Periodic Reporting for period 1 - ThePlantWaterPump (The Plant Water Pump)
Reporting period: 2022-10-01 to 2025-03-31
With global warming, climate zones are projected to shift poleward, and the frequency and intensity of droughts to increase, driving threats to crop production and ecosystems. Plant hydraulic traits play major roles in coping with such droughts, and process-based plant hydraulics (water flowing along decreasing pressure Ψp or total water potential Ψtot gradients) has newly been implemented in land surface models.
An enigma reported for the past 35 years is the observation of water flowing along increasing water potential gradients across roots. By combining the most advanced modelling tool from the emerging field of plant micro-hydrology with pioneering cell solute mapping data, I found that the current paradigm of water flow across roots of all vascular plants is incomplete: it lacks the impact of solute concentration (and thus negative osmotic potential Ψo) gradients across living cells. This gradient acts as a water pump as it reduces water tension without loading solutes in plant vasculature (xylem). Importantly, water tension adjustments in roots may have large impacts in leaves due to the tension-cavitation feedback along stems.
With The Plant Water Pump I will combine for the first time cutting-edge osmotic mapping and micro-hydrological modelling approaches to (1) characterize water status and osmotic responses to water deficit in diverse crop and tree species, (2) revolutionize the current paradigm of plant water uptake, and (3) increase the accuracy of plant water status functions for land surface models. By creating a continuum between key cell-scale variables and plant-scale water fluxes, this project lays the foundations for future multidisciplinary research encompassing plant physiology and ecohydrology. Besides its groundbreaking contribution to the fundamental understanding of plant water relations, this effort embodies a much-needed step toward the accurate forecasting of land water fluxes and decision support under future climates.
An enigma reported for the past 35 years is the observation of water flowing along increasing water potential gradients across roots. By combining the most advanced modelling tool from the emerging field of plant micro-hydrology with pioneering cell solute mapping data, I found that the current paradigm of water flow across roots of all vascular plants is incomplete: it lacks the impact of solute concentration (and thus negative osmotic potential Ψo) gradients across living cells. This gradient acts as a water pump as it reduces water tension without loading solutes in plant vasculature (xylem). Importantly, water tension adjustments in roots may have large impacts in leaves due to the tension-cavitation feedback along stems.
With The Plant Water Pump I will combine for the first time cutting-edge osmotic mapping and micro-hydrological modelling approaches to (1) characterize water status and osmotic responses to water deficit in diverse crop and tree species, (2) revolutionize the current paradigm of plant water uptake, and (3) increase the accuracy of plant water status functions for land surface models. By creating a continuum between key cell-scale variables and plant-scale water fluxes, this project lays the foundations for future multidisciplinary research encompassing plant physiology and ecohydrology. Besides its groundbreaking contribution to the fundamental understanding of plant water relations, this effort embodies a much-needed step toward the accurate forecasting of land water fluxes and decision support under future climates.
In the framework of model development tasks (WP1), 3 research articles on the topic of developing new upscaled solutions of water flow in the soil around root segments and analysing their quantitative impact on plant scale water availability were published with international collaborators (Vanderborght et al., 2024a,b; Schnepf et al., 2023). Such solutions will be central in accounting for the impact of soil hydraulics in experiments from WP2 and WP4. A new module was also implemented in the root hydraulic anatomy model MECHA, that allows for radially asymmetric patterns of plasmodesmata opening and of cell membrane permeability to be simulated. These parameters are key factors to fit heavy water distribution in roots observed by collaborators at University of Nottingham, and would impact the efficiency of the water pump according to our calculations. The manuscript in preparation will be submitted to Science end of 2024.
In the framework of model parametrization tasks (WP3), a collaborative review article on the diversity of root hydraulic parameter values across plant functional types and water deficit regimes was published (Baca Cabrera et al., 2024). This publication came with an online open access library of plant hydraulic parameters, which will be fed with the values estimated in WP2, WP3 and WP4.
In the framework of WP2, soil and plant hydraulic responses were monitored in water deficit experiments in maize and tomato, as well as trees including almond, pistachio and olive in field. In this report, we put the focus on results from soil-grown maize plants (analyses are ongoing for other plants, like tomato in hydroponics under PEG and NaCl treatments). Soil matric potential (ψsoil, using TEROS21 sensors), leaf water potential (ψleaf, using PSY1 sensors), and plant transpiration rate (Tr, using weighing scales) were continuously monitored under 3 treatments with target ψsoil values of -0.025 MPa (T1), -0.4 MPa (T2) and -0.8 MPa (T3), respectively, followed by a recovery phase for the droughted plants. The water deficit treatments induced lower transpiration rate and leaf water potential than in well-watered plants, while upon recovery they progressively converged. Specific algorithms were also created to clean and filter Tr and ψleaf data (see section 1.2). Vapor pressure deficit (VPD) and temperature were also recorded during the experiment.
