||Southwestern USA experienced an estimated 40 to 95% mortality of piñon pine (Pinus edulis) and 2-25% mortality of juniper (Juniperus monosperma) (Breshears et al. 2005) during a recent prolonged drought (1998-2004). This mortality has significantly decreased woody plant cover and altered species distributions throughout the region. Although mortality has ultimately been attributed to bark beetle infestation (e.g. Ips confuses), the common belief is that a hydraulic mechanism underlies piñon susceptibility to attack. Climate models agree that future droughts may become more frequent and severe in the southwest. Thus, in the future there is a high likelihood of conditions favoring mortality events of similar or greater magnitude than the current event. Unfortunately, we lack the mechanistic knowledge required to predict responses of woodland ecosystems to future droughts. Therefore, our goal is to determine the hydraulic mechanisms of piñon and juniper survival and mortality during drought in the southwestern USA. Theoretical developments coupled with new results from the first two years of this study support a logical and testable set of hypotheses regarding the mechanisms underlying widespread vegetation mortality (McDowell et al. 2008). Our primary hypothesis is that drought causes the more “mesic”, cavitation sensitive, isohydric species such as piñon to succumb to carbon starvation associated with prolonged stomatal closure. These trees ultimately succumb to the negative carbon balance or to attack from biotic agents, whichever occurs first. In contrast, the more “xeric”, cavitation resistant, anisohydric species such as juniper are more likely to die of hydraulic failure due to prolonged maintenance of stomatal conductance and transpiration during drought. During periods of relatively wet climate, hydraulic theory predicts these species-specific traits will allow piñon a growth advantage over juniper.Thus far our project has been a success. Installation of the replicated (3X) rainfall reduction, addition and control treatment infrastructures were complete in August 2007 (1.5 years after project initiation), and ~90% of our planned measurements were operational at that time. We collected approximately one year of pre-treatment data, including dendrometer band assessment of growth, allometric harvests (off-plot) for leaf area, sapflow based transpiration and conductance, soil moisture, soil water potential, plant water potential, δ13C of carbohydrates, and δ18O of soil and xylem water for determination of functional rooting depth. Post-treatment shallow soil water content, soil water potential and sapflow data suggest a dramatic response to the drought treatments. The breadth of data sets required to adequately test the hydraulic failure hypotheses has required the full commitment of our research team. To adequately address the carbon starvation hypothesis and to publish results in a timely fashion, we propose an expanded effort in the next five years. We plan to continue our focus on hydraulic mechanisms underpinning piñon and juniper survival and mortality, but would like to expand our effort to include 1) rigorous empirical tests of the carbon starvation hypothesis and 2) development of the hydraulic model and its integration with autotrophic carbon balance, allowing hypothesis tests as well as sensitivity analyses of the likelihood of piñon and juniper responses to other variables that we cannot manipulate, such as temperature and humidity. This project will enhance our understanding of plant mortality and survival in both isohydric and anisohydric species and improve our ability to predict future mortality.