97th ESA Annual Meeting (August 5 -- 10, 2012)

COS 138-3 - The water sourcing strategy of drought affected temperate grasslands

Thursday, August 9, 2012: 8:40 AM
E146, Oregon Convention Center
Ulrich E. Prechsl1, Ansgar Kahmen1, Albin Hammerle2, Susanne Burri3, Anna K. Gilgen4 and Nina Buchmann5, (1)Institute of Agricultural Sciences, ETH Zurich, Zurich, Switzerland, (2)Institute of Ecology, University of Innsbruck, Innsbruck, Austria, (3)Department of Environmental Systems Science, Institute of Agricultural Sciences, ETH Zürich,, Zurich, Switzerland, (4)Institute of Plant Sciences and Oeschger Centre for Climate Change Research, University of Bern, Bern, Switzerland, (5)Institute of Agricultural Science, ETH Zurich, Zurich, Switzerland
Background/Question/Methods

In Central Europe, temperate grasslands are agroecosystems of high agricultural, ecological and economic importance. Growth, distribution and yield of this vegetation type depend strongly on the sufficiency of water supply in late spring/early summer. Until now, water has never been considered as a limiting factor. However, climate models project a change in the future precipitation patterns: until 2070, a reduction in summer precipitation of about 20% (compared to 1961-1990) is projected for north of the Alps. In addition, the frequency of extreme drought periods is likely to increase.
The aim of this study was to investigate if herbaceous grassland species adapt their water sourcing to (simulated) drought by exploring deeper soil layers to compensate the decreasing soil moisture. Summer drought was simulated by using transparent shelters (3m x 3.5m) at three different Swiss sites along an altitudinal gradient (400m to 2000m).

Water of soil and plant root crown (xylem) samples was cryogenically extracted. Soil, plant and precipitation waters were analyzed for δ18O using isotope ratio mass spectrometry. Standing belowground biomass was sampled regularly to determine possible shifts in the root mass distribution.

Results/Conclusions
The δ18O values of drought affected plants differed significantly from those of control plants, and often were similar to those of upper soil layers, although upper soil layers were very dry. This pattern occurred, at the three sites and during three years in different intensity. Particularly, in the recovery periods, i.e., about 4 weeks after removing the shelters, a very strong relationship with the δ18O value of precipitation could be found. Our results were confirmed by an additional evaluation with a Bayesian calibrated mixing model. Belowground biomass distribution supported these results based on the stable isotopes approach: Under drought, a highly dynamic and strong increase of root mass was observed in the top soil. Thus, in contrast to the hypothesized drought adaption to deeper rooting patterns, temperate grassland species shifted root growth and water uptake to the top soil, supposedly towards the next most likely water source: the next precipitation event.