Soil and plant osmotic responses to water deficit were also monitored at each harvest stage in WP2. With the soil getting drier, soil osmotic potential decreased in the T2 and T3 treatments, while differences disappeared in the recovery phase. The soil osmotic potential was more negative than expected, even under well-watered conditions (T1), which could be due to the method used to extract the soil solutes. In future experiments, centrifuge and pressure plates methods will be tested as alternatives to the current water extraction protocol. Osmotic potential was also monitored in bulk roots. The only significant reduction in root osmotic potential was observed between mild water deficit conditions versus root osmotic potential before and after the drought phase. Xylem sap osmotic potential did not significantly differ between well-watered and water deficit conditions and also remained stable during the recovery phase. The difference with the pre-drought phase could be due to the fact that the xylem sap sampling duration and volume was shorter in the pre-drought phase. The methods will be investigated further in the coming months.
Other plant responses measured at harvest times were the root hydraulic conductance (Kr, kg kPa-1 s-1), which is essential for water supply to the shoot, and the root hydraulic conductivity (Lpr). From pre-drought to recovery phases, Kr and Lpr tended to decrease, while no significant effect of the treatments was observed.
Overall, similar results were observed in the tomato experiments (Lpr decreasing with age, without impact from the treatment; bulk root osmotic potential becoming more negative under water deficit; reduced transpiration rate and leaf water potential under water deficit). A manuscript with data from both maize and tomato experiments is in preparation. It will present the observed plant hydraulic and osmotic responses to water deficit, as well as inverse modelling results meant to test a hypothesis on the temporal dynamics of variables measured discretely at harvest times.
In the framework of model parametrization tasks (WP3), a collaborative review article on the diversity of root hydraulic parameter values across plant functional types and water deficit regimes was published (Baca Cabrera et al., 2024). This publication came with an online open access library of plant hydraulic parameters, which will be fed with the values estimated in WP2, WP3 and WP4.
In the framework of WP2, soil and plant hydraulic responses were monitored in water deficit experiments in maize and tomato, as well as trees including almond, pistachio and olive in field. In this report, we put the focus on results from soil-grown maize plants (analyses are ongoing for other plants, like tomato in hydroponics under PEG and NaCl treatments). Soil matric potential (ψsoil, using TEROS21 sensors), leaf water potential (ψleaf, using PSY1 sensors), and plant transpiration rate (Tr, using weighing scales) were continuously monitored under 3 treatments with target ψsoil values of -0.025 MPa (T1), -0.4 MPa (T2) and -0.8 MPa (T3), respectively, followed by a recovery phase for the droughted plants. The water deficit treatments induced lower transpiration rate and leaf water potential than in well-watered plants, while upon recovery they progressively converged. Specific algorithms were also created to clean and filter Tr and ψleaf data (see section 1.2). Vapor pressure deficit (VPD) and temperature were also recorded during the experiment.
Soil and plant osmotic responses to water deficit were also monitored at each harvest stage in WP2. With the soil getting drier, soil osmotic potential decreased in the T2 and T3 treatments, while differences disappeared in the recovery phase. The soil osmotic potential was more negative than expected, even under well-watered conditions (T1), which could be due to the method used to extract the soil solutes. In future experiments, centrifuge and pressure plates methods will be tested as alternatives to the current water extraction protocol. Osmotic potential was also monitored in bulk roots. The only significant reduction in root osmotic potential was observed between mild water deficit conditions versus root osmotic potential before and after the drought phase. Xylem sap osmotic potential did not significantly differ between well-watered and water deficit conditions and also remained stable during the recovery phase. The difference with the pre-drought phase could be due to the fact that the xylem sap sampling duration and volume was shorter in the pre-drought phase. The methods will be investigated further in the coming months.
Other plant responses measured at harvest times were the root hydraulic conductance (Kr, kg kPa-1 s-1), which is essential for water supply to the shoot, and the root hydraulic conductivity (Lpr). From pre-drought to recovery phases, Kr and Lpr tended to decrease, while no significant effect of the treatments was observed.
Overall, similar results were observed in the tomato experiments (Lpr decreasing with age, without impact from the treatment; bulk root osmotic potential becoming more negative under water deficit; reduced transpiration rate and leaf water potential under water deficit). A manuscript with data from both maize and tomato experiments is in preparation. It will present the observed plant hydraulic and osmotic responses to water deficit, as well as inverse modelling results meant to test a hypothesis on the temporal dynamics of variables measured discretely at harvest times.
In substance, we consider the preliminary results in maize and tomato as extremely promising for two main reasons: (i) plant transpiration was maintained under strong water deficit despite the reversed water potential gradient (e.g. xylem Ψtot approached -1.3 MPa while soil Ψtot reached below -1.5 MPa), and (ii) osmotic potential appears to be clearly more negative in the cortex than in the stele parenchyma (Fig. 5b), which supports our new theory to solve the uphill water flow enigma (Action 3.3).
Besides the 4 publications in peer-reviewed scientific journals (section 1.1) and related to WP1 and WP3, the team members had the chance to present their work on the ERC project at 9 conferences and workshops in 2023 and 2024 (listed below), as well as in multiple informal presentations (e.g. within the research groups at UCLouvain and UCDavis, not listed below). These dissemination activities are central in order to nourish the project through collective intelligence and networking.
Besides the 4 publications in peer-reviewed scientific journals (section 1.1) and related to WP1 and WP3, the team members had the chance to present their work on the ERC project at 9 conferences and workshops in 2023 and 2024 (listed below), as well as in multiple informal presentations (e.g. within the research groups at UCLouvain and UCDavis, not listed below). These dissemination activities are central in order to nourish the project through collective intelligence and networking.