Current Projects: Rain Effects on Microwave Backscatter from Water

INTRODUCTION

The effect of rain on microwave backscatter has been measured both from the ocean surface and from river surfaces.  Results show that rain increases backscatter from the ocean surface at Ku-band (14 GHz) for incidence angles greater than about 25 degrees.  For lower incidence angles, down to about 15 degrees, rain has little effect on the level of backscatter.  Modeling of the damping of short gravity waves and the generation of ring waves suggests that rain may significantly suppress backscatter from the sea at Ku-band for still lower incidence angles.  An auxiliary result of our rain measurements is that rain rate can be rather well determined from a continuous wave microwave system by observing the backscatter from rain drops.

THE KWAJALEIN EXPERIMENT

During the Kwajalein Experiment of 1999, we mounted a CW, Ku-band microwave system on the bow mast of the NOAA R/V Ronald H. Brown to measure the effects of rainfall on backscatter from the ocean.  Figure 1 shows our setup on the ship.  The microwave system was coherent and dual polarized so that Doppler spectra at both HH and VV polarizations could be recorded simultaneously.

Figure 1.  The APL/UW coherent, CW, Ku-band microwave system mounted on the bow mast of the R/V Ronald H. Brown during the Kwajalein Experiment of 1999.

The ship was at sea in the equatorial Pacific during the months of July and August, 1999.  The ship was primarily stationary with its bow directed mostly into the wind.  Our system was mounted to look 50 degrees to port of the bow so data were taken with a range of azimuth angles from about 0 to 90 degrees.  Figure 2 shows cross sections obtained from the data after removal of backscatter from raindrops.  Cross sections have been averaged over all azimuth angles.  Clearly rain increases the cross section of the sea surface at Ku-band and incidence angles above 30 degrees.  At lower incidence angles, our data suggest that cross sections may become slightly reduced.

Figure 2.  The normalized radar cross section, σ_o, versus incidence angle for no rain (solid line), and rain rates of 0.63 - 5 mm/hr, 5-20 mm/hr, and > 20 mm/hr.  Error bars represent 95% confidence limits.

When plotted against wind speed, our data indicated that cross sections during rain are nearly independent of wind speed for low to moderate winds.  This is illustrated in Figure 3.

Figure 3.  The dependence of σ_o on wind speed at 10 meters for the higher scatterometer incidence angles.  Error bars represent 95% confidence limits.  The mean cross sections during rain have been offset slightly in Un for ease of viewing.  Red circles - 0.63 < R < 5 mm/hr; Blue asterisks - 5 < R < 20 mm/hr; Black diamonds - R > 20 mm/hr.

One further result from these measurements is the observation that rain rate has a good correlation with the backscatter from raindrops in these measurements.  Figure 4 shows this effect.  Additional information on these experiments may be found in Contreras et.al., 2003.

Figure 4.  Fraction of backscatter due to rain divided by illuminated volume versus rain rate.  Error bars are 95% confidence limits.

MODELING RAIN EFFECTS ON SEA RETURN