HIGH RESOLUTION WIND AND WAVE MEASUREMENTS FROM TerraSAR-X IN COMPARISON TO MARINE FORECAST

Similar documents
First studies with the high-resolution coupled wave current model CWAM and other aspects of the project Sea State Monitor

FUSION OF SYNTHETIC APERTURE RADAR AND OPTICAL SATELLITE DATA FOR UNDERWATER TOPOGRAPHY ESTIMATION IN COASTAL AREAS

Polar Activities at DLR Maritime Security Lab Bremen in the Projects EISTAK and EMS

Using Satellite Spectral Wave Data for Wave Energy Resource Characterization

WaMoS II Wave Monitoring System

Development of SAR-Derived Ocean Surface Winds at NOAA/NESDIS

EVALUATION OF ENVISAT ASAR WAVE MODE RETRIEVAL ALGORITHMS FOR SEA-STATE FORECASTING AND WAVE CLIMATE ASSESSMENT

Sentinel-1A Ocean Level-2 Products Validation Strategy

Geophysical Model Functions for the Retrieval of Ocean Surface Winds

On the assimilation of SAR wave spectra of S-1A in the wave model MFWAM

Validation of 12.5 km Resolution Coastal Winds. Barry Vanhoff, COAS/OSU Funding by NASA/NOAA

Offshore wind resource mapping in Europe from satellites

Ocean Wave Parameters Retrieval from TerraSAR-X Images Validated against Buoy Measurements and Model Results

WAVE FORECASTING FOR OFFSHORE WIND FARMS

JCOMM Technical Workshop on Wave Measurements from Buoys

830 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 48, NO. 2, FEBRUARY /$ IEEE

ENVISAT WIND AND WAVE PRODUCTS: MONITORING, VALIDATION AND ASSIMILATION

THE QUALITY OF THE ASCAT 12.5 KM WIND PRODUCT

Jackie May* Mark Bourassa. * Current affilitation: QinetiQ-NA

Determination of Nearshore Wave Conditions and Bathymetry from X-Band Radar Systems

SEA-LEVEL AND SEA-STATE MEASUREMENTS WITH RADAR LEVEL SENSORS. Dr. Ulrich Barjenbruch 1 and Jens Wilhelmi 2

Comparison of data and model predictions of current, wave and radar cross-section modulation by seabed sand waves

The BASE-platform project: Deriving the bathymetry from combined satellite data

Scales of Atmospheric Motion Scale Length Scale (m) Time Scale (sec) Systems/Importance Molecular (neglected)

Determination Of Nearshore Wave Conditions And Bathymetry From X-Band Radar Systems

Reprocessed QuikSCAT (V04) Wind Vectors with Ku-2011 Geophysical Model Function

Wind Direction Analysis over the Ocean using SAR Imagery

PRELIMINARY STUDY ON DEVELOPING AN L-BAND WIND RETRIEVAL MODEL FUNCTION USING ALOS/PALSAR

GLOBAL VALIDATION AND ASSIMILATION OF ENVISAT ASAR WAVE MODE SPECTRA

Air-Sea Interaction Spar Buoy Systems

An Atlas of Oceanic Internal Solitary Waves (May 2002) by Global Ocean Associates Prepared for the Office of Naval Research - Code 322PO

IDENTIFICATION OF WIND SEA AND SWELL EVENTS AND SWELL EVENTS PARAMETERIZATION OFF WEST AFRICA. K. Agbéko KPOGO-NUWOKLO

SPATIAL AND TEMPORAL VARIATIONS OF INTERNAL WAVES IN THE NORTHERN SOUTH CHINA SEA

Dynamic validation of Globwave SAR wave spectra data using an observation-based swell model. R. Husson and F. Collard

RIVET Satellite Remote Sensing and Small Scale Wave Process Analysis

Flow separation and lee-waves in the marine atmosphere

Open Ocean Storm Waves in the Arctic

Application of Simulating WAves Nearshore (SWAN) model for wave simulation in Gulf of Thailand

7 YEARS METEOMAST AMRUMBANK WEST

Spatial characteristics of ocean surface waves

Available online at ScienceDirect. Procedia Engineering 116 (2015 )

Coastal Wave Studies FY13 Summary Report

Appendix D: SWAN Wave Modelling

Wave Energy Atlas in Vietnam

SENSOR SYNERGY OF ACTIVE AND PASSIVE MICROWAVE INSTRUMENTS FOR OBSERVATIONS OF MARINE SURFACE WINDS

Wave energy converter effects on wave and sediment circulation

RAPid Image Exploitation Resource (RAPIER ) Ship Detection System

ERS WAVE MISSION REPROCESSING- QC SUPPORT ENVISAT MISSION EXTENSION SUPPORT

G. Meadows, H. Purcell and L. Meadows University of Michigan

High resolution wind fields over the Black Sea derived from Envisat ASAR data using an advanced wind retrieval algorithm

