1
APPLICATIONS OF OPERATIONAL OCEANOGRAPHY WITHIN THE UK Mike Bell 1 , Nick Ashton 1 , Paul Coverdale 2 , Adrian Hines 1 , Tim Johns 1 , Torleif Lothe 3 , Jon Rees 4 , Adam Scaife 1 , Robin Stephens 5 , Derek Swannick 1 © Crown copyright 07/0XXX Met Office and the Met Office logo are registered trademarks Met Office FitzRoy Road, Exeter, Devon, EX1 3PB United Kingdom Tel: +44 (0)1392 886434 Fax: 01392 885681 Email: [email protected] 1. Overview The Met Office has a number of operational marine prediction systems (see §2) The Royal Navy makes use of many of these capabilities (§3) The Environment Agency use storm surge predictions (§4) Cefas predict spill drift & dispersion in UK waters (§5) Surface met and deep ocean current hindcasts support the oil & gas industry (§6) Bottom water temperature predictions support gas pipelines in the North Sea (§7) Seasonal forecasts for north-west Europe have been improved recently by higher resolution ocean models (§ 8) Short-range weather forecasts use SST analyses and will use coupled models (§9) 4. Use of storm surge predictions 1 Met Office, Exeter, UK 2 Royal Navy, Portsmouth, UK 3 Gassco, Stavangar, Norway 4 Cefas, Lowestoft, UK 5 BMT, Teddington, UK 7. Norwegian Gas Transport System Forecasts 2. Operational prediction systems 5. Spill drift and dispersion in UK waters Surface wave and 3D current hindcasts are essential information for safe, economic and environmentally responsible decision making within the O&G industry. “The use of ocean model predictions is fundamental to the Royal Navy’s understanding of the underwater environment, and aids decision makers on the frontline to gain a battle-winning edge at tactical and operational levels” • Tests of global short-range coupled atmos-ocean NWP systems show improved performance in the tropics (±30º) – see figures • First tests of a weakly coupled atmos- ocean global data assimilation system have been successful • 12 km global atmos-ice-wave-ocean NWP systems are expected to be affordable within 3 years • A 1.5 km grid atmos-river-ocean-wave demonstration prediction system will be developed for UK waters over next 2-3 years • The figures show rms errors for the tropical (± 30º) atmos from coupled (red) and uncoupled (blue) trials and higher resolution operational forecasts (green & yellow) 3. Royal Navy • The new GloSea5 seasonal forecast system uses much higher resolution (table) than the previous one • The cold winter bias in the North Atlantic SST (upper plots) is much reduced (red circles) • The bias in the winter westerlies (lower plots) is much reduced (for 60º W - 10º E) • The frequency of blocking over N Europe is greatly improved • Winter forecasts of NAO from early Nov have correlation score of 0.62 • Details in Scaife et al. 2011 (GRL) 1: Number of: L = layers; F = frequencies, D = directions ; E = ensemble members 2: NWES = north-west European shelf (out to 30º W) Current speed predictions (cm/s) for the west African continental shelf break at 1000 m depth showing Kelvin waves propagating from the equator along the slope. Effective management of the main phases in the life-cycle of offshore O&G operations requires relevant information about the surface waves, swell and ocean currents to which the structures and operators are subjected and which will transport and disperse any oil and gas that is discharged into the ocean. • Accurate surface wind and wave hindcasts and predictions have been supplied to the O&G industry for many years. The Met Office provides met- ocean support for the O&G industry within the North Sea through Aberdeen Weather Centre with gross turnover of £4Mpa. • Ocean current hindcasts and predictions are less accurate but their quality is improving and they now provide valuable complementary information to in situ current meter measurements. • The Met Office has teamed up with BMT to generate and supply 20-year hindcasts of 3D ocean