|
Titel |
Estimation of the ecosystem CO2 flux uncertainties over the Europe from ICOS atmospheric CO2 measurement network |
VerfasserIn |
Nikolay Kadygrov, Gregoire Broquet, Frédéric Chevallier, Leonard Rivier, Jean-Daniel Paris, Philippe Ciais |
Konferenz |
EGU General Assembly 2013
|
Medientyp |
Artikel
|
Sprache |
Englisch
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 15 (2013) |
Datensatznummer |
250083968
|
|
|
|
Zusammenfassung |
We present the utility assessment of Integrated Carbon Observing System (ICOS)
atmospheric observation network in constraining the biogenic surface CO2fluxes over the
Europe through the atmospheric inversion. A mesoscale inversion system was used to find
uncertainties in the 6 hourly Net Ecosystem Exchange (NEE) fluxes with 0.5o spatial
resolution for two week period and to estimate the uncertainties of monthly CO2 fluxes when
assimilating synthetic hourly observations from ICOS observation sites. The estimations of
the inverted flux uncertainties are analyzed on the model grid scale, country level and for the
whole European domain. The sensitivities of the results to inversion parameters were
tested. We checked the dependency of potential flux uncertainty reduction on the
extension of observation network and on the different parameters of the inversion
setup such as background covariance matrix parameterization and observation error
reflecting uncertainties in the ecosystem models used as a prior knowledge and in
the atmospheric transport model. The experiments were conducted for July and
December of 2007 showing heterogeneous seasonal structure of the uncertainty
reduction. We find that assimilating the data from the network with 23 observation sites
significantly reduces the uncertainty of the European biogenic two-week fluxes from the
biogenic ecosystem model by 50% to the value of ~ 0.51PgC yr-1 (over the land area
of 6.8*106 km2) for July 2007. The error reduction for December 2007 is found
to be higher and estimated as 66% with the value for posterior error ~ 0.31PgC
yr-1. The expansion of ICOS network to 66 atmospheric sites further reduce the
uncertainties of two-week fluxes by 64% (~0.39 PgC yr-1) and 79% (~0.18 PgC
yr-1) relative to the prior errors for July and December, respectively. The inversion
shows good performance reducing the prior flux uncertainties up to 65 % on the
grid scale in the vicinity of the station both for summer and winter fluxes. Having
dense ICOS atmospheric observation network we should be able to effectively
reduce the monthly CO2 flux uncertainties on the country scale up to 50-80 % for
specific country and significantly improve our knowledge about European CO2budget. |
|
|
|
|
|