Maritime Forum

Maritime Forum Themes

Close

Oil Leak Simulation - EMODnet Sea-Basin Checkpoint Challenge

Published on: Tue, 26/04/2016 - 09:56
Table of Contents
    As part of a stress test on marine data, two days ago teams were asked to assess the impact of simulated oil spills in six European seabasins. All of them managed to deliver assessments of the fate of the oil – dispersed, floating, evaporated, beached etc despite the short notice. Interesting results were the influence of ice in the Arctic, the impact on fishing in the Baltic, the differences in dispersion between the Atlantic and Black Sea and the threat to coastal habitats in the North Sea and Mediterranean. In all cases it is the wind that largely dictates the movement of the slick and the type of oil that dictates the evaporation rate. As in a real spill, the amount of information regarding the fate and trajectory of the oil increases rapidly over the first few days, as more information and resources come into play from the scenario, and information comes back from the field.

    The Checkpoint Oil Leak Challenges were launched on 10/05/2016 – an overview of all spills can be found at the end of the post.  

    Results

    This section will be updated with regular summaries of results of each of the checkpoints as they become available after 24 hours and after 72 hours.  

    Arctic

    At 10:27 (BST) on 10/05/2016, the Arctic Checkpoint was alerted that an explosion had occured at 08:15 (CET) on the Prirazlomnaya Platform, 60km off the coast in the Pechora Sea (see figure). Oil is leaking subsurface, at a rate of 800m³/day: this is expected to be reduced to 500 m³/day following emergency repairs with 24 hours, with the leak being stopped completely within 72 hours.

    Method and assumptions

    Initial modelling (results on 11/05/2016) was carried out using ASA’s OILMAPTM model. Because of the entrainment of the oil in the ice, the hydrodynamic data has been made equal to the ice velocity fields and wind action has been switched off. For later updates (on 12/05 and 13/05/2016) the oil spill forecast are based on SINTEF's OSCAR Model.

    It was assumed that the exact location of the platform was 69°16'4.44"N, 57°16'50.48"E, with a water depth of about 20m. It was also assumed that the oil was ARCO (Arctic Crude) due to the location and SG (0.910 kg/m3 ) with increased concentration of sulfur and low paraffin content.

    Results on 11/05/2016

    The oil is rising into an area that is currently an ice field meaning the oil will be entrained in the ice. The ice forecast predicts an westerly flow at approximately 0.3 m/s and no shoreline impact is expected. 

    Taking into consideration the thickness of the ice sheet, and the current location of the spill, it would be difficult to man a traditional response. The best option would be to monitor and track the oil under the ice sheet and be prepared to act once the ice has started to melt making the oil accessible. Some attempts to locate the oil under the ice and cut trenches for skimmers could be made, but this would depend on the availability of response vessels, and their ability to get through the ice. As such, containment options would be limited. In situ burning could be considered, although accelerants would have to be used.

    Figure: Location of the leaking oil platform in the 60km off the coast in the Pechora Sea

     

    Results on 13/05/2016

    The oil spill is in a complex, ice infested environment that reduces the reliability of any oil spill forecast.  In this case, the Prirazlomnaya Platform is in pack ice flowing westward at upto 0.5 m/s. The oil, which is lighter than water, can be expected to rise to the underside of the ice pack where it will be trapped into the moving ice sheet.  Over the last 72 hours, more and more oil was transported north-westward as the ice sheet flows over the top of where the leak is. The ice pack is currently breaking up, making an accurate prediction of where the oil can be recovered very difficult.

    Figure: Surface oil shown overlaid with a plot of maximum subsurface concentrations and a cross section of subsurface oil concentration at end of f Day 3 (13-May-2016 08:15 CET). Cross section is for arrow shown on map. (SINTEF OSCAR model, output updated 1100 12 May2016)

    Full reports and updates from the Arctic Checkpoint:

     

    Baltic Sea

    At 8:15 CET this morning (10/05/2016), a borehole located at (55,39974N ; 18,72303E) started to leak oil at a rate of 2500 barrels/day for a period of 3 days (1BBL=159L), which is equivalent to about 15tons/hour.