Coastal Sediment Transport Modeling Ocean Beach & San Francisco Bight, CA

Validation of 12.5 km and Super-High Resolution (2-5 km)

Nearshore Wind-Wave Forecasting at the Oregon Coast. Gabriel García, H. Tuba Özkan-Haller, Peter Ruggiero November 16, 2011

High-Resolution Measurement-Based Phase-Resolved Prediction of Ocean Wavefields

Metocean criteria for fatigue assessment. Rafael V. Schiller 5th COPEDI Seminar, Oct 8th 2014.

Field Evaluation of the Wave Module for NDBC s New Self-Contained Ocean Observing Payload (SCOOP) on Modified NDBC Hulls

Marine Renewables Industry Association. Marine Renewables Industry: Requirements for Oceanographic Measurements, Data Processing and Modelling

Offshore wind mapping Mediterranean area using SAR

DUXBURY WAVE MODELING STUDY

WAVE PERIOD FORECASTING AND HINDCASTING INVESTIGATIONS FOR THE IMPROVEMENT OF

Synthetic Aperture Radar imaging of Polar Lows

Coastal Wave Energy Dissipation: Observations and Modeling

Directional Wave Spectra from Video Images Data and SWAN Model. Keywords: Directional wave spectra; SWAN; video images; pixels

E. Agu, M. Kasperski Ruhr-University Bochum Department of Civil and Environmental Engineering Sciences

An Atlas of Oceanic Internal Solitary Waves (February 2004) by Global Ocean Associates Prepared for Office of Naval Research Code 322 PO

NUMERICAL STUDY OF WIND-GENERATED WAVES IN THE RED SEA

Boost Your Skills with On-Site Courses Tailored to Your Needs

PHYSICAL AND NUMERICAL MODELLING OF WAVE FIELD IN FRONT OF THE CONTAINER TERMINAL PEAR - PORT OF RIJEKA (ADRIATIC SEA)

Analyses of Scatterometer and SAR Data at the University of Hamburg

Generalized Wave-Ray Approach for Propagation on a Sphere and Its Application to Swell Prediction

Modelling and Simulation of Environmental Disturbances

COMPARISON OF CONTEMPORANEOUS WAVE MEASUREMENTS WITH A SAAB WAVERADAR REX AND A DATAWELL DIRECTIONAL WAVERIDER BUOY

Lifting satellite winds from 10 m to hub-height

Implementation of the new French operational coastal wave forecasting system and application to a wavecurrent interaction study

Wave Transformation Modeling with Bottom Friction Applied to the Southeast Oahu Reefs. Mary A. Cialone and Jane McKee Smith

Available online at ScienceDirect. Energy Procedia 53 (2014 )

SINGULAR WAVES, PROPAGATION AND PROGNOSIS. H. Günther, W. Rosenthal

Ocean Waves and Surf Forecasting: Wave Climate and Forecasting

Monitoring Conditions Offshore with Satellites

MOTUS Wave Buoys. Powered By the Aanderaa MOTUS Directional Wave Sensor

RapidScat wind validation report

Advancements in scatterometer wind processing

Ocean Wave Forecasting

CHAPTER 6 DISCUSSION ON WAVE PREDICTION METHODS

IMPROVED OIL SLICK IDENTIFICATION USING CMOD5 MODEL FOR WIND SPEED EVALUATION ON SAR IMAGES

Introduction EU-Norsewind

Wave Transformation, Prediction, and Analysis at Kaumalapau Harbor, Lanai, Hawaii

FULL-WAVEFORM INVERSION

Dynamics and variability of surface wind speed and divergence over mid-latitude ocean fronts

Wave forecasting at ECMWF

An IOOS Operational Wave Observation Plan Supported by NOAA IOOS Program & USACE

A study of advection of short wind waves by long waves from surface slope images

Pathways Interns: Annika O Dea, Ian Conery, Andrea Albright

WAVE CLIMATE IN DEEP AND COASTAL WATERS: NUMERICAL MODELS

Wind Power. Kevin Clifford METR 112 April 19, 2011

Review of Equivalent Neutral Winds and Stress

Wave Generation. Chapter Wave Generation

Swell (wind) fields obtained from Satellite Observation CLS Christian Ortega - Mission blue Journées Annuelles des Hydrocarbures 2013

An Atlas of Oceanic Internal Solitary Waves (February 2004) by Global Ocean Associates Prepared for Office of Naval Research Code 322 PO

from a decade of CCD temperature data

Transcription:

