Oil spill drift forecast and operational oceanography systems : the Prestige case and others…
Nicolas Ferry Pierre Daniel
Mercator-Ocean
Météo-France
Training Roses project, Satellite data for oil spill monitoring
Starlab, Barcelona, 3 June 2004
Outline
• MOTHY operational oil spill prediction system
• How to improve this system with operational oceanography systems ? - example with the Mercator system • Incidents : • Prestige 2002 (Off Galicia) • Lyria 1993 (Med. Sea)
MOTHY
: Modèle Océanique de Transport d’Hydrocarbures)
A pollutant drift model operated since 1994 Intensive use for Erika (1999) and Prestige (2002) events
France
Spain
MOTHY operational oil spill prediction system
A trajectory model Includes a hydrodynamic ocean model With direct access to weather (wind) forecast
MOTHY
MOTHY operational oil spill prediction system
Hydrodynamic model
Horizontal mesh: ~1.5 km resolution What does control the surface currents ?
- the bathymetry strong currents near the shelves / coast - the high frequency temporal and spatial variations of wind - the tides ~12h25’ period, strong impact near the coast : grounding - the large scale currents But also : the oil viscosity, density, its solubility in the water or its ability to evaporate…
Surface drift speed
What governs the oil drift ?
- The wind stress 2~3 % of the surface wind speed
20 km/h wind speed a drift of ~ 9 km/day - The ocean currents : 100% of the surface current 5 cm/s current a drift of ~ 4 km/day
• the wind field strongly impacts the oil spill prediction • Difficult to perform long range forecast for oil spill drift ~7-10d
6
Some examples :
The large scale ocean currents alone : the Prestige case
SPAIN
work by Sophie Besnard, Mercator
7
The impact of the surface wind : the Spanish trial in Gijón
Mean wind speed 8-13 May 2004 Buoy #2 Mercator
% of the wind speed
Some examples :
buoy#2
SPAIN
work by Sophie Besnard, Mercator
8
Prestige : oil spilled on the track
•
Prestige : November 2002
Prestige : oil spilled on the track
MOTHY operational alone
MERCATOR surface current (3 m) only
MOTHY and operational oceanography systems
How to improve this system with operational oceanography systems?
Use both an oil spill model and operational oceanography systems : combination of large scale currents from an operational oceanography system (Mercator) and the surface drift from MOTHY
Operational Oceanography system
Operational Oceanography system
How to include large scale currents into MOTHY ?
MOTHY: wind and tide currents, medium to high frequency spectrum. MERCATOR: wind current (low to medium frequency) and large scale current. How to use such a current in MOTHY, without counting 2 times the effect of the wind ? Current at the base of the mixed layer. For example: 100 m depth on the Prestige area
Prestige : oil spilled on the track
MOTHY operational MOTHY + Mercator 103m
In green, areas with positive impact of Mercator currents In red, areas where MOTHY alone is better
Prestige : oil spilled on the track
December 13, 2002
MOTHY operational
= observations
MOTHY is late. Addition of Mercator current at 103 m improves the simulation.
MOTHY and operational oceanography systems
MOTHY + Mercator 103m
MERCATOR and FOAM currents
Prestige
Prestige : oil spilled on the track
December 13, 2002
Observations: black triangles MOTHY alone: blue MOTHY+Mercator green MOTHY+FOAM: red
MOTHY and operational oceanography systems
But there is a new MERCATOR version
From MERCATOR Newsletter N°12
Oil spill drift at depth (3000m) ??
The PRESTIGE wreck : lies at 3500m depth, 125t fuel /day rejected Potential use of the Mercator deep currents analyses / forecasts :
Current at 3000m depth :
Lyria (1993)
Lyria
Lyria incident (1993): 3 weeks drift
Without large scale current
With climatology from in situ measurements (MODB)
With climatology from an ocean model (Mercator)
+ satellite altimetry (ERS1+TP)
Lyria
Lyria, August 17, 1993
Lyria
Lyria, August 23, 1993
Lyria
Summary
Direct use of operational oceanography system needs: • Hyper refinement in the first meters below the surface. • Use of high resolution wind forecasts • High resolution grid near the coast (1/60°) • Combine currents from an operational ocean system and MOTHY • Current at 100 m depth improve the drift, but this is not perfect • A need for new product: a current at the base of the mixed layer, other ? • Strong impact of the wind products used (MOTHY, MERCATOR) • Strong impact of the data assimilated (SLA, SST) on the forecast currents from Operational Oceanographic Systems
Acknowledgment
This work was supported by the Marine and Oceanography section at Météo-France and the Mercator-Ocean team. A large number of individuals help us in this work. We would particularly thank: P. Daniel (Météo France) N. Pinardi and C. Fratianni (INGV, Italy) M. Bell and M. Holt (Met Office, UK) M. Jouhan and P. Lazure (IFREMER) G. Caniaux (Météo-France, CNRM)
http://www.meteorologie.eu.org/mothy/ http://www.mercator-ocean.fr