Size-Mediated Tree Transpiration Along Soil Drainage Gradients in a Boreal Black Spruce Forest Wildfire Chronosequence

Home , Larix laricina, Picea mariana, Ptilium (plant)

Tree Physiology 32, 599–611 doi:10.1093/treephys/tps021

Research paper

Size-mediated tree transpiration along soil drainage gradients in a boreal black spruce forest wildfire chronosequence

J.L. Angstmann1,3, B.E. Ewers1 and H. Kwon2 Downloaded from

1Department of Botany, Program in Ecology, University of Wyoming, 1000 E. University Avenue, Laramie, WY 82072, USA; 2Yonsei University, Seoul, Korea; 3Corresponding author ([email protected])

Received September 23, 2011; accepted February 24, 2012; published online April 25, 2012; handling Editor Nathan Phillips http://treephys.oxfordjournals.org/

Boreal forests are crucial to climate change predictions because of their large land area and ability to sequester and store carbon, which is controlled by water availability. Heterogeneity of these forests is predicted to increase with climate change through more frequent wildfires, warmer, longer growing seasons and potential drainage of forested wetlands. This study aims at quantifying controls over tree transpiration with drainage condition, stand age and species in a central Canadian black spruce boreal forest. Heat dissipation sensors were installed in 2007 and data were collected through 2008 on 118 trees (69 Picea mariana (Mill.) Britton, Sterns & Poggenb. (black spruce), 25 Populus tremuloides Michx. (trembling aspen), 19 Pinus banksiana Lamb. (jack pine), 3 Larix laricina (Du Roi) K. Koch (tamarack) and 2 Salix spp. (willow)) at four at University of Wyoming Libraries on June 27, 2016 stand ages (18, 43, 77 and 157 years old) each containing a well- and poorly-drained stand. Transpiration estimates from sap flux were expressed per unit xylem area, JS, per unit ground area, EC and per unit leaf area, EL, using sapwood (AS) and leaf

(AL) area calculated from stand- and species-specific allometry. Soil drainage differences in transpiration were variable; only the 43- and 157-year-old poorly-drained stands had ~ 50% higher total stand EC than well-drained locations. Total stand EC tended to decrease with stand age after an initial increase between the 18- and 43-year-old stands. Soil drainage differences in transpiration were controlled primarily by short-term physiological drivers such as vapor pressure deficit and soil moisture whereas stand age differences were controlled by successional species shifts and changes in tree size (i.e., AS). Future predictions of boreal climate change must include stand age, species and soil drainage heterogeneity to avoid biased estimates of forest water loss and latent energy exchanges.

Keywords: chronosequence, Picea mariana, Pinus banksiana, Populus tremuloides, poorly drained, soil moisture.

Introduction Bond-Lamberty et al. 2005, 2009). Poorly-drained soils result Ecosystem water fluxes have ecological and socioeconomic from the pooling of snow melt in lower-lying stands that have consequences including controls over carbon cycling, water Sphagnum-insulated permafrost, inhibiting water drainage into availability for plants and rivers, and subsequent agricultural deeper soil layers (Gorham 1991, Bond-Lamberty et al. and hydroelectric power production (Mimikou and Baltas 1997, 2004a). Paludification, or the buildup of deep organic soils as Jackson et al. 2005, Bonan 2008). Boreal forests are of par- a result of cold, anoxic conditions, results in dwarfed tree mor- ticular interest in climate change studies because of their large phology (Roy et al. 1999, Lavoie et al. 2005). While some spe- land area, location in northern latitudes (Manabe and Wetherald cies have shown adaptation and/or acclimation to flooded 1980, Goulden et al. 1998, Stocks et al. 1998), ability to conditions, such as lenticel and adventitious root (i.e., root sequester and store carbon in both aboveground plant matter structures originating from the stem) formation and symbiotic and belowground organic peat soils (Drushka 2003) and the relations with mycorrhizae, little is known about the influence presence of both soil water excess and drought (Grant 2004, of these structures on ecosystem fluxes of water and carbon in

© The Author 2012. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] 600 Angstmann et al. the boreal forest (LeBarron 1945, Vartapetian and Jackson because of unlimited root water access and subsequent high 1997, Fougnies et al. 2007). While the mechanisms behind flows of water through the transpirational stream C( ˇermák et al. paludification are well known, its impact on tree transpiration 1982, Roy et al. 1999, Herrera et al. 2008) while differences has not been investigated with changes in stand age and in EL and EC are primarily controlled by age-mediated (Dawson species. 1996, Alsheimer et al. 1998, Ryan et al. 2000, Ewers et al. Failure to account for drastic changes in topography-driven 2002, Moore et al. 2004, Meinzer et al. 2005) and soil drain- soil moisture in models quantifying carbon, water and energy age-mediated (Akeroyd et al. 1998, Kreuzwieser et al. 2002) fluxes may result in biased predictions. This is especially true differences in tree size. Ewers et al. (2005) hypothesized that with models quantifying parameters under various global early successional species in well-drained boreal stands would change scenarios because well- and poorly-drained stands will undergo increasing EL and EC with age, late-successional spe- be dichotomously affected by global change. Under a warmer, cies would experience decreasing EL and increasing EC, with drier climate with increased boreal fire frequency, well-drained the EL decrease attributed to a decrease in AS:AL found in coni- stands will likely become younger, more diverse and drought fers with a low AS:AL and long-lived leaves (McDowell et al. Downloaded from stressed, whereas poorly-drained stands will undergo soil 2002). Furthermore, trees in poorly-drained conditions are drainage through permafrost melt increasing production and expected to have lower EC and EL than well-drained trees due decomposition (Goulden et al. 1998, Stocks et al. 1998, to lower AS and AL (Oren et al. 1999, Roy et al. 1999, Islam and

