Annual grassland resource pools and fluxes: sensitivity to precipitation and dry periods on two contrasting soils

In ecosystems throughout the world climate models project increased variability in precipitation patterns that may strongly affect the above- and below-ground processes that control carbon, water, and nutrient cycles. Uncertainty about how plant and soil processes respond to wet and dry periods at different times in the growing season is a barrier to understanding how changing rainfall patterns will affect ecosystem function in annual grasslands. We used mesocosm systems to test the sensitivity to mid- and late-season dry periods of twenty response variables related to nitrogen, carbon, and water cycling in Avena barbata monocultures. We compared the responses of individual variables and of grassland systems under low and high cumulative rain treatments and between two contrasting soil types.

Analysis of individual response variables demonstrated strong seasonal patterns: most soil and plant resource pools and fluxes changed between mid- and late-season, and many had larger responses to the late-season dry period. Under increasingly variable precipitation regimes, specific resource pools and fluxes such as soil nitrate and ecosystem CO₂ flux may be more strongly affected when dry periods occur later in the growing season. Individual responses variables were also used as state variables in a principal components analysis of changes in grassland functional states between treatments and over time. There were dramatic functional state shifts between the mid- and late-season for both soil types, driven by changes in canopy height, leaf soluble protein, soil nitrate, gross N mineralization, and leaf N. However, we did not find evidence that the functional state of grassland systems was affected by rainfall patterns, indicating that interactions among below- and above-ground processes resulted in system-level resistance to changes in soil moisture. A strong association between plant canopy size and ecosystem functional state suggests that responses of plant phenology and growth to climate change may predict changes in ecosystem function.