Predicting stomatal closure and turgor loss in woody plants using predawn and midday 1 water potential 2

39 Knowledge about physiological stress thresholds provides crucial information about plant 40 performance and survival under drought. In this study we report on the triphasic nature of the 41 relationship between plant water potential (  ) at predawn and midday and describe a method 42 that predicts  at stomatal closure and turgor loss exclusively from this water potential (WP) 43 curve. The method is based on a piecewise linear regression model that was developed to 44 predict the boundaries (termed  1 and  2 ) separating the three phases of the curve and 45 corresponding slope values. The method was tested for three economically important woody 46 species. For all species, midday  was much more negative than predawn  during phase I 47 (mild drought), reductions in midday  were minor while predawn  continued to decline during 48 phase II (moderate drought), and midday and predawn  reached similar values during phase- 49 III (severe drought). Corresponding measurement of leaf gas exchange indicated that boundary 50  1 between phase I and II coincided with  at stomatal closure. Data from pressure-volume 51 curves demonstrated that boundary  2 between phase II and III predicted  at leaf turgor loss. 52 The WP curve method described here is an advanced application of the Scholander-type 53 pressure chamber to categorize plant dehydration under drought into three distinct phases and 54 to predict  -thresholds of stomatal closure and turgor loss. 55


Introduction 204
Leaf gas exchange was measured for grapevine ( Figure 7). This showed that boundary  1 205 between phase-I and -II matched the -threshold at which leaf gas exchange was substantially 206 reduced ( Figure 7). Following an initial increase of g s ( Figure 7A) and A ( Figure 7B), that 207 reached a maximum at  pd of around -0.4 MPa, g s and A declined by 93% and 67%, 208 respectively, thereafter reaching values of close to zero at boundary  1 . 209 210 Almond 211 Similar to data collected for potted walnut (see Figure 2) and grapevine (see Figure 6) plants, 212 data collected for almond trees showed that the relationship of  pd and  md exhibited a triphasic 213 curve shape (Figure 8). The calculated boundary values using our statistical method were  1 = -214 1.37 MPa and  2 = -1.94 MPa (Figure 9, Table 1). See Supplemental Table S2 for statistical predominantly regulated by such a passive hydraulic signal. The authors came to this 281 conclusion because leaf ABA increased only after complete stomatal closure, and it was 282 hypothesized that this is of importance for long-term drought recovery to facilitate xylem 283 embolism repair by forcing stomatal closure when the soil is rehydrated. In summary, and in the 284 context of collecting WP curves for various woody plants, we speculate that if leaf ABA 285 accumulation is observed in phase-I, this points to a predominant ABA-mediated mechanism 286 driving stomatal closure, whereas if leaf ABA accumulation is observed in phase-II this points to 287 a predominant pressure-driven (passive) mechanism driving stomatal closure. 288 289 Atmospheric evaporative demand affects the rate of transpiration (see introduction; Klepper, 290 1968). Gollan et al. (1985) performed an extensive study on the relationship of vapor-pressure-291 deficit (VPD), determined on leaves with in-situ psychrometers and soil water content on leaf 292 gas exchange. The authors showed that the relationship of and leaf gas exchange is 293 dependent on VPD. During increasing drought, their data point to a sharp drop-off in g s at high 294 VPD (25 Pa kPa -1 ) whereas the reduction in g s was gradual until reaching a minimum (as 295 observed in this study) at low VPD (10 Pa kPa -1 ). Due to natural fluctuations in VPD and its 296 effect on transpiration, plant  can vary between days and over the course of a day for the 297 same soil water status. For this reason, plant  at midday (as measured on an equilibrated leaf 298 using a pressure chamber, i.e. ' stem ') is most meaningful as a water stress indicator over the 299 growing season when compared to baseline values under well-irrigated conditions (Turner NC, 300 1990;Shackel et al., 1997). Based on our findings, we hypothesize that effects of VPD on the 301 shape of the WP curve are most pronounced during phase-I prior to stomatal closure, and high 302 versus low VPD conditions result in either a steeper (higher transpiration) or shallower (lower 303 transpiration) slope  1 . On the other hand, and for a given plant species, we speculate that the 304 effect of VPD on boundary  1 is negligible if stomatal closure is predominantly driven by soil-to-305 root interactions (see previous paragraph). 