    The accident area is about 119 km northeast of Polish coast (Wladyslawowo) and 153 km west of Klaipeda, where there are Spatial Protected Areas in Natura2000. The oil spill site has a water depth of about 86 meters.

    Figure: location of the oil spill event (the location of the yellow star) and Natura2000 Areas from EMODnet Human Activities portal 

    Method and assumptions

    A continuous release of crude oil at the bottom of the sea is assumed in this event.

    Two operational oil spill system are used for this case study: (1) SMHI SEATRACK (https://stw-tst.smhi.se/) system to predict the drift of the oil spill with Hiromb as the ocean model and meteorological data from the ECMWF global model and HIRLAM regional NWP model; (2) DMI uses the BSHdmod to calculate oil drift and weathering, an add-on module to the hydro-dynamical model HBM developed at the Bundesamt für Seeschiffahrt und Hydrographie, Hamburg, Germany.The DMI model uses the full 3-D flow field. Input data used for the oil drift forecast include wind forecasts from DMI operational forecasting model HIRLAM (hourly, 3km resolution) , ECMWF (for forecast longer than 54hours), and current forecasts from HBM (15minutes, 5 km resolution).

    Results on 11/05/2016

    The accident area is dominated by a high pressure system in the afternoon of 10 May. Good weather and sea state conditions will last to the noon of 11 May. The winds will then increase up to Scale 6 and waves to 1-2 meters in the Polish waters in the following 24 hours. The weather and sea state conditions are in general good for oil spill combatting activities.

    Forecasts reveal that due to the calm weather condition, the oil does not drift too far from the incident site. The drift distance show that during the first day, the oil drifts less than 10km, and at 96h after the spill, the oil will drift away about 60km. As this is a spill at sea bottom, the vertical movement of the oil is also important. A major fraction of oil is moving to the surface in the first 24hours and % evaporated reaches 30% which appears high.

    Based on the forecasts made by DMI and SMHI, there will be no oil landing on the coast, and no impacts on the SPA areas in Natura2000. However, the impact on the marine ecosystems (especially benthic community and fishery) should be significant, and should be further investigated. The Baltic fishery is vulnerable as it has already experienced a decline in last 30 years.

    Full reports and updates from the Baltic Checkpoint:

     

    Atlantic

    At 8:15 CET on 10/05/2016) an oil tanker (located at LAT:47,330945;LON:-4,366687) reported it was breaking in several parts. By10:00 CET the ship had released all of its 8,000 tons of heavy fuel oil n°6. 

    Method and assumptions

    It was assumed that the oil was released at the surface. According to the oil profiles available in OSCAR, the oil was assumed to be similar to the HEAVY FUEL OIL profile, with an American Petroleum Institute (API) of 15.1.

    The checkpoint used the OSCAR (Oil Spill Contingency and Response) model developed by SINTEF for simulating oil spill fate at sea and operated by CLS under license. It is able to evaluate the evolution of oil on the water surface and water column and along shorelines. OSCAR computes the fate and weathering of oil, and potential biological effects. The Copernicus marine service forecasting service and the AVISO tide product are used for current forecast and the wind is extracted from NCEP/NOAA. As the currents are available for a five day period, the first report will provide forecast of the oil spill behaviour until 15/05/2016. The Sea bed substrate is taken from the EMODnet Seabed Habitats database. Information relative to coastal environment and human activities from different sources will be gathered to produce impact maps. QGIS will be used to display these different sets of data and to cross them with the oil spill forecast.

    Results on 11/05/2016

    During the first three days, the oil slick mostly remains contained around the release site. From that moment on, the oil slick spreads toward South-West under the influence of strong land breeze. It can be noted that the currents in the region are mostly driven by the tidal component. As a result, they don’t present a dominant tendency that could influence the oil slick behaviour.  

    During the first three days, the oil slick mainly remains at the surface. From that moment on, dispersion is the process responsible for most oil removal. Dispersion refers to one of the natural process of dissipation of an oil spill at sea: waves and turbulence at the sea surface can cause all or part of a slick to break up into fragments and droplets of varying sizes. These become mixed into the upper levels of the water column. As a result, after 80 hours the oil slick starts spreading reaching a maximum area of 5600 km².

     

    Figure: Chronological spreading of the surface oil spill