A Biophysically Based Coupled Model Approach For the Assessment of Canopy Processes Under Climate Change Conditions

The central questions of this thesis are concerned with the
investigation of vegetation related landsurface parameters under
the impact of changing climate conditions. The spatial extent of
the study is limited to the borders of the Upper Danube drainage
basin, according to the requirements of the cooperative Project
GLOWA-Danube (Global change of the Hydrological Cycle), funded by
the German Ministry of Education and research (BMB+F). Current
publications are indicating that the dynamic behaviour of the
vegetation cover often is inadequately accounted for in studies
that are investigating the impacts of climate change with respect
to the landsurface water cycle. In order to enable a dynamic
feedback between the animate land cover and the atmosphere, which
might be sensitive enough to trace active reactions of the
vegetation cover on changing climatic conditions, the physically
based land surface process model PROMET (Process of Radiation Mass
and Energy Transfer Model) was enhanced by an explicit description
of the growth activity of different plant types. The introduced
model approach was tested against measured data for a variety of
parameters. The different validation efforts all returned good to
very good results. It therefore can be stated that the model
soundly demonstrated its capability concerning the precise
reproduction of a variety of structural landsurface variables on
different scales under observed climatic conditions. An application
of the model to the calculation of climate scenarios therefore
seems appropriate. In order to enable comparability with
international research approaches, the internationally acknowledged
global change scenarios developed by the Intergovernmental Panel on
Climate Change (IPCC), are basically applied. The moderate A1B
emissions scenario, which is based on the assumption of a balanced
future development of different energy technologies, was selected
and modified by a regional impact factor that is assumed to apply
to the local situation of the Upper Danube catchment. Being applied
to the regionally adapted IPCC A1B climate scenario, the model
returned clear statements, projecting a possible future development
of selected landsurface parameters within the Upper Danube area.
Concerning the phenological behaviour of forest trees, the model
simulated a strong trend towards earlier incidence of the leaf
emergence of deciduous as well as of the mayshoot of coniferous
trees, contributing to a significant elongation of the vegetation
period. These longer phases of active growth in combination with
the rising temperatures and the elevated supply of atmospheric
carbon dioxide led to an increase of biological activity in the
model results that manifested in increasing rates of biomass
accumulation for the Upper Danube area. The increased biological
activity in combination with the strong decrease of summer
precipitation, which was assumed in the climate scenario, again led
to an escalating frequency of drought stress events in the Upper
Danube Basin. Not only the average count of water stress events per
year was modelled to increase, but also a spatial extension of the
regions that are affected by drought stress was mapped by the
model. This general increase of water stress and the significant
decrease of summer precipitation entailed a slight decline of the
transpiration and evapotranspiration of the Upper Danube area in
the scenario results. The modelled decline of the summer
precipitation also resulted in a noticeable decrease of the
modelled average discharge rates at the main gauge of the basin.
The base flow rates during the summer months thereby are likely to
be primarily affected. Since the model results for the scenario
period featured temporal and spatial variations and standard
deviations that were closely matching the statistics of the
reference period, while at the same time they showed clear trends
though they were avoiding extreme realizations, the scenario
assumptions can be considered to be reliable. The baseline
scenario, which was spot-checked for a set of reference proxels,
did not return any trends as expected, indicating that the observed
future trends are not of systematic origin. The further development
of the introduced model approach is an appealing challenge, which
might considerably contribute to the improvement of computer aided
decision support systems. It can be assumed that the progress of
the development of physically based models due to a more profound
understanding of the processes on one hand and the sophistication
and refinement of the model algorithms that result from the
increase of knowledge on the other, may contribute to the
development of reliable systems, that will be able to sustainably
assist humanity with the handling of future environmental
challenges. The author gratefully acknowledges the finacial support
of the German Research Foundation (DFG) in the frame of the project
"Coupled Analysis of Vegetation Chlorophyll and Water Content Using
Hyperspectral, Bidirectional Remote Sensing".