Xylem hydraulic safety margins inwoody plants

Functional Ecology 2009, 23,922-930 doi: 10.1111/j.1365-2435.2009.01577.x

Xylem hydraulic safety margins in woody plants:coordination of stomatal control of xylem tensionwith hydraulic capacitanceFrederick C. Meinzer*,1, Daniel M. Johnson1, Barbara Lachenbruch2, Katherine A. McCulloh2

and David R. Woodruff1

1USDA Forest Service, Pacific Northwest Research Station, 3200 SW Jefferson Way, Corvallis, Oregon 97331, USA;and 2Department of Wood Science and Engineering, Oregon State University, Corvallis, Oregon 97331, USA

Summary

1. The xylem pressure inducing 50% loss of hydraulic conductivity due to embolism (P50) iswidely used for comparisons of xylem vulnerability among species and across aridity gradients.However, despite its utility as an index of resistance to catastrophic xylem failure under extremedrought, P5o may have no special physiological relevance in the context of stomatal regulationof daily minimum xylem pressure and avoidance of hydraulic failure under non-extreme condi-tions. Moreover, few studies of hydraulic architecture have accounted for the buffering influenceof tissue hydraulic capacitance on daily fluctuations in xylem pressure in intact plants.2. We used data from 104 coniferous and angiosperm species representing a range of woodygrowth forms and habitat types to evaluate trends in three alternative xylem hydraulic safetymargins based on features of their stem xylem vulnerability curves and regulation of daily mini-mum stem water potential (Ystem min) under non-extreme conditions: (i)Ystem min - P50, (ii)Y stem min - Pe, the difference between Ystem min and the threshold xylem pressure at which lossof conductivity begins to increase rapidly (Pe) and (iii) Pe - P50 an estimate of the steepness ofthe vulnerability curve between Pe and P50. Additionally, we assessed relationships betweenxylem capacitance, species-specific set-points for daily minimum stem water potential andhydraulic safety margins in a subset of species for which relevant data were available.3. The three types of hydraulic safety margin defined increased with decreasing species-specificset-points for Y stem min, suggesting a diminishing role of stem capacitance in slowing fluctuationsin xylem pressure as Ystem min became more negative. The trends in hydraulic safety were similaramong coniferous and angiosperm species native to diverse habitat types.4. Our results suggest that here is a continuum of relative reliance on different mechanisms thatconfer hydraulic safety under dynamic conditions. Species with low capacitance and denserwood experience greater daily maximum xylem tension and appear to rely primarily on xylemstructural features to avoid embolism, whereas in species with high capacitance and low wooddensity avoidance of embolism appears to be achieved primarily via reliance on transient releaseof stored water to constrain transpiration-induced fluctuations in xylem tension.

Key-words: hydraulic architecture, xylem embolism, plant-water relations, transpiration

*Correspondence author. E-mail: [email protected]

pathway upstream (Meinzer & Grantz 1990; Cochard et al.2000; Hubbard et al. 2001; Mencuccini 2003). Thus, undernon-extreme conditions stomatal regulation of leaf waterpotential (YL) presumably constrains xylem pressure instems to a range that does not result in excessive loss of stemconductivity from embolism. However, the hydraulic resis-tance of leaves is substantial (Sack & Holbrook 2006) anddynamic over a daily time scale (Bucci et al. 2003; Brodribb& Holbrook 2004a; Woodruff et al. 2007), resulting in asubstantial and variable transpiration-induced disequilib-

Introduction

In a broad sense, stomata regulate transpiration to avoiddamaging levels of leaf dehydration and to avoid hydraulicfailure in the xylem of the stems to which the leaves areattached. As a consequence, stomatal control of vapourphase conductance is closely coordinated with dynamicchanges in the hydraulic properties of the water transport

© 2009 The Authors. Journal compilation © 2009 British Ecological Society

mailto:[email protected]

esipp

Text Box

This file was created by scanning the printed publication. Text errors identified by the software have been corrected: however some errors may remain.

Xylem hydraulic safety margins 923

rium between YL and stem xylem pressure (Begg & Turner1970; Ritchie & Hinckley 1971). Therefore, the bulk Y oftranspiring leaves is often difficult to relate to the magni-tude of xylem pressure in the adjacent stem and its impacton hydraulic function. In order to understand the potentialimpact of stomatal behaviour on stem hydraulics of field-grown plants, stem water potential (Ystem) must beestimated using techniques such as measurement of Y ofnon-transpiring leaves or shoot tips (Melcher et al. 2001;Bucci et al. 2004a) or stem psychrometry ( Dixon & Tyree1984; Scholz et al. 2007). Although published measure-ments of YL are abundant, estimates of Y stem in transpiringplants are scarce, making it difficult to draw general conclu-sions about relationships between stomatal behaviour andstem hydraulic function.

The resistance of xylem to loss of function from cavitationand embolism is often assessed by generating hydraulic vul-nerability curves describing the relationship between xylempressure and loss of conductivity (e.g, Fig. 1). Vulnerabilitycurves typically have a sigmoid shape with loss of conductiv-ity initially increasing gradually as xylem pressure decreasesfollowed by an abrupt transition to a much steeper, nearlylinear phase, ending with a more gradual phase as loss of con-ductivity approaches 100%. The xylem pressure correspond-ing to 50% loss of conductivity (P50) is widely used forcomparisons ofxylem vulnerability across species, within dif-ferent parts of the same individual and within species acrossenvironmental gradients (e.g. Melcher et al. 2001; Maheraliet al. 2004, 2006). Another important, but less widely usedpoint on the vulnerability curve is the threshold xylem pres-sure at which loss of conductivity begins to increase rapidly,often referred to as the air entry pressure (Pe). The air entrypressure can be estimated from the x-intercept of a tangent(Fig. 1, dotted line) drawn through the midpoint ofa sigmoidfunction fitted to the vulnerability curve data (Domec &Gartner 2001). Although specific values of P50 are widely

reported, values of Pe are much less frequently given aroften have to be estimated from published graphs of vulnerbility curves.

