Derivation of trace gas information combining differential optical absorption spectroscopy with radiative transfer modelling [Elektronische Ressource] / presented by Christoph v. Friedeburg

Christoph V. Friedeburg

Sujets

• Physik
• Astronomie

Informations

• Publié par: ruprecht-karls-universitat_heidelberg
• Publié le: 01 janvier 2003
• Nombre de lectures: 20
• Langue: Deutsch
• Poids de l'ouvrage: 7 Mo

Extrait

Dissertation
submitted to the

combined Faculties for the Natural Sciences and for Mathematics
of the Ruperto-Carola University of Heidelberg, Germany

for the degree of
Doctor of Natural Sciences
































presented by

Diplom-Physicist Christoph v. Friedeburg
born in Frankfurt/Main

rdOral examination: July 23 2003
Derivation of Trace Gas Information

combining

Differential Optical Absorption Spectroscopy with

Radiative Transfer Modelling




































Referees: Prof. Dr. Ulrich Platt


Prof. Dr. Bernd Jähne
Zusammenfassung

Ermittelung von Spurengasinformation durch Kombination der Differentiellen
Optischen Absorptionsspektroskopie mit Strahlungstransportmodellierung

Die Differentielle Optische Absorptionsspektroskopie (DOAS) ist als
Fernerkundungsmethode zur Messung atmosphärischer Komponenten etabliert. Zur
Ermittelung quantitativer Information über Spurengasverteilungen ist eine
Kombination neuer Meßtechniken, z.b. Multi-Axis DOAS, mit realitätsnaher
Strahlungstransportsimulation erforderlich.
In dieser Arbeit wird das dreidimensionale sphärische Strahlungstransportmodell
TRACY auf der Basis der Monte Carlo Methode zur Anwendungsreife gebracht und
für die Interpretation mehrerer eigener und fremder Messungen verschiedener DOAS
Plattformen angewandt. Im Gegensatz zu bisherigen analytischen Modellen sind keine
Vereinfachungen der Geometrie oder der physikalischen Vorgänge erforderlich.
Zusätzliche Ausgabeparameter, die sich auf Eigenschaften der Lichtstreuung
beziehen, gestatten ein Verständnis der spektroskopischen Modellergebnisse und
Aussagen über die Empfindlichkeit der Meßgeometrien für bestimmte
Höhenprofilformen.
Speziell untersucht wurde die entscheidende Rolle der Aerosole im
Strahlungstransport. Sie können Messungen derart beeinflussen, daß jedwede
geometrische Näherung der Meßempfindlichkeit zu falschen Aussagen führt.
Geeignete Meßgrößenkombinationen, inkl. O -Absorption, lassen in Verbindung mit 4
Modellierung der entsprechenden Szenarien jedoch Rückschlüsse auf die Verteilung
der Aerosole zu. Sie und Kenntnis über Quellen und Senken eines Spurenstoffes
erlaubt die Projektion mehrerer Unbekannter auf wenige Meßgrößen, was eine
Ermittelung quantitativer Verteilungen ermöglicht.


Abstract

Derivation of Trace Gas Information combining Differential Optical Absorption
Spectroscopy with Radiative Transfer Modelling

The Differential Optical Absorption Spectroscopy (DOAS) is an established remote
sensing technique for atmospheric constituent probing. To derive quantitative
distribution data of trace gas distributions it is necessary to combine novel
measurement techniques like Multi-Axis DOAS with realistic radiative transfer
modelling. In this thesis three three-dimensional spherical Monte Carlo based
radiative transfer model TRACY is brought to operational status and employed for the
interpretation of several own and existing measurements with different DOAS
platforms. In contrast to established models, no approximations and simplifications of
geometry or physical processes are needed. Additional model output parameters
describing the scattering of the light allow for the understanding of the spectroscopic
model results and for conclusions on the geometrie’s sensitivity to certain trace gas
distribution shapes. The decisive role of the aerosols in radiative transfer was
investigated. They were found to influence measurement to an extent rendering any
geometric approximation of the measurement sensitivity invalid. Combination of
measured quantities including O absorption with modelling allow for conclusions on 4
their abundance. They, in conjunction with knowledge on sources and sinks of the
considered species, allow for the projection of unknown parameters onto measurable
quantities, which facilitates the derivation of quantitative distributions.
Contents i
Contents
1 Introduction 1

2 Atmosphere 3
2.1 Vertical structure 3
2.1.1 Hydrostatic equation 3
2.1.2 Molecular-viscous layer and Prandtl’s layer 3
2.1.3 Ekman layer and the planetary mixing layer 4
2.1.4 The troposphere 4
2.1.5 The stratosphere 4
2.1.6 The mesosphere, thermosphere and exosphere 5
2.1.7 The ionosphere and the heterosphere 5
2.2 Composition 7
2.2.1 Trace gases 7
2.2.2 Aerosols and clouds 8