Schuur et al. 2008). Knowledge of the physiological and MacDonald 2004) but a similar pattern with age because AS:AL resource allocation controls over transpiration will refine pre- is constant regardless of soil drainage condition (e.g., no sig- http://treephys.oxfordjournals.org/ dictive models by accounting for divergent stand responses of nificant effect of soil drainage on allometric relationships; individual physiology and resources allocation patterns. Bond-Lamberty et al. 2002b). Short-term physiological controls over water and carbon We aim to explain the variability of tree transpiration across fluxes come from stomata which respond indirectly to vapor a soil drainage gradient by measuring sap flux of three boreal pressure deficit D( ) by plant regulation of minimum leaf water species (two early-successional and one late-successional) potential to avoid excessive cavitation during drought periods across four stand ages, each containing a well- and poorly- (Cowan and Farquhar 1977, Meinzer and Grantz 1991, Mott drained stand. We test the following predictions from data and Parkhurst 1991, Baldocchi 1997, Franks 2004). Extremes collected during the 2007 and 2008 growing seasons: at University of Wyoming Libraries on June 27, 2016 in soil moisture can also influence transpiration rates where (i) irrespective of stand age, well-drained stands will have both drought and flooded conditions inhibit root uptake of higher EL and EC than poorly-drained stands because reduc- water due to high negative soil water potentials or cold, anoxic tions in total AS and L will overwhelm higher JS in poorly- and/or nutrient-poor soil conditions, respectively (Gorham drained stands resulting from unlimited water availability and

1991, Wang et al. 2003a, Brodribb et al. 2005). Temporal (ii) younger stands will have higher EL and lower EC than older changes in transpiration are primarily driven by environmental stands because stand-level decreases in AS:AL with age are factors such as D and photosynthetic photon flux density Q( ) expected to occur in this ecosystem, which is dominated by where transpiration rates typically show a saturation response long-lived, low AS:AL coniferous black spruce trees. to increasing D and Q due to stomatal response to regulate minimum leaf water potential (Monteith 1995, Cowan and Farquhar 1997, Oren et al. 1999, Pataki et al. 2000, Franks Methods 2004, Ewers et al. 2005). Regulation of transpiration via sub- daily and daily plant physiological responses results in longer- Stand description term partitioning of resources that is highly dependent on the This study was conducted during the 2007 and 2008 air and soil environment. growing seasons in the central Canadian boreal forest in Consequently, long-term resource partitioning influences northern Manitoba, CA (55.8786°N–55.9137°N, transpiration rates and can be quantified by expressing transpi- 98.3818°W–98.9793°W), ~40 km west of Thompson, MB. ration at different scales within trees. For example, sap flux per Soils were sedimentary deposits from glacial Lake Agassiz unit xylem area (JS) rescaled to transpiration per unit ground with montmorillonite clay soil in the well-drained stands 3 −3 area, EC, or per unit leaf area, EL, by multiplication with sap- (volumetric soil moisture range 0.08–0.72 cm cm ) and wood to ground area (AS:AG) or sapwood to leaf area area overlain with a deep organic peat layer dependent on stand age

(AS:AL) ratios, respectively, quantifies the influence of tree- and in the poorly-drained stands (volumetric soil moisture range forest-related parameters on transpiration rates (Whitehead 0.42–0.76 cm3 cm−3) (Bond-Lamberty et al. 2002b). Mean annual and Jarvis 1981, Vertessy et al. 1995, Ewers et al. 1999, precipitation and temperature were 439 mm and 0.8 °C, respec-

Hatton and Wu 1995). JS in flood-tolerant species such as tively, with mean January and July air temperatures of −19.7 and black spruce is often increased under saturated soil conditions 16.5 °C (Bond-Lamberty 2002a, Environment Canada 2009).

Tree Physiology Volume 32, 2012 Boreal forest tree transpiration 601

A wildfire chronosequence established in previous research (C. Jens. ex Russ.) C. Jens. in Tolf and S. capillifolium (Ehrh.) was utilized and spans four major stages of boreal succession Hedw.) (Bond-Lamberty and Gower 2006). All forested stands, omitting tree establishment: (i) 1989—maximum species both well- and poorly-drained, were burned by a stand- diversity, little paludification (18 years old from sensor installa- replacing fire during the years indicated. tion in 2007), (ii) 1964—maximum species diversity, exten- sive paludification (43 years old), (iii) 1930—black Micrometeorological measurements spruce-dominated, full canopy closure in the well-drained A total of eight stands were sampled, one well-drained and one stand and complete paludification in the poorly-drained stand poorly-drained stand in each of four stand ages from the chro- (77 years old), and (iv) 1850—black spruce-dominated, can- nosequence (Figure 1). Well- and poorly-drained zones were opy self-thinning in the well-drained stand (157 years old) determined visually noting the presence of pooled water, (Bond-Lamberty et al. 2002a, 2002b, 2002c, 2004a, 2004b, Sphagnum moss and Ledum palustre L. (Labrador tea) (both Wang et al. 2003a, 2003b). Mid-successional well-drained typical of poorly-drained bogs) or feather moss (typical of stands younger than 60 years consist of three dominant spe- well-drained forests), and changes in tree size and species Downloaded from cies, Picea mariana (Mill.) Britton, Sterns & Poggenb. (black types. Zones were then verified with measurements of soil spruce), Pinus banksiana Lamb. (jack pine) and Populus moisture and peat depth, where well-drained stands had highly tremuloides Michx. (trembling aspen), with forest-floor vegeta- variable soil moisture and shallower peat depths (5–15 cm) tion dominated by grasses (only in the youngest stands), than poorly-drained stands (20–50 cm peat depth). Plots were