Despite the utility of P50 as an index of resistance to catstrophic xylem failure under extreme drought, it may have nospecial physiological relevance in the context of stomatal reg-ulation of xylem tension under most conditions. There isincreasing evidence that under non-extreme conditions sto-matal control of transpiration limits maximum xylem tensionto a value close to Pe for stem tissue to which the leaves areattached (Sparks & Black 1999; Brodribb et al, 2003; Domecet al. 2008). Runaway embolism should become increasinglikely once xylem pressure falls below Pe, reaching the stetportion of the vulnerability curve because tissue hydraulcapacitance (C) is largely depleted (Meinzer et al. 2008a) arno longer sufficient to adequately buffer fluctuations in tesion induced by rapid changes in transpiration (see belowThus it may be unlikely that stomatal regulation allows xylepressure to fall substantially below species-specific valuesPe, except during periods of extreme drought that ultimatelead to branch dieback and shedding (Sperry et al. 1993Sparks & Black 1999; Nardini & Salleo 2000). Both isohydrand anisohydric species would face a similar risk of runawayembolism as xylem pressure falls below Pe unless the slopes ofthe steeper portion of vulnerability curves systematically dif-fer among species with these two modes of regulation of Y.

Although considerable attention has been paid to xylesafety and efficiency as components of hydraulic architecturC is an additional and somewhat neglected component ofhydraulic architecture that merits further study. Capacitanceof plant tissues can play an important role in slowing changein xylem pressure following transpiration-induced changesxylem water flux (Cowan 1972; Phillips et al. 1997, 2004;Goldstein et al. 1998; Meinzer et al. 2004, 2008a; Scholz et al.

2007; Holtta et al. 2009). Using an Ohm's law analogy, thebuffering effect of C on fluctuations in xylem pressure can bequantified in terms of the time constant (t), or time requiredfor xylem pressure to undergo 63% of its total change follow-ing a change in water flux. In practice, t can be calculated,the product of hydraulic resistance and capacitance (R x C)between a reference point within the soil and a reference pointwithin the plant. The time constant defined in this manner isthus a dynamic rather than static property of the hydraulicpathway and is expected to fluctuate over the course of a dayas changes in tissue hydration cause C and possibly R to vary(Meinzer et al. 2003, 2008a; Scholz et al. 2007). Given thattime constants for stomatal responses to changing environ-mental variables range from a few minutes to 10 min or more(Ceulemans et al. 1989; Jarvis et al. 1999; Allen & Pearcy2000; Powles et a1. 2006), C through its impact on 't for train-spiration-induced fluctuations in xylem pressure may be criti-cal for avoiding sharp decreases in xylem pressure andrunaway embolism following transient increases in transpira-tion. Of course, sharp increases in R would increase 1:, but atthe cost of catastrophic xylem failure and ensuing lethal levelsof dehydration. Estimates of t for whole plants and plantparts are scarce, but available data for branches and whole

© 2009 The Authors. Journal compilation © 2009 British Ecological Society, Functional Ecology, 23, 922-930

924 F. C. Meinzer et al.

trees span a range from about 5 min to > 2 h (Phillips et al.1997,2004; Meinzer et al. 2004).

Our principal objectives in this study were to evaluate vari-ous measures of xylem hydraulic safety margins across abroad range of woody plant species and to further assessinteractions between woody tissue C and hydraulic safety in asubset of nine tropical tree species for which relevant datawere available (Meinzer et al. 2008a). Three alternativehydraulic safety margins were defined: the difference betweenminimum stem water potential (Y stem min) and P50, the differ-ence between Y stem min and Pe and the difference between Peand P50, an estimate of the steepness of the vulnerability curvebetween Pe and P50 (Fig. 1). In order to identify key pointsfor stomatal limitation of minimum xylem pressure when sto-matal control was still fully effective, data collected onlyunder non-extreme conditions for a given species and habitattype were included. Based on a previous study of tropicaltrees (Meinzer et al. 2008a), we hypothesized that safety mar-gins would increase with decreasing Ystem min and that thispattern would be associated with a declining contribution ofC to buffering of transpiration-induced fluctuations in xylempressure.

Materials and methods

Data on xylem vulnerability to embolism, stem capacitance and

Ystem min of woody species were obtained from the literature. Anextensive, but not exhaustive, search of the literature yielded partialor complete data sets for 104woody species representing a wide rangeofxylcm anatomy and phylogeny (see Table S1 in Supporting Infor-mation). Only data collected from field-grown plants were consid-ered. Data obtained from experiments involving imposition of severedrought were not included. The following additional criteria forincorporation of data were applied: Three measures of xylem vulnera-bility to embolism were estimated from standard hydraulic vulnera-bility curves consisting of plots of percent loss of hydraulicconductivity (PLC) vs. stem xylem pressure (Fig. 1), or from tables ofdata listing key points along vulnerability curves. The three measuresof vulnerability were (i) the xylem pressure corresponding to 50% loss

of hydraulic conductivity (P50), (ii) the air entry threshold (Pe) corre-sponding to about 12% loss of conductivity wherein species with pro-nounced sigmoid vulnerability curves PLC begins to rise sharply withdeclining xylem pressure (Domec & Gartner 2001) and (iii) Pe -P50

a measure of how steeply PLC rises once Pe has been reached. Largervalues of Pe - P50 imply a more gradual rise of PLC once xylem pres-sure has fallen below P e Stem sapwood capacitance expressed perunit sapwood volume (kg m-3 MPa-1) for nine co-occurring tropicaltree species was obtained from Meinzer et al. (2008a).

