Master Thesis

Modeling of the energy exchange above lake and land surfaces

Elisabeth Thiem (08/2010-02/2011)

Support: Wolfgang Babel, Thomas Foken

Geoökologie, Univ. Bayreuth

Models help us understand complex systems by representing those systems in simplified terms. They can provide input for other models and are the only tool that allows predictions. Another application of models is gap-filling. Models rely on the quality of the input data. Observational data contain gaps for very different reasons. Nonetheless data are required for the assessment of ecosystem budgets such as the net ecosystem exchange or as input for simulations schemes. The outcome of both depends greatly on the input data quality. A gap-filling procedure that is able to fill data gaps realistically and therefore enhances input data quality might influence the results of simulation schemes and the reliability of budget assessments positively. In this study, two models are evaluated as gap-filling schemes for turbulent flux data above lake and land surfaces. A Soil-Vegetation-Atmosphere (SVAT) scheme is applied and evaluated with respect to its performance of turbulent fluxes above a Tibetan meadow surface. The performance of a hydrodynamical multilayer model, applied for the simulation of fluxes above a lake surface is also evaluated. Their performance is then related to the reliability of input data, which were gathered during a field campaign on the Tibetan Plateau in the summer 2009. Both models performed very well, which was proven by high correlation coefficients. A sensitivity assessment of the water surface model revealed a high influence of a newly implemented parameterization for shallow water. Obvious errors of the input data, especially concerning the lake temperature were not found. An assessment of the heat budget over water failed due to insufficient data. Requirements for the direct flux measurements with the eddy-covariance method were not always fulfilled at the study area. In consequence, input data might not be able to represent “truth”. Model performance could not be fully assessed, mainly because no complete diurnal cycle for the turbulent fluxes either above the lake or above the land surface was measured, due to the geographically modified flows.

last modified 2012-12-06