Scattering from Surf Zone Waves

Transcription

1 Scattering from Surf Zone Waves Experimental Results and Modeling Patricio, Merrick Haller, Bill Plant Associate Professor Universidad Técnica Federico Santa María Chile

2 Motivation Wave Breaking 2

3 Background on this Work Multi Remote SENsing Surfzone Observations: MR-SENSO 6 weeks at Duck, NC, in 8 3 SENSORS GEOREFERENCED and SYNCHRONIZED ARGUS III Electro Optical Data Marine Radar X-Band, HH, log receiver PPI RiverRad X-Band, HH/VV Coherent Staring Identification of Breaking Stage et al, 2011, IEEE TGRS Microwave Scattering Analysis Scattering Model et al, 2014, JGR-O 3

4 RiverRad look Directions 4

5 Phase 1: Identification of Breaking Target Detection using coupled Constant False Alarm approach 600 Video Run9 600 Marine Radar Cross shore distance, m Cross shore distance, m Alongshore distance, m Alongshore distance, m Video Intensity, I NRCS, db 5

6 Phase 1: Identification of Breaking Target Detection using coupled Constant False Alarm approach 600 Video Run9 600 Marine Radar Cross shore distance, m Cross shore distance, m Alongshore distance, m Alongshore distance, m Non breaking Video Intensity, I

7 Phase 1: Identification of Breaking Target Detection using coupled Constant False Alarm approach 600 Video Run9 600 Marine Radar Cross shore distance, m Cross shore distance, m Alongshore distance, m Alongshore distance, m Remnant Foam Assumed not dynamically relevant Video Intensity, I

8 Phase 1: Identification of Breaking Target Detection using coupled Constant False Alarm approach 600 Video Run9 600 Marine Radar Cross shore distance, m Cross shore distance, m Alongshore distance, m Alongshore distance, m Video Intensity, I 150 Steepening Wave

9 Phase 1: Identification of Breaking Target Detection using coupled Constant False Alarm approach 600 Video Run9 600 Marine Radar Cross shore distance, m Cross shore distance, m Alongshore distance, m Alongshore distance, m Breaking Wave Video Intensity, I

10 Well, how does it work? Breaking Steep Foam 10

11 Phase 2: Microwave Scattering Goal: Identify the backscattering characteristics of breaking waves Improve our understanding of the scattering mechanisms urnal of Geophysical Research: Oceans 10.2/2014J We can compare the RiverRad data for all wave stages NRCS Doppler Spectra Doppler offset RiverRad settings 9.36 GHz 2 min 128 range 7.5 m 0.53 s sampling 31 two-minute runs were used 11

12 Enviromental conditions Journal of Geophysical Research: Oceans 10.2/2014JC Figure 2. Wave and environmental conditions during data collection. Vertical gray lines indicate the RiverRad runs used in the analysis. (a) Significant wave height, H m0 ; (b) peak wave period, T p ; (c) mean wave direction relative to the FRF x axis; (d) wind speed, U; and (e) wind direction relative to the FRF x axis. 12

13 Detection Area coverage Journal of Geophysical Research: Oceans 10.2/2014JC00988 Figure 2. Wave and environmental conditions during data collection. Vertical gray lines indicate the RiverRad runs used in the analysis. ( Significant wave height, H m0 ; (b) peak wave period, T p ; (c) mean wave direction relative to the FRF x axis; (d) wind speed, U; and (e) wind direction relative to the FRF x axis. 13

14 Wave by Wave NRCS Evolution: Energetic Cond. 14

15 Wave by Wave NRCS Evolution: Low Energy Cond. Figure 7. Space-time evolution of a shoaling wave for mild wave conditions at / 2 (collection 24868). Same key as Figure 5. N ET AL. VC American Geophysical Union. All Rights Reserved. 15

16 Journal of Geophysical Research: Oceans Ensemble Results 10.2/2014JC

17 Journal of Geophysical Research: Oceans Environmental Dependency (or 10.2/2014JC lack of) 2. Median NRCS of breaking waves classified by environmental parameters magnitude. Rows correspond to significant wave height Hm 0, mean wave direction, wind speed, and 17

