PhD Thesis

Kohlendioxid-, Wasserdampf- und Energieaustausch eines Fichtenwaldes in Mittelgebirgslage in Nordostbayern

Corinna Rebmann (01/1996-12/2003)

Support: Thomas Foken

Die Arbeit wurde vorwiegend am LS Pflanzenökologie und am Max-Planck-Institut für Biogeochemie in Jena angefertigt, Betreuer: Prof. Dr. E.-D.Schulze

Eddy covariance measurements were performed above a spruce forest in the Fichtelgebirge in Northern Bavaria in the years 1997 through 1999. The measurements were embedded in the EUROFLUX-project with the goal to investigate carbon dioxide and water vapour fluxes of European forests and their interactions with the Climate System on a long term scale. The investigation of carbon fluxes and the contribution of the two components assimilation and respiration is of immense interest in ecosystem and climate research as anthropogenic influences may change the carbon balance of ecosystems in an unknown direction and magnitude. Half-hourly fluxes of carbon dioxide, water vapour, sensible heat and momentum were derived in the Lehstenbach water catchment, a main research area of the Bayreuth Institute of Terrestrial Ecosystem Research (BITÖK). The measurements were performed in two heights at a tower located in a 45-year old spruce (Picea abies (L.) Karst.) stand. Additional eddy covariance measurements were temporarily carried out below the canopy with the intention to separate the contribution of the soil to the total ecosystem fluxes. The influence of the heterogeneous terrain was investigated by separating the fluxes according to the wind direction and by detailed footprint analysis. Footprint analysis provide information about the source areas of the fluxes. Under stable atmospheric conditions these source areas are larger and more far away from the measurement point compared to neutral and unstable conditions. Therefore the second measurement height at the lower level was used to determine the night-time fluxes. To account for the non-ideal conditions that underlie the basic theory of eddy covariance measurements quality tests were performed for all half-hourly fluxes. Parameterisations for ’ high quality‘ CO2-fluxes with well developed turbulence were derived depending on global radiation and temperature. Di.erent gap filling procedures from these parameterisations were compared when annual sums of the net ecosystem exchange (NEE) of carbon dioxide were determined. The di.erent gap filling procedures resulted in larger di.erences (up to 114 gCm-2) of the annual sums compared to the maximum inter-annual variation (31 gCm-2). As an average of the four methods used, yearly integrated sums of the NEE amounted to -84gCm-2 in 1997, -82 gCm-2 in 1998 and -91gCm-2 in 1999. These numbers indicate that the ecosystem is only a small sink for carbon compared to other spruce forest ecosystems in similar latitudes with yearly NEEs of about -600 gCm-2. The small di.erence in annual sums was surprising because of the di.erent weather patterns but were also confirmed by very similar sums of the evapotranspiration of the three years. Total ecosystem evaporation ranged between 311mm in 1997, the dry year, and 119 341mm in 1999. The total ecosystem respiration was determined by extrapolating night-time fluxes to day-time hours and also by extrapolating the ecosystem respiration during daytime from the intercept of light response functions. The latter results should reflect that daytime respiration is reduced because of light-induced inhibition of leaf respiration. The monthly sums from both methods di.er on average by about 8%, and all sums compare well with the ecosystem respiration derived from up-scaled chamber measurements of soil, wood and foliage respiration and modelling of these components. Annual sums of respiration ranged from 1092 to 1280 gCm-2, depending on the method used. Gross primary production, derived as di.erence of NEE and total ecosystem respiration yielded annual sums between -1323 gCm-2 and -1208 gCm-2. These numbers can roughly be compared to NPP measurements determined in the Lehstenbach catchment in stands with di.erent age classes for the years 1984 — 1994. If the NPP of both, trees and understorey vegetation is taken into account, a di.erence of about 15% was determined. As carbon accumulation in the soil is not considered, these values fit very well if also measurement and gap filling uncertainties and the heterogeneous landscape are concerned.

last modified 2003-12-22