target any significant fresh releases of oil from the tanker. Once these had been successfully treated with dis-persant then a secondary target was the larger patches of more weathered oil further offshore. These patches were probably the result of oil releases at low tide during the hours of darkness which could not immediately be treated with dispersants.
MONITORING THE EFFECTIVENESS OF THE DISPERSANT RESPONSE THROUGH FLUO-ROMETRY AND REMOTE SENSING
Sometimes, as was seen for the application of dispersant to fresh Forties Blend being released form the SEA EMPRESS, aerial observation is sufficient as the dispersant can produce a visible plume of dispersed oil just below the water surface. However, as reported in NETCEN field trials (Lunel 1994, 1995) and incorporated into the IMO/UNEP guidelines on oil spill dispersant application (IMO 1995), field experiments have shown that enhanced oil concentrations can occur in the water column following dispersant use, without the appearance of plumes. This indicates that dispersion can occur in the absence of the usual visual indications, and therefore the IMO guidelines suggest that subsurface oil concentrations should be measured in addition to conventional aerial observations.
The SEA EMPRESS incident showed that this subsurface monitoring is essential once the oil had weathered for more than 12-18 hours. By this point surface sampling had shown that 40% had evaporated and that a 70% water-in-oil emulsion had formed. In the absence of dispersants natural dispersion still resulting an exposure of the marine environment to dispersed oil although elevated concentrations were mostly restricted to the top 1 m of the water column (Fig. 3) . On emulsification after 12-48 hours, the natural dispersion process slowed signif-icantly. When dispersant was applied to emulsions of Forties Blend, the first application of dispersant tended to break the emulsion while subsequent additions increased the concentrations of dispersed oil (Fig. 4) . This was consistent with previous trials in the North Sea with Forties (Lunel & Lewis 1994, Walker & Lunel 1995) and other North Sea crudes (Lewis et al. 1995 ; Brandvik et al. 1995) when the dispersant operation was successful in breaking the water-in-oil emulsion and then dispersing it. Monitoring of the surface oil properties by boat was critical for the dispersant operation to treat the emulsified oil since it is very difficult to get a clear indication of the extent of emulsion break down from visual observations from the aircraft. NETCEN scientists were able to inform MPCU of the demulsifying effect of the dispersant even though dispersed oil concentrations were not elevated greatly immediately following treatment.
MONITORING THE RATES OF DILUTION OF THE DISPERSED OIL
The initial monitoring of oil concentrations in the water column was carried out with a focus of determining the efficiency of the dispersant operation. Evidence from NETCEN monitoring programmes has shown that the dis-persant was effective in removing oil from the sea surface. The hazard posed by a particular pollution event is a combination of the toxicity arising from the concentration of the pollutant and the length of exposure to the elevated concentrations. The potential for environrnental disbenefit is in the possible generation of oil concen-trations in the water column which :
・ are acutely toxic
・ cause long term tainting of commercially important fish and shellfish
There is much debate on the threshold of oil concentration where toxic effects are likely to be observed. The 2 major issues being firstly, that different components of the oil have very different toxic effects (the light aromatic compounds such as benzene and xylenes having the greatest effect), secondly, different species and life stages have a wide range of tolerances to dispersed oil. A workshop was held in the USA in 1995 on the impli-cations of the use of dispersants since the USA is increasingly considering dispersants as one of their response options to oil spills. The workshop addressed concerns over the potential toxicity of dispersed oil has limited the use of dispersants in responding to major oil spills. This included many of the main researchers in the oil spill field. One of the workshop's conclusions was that"the available acute(short-timescale) toxicological data support the conclusion that, at water column concentrations at or below 10 ppm dispersed oil for 2-4 hour exposure duration, adverse ecological effects are not expected".
During the dispersant operation at the SEA EMPRESS dispersed oil concentrations only locally exceeded 10