Daily Ystem min was obtained from studies in which Y was deter-mined with a pressure chamber on covered, non-transpiring leaves(Begg & Turner 1970; Bucci et al. 2004a) and from studies in which Ywas measured on excised previously transpiring leafy twigs(Table S1). The covered leaf method is preferable for estimating Ystembecause the hydraulic resistance of leaves often results in substantialtranspiration-induced disequilibrium between stem and leaf Y.However, the occurrence of daily minimum plant Y is typically asso- of transpiration, which wouldconstrain the degree of disequilibrium between leaf and stem Y.

In three co-occurring tropical tree species for which published datawere available (Meinzer et al. 2008a), relative values of stem C and

hydraulic resistance (R) in relation to Ystem were used to estimate rela- .tive time constants (t = R x C) for changes in xylem pressure follow-ing changes in xylem water flux. Relative rather than absolute valuesof t were used because the data necessary to express both C and R inunits that would yield t in time units only were not available. The pur-pose of this exercise was to determine whether P 50 and Ystem min of dif-ferent species occurred at similar points along the trajectory of t asYstem declined.

The significance of relationships between Ystem and Pe , P50 andPe - P50 and between P50 and Pe was evaluated using one-way ANO·

VA. Differences in the dependence of P50 on Pe among three groups ofspecies provisionally identified post hoc were assessed by pair-wiseANCOVA after it was determined that the regressions were heteroge-neous (F = 12·5, P < 0·0001).

Results

Observations compiled for rune coniferous and 28 angio-sperm species showed that daily Ystem minmeasured when soilwater deficits were non-extreme was tightly coordinated withboth Pe and P50 a similar manner across species and acrossboth conifers and angiosperms (P < 0·0001, Fig. 2). Therelationship between Pe and Ystem min was close to the 1:1 lineover a relatively wide range of species values of' Y stem min. Forspecies having Y stemmin more negative than about -3 MPa,Pe began to decline noticeably more quickly than Y stem min. Inthe species with the least negative values of' Y stem min, Pe wasless negative than Y stem min- In contrast, P50 was always morenegative than Y stem min, and decreased more than 2·5 timesfaster than Y stemmin with declining species-specific values of

Y stem min-The hydraulic safety margin expressed as Ystem min - P50

was always positive and increased about 10-fold over therange of Y stem min observed (Fig. 3a). The safety marginexpressed as Y stem min - Pe was negative or near zero for spe-cies with Y stem min less negative than about -2·5 MPa andincreased to about 4 MPa in species with the lowest values ofY stem min observed (Fig. 3b). A third safety margin, Pe - P50based on species-specific xylem vulnerability curve character-istics alone, increased from about 1 to 4 MPa as Y stem mindeclined from about -0·5 to -4·5 MPa (P < 0·0001, Fig. 3c).Greater values of Pe - P50 indicate a more gradual loss ofhydraulic conductivity as xylem pressure falls below Pe.

Xylem vulnerability curve data were found for more speciesthan were data for '¥stem. Variation in P50 was linearly relatedto that of Pe, but three distinct groups of species appeared toemerge based on trajectories of regressions fitted to relation-ships between Pe and PSG (P < 0·0001 for all regressions,Fig. 4). Among 16 species of conifers and 47 species of angio-sperms native to wet, mesic and seasonally dry habitats, theslope of the relationship between P50 and Pe was 1·42, indicat-ing that the slope of the steeper portion of the vulnerabilitycurve became more gradual as Pe declined (Fig. 4a). In twoother groups of species, the slopes of the relationshipsbetween P50 and Pe were not significantly different from 1·0,indicating that the difference between Pe and PSG was con-stant over the entire range of Pe observed (Fig. 4b,c). In 12members of Cupressaceae (e.g. Juniperus spp.) native to



semi-arid and arid zones and 14 angiosperm species native toMediterranean climate zones, P50 was about 4·5 MPa morenegative than Pe over a broad range of Pe (Fig. 4b). Among15 vesse1ess angiosperm species, P50 was only about 1 MPamore negative than Pe (Fig. 4c). The slopes of the regressionsin Fig. 4b,c were both significantly different from that inFig. 4a (P< 0-01) and they differed significantly in theiroffset from the 1:1line (P < 0.0001).

Minimum Ystem showed a strong positive correlation(P = 0.0002) with sapwood capacitance among nine tropicaltree species (Fig. Sa). Minimum daily Y stem was about 1 MPaless negative in the species with the highest C(410 kg m-3 MPa-l) than in the species with the lowest C(83 kg m-3 MPa-I), consistent with the buffering effect onxylem pressure of transient withdrawal of water from internalstorage. Among the same nine species, the difference betweenPe and P50 decreased in a linear fashion (P < 0.0001) withincreasing sapwood C (Fig. 5b), indicating that the slope ofthe steeper portion of the xylem vulnerability curve was moreabrupt in species with higher C.

