Non-native Tree In A Dry Coastal Area In Hawai'i Has High Transpiration .

1y ago
13 Views
2 Downloads
557.24 KB
11 Pages
Last View : 1m ago
Last Download : 2m ago
Upload by : Tripp Mcmullen
Transcription

ECOHYDROLOGYEcohydrol. 9, 1166–1176 (2016)Published online 23 December 2015 in Wiley Online Library(wileyonlinelibrary.com) DOI: 10.1002/eco.1715Non-native tree in a dry coastal area in Hawai’i has hightranspiration but restricts water use despite phreatophytic traitYoshiyuki Miyazawa,1,2* Bruce D. Dudley,3,4,5 R. Flint Hughes,3 Joshua Vandemark,3,4Susan Cordell,3 Michael A. Nullet,1 Rebecca Ostertag4 and Thomas W. Giambelluca1251Department of Geography, University of Hawai’i at Mānoa, Honolulu, HI, 96822, USAResearch Institute for East Asia Environments, Kyushu University, 744 Motooka, Fukuoka 819-0395, Japan3Institute of Pacific Islands Forestry, USDA Forest Service, 60 Nowelo Street, Hilo, HI, 96720, USA4Department of Biology, University of Hawai’i at Hilo, 200 W. Kawili Street, Hilo, HI, 96720, USANational Institute of Water and Atmospheric Research, 10 Kyle Street, Riccarton, Christchurch, New ZealandABSTRACTIn arid systems, phreatophytes are often among the most effective invaders because of their capacity to access and exploitgroundwater resources otherwise unavailable to native vegetation. On Hawai’i Island, a non-native phreatophyte, Prosopispallida (kiawe), has invaded extensive dry lowland areas following its introduction in the 19th century. To better understand theinfluence of this invader on the host ecosystem, and vice versa, we determined transpiration characteristics of P. pallida bymonitoring sap flux of mature individuals over an 18-month period on the arid leeward coast of Hawai’i Island. Daily sap fluxincreased with increasing atmospheric vapour pressure deficit (D) and exhibited no clear response related to rainfall events orprolonged drought. Annual transpiration (i.e. 308 mm) was 80% higher than rainfall. Stomatal conductance decreased withincreasing vapour pressure deficits more slowly than the theoretical stomatal conductance–D relationship that assumes stomatalregulation of leaf water potential above a critical level. While daily peak stomatal conductance was relatively high, it nonethelessappeared limited by high intrinsic water use efficiency (net photosynthetic rate per stomatal conductance), suggesting a constraintto unlimited groundwater absorption. In this young volcanic environment where rainfall is low and quite episodic, the success ofthis species likely stems from its anisohydric strategy even at the risk of cavitation, and it is altering the hydrological cycling ofthe areas where it is dominant. Copyright 2015 John Wiley & Sons, Ltd.KEY WORDScoastal groundwater; carbon assimilation; invasive species; sap flow; stomatal controlReceived 12 February 2015; Revised 30 November 2015; Accepted 1 December 2015INTRODUCTIONNon-native species invasions have been definitively linked tochanges in energy, water, carbon and nutrient cycles in manyecosystems (Vitousek et al., 1987; D’Antonio and Vitousek,1992). In arid systems, phreatophytes are often among the mosteffective invaders because they are able to access groundwatertypically unavailable to native vegetation (Smith et al., 1998;Stromberg et al., 2007; Milton and Dean, 2010). Access to, anduse of, groundwater provides a considerable competitiveadvantage because it allows trees to transpire, assimilate carbonand thereby increase live biomass under limited rainfallconditions that typically occur during the dry seasons in savannaand tropical seasonal ecosystems (Hutley et al., 2000; Cleverlyet al., 2006; O’Grady et al., 2009; Pfautsch et al., 2011), whilephotosynthesis and transpiration of non-phreatophytic native*Correspondence to: Yoshiyuki Miyazawa, Research Institute for EastAsia Environments, Kyushu University, 744 Motooka, Fukuoka819-0395, Japan. E-mail: sclerophyll@gmail.comCopyright 2015 John Wiley & Sons, Ltd.plants are constrained by the difficulties in water uptake fromrelatively dry soil. As such, restrictions on water absorption fromthe dry soil of non-phreatophytic native plants under droughtconditions help to determine the degree to which invasivephreatophytes alter local hydrological processes (Hultine andBush, 2011), nutrient cycling (Dudley et al., 2014) andultimately plant community structure (Breshears et al., 2005).Hawai’i Island is an excellent place to examine theimpacts of invasive species on ecosystem function.Communities of plants and animals in the Hawaiian Islandshave been profoundly altered by introduced speciesbrought from other regions of the world, especially sincethe late 18th century (Cuddihy and Stone, 1990). Today,approximately half of Hawai’i’s flora is non-native(Wagner et al., 1999). In addition, the chemical homogeneity of the lava that forms the basis of Hawai’i’s soilsubstrates provides uniform settings in which to examineimpacts of non-native species on processes such as nutrientavailability (Vitousek et al., 1987; Hughes and Denslow,2005), carbon cycling (Sandquist and Cordell, 2007; Litton

