Experimental Evidence of the Impacts of Different Root-Pruning Intensities on Post-Transplant Water Potential and Xylem Hydraulic Conductivity in Two Tree Species

Plant transplantation is a technique used to artificially position plants for purposes such as stable crop production, rapid forest regeneration, and urban greening. However, there remains a pressing need to develop labor-efficient and economically viable transplantation methods (Korbik et al. 2025). In the case of trees, saplings for transplantation may be cultivated over several years in containers or in field nurseries before being relocated to different regions (Struve 2009), making them especially susceptible to transplant shock.

Transplant shock is generally induced by the loss of a large portion of the root system during transplanting (i.e., root pruning) (Watson 1985). In particular, during the transplantation of balled and burlapped stock, up to 95% of the total root system may be lost (Harris and Bassuk 1993), which has been reported to result in a substantial reduction in water uptake after transplantation (Poni et al. 1992; Wang et al. 2014; Dong et al. 2016; Benson et al. 2019). Currently, a variety of transplantation methods are used, including bare-root stock, balled and burlapped stock, container-grown stock, and fabric container stock. Among these methods, container-grown stock is relatively expensive but has the advantage of allowing transplantation without root pruning (Harris and Bassuk 1993). On the other hand, the other methods involve varying levels of root loss depending on the technique, but all require root pruning and therefore may impose substantial stress on saplings. Consequently, no single transplantation method is optimal for all situations. Survival rates after transplantation and recovery from transplant shock are thought to be influenced by multiple factors, including tree species, environmental conditions, physiological status of the saplings, timing of transplantation, cultivation management, and root system type (Bevington and Castle 1985; Fare et al. 1985; Gilman 1990; Watson 2005). Therefore, it is necessary to quantitatively evaluate the effects of changes to the water-conducting pathway caused by root loss and the susceptibility to water stress arising from damage to the xylem water transport system. Accordingly, this study aimed to establish a method for quantitatively evaluating the effects of different intensities of root loss/removal on the water transport system.

Woody plants draw water up from the soil through their xylem conduits (tracheids and vessels). Since this water uptake is a consequence of a negative pressure that pulls the water upwards (Tyree and Zimmermann 2002; Umebayashi et al. 2016), the water column in the conduits is vulnerable to embolism. Therefore, at the time of root pruning, it is considered that embolism occurs at the cut sites and that the embolized area expands due to an increase in xylem negative pressure resulting from reduced water uptake caused by the root loss.

The vulnerability of xylem water transport pathways can be quantitatively assessed using a vulnerability curve, which illustrates the decline in hydraulic conductance with decreasing xylem water potential (Sperry and Tyree 1988; Bush et al. 2010; Schreiber et al. 2013). By examining the reduction in hydraulic conductance in transplanted individuals with root pruning, it is expected that tree species with high transplant tolerance can be identified.

In this study, we conducted a transplant experiment on two species growing in the Northern regions of Japan. As shown in Figure 1, 3 levels of root-pruning intensity were established based on basal root diameter, and the effects on water transport after transplantation were examined. One species was Betula maximowicziana, a broadleaf tree commonly found in riparian forests, and the other was Larix kaempferi, a deciduous conifer. These two species were selected in order to determine the effects of root pruning and water transport on transplant shock that comes from sapling production and planting for a coniferous species used in afforestation in Japan and a broadleaf species vulnerable to water stress. To establish a method for evaluating water transport function in trees transplanted after root pruning, we focused on (1) investigating the influence of root pruning on hydraulic conductivity (K_s) and mortality rate, and (2) examining relationships between species vulnerability to embolism, midday water potential, and the survival of saplings with different levels of root pruning.

• Download figure
• Open in new tab
• Download PowerPoint

Figure 1.

A methodological diagram of root pruning before the transplant (see also Figures S1A–C for representative photographs). (A*) The maximal diameter of the stem at ground level. (A) The dotted line shows the cut position for a 100% root loss on the main root (the same diameter as A*). (B) The dotted line shows the cut position for a 75% root loss (three-quarters of the diameter of A). (C) The dotted line shows the cut position for a 50% root loss (one-half diameter of A). (a) The dotted line shows the position for a 100% root loss on a lateral root (the base diameter of that lateral root). (b) The dotted line shows the cut position for a 75% root loss on a lateral root (three-quarters of the diameter of that lateral root). (c) The dotted line shows the position for a 50% root loss on a lateral root (one-half diameter of that lateral root). (s) Finer roots with diameters below 1 mm.