306 307 Leaf turgor loss 308 Our data provide evidence that boundary  2 separating phases-II and -III of the WP curve 309 predicts  TLP . For walnut (variety 'Cisco'), average  TLP determined from pressure-volume (PV) 310 curves of -1.39 MPa (Supplemental Table S1) was only slightly more negative as compared to -1.33 MPa (Table 1). For almond, Torrecillas et al. (1996) determined a  TLP of around -2.2 MPa 314 (cultivar 'Garrigues') and -2.3 MPa (cultivar 'Ramilete') for well-watered trees as compared to 315 our  2 of -1.9 MPa (cultivar 'Nonpareil') (Table 1). Together, this suggests that  TLP can be 316 predicted from the triphasic WP curve using our PLR model for a variety of woody species. 317 318 For walnut trees subjected to the slow drydown, we were able to collect sufficient data points to 319 elucidate the pattern of drought-induced variations in leaf turgor (i.e. P =  -; Jones and 320 Turner , 1978). To calculate P, we assumed that  ≈  symplast and that the reflection coefficient for 321 solutes () was unity. If we consider the remaining apoplastic sap in the centrifuged leaf tissue 322 resulted in a dilution effect (Wardlaw, 2005), then our measured  underestimated true  symplast . 323 Hence, true P should be slightly lower than our calculated P, which would explain why our 324 calculated P was 0.3 MPa and not 0 MPa at  2 of -1.31 MPa (see Figure 4A). On the other 325 hand,  o at full hydration obtained from PV curves was 1.1 MPa (Supplemental Table S1) but 326 at predawn  of 0 MPa was around 1.7 MPa ( Figure 4A), which points to a possible 327 overestimation of  symplast . The indirect approach used here provides a relatively easy way to 328 obtain information on drought-induced P changes, but if experimentally feasible, P and  symplast 329 are best determined directly using a combination of cell pressure probing and picolitre 330 osmometry (Tomos and Leigh, 1999;Fricke and Peters, 2002;Knipfer et al., 2014). 331 332 Although controversial, it has been reported that negative turgor exists in plant cells according 333 to indirect measurements from PV curves (Tyree, 1976(Tyree, , 1979Rhizopoulou, 1997, Ding et al., 334 2014. Our P data obtained indirectly by  - also point to the existence of negative P which 335 occurred in phase-III following the turgor loss point. However, one factor that may explain the 336 measurement of negative turgor when determined indirectly is , which provides a measure of 337 solute permeability/leakage of the cell membrane (Staverman, 1951;Knipfer et al., 2014). 338 Drought stress results in increased solute leakage and modulation of the physical state of the 339 membrane (Blum and Ebercon, 1981;Premachandra and Shimada, 1987;Couchod et al., severe under increasing drought stress,  would become smaller and smaller and the term  343 goes towards zero, which in turn would result in a calculated P that may not reach negative 344 values. Comparing the slow versus fast drydown, our data suggest that P reaches values closer 345 to zero during the fast drydown at boundary  1 . This may be due to generally higher  values in versus weeks). However, the biophysical properties of the cell membrane under various levels 350 of drought stress and in response to the type of the drought experiment remain unknown, and 351 only direct P measurements using a cell pressure probe would allow us to resolve these open 352 questions (see previous paragraph). 353 354

Pressure-volume curve 355
The Scholander-type pressure chamber has been used successfully to generate pressure-356 volume (PV) curves for determination of tissue properties such as  TLP , bulk modulus of 357 elasticity, and  o at full hydration (Tyree andHammel, 1972, Ding et al. 2014). The advantage of 358 the PV curve is that  TLP can be determined from a single leaf measurement. The disadvantage 359 is that generating a PV curve can be time-consuming (>10h) depending on the speed of leaf 360 dehydration and requires accessibility to an analytical digital balance to determine relative water 361 content; one technical difficulty is finding the right time interval for progressive leaf dehydration 362 and data collection. 363 364

