Hindcast of a Japan Sea oil spill

An oil spill model was applied to the Nakhodka tanker spill accident that occurred in the Japan Sea in January 1997. The amount of oil spilled was estimated to be around 5000 kl, released over 1 day. Under a 2-m wave height condition, and a 3.5% of drift factor, the model simulated the oil slick to hit the shoreline after 6 days. This was in good agreement with the observed conditions. After drifting to the shoreline, the oil slick moved northeastward with the current. In the model, the simulation where the shoreline absorbs 100% of stranded oil failed to reproduce the actual oil slick trajectory. The simulation in which oil resuspended after stranding indicated a similar trend to the actual case. Therefore, it is likely that a considerable amount of oil that hit the shoreline may have returned to the sea and moved with the current. The effects of current pattern and wind drift angle on the oil slick trajectory were also examined. It is suggested that the wind parameters were of prime importance in reproducing a realistic distribution.     

The technology windows-of-opportunity for marine oil spill response as related to oil weathering and operations

This paper identifies and estimates time periods as ‘windows-of-opportunity’ where specific response methods, technologies, equipment, or products are more effective in clean-up operations for several oils. These windows have been estimated utilizing oil weathering and technology performance data as tools to optimize effectiveness in marine oil spill response decision-making. The windows will also provide data for action or no-action alternatives. Crude oils and oil products differ greatly in physical and chemical properties, and these properties tend to change significantly during and after a spill with oil aging (weathering). Such properties have a direct bearing on oil recovery operations, influencing the selection of response methods and technologies applicable for clean up, including their effectiveness and capacity, which can influence the time and cost of operations and the effects on natural resources.The changes and variations in physical and chemical properties over time can be modeled using data from weathering studies of specific oils. When combined with performance data for various equipment and materials, tested over a range of weathering stages of oils, windows-of-opportunity can be estimated for spill response decision-making. Under experimental conditions discussed in this paper, windows-of-opportunity have been identified and estimated for four oils (for which data are available) under a given set of representative environmental conditions. These ‘generic’ windows have been delineated for the general categories of spill response namely: (1) dispersants, (2) in situ burning, (3) booms, (4) skimmers, (5) sorbents, and (6) oil-water separators. To estimate windows-of-opportunity for the above technologies (except booms), the IKU Oil Weathering Model was utilized to predict relationships—with 5 m s⁻¹ wind speed and seawater temperatures of 15°C.The window-of-opportunity for the dispersant (Corexit 9527^®) with Alaska North Slope (ANS) oil was estimated from laboratory data to be the first 26 h. A period of ‘reduced’ dispersibility, was estimated to last from 26–120 h. The oil was considered to be no longer dispersible if treated for the first time after 120 h. The most effective time window for dispersing Bonnic Light was 0–2 h, the time period of reduced dispersibility was 2–4 h, and after 4 h the oil was estimated to be no longer dispersible. These windows-of-opportunity are based on the most effective use of a dispersant estimated from laboratory dispersant effectiveness studies using fresh and weathered oils. Laboratory dispersant effectiveness data cannot be directly utilized to predict dispersant performance during spill response, however, laboratory results are of value for estimating viscosity and pour point limitations and for guiding the selection of an appropriate product during contingency planning and response. In addition, the window of opportunity for a dispersant may be lengthened if the dispersant contains an emulsion breaking agent or multiple applications of dispersant are utilized. Therefore, a long-term emulsion breaking effect may increase the effectiveness of a dispersant and lengthen the window-of-opportunity.The window-of-opportunity of in situ burning (based upon time required for an oil to form an emulsion with 50% water content) was estimated to be approximately 0–36 h for ANS oil and 0–1 h for Bonnie Light oil after being spilled. The estimation of windows-of-opportunity for offshore booms is constrained by the fact that many booms available on the market undergo submergence at speeds of less than 2 knots. The data suggest that booms with buoyancy to weight ratios less than 8:1 may submerge at speeds within the envelope in which they could be expected to operate. This submergence is an indication of poor wave conformance, caused by reduction of freeboard and reserve net buoyancy within the range of operation. The windows-of-opportunity for two selected skimming principles (disk and brush), were estimated using modeled oil viscosity data for BCF 17 and BCF 24 in combination with experimental performance data developed as a function of viscosity. These windows were estimated to be within 3–10 h (disk skimmer) and after 10 h (brush skimmer) for BCF 17. Whereas for BCF 24, it is within 2–3 d (disk skimmer) and after 3 d (brush skimmer).For sorbents, an upper viscosity limit for an effective and practical use has in studies been found to be approximately 15,000 cP, which is the viscosity range of some Bunker C oils. Using viscosity data for the relative heavy oils, BCF 17 and BCF 24 (API gravity 17 and 24), the time windows for a sorbent (polyamine flakes) was estimated to be 0–4 and 0–10 d, respectively. With BCF 24, the effectiveness of polyamine flakes, was reduced to 50% after 36 h, although it continued to adsorb for up to 10 d. For BCF 17, the effectiveness of polyamine flakes was reduced to 50% after 12 h, although it continued to adsorb for up to 4 d. The windows-of-opportunity for several centrifuged separators based upon the time period to close the density gap between weathered oils and seawater to less than 0.025 g ml⁻¹ (which is expected to be an end-point for effective use of centrifugal separation technology), were estimated to be 0–18 (ANS) and 0–24 h (Bonnie Light) after the spill. Utilizing the windows-of-opportunity concept, the combined information from a dynamic oil weathering model and a performance technology data base can become a decision-making tool; identifying and defining the windows of effectiveness of different response methods and equipment under given environmental conditions. Specific research and development needs are identified as related to further delineation of windows-of-opportunity.     

