2.1. Discrepancies between theoretical and experimental measurements of xylem conductance

The Hagen–Poiseuille equation has been used to estimate xylem conductance and sap flow rate for many decades (Doussan, Reference Doussan1998; Frensch & Steudle, Reference Frensch and Steudle1989; Landsberg & Fowkes, Reference Landsberg and Fowkes1978). In its simplest form, the axial hydraulic conductance (m⁴ s ⁻¹ MPa⁻¹) of a single vessel element can be calculated by:

$$\begin{align*}Lax=\pi \kern0.24em {r}^4\div 8\;\eta\end{align*}$$

where r is the radius of a xylem vessel (m) and $\eta$ is the viscosity of water (MPa s ⁻¹ ). This method has the great advantage of being accessible to a wide range of labs equipped with a microscope, which enables visualization of root cross-sections. However, it has been observed on many occasions that the theoretical Hagen–Poiseuille computation of xylem conductance is larger than the conductance measured experimentally (Bouda et al., Reference Bouda, Windt, McElrone and Brodersen2019; Boursiac et al., Reference Boursiac, Pradal, Bauget, Lucas, Delivorias, Godin and Maurel2022; Frensch & Steudle, Reference Frensch and Steudle1989; Landsberg & Fowkes, Reference Landsberg and Fowkes1978). A first example comes from measurements of the axial hydraulic properties of small, unbranched excised root segments using root pressure probes, in which loss of resistance before and after cutting a segment is used to infer the axial conductance of the removed portion (Frensch & Steudle, Reference Frensch and Steudle1989; Meunier et al., Reference Meunier, Zarebanadkouki, Ahmed, Carminati, Couvreur and Javaux2018). Using this method, Landsberg and Fowkes (Reference Landsberg and Fowkes1978) suggested that the actual axial conductance of grass roots was two to three times smaller than the theoretical one. Furthermore, Frensch and Steudle (Reference Frensch and Steudle1989) observed up to a five fold discrepancy in the upper part (>14 cm) of single maize roots, where measured resistance exceeded theoretical estimates.

More recently, the ‘Cut and Flow’ method was developed by Boursiac et al. (Reference Boursiac, Pradal, Bauget, Lucas, Delivorias, Godin and Maurel2022) to determine simultaneously both radial and axial conductivities of a fully developed root system. This ‘model-assisted’ phenotyping combines the use of the root hydraulic architecture model ‘Hydroroot’ (Boursiac et al., Reference Boursiac, Pradal, Bauget, Lucas, Delivorias, Godin and Maurel2022) and measurements of the whole root-system conductance with a pressure chamber (Boursiac et al., Reference Boursiac, Pradal, Bauget, Lucas, Delivorias, Godin and Maurel2022). The pressure-induced sap flow of one entire root system was measured in intact plants and after successive cuts from the tips. After reconstitution of the exact architecture, radial and axial hydraulic properties of the RSA were obtained from an optimization procedure of the Hydroroot model parameters in order to match the various sap flow measurements. Boursiac et al. (Reference Boursiac, Pradal, Bauget, Lucas, Delivorias, Godin and Maurel2022) measured axial conductance values four- to sixfold lower than theoretical estimates along the entire Arabidopsis roots and showed through simulations that even a small decrease in xylem conductivity affects sap flow throughout the root system, with varying impact depending on the distance from the base.

Magnetic resonance imaging has also been used to investigate the functional status of the xylem in planta (Bouda et al., Reference Bouda, Windt, McElrone and Brodersen2019; Buy et al., Reference Buy, Le Floch, Tang, Sidiboulenouar, Zanca, Canadas, Nativel, Cardoso, Alibert, Dupont, Ambard, Maurel, Verdeil, Bertin, Goze-Bac and Coillot2018). In grapevine, this non-destructive technique was combined with modelling and sap flow measurements to study the xylem network within a stem segment (Bouda et al., Reference Bouda, Windt, McElrone and Brodersen2019). Simulations using Hagen–Poiseuille equation overestimated flow rates in larger vessels and underpredicted it in smaller vessels. This observation is likely to be true for roots as well, highlighting the complexity of xylem conductive tissues.