The dependence of relative C and relative hydraulic resis-tance on Ystem was characterized for three tropical treespecies with high, intermediate and low values of sapwood Cand daily Ystem min (Fig. 6). As expected, relative C initiallydeclined sharply as Ystem decreased followed by an asymp-totic approach to a minimum value of C, whereas the initialincrease in relative R associated with embolism was gradual



followed by a steep rise. Regardless of species-specific valuesof Ystem min and C, Ystem min occurred near or just after thetransition from a steep to gradual decline in C as Ystemdeclined, and was associated with a relatively narrow range ofincrease in relative resistance (30-39%). Overall, the patternsin Fig. 6 implied that the relative time constant (R x C) for aresponse in xylem tension following a change in xylem water

Discussion

Even under non-extreme conditions of adequate soil moistureand moderate vapour pressure deficit, the potential risk ofcatastrophic xylem failure is high, unless stomatal behaviouris tightly coordinated and integrated with the properties ofthe hydraulic pathway upstream from the leaves as theychange over the course of the day. We found a notable degreeof convergence in trends of xylem hydraulic safety over abroad range of Ystem min among 37 coniferous and angio-sperm species native to diverse habitat types. The three typesof hydraulic safety margin defined increased with decreasingspecies-specific values of Ystem min, This pattern probablyreflects a diminishing role of C in slowing fluctuations inYstem as Ystembecomes more negative. Nevertheless, stomatalcontrol interacts with C to determine species-specific set-



points for Ystemmin, and therefore hydraulic safety margins.Overall, the patterns observed were consistent with earliersuggestions that although coordination between stomatalbehaviour and stem hydraulic capacity generally avoids cata-strophic hydraulic failure, tolerance of a certain degree ofembolism may enhance productivity under some conditions,especially if loss of stem hydraulic function is reversible over arelatively short time scale (Jones & Sutherland 1991; Vogt2001; Brodribb & Holbrook 2004b; Sperry 2004).

Taken together, our results suggest that here is a contin-uum of relative reliance on different mechanisms that conferhydraulic safety under dynamic conditions. Species with lowcapacitance and denser wood experience greater daily maxi-mum xylem tension and appear to rely primarily on xylemstructural features to avoid embolism, whereas in species withhigh capacitance and low wood density, avoidance of embo-lism appears to be achieved primarily via reliance on tran-sient release of stored water to constrain fluctuations in xylemtension.

Fig. 7. Calculated relative time constants for changes in stem waterpotential in response to changes in xylem water flux for three co-occurring tropical tree species. Time constants were estimated as theproduct of resislance and capacitance (R x C) from the data shown inFig. 6. Vertical lines represent daily minimum stem water potential(Ystem min) and the stem xylem pressure corresponding to 50% loss ofhydraulic conductivity (Pso). Horizontal dashed lines represent tan-gents drawn through the nearly linear portions of the curves at lowvalues of stem water potential.

STOMATAL CONTROL OF STEM XYLEM PRESSURE

Minimum Ystemwas close to Pe over a wide range of' Ystem minfrom about -0,5 to -3 MPa (Fig. 2). This finding is consis-tent with earlier reports that when soil water availability isadequate, stomata constrain 'l'stem to minimum values at orless negative than Pe (Sparks & Black 1999; Brodribb et al.2003; Domec et al. 2008). In species with Pe below about-4 MPa, stomatal regulation of' Ystem was more conservativeresulting in a safety margin (Ystem min - Pe)of nearly 4 MPain the species with the lowest values of Pe, (cf. Figs 2a and 3b).Stomatal regulation of stem xylem pressure at values substan-tially more negative than Pe implies an increased risk of run-away embolism as xylem pressure begins to traverse the steep



portion of the sigmoid hydraulic vulnerability curve wheretissue C has effectively been exhausted, causing a sharp reduc-tion in the time constant for changes in xylem pressure inresponse to changes in transpiration. The effectiveness of sto-mata in constraining stem xylem pressure diminishes as thetime constant for flow-induced changes in xylem pressuredecreases relative to that for reductions in stomatal conduc-tance. The trend towards higher values of Ystem min - Pe inspecies with lower Ystem min is consistent with this interpreta-tion because the relative role of C in buffering changes inYstem diminishes as Ystem declines (Figs 5 and 7). The specificmechanisms by which stomata respond to and control YL toresult in species-specific values of Ystem minare not known.Regardless of the mechanisms, the coordination of leaf andstem Y set-points optimizes reliance on hydraulic capacitanceof stem tissue (Meinzer et al. 2008a).

The data presented for daily Ystem min represent pressurechamber measurements on non-transpiring leaves attached totranspiring shoots as well as conventional measurements ontranspiring leaves or leafy shoots. However, the hydraulicresistance of leaves is substantial, leading to pronounceduncoupling or disequilibrium between leaf and stem Y intranspiring shoots (Begg & Turner 1970; Meinzer et al. 2001;Melcher et al. 2001; Bucci et al. 2004a). Thus, a violation ofour assumption that YL is approximately equal to Ystem whenYL is at its daily minimum value and stomata have partlyclosed (see Materials and methods) would yield estimates ofYstemmin that are too negative. If actual species values ofYstem min were less negative than those shown in some cases,the overall trend of increasing hydraulic safety margins withdeclining Ystem min would remain the same, but stomatal regu-lation of Ystem min relation to Pe and other key points for stemhydraulic function would appear to be even more conserva-tive. Nevertheless, most of the data for Ystem min> -2 MPawere obtained from non-transpiring leaves (Table S1) rein-forcing the inference that the safety margin defined as Ystem

min - Pe was actually negative for values of Ystem min> -2 MPa: The impact of stomatal behaviour on the hydraulicfunction of stems to which leaves are attached cannot beassessed fully without knowledge of Ystem, but explicitattempts to estimate Ystem using techniques such as pressurechamber measurements on non-transpiring leaves or stempsychrometry are rather scarce in the literature. Studies aimedat describing the integration of stomatal and stem hydraulicfunction in intact plants should incorporate methodsdesigned to estimate the Y of the stems supporting the leaves.