13 3. Differential Optical Absorption Spectroscopy
3.1 Principle 15
3.2 Apparatus-related influences 17
3.2.1 Convolution with the Instrument Function 17
3.2.2 Discretization 18
3.3 Numerical evaluation 19
3.3.1 Reference spectra 20
3.3.2 Numerical fitting 21
3.3.3 Error calculation 22
3.3.4 Influence of noise 23
3.3.5 Residual structure 23
3.3.6 Mutual reference spectra compensation 24
3.4 Air Mass Factor 24
3.4.1 Classical Air Mass Factor 24
3.4.2 Box AMF 25

4. Atmospheric scattering 29
4.1 Rayleigh scattering 29
4.1.1 Molecular dipole radiation 29
4.1.2 Rayleigh cross section 31
4.1.3 Rayleigh phase function 32
4.1.4 Polarization 33
4.1.5 Raman scattering 34
4.2 Aerosol scattering 35
4.2.1 Mie scattering 35 ii Contents
4.2.2 Mie cross section 35
4.2.3 Mie phase function 36
4.2.4 Aerosol absorption 38
4.3 Surface scattering 38
4.4 Cloud scattering 39
4.5 Principal mechanisms of light scattering in the atmosphere 40

5. Radiative transfer equation 45
5.1 Radiometric quantities 45
5.1.1 Photon distribution function 45
5.1.2 Flux and intensity 46
5.1.3 Radiance 46
5.1.4 Irradiance 46
5.1.5 Extraterrestrial solar irradiance 48
5.2 Radiative transfer equation 48
5.3 Analytical solution approaches to the RTE 50
5.3.1 Two Stream Approximation 50
5.3.2 Discrete Ordinate Method 51
5.3.3 Finite Difference Method 52
5.3.4 Other Methods 53
5.3.5 Raytracing Methods 53
5.4 The Monte Carlo approach to the RTE 53
5.4.1 Relation between MC and the RTE 54
5.4.2 Backward Monte Carlo technique 57

6. The 3D Monte Carlo RTM „Tracy“ 59
6.1 Previous RTM used for DOAS 59
6.2 Motivation and requirements 59
6.3 The backward Monte Carlo implementation 60
6.4 Geometrical structure 62
6.4.1 Coordinate system and „voxel“ structure 62
6.4.2 Locations and directions 63
6.5 Photon unit concept 64
6.6 Principal algorithms 65
6.6.1 Decision between multiple scatterers 65
6.6.2 Path points 67
6.6.3 Raytracer 67
6.6.4 Result calculation 69
6.7 Error estimate 70
6.8 Influence of refraction 73
6.9 Runtime estimate 73

7. Validation 75
7.1 Zenith-sky NO2 and BrO with diurnal variation 75
7.2 Off-Axis HCHO 78
7.3 Off-Axis NO 81 2

8. Application to absolute radiometry 85 Contents iii
8.1 Calibrated scanning spectro radiometer 86
8.1.1 Incoupling optics 86
8.1.2 Spectrograph 86
8.1.3 Photo multiplier 87
8.1.4 Energy supply, controlling and detection electronics 87
8.1.5 Control software 87
8.1.6 Calibration 88
8.1.7 Calculation of photolysis frequencies 89
8.2 Conversion of “Tracy” output to radiances 91
8.3 Modeled radiances 92
8.4 Addition of polarization 93
8.5. Polarization characteristics of the SR with a polarization filter 95

97 9. Application to ground based DOAS
9.1 Sequential MAX-DOAS of BrO Alert/Canada 2000 97
9.1.1 Role of tropospheric BrO and campaign objectives 97
9.1.2 Instrument 100
9.1.3 Measured SCD 102
9.1.4 AMF investigation 103
9.1.5 Conclusions on the vertical BrO profile 117
9.2 Simultaneous MAX-DOAS of NO2 Heidelberg/Germany 2002 118
9.2.1 Role of NO and campaign objectives 118 2
9.2.2 Instrumental setup 120
9.2.3 Evaluation procedure and measured NO SCD 125 2
9.3 Simultaneous MAX-DOAS of HCHO and O , Milano/Italy 2002 137 4
9.3.1 FORMAT campaign and the Chemistry of HCHO 137
9.3.2 Instrumental Setup 138
9.3.3 Comparison between HCHO measurements and model SCD 139
9.3.4 Comparison between O measuremodel SCD 140 4

10. Aircraft measurements of NO 143 2
10.1 Instrument 143
10.2 Measured SCD