currents for the West African O&G industry. Model Type Model name Area Grid spacing No 1 Range (days) Reference SST & sea-ice analysis OSTIA Global 1/20º 1 0 Donlon et al. 2012 Rem Sens Env Storm surge CS3X NWES 2 12km 24E 6 Flowerdew et al. 2012 Q J Roy Met Soc Surface waves Wave- Watch III Global 35 km 30F, 24D 6 Li & Saulter 2013 Proc 13 Int Work. Wave Hind. Euro 12 km 6 UK 4 km 2.5 Deep ocean FOAM deep Global 1/4º 75L 7 Storkey et al. 2010 J Oper. Ocean. Blockley et al. 2013 N Atlantic, Indian 1/12º 50L 7 Shelf- seas FOAM shelf NWES 2 7 km 33L 5 O’Dea et al. 2012 J. Oper. Ocean. Arabian Gulf 4 km 30L 5 Hyder et al. 2013 J. Oper. Ocean. Seasonal GloSea5 Global 1/4º O 60 km A 75L 220 MacLachlan et al. 2013 Q J Roy Met Soc (sub) SALINITY WATER TEMPERATURE CHLOROPHYLL (BIOLUMINESENCE) WATER CLARITY MARINE MAMMAL DENSITY AMBIENT NOISE WIND WAVES Global surface wave and Forecasting Ocean Assimilation Model (FOAM) predictions are used for day-to-day operational forecasting of the water column around the world for the Royal Navy. The detail they provide about oceanographic features gives tactical commanders the knowledge to make informed decisions about maritime safety and routeing, and to deploy ships, aircraft and submarines and their weapons and sensors effectively. 6. Support for oil industry • Operated by Gassco, Norway; 8000 km sea-bed pipelines; riser platforms and processing facilities 6 receiving terminals in 4 countries; • Transports 20% of gas used in Europe • Will handle a larger production than the Norwegian oil industry within a few years Gas temperature (and therefore volume and pressure) are proportional to the sea bed temperature (SBT) Capacity planning and sales hence sensitive to SBT Leakage detection systems critically dependent on SBT Optimal use of additives to prevent formation of hydrate (=‘Gas Ice’), depends on SBT 8. Seasonal predictions 9. Short-range weather forecasts Temperature 850hPa JJA [K] 1.4 1.0 0.6 T+ day 1 2 3 4 5 [mb] 0 500 1000 5 10 15 20 25 30 35 (m) Geopot height 5 day f/c JJA Coupled 50 km (N256) Atmos 50 km (control) Atmos 25 km (opnl) Atmos 40 km (opnl) GloSea4 GloSea5 Grid GloSea4 GloSea5 Atmos @150 km @60 km Ocean 0.25º • The Flood Forecast Centre in the Met Office Operations Centre is jointly staffed by the Environment Agency (EA) and the Met Office • It provides guidance to the EA regional centres on coastal flooding • Coastal flooding is still on the national risk register. • Low water forecasts are also important to warn ships entering port of the risks of grounding • Surge predictions are generated using 24 member ensembles of weather forecasts to drive 12 km barotropic surge model predictions for the NW European shelf • The figures show a low pressure event when the danger of flooding at Millport was first forecast 5 days ahead and consistently forecast 3 days ahead Development of higher resolution (e.g. 2.2 km grid) atmos-wave-ocean surge predictions is highly desirable Red = flood warning level (depends on tides) Green = short-range surge ensemble Blue = long-range surge ensemble Cyan = deterministic hindcast Black = raw observations Synoptic situation at 00Z on 8 Nov (at time of predicted coastal flooding) Millport • Cefas has a 24/7 multi- disciplinary Emergency Response Team for oil/chemical and other spills in UK waters • It has roles in modelling (<1 hour response), monitoring & providing advice (e.g. on the use of dispersants) • It uses various oil & chemical models & inputs (see figure) and receives operational wind, wave & current data feeds from the Met Office • It has a good track record with Napoli, Ice Prince (wood spill pictured) and other incidents • Predictions for deep water north-west of Shetland require additional challenging capabilities Tidal Database Met Office Winds MyOcean Flow fields Model selection and parameterisation Cefas Spill ASA Oilmap Sintef OSCAR Analysis and Interpretation