HYDRAULIC SAFETY MARGINS

The air-seeding pressure, P50 is widely used as a comparativeindex of xylem hydraulic safety. Physiologically, however,attainment of P50 indicates that nearly catastrophic hydraulicfailure has already occurred and that the risk of further run-away embolism is acute because xylem pressure is operatingalong the steepest portion of the vulnerability curve (Tyree &Sperry 1988). Thus, the relevance of P50 as an indicator ofresistance to hydraulic failure may be restricted to episodic

extreme drought conditions under which it becomes physi-cally impossible for stomata to constrain xylem pressureabove or around Pe. These conditions are expected to lead tobranch dieback and shedding, and in some cases death of theentire plant, which would ultimately limit species distribu-tions along aridity gradients. In the present study we haveidentified three alternative hydraulic safety margins refer-enced to species-specific values of daily Ystem min set by stoma-tal control of transpiration (Fig. 3). All three types of safetymargins increased with declining species-specific values ofYstemmin, Interestingly, the safety margin defined as Ystem

min- Pe was negative in species having values of Ystem mindown to about -2 MPa, implying that branches of thesespecies may regularly experience a substantial loss of theirconductivity even when soil water availability is not severelyrestricted. Consistent with this, native loss of conductivitywas about 36% among 20 co-occurring tropical tree specieshaving mean daily values of Ystem min between -0,7 and-1·5 MPa (Santiago et al. 2004; Meinzer et al. 2008a). Evenhigher levels of native embolism were found in two co-occur-ring Mediterranean Quercus species (Tognetti et al. 1998).The extent to which daily losses of conductivity may bereversed overnight is unclear because branch samples in thestudies cited were collected after the onset of transpiration inthe morning. However, results of other studies suggest that acertain fraction of stem embolism may be readily reversibleover diel and seasonal cycles (e.g. Zwieniecki & Holbrook1998; Melcher et al. 2001; Vogt 2001) and that in leavesembolism reversal may be even more rapid with recovery ofhydraulic conductance occurring during the afternoon whilexylem pressure is still well below zero (Bucci et al. 2003; Bro-dribb & Holbrook 2004a; Woodruff et al. 2007).

In contrast to Ystem min - Pe, Ystemmin- P50 was alwayspositive and increased by a factor of about 10 over theobserved range of' Ystem min. The absence of negative values ofYstem-min - P50 reinforces the notion that stomatal regulationnormally prevents stem xylem pressure from traversing thesteep portion of the vulnerability curve. Increasing values ofthe safety margin defined as Pe - P50 indicate a progressivelymore gradual slope of the vulnerability curve once the airentry threshold has been crossed, which can be regarded as asecond line of defence against catastrophic xylem failure oncestomatal regulation can no longer prevent xylem pressurefrom entering the steeper portion of the vulnerability curve..The specific xylem structural features responsible for theobserved behaviour of Pe - P50 are not known, but could cer-tainly be examined. Most likely they relate to the incidence ofsafety features in the population of cells contributing toembolism avoidance, which could include not only tracheidsand vessels, but also 'parenchyma and fibre cells.

It is interesting that when all vulnerability curve data werepooled, two types of relationships between Pe, and P50emerged, one in which there was an increasing departure fromthe 1:1 relationship with declining Pe (Fig. 4a), and one inwhich there was a constant offset from the 1:1 line (Fig. 4b,c).The patterns observed suggest that phylogeny and evolution-ary adaptations to certain habitat types were important deter-

© 2009 The Authors. Journal compilation © 2009 British Ecological Society, Functional Ecology, 23, 922930

minants of relationships between Pe and P50. Among 15vesseless angiosperms, Pe - P50 remained nearly constant atabout 1 MPa over a 3·5 MPa range of Pe, whereas Pe - P50

was about 4·5 MPa over an 8 MPa range of Pe in 28 speciesof Cupressaceae and angiosperms native largely to arid andMediterranean climates zones. In other conifers and angio-sperms native to more mesic climate zones Pe - P50 increasedwith declining Pe. The species represented in Fig. 4 do notentirely overlap with those in Fig. 3 because published dataon Ystem were not available for many of the species in Fig. 4.The overall trends in hydraulic safety margins were consistentwith a diminishing role of C as Ystem declined (Fig. 5),increasing the likelihood that stomatal responses would notbe rapid enough to avoid catastrophic hydraulic failure unlesssafety margins increased or were consistently large.