NON-NATIVE PHREATOPHYTE IN DRY AREA RESTRICTS TRANSPIRATIONet al., 2008) and hydrological cycling (Cordell andSandquist, 2008) in native ecosystems.Throughout the dry coastal shrublands of leeward Hawai’i,many native plant communities have been replaced by invasivegrasses and non-native N-fixing tree species, particularlyLeucaena leucocephala and the phreatophytic tree Prosopispallida (D’Antonio and Vitousek, 1992; Wagner et al., 1999).P. pallida, known as ‘kiawe’ in Hawai’i and mesquite elsewhere,is native to Peru, Colombia and Ecuador and naturalized in PuertoRico and Australia as well as Hawai’i, where it is the morecommon of the two Prosopis species introduced to the HawaiianIslands from South America (Gallaher and Merlin, 2010).Following its introduction at the end of the 19th century,P. pallida has proliferated across leeward coastal areas,including much of the leeward coast of Hawai’i Island whereit has altered the ecology and possibly the hydrology of thesedry forest ecosystems (Gallaher and Merlin, 2010). It typicallydominates dry, disturbed, low-elevation areas, often abuttingleeward beaches and occupying raised limestone reefs, dryslopes and gulches, and degraded dry forests from sea level to 600 m above sea level (Wagner et al., 1999).Because of the high permeability of the soils andunderlying basalt rock, groundwater flow constitutes theprimary means of land-to-sea freshwater transport onleeward coasts of Hawai’i Island. This groundwater,derived primarily from high-altitude recharge, is a criticalresource to coastal human communities as well asterrestrial flora and fauna. Groundwater also transfersterrestrially derived nutrients to marine environments(Knee et al., 2010), helping to sustain economically andculturally important marine ecosystems along these coasts(Duarte et al., 2010). Unfettered groundwater use byP. pallida, a species capable of high rates of transpiration(Scott et al., 2004), has the potential to reduce freshwaterinputs and alter nutrient transport to near-coastal marineenvironments.Dudley et al. (2014) documented P. pallida’s access togroundwater along leeward coasts and found that thisaccess resulted in considerable increases in the size anddensity of P. pallida individuals. However, high stablecarbon isotope ratio, low predawn leaf water potential andmidday leaf water potential values close to the critical levelfor cavitation in other Prosopis species (Dzikiti et al.,2013) suggest that groundwater uptake may be constrainedto lower levels than those demanded by the transpiringfoliage and that transpiration rates are likely to be differentfrom those of trees under wet soil conditions, in bothmagnitude and the response to the varying environmentaldrivers. During drought periods, many species elsewhereexhibit ecophysiological traits that maintain transpirationrate at low levels and have specific responses to waterrestriction, characterized by reduced stomatal conductanceunder increasing atmospheric evaporative demand in orderto maintain the minimum leaf water potential at a constantCopyright 2015 John Wiley & Sons, Ltd.1167level (isohydric stomatal control, Oren et al., 1999). Leavesreduce stomatal conductance at low leaf water potential(Epron and Dreyer, 1993; Sala and Tenhunen, 1996) toavoid increasing transpiration, which would result inreduction in leaf water potential beyond the minimumlevel and avoidance of cavitation (McDowell et al., 2008a),especially under high air vapour pressure deficits that occurnear midday. This response to drought is also characterizedby sharp increases in transpiration and stomatal conductance when leaf water potential rises following episodicrainfall events, as seen in other Prosopis species (Scottet al., 2006; Dzikiti et al., 2013).A potential alternative response to restricted groundwater access is to maximize stomatal conductance, continuehigh transpiration rates and further reduce leaf waterpotential, a strategy that increases transpiration rates,carbon gain (Farquhar and Sharkey, 1982) and the risk ofcavitation (McDowell et al., 2008a; Manzoni et al., 2013).If P. pallida exhibits this latter type of stomatal control(anisohydric stomatal control), stomatal conductanceshould not decline during rainless periods, nor shouldtranspiration rates increase in response to rainfall events,allowing higher transpiration rates than the level that isexpected to be achieved under isohydric stomatal control(Oren et al., 1999).In this study, we monitored the sap flux of P. pallidastands with ample access to shallow groundwater andcompared the annual transpiration rate with annualrainfall to evaluate the influence of P. pallida invasionon local hydrology. We examined several hypotheses,under the basic assumption that transpiration characteristics of P. pallida follow those of other Prosopisspecies (isohydric stomatal control and reduced transpiration rates during rainless periods) (Scott et al.,2006; Dzikiti et al., 2013; but see Martínez-Vilaltaet al., 2014). First, we hypothesized that sap fluxincreases with vapour pressure deficit but that itsincrease is saturated at low atmospheric vapour pressuredeficit on diurnal and longer timescales. Second, wehypothesized that stomatal conductance declines withincreasing vapour pressure deficit and maintains astomatal conductance–vapour pressure deficit relationship that corresponds to the theoretical relationship foroptimum maintenance of leaf water potential (Orenet al., 1999). Third, we hypothesized that sap flux andthe stomatal conductance–vapour pressure deficit relationship would change substantially in response torainfall events and that the leaf-level ecophysiologicaltraits would vary in close relation to stomatalconductance (Collatz et al., 1991; Katul et al., 2003).In light of projected future decreases in rainfall in dryleeward areas of Hawai’i in response to global climatewarming (Elison Timm et al., 2015), the response ofP. pallida stomatal control to drought and episodic rainfallEcohydrol. 9, 1166–1176 (2016)