feather moss (e.g., Pleurozium schreberi (Brid.), Ptilium crista- selected at a minimum distance of 50 m from the edge of http://treephys.oxfordjournals.org/ castrensis (Hedw.) De Not. and Hylocomium splendens (Hedw.) these drainage zones to ensure that data were only collected Schimp. In B.S.G.) and bare soil (Bond-Lamberty et al. 2004a, from the drainage condition of interest (Angstmann et al. 2006). Late successional well-drained stands are black spruce- accepted). Within each stand a meteorological tower was dominated with scattered trembling aspen in canopy gaps and installed to monitor environmental variables including vapor feather moss-dominated forest floor. Poorly-drained stands of pressure deficit D( ), air temperature (HMP45 Temperature and all ages, with a difference in elevation from well-drained stands Relative Humidity Probe, Campbell Scientific, Logan, UT, USA), of <5 m, are dominated by dwarfed black spruce and scattered photosynthetic photon flux density Q ( ) (LI190SB Quantum jack pine and tamarack (Salix spp. in the youngest poorly- Sensor, Campbell Scientific), wind speed (CSAT3 Sonic at University of Wyoming Libraries on June 27, 2016 drained stand) with the forest floor dominated by Sphagnum Anemometer, Campbell Scientific) and soil moisture (ML2x spp. hummocks and hollows (e.g., Sphagnum fuscum (Shimp.) ThetaProbe Soil Moisture Sensor, Delta-T Devices, Cambridge, Klinggr., S. riparium Ångstr., S. warnstorfii Russ., S. angustifolium UK). Atmospheric data from the 77-year-old stand were used

Figure 1. Diagram showing the sampling design of the eight stands sampled, four stand ages, each consisting of a well- and poorly-drained stand. Sizes of tree icons represent relative sizes of each species at each age and drainage condition. Triangle-topped icons represent black spruce, flat triangle-topped icons are jack pine and circle-topped icons show trembling aspen trees. Arrows indicate predicted magnitude of total tree transpiration for all species located at each stand.

Tree Physiology Online at http://www.treephys.oxfordjournals.org 602 Angstmann et al. for all analyses in this study because no significant differences dependent on stand age and stem diameter-at-breast-height, among stands were found except for soil moisture for which all but not soil drainage condition. Leaf phenology did not influ- site data were used (Barker et al. 2009). ence leaf area estimates using allometrics because no residu- Heat dissipation sensors, 20 mm in length (Granier 1987), als were found in response curves of transpiration to D or Q in were installed in 8–10 trees of each species if present in the time (Ewers et al. 2007). Transpiration expressed as transpira- −2 −1 stand and were chosen from the full range of diameter-at- tion per unit ground area (EC) (g m AG s ) and transpiration −2 −1 breast-height representing size classes from the smallest to per unit leaf area (EL) (g m AL s ) were calculated according the largest tree of each species. Heat dissipation sensors were to Pataki et al. (2000). consistently placed on the north side of the tree at breast height, and covered with aluminum casing to reduce thermal Relationship of transpiration to leaf water status gradients. Smaller trees, such as those located in poorly- Predawn and midday leaf water potentials were measured by drained and younger stands, required the use of 10 mm length excising leaves, placing them in a plastic bag and using the sensors (use of standard calibration justified by Braun and pressure chamber technique within 30 min of collection (Model Downloaded from Schmid 1999, McCulloh et al. 2007), foam insulation covered 600, PMS Instrument Co. Albany, OR, USA) (Scholander et al. with aluminum casing and installation at ground level where 1964, Hinckley et al. 1978). Leaf water potential was m easured the height of allometry measurements were taken. Preliminary every month of the growing season over 3 years for 3–5 trees measurements showed significant thermal effects (tempera- of each species at each stand (total of 101 samples each ture difference in unheated probes =0.78 (±0.78) °C) so month) and averaged by year for comparison with average http://treephys.oxfordjournals.org/ probes were moved to 0.5 m height and thoroughly insulated daily transpiration. Soil water potential can be approximated (temperature difference in unheated probes =0.08 (±0.03) °C) from predawn leaf water potential if trees are not experiencing (Goulden and Field 1994). Maximum temperature differences night-time transpiration or stem refill Scholander( et al. 1964, for use in the empirical equation of Granier (1987) were cho- Waring and Cleary 1967, Dawson et al. 2007). sen so that night-time transpiration and stem refill were accounted for, considering minimum temperature differences Estimating temporal drivers of transpiration over multiple days (Oishi et al. 2008). Previous studies at D and Q were partitioned into 0.2 kPa and 50 µmol m−2 s−1 these stands found no significant radial or circumferential bins, respectively, and EC was averaged by those bins and used at University of Wyoming Libraries on June 27, 2016 trends in the sapwood of black spruce, jack pine and trembling to quantify responses of transpiration to D and Q. Exponential aspen because sensors extended deeper than active sapwood response curves used the following equation: (Ewers et al. 2005); therefore, intra-tree scaling of sap flux ()1e−bx (1) was conducted using the methods outlined in Clearwater et al. ya=− (1999). Sap flux and meteorological data were collected every 30 s and averaged every 30 min using AM16/32 multiplexors where a represents the upper limit of the curve and b is the and CR1000 dataloggers (Campbell Scientific) Barker ( et al. slope. Curves were forced through zero following the assump- 2009). Gaps were present in the data when sap flux sensors tion that night-time transpiration is driven by D and confirmed failed or were shut off during battery drainage and rain days with modeled intercepts and average daily night-time values were eliminated from the data set because heat dissipation that were not significantly different from zero for any species sensors have poor signal-to-noise ratios during rainfall (Phillips or site condition (P < 0.0001); R2 values of curves were signifi- and Oren 1998). Daily sums of each tree were calculated by cantly higher than linear fits to the same data and produced adding up all 48 half-hourly values of transpiration on each day better residual patterns (data not shown). from 4.5 h to the next 4.0 h of the day to include potential water storage (Phillips and Oren 1998). Days that contained Determination of age, species and drainage condition missing data (i.e., <48 half-hourly values) were not used in transpiration differences calculations of daily average transpiration (~46%). We acknowledge pseudoreplication when using trees as independent units of replication in this study (Hurlbert 1984). Calculation of expressed daily transpiration However, the difficulty in locating and having resources to