Maersk Navigator oil spill in the great channel (Andaman Sea) in January 1993 and its environmental impact

Observations on oil slicks, tar residues and dissolved petroleum hydrocarbons (DPH) shortly after the oil spill resulting from the tanker accident in January 1993 showed negligible impact on the Indian EEZ of the Great Channel (Andaman Sea). DPH were between 0.31 and 1.85 μg l⁻¹ in the area examined. Tar residues were absent throughout the study area. Prevailing NE wind with resultant SW surface current appears to have pushed the oil patches out towards the open Indian Ocean.A follow-up survey of the same area was carried out in September-October 1993 and observations similar to those made during the earlier survey were recorded. The zooplankton biomass had increased considerably during the interval between the two surveys, but this was probably due to seasonal changes and natural variability.The spill did not cause any perceptible impact on the environment.     

Quantitative analysis of alternate oil spill response strategies using OSCAR

A three-dimensional numerical model of the physical and chemical behavior and fate of spilled oil has been coupled to a model of oil spill response actions. This coupled system of models for Oil Spill Contingency and Response (OSCAR), provides a tool for quantitative, objective assessment of alternative oil spill response strategies. Criteria for response effectiveness can be either physical (‘How much oil comes ashore?’ or ‘How much oil have we recovered?’) or biological (‘How many biologically sensitive areas were affected?’ or ‘What exposures will fish eggs and larvae experience in the water column?’). The oil spill combat module in the simulator represents individual sets of equipment, with capabilities and deployment strategies being specified explicitly by the user. The coupling to the oil spill model allows the mass balance of the spill to be affected appropriately in space and time by the cleanup operation as the simulation proceeds. An example application is described to demonstrate system capabilities, which include evaluation of the potential for both surface and subsurface environmental effects. This quantitative, objective approach to analysis of alternative response strategies provides a useful tool for designing more optimal, functional, rational, and cost-effective oil spill contingency solutions for offshore platforms, and coastal terminals and refineries.     