Water-potential curve 365
The WP curve method allows predicting  at stomatal closure and  TLP exclusively from 366 measurements of  at predawn and midday using a Scholander-type pressure chamber. This 367 can be especially useful under remote field conditions or during research operations with limited 368 access to a leaf gas exchange system, analytical balance, and laboratory space. Moreover, the 369 WP curve method will provide for a time-integrative measurement of at stomatal closure and 370  TLP when plant  at predawn and midday is collected on mature leaves over the growing 371 season. In this case, it is recommended that plant  is measured during the phenological 372 timeframe following leaf maturation and prior senescence to ensure that leaf cells, xylem and One alternative procedure to save time is to skip the equilibration step and immediately proceed 392 from step 1 to 3 ( Figure 10). However, excluding the equilibration step 2 provides a less 393 accurate estimate of P x since leaf internal water potential gradients between apoplast and 394 symplast are not minimized prior to leaf excision (Shackel et al., 1997). Another alternative 395 procedure is to seal the excised leaf in a second plastic bag (i.e. minimize leaf water loss as 396 much as possible) and store the sample at around 4 o C for up to 24 hours prior to measurement 397 of  ( Figure 10); the assumption is that leaf internal water potentials are maintained constant 398 during the storage period, but this should ideally be tested first on a subset of plants through 399 frequent  measurements during the storage period. 400

401
Our data show that, opposite to a fast drydown, a controlled and slow drydown (weeks) has the 402 advantage of collecting a higher number of data points during all three phases of the WP curve 403 because predawn  can be determined for a wide range of soil moisture contents. When 404 establishing the WP curve, predawn  can be interpreted as a measurement of soil  when (i.e. days versus weeks) can result in a more or less severe stress response for a given level of 424 soil moisture. A slow drydown (weeks) would provide time for stress adaptions linked to plant 425 anatomical changes, for example, root suberization to minimize water loss back to the soil and 426 reductions in vessel diameters to secure long-distance transport capacity (Barrios-Masias et al., 427 2015;Knipfer et al. 2015Knipfer et al. , 2020. Here, we conducted a slow and fast drydown experiment for 428 walnut trees to test if the WP curve (i) remains triphasic and (ii) predicts stomatal closure in both 429 types of experiments. Our data show that the WP curve was triphasic in both drydown 430 experiments but the character of the WP curve differed between the slow and fast drydown and 431 boundary values  1 and  2 were specific to the drydown experiment (see Table 1). Boundary  1 432 predicted the shift in -threshold of stomatal closure in both types of drydown experiments. We 433 speculate that boundary 1 was less negative during the fast drydown, i.e. earlier stomatal 434 closure, because the time period for anatomical adaptations -which would aid to maintain plant 435 performance for the imposed level of stress -was too short. Data by Knipfer et al. (2020) show 436 that walnut fine roots develop a multi-seriate endodermis in response to a slow drydown, and stomatal closure, ii) root hydraulic conductivity declines progressively during phase-I and -II and 453 reaches a minimum at boundary  2 at a -threshold that corresponds to leaf turgor loss, iii) 454 vessel cavitation is initiated at boundary  2 and cavitated vessels accumulate during phase-III. 455 This example demonstrates that our WP method can be used to categorize the sequence of 456 physiological and anatomical events that occur under progressive drought stress into three 457 distinct phases. Moreover, we propose that the WP curve method can assist in the 458 determination of -thresholds that mark the breakdown of the soil-to-root hydraulic continuum 459 Initial data inspection showed that a simple linear model did not appropriately describe the 466 relationship of plant water potential () at predawn (subscript 'pd') and midday (subscript 'md') 467 The PLR model was used to calculate the boundaries between linear phases ( 1 = phase-I to -471 II,  2 = phase-II to -III) and corresponding slope values ( 1 = phase-I,  2 = phase-II,  3 = phase-472 III;  =intercept) (Eq. 1). 473