APPLICATIONS OF OPERATIONAL OCEANOGRAPHY WITHIN THE …godae-data/Symposium/GOV-posters/... · 2019. 3. 1. · The Met Office has a number of operational marine prediction systems

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: APPLICATIONS OF OPERATIONAL OCEANOGRAPHY WITHIN THE …godae-data/Symposium/GOV-posters/... · 2019. 3. 1. · The Met Office has a number of operational marine prediction systems

APPLICATIONS OF OPERATIONAL OCEANOGRAPHY

WITHIN THE UKMike Bell1, Nick Ashton1, Paul Coverdale2, Adrian Hines1, Tim Johns1,

Torleif Lothe3, Jon Rees4, Adam Scaife1, Robin Stephens5, Derek Swannick1

© Crown copyright 07/0XXX Met Office and the Met Office logo are registered trademarks

Met Office FitzRoy Road, Exeter, Devon, EX1 3PB United Kingdom

Tel: +44 (0)1392 886434 Fax: 01392 885681

Email: [email protected]

1. Overview

The Met Office has a number of operational marine prediction systems (see §2)

The Royal Navy makes use of many of these capabilities

(§3)

The Environment Agency use storm surge predictions (§4)

Cefas predict spill drift & dispersion in UK waters (§5)

Surface met and deep ocean current hindcasts support the

oil & gas industry (§6)

Bottom water temperature predictions support gas

pipelines in the North Sea (§7)

Seasonal forecasts for north-west Europe have been

improved recently by higher resolution ocean models

(§ 8)

Short-range weather forecasts use SST analyses and will

use coupled models (§9)

4. Use of storm surge predictions

1Met Office, Exeter, UK2Royal Navy, Portsmouth, UK3Gassco, Stavangar, Norway4Cefas, Lowestoft, UK5BMT, Teddington, UK

7. Norwegian Gas Transport System

Forecasts

2. Operational prediction systems

5. Spill drift and dispersion in UK waters

Surface wave and 3D current hindcasts are essential information for safe, economic and environmentally responsible decision making within the O&G industry.

“The use of ocean model predictions is fundamental to the Royal Navy’s understanding of the underwater environment, and aids decision makers on the frontline to gain a battle-winning edge at tactical and operational levels”

• Tests of global short-range coupled atmos-ocean NWP systems show improved performance in the tropics(±30º) – see figures

• First tests of a weakly coupled atmos-ocean global data assimilation system have been successful

• 12 km global atmos-ice-wave-ocean NWP systems are expected to be affordable within 3 years

• A 1.5 km grid atmos-river-ocean-wave demonstration prediction system will be developed for UK waters over next 2-3 years

• The figures show rms errors for the tropical (± 30º) atmos from coupled (red) and uncoupled (blue) trials and higher resolution operational forecasts (green & yellow)

3. Royal Navy

• The new GloSea5 seasonal forecast system uses much higher resolution (table) than the previous one• The cold winter bias in the North Atlantic SST (upper plots) is much reduced (red circles)• The bias in the winter westerlies (lower plots) is much reduced (for 60º W − 10º E)• The frequency of blocking over N Europe is greatly improved• Winter forecasts of NAO from early Nov have correlation score of 0.62• Details in Scaife et al. 2011 (GRL)

1: Number of: L = layers; F = frequencies, D = directions ; E = ensemble members 2: NWES = north-west European shelf (out to 30º W)

Current speed predictions (cm/s) for the west African continental shelf break at 1000 m depth showing Kelvin waves propagating from the equator along the slope.

• Effective management of the main phases in the life-cycle of offshore O&G operations requires relevant information about the surface waves, swell and ocean currents to which the structures and operators are subjected and which will transport and disperse any oil and gas that is discharged into the ocean. • Accurate surface wind and wave hindcasts and predictions have been supplied to the O&G industry for many years. The Met Office provides met-ocean support for the O&G industry within the North Sea through Aberdeen Weather Centre with gross turnover of £4Mpa. • Ocean current hindcasts and predictions are less accurate but their quality is improving and they now provide valuable complementary information to in situ current meter measurements.• The Met Office has teamed up with BMT to generate and supply 20-year hindcasts of 3D ocean currents for the West African O&G industry.