2.2. Influence of topology on axial conductance

Evidence showing that axial conductance is often nearly an order of magnitude lower than theoretical estimates raises the question of whether it is truly non-limiting for root water uptake. In fact, it has been proposed that the interplay between xylem vessel morphology and the topological organization of the root system imposes constraints on axial water flow (Bouda et al., Reference Bouda, Brodersen and Saiers2018). Bouda et al. (Reference Bouda, Brodersen and Saiers2018) showed that, in some cases, root system conductance is more sensitive to axial than radial conductance in absorbing roots, demonstrating that the extent to which axial conductance limits water uptake depends strongly on root network topology and on root length (Figure 1). Furthermore, large differences in xylem water flow have been reported among root types (Meunier et al., Reference Meunier, Zarebanadkouki, Ahmed, Carminati, Couvreur and Javaux2018), and along the axes of primary roots (Pierret et al., Reference Pierret, Doussan and Pagès2006), which can lead to strong variations in water transport limitations depending on the root system architecture. Limitations in axial conductance caused by reductions in metaxylem vessel number and diameter with increasing root depth, for instance, may ultimately influence root water uptake (Clément et al., Reference Clément, Schneider, Dresbøll, Lynch and Thorup-Kristensen2022; Strock et al., Reference Strock, Burridge, Niemiec, Brown and Lynch2021).

Figure 1.

Overview of architectural, anatomical and functional differences in terms of water flow between dicot and monocot-like root systems. The anatomical cross-sections represent different maturation stages along the primary root axis.

Comparative analyses of the distinct root architectures of dicot and monocot species offer deeper insights into how axial limitations emerge in different root systems. Typically, dicots (such as soybean in Figure 2) have a taproot system, with a primary root bearing the complete root system. Their secondary roots tend to be long and branched to the second or third order. All water taken up by the root system is funnelled towards the unique primary root, which must assure its transport to the shoot. Secondary growth and the development of secondary metaxylem in the primary root contribute, in part, to enabling this function. Monocot species (such as wheat in Figure 2) tend to form fibrous root systems. These are composed of a multitude of first-order root axes, originating either from the seed (primary and seminal roots) or shoot nodes (brace and crown roots) that do not undergo secondary growth. They typically bear lateral roots that are relatively short, usually without higher-order roots. In such root systems, water uptake can follow multiple independent pathways to reach the shoot. Therefore, contrast in topology and anatomy between monocot and dicot species are two important factors that may contribute to differences in the relative importance of axial conductance within the whole root system conductance.

Figure 2.

The effect of axial conductance (k _x) and radial conductivity (k _r) changes on whole root system conductance (K _rs) for dicot (soybean, left panels) and monocot (wheat, right panels) species. (a) Effect of k _x and k _r changes (increase or decrease) on modelled K _rs at different root system ages. Based on a default parametrization (Doussan et al., Reference Doussan1998, Baca Cabrera et al., Reference Baca Cabrera, Vanderborght, Couvreur, Behrend, Gaiser, Nguyen and Lobet2024), k _x or k _r were lowered or increased by one order of magnitude for all root types. (b) Heat map with the effect of k _r/k _x ratio changes on K _rs at different root system ages (yellow indicates higher, magenta lower K _rs values). (c) Contrasting root system architecture at the end of the simulations.

To illustrate the effect of these contrasted topologies on the whole root system conductance (Krs) and test potential axial conductance limitations to water transport, we simulated root system architectures of a dicot (soybean) and a monocot (wheat) using the whole-plant model CPlantBox (Giraud et al., Reference Giraud, Gall, Harings, Javaux, Leitner, Meunier, Rothfuss, van Dusschoten, Vanderborght, Vereecken, Lobet and Schnepf2023), with identical segment-scale parametrization of radial (k _r) and axial hydraulic (k _x) properties (Baca Cabrera et al., Reference Baca Cabrera, Vanderborght, Couvreur, Behrend, Gaiser, Nguyen and Lobet2024; Figure 2 and Supplementary Methods S1). As shown in Meunier et al. (Reference Meunier, Heymans, Draye, Couvreur, Javaux and Lobet2020) and Baca Cabrera et al. (Reference Baca Cabrera, Vanderborght, Couvreur, Behrend, Gaiser, Nguyen and Lobet2024, Reference Baca Cabrera, Vanderborght, Boursiac, Behrend, Gaiser, Nguyen and Lobet2025), Krs quickly reaches a maximum value in both species, even though the root system is still growing. A sensitivity analysis was performed by modifying k _x and k _r by an order of magnitude relative to the default parametrization. Interestingly, increasing or decreasing k _x and k _r had a similar impact on K _rs, challenging the common paradigm that k _r is the predominant limitation to root water uptake (Figure 2a). This effect was particularly pronounced in soybean, where changes in k _r and k _x affected K _rs almost identically. In contrast, the effect was less pronounced in wheat, where k _r changes had a stronger influence on K _rs than k _x changes, although axial flow limitation was still present. This difference likely reflects that in dicots, where all water must pass through the primary root to reach the shoot, the axial conductance of that root can quickly become a bottleneck. In monocots, changes in k _x have less impact on the root system conductivity, with k _r remaining the main limitation. Since each first-order axis transports only a fraction of water to the shoot, their individual importance is lower. Redundancy root axes and continuous growth maintain overall water uptake capacity.