474
To parameterize the PLR model, an estimate of the transition points between phase-I and -II ( 1 ) 475 and phase-II and -III ( 2 ) was obtained mathematically as follows. First, a smoothed line was 476 fitted to the relationship of  pd and  md that best described the data pattern. Subsequently This allowed for fitting a continuous piecewise linear function to corresponding data of  pd and 484  md for a specified number of three line segments. Our method used a limited memory 485 Broyden-Fletcher-Goldfarb-Shanno (LM-BFGS) algorithm for bound constrained optimization to 486 obtain a statistical solution of boundary values  1 and  2 from the initial estimates of  1 and  2 . 487 Standard errors and p-values corresponding to output parameters of the PLR model were the 488 result of using this optimization procedure to find boundaries  1 and  2 that best satisfied the 489 specified number of linear segments; standard errors were obtained following the derivation of Old Davis Road). Trees were irrigated by supplying water to the top of the soil every two days 512 and maintained well-watered for 3 months after transplanting to ensure sufficient time for tree 513 establishment. All physiological measurements were performed on mature leaves of current-514 year shoots. A temperature and relative humidity sensor (HMP50; Vaisala, Woburn, MA) was 515 installed at the plot site to monitor vapour-pressure deficit (VPD, see Supplemental Figure S5). 516 To investigate possible effects of type of drought experiment, trees were subjected to a slow 517 drydown (i.e. weeks, adjustment of irrigation) or fast drydown (i.e. days, no supplemental 518 irrigation): The slow drydown experiment was performed on n=65 trees. Irrigation was adjusted 519 based on estimates of bulk soil moisture (SM, = weight H2O / weight H2O-pot-capacity x 100%) as 520 calculated from pot weights (for details see Knipfer et al., 2020). A subset of n=27 plants was 521 located on mini-weighing lysimeters to continuously monitor pot weight and SM (see 522 Supplemental Figure S6 for representative data of two individuals); to account for temporal 523 effects, trees were either maintained well-watered or subjected to a drydown. At 103 days after 524 transplanting, measurements were performed on n=27 trees with SM of individuals ranging from 525 transplanting, measurements were carried out on n=36 trees with SM ranging from 58 to 87% performed on n=30 trees with SM ranging from 44 to 75% w/w (VPD ranged from 1.7 to 2.2 529 kPa). The fast drydown experiment was performed on n=6 trees that were maintained well-530 watered until the start of the drydown. Supplemental irrigation was stopped at 210 days after 531 transplanting. Trees were analyzed at 210 (SM ranging from 73 to 100% w/w), 213 (69% to 96% 532 w/w), 216 (54% to 86% w/w), 218 (49% to 77% w/w) and 220 (47% to 72% w/w) days after 533 transplanting. 534 535 Water potential -A pressure chamber (PMS Instrument Company, Model 1505D, Albany, 536 Oregon, USA) was used to measure plant  following leaf covering and equilibration (i.e. 537 ' stem '). Measurements were performed on a leaflet of a mature leaf that was covered with 538 aluminum foil and equilibrated for >1h using a sealed plastic bag. Following excision of the 539 leaflet at the petiolule, the plastic bag was removed, and leaflets still covered with foil (i.e. to 540 exclude effects of transpiration; Turner and Jones, 1980) were inserted into the pressure 541 chamber. The pressure in the chamber was raised slowly at a constant rate (about 0.01 MPa 542 per seconds) and pressure was recorded when a water meniscus started to form on the cut 543 petiolule (midvein of leaflet) surface. For the same plant,  at predawn was measured prior to 544 sunrise between 4 AM to 6 AM prior to sunrise and  at midday was measured between 11 AM 545 to 1 PM pacific daylight time. Watering was always completed the day before measurement of 546 to allow for soil water distribution. 