Experimental Design

Our experimental design consists of three fully crossed factors (warming, increased winter precipitation, and N addition) in a completely randomized design, for a total of eight treatment combinations, with five replicates of each treatment combination, for a total of 40 plots. Each plot is 3 x 3.5 m. All plots contain B. eriopoda, B. gracilis and G. sarothrae. Our nighttime warming treatment is imposed using lightweight aluminum fabric shelters (mounted on rollers similar to a window shade) that are drawn across the warming plots each night to trap outgoing longwave radiation. The dataloggers controlling shelter movements are programmed to retract the shelters on nights when wind speeds exceed a threshold value (to prevent damage to shelters) and when rain is detected by a rain gauge or snow is detected by a leaf wetness sensor (to prevent an unintended rainout effect).

Each winter we impose an El Nino-like rainfall regime (50% increase over long-term average for non-El Nino years) using an irrigation system and RO water. El Nino rains are added in 6 experimental storm events that mimic actual El Nino winter-storm event size and frequency. From January-March, there are 4x5mm applications, 1x10mm application and 1x20mm application. During El Nino years we use ambient rainfall and do not impose experimental rainfall events. For N deposition, we add 2.0 g m-2 y-1 of N in the form of NH4NO3 because NH4 and NO3 contribute approximately equally to N deposition at SNWR (57% NH4 and 43% NO3; Bez et al., 2007). The NH4NO3 is dissolved in 12 liters of deionized water, equivalent to a 1 mm rainfall event, and applied with a backpack sprayer prior to the summer monsoon. Control plots receive the same amount of deionized water.

Air and soil temperature are measured in each plot with copper constantan thermocouples positioned 20 cm above ground, and 2 and 10 cm below the soil surface in both grass-covered and soil/crust areas. Measurements are taken at 15 minute intervals.