First continuous middle-atmospheric wind profile measurements with a ground-based microwave Doppler-spectro-radiometer

Transcription

1 First continuous middle-atmospheric wind profile measurements with a ground-based microwave Doppler-spectro-radiometer Rolf Rüfenacht 1,N.Kämpfer 1, A. Murk 1, P. Eriksson 2,S.A.Buehler 3 1 Institute of Applied Physics, University of Bern, Switzerland 2 Chalmers University of Technology, Sweden 3 Luleå University of Technology, Sweden, now at University of Hamburg, Germany URSI F Symposium 2013

2 Why should we measure middle-atmospheric wind? Wind and temperature are key variables to describe the atmosphere Understand the dynamics of the middle atmosphere Understand the coupling between the lower and upper atmosphere Interpretation of transport processes

3 What wind pattern do we expect?

4 Who is measuring middle-atmospheric wind?

5 Who is measuring middle-atmospheric wind?

6 Who is measuring middle-atmospheric wind?

7 The measurement principle Doppler effect: ν = v LOS c ν 0 where v LOS = v cos with : elevation angle Frequency ν 0 = GHz from the rotational transition of the ozone molecule Ozone emission line is pressure broadened altitude information

8 Atmospheric emission spectrum in the microwave region

9 Influence of wind on the ozone emission spectrum 130 no wind Brightness temperature [K] Frequency [GHz]

10 Influence of wind on the ozone emission spectrum Brightness temperature [K] no wind wind km Frequency [GHz]

11 Influence of wind on the ozone emission spectrum Brightness temperature [K] no wind wind km wind km Frequency [GHz]

12 Influence of wind on the ozone emission spectrum Brightness temperature [K] no wind wind km wind km wind km Frequency [GHz]

13 The instrument

14 The instrument

15 Receiver noise with or without RF electronics without LNA and filter: with LNA and filter: T N = 880 K DSB ( 1760 K SSB) T N = 740 K SSB

16 Receiver noise with or without RF electronics without LNA and filter: with LNA and filter: T N = 880 K DSB ( 1760 K SSB) T N = 740 K SSB 2.4 times better signal to noise ratio in spectra 2.4 times smaller error for wind measurements 5.6 times less integration time for the same wind error

17 Measured frequency stability Frequency instabilities during one measurement cycle (1 minute) Allan std influence on IF accuracy LO for RF mixer 7 Hz (@ 72 GHz) 14 Hz IF-converter LO 0.2 Hz (@ 3.7 GHz) 0.2 Hz Total frequency instability < 15 khz ˆ 0.03 m/s Long term frequency drifts are larger but do not influence the accuracy of the wind measurements

18 Zonal wind observed in Provence 2012/13 Comparison with model data 10 2 Zonal wind WIRA Wind speed [m/s] 27 Nov 17 Dec 06 Jan 26 Jan 15 Feb 07 Mar 27 Mar 16 Apr 06 May Zonal wind ECMWF Wind speed [m/s] 27 Nov 17 Dec 06 Jan 26 Jan 15 Feb 07 Mar 27 Mar 16 Apr 06 May 100

19 Scatter plots WIRA measurements vs. ECMWF model Zonal wind Provence campaign 150 Altitude: km 150 Altitude: km WIRA [m/s] WIRA [m/s] Altitude: km 150 Altitude: km WIRA [m/s] WIRA [m/s] ECMWF [m/s] ECMWF [m/s]

20 First measurements from the tropics Measurements at Maïdo observatory, La Reunion Zonal wind WIRA Wind speed [m/s] 01 Sep 03 Sep 05 Sep 07 Sep 09 Sep 11 Sep 13 Sep 15 Sep 17 Sep Meridional wind WIRA Wind speed [m/s] 01 Sep 03 Sep 05 Sep 07 Sep 09 Sep 11 Sep 13 Sep 15 Sep 17 Sep 50

21 Conclusions Microwave Doppler wind radiometry is a unique and powerful tool to continuously measure middle-atmospheric wind profiles

22 Conclusions Microwave Doppler wind radiometry is a unique and powerful tool to continuously measure middle-atmospheric wind profiles Key characteristics of the technique: + sensitive to the altitude range between 30 and 70 km + operational 24 hours per day (independent of night/day) + relatively unaffected by cloud cover + transportable instrument + radiometers can be operated remotely with little maintenance needed weak signal receiver noise limits sensitivity

23

24 Averaging kernels for wind ^x = x a + A(x x a ) A = ^x x Rows of wind averaging kernel matrix

25 Error of zonal wind observed in Provence 2012/ Wind error [m/s] 5 27 Nov 17 Dec 06 Jan 26 Jan 15 Feb 07 Mar 27 Mar 16 Apr 06 May 0

26 Meridional wind observed in Provence 2012/13 Comparison with model data 10 2 Meidional wind WIRA Wind speed [m/s] 27 Nov 17 Dec 06 Jan 26 Jan 15 Feb 07 Mar 27 Mar 16 Apr 06 May Meridional wind ECMWF Wind speed [m/s] 27 Nov 17 Dec 06 Jan 26 Jan 15 Feb 07 Mar 27 Mar 16 Apr 06 May 100

27 Zonal wind observed in Bern

28 Zonal wind observed in Sodankylä

29 Average information content Wind profile retrieval double sideband single sideband effective rank signal degrees of freedom Shannon information cont. 3.0 bits 7.8 bits Ozone profile retrieval double sideband single sideband effective rank signal degrees of freedom Shannon information cont bits 17.6 bits

30 Ozone profile retrieval 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east 10 1 GROMOS Aura MLS WIRA west WIRA east O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm] O3 VMR [ppm]

31 Comparison with GROMOS and MLS WIRA west GROMOS WIRA east GROMOS WIRA west MLS WIRA east MLS MLS GROMOS Difference ozone profiles [ppm] WIRA west GROMOS WIRA east GROMOS WIRA west MLS WIRA east MLS MLS GROMOS Relative difference ozone profiles

32 Averaging kernels for O3 retrieval Altitude [km] Amplitude of averaging kernels

33 Atmospheric pressure Altitude [km]

34 Shape of the ozone emission line Planck brightness temperature [K] Elevation Frequency [GHz]

35 Sharpness of the emission line Temperature span [K] Elevation angle [ ]