Surface Pressure Fluctuations Produced by the Total Solar Eclipse of 1 August 2008 Marty J. 1 , Dalaudier F. 2 , Ponceau D. 3 , Blanc E. 3 , Munkhuu U. 4 1 CTBTO, Vienna, Austria 2 LATMOS, IPSL, UPMC ‐ Paris 6, France 3 CEA, DIF, DAM, Arpajon, France 4 RCAG, Ulaanbaatar, Mongolia
Bow-Wave Model Theory Wake Moon shadow UV IR Chimonas (1970), Fritts and Luo (1993) Mesosphere Pressure height [km] O 3 Stratosphere Troposphere H 2 O Heating rate profile [K.day ‐ 1 ]
Previous Observations Possible detection 16 h False detection No detection 8 h Seykora et al. (1985) Wave Period 4 h Jones et al. (1992) 2 h Anderson et al. (1972) Venkatachari et al. (1982) 1 h Goodwin et Hobson (1978) 20 min Jones et Bogart (1975) Beckman et Clucas (1973) Farges et al. (2003) Schödel et al. (1973) 10 min Anderson et Keefer (1975) Jones (1999) Jones (1976) 5000 10000 50 100 500 1000 Distance from central line of the eclipse [km]
An Unsolved Problem Observations < ‐ > Modeling 100 Seykora et al. (1985) Amplitude [Pa] 10 Farges et al. (2003) Model – Chimonas (1970) 1 Goodwin et Hobson (1978) 0.1 4 h 10 h 10 min 20 min 1 h 2 h Period Can the passage of a solar eclipse produce detectable surface pressure fluctuations? • Are the existing models realistic enough ? • Are the pressure fluctuations too small to be detected at the ground ? • Are the measurement systems adapted to the expected signals ? • Is the recording duration too short to differentiate the waves produced by the solar eclipse from other waves ? Julien Marty Ph.D. defense 12 October 2010
New Linear Spectral Numerical Model Model description • Linearization of fundamental fluid equations without source term (Fritts, 2003) : �. exp � �� � �� � �� � �� � � � � � � � �, �, � � � � � � �. � with and 2� � � � � �� � 1 � 4� � 1 � Fully compressible dispersion relation for gravity waves � 2 � � � � � � � � � � � � � � � � � � � � � � � � 1 � � � � � � � � � � � 1 with and � � � � � Buoyancy frequency • Linear spectral model (Fourier Transform) ‐ > Constant dispersion relation parameters Density scale height Sound velocity Coriolis parameter • Propagation of normalized pressure fluctuation (amplitude do not depend on altitude) � � � � � ��/�� φ′ �, �, �, � � �′ �, �, �, � Normalized Exponential growth with altitude
New Linear Spectral Numerical Model Source effects • Free propagation : wavefield can be directly estimated at any time • Impulsive source : instantaneously modify the solution • Continuous source : source terms are time ‐ discretized and handled as impulsive modifications of the solution Model benefits Source spatial and temporal evolution Linearity Light (programming and calculation time) Wave field phase Wave perturbation variables (polarization equations) Ground reflection J. Marty and F. Dalaudier, 2010: Linear spectral numerical model for internal gravity wave propagation. J. Atmos. Sci., 67:1632–1642.
