Spaceborne synthetic aperture radar SAR applications in marine

Spaceborne synthetic aperture radar (SAR) applications in marine meteorology Xiaofeng Li NOAA/NESDIS Acknowledgment Zheng Weizhong and William Pichel, NOAA/NESDIS Xiaofeng Yang, Institute of Remote Sensing Application, CAS NOAA Gilmore Creek Ground Station, Alaska

Outline Motivation SAR Observation and WRF model simulation of Marine Atmospheric Boundary Layer (MABL) phenomena 1) Atmospheric Gravity Waves 2) Atmospheric Vortex Streets 3) Katabatic Winds 4) Storm pattern Summary

Motivation SAR Provides high resolution (tens to hundreds m) all day/night synoptic observation of ocean surface. Community weather model now provides time series of sub-km actual atmospheric circulation fields. Synergy of the two to understand the actual atmospheric conditions that lead To the generation and evolution of these phenomena in the MABL.

SAR applications in Marine meteorology Ø atmospheric lee waves atmospheric disturbances atmospheric boundary layer rolls atmospheric fronts katabatic winds, gap winds Hurricanes island wakes cold air outbreaks Storms atmospheric vortex streets Land breeze Cloud line Ø … Ø Ø Ø

1. Atmospheric Gravity Wave Observations Upstream Waves: Waves propagates against the mean flow Lee wave has two types of wave patterns: 1. the transverse wave type where the wave crests are nearly perpendicular to the wind direction 2. (2) the diverging wave type where the wave crests are orientated outwards from the center of the wake.

1. Atmospheric Gravity Wave Observations Upstream waves

1. Atmospheric Gravity Wave Observations Upstream waves (2) Wind variations are between 8 (wave trough) and 16 m/s (wave crest).

1. Atmospheric Gravity Wave Observations Upstream waves 3 06/06/2001 @COPYRIGHT CSA The number of wave crests increases from 3 to 8 in 4. 5 hrs.

1. Atmospheric Gravity Wave Observations Upstream AGW can be understood by solving the Fe. Kd. V Model with Froude number f=1. 1 Li, et al. 2004, JGR

1. Atmospheric Gravity Wave Observations Lee Wave (transverse waves) W in d. D ire cti o n A RADARSAT Scan. SAR wide B SAR Western Alaska Image center (60 N, 145 W) 03: 10: 05 GMT, June 12, 2001. (© Canadian Space Agency, 2001)

1. Atmospheric Gravity Wave Observations W in d Di re cti on Lee Wave (Diverging waves) RADARSAT Scan. SAR wide image shows the island lee wave pattern in the Gulf of Alaska. • Image center: (55 N, 154 W) • Time 16: 29: 42 UTC, June 27, 1997. (© CSA, 1997)

1. Atmospheric Gravity Wave Observations W ind Di rec tio n Lee Wave (diverging waves, Example 2) Constant lee wave phase lines Chunchuzov, Vachon, and Li, Remote Sensing of Environment, 74, 343 -361, 2000.

1. Atmospheric Gravity Wave Observations Lee Wave 3

1. Atmospheric Gravity Wave Observations Lee wave simulation with WRF Weather Research and Forecasting NCAR Community model Two-way nested domains: D 1: 9 km 280 x 251 D 2: 3 km 310 x 301 D 3: 1 km 271 x 262 Vertical levels: 27 Model parameterization: • Rapid Radiative Transfer Model (RRTM) longwave radiation • Dudhia shortwave radiation • Yonsei University (YSU) planetary boundary layer scheme • WRF Single Moment 3‐Class (WSM) cloudmicrophysics, • Kain‐Fritsch (new Eta) convective parameterization employed only in coarse and medium domains, • Monin‐Obukhov surface‐layer scheme • Noah land‐surface model. Boundary conditions are specified by linearly interpolating the NCEP 6 hourly final analyses (FNL) at a resolution of 1° × 1° degree. The digital elevation of 30 s USGS topography data.

1. Atmospheric Gravity Wave Observations WRF Simulation results at the imaging time Vertical velocity Horizontal velocity

1. Atmospheric Gravity Wave Observations Lee Wave 3 Standing Waves MODIS true color image acquired about 8. 5 h prior to the Envisat ASAR pass in Google Earth.