1168Y. MIYAZAWA ET AL.will likely be important in dictating future ecosystemlevel impacts of this species.MATERIALS AND METHODSStudy site and speciesOur study site was located near Kīholo Bay, Hawai’i Island,Hawai’i (19·85466 N, 155·92416 W), on the coastal marginof a basaltic formation derived from a 3000- to 5000-year-oldlava flow emanating from Hualālai Volcano (Wolfe andMorris, 1996). The substrate is poorly weathered, largelyowing to low rainfall, and soil at the study site is shallow anddiscontinuous, with a mean depth of less than 5 cm (Dudleyet al., 2014). Roots extend into the ground through cracks inthe rock. Kīholo Bay receives considerable shallow submarine groundwater discharge near its shoreline via freshwaterflow from montane areas (Peterson et al., 2009). Dominantvegetation consists of P. pallida stands and ephemeralherbaceous species, primarily Sida fallax (’Ilima) andWaltheria indica (’Uhaloa), which partially cover the areafollowing heavy rain events. Although P. pallida trees mayattain heights 20 m in areas with abundant moisture(Y. Miyazawa, personal observation), stands surveyed alongthe coast of Kīholo Bay were approximately 8 m in height.Our study site was established in a closed P. pallida stand inMay 2010 at the same location as Dudley et al. (2014). Annualrainfall of the study site averaged 270 mm over the 30-yearperiod ending in 2007 (Giambelluca et al., 2013). BetweenAugust 2010 and August 2011, annual air temperatureaveraged 25 C, and annual rainfall was 188 mm, approximately two-thirds of the long-term annual average for the area.A weather station equipped with a pyranometer (CMP3, Kipp& Zonen, Delft, the Netherlands), a photosynthetically activeradiation (PAR) sensor (LI-190SB, Li-Cor, Lincoln, NE,USA), a tipping-bucket rain gauge (TE525, Texas Electronics,Dallas, TX, USA), air temperature–humidity sensors(HMP45C, Vaisala, Vantaa, Finland) and an anemometer(014a, Met One Instruments, Grants Pass, OR, USA) wasestablished in an open area adjacent to the sap flow monitoringsite. Sensors were monitored by a data logger (CR3000,Campbell Scientific, Logan, UT, USA), which sampled dataat a 10-s interval and recorded mean values (sums in the caseof rainfall) at a 10-min interval. Because of a power supplyinterruption, weather data were not collected for 24 days inJanuary 2011. Air temperature and relative humidity datawere used to calculate atmospheric vapour pressure deficit (D,kPa). Because transpiration is linked to D in daylight, mean Dwas calculated for both daytime (6:00–18:00 hours) andcomplete 24-h periods.In order to scale up transpiration (E, g m 2 s 1 ormm day 1) from mean stand xylem sap flux (Js, g m 2 s 1)as E Js stand-scale sapwood area/ground area, wemeasured P. pallida stem metrics in five 20-m radius plotsCopyright 2015 John Wiley & Sons, Ltd.established within coastal P. pallida stands known to haveaccess to groundwater as determined by oxygen isotopemeasurements (Dudley et al., 2014). Within each plot,diameters of all stems present at 2·5 m above the groundwere measured at that point – a height coinciding with thelocation at which sap flow sensors were installed inselected trees. We estimated stand-level sapwood area froma regression relationship that predicted sap wood area fromstem diameter and was derived from 72 P. pallida stemcross sections ranging in diameter from 2·9 to 59 cm at2·5 m above the ground (sapwood area 2·18 diameter0·954,r2 0·733, p 0·001, Figure S1). For details about thecalculation of sapwood area, see Supporting Information 1.Leaf area index (LAI, m2 leaf m 2 ground) of the standwas periodically measured using paired LAI-2000 instruments (Li-Cor); one sensor was located in an open site, andthe second at designated points within P. pallida stands. Ineach of the five plots, we measured LAI at four permanentpoints, facing towards, and equidistant from, the plotcentre. During any given measurement, the sensor in theopen area collected data at the same time and in the samedirection as the within-canopy sensor. Measurements werealways collected during overcast sky conditions.Sap flow measurementsThe Js (g m 2 s 1) was estimated using Granier-type sapflux sensors (Granier, 1987) modified slightly with respectto the length of the probes (James et al., 2002). A datalogger (CR1000, Campbell Scientific) attached to amultiplexer (AM16/32, Campbell Scientific) scanned theoutput from the sensors at a 30-s interval, and averagevalues were recorded at 10-min intervals. We installed sapflux sensors at a height of 2·5 m because bark and stems atlower positions were observed to be frequently damaged byferal goats. Sensors were installed on 17 selected trees inMay 2010. Details about the sensors, installation andthe calculation of sap flux are given in SupportingInformation 2.As Js and E are functions of atmospheric D and leafstomatal conductance (Jarvis and McNaughton, 1986),stomatal conductance was calculated on a unit leaf areabasis (Gs, mol m 2 leaf s 1) and on a unit sapwood areabasis (Gs sap, m s 1, Meinzer et al., 2013) during periodsfor which LAI data were not available, based on Js and D(Supporting Information 3). We determined the relationship between Gs and D by performing boundary lineanalysis under high solar radiation (solarradiation 600 W m 2) (Supporting Information 3). Theselected Gs points were regressed against ln(D):Gs ¼ Gsref –m lnðDÞ(1)where m is the slope Gs/ lnD (mol kPa 1 m 2 s 1) andGsref is the intercept at D 1 kPa (mol m 2 s 1). The ratioEcohydrol. 9, 1166–1176 (2016)