The role of wind and emulsification in modelling oil spill and surface drifter trajectories

Laboratory and field data suggest that the movement of spilled oil at sea is in general a three-dimensional phenomenon in physical space, whereas trajectories of undrogued surface drifters are more susceptible to two-dimensional analysis. These conclusions are consistent with the intermittent failure of two-dimensional surface models to simulate the trajectories of spilled oil, although such models may be more successful with data from surface drifters. A physical explanation is presented, and a model that incorporates the key portions of the governing processes is described and tested against data from experimental oil spills at sea. Observations suggest that emulsified surface oil will drift down wind at speeds in excess of 3% of the windspeed. When surface turbulence drives oil subsurface for a significant fraction of time, however, net transport speeds are considerably less and significantly to the right of the wind in the northern hemisphere.     

COASTMAP: An integrated monitoring and modeling system to support oil spill response

One of the most difficult tasks in oil spill response modeling is to provide accurate estimates of the currents and winds during the spill event. This is typically done in an ad-hoc, subjective manner combining very limited field observations with simplified hydrodynamic and meteorological models. As an alternative an integrated environmental monitoring and modeling system, called COASTMAP, is presented. COASTMAP allows the user to collect, manipulate, display, and archive real-time environmental data through an embedded geographic information system and environmental data management tools; to perform simulations with a suite of environmental models (e.g. hydrodynamics, meteorological) in order to predict dynamics in the operational area and to assimilate real-time data into the models to allow hindcasting, nowcasting and forecasting. COASTMAP, operational on a personal computer, is controlled by mouse/keyboard through a series of menus and uses color graphics to present model predictions (plots, graphs, animations) and the results of data analyses. The software is designed using a shell based architecture making application to any geographic location simple and straightforward.In the present paper, COASTMAP is linked with OILMAP to provide a fully operational, real-time system that allows prediction of circulation, winds and oil spill trajectory and fate for estuarine and coastal sea areas. System performance is illustrated by the simulation of the trajectory of oil tracking buoys during two experiments performed in the lower west passage of Narragansett Bay. Simulation results using several forecast procedures, with/without real-time data, are presented.     

Oil and water separation in marine oil spill clean-up operations

This paper discusses the changes in spilled oil properties over time and how these changes affect differential density separation. It presents methods to improve differential density, and operational effectiveness when oil-water separation is incorporated in a recovery system. Separators function because of the difference in density between oil and seawater. As an oil weathers this difference decreases, because the oil density increases as the lighter components evaporate. The density also increases as the oil incorporates water droplets to form a water-in-oil emulsion. These changes occur simultaneously during weathering and reduce the effectiveness of separators. Today, the state-of-the-art technologies have limited capabilities for separating spilled marine oil that has weathered.For separation of emulsified water in an emulsion, the viscosity of the oil will have a significant impact on drag forces, reducing the effect of gravity or centrifugal separation. Since water content in an emulsion greatly increases the clean up volume (which can contain as much as two to five times as much water as the volume of recovered oil), it is equally important to remove water from an emulsion as to remove free water recovered owing to low skimmer effectiveness. Removal of both free water and water from an emulsion, has the potential to increase effective skimming time, recovery effectiveness and capacity, and facilitate waste handling and disposal. Therefore, effective oil and water separation in marine oil spill clean-up operations may be a more critical process than credited because it can mean that fewer resources are needed to clean up an oil spill with subsequent effects on capital investment and basic stand-by and operating costs for a spill response organization.A large increase in continuous skimming time and recovery has been demonstrated for total water (free and emulsified water) separation. Assuming a 200 m³ storage tank, 100 m³ h⁻¹ skimmer capacity, 25% skimmer effectiveness, and 80% water content in the emulsion, the time of continuous operation (before discharge of oil residue is needed), increases from 2 to 40 h and recovery of oil residue from 10 to 200 m³.Use of emulsion breakers to enhance and accelerate the separation process may, in some cases, be a rapid and cost effective method to separate crude oil emulsions. Decrease of water content in an emulsion, by heating or use of emulsion breakers and subsequent reduction in viscosity, may improve pumpability, reduce transfer and discharge time, and can reduce oily waste handling, and disposal costs by a factor of 10. However, effective use of emulsion breakers is dependant on the effectiveness of the product, oil properties, application methods and time of application after a spill.     