CAPACITANCE AND HYDRAULIC SAFETY

Stomata regulate xylem pressure under dynamic conditions inwhich water released into the transpiration stream via dischargeof capacitance comes into play. Therefore, C must be consid-ered in attempts to understand how intact plants avoid hydrau-lic failure under the dynamic conditions that prevail in the field.The buffering influence of C can be viewed as a dynamic com-ponent of overall hydraulic safety because it varies with Ystemover daily and seasonal cycles. In contrast, properties such asPe and P50 are static over the daily time scale of stomatal regula-tion. The patterns reported here are consistent with a partialreliance on C to confer a margin of hydraulic safety rather thansole reliance on xylem structural features that govern propertiessuch as Pe and P50. Species-specific values of Ystem min and Cwere positively correlated (Fig. Sa). Additionally, previousstudies have noted positive correlations between C and P5() andbetween C and Ystem min - P50 (Domec & Gartner 2001; Prattet al. 2007; Meinzer et al. 2008a). Even the slope of the steeperportion of the vulnerability curve (Pe - P50) was steepest in spe-cieswith the highest C (Fig. 5b). Thus, declining species-specificvalues of C imply increasing reliance on xylem structural fea-rures that confer adequate hydraulic safety margins via changesin features such as Pe, P50 and Pe - P50.

Capacitive release of water via xylem cavitation can con-tribute to buffering of fluctuations in xylem pressure despitethe negative impact of cavitation on hydraulic conductivity(Holtta et al. 2009). However, capacitive release of waterfrom wood does not necessarily imply a trade-off involvingreduced conducting efficiency. In conifers, cavitation of late-wood tracheids can supply water to the transpiration streamwithout a significant loss of sapwood hydraulic conductivitybecause the conductivity of the more cavitation-resistant ear-lywood tracheids is about an order of magnitude greater thanthat oflatewood tracheids (Domec & Gartner 2001, 2002). Indiffuse-porous angiosperms, living axial and ray parenchymacan constitute up to 40-75% of sapwood volume (Panshin &de Zeeuw 1980; Chapotin et al. 2006) making these tissues apotentially important source of C. In several tropical tree spe-cies water released via cavitation of vessels accounted for onlyabout 15% of total daily reliance on C (Meinzer et al. 2008a).

Xylem hydraulic safely margins 929

When stem tissue is fully hydrated, species with higher val-ues of C are expected to show greater time constants forchanges in xylem pressure in response to changes in transpira-tion. These attributes would partly compensate for relativelylong time constants of several minutes for stomatal closure(Ceulemans et al. 1989; Jarvis et al. 1999; Allen & Pearcy2000; Powles et al. 2006), permitting safety margins based onfeatures such as Pe and P50 to appear small. However, asYstem min and Pe are approached, the hydraulic systembecomes nearly inelastic because C is essentially exhausted(Figs 6 and 7) causing time constants for transpiration-induced changes in xylem pressure to approach minimum val-ues, thereby making it difficult for stomata to respond quicklyenough to dampen abrupt decreases in xylem pressure thatcould trigger rampant embolism. The inverse relationshipbetween dynamic and static components of hydraulic safetyrepresents a trade-off whose common denominator is likelyto be wood density. Several studies have noted inverse rela-tionships between wood density and C (Meinzer et al. 2003,2008b; Pratt et at. 2007; Scholz et al. 2007), minimum shootY as well as variation in shoot Y (Stratton et al. 2000; Mein-zer 2003; Bucci et al. 2004b; Jacobsen et al. 2007, 2008),xylem specific conductivity (Stratton et al. 2000; Bucci et al.2004b; Meinzer et al. 2008b) and P50 (Hacke et al. 2001; Prattel al. 2007). However, some studies have not detected adependence of some of these traits on wood density (e.g. Jac-obsen et al. 2007; Meinzer et al.2008b).

Acknowledgements

This research was supported by National Science Foundation grants IBN9905012 and IOB 0544470.

References

Allen, M.T. & Pearcy, R.W. (2000) Stomatal behavior and photosyntheticperformance under dynamic light regimes in a seasonally dry tropical rainforest. Oecologia, 122, 470-478.

Bcgg, J.E. & Turner, N.C. (1970) Water pntential gradients in field tobacco.Plalll Physiology, 46, 343-346.

Brodribb, T.J. & Holbrook, N.M. (2004 ••) Diurnal depression of leaf hydraulicconductance in a tropical tree species. Plant Cell and Environment, 27,820-827.

Brodribb, T.J. & Holbrook, N.M. (2004b) Stomatal protection against hydrau-lic failure: a comparison of coexisting ferns and angiosperms. New Phytolo-gist, 162, 663-670.

Brodribb, TJ., Holbrook, N.M ..•Edwards, E.1. & Gutierrez, M.V. (2003) Rela-tions between stomatal closure, leaf turgor and xylem vulnerability in eighttropical dry forest trees. Plant, Cell and Environment, 26,443-450.

Bucci, S.J., Scholz, F.G., Goldstein, G., Meinzer, FiC. & Sternberg, L.D.L.(2003) Dynamic changes in hydraulic conductivity in petioles of two savannatree species: factors and mechanisms contributing 10 the refilling of ernbol-izcd vessels. Plant, Cell and Environment, 26, 1633-1645.

Bucci, S.J., Scholz, F.G., Goldstein, G., Meinzer, F.C., Hinojosa, J.A., Hoff-mann, W.A. & Franco, A.C. (2004a) Processes preventing nocturnal equili-bra tion between leaf and soil water polential in tropical savanna woodyspecies. Tree Physiology, 24, 1119-1127.

Bucci, S.J., Goldstein, G., Meinzer, F.e., Scholz, F.G., Franco, A.C. & Busta-mante, M. (2004b) Functional convergence in hydraulic architecture andwater relations of tropical savanna trees: from leaf to whole plant. TreePhysiology, 24, 891 899.

