Ocean Wave Forecasting at ECMWF Jean-Raymond Bidlot Marine Aspects Section Predictability Division of the Research Department European Centre for Medium-range Weather Forecasts (E.C.M.W.F.) Reading, UK Slide 1
Ocean waves: We are dealing with wind generated waves at the surface of the oceans, from gentle to rough Slide 2
Ocean Waves Forcing: wind earthquake moon/sun Restoring: surface tension gravity Coriolis force 10.0 1.0 0.03 3x10-3 2x10-5 1x10-5 Frequency (Hz) Slide 3
What we are dealing with? Water surface elevation, η Wave Period, T Wave Length, λ Wave Height, H Slide 4
Wave Spectrum l The irregular water surface can be decomposed into (infinite) number of simple sinusoidal components with different frequencies (f) and propagation directions (θ ). l The distribution of wave energy among those components is called: wave spectrum, F(f,θ). Slide 5
Ocean Wave Modelling l Modern ocean wave prediction systems are based on statistical description of oceans waves (i.e. ensemble average of individual waves). l The sea state is described by the two-dimensional wave spectrum F(f,θ). Slide 6
Ocean Wave Modelling l For example, the mean variance of the sea surface elevation η due to waves is given by: 2 η F ( f, θ) dfdθ = l The mean energy associated with those waves is: energy l The statistical measure for wave height, called the significant wave height (H s ): H s = 4 2 = ρ g η w η 2 The term significant wave height is historical as this value appeared to be well correlated with visual estimates of wave height from experienced observers. It can be shown to correspond to the average 1/3 rd highest waves (H 1/3 ). Slide 7
Ocean Wave Modelling l The ocean wave modelling at ECMWF is based on the wave mode WAM cycle 4 (Komen et al. 1994), albeit with frequent improvements (Janssen 2007: ECMWF Tech. Memo 529.). l Products from different configurations of WAM are currently available at ECMWF. l Wave model wave page: http://www.ecmwf.int/products/forecasts/wavecharts/index.html#forecasts l General documentation: http://www.ecmwf.int/research/ifsdocs/cy36r1/index.html Slide 8
ECMWF Wave Model Configurations Global models l Global from 81 S to 90 N, including all inland seas. l Coupled to the atmospheric model (IFS) with feedback of the sea surface roughness change due to waves. l The interface between WAM and the IFS has been generalised to include air density and gustiness effects on wave growth and more recently neutral winds. l Data assimilation Jason-2 altimeter wave heights. 70 N 50 N 30 N 10 N 10 S 30 S 40 S 40 S 50 S 60 S 60 S 70 S Tuesday 14 March 2006 00UTC ECMWF Forecast t+36 VT: Wednesday 15 March 2006 12UTC Surface: significant wave height 20 E 60 N 60 N 40 N 40 N 20 N 20 N 0 0 20 S 20 S 20 E 40 E 40 E 60 E Forecast wave height on 15/03/2006 12UTC. Atmospheric model 60 E 80 E 80 E 100 E 100 E 120 E 120 E 140 E 140 E 160 E 160 E 180 180 160 W 160 W 140 W 140 W 120 W neutral wind wind gustiness air density 120 W 100 W roughness 100 W 80 W 80 W 60 W 60 W 40 W 40 W 20 W 20 W Wave model 70 N 50 N 30 N 10 N 10 S 30 S 50 S 70 S 10.29 9.75 9 8.25 7.5 6.75 6 5.25 4.5 3.75 3 2.25 1.5 0.75 0 Slide 9
ECMWF Wave Model Configurations Deterministic model l 28 km grid spacing. l 36 frequencies. l 36 directions. l Coupled to the TL1279 model. l Analysis every 6 hrs and 10 day forecasts from 0 and 12Z. Probabilistic forecasts (EPS) l 55 km grid spacing. l 30 25 frequencies *. l 24 12 directions *. l Coupled to TL639 TL319 model *. l (50+1) (10+5) day forecasts from 0 and 12Z (monthly once a week). * Change in resolutions after 10 days NB: also in seasonal forecast at lower resolutions Slide 10
ECMWF Wave Model Configurations Interim reanalysis (1979 to present) (as a follow-up to ERA40 (45 year reanalysis) l 1.0 x1.0. l 30 frequencies. l 24 directions. l Coupled to TL255 model l Production is ongoing. l Very satisfactory performance: Comparison with buoys: http://www.ecmwf.int/research/era/do/get/era-interim ECMWF Newsletters No. 110 (Winter 2006/07) & 111 (Spring 2007) http://www.ecmwf/publications/newsletters Slide 11 ERA-40 Operations ERA Interim
