-Temporal variability in precipitation, which will increase in future, modifies ecosystem properties in semi-natural grassland communities (treatments 2008-2012)
-timing of extreme events is more important for ecosystem functioning than magnitude of the events (treatments 2008-2012)
-land use intensity (mowing regime, fertilization) interacts non-linearly with variability of precipitation (treatments (2008)-2010-2012)
-warming in summer and warming in winter interact non-additively with the precipitation treatments (treatments 2009/10-2018/19, starting from 2014 sampling (species-specific cover estimates and total biomass twice per year) only in N1, N5, N6 in CA & D1 – manipulations continue in all blocks)
winter rainfall addition interacts non-additively with the precipitation treatments (treatments winter 09/10 until winter 2012/13)
Aim: Test the effects of altered temporal rainfall variability (including drought extremes at different times in the year) in interaction with land use schemes or winter warming or winter rainfall addition or summer warming on ecosystem performance.
Location: Ecological-Botanical Garden, University of Bayreuth, Germany (49°55’19”N, 11°34’55”E, 365 m asl).
Climate: The regional climate is characterized as temperate and moderately continental with a mean annual air temperature of 8.2 °C and 724 mm of mean annual precipitation (1971-2000, data from German Weather Service).
Plant community: A semi-natural grassland which has not been ploughed for at least 25 years and not fertilized for more than 20 years prior to the installation of the experiment in 2008. Until the start of the EVENT II experiment, the meadow was mown twice a year for hay production. The semi-natural grassland community is dominated by tall grasses such as Alopecurus pratensis L. (meadow foxtail) and Arrhenatherum elatius (L.) P. Beauv. ex J. Presl & C. Presl (tall oat-grass) and belongs to the Galio molluginis-Alopecuretum pratensis Hundt (1954) 1968).
Substrate: The soil of the experiment is classified as Gleysol (Glaser et al., 2013). The homogeneous, loamy Ap horizon (42% sand, 43% silt, 15% clay) has a depth of 30 cm followed by a clayey Bg horizon. The groundwater table drops to -1.5 to -2 m during summer and can reach up to -30 cm in winter and after longer rainfall periods. Main rooting zone is within the upper 15 cm, hardly any roots reach the B horizon. The mean pH of the topsoil is 4.1 (1 M KCl). Permanent wilting point is around 10 vol.% soil moisture content.
Experimental design: five replications of five precipitation treatments. Within each precipitation treatment six plots (1.5 m x 1.5 m size) receive different warming, winter rainfall and management manipulations.
Summer Precipitation Manipulations consist of 1000-year extreme drought events according to local climate data series (42 days without rainfall; rain out shelters) and different control manipulations (ambient control with and without rain-out shelters, longterm weekly average precipitation).
Since 2009 the annual sum of precipitation is kept constant between the treatments by adding the amount that CM received artificially at four times per year (before D1, after D1, after D2 and end of September) to all other treatments!
D1: Spring Drought:
D2: Summer Drought:
CM: Weekly average precipitation: add missing amount if weekly precipitation is less than long-term average for the same week
CA: Ambient Control: no manipulation
XX: roof artifact control: during D1 rain-out shelters are installed and natural rainfall is added below the shelters twice a week.
Within block manipulations
Winter rain: 15 l/m² (=mm) in mid November, mid December, mid January, mid February (60 mm total)
N1: mowed two times per year, no further manipulation
N2: mowed two times per year, winter rain from winter 09/10 until winter 2012/13 (no manipulation before)
N3: land use scenarios since 2010 (four times mowed and no further manipulation before):
four subplots 0.75 x 0.75 m² separated by stainless steel frames down to -25 cm.
- F2: fertilized at the 30th day of the D1 manipulation (except for D2 which is fertilized at the 30th day of the D2 manipulation) with NPK-fertilizer “Linzer Top 12/12/17”, 14 g/Plot; mowed two times (end of drought D1 (except for D2 which is mowed at the end of drought D2) and September)
- F4: same fertilization as F2; mowed two times (10 days after end of drought D1 (except for D2 which is mowed 10 days after the end of drought D2) and September)
- U2: no fertilization; mowed according to F2
- U4: no fertilization; mowed according to F4
see Figure 2 for details
N4: mowed four times per year, winter rain from winter 09/10 until winter 2012/13 (no manipulation before)
N5 mowed two times, winter warming by IR-heaters (250 W, 80 cm height) from October to End of March (please note that all other plots are equipped with dummies of the heating facilities) since October 2009 (no warming in winter 2013/14!)
N6 mowed two times, summer warming by IR-heaters from April to End of September since 2010 (warming in 2014 started in January!)
Response parameters until 2013:
-aboveground biomass by species
-soil moisture in all N1 until 2010 and in all plots except N3 from 2010 onwards
-soil and air temperature in selected plots
-soil enzyme activity (GSF)
-N-content in soil, 15N experiment (Glaser)
-Insects in 2008 (Schädler)
-Soil respiration in selected plots since 2010
-Chlorophyll content in selected plots since winter 2009/10
-Micrometeorology in climate manipulations 2012 (Foken)
Response parameters from 2014 on:
-species specific cover and total biomass twice per year only in N1, N5, N6 in CA & D1 (30 plots total).