18 Journal of Geophysical Research: Oceans Doppler Spectra Journal of Geophysical 10.2/2014JC S M Research:U Oceans Figure 3. Sample time-space maps for collection (a) Marine radar calibrated sure showing the RiverRad look direction as dotted line. Arcs correspond to radial d (green), and remnant foam (cyan) areas. Gray scale corresponds to the microwave Time is given in seconds rela corresponds to the horizonta range). Figure 13. Mean Doppler spectra, db. (a) Collection at HH, (b) HH, and (c) VV forground collection 24868, where white line indicates range detected. (d g) Plots of mean Doppler spectra at ranges indicated by white arrows in plots (a c). Blue lines correspond to VV and red line Black arrows indicate the Bragg frequency. Figure 3 shows sample timeriverrad r0hh and r0vv, alon almost all of the individual w images are very similar. How than r0hh (compare Figure 3 would be expected within C Mean Doppler spectra, db. (a) Collection at HH, (b) HH, and (c) VV for collection 24868, where white line indicates ranges where at least one breaking event was d g) Plots of mean Doppler spectra at ranges indicated by white arrows in plots (a c). Blue lines correspond to VV and red lines to HH. Thicker lines correspond to collection ck arrows indicate the Bragg frequency Breaking Detection Figure 4 shows sample resul demarcating wave breaking,! ET AL. C American G V CATALAN space maps clearly show the pres ence of propagating waves, espefigure 14. Doppler spectra variability at r m, for collections where / < 15!. Shaded 18

19 Doppler Offset Figure 15. Time-space maps of surface radial velocities (m/s) derived from the Doppler offset, for collection (a) HH and (b) VV. Thin contours denote locations identified as breaking, and dashed contours correspond to the individual wave identified in section High Energy Low Energy Figure 16. Time-space maps of surface radial velocities (m/s) derived from the Doppler offset, for collection Same key as Figure

20 Comparison with Linear Theory Linear Phase Speed Orbital Velocity 20

21 Summary so far Breaking waves source of large backscattered power, several db above non breaking waves Maximum NRCS seems to have a limiting value Weak dependency on viewing geometry (θg,φ) Subtle polarization dependency No discernible dependency on environmental parameters Large polarization ratio but not necessarily greater than 1 Few breaking events suffice to broaden the Doppler spectrum Peak speeds well correlated with speed of the carrier wave 21

22 How does it compare with other data? Lambertian Sactterer Decay 22

23 Scattering model Previous research assumed surface scattering e.g. Bragg, specular, etc. Roller has a complex structure and its morphology is usually disregarded (e.g. Coakley et al., 1) Proposition: Model based on Volumetric Scattering Roller as a two-phase medium (air + water) Scattering from water droplets (Mie regime) 23

24 EM model Roller is a collection of water droplets scattering in the Mie regime Multiple scattering interactions among particles Interactions using the Quasi-Crystalline Approximation Allows estimation of extinction, absorption and scattering coefficients Wave propagation through media using Dense Media Radiative Transfer Theory Interactions with boundaries Tsang et al., 7 Model parameters Droplet size cm Seawater permitivitty Volume fraction (max 50 %) Stickiness parameter z=0 z= d θ io θ so ε0 ε1 ε2 24

25 Results Thin but finite layers of droplets suffice to explain scattering levels Non Sticky Sticky Albedo shows that small volume fractions and relative small particles O(1)cm can yield large reflectivity Weak dependency on grazing angle Subtle polarization dependency. rnal of Geophysical Research: Oceans 10.2/2014JC Albedo Volume fraction, % 0 Albedo Model Results: Backscattering coe 1 2 Diameter, cm Albedo cient Volume fraction, % Volume fraction, % Conceptual model Diameter, cm Diameter, cm! O A Catalán (OSU) Microwave Scattering from Surf Zone Waves No 0 (HH) (VV) d=1cm f v =5% d=1cm f v =20% 20 d=0.1cm f v =5% σ 0, db d=0.1cm f v =20% 40 Figure 18. Grazing angle dependency of NRCS from various data sets. Line corresponds to grazing angle dependency of a Lambertian scattering surface. 60 turbulent front face that have a range of speeds about their spatial mean, which is the local wave celerity [e.g., Fuchs et al., 1999; Coakley et al., 1; Farquharson et al., 5]. All locations where breaking waves were identified, this shift and broadening were identified in the Doppler spectra. Herein, we have shown that for surf zone waves the presence of even just one breaking event (during a 2 min record) is sufficient to funda Grazing Angle Grazing Angle 25

26 Model summary Volume scattering can be relevant Small volume fractions suffice (upper roller layers) Complex formulation but requires physical parameters only Can we measure a droplet distribution to validate this? Model output shows good agreement with measured data Median NRCS Grazing angle dependency Small polarization ratios Roller travels with the wave at its phase speed. 26

27 27