Plant whispers : monitoring plant water potential to quantify plant water needs and health status during drought

Demand for water to irrigate crops is increasing due to rising global temperature and population growth, with 70% of fresh water already diverted to irrigation. This is expected to exacerbate water scarcity and increase water stress in many crop production regions. To manage these challenges, there is a need to identify ways to precisely manage agricultural water use and understand the impact of limited water availability on crop performance. One way to improve irrigation efficiency is to restrict water supply to plants while minimizing the impact on crop production. This requires real-time monitoring of plant water use and water stress, combined with knowledge of crop-specific physiological requirements and critical thresholds under water stress.

Current irrigation scheduling practices have primarily focused on soil-based moisture monitoring, but this approach does not sense the way moisture stress is detected by plants, and it is therefore not sensitive to plant physiology nor soil or root heterogeneity. In this thesis, I investigate the potential for plant monitoring systems based on plant physiological measurements, determined through continuous field-based monitoring of plant water potential, or more specifically, stem water potential (Ψ_stem). Ψ_stem is a common physiological measurement which indicates the tension that drives water flow within the plant. The magnitude of this tension reflects the interactions among aspects of plant hydraulic architecture as well as soil water supply around the roots and transpirational water demands, thus making Ψ_stem an excellent candidate for monitoring crops under field conditions. In this thesis, I examined the applicability of using Ψ_stem to continuously monitor plant water use and plant health status by defining different physiological thresholds during water stress.

In chapters 2 and 3 I tested the viability of using Ψ_stem to monitor plant transpiration under field conditions in various crop and woody species with distinct morphological, anatomical and phylogenetic identities such as Triticum aestivum (grassy monocot), Tanacetum cinerariifolium (daisy perennial herb), Zea mays (grassy monocot) and Callitris rhomboidea (hardy conifer). This approach requires the hydraulic supply (root to stem hydraulic conductance, Kr-s) to remain constant with changing transpiration. I verified whether such an approach is viable by investigating whether Kr-s remains constant under a wide range of daytime transpiration rates. Optical dendrometers with high temporal resolution were used to continuously monitor leaf and petiole shrinkage as a proxy of Ψ_stem. Kr-s remained relatively stable with changing transpiration throughout the day under variable natural conditions in all species. Thus, these results indicate that optically derived Ψ_stem dynamics can reliably infer plant water use in situ under nonlimiting water conditions. The ability to continuously monitor Ψ_stem with optical dendrometry allowed both instantaneous and daily plant water transpiration to be monitored at high accuracy and under variable ambient conditions.

Having established the functionality of Ψ_stem for monitoring water use and water stress levels, using optical dendrometers, in Chapter 4 I investigated the potential of Ψ_stem to identify thresholds in water stress that have significant impacts on plant water relations and gas exchange. Efficient water transport from the roots to the shoots is critical for maintaining open stomata, photosynthesis, and growth. The physiology of aboveground portion of this pathway is well-described, yet much of the dynamic response of root hydraulics and its influence on gas exchange during soil drying and recovery remains uncertain. Therefore, in chapter 4, I examined the sensitivity of root hydraulic conductance (Kr) to declining soil water and coordination with canopy diffusive conductance (g_c) during exposure to water deficits sufficient to close stomata but not cause xylem cavitation. This was done in two species with contrasting root systems: T. cinerariifolium with herbaceous roots and C. rhomboidea with woody roots. Kr vulnerability to increasing water stress was measured non-invasively by monitoring the kinetics of plant hydration after rewatering from different water stress levels. The relaxation kinetics of Ψ_stem in intact hydrating plants was recorded at high temporal resolution using optical dendrometers. Kr in both species was similarly highly sensitive to moderate soil water deficit, with depressions occurring concomitantly with declining g_c after Ψ_stem fell below -1 MPa. The speed and extent of Kr recovery after rewatering, however, varied between species depending on root type and water stress severity, with herbaceous roots faster to recover. Recovery of g_c followed a similar trend to Kr in both species, hence Kr was suggested to play a fundamental role in steering the dynamics of g_c under dynamic water stress.

A high and reversible sensitivity of roots to moderate water stress was shown in chapter 4, but sustained drought after stomatal closure may result in further decreases in Ψ_stem causing irreversible damage to the vascular system due to xylem cavitation, leading to highly undesirable consequences for crop production. In Chapter 5, I examined the potential to use Ψ_stem to quantify key thresholds causing permanent damage to crop productivity. Relationships between xylem cavitation and tissue damage have been widely characterized in vegetative tissues, such as stems, roots and leaves. By contrast, nothing is known about whether flowers are also damaged by xylem cavitation during drought. In chapter 5, I examined if, and when, flowers suffer cavitation damage compared with neighbouring leaves using the optical technique coupled with in-situ quantification of tissue damage during acute water stress in the daisy herbaceous T. cinerariifolium. I also determined the relative transpirational cost of flowers during drought. Interestingly, flowers were found to be more vulnerable to xylem cavitation than leaves as they cavitated rapidly and completely soon after stomatal closure and before incipient xylem cavitation in leaves. This xylem vulnerability segmentation was confirmed by observations that most flowers died after reaching Ψ_stem causing significant cavitation in the flowering stem xylem while leaves were minimally damaged. This clearly indicates a direct mechanistic link between xylem cavitation and flower death and confirm that the water supply to flowers is mainly via the xylem. Flowers were found to be a major source of water loss after drought induced stomatal closure due to their leakier cuticle compared with leaves, hence their isolation by cavitation may provide a means of delaying Ψ_stem decline by reducing excessive water loss and preserving perennial vegetative organs essential for ongoing survival and future reproduction from drought damage.

The findings of this thesis demonstrate the enormous potential of using Ψ_stem as a tool for monitoring plant water use and magnitudes of water stress. In combination with critical levels or thresholds related to both reversible and irreversible loss of function in plants identified here, optical dendrometers provide a continuous record of plant performance. These findings will have substantial implications for both agricultural and natural ecosystems. Continuous monitoring of Ψ_stem with optical dendrometers has the potential to vastly improve the ability of a diversity of crop managers to quantify the dynamic water requirements of crop species through monitoring of hydration status in situ and inform the risk or degree of damage in both vegetative and reproductive organs under field conditions. Employing the monitoring technique which I present here will help to improve irrigation decision making and forecast the likelihood of crop failure, especially under condition of limited water supply.

The findings of this thesis also greatly underscore the need to incorporate Kr dynamics and flower vulnerability into dynamic global vegetation models to better predict the impact of future climate on carbon gain and species generational survival or plant reproduction success, respectively.