The understanding of processes affecting plant water availability has fundamental and applied implications. Recent studies have recognized the key role of roots in promoting acclimation to different types of stress; mainly through preferential growth and control of hydraulic properties that regulate transpiration . A better understanding of root response is, therefore, key for understanding water fluxes through the soil-plant-atmosphere continuum. Accordingly, here we examine the effect of root growth and plant hydraulic conductance on water availability for canopy transpiration of young walnut trees under different levels of water stress.The study was conducted from April 2015 to July 2015, using nine 8-month-old potted walnut trees cv. Chandler, grafted onto Paradox root stock in an experimental greenhouse at the University of California, Davis. Plants were grown in 0.02 m3 pots filled with a 1:3 mixture of a fine sand and organic compost. As the experiment was conducted over a short period and the plants were young, the size of the pots was considered suitable. Pots were kept covered with aluminum foil to avoid soil evaporation and their transparent walls were covered with plastic sheets that were black inside and white outside, to protect roots from light exposure. All pots were maintained at field capacity for at least a week before the beginning of each 10-days period experiment. Replicates were monitored over time due to the careful tracking of soil-plant properties and limited availability of leaf psychrometers and high precision weighing scales for all individuals. Hence,macetas 30l the experiment was replicated using three different plants per treatment monitored over 10- days in three different time periods , for a total of nine receiving one of the irrigation treatments and three control plants.
While temporal replications integrate the effect of different insolation and temperature conditions in the greenhouse at each 10-day sampling event, we expect to observe consistent shifts between T100, T75, T50 throughout the experiment.Stem water potential was measured on expanded terminal leaflets located close to the trunk, every 15 min and averaged to hourly values, with a psychrometer/hygrometer , model PSY-1 . The leaflet equipped with the psychrometer was fully covered with an insulation capsule limiting temperature fluctuations . As the monitored leaf did not transpire, the measurement was representative of stem rather than leaf water potential. An independent measurement of stem water potential was carried out weekly on fully expanded leaflets with a pressure chamber . Prior to this destructive measurement, leaflets were enclosed in foil-laminate bags for at least 10 min . Plant transpiration rate was quantified by automatic weighing of pots on a high precision weighing scale every ten minutes, averaged to hourly values. Draining water was collected daily in plastic reservoirs attached laterally to the bottom of the pots by flexible rubber tubing. Hence, the weight of leaching water did not affect the weighing scale reading until its collection. Both the added irrigation water and collected leachate were weighed and removed from the water balance in order to evaluate the weight loss due to TR . Bulk soil water potential at soil-root interface was monitored by one tensiometer per pot, placed at approximately the midpoint of the root system at 0.2 m depth, and recording data every ten minutes to generate average hourly values. Its porous ceramic cup was connected through a water-filled PVC tube and a smaller acrylic glass tube equipped with a pressure transducer. A rubber cap on top of the tensiometer ensured its air tightness.
All plant and soil measurements were continuously recorded with a data logger located inside the greenhouse. Hourly average air temperature and relative humidity were obtained in an automatic micrometeorological station placed inside the greenhouse. The reference evapotranspiration was obtained by use of an atmometer Model E , that gives one pulse at each 0.254 mm of evaporated water . Hourly vapor pressure deficit was estimated by the difference between saturated and actual vapor pressure. Saturated vapor pressure was calculated using air temperature based on the Tetens formula . Actual vapor pressure was obtained by saturated vapor pressure multiplied by fractional humidity. We used an empirical water stress indicator based on plant relative transpiration . For each plant, the potential daily transpiration was estimated as a product of the plant standard daily transpiration by the ratio of the actual daily transpiration to TD* of the unstressed plant . The water stress indicator was simply calculated as the ratio of TD to plant potential daily transpiration. An undisturbed leaf was harvested and water extracted using a custom-made cryogenic distillation system suitable for isotopic analysis, adapted from previous studies of this kind . Briefly, the leaves were transferred to individually cut 1.27 cm diameter pyrex tubes where the leaf material was held in place by stainless steel wool. After attachment to a vacuum manifold, leaves were frozen in liquid nitrogen and air evacuated to 100 mTorr. The tube was then flame sealed to preserve the vacuum, and subjected to gravity assisted cryogenic distillation, the top of the tube at 110° C, bottom at −20° C. After distillation, the tube was removed and ice water isolated by flame sealing the tube again to separate water and leaf material. Leaf material was separated and ground to a powder using liquid nitrogen in a mortar and pestle. 3 mg samples were submitted for δ13C determination at the UC Davis Stable Isotope Facility by continuous flow GC-IRMS on a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20- 20 isotope ratio mass spectrometer . The water samples were transferred to 2 mL vials and was analyzed for δD by equilibration with water vapor and added hydrogen gas, assisted by a platinum black powder catalyst. Next, CO2 was added to the system and equilibrated with water vapor for δ18O analysis.