SAR Maritime Applications German Aerospace Center (DLR) Remote Sensing Technology Institute Maritime Security Lab HIGH RESOLUTION WIND AND WAVE MEASUREMENTS FROM TerraSAR-X IN COMPARISON TO MARINE FORECAST Susanne Lehner, Andrey Pleskachevsky German Aerospace Center (DLR) Jens Kieser Thomas Bruns German Weather Service (DWD) Approach Sea State Processor Examples

Recently: Wave Experiment NOVA SE University 2015-11-05 11:26 UTC NOAA WAVEWATCH III 2km subscene TS-X H S mean value 0.99m 0.92m 1.03m NOAA MIA01 11:00 UTC H S =0.92m (26.25N 79.50W) NOAA buoys

www.dlr.de/fl Folie 3 > Vortrag > Autor Dokumentname > Datum Introduction: Sea state in coastal areas ) German Bight: high temporal and spatial variability of sea state, interaction waves/bathymetry/currents high activity: shipping and building offshore constructions AIS messages mapped on TerraSAR-X image German Bight German Bight wind parks condition: wave height Hs < 1.3m condition: wave height<1.3m accurate forecast necessary wave rider buoy FiNO-1 Dec. 2013 1

Coastal Wave Model (CWAM) NEW! increment longitude: 0.01389 increment latitude: 0.00833 coupled with the ocean circulation model (HBM) that provides current and water depth European Wave Model (EWAM) CONVENTIONAL Wind input from COSMO-EU, spectral resolution: 36 directions, 30 frequencies increment longitude: 0.1 increment latitude: 0.05 4 First studies with the high-resolution wave current model CWAM, Jens.Kieser@dwd.de, Banff, 2013-10-31

www.dlr.de/fl Folie 5 > Vortrag > Autor Dokumentname > Datum Introduction: SEEGANGSMONITOR-Project Remote sensing for validation of forecast model DWD-German Meteorological Service: Sea sate Forecast - improvement for coastal regions DLR: Remote Sensing - Sea State Data for Validations Development of a new coastal wave model 900m resolution, interactively coupled with BSH circulation model 2

1. Approach Satellites & Data Methods & Objectives

Satellites: X-band SAR TerraSAR-X and TanDEM-X - Radar signal penetrates clouds - No sun light Helgoland is necessary Helgoland German Bight 35m Resolution 2010 Cuxhaven Wilhelmshaven Bremerhaven 10km Bremen SpotLight Stripmap ScanSAR Wide ScanSAR 10km 30km 100km 210km 1m Resolution 3m Resolution 16m Resolution 35m Resolution 3

Developments: Wind and Waves from TerraSAR-X Products for NRT Algorithm and processing requirements - raster analysis, fast, robust, automatic TerraSAR-X wind and waves from TerraSAR-X TanDEM-X Synergie NRT NZ 4

2. Sea Sate Processor Principe Deep water application

XWAVE empirical algorithm: GMF principle and structure sub-scene TS-X Stripmap m = I( i, j) mean mean signal intensity I signal modulation m I S = FFT (m) image spectra S sub scene E = k = min k = max S( k, θ ) dkdθ filtering, L min, L max Local Wind (XMOD) XWAVE GMF function Parameters: H s, L p,t p 2D image spectrum wave number domain 100 m 50 m 1D spectrum frequency domain buoy spectrum integrated image spectrum

XWAVE empirical algorithm: Deep water version for open ocean regions Empirical Geophysical Model Function Tuning H s = x1 E( 1.0 + cos( α)) + x2 XWAVE (first guess) Hs Hs 1m 4m 5m 8m = a 1 = a 4 E tan( θ ) + a U E tan( θ ) + a U 2 5 10 XMOD 10 XMOD + a 3 + a 6 cos( α) + a 7 XWAVE-1 deep water, sea state does not visible change spatially NOAA Buoys used for tuning Examples 5

3. Sea State Processor: GMF Adaption for Coastal Application sea state safe estimation in raster mode Data: Acquisition concept for German Bight Filtering artefacts before analysis GMF Improvements Examples of tuning / explanation

www.dlr.de/fl Folie 13 > Vortrag > Autor Dokumentname > Datum Date acquisition concept: TerraSAR-X Strips covering DWD model domain 6 Buoys in DWD Model domain German Bight North Sea German Bight Example: TS-X Scene with 5 images 30km One TerraSAR-X StripMap scene strip cowers 30km x 50-300km To date: 51 TerraSAR-X Scenes (overflights/events/days 2013-2015) with 188 Images and 81 buoy collocations 6

