 |
| Titel |
Analysing global ecosystem CO2 uptake capacity with plant trait data |
| VerfasserIn |
Martine van de Weg, Talie Sadat Musavi, Peter Van Bodegom, Jens Kattge, Miguel Mahecha, Markus Reichstein, Michael Bahn |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250096160
|
| Publikation (Nr.) |
 EGU/EGU2014-11653.pdf |
|
|
|
|
|
| Zusammenfassung |
| Given the modulating role of vegetation in the global carbon cycle, there is a demand for
simple and general scaling relationships of vegetation characteristics and ecosystem
CO2-uptake and emissions. On a leaf level, is it well established that plant trait foliar nitrogen
(N) relates strongly with leaf level CO2. Furthermore, ecosystem productivity or CO2 uptake
capacity have been related directly with whole-canopy N concentrations for a variety of
ecosystems such as grasslands, and boreal, temporal and tropical forests. However,
studies on the global validity of these leaf and ecosystem level relationships have
been lacking up to date. The arrival of the large plant trait database TRY database
offers the opportunity to link plant trait and ecosystem functioning on a global
scale.
In this study, we used CO2 flux data from the FLUXNET database, with plant trait
(Narea) data from TRY and Narea measurements from a selection of FLUXNET sites as
well. For 83 global FLUXNET sites, which had information available on species
composition, we derived the light saturated gross primary productivity (GPP1000). We used
MODIS LAI and fPAR, together with the species’ relative height and abundance data, to
up-scale the TRY derived Narea values to a canopy value per site (Ncanopy). For this
calculation we assumed that top canopy leaves contribute more to CO2 uptake, and used a
Lambert-Beer canopy light extinction principle to weigh the relative contribution per species
to the final Ncanopy value. For our analyses, we divided the sites in five different vegetation
classes: broad leaved forests, needle leaved forests, grasslands, crops and (sub)arctic
non-forest vegetation.
Site-measured Nareadata corroborated well with TRY derived Narea data, giving
confidence in using a database such as TRY for global analyses like ours. Ncanopy alone
explained 18 % of the observed variation in maximum (90th percentile) GPP1000 with a
linear model. When adding the different vegetation types as a covariate, however, this
increased to explaining 72% of the observed variation. Furthermore, the Ncanopy - LAI
patterns suggest a global ecosystem level strategy of relatively high N-investment per leaf
area when little N is available and additional investment in leaf area only when more N is
present.
Overall, our study shows that including Narea and vegetation type information can be
very informative when analyzing the global GPP patterns in addition to using LAI alone,
underlining the potential of including traits in modeling global CO2 uptake. Our results
furthermore point out the importance of putting our focus beyond leaf level and
consider whole canopy strategies for explaining ecosystem level CO2 exchange. |
|
|
|
|
|
|