Modeling surface-atmosphere exchange of trace gases and energy within and above the Amazon rain forest [Elektronische Ressource] / Eric Simon

Informations

• Publié par: johannes_gutenberg-universitat_mainz
• Publié le: 01 janvier 2004
• Nombre de lectures: 8
• Langue: English
• Poids de l'ouvrage: 8 Mo

Extrait

Modeling Surface-Atmosphere
Exchange of Trace Gases and
Energy within and above the
Amazon Rain Forest
D I S S E R T A T I O N
Zur Erlangung des Grades
Doktor der Naturwissenschaften
Am Fachbereich Biologie
Der Johannes Gutenberg-Universitat˜ Mainz
Eric Simon
geboren am 23.2.1972
in Bernkastel-Kues
Mainz, 13. September 2004Jahr der mundlichen Prufung: 2004˜ ˜
D77 - Mainzer Dissertation\... die Tranen von gestern wird die Sonne trocknen,˜
und die Spuren der Verzwei ung wird der Wind verweh’n ..."
(Ton-Steine-Scherben)
iiiPreface
The present dissertation is a result of more than three years of work which I car-
ried out at the Max Planck Institute for Chemistry in Mainz. The dissertation
focuses on the biogeochemistry of the vegetation layer (canopy) and the interac-
tions between physiological and environmental processes which afiect the climate
and chemistry of the lower atmosphere. A main task is the quantiflcation of
vertical exchange of trace gases and energy with theoretical concepts considering
the links and feedbacks between the partitioning of energy at the leaf surface,
the uptake of CO , the emission of biogenic volatile organic compounds (VOC),2
the dry deposition of ozone, the vertical transport within the canopy and the
ecosystem net exchange (vertical ?uxes). This is achieved by implementing a so-
phisticated multi-layer canopy scheme of energy and trace gas exchange which is
combined with a Lagrangian description of vertical transport within the canopy.
Thereby, extensive data sets of fleld measurements in Amazonia are considered.
Additionally, an alternative approach is applied to infer the link between physi-
ological parameters of CO and H O exchange and the emission of isoprene and2 2
monoterpenes, representing the most important non-methane VOC’s. The trop-
ical Amazon rain forest is still the largest forest ecosystem on earth and plays
a particular role in the global climate. Therefore, a very detailed description
and discussion of the canopy biochemistry and vertical exchange characteristics
is given. Although the applied techniques and concepts are also applicable to
other ecosystem types, the present work is restricted to the tropical rain forest
canopy and does not consider the efiects of deforestation and land-use change.
The work is organized in six chapters: Chapter 1 contains a general intro-
duction describing brie y the particular role of the terrestrial vegetation in the
global climate system and current trends of climate change. After a description
of methods and models of surface exchange, it leads to a comparison of difier-
ent canopy model types and the motivation for the present work. The canopy
model which has been developed for the present study, is described in the second
part of Chapter 1. In Chapter 2 the parameterizations for soil surface exchange,
vcanopy structure, the proflle of horizontal wind speed, and the biochemistry of
the rain forest canopy are inferred using available data sets from difierent inten-
sive campaigns in Amazonia. Additionally, the calculations related to radiation
attenuation and leaf surface exchange are evaluated and the uncertainty of key
parameters is assessed. The vertical transport within the multi-layer scheme is
simulated with a Lagrangian approach which is parameterized and evaluated in
detail in Chapter 3 using in-canopy turbulence measurements and observations of
222 222Radon as a tracer of vertical dispersion, respectively. The radioactive Radon
is an inert gas emitted from natural soils. Special emphasis is given on nighttime
conditions where common micrometeorological methods fail to produce reliable
results (Eddy Covariance technique,! EC). In Chapter 4, the canopy scheme is
applied to late wet and late dry season conditions at a remote rain forest site in
Rondon^ ia, southwest Amazonia. Model predicted in-canopy proflles of CO , H O,2 2
ozone, and isoprene concentration for day- and nighttime conditions are compared
to observations. Predicted ?uxes of sensible and latent heat, CO and ozone are2
compared to EC measurements considering the efiect of canopy storage. The sen-
sitivity of model predictions to key parameters’ uncertainties and the observed
seasonality of net assimilation and transpiration are assessed. An alternative ap-
proach to predict the emission of VOC’s is presented in Chapter 5. Isoprene and
