MMSE approach for estimating radar Doppler spectrum

Accurate radar observation of precipitation at low altitudes is important for many practical uses. Rainfall rate calculated from radar reflectivity factor or dual-polarization radar parameters at a lower altitude indicates better agreement with a ground-based devise such as a rain gauge and a disdrometer [1], [2], [3], and is expected as a valid input to a flood monitoring system [4]. In airport operation, Doppler velocities at altitudes especially less than 500 m contribute to enhance safety of landing and take-off aircrafts [5]. Since radar measurements at low elevations are heavily affected by ground clutter, many methods for filtering ground clutter have been proposed and utilized as a standard function of a weather radar [6], [7]. These filtering methods essentially filter around-zero velocity components, resulting in non-zero components corresponding to precipitation can be extracted. However, in a case of observing a low elevation which frequently includes strong ground clutter, the non-zero components of precipitation is not detectable due to clutter sidelobes, as described in Panels (a) and (b) of Figure 1.

In this paper, an Doppler spectrum estimation method via minimum mean square error (MMSE) to obtain the practically useful precipitation signals at a low elevation by suppressing clutter sidelobes is proposed. A conceptual figure of the proposed method is shown in Figure 1 (c), which indicates that non-zero components of precipitation becomes detectable by reducing sidelobes due to around-zero components of ground clutter. A significant feature of this research is that the proposed method can be applied instead of the Fourier method which is the most conventional method to estimate Doppler spectrum. Therefore, any existing ground clutter filtering methods can be applied to a sidelobe-less Doppler spectrum calculated by the proposed method. In the sidelobe-less Doppler spectrum, ground clutter appears sharper than one by the Fourier method, and is easier to be eliminated by any existing filtering methods. In addition, more accurate radar reflectivity factor, radial velocity, and velocity dispersion can be calculated from the sidelobe-less precipitation components. In Figure 2, an example of simulation result is shown. Compared with the simulation truth, the proposed method outputs a Doppler spectrum with less sidelobes and higher accuracy than the conventional Fourier method. In the presentation, methodology, simulation results, and evaluation results of the proposed method will be explained.

 

 

Figure 1: Concept of sidelobe reduction on a radar Doppler spectrum. a) a case in which precipitation components can easily be extracted; b) precipitation components are obscured by sidelobes of strong ground clutter; and c) the clutter sidelobes are suppressed by the proposed method.

 

 

Figure 2: An example of simulation results

 

[1] K. S. Gage, C. R. Williams, W. L. Clark, P. E. Johnston, and D. A. Carter, 2002: Profiler Contributions to Tropical Rainfall Measuring Mission (TRMM) Ground Validation Field Campaigns. J. Atmos. Oceanic Technol., 19, 843–863.

[2] C. R. Williams, A. B. White, K. S. Gage, and F. M. Ralph, 2007: Vertical Structure of Precipitation and Related Microphysics Observed by NOAA Profilers and TRMM during NAME 2004. J. Climate, 20, 1693–1712.

[3] E. Yoshikawa, S. Kida, S. Yoshida, T. Morimoto, T. Ushio, and Zen Kawasaki, 2010: Vertical structure of raindrop size distribution in lower atmospheric boundary layer. Geophysical Research Letters, 37, L20802.

[4] V. Chandrasekar, Y. Wang, and H. Chen, 2012: The CASA quantitative precipitation estimation system: a five year validation study. Nat. Hazards Earth Syst. Sci., 12, 2811–2820.

[5] N. Matayoshi, T. Iijima, E. Yoshikawa, and T. Ushio, Development of Low-level Turbulence Advisory System for Aircraft Operation. 29th Congress of the International Council of the Aeronautical Sciences, St. Petersburg, Russia, September 2014.

[6] R. J. Doviak and D. S. Zrnic, Doppler Radar and Weather Observations. San Diego, CA: Academic, 1993,

[7] A. D. Siggia, and R. E. Passarelli, Jr., Gaussian Model Adaptive Processing (GMAP) for Improved Ground Clutter Cancellation and Moment Calculation. ERAD 2004.