The integration of trajectory models and analysis into spill response information systems

The spill response community is engaged in a technological rush towards computer-based, information-synthesis systems. Typically, they are modeled after many successful ‘incident command’ or ‘command and control’ systems that rely on micro- or mini-computer technology that is friendly and graphically oriented. Virtually all of these systems offer spill trajectory modeling components. What is typically lacking in this modeling output is any reliable way to estimate the uncertainty. This means that advice derived from the models is of questionable value, and when integrated into a complex response plan, the propagation of errors could seriously compromise the usefulness of results. It is shown that no single trajectory model run can provide the necessary information to respond in an optimal, ‘minimum regret’ strategy. However, a well-defined series of model runs used as the basis for trajectory analysis can provide the required information. A discussion of options suggests that the adoption of a minimum standard analysis procedure would significantly improve the ability of integrated response systems to use the predictions of oil distributions.     

RADARSAT SAR for oil spill response

RADARSAT synthetic aperture radar imagery has been successfully classified to delineate oil slicks on water using training areas for various degrees of oil coverage located within each image. Three and four class schemes have been tested with imagery from the Nakhodka and Milford Haven spills. An interactive graphical editor has been developed using the classified images to re-initialize the SPILLSIM oil spill model during a simulation.     

Assimilation of radar measured surface current fields into a numerical model for oil spill modelling

In summer 1996, surface current fields were measured in the central Strait of Georgia using the SeaSonde, a portable shore-based HF radar system. The objective of the study was to assess the feasibility of blending SeaSonde currents with numerical model fields to fill gaps in the measured fields and reduce noise. Our eventual goal is to assimilate the blended SeaSonde fields into a three-dimensional numerical model to improve short-range current forecasts.The first part of the study involved using the blended current fields as input to a set of auto-regressive moving-average (ARMA) statistical models. The blended fields were found to give forecasts comparable in terms of RMS error with forecasts using raw SeaSonde measurements. Moreover, the blended fields were smoother and more spatially complete than the raw data. The second part of the study examined the suitability of using the SeaSonde current fields to update the surface layer of Seaconsult's C3 hydrodynamic model. A simple nudging method is presented as an economical way to drive the model surface layer using operationally gathered SeaSonde information.     

A Numerical Simulation of an Oil Spill in Tokyo Bay

An oil spill accident happened in Tokyo Bay on 2 July 1997. About 1500 m³ of crude oil was released on the sea surface from the Japanese tanker Diamond Grace. An oil spill model is applied to simulate the fate of spilled oil. The Lagrangian discrete-parcel method is used in the model. The model considers current advection, horizontal diffusion, mechanical spreading, evaporation, dissolution and entrainment in simulating the oil slick transformation. It can calculate the time evolution of the partition of spilled oil on the water surface, in the water column and the sedimentation on the bottom. A continuous source at constant rate is set up as a tanker off the coast of Yokohama. The grid size is 1 km in the calculation domain. The residual flow simulated by a 3-D hydraulic model and observed wind data are used for advection. The simulated distribution of oil spreading agrees well with observations from satellite remote-sensing.     