Sapwood area (AS), leaf area index (L) and sapwood-to-leaf install sap flux sensors in multiple stands for adequate stand area ratio (AS:AL) were calculated from diameter-at-breast- age and drainage replications limits our ability to provide true height or diameter at 10 cm above the ground of each tree replication as seen in many other ecological studies. Moreover, using previously determined specific leaf area and allometry Bond-Lamberty et al. (2006) showed that the stands selected (Bond-Lamberty et al. 2002a, 2002b). Allometric equations in this study are representative of stands of similar age and and specific leaf area values were quantified in these previous drainage condition throughout the region sampled through studies for the same sites in the current study and were helicopter flights. In this study, variation among individual trees

Tree Physiology Volume 32, 2012 Boreal forest tree transpiration 603 is greater than between stands as has been found in other sap Table 1. Total number of trees measured (n), tree density, sapwood flux studies Ewers( et al. 1999); therefore, we are confident area (AS), leaf area index (L) and sapwood-to-leaf area ratio (AS:AL) measured in 2007 for stand age, drainage and species. that the analysis of variance (ANOVA) adequately explains differences between stand age and drainage and not just trees. n Tree AS L AS:AL 2 −1 2 −2 2 −2 Drainage, stand age and species differences of expressed density (m ha ) (m m ) (cm m ) (# ha−1) values of daily transpiration (i.e., EC and EL) were analyzed using the PROC MIXED function with repeated measures Black spruce Well drained in SAS (SAS 9.1.3 2005, SAS Institute Inc., Cary, NC, USA). 18 years old 93 24,166 1.51 0.178 2.74 Assumptions of normality and equal variances were tested in 43 years old 230 29,285 6.93 3.48 2.67 SAS resulting in log transformations of AS, AL, JS, EC and EL to 77 years old 103 36,429 17.5 6.46 2.95 meet the assumption of equal variances. A first-order autore- 157 years old 199 25,337 11.3 4.51 2.51 gressive moving average time autocorrelation function was Poorly drained used to model the error structure due to temporal autocorrela- 18 years old 107 27,803 1.62 0.632 2.68 43 years old 313 48,888 7.32 2.63 3.51 Downloaded from tion from repeated measures on the same trees. This autocor- 77 years old 96 12,223 8.16 2.49 3.96 relation function was determined through minimization of the 157 years old 61 21,574 7.40 1.51 4.90 Akaike information criterion after running the analysis through Trembling aspen various autocorrelation functions (Konishi and Kitagawa 1996). Well drained 18 years old 148 38,457 3.96 0.490 3.79 Tukey’s honestly significant difference post-hoc test of mean http://treephys.oxfordjournals.org/ 43 years old 22 2801 2.30 1.08 3.17 differences (α < 0.05) was used to determine significant dif- 77 years old 3 1061 0.036 0.083 0.54 ferences among treatment effects and interaction effects. Poorly drained To further support the use of ANOVA procedures on the 18 years old 109 28,323 0.947 0.269 3.93 continuous variable stand age, we also analyzed data using Jack pine linear regression analysis with the glm() function in the R Well drained 18 years old 4 1039 0.077 0.007 4.71 Statistical Package (R Development Core Team, Vienna, 43 years old 22 2801 3.66 0.611 5.22 Austria) and compared slopes of a one-parameter model (i.e., Poorly drained stand age) and a two-parameter model (stand age and drain- 18 years old 2 520 0.250 0.130 1.81 at University of Wyoming Libraries on June 27, 2016 age condition). Results from this analysis produced the same 43 years old 3 469 0.931 0.150 6.17 differences in stands and drainage conditions as the ANOVA Tamarack Poorly drained analysis and are not shown here because the power of these 43 years old 1 156 1.44 0.015 97.1 statistical tests was lower than that of the ANOVA approach 157 years old 7 2476 1.15 0.084 10.2 (i.e., n = 2 for stand age and n = 4 for drainage condition). Willow Data are shown as yearly averaged total daily transpiration for Well drained simplification and because monthly and daily averaged analy- 18 years old 39 10,134 0.027 0.003 2.73 Poorly drained ses resulted in comparable results. No significant differences 18 years old 47 12,213 0.807 0.098 7.27 were found between the 2007 and 2008 growing seasons in terms of stand age, species and drainage condition; therefore, both years are reported to increase the statistical power of Ewers et al. 2005). Relationships between drainage condition each analysis. Furthermore, boreal systems are composed and tree- and stand-related parameters depended on the age mainly of coniferous species with long leaf life spans and have of the stand. Well-drained stands—considering the dominant short growing seasons; therefore, leaf area does not change species black spruce only—had higher tree density, AS and L significantly for the seasonal time period studied (May through than poorly-drained stands but 24–78% lower AS:AL, except in August). the two youngest stands where the pattern was reversed (only

the 18-year-old stand for L and AS:AL) (Table 1). Age-controlled differences in black spruce stand parameters were present Results with well-drained stands showing an increase and subsequent decrease in (i) tree density at 77 years of age, (ii) AS and L at Tree- and stand-related characteristics among drainage 157 years and (iii) a 7% decrease in AS:AL across the chrono- condition sequence (Table 1). Poorly-drained stands showed a general

Stand tree density, sapwood area (AS), leaf area index (L) and decrease in tree density, an increase in AS and L with age until sapwood-to-leaf area ratio (AS:AL) measured during 2007 the oldest stand and an increase in AS:AL of ~77% over the retained values similar to those published for these same sites chronosequence (Table 1). When considering all species in the in previous studies (Table 1; Bond-Lamberty et al. 2002b, stand, well-drained stands had greater AS than poorly-drained