Ceulernans, R., Hinckley, T.M. & Impens, I. (1989) Stomatal response ofhybrid poplar to incident light, sudden darkening and leaf excision. Physiolo-gia Plantonsn, 75,174-182.



Chapotin, S.M., Razanamcharizaka, J.H. & Holbrook, N.M. (2006) A biorne-chanical perspective on the role of large stem volume and high water contentin baobab trees (Adansonia spp., Bombacaceae). American Journal of Bot-ally, 93, 1251-1264.

Cochard, H., Martin, R., Gross, P. & Bogeat-Triboulot, M.B. (2000) Tempera-lure effects on hydraulic conductance and water relations of Quercus roburL. Journal of Experimental Botany 51,1255·1259.

Cowan, LR. (1972) An electrical analogue of evaporation from. and flow ofwaterin plants. Planta, 106, 221-226.

Dixon, M.A. & Tyree, M.T. (1984) A new stem hygrometer, corrected for tem-perature-gradients and calibrated against the pressure bomb. Plant, Cell andEnviromnent, 7, 693 697.

Domee, J.-C. & Gartner, B.L. (2001) Cavitation and water storage capacity in bolexylem segments of mature and young Douglas-fir trees. Trees, 15, 204·214.

Domec, .f-C. & Gartner, B.L. (2002) How do water transport and storage differin coniferous earlywood and latewood? Journal of Experimental Botany 53,2369 2379.

Domcc, J.-C., Lachcnbruch, B., Meinzer, F.C., Woodruff, D.R., Warren, J.M.& McCulloh, K.A. (2008) Maximum height in a conifer is associated withconflicting requirements for xylem design. Proceedings of the National Acad-emy of Sciences of the United States of America, 105, 12069-12074.

Goldstein, G., Andrade, .l.L., Meinzer, F.e.,Holbrook, N.M., Caveiier,J.,Jack-son, P. &Celis, A. (1998) Stem water storage and diurnal patterns of water uscin tropical forest canopy trees. Plant, Cell and Environment., 21, 397--406.

Hacke, U.G., Sperry, J.S., Pockman, W.T., Davis, S.D. & McCulloh, K.A.(2001) Trends in wood density and structure are linked to prevention ofxylem implosion by negative pressure. Oecologia, 126,457-461.

Holtta, T., Cochard, H., Nikinmaa, E. & Mencuccini, M. (2009) Capacitiveeffect of cavitation in xylem conduits: results from a dynamic model. Plant,Cell and Environment, 32, 1Q.-21.

Hubbard, R.M., Ryan, M.G., Stiller, Y. & Sperry, J.S. (2001) Stomatal conduc-tance and photosynthesis vary linearly with plant hydraulic conductance inponderosa pine. Plant, Cell and Environment, 24, 113-121.

Jacobsen, A.L., Agenbag, L., Esler, K.1., Prall, R.B., Ewers, F.W. & Davis,S.D. (2007) Xylem density, biomechanics and anatomical traits correlatewith water stress in 17 evergreen shrub species of the Mediterranean-type cli-mate region of South Africa. Journal of Ecology, 95, 171-183.

Jacobsen, AL., Pratt, R.B., Davis, S.D. & Ewers, F.W. (2008) Comparativecommunity physiology: nonconvergence in water relations among threesemi-arid shrub communities. Nell' Phytologist, 180, 10Q.-lB.

Jarvis, A.J., Young, P.e., Taylor, C.J. & Davies, W.J. (1999) An analysis of thedynamic response of stomatal conductance to a reduction in humidity overleaves of Cedrella cordata. Plant, Cell and Environment, 22, 913-924.

Jones, H.G. & Sutherland, R.A. (1991) Stomatal control of xylem embolism.Plant, Cell and Environment, 14, 607-612.

Maherali, H., Pockman, W.T. & Jackson, R.B. (2004) Adaptive variation in thevulnerability of woody plants to xylem cavitation. Ecology, 85, 2184-2199.

Maherali, H., Moura, cr., Caldeira, M.C., Willson, c.r & Jackson, R.B. (2006)Functional coordination between leaf gas exchange and vulnerability to xylemcavitation in temperate forest trees. Piant, Cell and Environment, 29, 571-583.

Meinzer, F.e. (2003) Functional convergence in plant responses to the environ-ment. Oecologia, 134, 1-11.

Meinzer, F.e. & Grantz, D.A. (1990) Stomatal and hydraulic conductance ingrowing sugarcane: stomatal adjustment to water transport capacity. Plant,Cell and Environment, 13,383-388.

Meinzer, F.e., Clearwater, M.J. & Goldstein, G. (2001) Water transport intrees: current perspectives. new insights and some controversies. Environ-mental and Experimental Botany, 45, 239 262.

Meinzer, F.e.,James, S.A., Goldstein, G. & Woodruff, D.R. (2003) Whole-treewater transport scales with sapwood capacitance in tropical forest canopytrees. Plant, Cell and Environment, 26, 1147 1155.

Meinzer, F.C., James, S.A. & Goldstein, G. (2004) Dynamics of transpiration,sap flow and use of stored water in tropical forest canopy trees. Tree Physiol-ogy, 24, 901-909.

Meinzer, F.e., Woodruff, D.R., Domcc, J.-c., Goldstein, G., Campanello,P.l.,Gatti, M.G. & Villalobos-Vega, R. (2008a) Coordination of leaf and stemwater transport properties in tropical forest trees. Oecologia. 156, 31-41.

