence time of the oil in the sediments was estimated to be less than 1 year. Concentrations of oil in sediments at both sampling times around Harriet A were within the range of recently published control (unoiled reference) sites in the Caribbean.

Our sensitive analytical methods and effective sampling strategy defined the distribution of oil in sediments around the platform. While it is considered unlikely that benthic ecosystems would be adversely impacted, the team concluded that further assessment at this site should emphasize: processes of bio-accumulation, potential water column toxicity, and the further study of benthic communities. Future chemical studies could focus on photooxidation and evaporation as the likely major mechanisms for removing hydrocarbons from the surface seawaters.

The collaborative application of oceanographic, geochemical and modelling techniques is providing effective feedback to industry on the impact of their operational strategies in Australia's pristine offshore coastal regions. A future report will be about verifying dispersion models for platform discharges in Bass Strait.

Organic chemistry lab.