Tree Physiology Online at http://www.treephys.oxfordjournals.org 604 Angstmann et al.

stands over all ages, but the patterns of tree density, L and Table 2. Mean predawn and midday water potential (ΨL) (MPa) aver- A :A between drainage condition were dependent on stand aged for the 2007 and 2008 growing seasons. Values in parentheses S L are one standard deviation from the mean. Letters and asterisks repre- age. Generally, tree density and L were greater in the well- sent significant differences among stand age and drainage, respec- drained stands and AS:AL was ~10% greater in the poorly- tively. Differences were considered significant at aP value < 0.05. drained stands (any anomalies were in the youngest two n Predawn Ψ (-kPa) Midday Ψ (-kPa) stands). Age-controlled differences in these parameters for all L L species generally decreased with age in both the well- and the Black spruce poorly-drained stands except for A which showed an increase Well drained S 18 years old 20 0.53 (0.07)a 1.5 (0.10)a* for both drainage types and L which indicated an increase for 43 years old 25 0.41 (0.04)a 1.5 (0.06)a the well-drained stands only. 77 years old 25 0.55 (0.03)a 1.4 (0.10)a 157 years old 18 0.54 (0.05)a 1.6 (0.10)a Physiological controls of transpiration Poorly drained No significant differences in predawn and midday leaf water 18 years old 20 0.50 0.22)ab 1.9 (0.11)a 43 years old 25 0.55 (0.08)a 1.5 (0.09)b Downloaded from potential were found between leaves bagged on the tree and 77 years old 24 0.45 (0.04)b 1.4 (0.13)b non-bagged leaves (bagged: n = 147, F = 0.793, P = 0.375; 157 years old 18 0.43 (0.03)b 1.3 (0.11)b and non-bagged: n = 147, F = 2.2153, P = 0.139) or with Trembling aspen height in the canopy (low: n = 34, F = 2.15, P = 0.152; and Well drained 18 years old 20 0.37 (0.03)a 1.7 (0.16)a high: n = 34, F = 0.0469, P = 0.830), indicating differences in http://treephys.oxfordjournals.org/ 43 years old 25 0.33 (0.04)a 1.5 (0.07)ab leaf hydraulic resistances were minimal and allowing pooling of 77 years old 18 0.47 (0.03)b 1.3 (0.13)b all samples for each month of collection (Turner and Long Poorly drained 1980, Ryan and Yoder 1997). Night-time transpiration was 18 years old 20 0.32 (0.02) 1.7 (0.20) considered absent on days when water potential was mea- Jack pine sured because there were no significant differences between Well drained 18 years old 20 0.52 (0.03)a 1.1 (0.10)a bagged and non-bagged leaves and night-time D was <0.2 kPa. 43 years old 25 0.57 (0.04)a 1.5 (0.08)b Differences in transpiration rates among stand age and drain- Poorly drained age were not driven by predawn and midday ΨL as indicated by 18 years old 20 0.48 (0.03)a 1.0 (0.10)a at University of Wyoming Libraries on June 27, 2016 the lack of significant differences (Table 2). However, species 43 years old 15 0.50 (0.05)a 1.1 (0.12)a differences in transpiration were, at least partially, driven by predawn and midday ΨL differences (Table 2). D and Q were linearly correlated at all stands when Q is trembling aspen and jack pine were greater in the well-drained lagged with respect to D by 150 min (y = 0.286 + 2.947x, than in the poorly-drained stands (Figure 3a, b, e and f). In 2 R = 0.57, P < 0.0001). Preliminary analysis of black spruce contrast to our first prediction, maximum half-hourlyE C was up half-hourly JS, EC and EL regressed onto D and lagged Q also to 50% less in well- than in poorly-drained stands for 157-, showed significant relationships in all cases P( < 0.0001), with 43-, and 18-year-old black spruce stands (Figure 3a and b). less-biased residual patterns between scaled transpiration and Specifically, statistically significant differences between drain- D (R2 between 0.87 and 0.99) than with lagged Q (R2 between age conditions for model parameter a from Eq. (3)—represent-

0.66 and 0.99). Clear diurnal patterns of JS are present and ing the upper limit of EC—were present for black spruce in the show distinct peaks during high D and Q times (Figure 2). 43-year-old stand, trembling aspen in the 18-year-old stand

Well-drained black spruce had 42% lower midday JS than and 43- and 18-year-old stands for jack pine (P < 0.001). poorly-drained black spruce midday JS (Figure 2a and b), jack Parameter b, representing the rate of EC response to D and Q, pine midday JS was 20% lower in well-drained stands was significantly greater in the 43-year-old trembling aspen

(Figure 2d) and well-drained trembling aspen midday JS was relationship (Figure 3c). 33% higher than poorly-drained stands (Figure 2c). For parameter a, (i) black spruce in the 157-year-old well- Significant saturation responses (Eq. (3)) were present drained stand was significantly lower than the 43- and 77-year- between average half-hourly EC and binned D and Q for all old well-drained stands, (ii) black spruce in the 43-year-old ages, drainage and species. The maximum half-hourly EC stand was significantly greater than all other stands, (iii) ranged from (i) 0.006–0.018 and 0.004–0.029 g m−2 s−1 for 77-year-old black spruce was higher than the 157- and black spruce; (ii) 0.001–0.026 and 0.003 g m−2 s−1 for trem- 18-year-old stands, (iv) all stand ages were significantly differ- bling aspen; and (iii) 0.001–0.016 and 0.001–0.003 g m−2 s−1 ent for well-drained trembling aspen (P < 0.001, data not for jack pine in well- and poorly-drained stands, respectively shown), and (v) 43-year-old jack pine was greater than (Figure 3). As expected in our first prediction, maximum 18-year-old (Figure 3). Parameter b (i.e., slope) was only

half-hourly EC for 77-year-old black spruce and all ages of both significantly different in the 43-year-old stand for trembling