547 548 Leaf gas exchange -Stomatal conductance (g s ) and CO 2 assimilation rate (A) were measured 549 between 11 AM and 1 PM using a LICOR-6800 gas exchange system (fan speed at 10.000 550 rpm, leaf temperature at 24.5 º C, CO 2 sample at 400 ppm, and 1500 mol m -2 s -1 light intensity). 551 One non-shaded leaflet of a mature leaf was measured on each sapling that was in proximity to 552 the leaflet used for measurements of  (as described above). Since the leaf area inserted in the 553 cuvette of gas exchange system during measurement occupies many stomata and the response 554 of individual stomates within this area can be more or less coordinated and homogenous, the 555 point of stomatal closure was defined as the point where g s and A reach a minimum. 556 Leaflet fresh weight was measured with a digital balance before and after each water potential 577 step measurement (PMS Instrument Company, Model 1505D; Albany, Oregon, USA). 578 Measurements were repeated until five measurements after the turgor loss point ( TLP ). 579

Grapevine experiments 581
Experiments were performed in 2019 on 48 grafted vines (variety Chardonnay on rootstock 5C, 582 420A, Riparia Gloire, 101-14, Ramsey, 140Ru, 1103P, and 110R). Vines were planted in 6-liter 583 pots. Pots were filled with similar amounts of soil mix (approximately 75% coconut coir and 25% 584 perlite) by leaving a gap of around 2 cm to the upper edge of the pot. Growth was maintained in 585 a greenhouse on the UC Davis campus. Vines were allowed to establish for three months prior 586 to data collection. During the establishment period, the vines were pruned to a single shoot, 587 which was staked and tied after reaching approximately 0.5m in length. The pots were weighed 588 and irrigated by supplying water to the top of the soil three times per week to a target weight 'PROC TRANSREG' (sm = smoothing parameter) procedure in SAS (version 9.2, SAS Institute, 637 Cary, NC, USA). A statistical comparison of model output parameters was performed using a Z-638 test (see Supplemental Table S2).  Table S1: Turgor loss point determined from pressure-volume curves for walnut trees. 648 Table S2: Z-test used for statistical comparison of output parameters shown in Table 1. n/a n/a 24 n/a n/a 80 n/a n/a 336 n/a n/a R 2  0.91 n/a n/a 0.90 n/a n/a 0.62 n/a n/a 0.95 n/a n/a  Roman numerals I to III designate the three phases of the water potential curve. Vertically solid 701 lines are the boundaries between phase-I and -II ( 1 ) and phase-II and -III ( 2 ) and 702 corresponding standard errors are indicated in gray color. Model output parameters are 703 summarized in Table 1. 704 Figure 3: Relationship between predawn  and leaf gas exchange (stomatal conductance or 705 CO 2 assimilation rate) for walnut trees. Trees (variety 'Cisco' grafted on rootstock RX1, VX211 706

Figure legends
and Vlach) were subjected to a slow drydown (in A and B) or fast drydown (in C and D). (C, D) 707 During the fast drydown, same symbols indicate data collected for the same tree. Roman 708 numerals I to III designate the three phases of the water potential curve (corresponding to 709  During the fast drydown, same symbols indicate data collected for the same tree. Roman 718 numerals I to III designate the three phases of the water potential curve (corresponding to 719 ± 0.14 MPa as measured from pressure-volume curves (see Supplemental Table S1). (C, D) 724 Due to the limited amount of data points collected during the fast drydown for phase-II and -III a 725 fitted line is not included. 726 Vertically solid lines are the boundaries between phase-I and -II ( 1 ) and phase-II and -III ( 2 ) 738 and corresponding standard errors are indicated in gray color. Model output parameters are 739 summarized in Table 1. 740 Figure 7: Relationship between predawn  and leaf gas exchange (stomatal conductance or 741 CO 2 assimilation rate) for grapevine (variety 'Chardonnay' grafted on rootstock 110R, 1103, 742 140RU, 5C, RG, Ramsey, 101-14 and 420) corresponding to Figures 5 and 6. The dashed line 743 is a smoothed line (smoothing factor of sm=60) that best described the pattern of data points.