溢油监测技术在石油石化企业环境风险防控中的应用

针对石油石化企业的溢油风险,提出企业在厂区雨水系统、外排口、涉水生产设施、环境敏感受体、溢油事故应急处置5类场景下的溢油监测需求,总结了溢油监测技术的类型和特点,介绍了可见光、红外、紫外、荧光、高光谱、微波辐射、雷达、电磁能量吸收等溢油监测技术的应用现状和优缺点。提出:企业溢油监测系统可分为企业内部溢油风险分级管控监测、企业边界的溢油风险报警监测、敏感环境监视的风险预警监测、溢油事故应急救援的溢油处置监测4个层次的运行模式。     

How shell cleaned up a 400,000-gallon oil spill

On April 23, 1988, approximately 9,500 barrels (400,000 gallons) of San Joaquin Valley crude oil leaked from an aboveground storage tank at Shell Oil Company's Martinez Manufacturing Complex in Martinez, California and entered Suisun Bay, an important recreation area. This article describes the remediation techniques Shell used to protect and clean up the Bay's oiled marshes, sloughs, rocky shores, marinas, and sandy beaches, and discusses the main methods of oil spill response, site-specific factors that must be considered in choosing remediation techniques, the interaction between Shell and government agencies, and the costs associated with the spill. The cleanup's total cost was approximately $8.3 million, which did not include private claims and claims handling costs; Shell also signed a separate consent decree for $19.75 million with the state of California and the federal government. This spill and its aftermath emphasize the need for preparation that facilitates response actions, improves the chances for cooperation between responsible parties and government agencies, minimizes the time needed for remediation, lowers cleanup costs, and limits natural resource damage claims and penalties.     

Current predictions for oil spill models

Baroclinic currents for flow along the North Coast of British Columbia were modelled using a finite element approach. Observational data from Loran-C drifters were used to get surface truth data. Least squares fit was applied to both the wind-driven current and baroclinic current compared with the drifter velocities in Dixon Entrance and Hecate Straits. The current data achieved was found to be useful for oil spill modelling.     

Comparative occurrence rates for offshore oil spills

Estimates of occurrence rates for offshore oil spills are useful for analysis of potential oil spill impacts and for oil spill response contingency planning. As the Oil Pollution Act of 1990 (U.S. Public Law 101–380, 18 August 1990) becomes fully implemented, estimates of oil spill occurrence will become even more important to natural resource trustees and to responsible parties involved in oil and gas activities. Oil spill occurrence rate estimates have been revised based on U.S. Outer Continental Shelf platform and pipeline spill data (1964–1992) and worldwide tanker spill data (1974–1992). These spill rates are expressed and normalized in terms of number of spills per volume of crude oil handled. The revisions indicate that estimates for the platform spill occurrence rates declined, the pipeline spill occurrence rates increased, and the worldwide tanker spill occurrence rates remained unchanged. Calculated for the first time were estimates of tanker and barge spill rates for spills occuring in U.S. waters, and spill occurrence rates for spills of North Slope crude oil transported by tanker from Valdez, Alaska. All estimates of spill occurrence rates were restricted to spills greater than or equal to 159 m³ (1000 barrels).     

Computer Modeling of Oil Spill Trajectories With a High Accuracy Method

This paper proposes a high accuracy numerical method to model oil spill trajectories using a particle-tracking algorithm. The Euler method, used to calculate oil trajectories, can give adequate solutions in most open ocean applications. However, this method may not predict accurate particle trajectories in certain highly non-uniform velocity fields near coastal zones or in river problems. Simple numerical experiments show that the Euler method may also introduce artificial numerical dispersion that could lead to overestimation of spill areas. This article proposes a fourth-order Runge–Kutta method with fourth-order velocity interpolation to calculate oil trajectories that minimize these problems. The algorithm is implemented in the OilTrack model to predict oil trajectories following the “Nissos Amorgos” oil spill accident that occurred in the Gulf of Venezuela in 1997. Despite lack of adequate field information, model results compare well with observations in the impacted area.     

Overview and future trends in oil spill remote sensing

Remote sensing has great potential to provide data to improve oil spill response efforts. There are a number of sensors available that have been proven capable of detecting oil on water and measuring some of its properties. There is no single sensor that provides all the data needed, and hence a combination of sensors must be used. Even if finances and aircraft load capacity were unlimited, there are still many parameters of an oil slick that cannot be measured by remote sensing. This paper describes the cyrrently available sensors and their method of operation and outlines some new developments that have the potential to increase the amount of data available from an airborne remote sensing operation.     