Tree Physiology Volume 32, 2012 Boreal forest tree transpiration 605

excess. This prediction was not supported in well-drained

black spruce which did not have significantly higher daily EC than poorly-drained trees except in the 77-year-old stand

where well-drained daily black spruce EC was ~50% higher than poorly-drained (Figure 4b). Well-drained trembling aspen and the 43-year-old well-drained jack pine had significantly

greater daily EC than poorly-drained stands as expected in our

first prediction (Figure4 d and e). Daily JS was higher and stand

AS lower in poorly-drained stands, suggesting that daily EC was

not primarily controlled by AS because EC was still higher in the majority of poorly-drained stands (data not shown; Table 1).

Prediction (ii), that daily EC will be greater whereas daily EL will be lower in older stands was only partially supported by Downloaded from the data. Stand age effects of black spruce leaf-level transpiration within each drainage condition suggest that younger

daily EL is greater than older stand EL in well- and poorly-drained conditions respectively, supporting prediction (ii) (Figure 4a).

This pattern is also seen in well-drained trembling aspen with http://treephys.oxfordjournals.org/ age; however, the opposite is found in both well- and poorly-

drained jack pine where older stands tend to have greater EL

(Figure 4c and e). The decrease in daily EL with age does not seem to be driven singularly by L, which shows variable or gradual decrease in well- and poorly-drained stands, respectively (Table 1). In contrast to prediction (ii), black spruce daily

Figure 2. Diurnal patterns of JS and environmental drivers D and Q for EC was lower in the older than in younger well- and poorly- a 5-day period during the 2007 growing season for various stand drained stands, except in the 18-year-old poorly-drained ages for (a) well-drained black spruce, (b) poorly-drained black at University of Wyoming Libraries on June 27, 2016 spruce, (c) well- and poorly-drained trembling aspen, (d) well- and stand (Figure 4b). This pattern was also seen for well-drained poorly-drained jack pine, and (e) D and Q. Well-drained stands are trembling aspen but not for well-drained jack pine where daily represented by black lines and poorly-drained stands by gray. EC increased with age as predicted (Figure 4d and f). The

changes in daily EC with age correspond with similar changes aspen, being steeper in slope than in other stand ages in AS in both well- and poorly-drained stands, suggesting that

(P < 0.001, data not shown). age differences in daily EC are driven primarily by AS (Figure 4c, Table 1). Resource allocation controls of transpiration Variability in stand total daily E Our first prediction inferred that daily EL would be greater in C the well-drained stands than in the poorly-drained stands as a Controls over differences in total stand daily EC between soil result of differences between AS and AL. Contradicting this pre- drainage conditions were site-specific with some sites being diction, poorly-drained black spruce in all stands had 15–80% more influenced by long-term resource allocation and others higher daily EL than well-drained stands, although the drainage by short-term physiological responses. In the youngest differences in the 77-year-old stand were not significant 18-year-old stand, higher daily total EC mirrored an increase in 2 −1 (Figure 4a). Further contradicting prediction (i), poorly-drained total stand AS from 2.8 to 5.6 m ha (primarily due to trem- jack pine had 34% greater daily EL than well-drained trees bling aspen) from poorly- to well-drained stands (Figure 5,

(Figure 4e). Supporting this prediction, trembling aspen had Table 1). We note that total daily EC was likely underestimated higher daily EL in the 18-year-old well-drained stand than in the in the 18-year-old well-drained stand because black spruce poorly-drained stand, although not significant (Figure 4c). transpiration was not measured due to small tree size. Addition

Daily JS was significantly greater in poorly-drained black spruce of black spruce measurements would make the non-significant and jack pine than in well-drained stands, suggesting that EL difference in total daily EC of the 18-year-old stand significant. was not controlled by L which was lower in poorly-drained Similarly, the 77-year-old well-drained stand had higher daily stands (data not shown; Table 1). total EC, reflecting a coincident increase in total stand AS 2 −1 2 −1 Prediction (i) also forecasted that average daily EC would be (i.e., 17.5 m ha as opposed to 8.2 m ha in the poorly- greater in well-drained stands than in poorly-drained stands drained stand) (Figure 5, Table 1). Alternatively, daily total EC in because of a decrease in tree size (i.e., AS) with soil moisture the 43- and 157-year-old stands was higher in the poorly-

Tree Physiology Online at http://www.treephys.oxfordjournals.org 606 Angstmann et al. Downloaded from http://treephys.oxfordjournals.org/ at University of Wyoming Libraries on June 27, 2016

Figure 3. Mean response of transpiration per unit ground area (EC) to D and Q for (a, e) well-drained black spruce, (b, f) poorly-drained black spruce, (c, g) trembling aspen and (d, h) jack pine. Symbols represent average transpiration for each stand age with closed and open symbols showing well- and poorly-drained stands, respectively. Standard error was less than 5% of the mean data value (e.g. standard error = well-drained black spruce ≤ 0.0001; poorly-drained black spruce ≤ 0.001; trembling aspen ≤ 0.0002; and jack pine ≤ 0.0004).

drained condition, opposite to what would be expected based soil conditions and poorly-drained total stand daily EC decreas- upon total stand AS data (Figure 5, Table 1). In these two ing between the 43- and 77-year-old stand followed by a dis- stands, total stand AS was up to 47% higher in the well-drained tinct increase in the oldest stand (Figure 5). Variability in than in the poorly-drained stands, suggesting that physiologi- species contribution to total daily EC is high in the two young- cal drivers of EC independent of resource allocation effects est stands where broad-leaved species contribute up to 50 may be driving these differences. and 85% of total transpiration in poorly- and well-drained

