 |
| Titel |
Integrative comparison of CO2 flux measurements at a boreal peatland by chamber and eddy covariance methods with LPJ-GUESS model output |
| VerfasserIn |
Julia Schneider, Michal Gažovič, Lars Kutzbach, Paul Miller, Sannamaija Susiluoto, Tarmo Virtanen, Martin Wilmking, Jiabing Wu |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250034530
|
|
|
|
|
|
| Zusammenfassung |
| Many studies have been conducted to quantify the exchange of CO2 between the atmosphere
and the biosphere and to analyse the underlying processes at different spatial and
temporal scales. The chamber method and the eddy covariance (EC) technique
are the two measurement techniques which are mainly used to determine the CO2
fluxes. The chamber measurements provide discontinuous flux data on the plot scale
(10-2-100 m2) while the EC measurements provide under ideal conditions continuous
flux data on the ecosystem scale (104-106 m2). In general, measured CO2 flux
estimates of the plot- and ecosystem-scale over homogenous areas are comparable.
Over heterogeneous areas, however, large differences between EC and chamber
measurements can occur. This study focuses first on the comparability of CO2 flux
measurements conducted with those two techniques at a heterogeneous patterned boreal
peatland (seven microform types) in Northwest Russia from April to October 2008 and
then compares measured fluxes with CO2 flux estimates of the LPJ-GUESS. Both
measurement methods were carried out simultaneously to allow for comparison
by empirical modelling of the flux time series. Three different approaches were
used to integrate the plot-scale chamber measurements with the larger-scale EC
measurements: 1. upscaling based on average microform distribution and the mean NEE
flux for each microform type over the investigated period, 2. upscaling based on
areal weighting which accounts for main wind direction and 3. upscaling based
on footprint modelling which simulates the varying source fraction of the CO2
fluxes. First results indicate a substantial discrepancy between the flux estimated
from EC data and estimates obtained by upscaling from chamber measurements. At
larger scales, LPJ-GUESS is a process-based model of vegetation dynamics and
land-atmosphere carbon and water exchanges and is suitable for regional (103-105 km2) to
continental (106-107 km2) simulations on the time scale from days to millennia.
We will discuss the outputs of the LPJ-GUESS model for the investigated region
and compare the model flux estimates to the measured and upscaled CO2 fluxes. |
|
|
|
|
|
|