Modeling the fate of oil spills at sea

A numerical model for the simulation of the physicochemical weathering processes of an oil spill at sea is presented based on state-of-the-art models. The model includes the most significant processes: spreading, evaporation, dispersion into the water column, emulsification and the change in viscosity and density. These processes depend on each other and are allowed to vary simultaneously since processes are described by a set of differential equations, solved by a fourth-order Runge-Kutta method. Numerical examples are given, in order to test the results obtained, and compared with available experimental data in the literature. The model predicts well the variation of water incorporation, density and viscosity but seems to overestimate the fraction evaporated. However more experimental data are needed to calibrate and validate the model since differences in the composition of the simulated oil and the samples from which experimental data are taken may occur in evaporation studies. The model is suitable to join other modules for the prediction of the spill trajectory by advection due to winds and currents and sub-sea transport.     

An approach to evaluate the application of the vulnerability index for oil spills in tropical Red Sea environments

The Egyptian national marine oil pollution contingency plan was urgently initiated after the Nabila oil spill in 1982, to provide an estimate of its environmental effects on the Egyptian Red Sea coastal areas and to determine geomorphological features and cuastal processes, together with physical, chemical and biological baseline data for this tropical environment.The ‘Vulnerability Index’ (VI) was applied to evaluate and calibrate the effect of the Nabila oil spill on the Egyptian Red Sea Coastal area. A detailed in situ coastal survey was conducted during two visits in November 1982 and May 1983 to 80 shore sites from Suez to Ras Banas to monitor the oil pollution and to apply the ‘Vulnerability Index’. A comparative assessment of the index over time by comparing it with a quick ground inspection in November 1993 to some sites to evaluate the applicability of this index for oil spills in such environments. In addition, the physical effects of fresh and weathered crude oil and/with dispersant on water filtration by different beaches were preliminary studied.The geomorphological/Vulnerability Index results show that most of the Egyptian Red Sea coastal environments have medium to high vulnerability to immediate and medium term oil spill damage. The oil pollution spread estimated to be 250 km south of the oil spill and about 200 km north of it. The quantity of oil along the shoreline was reduced by about 60% due to natural and authorities clean up. The third survey after 11 years showed that the VI could be used as a predictive tool for assessment of oil spill effects on such tropical environments.     

Oil Erosion in an Annular Flume by Seawater of Varying Turbidities: A Critical Bed Shear Stress Approach

Laboratory experiments were conducted in an annular flume using Hibernia crude oil to determine: (1) the critical shear stress (τ_c) necessary to remove stranded oil from a surface by resuspension and (2) the effect of suspended sediment concentrations (SSCs) on the oil erosion processes. Two types of erosion were evident: Type I––solution and erosion of soluble aromatics; and Type II––mass erosion of visible droplets. In particulate free seawater at 13 °C, the Type II erosion threshold τ_cII is 5.0 Pa. This is equivalent to a mean current velocity (U_y) of 0.55 m s⁻¹. At U_y values <0.55 m s⁻¹, Type I erosion occurred as shown by the increase of oil concentrations without visible erosion of the oil surface. Temperature has a strong control on the threshold and rate of oil erosion: the threshold for Type I erosion at 4 °C was higher and erosion rate lower than at 13 °C. No Type II erosion was observed at 4 °C. SSCs also affects the entrainment of oil. Oil erosion was most efficient at moderate SSCs. At very high SSCs, turbulence suppression and drag reduction became effective and oil erosion rate decreased. SSC at 200–250 mg l⁻¹ were observed to give maximum erosion efficiency and is therefore suggested as the optimal concentration for erosion and elimination of heavy crude oil at a water temperature of 13 °C.