Age differences in total stand daily EC were present and conditions, respectively. Initial increases in total stand daily EC showed two distinct patterns depending on drainage condition occurred from the 18- to the 43-year-old stands for both with EC decreasing after the 43-year-old stand in well-drained well- and poorly-drained conditions due to species shifts

Tree Physiology Volume 32, 2012 Boreal forest tree transpiration 607 Downloaded from http://treephys.oxfordjournals.org/ at University of Wyoming Libraries on June 27, 2016

Figure 4. Transpiration per unit leaf area, EL, and transpiration per unit ground area, EC, for three species, black spruce, trembling aspen and jack pine. Letters above the figures represent significant differences among stand age for well- and poorly-drained sites individually. Asterisks represent significant differences between drainage conditions within one stand age. Error bars represent standard error around the mean value. P value ≤ 0.05.

−1 Figure 5. Average total daily transpiration, EC (mm day ) for all species at well-drained (WD) and poorly-drained (PD) stands.

Tree Physiology Online at http://www.treephys.oxfordjournals.org 608 Angstmann et al.

caused by succession as a result of addition of black spruce ally supported for EL but not EC. Lower EL in older stands is not into the chronosequence and removal of trembling aspen and surprising as many studies have found that photosynthetic jack pine biomass with canopy closure and/or paludification capacity decreases (Hom and Oechel 1983, Oleksyn et al. intensifies (Figure 5). In the well-drained stands, total stand 1997, Mediavilla and Escudero 2003) and stomatal sensitivity daily EC decreased from the 43- to the 77-year-old stand to soil and atmospheric drought increases with stand age through the removal of jack pine from the chronosequence as (Tretiach 1993, Kolb and Stone 2000, Mediavilla and Escudero trembling aspen and black spruce EC was not significantly dif- 2003). A decrease in EC in poorly-drained stands may be a ferent between the two ages (Figure 5). The final decrease in result of stomatal closure in response to increasing cold, anoxic total stand daily EC was controlled by both removal of aspen soil conditions occurring under paludification processes which from the successional chronosequence and age-induced tree may overshadow the influence of minor increases in AS with size differences in black spruce AS (Figure 5, Table 1). Changes age. Negative effects of paludification on productivity have described above for poorly-drained total stand daily EC were been previously described in northern forests (Simard et al. also controlled by AS (Figure 5, Table 1). 2007) and have more recently been attributed to soil tempera- Downloaded from ture as opposed to anoxia (Wolken et al. 2011). Age-related decline in stomatal conductance and transpiration have been Discussion documented in a variety of species including but not limited to

Our first prediction—that EL and EC would both be greater in Picea abies (L.) Karst. (Alsheimer et al. 1998), Eucalyptus reg- well- than in poorly-drained stands resulting from reductions in nans F. Muell. (Vertessy et al. 2001), Pinus pinaster Aiton http://treephys.oxfordjournals.org/

AS and L in anoxic poorly-drained soil conditions—was only (Delzon and Loustau 2005) and tropical deciduous and conif- partially supported by our data. The higher AS and L in well- erous species (Sobrado 1994). drained stands only controlled transpiration differences for We found that stand age differences in tree transpiration trembling aspen and jack pine, but did not seem to influence are primarily controlled by successional species changes black spruce transpiration which was still greater in the poorly- and secondarily by tree size parameters (i.e., AS and L) drained stand regardless of changes in tree size (Figure 4a whereas transpiration differences between soil drainage and b, Table 1). These results suggest that physiological conditions are controlled by short-term physiological responses of black spruce to D, Q and/or soil moisture responses and only by tree size parameters in certain spe- at University of Wyoming Libraries on June 27, 2016 (although not leaf water potential) may control drainage differ- cies such as trembling aspen. These findings provide com- ences in stand water fluxes more than resource allocation pat- pelling evidence that consideration of age and drainage terns. Greater transpiration rates resulting from a higher water heterogeneity in boreal forests will impact predictions of car- table have been confirmed in many other species including bon and water fluxes in the face of global change. We specu- Thuja plicata Donn ex D. Don (western red cedar) (Ewers et al. late that warmer, drier growing seasons will result in younger, 2007), Quercus alba L (white oak) Quercus michauxii Nutt. more diverse stands due to increased wildfire in the well- (swamp chestnut oak) (Parker 1950) and Taxodium distichum drained portions of the ecosystem resulting in higher stand- (L.) Rich. (bald cypress) (Pezeshki et al. 1996). Notwithstanding, level transpiration rates. Drainage of poorly-drained stands black spruce and other coniferous species may be negatively as a result of warming of the permafrost layer will likely influenced by prolonged flooding causing root dieback, nutri- increase total stand transpiration in all but the oldest stands, ent limitation and decreased stomatal conductance (Zaerr which are uncommon in this ecosystem. Furthermore, 1983, Lamhamedi and Bernier 1994, Islam et al. 2003), mak- increased transpiration in poorly-drained stands will create a ing the higher transpiration rates for black spruce seen in this positive feedback cycle of additional draining of soil mois- study an anomaly that is supported by energy-balance estima- ture, which may lead to increased wildfires, increased tran- tions for these stands (Barker et al. 2009). The unexpected spiration and photosynthesis, less paludification and a response of high black spruce transpiration rates to prolonged decrease in belowground carbon sequestration and storage flooding could be due to the formation of hypertrophied lenti- (Black and Bliss 1978, Bonan et al. 1992, Goulden et al. cels on submerged roots and stems (Kozlowski 1997), or the 1998, Heimann and Reichstein 2008, Ise et al. 2008). production of adventitious roots (LeBarron 1945, Kasischke Depending on the spatial scale of interest, increased fire fre- and Johnstone 2005). Trembling aspen and jack pine, on the quencies can lead to more homogeneous forest age struc- other hand, are moderately flood tolerant with aspen entering ture at regional scales, but increase stand-scale heterogeneity dormancy and jack pine experiencing decreased productivity by influencing species composition and paludification levels. in prolonged flooded conditions Dickman( et al. 2001, Apostol Alternatively, warmer, drier periods of permafrost melt, wet- and Zwaizek 2003). land drainage and subsequent increases in decomposition