1. Atmospheric Gravity Wave Observations Lee Wave 4 Not trapped in the MABL

1. Atmospheric Gravity Wave Observations Lee Wave 4 Comparison of WRF simulated surface wind speed with the independent SAR measurements

1. Atmospheric Gravity Wave Observations AGW Summary Ø AGM patterns can be identified on SAR image. Ø Most of the waves generated by flow over bump observed by SAR are on the lee side of the bump, and has two types of wave patterns. Ø The terrain-forced stmospheric lee waves can be simulated using the WRF model. Model reveals evolution of the AGW. Ø Upstream waves happen when Froude number is close to 1. They are elevation waves. They are explained well with Fe. KDV model.

2. Atmospheric Vortex Streets (AVS) Atmospheric vortex-street patterns on the lee side of a flow obstacle l double row of counter rotating vortex pairs shedding alternately near the edge of the obstacle l No observation in the atmosphere prior to the satellite age due to AVS scale (100 -400 km): too small to be delineated by a synoptic observation network and too large to be observed by a single station l h/a is the basic property of an atmospheric vortex streets l Uo is the wind speed, Ue is the vortex shedding velocity

2. Atmospheric Vortex Streets (AVS) Ø NASA space shuttle picture l cloud pattern (1) Ø NASA space shuttle picture l cloud pattern (2)

2. Atmospheric Vortex Streets (AVS) Landsat-7 (Defelice et al. Vol. 81, No. 5, 1047 -1048, BAMS, 2000)-cloud pattern Aircraft photo (Ferrier et al. Vol. 17, No. 1, 1 -8, International J. of Remote Sensing, 1996)- sediment pattern SAR Wind Pattern. IEEE/TGARSS Feb. 2004

ATMOSPHERIC VORTEX STREETS OBSERVED ON REMOTE SENSING IMAGES (3) Ø RADARSAT synthetic aperture radar (SAR) image (© CSA, 1999) Li, et al. 2002, GRL Ø RADARSAT synthetic aperture radar (SAR) image (© CSA, 1999)

2. Atmospheric Vortex Streets (AVS) SAR observation of AVS Li, et al. 2008, Sensors

2. Atmospheric Vortex Streets (AVS) WRF Simulation of AVS Li, et al. 2008, Sensors

2. Atmospheric Vortex Streets (AVS) WRF Simulation of AVS • Vortex shedding rate and propagation velocity • Reynolds Number (50 -250) • Frude Number (0. 15 -0. 4) • Vorticity and energy dissipation

3. Katabatic Wind katabatic wind is a gravitational air flow that descends from a high-elevation mountain down slope to lower elevation. Happens night and early morning hours in the winter season when the air mass over the mountain top becomes colder and heavier due to fast radiation cooling and when the land-sea temperature gradient is large.

3. Katabatic Wind hourly coastal horizontal wind speed

3. Katabatic Wind The black solid line is the topography; the red dash line represents the horizontal wind speed simulated by the model, the blue line represents the sea surface wind derived from SAR measurements, and the green dash line represents the near surface descending wind speed

3. Katabatic Wind • Summary: • The wind associated with the katabatic wind varies between 5 and 8 m/s • The katabatic wind finger extends up to 20 km offshore. • The 3 -layer nested weather model simulation captures the finger-like katabatic wind pattern observed in the SAR image. • Katabatic wind is the strongest at 14: 00 UTC when the land-sea temperature gradient is the largest. • The low-level model wind speed is very close to that derived from the SAR image. • The SAR image shows a more detailed finger-like katabatic wind structure than that shown in the • model simulation.

4 Storm and Precipitation

4 Storm and Precipitation Storm Summary: WRF can capture wind and rain pattern in the storm Heavy rain in the hurricanes damps NRCS. Moderate rain has less effect in C-band NRCS than wind does in the regular storm case

5 Summary We have set up WRF model to simulate SAR observed MABL phenomena WRF model captures the basic characteristics AGV, AVS, Katabatic and Storm The 1 -km resolution WRF model gives detailed structures of the phenomena This leads to the understanding of generation and evolution of these phenomena Synergy of SAR and WFR provides great opportunity for detailed study of MABL Phenomena