Identification of suites of traits that explains drought resistance and phenological patterns of plants in a semi-arid grassland community.

Grassland ecosystems are comprised of plants that occupy a wide array of phenological niches and vary considerably in their ability to resist the stress of seasonal soil-water deficits. Yet, the link between plant drought resistance and phenology remains unclear in perennial grassland ecosystems. To evaluate the role of soil water availability and plant drought tolerance in driving phenology, we measured leaf hydraulic conductance (Ksat), resistance to hydraulic failure (P50), leaf gas exchange, plant and soil water stable isotope ratios (δ18O), and several phenology metrics on ten perennial herbaceous species in mixed-grass prairie. The interaction between P50 and δ18O of xylem water explained 67% of differences in phenology, with lower P50 values associated with later season activity, but only among shallow-rooted species. In addition, stomatal control and high water-use efficiency also contributed to the late flowering and late senescence strategies of plants that had low P50 values and relied upon shallow soil water. Alternatively, plants with deeper roots did not possess drought-tolerant leaves, but had high hydraulic efficiency, contributing to their ability to efficiently move water longer distances while maintaining leaf water potential at relatively high values. The suites of traits that characterize these contrasting strategies provide a mechanistic link between phenology and plant-water relations; thus, these traits could help predict grassland community responses to changes in water availability, both temporally and vertically within the soil profile.

Impacts of warming and elevated CO2 on a semi-arid grassland are non-additive, shift with precipitation, and reverse over time.

It is unclear how elevated CO2 (eCO2 ) and the corresponding shifts in temperature and precipitation will interact to impact ecosystems over time. During a 7-year experiment in a semi-arid grassland, the response of plant biomass to eCO2 and warming was largely regulated by interannual precipitation, while the response of plant community composition was more sensitive to experiment duration. The combined effects of eCO2 and warming on aboveground plant biomass were less positive in 'wet' growing seasons, but total plant biomass was consistently stimulated by ~ 25% due to unique, supra-additive responses of roots. Independent of precipitation, the combined effects of eCO2 and warming on C3 graminoids became increasingly positive and supra-additive over time, reversing an initial shift toward C4 grasses. Soil resources also responded dynamically and non-additively to eCO2 and warming, shaping the plant responses. Our results suggest grasslands are poised for drastic changes in function and highlight the need for long-term, factorial experiments.