Meinzer, F.e., Campanello, P.l., Domec, J.e., Gatti, M.G., Goldstein, G., vm-alobos-Vega, R. & Woodruff, D.R. (2008h) Constraints on physiologicalfunction associated with branch architecture and wood density in tropicalforest trees. Tree Physiology, 28,1609 1617.

Melcher, PJ., Goldstein, G., Meinzer, F.C., Yount, D.E., Jones, T., Holbrook,N.M. & Huang, C.X. (2001) Water relations of coastal and estuarine Rhizo-phora mangle: xylem pressure potential and dynamics of embolism forma-tion and repair. Oecologia, 126, 182-192.

Mencuccini, M. (2003) The ecological significance of long-distance water trans-port: short-term regulation, long-term acclimation and the hydraulic costsof stature across plant life f011TIS. Plant Cell and Environment, 26, 163---182.

Nardini, A. & Salleo, S. (2000) Limitation of stomatal conductance by hydrau-Iictraits: sensing-orpreventing xylem cavitation? Trees, 15, 14-24.

Panshin, A.J. & de Zceuw, C. (1980) Textbook of Wood Technology. McGraw-Hill,NewYork.

Phillips, N., Nagchaudhuri, A., Oren, R. & Katul, G. (1997) Time constant forwater transport in loblolly pine trees estimated from time series of evapora-tive demand and stem sapflow. Trees, Il, 412 419.

Phillips, N.G., Oren, R., Licata, J. & Linder, S. (2004) Time series diagnosis oftree hydraulic characteristics. Tree Physiology, 24, 879-890.

Powles, J.E., Buckley, T.N., Nicotra, A.B. & Farquhar, G.D. (2006) Dynamicsof stomatal water relations following leaf excision. Plant, Cell and Environ-ment, 29,981-992.

Pratt, R.B., Jacobsen, AL., Ewers, F.W. & Davis, S.D. (2007) Relationshipsamong xylem transport, biomechanics and storage in stems and roots of nineRhamnaccac species ofthc California chaparral. Nell' Phytologist, 174, 787-798.

Ritchie, G.A. & Hinckley, T.M. (1971) Evidence for error in pressure-bombestimates of stem xylem potentials. Ecology, 30,534-536.

Sack, L. & Holbrook, N.M. (2006) Leaf hydraulics. Annual Review of PlantBiology, 57, 361-381.

Santiago, L.S., Goldstein, G., Meinzer, F.C., Fisher, J.B., Machado, K.,Woodruff, D. & Jones, T. (2004) Leaf photosynthetic traits scale withhydraulic conductivity and wood density in Panamanian forest canopy trees.Oecologia, 140, 543-550.

Scholz, F.G., Bucci, S.1., Goldstein, G.~ Meinzer, F.C., Franco, A.C. & Miral-les-Wilhelm, F. (2007) Biophysical properties and functional significance ofstem water storage tissues in neotropical savanna trees. Plant Cell and Envi-ronment, 30, 236--248.

Sparks, J.P. & Black, R.A. (1999) Regulation of water loss in populations ofPopulus trichocarpa: the role of srornatal control in preventing xylem cavita-tion. Tree Physiology, 19,453-459.

Sperry, J.S. (2004) Coordinating stomatal and xylem functioning - an evolu-tionary perspective. Nell' Phytoiogist, 162, 568-570.

Sperry, J.S., Alder, N.N. & Eastlack, S.E. (1993) The effect of reduced hydrau-lic conductance on stomatal conductance and xylem cavitation. Journal ofExperimental Botany, 44,1075--1082.

Stratton, L., Goldstein, G. & Meinzer, F.C. (2000) Stem water storage capacityand efficiency of water transport: their functional significance in a Hawaiiandry forest. Plant, Cell and Environment, 23, 99-106.

Tognetti, R., Longobucco, A. & Raschi, A. (1998) Vulnerahility of xylem toembolism in relation to plant hydraulic resistance in Quercus pubescens andQuercus ilex co-occurring in a Mediterranean coppice stand in central Italy.Nell' Phytologlst, 139,437-447.

Tyree, M.T. & Sperry, J.S. (1988) Do woody plants operate near the point ofcatastrophic xylem dysfunction caused by dynamic water stress? Plant Physi-ology, 88, 574-580.

Vogt, U.K. (2001) Hydraulic vulnerability, vessel refilling, and seasonal coursesof stem water potential of Sorb us aucuparia L. and Sambucus nigra L. Jour-nal ofExperimental Botany, 52, 1527-1536.

Woodruff, D.R., McCulloh, [CA., Warren, J.M., Meinzer, F.e. & Lachenb-ruch, B. (2007) Impacts of tree height on lcafhydraulic architecture and sto-matal control in Douglas-fir. Plant, Cell and Environment, 30, 559569.

Zwieniccki, M.A. & Holbrook, N.M. (1998) Diurnal variation in xylem hydraulicconductivity in white ash (Fraxinus americana L.), red maple (AceI' rubrum L.)and red spruoe (Picea rubens Sarg.). Plant, Cell and Environment, 21, 1173 1180.

Received 29 December 2008; accepted l3lvIarch 2009Handling Editor: David Whitehead

Supporting information

Additional supporting information may be found in the online ver-

sion of this article.

Table 81, Stem xylem vulnerability characteristics and daily mini-mum stem water potential.

Please note: Wiley-Blackwell is not responsible for the content or

functionality of any supporting information supplied by the authors.

Any queries (other than missing material) should be directed to the

corresponding author for the article.


Documents

Xylem hydraulic safety margins inwoody plants