How plants die of thirst: insights into mechanisms of tree drought mortality

We need to understand the mechanisms that cause tree death during drought to be able to identify which trees are most susceptible and develop management strategies under a changing climate. It has been hypothesized is that survival during drought is largely determined by the degree to which plants are isohydric (maintain constant tissue water potentials) or anisohydric (allow water potentials to change with changing water availability). Specifically, isohydric trees should more likely than anisohydric trees to survive a severe but short drought since they are better able to prevent dangerously low tissue water potentials. Isohydric trees, however, should be more susceptible to long, slow droughts as they close their stomata and then can succumb to carbon starvation. These expectations, however, are somewhat dependent upon the physiological mechanisms that govern isohydry/anisohydry; if governed by differences in leaf-level physiology, rather than being an artifact of rooting depth, then isohydric species may not be the most tolerant of drought. In 2011 we tested this hypothesis by conducting a dry-down experiment in the field on ten species planted together in common soil in large planters. The species were chosen to provide a range of drought strategies along the iso-anisohydric continuum. An automatic watering system was put in place to precisely control the rate of soil drying and thereby approximate natural drought conditions. Half of the twenty planters were allowed to dry down over 10 weeks, while the other half were maintained as controls. The most isohydric species (Liriodendron tulipifera, Liquidambar styraciflua, and Nyssa sylvatica) had the greatest and earliest mortality, however the drought is not likely to have been long enough to have resulted in depletion of sufficient carbohydrate stores to result in carbon starvation. Isohydry in these species seems to be a mechanism to protect what would otherwise be tissues very susceptible to desiccation. We also conducted a follow-up experiment in June through August of 2012 with four of the ten species. We made cuts into the stems to restrict water flow to the leaves so that we could rule out rooting depth as a cause of isohydry (the isohydric species tended to also have deep root systems). Isohydric species maintained their isohydric tendencies and kept fairly constant midday leaf water potentials until they reached severe levels of stress and subsequently died. This confirms that isohydry is a distinct physiological trait rather than an artifact of rooting depth.