Coastal applications: artifacts impacts spectral analysis Artefacts in SAR image impact spectra Task 1 - removing artefacts before analysis Sand banks Wave breaking Ships, Buoys, Wind parks Current fronts, ship wakes GMF is applicable for pure sea state case only: Pre-filtering of images is necessary for raster analysis Without pre-filtering Integrated energy and Hs can > 10 times overestimate real value wave breaking ship pure sea state wind park front dry sand bank

www.dlr.de/fl Folie 15 > Vortrag > Autor Dokumentname > Datum SAR imaging of ocean surface waves: a composition of different schemas for complete Hs Sea State Imaging SAR Image 1. Swell, weak wind wave height 0m-3m Bragg waves, III modulations Tends to real waves 2. Developed wind sea moderate wind wave height 1m-5m Bragg waves, III modulations + individual scatterers (step crests of wind waves, white capping scatterers) smearing, defocusing 3. Wave breaking Underwater hindrance or by wind > 15m/s wave height 0m-5m flying scatterers, high turbulent surface in breaking zone Strong streaks 4. Low wind sea, waves interact with currents and bathymetry wave height 0m-1m Noise due to amount of small non- systemized wave-targets, waves imaged hardly visible costal algorithm: more focus on last 3 types clutter effect 7

www.dlr.de/fl Folie 16 > Vortrag > Autor Dokumentname > Datum GMF improvement for coastal applications GMF Deep water Tuning Step 1: spatial, using model data to get Ansatz function Step 2: collocated, buoy data to correct coefficients new GMF_Coastal physical positive term +Spectral energy +wind impact a 1 a 5 - coefficients (constants) tuned B 1 B 5 - functions of spectral parameters fishing outliers different origin -short distortions -long distortions -extra large distortions B 1 - noise scaling of total energy E IS a 2 B 2 - wind impact according to the original XWAVE a 3 B 3 and a 4 B 4 - corrections for eliminating the impact of short (e.g. wave breaking induced) and long (e.g. wind streaks) structures in the SAR image. a 5 B 5 - correction for outliers produced by extra-large structures like sand banks or ship wakes 8

www.dlr.de/fl Folie 17 > Vortrag > Autor Dokumentname > Datum Integrating GMF into Sea State Processor: Whole automatic sea state processing for NRT Step-1: sub-scene pre-filtering (remove image intensity outliers like ships, buoys etc. based on local intensity statistics) Step-2: calculation of XMOD-2 wind Step-3: spectral analysis of subscene (FFT, integration and spectral parameters e.g. noise level) Step-4: new XWAVE-2 GMF (SSP core) Step-5: checking of results using wind speed and integrated spectral parameters Example of a NRT delivery per e-mail txt file: lon, lat, Hs 12

Statistics: Comparison spatial (model) and at locations (buoys) TS-X analysis with raster step of 3km 1 StripMap 30kmx50km 10x15 =150 subscenes comparison TS-X/wave model (spatial ) 21 overflights with 64 TS-X images Depth>5m comparison TS-X/Buoy (at location) 51 overflights with 188 TS-X images collocation +/-20min, subscene center +/- 5km from buoy location color shows wind speed XMOD-2 for each location analysed XWAVE_C TerraSAR-X, m XWAVE_C TerraSAR-X, m Hs model, m Hs buoy, m 13

Statistics: Comparison with buoys: all collocated data 6 Buoys in German Bight comparison SI=20% RMSE=0.25 New data since Dec. 2014 Data 2013-2014 14

4. Examples Different sea state at the same location Different wind condition in German Bight 15

www.dlr.de/fl > Vortrag > Autor Folie 21 Different condition at the same location in North Sea: Dokumentname > Datum Low wind and swell with improved GMF Low wind condition: wave-like strictures are hardly visible H S can be estimated based on noise properties of spectra and wind information. wind sea waves + interaction with currents Buoy XWAVE Wind 7m/s Hs 0.50m 0.46m Buoy ELBE 1km swell & wind sea Buoy XWAVE Wind 19m/s Hs 5.80m 5.72m 1km 16

Different wind conditions in German Bight with TerraSAR-X data using new XWAVE_C (coastal) Light breeze 1-3 m/s Beaufort 2 Moderate breeze 5 8 m/s Beaufort 4 Fresh/strong breeze 8 12 m/s Beaufort 5/6 2015-01-20 05:50 2015-02-05 05:59 2014-04-12 05:50

Summary High Resolution Spatial Distribution of Hs and Peak Parameters Tested over North Sea for Offshore Windfarming Scatter Hs < 20 % Available in NRT (<15 min after acquisition at station) Offer: To be tested over other areas 15

Thank you for your attention! Storm acquired: 10-12-2014 sequence of 6 TerraSAR-X images covers German Bight (300km!)