To evaluate species-specific vulnerability to embolism, we calculated the slopes of vulnerability curves and the water potential at which K_s decreased by 50% (P₅₀) in saplings of both species using the centrifuge method. Using these parameters, we tested the hypothesis that the slopes of vulnerability curves measured for each species and the magnitude of declines in K_s observed in transplanted saplings subjected to different root-pruning intensities could serve as useful indicators for selecting tree species suitable for transplantation. We predicted that B. maximowicziana, a riparian species that forms vessels, would exhibit increased declines in K_s and higher mortality with increasing root-pruning intensity. In contrast, L. kaempferi, which forms tracheids and has greater tolerance to water stress, was expected to show little decline in K_s even at higher levels of root pruning and consequently no mortality. In addition, we examined whether periodic measurements of water potential in both species could be used to quantitatively assess water stress conditions.

The vulnerability curves for stem segments of B. maximowicziana had significantly higher K_{s initial} (by a factor of 3.0) (Figure 2A) and were more vulnerable to embolism than the stem segments of L. kaempferi (Figure 2B). The percent loss of hydraulic conductivity (PLC) at a pressure of –1.5 MPa showed 23.7% ± 10.1% (mean ± SD) in the stem segments of B. maximowicziana and 3.0% ± 0.5% in L. kaempferi segments. The PLC of B. maximowicziana increased drastically at pressures lower than –1.5 MPa (85.1% ± 13.0% at a pressure of –2.0 MPa), while the PLC of L. kaempferi showed almost no increase up to –2.0 MPa (2.8% ± 8.0% at a pressure of –2.0 MPa). The P₅₀ values of B. maximowicziana and L. kaempferi were –1.72 MPa and –2.90 MPa, respectively (Table 2).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 2.

Hydraulic conductivity characteristics of the stems of the two species used in the centrifuge method. (A) Initial hydraulic conductivity divided by the stem cross-sectional area (K_{s initial}) in two species (t-test, P = 0.05). (B) Vulnerability curves of the two species. (•) indicates Larix kaempferi and (o) indicates Betula maximowicziana. Conductive capacity in the vulnerability curves is expressed as percentage loss of conductivity (PLC) relative to the initial conductivity shown in (A). Error bars represent ±1 SD for n = 5 samples. Data were fitted with a Weibull function.

Dead saplings of either species were not detected in June (Figure S2). However, many fallen leaves were observed in July in saplings of B. maximowicziana with 100% root loss (Figure S2C). On 2016 August 2, midday xylem water potential reached –1.6 MPa ± 0.3 MPa in saplings of B. maximowicziana with 75% root loss (Table 3) (Figure S2D). All saplings with 100% root loss had died by that time. The mortality rate recorded on October 3 differed widely among treatments in both species. In B. maximowicziana, 3 saplings of 75% root loss and 4 saplings of 50% root loss survived. Thus, for B. maximowicziana, mortality rate increased with the increased level of root pruning. The water potential in the surviving B. maximowicziana saplings with 75% root loss recovered at the pressure of –1.3 MPa ± 0.1 MPa. On the other hand, the decrease in water potential of L. kaempferi saplings during the experimental period was not large. All saplings of L. kaempferi survived irrespective of the level of root pruning (Table 1).

New roots were observed in all surviving saplings. The loss in K_s was detected in surviving B. maximowicziana saplings with pruned roots, but the decrease did not occur in L. kaempferi, regardless of the level of root pruning (Figure 3). In B. maximowicziana saplings, no significant difference in K_s was observed between 50% and 75% root loss (Figure 3).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 3.

Hydraulic conductivity of surviving saplings of the two species after transplanting under each root-pruning treatment. Error bars are ±1 SD for n = 3 to 5 samples. Dieback of all saplings was induced in 100% root loss B. maximowicziana saplings. To evaluate differences among root-pruning levels within each species, analysis of variance (ANOVA) was conducted using R statistical software (R Foundation, Vienna, Austria), followed by Tukey’s honestly significant difference (HSD) test as a post hoc analysis. For B. maximowicziana, different letters above each column indicate significant differences according to Tukey’s HSD test (P = 0.05). No significant differences were observed in L. kaempferi.

In this study, at one month after transplantation, the midday water potential values of saplings of neither species showed a decline sufficient to be associated with a reduction in K_s (Table 3). However, the decrease in midday water potential was induced in many saplings of B. maximowicziana, and saplings subjected to higher levels of root pruning tended to exhibit lower water potential values in August (Table 3). In addition, many saplings could not survive the later stages of this experiment, and the mortality rate increased with the increased level of root pruning in B. maximowicziana. In contrast, no reduction in K_s was observed in any of the transplanted L. kaempferi saplings. In healthy plants, water moves from roots to leaves as a consequence of pressure gradients, and water transport in saplings is mainly controlled by leaves. In transplanted saplings whose roots were subjected to pruning, a sufficient water supply cannot be achieved until the water transport pathway from soil to roots has recovered, and serious decreases in stomatal conductance and water potential are induced by increasing the level of root pruning (Benson et al. 2019). This concurs with the results of our research in which the transplanted saplings whose roots were pruned in mid-May were more vulnerable to water stress during midsummer. In addition, the number of embolized xylem conduits may have increased on the side that the plant was cut while root pruning (Torres-Ruiz et al. 2015).

