Quantifying carbon processes of the terrestrial biophere in a global atmospheric inversion based on atmospheric mixing ratio, remote sensing and meteorological data [Elektronische Ressource] / Bakr Badawy. Gutachter: Christiane Schmullius ; Martin Heimann

Informations

• Publié par: friedrich-schiller-universitat_jena
• Publié le: 01 janvier 2011
• Nombre de lectures: 13
• Langue: English
• Poids de l'ouvrage: 6 Mo

Extrait

Quantifying carbon processes of the terrestrial
biosphere in a global atmospheric inversion based
on atmospheric mixing ratio, remote sensing and
meteorological data
Dissertation
zur Erlangung des akademischen Grades doctor rerum naturalium
(Dr. rer. nat.)
durchgefuhrt am Max-Planck-Institut fur Biogeochemie Jena
vorgelegt dem Rat der ChemischGeowissenschaftlichen Fakult at
der
Friedrich-Schiller-Universit at Jena
vorgelegt von
Bakr Badawy
aus
Agypten
Jena,
2011Gutachter:
1. Prof. Dr. Christiane Schmullius, Friedrich-Schiller-Universit at Jena
2. Prof. Dr. Martin Heimann, Max-Planck-Institut fur Biogeochemie Jena
Datum der o entlichen Verteidung: 20.04.2011Selbstst andigkeitserkl arung
Ich erkl are, dass ich die vorliegende Arbeit selbstst andig und nur unter Ver-
wendung der angegebenen Hilfsmittel, pers onlichen Mitteilungen und Quellen
angefertigt habe.i
Contents
Acknowledgement vii
Abstract 1
Zusammenfassung 3
Acronyms 5
1 Introduction 7
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 Overview of Global Carbon Cycle . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3 Terrestrial biosphere carbon cycle processes . . . . . . . . . . . . . . . . . . 12
1.4 Methods of estimating terrestrial carbon uxes . . . . . . . . . . . . . . . . 15
1.4.1 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4.1.1 Forest inventory . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4.1.2 Eddy ux measurements . . . . . . . . . . . . . . . . . . . 16
1.4.2 Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.4.2.1 Bottom-up approach . . . . . . . . . . . . . . . . . . . . . . 17
1.4.2.2 Top-downh . . . . . . . . . . . . . . . . . . . . . . 18
1.5 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.6 Objectives of this thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.7 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.8 Thesis outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2 Simple Diagnostic Photosynthesis and Respiration Model (SDPRM) 27
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2.1 GIMMS NDVI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.2.2 Land Cover Classi cation . . . . . . . . . . . . . . . . . . . . . . . . 31
2.2.3 fAPAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.2.4 Meteorological Data . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.3 Ecosystem Respiration Model . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.3.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.3.2 Model Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.4 Photosynthesis model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41ii Contents
2.4.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.4.2 Model Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.5 Implementation Details and Results . . . . . . . . . . . . . . . . . . . . . . 45
2.5.1 Experiment-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.5.2 Expt-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.5.3 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . 73
3 Atmospheric CO Inversion 752
3.1 Overview of STD-inv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.1.1 The linear ux model of STD-inv . . . . . . . . . . . . . . . . . . . . 78
3.1.1.1 Fixed and adjustable terms of STD-inv . . . . . . . . . . . 78
3.1.2 Atmospheric transport model . . . . . . . . . . . . . . . . . . . . . . 80
3.1.3 CO concentration data . . . . . . . . . . . . . . . . . . . . . . . . . 802
3.2 The coupled inversion system (SDPRM-inv) . . . . . . . . . . . . . . . . . . 81
3.2.1 Linearization of R . . . . . . . . . . . . . . . . . . . . . . . . . . 85eco
3.2.2 of GPP . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.2.3 SDPRM-inv set-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.2.3.1 The main di erence between STD-inv and SDPRM-inv . . 91
3.2.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.2.4.1 Time series of the estimated uxes . . . . . . . . . . . . . . 92