Model Comparison - Stratospheric Source z= 80km, w’ [cm.s ‐ 1 ], Δ w’ = 0.1 z= 80km, w’ [cm.s ‐ 1 ], Δ w’ = 0.05 Stationary solutions Shape and intensity of the wavefield fairly similar Asymptotic solution reached after 40 h → Solar eclipses only last 2 ‐ 3 h at Earth’s surface Fritts and Luo (1993) Marty and Dalaudier (2010) z= 80km, w’ [cm.s ‐ 1 ] z= 80km, w’ [cm.s ‐ 1 ] Solutions in good agreement despite the use of constant atmospheric parameters Marty and Dalaudier (2010) Eckermann et al. (2007)
Surface Pressure Fluctuation P [Pa] Eckermann et al. (2007) Marty and Dalaudier (2010) with stratospheric source P [Pa] The main surface pressure perturbation cannot be explained by the stratospheric cooling The tropospheric cooling is likely to be the predominant source Marty and Dalaudier (2010) with tropospheric source
Total Solar Eclipse of 1 August 2008 (200 km) (70 km) (320 km) (1060 km) (1650 km)
Expected Wave Periods Possible detection 16 h False detection Expected period for the tropospheric source No detection 8 h Expected period for the stratospheric source Seykora et al. (1985) Wave Period 4 h Jones et al. (1992) 2 h Anderson et al. (1972) Venkatachari et al. (1982) 1 h Goodwin et Hobson (1978) 20 min Jones et Bogart (1975) Beckman et Clucas (1973) Farges et al. (2003) Schödel et al. (1973) 10 min Anderson et Keefer (1975) Jones (1999) Jones (1976) 5000 10000 50 100 500 1000 Distance from central line of the eclipse [km]
Mongolia 2008 (M2008) Measurement Campaign Objectives • Study the response of infrasound measurement systems in the gravity wave frequency band • Characterize gravity events and identify sources (network operational for 20 days) • Detect the pressure fluctuations produced by the passage of the 1 August 2008 solar eclipse (determine source, propagation mode and spatial and temporal evolution) External Sonic 50 km temperature anemometer 8 m
Validity of Pressure Measurements in GW Frequency Band ? Acoustic cut ‐ off period The sensor self ‐ noise, thermal Coriolis Buoyancy Cut ‐ off period of susceptibility and transfer period period pressure sensors function do not significantly IMS band affect the pressure signals in the ~17 h 8 min 5 min 100 s 50 s 0.25 s 0.05 s gravity wave frequency band Period Gravity waves Infrasonic waves J. Marty, D. Ponceau, and F. Dalaudier, 2010: Using the International Monitoring System infrasound network to study gravity waves, Geophys. Res. Lett., 37, L19802. • In case M2008 experiment, sensor less B1 TS = 2.8 Pa.K ‐ 1 25 dB protected from temperature variations > 30 dB • MicrobarometerTS < 10 Pa.K ‐ 1 • Thermal susceptibility (TS) can affect pressure signals in the low ‐ frequency part of the GW band Need to correct pressure signals from influence of temperature
Thermal Susceptibility Correction Excellent correction from Evaluation of the correction temperature effects Raw signal Corrected with Corrected with manufacturer TS re ‐ evaluated TS Manufacturer TS estimated linearly from 2 measures at ‐ 25 ° C and 60 ° C RSD: 17.4 % RSD : 6.1 % RSD: 2.3 % Experimental protocol to improve TS estimation T : 14 → 30 ° C (steps = 2 ° C) T ≈ 19 ° C Calibrated temperature sensors Microbarometers used Temperature sensors used during M2008 campaign during M2008 campaign
Pressure Signal Analysis total eclipse time Diurnal + semidiurnal oscillation (Superimposed epoch method) eclipse signal obtained from model (tropospheric source) Cubic spline smoothing (meteorological changes)
Time-Frequency Analysis • Wavelet transform (Morlet) applied on a Fourier transform normalized by a factor corresponding to the spectrum slope • Detection of two wave packets with similar time ‐ frequency characteristics to those obtained from model total eclipse time Ballard et al. (1969) Quiroz et Henry (1973) Signal obtained from the model Randhawa (1974) (stratospheric source x 20) Schmidlin et Olsen (1984) Signal obtained from the model (tropospheric source x 3)
IMS Data Analysis
Comparison With Previous Observations
Conclusion • Development of a new linear spectral numerical model to simulate the propagation of internal gravity waves • Field experiment with broadband, high dynamic range and calibrated measurement systems • Characterization of sensor response and susceptibility to environment • Use of specific and original signal processing techniques • Validation and use of IMS Infrasound data down to 24 h ‐ period A Unique Result • The tropospheric cooling is the predominant source of surface pressure fluctuations after the passage of a solar eclipse • Worldwide detection of wave packets after the passage of the 1 August 2008 solar eclipse with similar time ‐ frequency characteristics as those obtained from modeling J. Marty, F. Dalaudier, D. Ponceau, E. Blanc, U. Munkhuu, 2013: Surface Pressure Fluctuations Produced by the Total Solar Eclipse of 1 August 2008. J. Atmos. Sci., 70, 809–823. J. Marty, D. Ponceau, and F. Dalaudier, 2010: Using the International Monitoring System infrasound network to study gravity waves, Geophys. Res. Lett., 37, L19802. J. Marty and F. Dalaudier, 2010: Linear spectral numerical model for internal gravity wave propagation. J. Atmos. Sci., 67:1632–1642.
Thanks for your attention !
Recommend
More recommend