Post-transplantation measurements of water potential in B. maximowicziana indicated that xylem negative pressure reached its maximum in August. Moreover, the decline in water potential became more pronounced with increasing levels of root pruning, potentially leading to a further reduction in K_s. Furthermore, once xylem dysfunction occurred in the stem, little recovery of K_s was observed even when water stress was alleviated in October (Table 2). In contrast, in L. kaempferi, even saplings subjected to the 100% root-pruning treatment in August maintained water potentials of –1.2 MPa ± 0.2 MPa, and no reduction in K_s was observed (Table 3).

Results from the vulnerability curves revealed that, in B. maximowicziana, a loss in K_s occurred under pressure conditions below –1.5 MPa (Table 2) (Figure 2). In saplings subjected to the 75% root-pruning treatment, the mean water potential in August reached –1.6 MPa, which is consistent with the vulnerability curve indicating the occurrence of stem embolism formation (i.e., an increase in PLC). In contrast, in the vulnerability curve of L. kaempferi, a decrease in K_s was detected only below –2.5 MPa (Figure 2B). These results suggest that species-specific vulnerability curves are useful indicators for selecting tree species with high tolerance to transplantation. In species vulnerable to water stress, increasing levels of root pruning enhance sensitivity to summer water stress, resulting in higher mortality rates (Tables 1 and 3). Therefore, for such species, it is desirable to select transplanting techniques that minimize root loss during transplantation.

Urli et al. (2013) reported that the lethal threshold of drought stress in broadleaved species occurs around P₈₈. Among the surviving B. maximowicziana saplings subjected to root pruning, the lowest K_s was observed in a sapling from the 75% root-pruning treatment, with a value of 0.20 kg MPa⁻¹ m⁻¹ s⁻¹ (Figure 3). This corresponds to an 88% reduction relative to the control K_{s max} (i.e., the value after refilling following K_s measurement; 1.60 ± 0.29 kg MPa⁻¹ m⁻¹ s⁻¹). In addition, our data clearly showed that the K_{s max} of B. maximowicziana tended to be lower in saplings subjected to the 50% and 75% root-pruning treatments (0.64 ± 0.15 and 0.76 ± 0.55 kg MPa⁻¹ m⁻¹ s⁻¹, respectively) than in the control. It is well known that tyloses and/or gums can form within embolized vessels (De Micco et al. 2016), and the observed reduction in K_{s max} suggests that embolisms may have already formed in the stem at the time of root pruning. Therefore, in species vulnerable to water stress, a reduction in K_{s max} may occur in individuals planted after root pruning, and evaluating hydraulic function solely based on PLC may lead to an underestimation of the decline in water transport capacity. In this context, assessment based on K_s is useful for understanding the hydraulic properties of saplings after transplantation. Future studies incorporating spatial analyses of embolized areas in saplings after root pruning would further improve our understanding of the relationship between reductions in K_s and survival.

In conclusion, our methods, including root pruning, were shown to be effective for evaluating species-specific tolerance to water stress that arises after transplantation. In L. kaempferi, which forms tracheids and is relatively tolerant to water stress, no reduction in K_s or mortality was observed in saplings subjected to root pruning. In contrast, in B. maximowicziana, which forms vessels and is more vulnerable to water stress, declines in water potential during midsummer and increases in mortality were observed with increasing levels of root pruning. In B. maximowicziana, the reduction in K_s was suggested to be induced not only by hydraulic disruption at the time of root pruning but also by summer water stress. Therefore, as hypothesized, the slope of the vulnerability curve, the magnitude of K_s reduction after transplantation, and the monitoring of post-transplantation water potential are useful indicators for evaluating post-transplant survival potential and provide important information for species selection in planting practices. Species with higher tolerance to water stress tend to exhibit higher survival rates after transplantation. Although this study focused solely on the effects of root pruning, it is well known that multiple factors, including seasonal conditions, weather, soil properties, and post-transplant management interact in complex ways during actual transplantation practices (Harris and Bassuk 1993). Therefore, further studies adopting more comprehensive perspectives are required (Brancalion and Holl 2020).

The authors reported no conflicts of interest.