Model

Type

Model

name

Area Grid

spacing

No1 Range

(days)

Reference

SST &

sea-ice

analysis

OSTIA Global 1/20º 1 0 Donlon et al. 2012

Rem Sens Env

Storm

surge

CS3X NWES2 12km 24E 6 Flowerdew et al. 2012

Q J Roy Met Soc

Surface

waves

Wave-

Watch

III

Global 35 km 30F,

24D

6 Li & Saulter 2013

Proc 13 Int Work. Wave

Hind.

Euro 12 km 6

UK 4 km 2.5

Deep

ocean

FOAM

deep

Global 1/4º 75L 7 Storkey et al. 2010

J Oper. Ocean.

Blockley et al. 2013

N Atlantic,

Indian

1/12º 50L 7

Shelf-

seas

FOAM

shelf

NWES2 7 km 33L 5 O’Dea et al. 2012

J. Oper. Ocean.

Arabian

Gulf

4 km 30L 5 Hyder et al. 2013

J. Oper. Ocean.

Seasonal GloSea5 Global 1/4º O

60 km A

75L 220 MacLachlan et al. 2013

Q J Roy Met Soc (sub)

SALINITY

WATER TEMPERATURE

CHLOROPHYLL (BIOLUMINESENCE) WATER CLARITY

MARINE MAMMAL DENSITY

AMBIENT NOISE

WIND WAVES

Global surface wave and Forecasting Ocean Assimilation Model (FOAM) predictions are used for day-to-day operational forecasting of the water column around the world for the Royal Navy. The detail they provide about oceanographic features gives tactical commanders the knowledge to make informed decisions about maritime safety and routeing, and to deploy ships, aircraft and submarines and their weapons and sensors effectively.

6. Support for oil industry

• Operated by Gassco, Norway; 8000 km sea-bed pipelines; riser platforms and processing facilities

• 6 receiving terminals in 4 countries;

• Transports 20% of gas used in Europe

• Will handle a larger production than the Norwegian oil industry within a few years

• Gas temperature (and therefore volume and pressure) are proportional to the sea bed temperature (SBT)

• Capacity planning and sales hence sensitive to SBT

• Leakage detection systems critically dependent on SBT

• Optimal use of additives to prevent formation of hydrate (=‘Gas Ice’), depends on SBT

8. Seasonal predictions 9. Short-range weather forecasts

Temperature 850hPa JJA[K]1.4

1.0

0.6T+ day 1 2 3 4 5

[mb]0

500

1000

5 10 15 20 25 30 35 (m)

Geopot height 5 day f/c JJA

Coupled 50 km (N256)Atmos 50 km (control)

Atmos 25 km (opnl) Atmos 40 km (opnl)

GloSea4 GloSea5

Grid GloSea4 GloSea5

Atmos @150 km @60 km

Ocean 1º 0.25º

• The Flood Forecast Centre in the Met Office Operations Centre is jointly staffed by the Environment Agency (EA) and the Met Office• It provides guidance to the EA regional centres on coastal flooding• Coastal flooding is still on the national risk register. • Low water forecasts are also important to warn ships entering port of the risks of grounding• Surge predictions are generated using 24 member ensembles of weather forecasts to drive 12 km barotropic surge model predictions for the NW European shelf • The figures show a low pressure event when the danger of flooding at Millport was first forecast 5 days ahead and consistently forecast 3 days ahead

• Development of higher

resolution (e.g. 2.2 km grid) atmos-wave-ocean surge predictions is highly desirable

Red = flood warning level (depends on tides)Green = short-range surge ensembleBlue = long-range surge ensembleCyan = deterministic hindcast Black = raw observations

Synoptic situation at 00Z on 8 Nov (at time of predicted coastal flooding)

Millport

• Cefas has a 24/7 multi-disciplinary Emergency Response Team for oil/chemical and other spills in UK waters

• It has roles in modelling (<1 hour response), monitoring & providing advice (e.g. on the use of dispersants)

• It uses various oil & chemical models & inputs (see figure) and receives operational wind, wave & current data feeds from the Met Office

• It has a good track record with Napoli, Ice Prince (wood spill pictured) and other incidents

• Predictions for deep water north-west of Shetland require additional challenging capabilities

Tidal Database

Met Office Winds

MyOcean Flow fields

Model selection and parameterisation

CefasSpill

ASA Oilmap

SintefOSCAR

Analysis and Interpretation