Wave Model Products The complete description of the sea state is given by the 2-D spectrum, however, it is a fairly large amount of data (e.g. 24x30 values at each grid point in the global model). It is therefore reduced to integrated quantities: Ø 1-D spectrum obtained by integrating the 2-D spectrum over all directions and/or over a frequency range. Wave model 2-D spectrum 1-D spectrum Slide 12
Wave Model Products When simple numbers are required, the following parameters are available: Ø The significant wave height (Hs). E( f ) peak Ø The peak period (period of the peak of the 1-D spectrum). area under spectrum = <η 2 > Ø Mean period(s) obtained from weighted integration of the 2-D spectrum. Ø Integrated mean direction. Ø Few others. f p (peak frequency) H s < f > (mean frequency) = 4 η 2 f T = 1 / f Complete list at: http://www.ecmwf.int/services/archive/d/parameters/order=/table=140/ Slide 13
Wave Model Products Use simple parameters: total wave height and mean propagation direction 10m winds and mean sea level pressure: Analysis : 14 February 2009, 00 UTC Wave height and mean direction: Analysis : 14 February 2009, 00 UTC Slide 14
Wave Model Products Wave periods: at the peak of the spectrum or in the mean PEAK PERIOD: Analysis : 14 February 2009, 00 UTC MEAN WAVE PERIOD: Analysis : 14 February 2009, 00 UTC Slide 15
Wave Model Products Wave height and mean direction: Analysis : 14 February 2009, 00 UTC PEAK PERIOD: Analysis : 14 February 2009, 00 UTC Slide 16
Wave Model Products Situation might be more complicated! 10m winds and mean sea level pressure: Analysis : 15 February 2009, 00 UTC Wave height and mean direction: Analysis : 15 February 2009, 00 UTC Slide 17
Wave Model Products Situation might be more complicated: Wave height and mean direction: Analysis : 15 February 2009, 00 UTC Slide 18
Wave Model Products A scheme is used to split the global wave fields into waves which are under the direct influence of the forcing wind, the so-called windsea or wind waves, and those waves that are no longer bound to the forcing wind, generally referred to as swell. Period and mean direction are also determined for these split fields. Wave height and windsea mean direction: Analysis : 15 February 2009, 00 UTC Wave height and swell mean direction: Analysis : 15 February 2009, 00 UTC Slide 19
Wave Model Products Windsea and swell: opposing sea Slide 20
Wave Model Products Windsea and swell: cross sea swell Slide 21
Wave Model Products on the web: Currently on our web: significant wave height and mean direction Slide 22
Wave Model Products on the web Currently on our web: mean wave period and direction Slide 23
e.g.: Wave Model Products on the web: For the Severe Weather Forecast Demonstration Projects (SWFDP) for South Africa, East Africa and the Pacific windsea and swell plots are also available: Swell wave height and direction Swell Mean period and direction Significant wave height and direction http://www.ecmwf.int/products/forecasts/d/charts/medium/special Slide 24
Wave Model Products: EPS From the EPS wave forecasts it is possible to derive probabilities for certain wave conditions. Tuesday 6 November 2001 12UTC ECMWF EPS Probability Forecast t+120 VT: Sunday 11 November 2001 12UTC Surface: significant wave height probability >8 25 W 20 W 15 W 10 W 5 W 0 5 E 10 E 15 E 20 E 25 E 54.47 25 E 50 50% MIKE HEIDRUN DRAUGEN 45 40 65 N 25 W 20 E 35 30 25 25% 20 60 N 55 N 14 15 E 15 10 Significant wave height (m) at Heidrun 12 5 20 W 15 W 10 W 5 W 0 5 E 10 E 06 Nov. 2001 12 UTC ECMWF EPS probability forecast t+120 Significant wave height above 8 m 10 8 6 0 8 m 4 ECMWF Newsletter 95 Autumn 2002 2 0 1 2 3 4 5 6 7 8 9 10 Slide 25 Forecast day