Additionally, the simulations indicate a non-linear effect of the radial-to-axial conductivity ratio (k _r/k _x) on K _rs. Across both dicot and monocot architectures, the highest K _rs values are observed at intermediate k _r/k _x ratios, while lower or higher ratios result in decreased K _rs (Figure 2b). These results underscore the importance of a coordinated balance between radial and axial hydraulic properties for optimal water uptake. They are in line with previous studies emphasizing the need for functional integration of both components in determining root water uptake capacity (Bouda et al., Reference Bouda, Brodersen and Saiers2018), and further challenge the common assumption that root water uptake is limited primarily by radial conductivity alone.

3.1. Influence of root axial conductance on plant water use

The plant hydraulic network can be described using a demand and supply scheme, in which the water supply from roots sustains shoot transpiration (Vadez et al., Reference Vadez, Grondin, Chenu, Henry, Laplaze, Millet and Carminati2024). In this scheme, if the water supply cannot match the water demand, stomatal closure will occur to avoid a large drop in leaf water potential (Tardieu et al., Reference Tardieu, Draye and Javaux2017). Stomatal sensitivity to declining leaf water potential has been used to classify plant water-use strategies, distinguishing water savers (or isohydric plants), which close their stomata in response to small drops in leaf water potential, from water spenders (or anisohydric plants), which tolerate larger drops before closure (Tardieu & Simonneau, Reference Tardieu and Simonneau1998). It has been hypothesized that reduced xylem diameter and the resulting decline in root axial and overall conductance are associated with water-saving strategies by limiting transpiration more quickly as soil dries or evaporative demand rises, thereby promoting more parsimonious water use and greater drought tolerance (Figure 3b; Burridge et al., Reference Burridge, Grondin and Vadez2022; Vadez et al., Reference Vadez, Grondin, Chenu, Henry, Laplaze, Millet and Carminati2024). In crops such as sorghum, pearl millet or maize, a direct link was established between water savings during the vegetative stage through transpiration restriction in response to the increasing evaporative demand, and yield maintenance under drought (Cooper et al., Reference Cooper, Gho, Leafgren, Tang and Messina2014; Sinclair et al., Reference Sinclair, Hammer and Van Oosterom2005; Vadez et al., Reference Vadez, Kholová, Yadav and Hash2013). However, direct evidence linking root hydraulics, and in particular root axial conductance, to whole plant water use remains sparse. This may be explained by the complexity of untangling the effect of root axial conductance from the whole plant hydraulics through both empirical and modelling approaches as water capture, flow and use integrate multiple architectural, anatomical and functional components operating throughout the plant (Burridge et al., Reference Burridge, Grondin and Vadez2022; Klein et al., Reference Klein, Schneider, Perkins, Brown and Lynch2020; Koehler et al., Reference Koehler, Schaum, Tung, Steiner, Tyborski, Wild, Akale, Pausch, Lueders, Wolfrum, Mueller, Vidal, Vahl, Groth, Eder, Ahmed and Carminati2023; Strock et al., Reference Strock, Burridge, Niemiec, Brown and Lynch2021). Another layer of complexity arises from the interaction between plant hydraulics and the hydraulics of the soil and rhizosphere, with the latter significantly influencing transpiration as the soil dries (Cai et al., Reference Cai, Ahmed, Abdalla and Carminati2022; Javaux & Carminati, Reference Javaux and Carminati2021; Koehler et al., Reference Koehler, Moser, Botezatu, Murugesan, Kaliamoorthy, Zarebanadkouki, Bienert, Bienert, Carminati, Kholová and Ahmed2022; Sperry et al., Reference Sperry, Stiller and Hacke2003). It follows that developing approaches to better understand the crop hydraulic architecture in its environment is important for defining appropriate strategies orienting drought breeding programs based on root hydraulics improvements.