NON-NATIVE PHREATOPHYTE IN DRY AREA RESTRICTS TRANSPIRATIONof m to Gsref was obtained for each month for comparisonwith the theoretical value (0·6) for isohydric leaves with perfectstomatal control of minimum leaf water potential (Oren et al.,1999). When LAI data were not available, a similar analysiswas carried out using Gs sap and obtained Gs sapref and msap.We employed Gsref and m obtained in the boundary lineanalysis to model Gs during midday (10:00–15:00 hours) fortwo cases with different assumptions on the stomatal control(m–Gsref relationship). In case 1, modelled Gs was calculatedas Gs Gsref mlnD. Similarly in case 2, which assumedisohydric stomatal control (i.e. m/Gsref 0·6), modelled Gswas computed as Gs Gsref 0·6GsreflnD. If P. pallida doesnot adopt isohydric stomatal control, i.e. if it exhibits speciesspecific m/Gsref values, modelled Gs in case 1 should differfrom modelled Gs in case 2; if modelled Gs in case 1 is largerthan modelled Gs in case 2, increased Gs of P. pallida shouldbe partly attributed to the species-specific stomatal controlrepresented by the m/Gsref value. Modelled Gs values in cases1 and 2 were compared to evaluate the extent to whichP. pallida exhibited enhanced or reduced Gs during middayby adopting species-specific stomatal control, rather thanisohydric stomatal control.Measurements of leaf ecophysiological traitsDuring periods preceded by rainfall events (January andAugust 2011 and July 2012), leaf-level ecophysiologicaltraits were measured using an LI-6400 portable photosynthesis system (Li-Cor). We obtained a relationship betweennet photosynthetic rate (A, μmol m 2 s 1) and intercellularCO2 concentration using intact leaves from the trees selectedfor sap flow measurements in order to determine theparameter described by Farquhar et al. (1980): the maximumrate of RuBP carboxylation at a common leaf temperature 25 C (Vcmax25, μmol m 2 s 1). Leaves were selected from thesun-exposed crown surface within reach (1–1·5 m height). Atotal of 8–12 leaves from each of the three trees were used todetermine Vcmax25 on each sample day.In addition to Vcmax25, we also determined leaf-levelintrinsic water use efficiency (iWUE, mmol mol 1) and theintercellular CO2 concentration relative to the air CO2concentration (Ci/Ca) for sunlit leaves and partially shadedleaves, respectively. Leaf-level iWUE was calculated as theratio of A to the stomatal conductance for water vapour (gsw,mol m 2 s 1). We derived mean iWUE from the diurnallyvarying leaf gas exchange rate data collected between 10:00and 15:00 hours when 70–80% of daily transpiration typicallyoccurs. Details about photosynthesis measurements, samplings and the calculation of the parameters are shown inSupporting Information 4.Statistical analysisWe used t-tests (Sokal and Rohlf, 1995) to compareVcmax25 and iWUE among periods that varied with regardCopyright 2015 John Wiley & Sons, Ltd.1169to the amount of preceding rainfall. Similarly, t-tests wereused to compare between sun-exposed leaves and shadedleaves.RESULTSEnvironmental variation and tree water statusSolar radiation (Figure 1a) exhibited a large annual cycle,reaching its lowest levels in winter. High rainfall eventsoccurred in winter–spring (December 2010–May 2011),and a period of very low rainfall occurred from June toOctober 2011. Monthly rainfall was 81 mm in December2010 but was less than 20 mm in other months (Figure 1a).Despite these variations in rainfall, LAI did not differbetween seasons (Figure 1d).Mean daytime and whole day D remained highthroughout the study period (Figure 1b), and neitherapproached zero, even during predawn hours.Sap flux and its response to the environmentDaily Js of P. pallida stand exhibited a clear annual cycle,with high levels in summer and lower levels in winter(Figure 1c). Low Js was observed in January–February,attaining only 60–70% of the levels exhibited in July–September. Annual stand-level E was 350 mm year 1(August 2010–July 2011) and exceeded annual rainfallby 160 mm year 1 (84%), indicating that water sourcesother than rainfall supported transpiration. Importantly, wedid not observe decreasing trends in Js during long rainlessperiods (June–September 2011) or sudden increases afterthe high rainfall event in December 2010.Js was positively correlated with varying seasonal patternsof D (Figure 2). The Js–D relationships did not differ betweenthe rainless period from June to September 2011 and the postrainfall period in December 2010. Patterns clearly indicatedthat variation in transpiration was a function of D and provideno evidence of saturation of the increase of Js with D inresponse to limitation by other factors.At a fine temporal resolution (i.e. 10 min), we observed anegative correlation (r2 0·646–0·922, p 0·01, in eachmonth) between D and Gs when upper envelopes of theGs–D relationship were compared (Figure 3a). Althoughthe ratio of msap to Gs sapref approached 0·6 in three of the14 months, the slope was less than 0·5 for seven months(Figure 4b). Gs sapref and msap/Gs sapref changed seasonallywithout a clear relationship with the preceding month’srainfall, suggesting that rainfall did not influence the Gs–Dor E–D relationship of P. pallida.Irrespective of rainfall during the preceding month, Gsdid not exhibit a clear midday depression at high PARlevels ( 1500 μmol m 2 s 1, Figure 4b). Reductions in Gsin the afternoons during July coincided with reductions inPAR. In February 2011 when m/Gsref was near 0·6,Ecohydrol. 9, 1166–1176 (2016)

