AIRS Department Home Page

   William J. Plant  
      Principal Research Scientist  
      AIRS Department  
      Applied Physics Laboratory  
      University of Washington  

   Office of Naval Research  


The most promising method to monitor the generation of internal waves in the region of the Luzon Strait between the two ridges routinely is remote sensing. A variety of satellite or aircraft sensors, visible and microwave, are capable of providing imagery of the area to observe internal waves. Visible sensors cannot operate at night or during many types of weather, limitations that are not present for microwave sensors. Therefore microwave sensors are a very attractive means of routinely monitoring internal wave generation.

Images showing internal wave signatures are available for both satellite and aircraft sensors but are mostly concentrated in the SCS, west of Luzon Strait. Information on the direction of propagation, speed, and distance between waves can immediately be obtained from such imagery and has shown itself to be very useful for the prediction of internal waves and for studying their generating mechanisms [1, 2]. However, at present it is not possible to determine reliably the displacement height or currents associated with the internal waves from the imagery. The project to be proposed will attempt to shed light on the relationship between microwave image intensity and the strength of internal waves in order to overcome this limitation.


Our work of the past three years in NLIWI and PhilEx has shown that present theories of this relationship do not work well for strong internal waves. These theories view the process responsible for internal wave surface signatures in microwave images to be modulation processes. The internal waves modulate long surface waves which in turn modulate the short wind waves responsible for microwave backscatter. This process, however, produces polarization properties of the backscattered microwaves which many times are different from those observed, especially when the microwaves strike the surface at relatively large incidence angles. This strongly suggests that the use of these theories will not be able to produce accurate estimates of the strength of internal waves in many important situations.

In fact, microwave backscatter from the ocean is always produced by a combination of
scattering from free wind waves and from the roughness generated by breaking waves. The balance between these two on the open ocean is such that effects of breaking waves can largely be neglected for moderate to low incidence angles. For vertical polarization, they may be neglected even at larger incidence angles. However, current gradients on the ocean surface, such as those produce by internal waves, increase the number of breaking waves, making their effect on microwave backscatter much greater. At present it is not known what fraction of internal wave microwave signatures is due to breaking waves; this undoubtedly depends on wind speed, polarization, and incidence angle. We believe that multi-bounce backscatter from breaking waves can explain the polarization features we have observed in microwave signatures of IWs. We will attempt to model this so that the variation in polarization properties with incidence angle and sea state can be understood. We will then apply these models to attempt to derive internal wave intensities in the Luzon Strait using a shipboard microwave Doppler radar and will compare the results with other instruments carried by the ship and with models of internal wave generation.


The following two figures illustrate the nature of the problem. Figure 1 shows airborne images of normalized radar cross section from our CORAR system taken off the New Jersey coast as part of NLIWI. The left image is HH polarization while the right is VV. The images were taken simultaneously. The incidence angle varies from 40o at the bottom of the images to 70o at the top. It is clear that near 40o, HH and VV cross sections in the intense regions of the internal wave are nearly equal. Above about 50o, however, HH cross sections are clearly larger than VV in the intense regions produced by the IWs.

Figure 2 shows this phenomenon carried to an extreme. This shipboard image from the ocean around the Philippine Islands shows that features probably associated with internal waves show up very strongly in the HH image but are hardly visible in the VV one. Neither the near equality of HH and VV cross sections in intense parts of the IW at 40o nor the fact that HH cross sections are larger than VV at higher incidence angles can be explained by present theories of internal wave microwave signatures.

Figure 1.  Images of internal waves on the New Jersey shelf obtained with the airborne CORAR. Left: HH polarization, Right: VV polarization.

Figure 2.  Images of current gradients near the Philippine Islands obtained with a coherent radar from the R/V Melville. Top: HH polarization, Bottom VV polarization.


1.  Zhao, Z. and M. H. Alford, Source and propagation of internal solitary waves in the northeastern South China Sea, J. Geophys. Res., 111, doi:10.1029/2006JC003644, 2006.

2.  Jackson, C., Internal wave detection using the moderate resolution imaging spectroradiometer (MODIS), J. Geophys. Res., 112, doi: 10.1029/2007JC004220, 2007.