3.2.4.2 Fit to Data . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.2.4.3 Optimized parameters . . . . . . . . . . . . . . . . . . . . . 106
3.2.4.4 Total land uxes . . . . . . . . . . . . . . . . . . . . . . . 110
3.2.4.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
4 Summary and Outlook 115
4.1 General summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . 115
4.2 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Bibliography 121iii
List of Figures
1.1 The variations in concentration of carbon dioxide (CO ) in the atmosphere2
during the last 400 thousand years . . . . . . . . . . . . . . . . . . . . . . . 8
1.2 Recent atmospheric CO concentrations at Mauna Loa Observatory . . . . 82
1.3 Recent CO concentrations and emissions. . . . . . . . . . . . . . . . . . . . 92
1.4 The global carbon cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.5 Model framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.1 The SYNMAP land cover dataset aggregated into seven major PFTs . . . . 31
2.2 fAPAR time series calculated from GIMMS NDVI . . . . . . . . . . . . . . 36
2.3 VPD and TMIN attenuation scalars . . . . . . . . . . . . . . . . . . . . . . 43
2.4 Land regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.5 Global map of annual GPP . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.6 map of annual Reco . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.7 Time series of NEE estimated by SDPRM and STD-inv (IAV, 3Bands) . . . 54
2.8 Time series of NEE by SDPRM, STD-inv and BIOME-BGCv1
(full variability, 3Bands) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.9 Time series of NEE estimated by SDPRM, STD-inv and
(IAV, 3Bands) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.10 Time series of NEE estimated by SDPRM, STD-inv and BIOME-BGCv1
(full variability, TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . 57
2.11 Time series of NEE estimated by SDPRM, STD-inv and
(IAV, TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.12 Time series of GPP estimated by SDPRM and BIOME-BGCv1 (full vari-
ability, TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
2.13 Time series of GPP estimated by SDPRM and (IAV,
TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.14 Time series of Reco estimated by SDPRM and BIOME-BGCv1 (full vari-
ability, TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.15 Time series of Reco estimated by SDPRM and (IAV,
TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
2.16 Time series of the GPP: GPP-default (Black), GV-only (Blue), GT-only
(Red), and GS-only (violet) (IAV, 3Bands) . . . . . . . . . . . . . . . . . . 67
2.17 Time series of the GPP: (Black), GV-only (Blue), GT-only
(Red), and GS-only (violet) (IAV, TransCom regions) . . . . . . . . . . . . 68iv List of Figures
2.18 TimeseriesoftheReco: Reco-default(Black), RT-only(Red), andRP-only
(Blue) (IAV, 3 Bands) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
2.19 TimeseriesoftheReco:(Black), RT-only(Red), andRP-only
(Blue) (IAV, TransCom regions) . . . . . . . . . . . . . . . . . . . . . . . . 70
2.20 The global distribution of the relative contribution of the climate variable
to the IAV of GPP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
2.21 The global of the relative contribution of the climate variable
on the IAV of Reco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
3.1 Locations of the stations measuring atmospheric CO concentrations.. . . . 832
3.2 CO data availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 842
3.3 Temperature dependence of the non-linearized and linearized model of Reco 87
3.4 Time series of CO uxes from SDPRM-inv, and STD-inv - (full variability,2
TransCom) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
3.5 Time series of CO uxes from SDPRM-inv, and STD-inv - (IAV, 3Bands) 952
3.6 Time series of CO uxes fromv, andv - (IAV, TransCom) 962
3.7 Time series of Reco and GPP - (full time variability, 3Bands) . . . . . . . . 97
3.8 Time series of Reco and GPP - (IAV, 3Bands) . . . . . . . . . . . . . . . . 98
3.9 Taylor diagrams for the mean seasonal cycle of the observed and simulated
CO concentrations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1022
3.10 Taylor diagrams for the IAV of the observed and simulated CO concentra-2
tions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
3.11 Observed and simul