1170Y. MIYAZAWA ET AL.Figure 1. Seasonal trends in (a) daily solar radiation (circles) and rainfall (bars), (b) whole day (open circles) and midday (10:00–14:00 hours, closedcircles) atmospheric vapour pressure deficit (D), (c) sap flux (Js) and (d) leaf area index (LAI) of the study site (mean standard deviation, n 5). Forsolar radiation (a), grey symbols show daily values adjusted to correct for the influence of shading by nearby trees (Section on Materials and Methods).modelled Gs in cases 1 and 2 exhibited similar diurnal trendsto the measured Gs; that is, values were similar to means 1standard deviation of Gs (Figure 4c). In July, a period when m/Gsref 0·4, modelled Gs in case 2 continued to decrease withincreasing D between 10:00 and 12:00 hours and wasconsistently lower than the measured Gs. Large differencesof modelled Gs values between cases 1 and 2 were due to thediffering assumptions for stomatal control inherent in thesetwo measures (i.e. anisohydric or isohydric stomatal control).Copyright 2015 John Wiley & Sons, Ltd.Photosynthesis measurementsPhotosynthetic capacity, or Vcmax25 of sun-exposed leaves,remained high and exhibited no differences betweenFebruary 2011 and August 2010–2011 despite largedifferences in antecedent rainfall amount between theseperiods (Figure 5). Seasonal trends and interannualvariation in Vcmax25 were less clear. Shaded and sunexposed leaves did differ with regard to Vcmax25 over eachmeasurement period (p 0·05) (Table I).Ecohydrol. 9, 1166–1176 (2016)