3.2. Influence of xylem vessel diameter on cavitation

Increasing capillary tension within the xylem may lead to cavitation damage. Cavitation represents the breaking of intermolecular water bonds that create embolisms within the xylem vessels and block the flow of water (Venturas et al., Reference Venturas, Sperry and Hacke2017). Embolism may further spread within the xylem conduit, causing hydraulic network failure and ultimately plant mortality (Mantova et al., Reference Mantova, Cochard, Burlett, Delzon, King, Rodriguez-Dominguez, Ahmed, Trueba and Torres-Ruiz2023). Xylem vulnerability to cavitation is typically measured as the cumulative proportion of xylem conductivity lost versus water tension, resulting in values such as P₅₀ that represent the xylem water potential threshold at which 50% of the xylem conductivity is lost due to cavitation (Choat et al., Reference Choat, Jansen, Brodribb, Cochard, Delzon, Bhaskar, Bucci, Feild, Gleason, Hacke, Jacobsen, Lens, Maherali, Martínez-Vilalta, Mayr, Mencuccini, Mitchell, Nardini, Pittermann and Zanne2012). For instance, the P₅₀ in wheat leaves is around −2.87 MPa (Johnson et al., Reference Johnson, Jordan and Brodribb2018), but it varies between organs, with leaves being more vulnerable than roots and peduncles (Harrison Day et al., Reference Harrison Day, Johnson, Tonet, Bourbia, Blackman and Brodribb2023). In wheat roots, xylem vulnerability varies largely among root types, with small lateral roots appearing more susceptible to cavitation than larger crown roots (Harrison Day et al., Reference Harrison Day, Johnson, Tonet, Bourbia, Blackman and Brodribb2023). In wet soils, it was proposed that grasses can generate sufficient positive root pressure to repair xylem embolism overnight, should they experience cavitation during the day due to excessive transpiration (Gleason et al., Reference Gleason, Wiggans, Bliss, Young, Cooper, Willi and Comas2017; Sperry et al., Reference Sperry, Stiller and Hacke2003). In drying soils, stomatal closure often precedes substantial losses in xylem conductivity, thereby preventing damage from cavitation (Brodribb & McAdam, Reference Brodribb and McAdam2017; Cochard et al., Reference Cochard, Coll, Le Roux and Améglio2002; Martin-StPaul et al., Reference Martin-StPaul, Delzon and Cochard2017). Furthermore, annual crops are usually grown during rainy seasons, soil water potential rarely reaches the permanent wilting point caused by soil hydraulic conductivity loss (Carminati & Javaux, Reference Carminati and Javaux2020). Therefore, cavitation is generally not considered critical in crops (Corso et al., Reference Corso, Delzon, Lamarque, Cochard, Torres-Ruiz, King and Brodribb2020). Yet, the safety margin, defined as the difference between the minimum midday water potential a plant can experience and the P₅₀ , is often very narrow (Choat et al., Reference Choat, Jansen, Brodribb, Cochard, Delzon, Bhaskar, Bucci, Feild, Gleason, Hacke, Jacobsen, Lens, Maherali, Martínez-Vilalta, Mayr, Mencuccini, Mitchell, Nardini, Pittermann and Zanne2012; Franklin et al., Reference Franklin, Fransson, Hofhansl, Jansen and Joshi2023). With the increasing frequency of drought and heat stress events, crops are likely to operate closer to this margin more frequently, potentially increasing the risk of cavitation (Brodribb et al., Reference Brodribb, Powers, Cochard and Choat2020; Buckley, Reference Buckley2005). Decreasing xylem diameter is generally thought to reduce the risk of cavitation (Figure 3c; Jacobsen et al., Reference Jacobsen, Brandon Pratt, Venturas, Hacke and Lens2019), although it would also decrease potential water flow sustaining transpiration, hence photosynthesis and growth. Other xylem features, such as xylem length, pit size and density or pit membrane permeability, may also play important roles in resistance to cavitation (Bouda et al., Reference Bouda, Windt, McElrone and Brodersen2019; Brodersen et al., Reference Brodersen, McElrone, Choat, Lee, Shackel and Matthews2013; Venturas et al., Reference Venturas, Sperry and Hacke2017). Xylem network organization was linked to a significant increase in stem resistance to embolism spread in grapevines, for instance (Wason et al., Reference Wason, Bouda, Lee, McElrone, Phillips, Shackel, Matthews and Brodersen2021). A better understanding of the effects of water deficit on cavitation in crops and the traits potentially affecting cavitation resistance is necessary in future research to clearly define in which drought scenarios hydraulic failure may be problematic for crop productivity. Recent development of non-invasive optical methods for observation of cavitation may represent useful tools for exploring the diversity of xylem vulnerability, potentially contributing to drought adaptation (Brodribb et al., Reference Brodribb, Skelton, McAdam, Bienaimé, Lucani and Marmottant2016).