 |
| Titel |
Fate of leaf-litter N in forest and grassland along a pedo-climatic gradient in south-western Siberia: an in situ 15N-labelling experiment |
| VerfasserIn |
Félix Brédoire, Bernd Zeller, Polina Nikitich, Pavel A. Barsukov, Olga Rusalimova, Mark R. Bakker, Arnaud Legout, Alexander Bashuk, Zachary E. Kayler, Delphine Derrien |
| Konferenz |
EGU General Assembly 2017
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 19 (2017) |
| Datensatznummer |
250149085
|
| Publikation (Nr.) |
 EGU/EGU2017-13401.pdf |
|
|
|
|
|
| Zusammenfassung |
| The suitability of Siberia for agriculture is expected to increase in the next decades due to
strong and rapid climatic changes, but little is known on the environmental drivers of soil
fertility there, especially nitrogen (N). Plant-available N is mainly derived from litter
decomposition. South-western (SW) Siberia is located on the transition between several
bioclimatic zones that are predicted to shift and extend along with climate change (steppe,
forest-steppe, sub-taiga). The soils of this region are formed on a common loess deposit but
they are submitted to different climatic conditions and vegetation cover. In the south of the
region, typically in steppe/forest-steppe, soil freezes over winter because of a relatively
shallow snow-pack, and water shortages are frequent in summer. In the north, typically in
sub-taiga, the soil is barely frozen in winter due to a thick snow-pack and sufficient soil
moisture in summer.
In this study, we characterized the dynamics of leaf litter decomposition and the transfer
of N from leaf litter to the soil and back to plants. Four sites were chosen along a climate
gradient (temperature, precipitation and snow depth). At each site, we applied 15N-labelled
leaf litter on the soil surface in experimental plots in an aspen (Populus tremula L.) forest and
in a grassland. Twice a year during three years, we tracked the 15N derived from the
decomposing labelled-litter in the organic layers, in the first 15 cm of the soil, and in
above-ground vegetation. Soil temperature and moisture were monitored at a daily timestep
over three years and soil water budgets were simulated (BILJOU model, Granier et al.
1999).
We observed contrasting patterns in the fate of litter-derived 15N between bioclimatic
zones. Over three years, along with faster decay rates, the release of leaf litter-N was faster in
sub-taiga than in forest-steppe. As such, higher quantities of 15N were transferred into the
soil in sub-taiga. The transfer was also deeper there, which might be related to a more intense
drainage because of higher snow levels, as inferred from soil water budget modelling.
Interestingly, this higher drainage seems to induce only a small loss of N from the system.
Such retention could result from soil physico-chemical properties (higher fine silt and oxides
contents) enhancing soil organic matter stabilization, and/or by the immobilization of N in
microbial metabolites.
We observed differing N dynamics between forest and grassland that can be
related to the different chemical composition of initial litter (tree leaves vs. grasses)
and plant–soil interactions. In general, N was retained in the first centimeters of
the mineral soil in grassland while the transfer was deeper in the forest soils. As
fine root exploration is denser in grassland topsoil than in forest topsoil, we infer
that an efficient uptake of N by grasses in the first soil layers limits N migration
down the profile. It is also possible that grasses are active earlier in the season than
trees and understorey species, i.e. at snow-melt when drainage is the most intense. |
|
|
|
|
|
|