monoterpene emissions of difierent tree species from difierent seasons growing in
difierent light environments are related to environmental and leaf physiological
parameters using a neuronal approach. The performance of difierent parameter
combinations serving as predictors of VOC emissions are compared to the results
of the quasi-standard emission algorithm for isoprene given by Guenther et al.
(1993). Chapter 6 ofiers a summary and the main conclusions of the dissertation.
Detailed descriptions of the canopy model calculations, the radon soil ?ux mea-
surements, the neuronal approach and the applied or derived parameterizations
are given in the Appendices, containing additionally a list of abbreviations and
symbols.
viContents
1 Introduction 1
1.1 Background and motivation . . . . . . . . . . . . . . . . . . . . . 1
1.2 Model description . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 General concept . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.2 Radiation absorption and surface exchange . . . . . . . . . 8
2 Parameterization of Amazon rain forest surface characteristics 13
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2 Material and method . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.1 Inferred parameterizations . . . . . . . . . . . . . . . . . . 14
2.2.2 Site description and fleld data . . . . . . . . . . . . . . . . 17
2.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3.1 Inferring a mean canopy structure for Amazon rain forest . 21
2.3.2 Inferring a normalized proflle of horizontal wind speed . . 23
2.3.3 Evaluation of calculations related to radiation . . . . . . . 24
2.3.4 Inferring parameters related to soil surface exchange . . . . 28
2.3.5 Inferring the light acclimation of photosynthetic capacity . 29
2.3.6 Evaluation of calculations related to leaf surface exchange 31
2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2223 On Lagrangian dispersion of Rn, H O, and CO 392 2
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2 Material and method . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2.1 Field data and measurement overview . . . . . . . . . . . . 42
3.2.2 Measurements of canopy structure at the Cuieiras site . . . 44
3.2.3 Radon measurements . . . . . . . . . . . . . . . . . . . . . 45
3.2.4 Implementation of the Localized Near-fleld theory (LNF) . 46
3.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . 48
3.3.1 Parameterization of in-canopy turbulence proflles . . . . . 48
vii3.3.2 Meteorological conditions at the Cuieiras site . . . . . . . 51
3.3.3 Observed radon soil ?uxes at the Cuieiras site . . . . . . . 52
2223.3.4 Forward modeling of Rn activity concentrations . . . . . 54
3.3.5 Efiective transfer velocities and timescales . . . . . . . . . 56
2223.3.6 Inverse predictions of Rn ?uxes . . . . . . . . . . . . . . 56
3.3.7 Net ?uxes for CO , latent and sensible heat . . . . . . . . 572
3.3.8 Diurnal source/sink distributions for CO and H O . . . . 602 2
3.3.9 Daily integrated net carbon exchange . . . . . . . . . . . . 62
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4 Modeling seasonal exchange of energy, CO , isoprene, and ozone 652
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.2 Material and method . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.2.1 Site location, period and fleld data . . . . . . . . . . . . . 67
4.2.2 Meteorological overview . . . . . . . . . . . . . . . . . . . 68
4.2.3 Model setup . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.2.4 Calculation of storage terms . . . . . . . . . . . . . . . . . 71
4.2.5 of isoprene emission and ozone deposition . . . 72
4.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . 73
4.3.1 Stable solutions for steady-state environmental conditions 73
4.3.2 Model sensitivity to key parameter uncertainty . . . . . . 76
4.3.3 Evaluating seasonal predictions of CO and energy exchange 792
4.3.4 Evaluatingal predictions of isoprene exchange . . . 86
4.3.5 Seasonal comparison and evaluation of predicted ozone ex-
change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.4 Further model application examples . . . . . . . . . . . . . . . . . 94
4.4.1 Global isoprene emissions from tropical rain forest . . . . . 94
4.4.2 Seasonal comparison of predicted climate change due to el-
evated atmospheric CO levels . . . . . . . . . . . . . . . . 952
4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5 Neuronal modeling of biogenic VOC emission 99
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
5.2 Field measurements and site description . . . . . . . . . . . . . . 102
5.3 Neuronal approach . . . . . . . . . . . . . . . . . . . . . . . . . . 103
5.3