Wave Model Products on the web Currently on our web: probability for set thresholds (2m) Slide 26
Wave Model Products on the web Currently on our web: probability for set thresholds (4m) Slide 27
Wave Model Products on the web Currently on our web: probability for set thresholds (6m) Slide 28
A bit more compact: Wave EPSgram: Like normal EPSgram but for wind direction, wind speed, significant wave height, mean wave direction and mean period. South of Grindavik, Iceland Each octant is coloured based on the distribution of the significant wave height associated with each mean direction. The coloured areas correspond to the fractional number of ensemble members with wave height in the range specified by the coloured ruler. Ocean waves at ECMWF Slide 29
EPS Wave Model Products on the web: SWFDP Pacific wave EPSgrams: Set of locations where wave EPSgram are available Slide 30
EPS Wave Model Products on the web: SWFDP East Africa wave EPSgrams: Set of locations where wave EPSgram are available Slide 31
Individual Waves, Significant Wave Height, H s, Maximum Individual Wave Height, H max, and Freak Wave Individual waves etc. H max H s = H 1/3 If H max > 2 H s freak wave event Slide 32
50 N Wave Model Products: Extreme Waves We have recently introduced a new parameter to estimate the height of the highest individual wave (H max ) one can expect: Friday 7 March 2008 00UTC ECMWF Forecast t+84 VT: Monday 10 March 2008 12UTC Surface: Significant wave height (Exp: 0001 ) 20 W 0 Hs 60 N 60 N 40 N 40 N 20 W March 10 th, 2008, 12UTC Forecasts fields from Friday 7 th March, 2008, 0 UTC 0 60 N 60 N 50 N Friday 7 March 2008 00UTC ECMWF Forecast t+84 VT: Monday 10 March 2008 12UTC Surface: (Exp: 0001 ) 20 W 0 40 N 40 N See ECMWF Tech Memo 288 for derivation and discussion http://www.ecmwf.int/publications/library/do/references/list/14 50 N 17.44 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 20 W Slide 33 Expected H max in 3 hour records 0 50 N 34.79 34 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
Large swell reaching la Réunion, May 2007: the aftermath Lagon de Trou d Eau Coastal flooding Fisherman who was swept out to sea rescued by helicopter. Saint-Leu. Photo Ludovic Lai-Yu Hotel hit hard Sea front destroyed All pictures courtesy of CLICANOO (www.clicanoo.com) Slide 34
Large Swell Reaching la Réunion: The Model Sig. wave height and mean propagation direction valid on May 12, 2007, 18UTC. The model has nicely predicted significant wave height in excess of 5m in the evening of the 12 th. analysis forecast t+30 7 H1/3 (m) Hs model analysis (m) 6 5 4 forecast t+54 forecast t+78 3 2 1 0 11/05/2007 12:00 12/05/2007 00:00 12/05/2007 12:00 13/05/2007 00:00 13/05/2007 12:00 14/05/2007 00:00 14/05/2007 12:00 Slide 35
Case study: long swell affecting the western Pacific in December 2008 Coastal flooding linked to high tide, barometric surge and long swell, e.g. Nukutoa on Takuu Atoll: Pictures courtesy of John Hunter Antarctic Climate & Ecosystems Hobart, Tasmania, Australia Slide 36
Case study: long swell affecting the western Pacific in December 2008 10m winds (arrows) Sfc pressure (contours) Sig. wave height (shading) Dec. 5 th, 2008, 0UTC Dec. 6 th, 2008, 0UTC Dec. 8 th, 2008, 0UTC Dec. 8 th, 2008, 0UTC Slide 37
Case study: long swell affecting the western Pacific in December 2008 Dec. 9 th, 2008, 12 UTC Windsea wave height and direction Swell wave height and direction Long swell reaching South! Windsea Mean period and direction Swell Mean period and direction Slide 38
Case study: long swell affecting the western Pacific in December 2008 Wave condition north of New Ireland (PNG07) for 2008, based on ECMWF analysis! Conditions were exceptional! Sig. wave height Dec. 10 th, 2008, 0 UTC Slide 39
Case study: long swell affecting the western Pacific in December 2008: EPS Wave Epsgram from Dec. 6 th, 2008 at PNG07 Wave Epsgram from Dec. 8 th, 2008 at PNG07 Slide 40
Questions? Slide 41 Duck, North Carolina