NON-NATIVE PHREATOPHYTE IN DRY AREA RESTRICTS TRANSPIRATION3four times in each period (Table II). In each period, incontrast to the diurnal trend in Gs, gsw of sun-exposedleaves showed strong midday depression to 30–50% of thepeak after 9:00 hours (data not shown). Sun-exposed leavesdid not exhibit differences in iWUE around middaybetween August and February 2011, despite the differingantecedent precipitation regimes between those periods.Leaves maintained high iWUE values characterized by lowCi/Ca (Table II). Despite differences in Vcmax25, gsw andPAR, sun-exposed leaves and partially shaded leaves didnot differ with respect to iWUE (Table II).Js (m3 m-2 day-1) 10mm 20mm 20mm21y 2.039[1-exp(-0.695x)]r2 0.6900121171DISCUSSION33.5Daytime D (kPa)Figure 2. The relationship between sap flux density (Js) and daytimeatmospheric vapour pressure deficit (D) for days with rainfall amount inthe precedent one month: 10 mm (open circles), 20 mm (closed circles)and 20 mm (closed triangles). The equation is for the regression line forall the pooled data.Access to groundwater for transpirationDespite the low rainfall of Kīholo Bay, fairly unweatheredpāhoehoe basalt substrate and very patchy, shallow soilsthat characterize these leeward coastal environments ofHawai’i Island, our results clearly indicate that P. pallidastands were able to successfully access and utilizegroundwater and subsequently transpire substantially( 80%) more water than was supplied via rainfall. Thiscapacity has been documented for other phreatophytespecies in other environments (Sala et al., 1996; Scottet al., 2006; Canham et al., 2012). Prosopis species areknown to extend their tap roots to great depths in theirnative habitat (Prosopis juliflora, Canadell et al., 1996) tomaintain transpiration, although soil conditions in thesenative conditions may differ considerably from the youngbasaltic substrates of Hawai’i Island.Because other woody species in nearby areas do notexhibit the capacity to access groundwater and sotypically reduce transpiration in response to dry periods(Sandquist and Cordell, 2007; Cordell and Sandquist,2008), our results support the idea that invasion ofP. pallida has altered the hydrological processes of thisregion, which is now characterized by a negative waterbudget through groundwater uptake and continuoustranspiration.Transpiration characteristics of Prosopis pallida trees inarid Kīholo BayFigure 3. (a) The relationship between stomatal conductance (Gs sap) andatmospheric vapour pressure deficit (D) for upper boundary data under 2non-limiting solar radiation (600 W m ). (b) The relationship between theslope of Equation 1 for stomatal conductance on a unit sapwood area basis(Gs sap), msap, and intercept, Gs sapref. Each point indicates the values forone month. The diagonal line with slope 0·6 shows the theoretical valuefor perfect stomatal control of minimum leaf water potential (Section onMaterials and Methods).Around midday, peak gsw of sun-exposed leaves washigher in February 2011 than in August 2012 and that ofpartially shaded leaves and was higher than Gs by three toCopyright 2015 John Wiley & Sons, Ltd.Low predawn leaf water potential is known to inducestomatal closure in many species (Pereira et al., 1987;Epron and Dreyer, 1993; Sala and Tenhunen, 1996) – amechanism to avoid hydraulic failure by restricting E(Sperry et al., 2002; McDowell et al., 2008a). Wehypothesized that leaves of P. pallida with low predawnleaf water potential would restrict Gs to low levels inmidday and avoid excessive increase in E under high D(Irvine et al., 1998; Williams et al., 1998). In contrast toour first hypothesis, transpiration by P. pallida wascharacterized by a positive Js–D relationship (Figure 2)Ecohydrol. 9, 1166–1176 (2016)