Water analysis was performed at the University of Miami by using multi-flow system connected to an Isoprime mass spectrometer . To standardize isotopic data, values are reported in del notation with reference standards as in the equation below. The visible root length was monitored weekly over five weeks from the beginning of each 10-days period experiment by combining root mapping on the transparent walls of the pots and observation of inner root length with minirhizotrons , which provide a nondestructive method for repeated root observations . In addition, weekly root length observations started five weeks before each 10-days period experiment in order to follow the Rl pattern through time. Minirhizotrons consisted of transparent acrylic tubes with an inner diameter of 50 mm, and wall thickness of 3 mm. We used one tube per pot, installed at an angle of 45°, and sealed with silicon. Analyses of Rl were performed weekly with a BTC minirhizotron digital image capture system , located inside the minirhizotron tube. Each observation consisted of systematically taking pictures at one-centimeter intervals from the top to the bottom of the pot in three dimensions, totaling approximately 90 pictures per tube. The Rootfly software was used to analyze root length semi-automatically.Analysis of covariance of linear regressions between hydrogen and oxygen isotope ratios of leaf water showed significant differences in intercept between treatments , but no differences in slope . All experimental pots were covered to suppress soil evaporation, therefore, differences between treatment regression lines relative to the source water line are attributed to changes in leaf transpiration. Differences in intercept tracked expected declines in transpiration rates under drought stress and are consistent with changes in iWUE inferred from carbon isotope ratios . There was no difference between T100 and T75 with respect to iWUE or d-excess, indicating physiological acclimation and maintenance of a steady balance between photosynthesis and transpiration. However, iWUE and dexcess of T50 trees was significantly different from the others, indicating low stomatal conductance .Snapshots of root growth over time are shown in Fig. 6. In general, under well-watered conditions, new roots started to grow before the old roots died and were more frequently observed . Large variability was recorded for relative external and internal patterns of root growth at each sampling event. However, the cumulative total and living root growth detected by the minirhizotron showed significant changes with greater growth observed in the well-watered treatment . Crucially,maceta 25 litros root growth patterns were proportionally and positively related with d-excess . This indicates the existence of a fundamental trade off between root growth and iWUE , by which canopy transpiration and root development can be estimated based on changes in leaf stable isotope ratios. It is important to note, however, that differences between T100 and T75 with respect to either root growth or iWUE were not statistically significant. Therefore, acclimation is possible at that level and high physiological stress seems to be required to study costs and benefits of such a trade off with respect to changes in water supply.Our observations confirmed the decreasing TR as a response of midday depressions of leaf water potential , showing the minimum ψstem in T50 between −1.0 MPa and −2.0 MPa, which was strongly and positively correlated with ψsoil, explaining low TR under deficit irrigation . Indeed, stomata are expected to be completely closed in walnut trees when leaf water potential reaches −1.6 MPa and similar ψstem values and associated stomatal closure have been previously reported in stressed walnut trees , as transpiration rates decrease to prevent leaf dehydration under moderate to high Tair and VPD .
Otherwise, the strong and positive correlation between TR and evaporative demand was noticed for well-watered plants , as observed in previous studies , followed by strong and moderate water limitation . Multiple lines of isotopic evidence integrate the effect of physiological responses to treatments during the entire experiment and corroborate a significant decline in TR under deficit irrigation. Leaf water regressions show significant deviation from source water with reducing water loss by transpiration earlier under water stress has also been recognized in peanut and pearl millet . Here, our results showed an early and rapid decline in transpiration followed by stabilization of water loss in stressed trees, which is consistent with the fraction of “transpirable” soil water general mechanism of declining TR and with the classic descriptions of the plant water stress function . The nonlinear decrease of TR as a function of ψsoil and ψstem can be seen as a water conservative strategy to prevent water loss and leaf dehydration long before being limited by water supply from the soil-root system . Such a strategy lowers the risk of hydraulic failure and increases the iWUE. Considering that the major part of the walnut orchards are located in areas periodically affected by drought and due to its high water requirement over seasons, this observed trend and its further understanding has a key role in the identification and use of relevant physiological traits in plant breeding programs, allowing greater water-use efficiency under deficit conditions. The observed values of Kh fall in the typical range reported for young tree species and annual crops . Our results highlight the decrease of Kh under moderate and strong water limitation . Water deficit is one of the most important factors affecting Kh , and its decline in response to decreasing stem water potential under water deficit has been reported in walnut at ψstem approaching −1.8 MPa due to cavitation . However, we observed reduced Kh long before reaching such negative stem water potentials . As our Kh only includes hydraulic resistances between the stem and the soil-root interface, its reduction might have been fostered by a combination of poor soil-root contact under lower soil water content and altered root permeability that were described in other species.It turns out that in the T75 treatment, a reduction of stomatal opening due to water limitation occurred long before transpiration was limited by Qavail. Functionally, such stomatal regulation might play the role of extra security margin against hydraulic failure and translate into a so-called water saving behavior at longer term . The results also suggest that the supply-demand view in plant transpiration modeling is inappropriate for walnut, so that more complex models are needed . Despite the significant effect of water deficit on various plant properties, root growth responses over time did not correlate with any other recorded variable, and could did thus not explain changes of Kh. However, our observations suggest that healthy roots rapidly shifted to decaying roots with the continuity of water stress, which means a reduction of root activity and less capacity to take up water .