1172Y. MIYAZAWA ET AL.Figure 4. Monthly mean diurnal cycles in (a and b) atmospheric vapour pressure deficit (D, solid line) and photosynthetically active radiation (PAR,broken line) and (c and d) measured stomatal conductance (Gs, bold line) and modelled Gs in case 1 (thin line) and case 2 (broken line). The grey areasindicate means standard deviation for D (a and b) and Gs (c and d). The plots shown are for February 2011 (a and c) and July 2011 (b and d). Thenumbers above the panels indicate the mean rainfall in the preceding month.Figure 5. Photosynthetic capacity (means standard deviation) ofProsopis pallida leaves at 25 C (Vcmax25) for sun-exposed leaves at theupper canopy surface (black bars) and deeply shaded leaves at the bottomof the canopy (white bars) obtained in periods with different antecedentrainfall amounts (high: January 2011, low: July 2011 and August 2012).Vertical bars indicate the standard deviation among measured leaves.Asterisks indicate significant difference at p 0·05 between sun-exposedleaves and the deeply shaded leaves (t-test).and the absence of midday depression (Figure 4), suggesting that water uptake from groundwater was sufficient tomeet the seasonally and diurnally variable D (Sala et al.,Copyright 2015 John Wiley & Sons, Ltd.1996; Hutley et al., 2000). Despite low predawn leaf waterpotential, P. pallida was able to take up water to meetdaytime demand by creating a high gradient of waterpotential between groundwater and leaf (Meinzer et al.,1995); resulting midday leaf water potential values ( 3 to 4 MPa, Dudley et al., 2014) were near the critical levelfor cavitation in other Prosopis species in South Africa(Dzikiti et al., 2013).Prosopis pallida exhibited anisohydric stomatal control,and the assumption of isohydric stomatal control(m 0·6Gsref) in our second hypothesis would considerablyunderestimate Gs and E of P. pallida during most periods(Figure 4d). During most periods, the ratio of m to Gsrefwas lower than the theoretical value for perfect control ofminimum leaf water potential: 0·6 (Oren et al., 1999)(Figure 3). This suggests that P. pallida controlled stomataat the risk of reducing leaf water potential with increasingD (McDowell et al., 2008a; Manzoni et al., 2013).Anisohydric behaviour of P. pallida leaves was supportedby leaf water potential values in the native region(Martínez-Vilalta et al., 2014) and at the site during theperiod of study; midday leaf water potential was notconstant but varied between approximately –2 and 4 MPa(Dudley et al., 2014). Similarly, high risk for cavitationcoincides with the observation of many dead branches andstems in this region. Even within a single indi

transpiration rates increase in response to rainfall events, allowing higher transpiration rates than the level that is expected to be achieved under isohydric stomatal control (Oren et al., 1999). In this study, we monitored the sap flux of P.pallida stands with ample access to shallow groundwater and compared the annual transpiration rate .

Related Documents:

Civic Style - Marker Symbols Ü Star 4 û Street Light 1 ú Street Light 2 ý Tag g Taxi Æb Train Station Þ Tree 1 òñðTree 2 õôóTree 3 Ý Tree 4 d Truck ëWreck Tree, Columnar Tree, Columnar Trunk Tree, Columnar Crown @ Tree, Vase-Shaped A Tree, Vase-Shaped Trunk B Tree, Vase-Shaped Crown C Tree, Conical D Tree, Conical Trunk E Tree, Conical Crown F Tree, Globe-Shaped G Tree, Globe .

NATIVE INSTRUMENTS GmbH Schlesische Str. 29-30 D-10997 Berlin Germany www.native-instruments.de NATIVE INSTRUMENTS North America, Inc. 6725 Sunset Boulevard 5th Floor Los Angeles, CA 90028 USA www.native-instruments.com NATIVE INSTRUMENTS K.K. YO Building 3F Jingumae 6-7-15, Shibuya-ku, Tokyo 150-0001 Japan www.native-instruments.co.jp NATIVE .

NATIVE INSTRUMENTS GmbH Schlesische Str. 29-30 D-10997 Berlin Germany www.native-instruments.de NATIVE INSTRUMENTS North America, Inc. 6725 Sunset Boulevard 5th Floor Los Angeles, CA 90028 USA www.native-instruments.com NATIVE INSTRUMENTS K.K. YO Building 3F Jingumae 6-7-15, Shibuya-ku, Tokyo 150-0001 Japan www.native-instruments.co.jp NATIVE .

Family tree File/directory tree Decision tree Organizational charts Discussion You have most likely encountered examples of trees: your family tree, the directory . An m-ary tree is one in which every internal vertex has no more than m children. 2. A full m-ary tree is a tree in which every

Search Tree (BST), Multiway Tree (Trie), atau Ternary Search Tree (TST). Pada makalah ini kita akan memfokuskan pembahasan pada Ternary Search Tree yang bisa dibilang menggabungkan sifat-sifat Binary Tree dan Multiway Tree. . II. DASAR TEORI II.A TERNARY SEARCH TREE Ternary Search Tr

Torres, Julie West, Speaking up! Adult ESL students' perceptions of native and non-native English speaking teachers. Master of Arts (English as a Second Language), December 2004, 89 pp., 2 tables, 2 figures, references, 38 titles. Research to date on the native versus non-native English speaker teacher (NEST

React-Native Apps JS components render as native ones Learn once, write everywhere 13 Android Android SDKs Native UI JS Runtime React Native 3rd Party Libs NPM Pkgs (e.g., React) Bridge Your App Your App (JS) (Native UI & Modules) iOS iOS SDKs Native UI JS Runtime React Native 3 Party Libs NPM Pkgs (e

Native Village of Port Heiden Lake and Peninsula Borough Meshik Inc. Bristol Bay Native Corporation Port Heiden Native Village of Port Lions Kodiak Island Borough Afognak Native Corp Koniag, Incorporated Port Lions Native Village of Ruby Yukon-Koyukuk Census Area Dineega Corp Doyon, Limited Ruby Native Village of Saint Michael Nome Census Area .