Prediction (ii)—that older stands will have higher EC but may offset the uptake of CO2 (Davidson and Janssens 2006). lower EL resulting from decreasing AS:AL with age—was gener- Additional negative consequences on the water availability

Tree Physiology Volume 32, 2012 Boreal forest tree transpiration 609 for hydroelectric power and agricultural irrigation processes Angstmann, J.L., B.E. Ewers and H. Kwon. Testing transpiration controls will be seen with increasing transpiration under global by quantifying spatial variability along a boreal black spruce forest drainage gradient. Ecohydrol. Ecohydrology, accepted. change. Apostol, K.G. and J.J. Zwaizek. 2003. Hypoxia affects root sodium and chloride concentrations and alters water conductance in salt-treated jack pine (Pinus banksiana) seedlings. Trees 17:251–257. Conclusion Baldocchi, D.D. 1997. Measuring and modeling carbon dioxide and water vapour exchange over a temperate broad-leaved forest during This study determined rates and drivers of transpiration rates in the 1995 summer drought. Plant Cell Environ. 20:1108–1122. three dominant tree species in the central Canadian boreal for- Barker, C.A., B.D. Amiro, H. Kwon, B.E. Ewers and J.L. Angstmann. est and how the contribution of each species changes with 2009. Evapotranspiration in intermediate-aged and mature fens and drainage condition. Short-term physiological drivers such as upland black spruce boreal forests. Ecohydrology 2:462–471. Black, R.A. and I.C. Bliss. 1978. Recovery sequence of Picea mariana– vapor pressure deficit and soil moisture controlled differences Vaccinium uliginosum forests after burning near Inuvik, Northwest between well- and poorly-drained stands whereas successional Territories, Canada. Can. J. Bot. 56:2020–2030. species changes and tree size (i.e., sapwood and leaf area) Bonan, G.B. 2008. Forests and climate change: forcings, feedbacks, Downloaded from controlled changes in water use with stand age. Future work and the climate benefits of forests. Science 320:1444–1449. Bonan, G.B., D. Pollard and S.L. Thompson. 1992. Effects of boreal should focus on modifying current models of water and carbon forest vegetation on global climate. Nature 359:716–718. fluxes to include accurate representation of stand-, species- Bond-Lamberty, B. and S.T. Gower. 2006. Estimation of stand-level and drainage-heterogeneity by incorporating tree- and stand- leaf area for boreal bryophytes. Oecologia 151:584–592.

related characteristics (i.e., sapwood area interactions with Bond-Lamberty, B., C. Wang and S.T. Gower. 2002a. Aboveground and http://treephys.oxfordjournals.org/ transpiration) and site status (i.e., saturated vs. unsaturated soil belowground biomass and sapwood area allometric equations for six boreal tree species of northern Manitoba. Can. J. For. Res. conditions). Inclusion of these parameters in regional models 32:1441–1450. will increase predictive understanding of the effects of global Bond-Lamberty, B., C.K. Wang, S.T. Gower and J. Norman. 2002b. Leaf change on this region (Bond-Lamberty et al. 2009). area dynamics of a boreal black spruce fire chronosequence. Tree Physiol. 22:993–1001. Bond-Lamberty, B., C. Wang and S.T. Gower. 2002c. Annual carbon Acknowledgments flux from woody debris for a boreal black spruce fire chrono sequence. J. Geophys. Res. 107:8220–8230. Thank you to collaborators at the University of Manitoba, Bond-Lamberty, B., C. Wang and S.T. Gower. 2004a. Net primary pro- at University of Wyoming Libraries on June 27, 2016 Dr Brian Amiro and Corrine Barker, and at the University of duction and net ecosystem production of a boreal black spruce wildfire chronosequence. Glob. Change Biol. 10:473–487. Wisconsin, Dr Ben Bond-Lamberty for assistance with site Bond-Lamberty, B., C. Wang and S.T. Gower. 2004b. Contribution of setup and data collection. A special thanks to Bruce Holmes, root respiration to soil surface CO2 flux in a boreal black spruce Regional Forester, Manitoba Department of Natural Resources, chronosequence. Tree Physiol. 24:1387–1395. and the Nisichawayasihk Cree Nation for permitting assistance Bond-Lamberty, B., C. Wang and S.T. Gower. 2005. Spatiotemporal measurement and modeling of stand-level boreal forest soil tem- and use of land for research purposes. We thank anonymous peratures. Agric. For. Metereol. 131:27–40. reviews and the journal associate editor, Dr Nathan Phillips, for Bond-Lamberty, B., S.T. Gower, M.L. Goulden and A. McMillan. 2006. comments that greatly improved the manuscript. Simulation of boreal black spruce chronosequences: comparison to field measurements and model evaluation. J. Geophys. Res. 111:G02014, doi: 10.1029/2005JG000123. Funding Bond-Lamberty, B., S.D. Peckham, S.T. Gower and B.E. Ewers. 2009. Effects of fire on regional evapotranspiration in the central Canadian Funding provided by a National Science Foundation Grant boreal forest. Glob. Change Biol. 15:1242–1254. DEB0515957. Braun, P. and J. Schmid. 1999. Sap flow measurements in grapevines (Vitis vinifera L.) 2. Granier measurements. Plant Soil 215:47–55. Brodribb, T.J., N.M. Holbrook, M.A. Zwieniecki and B. Palma. 2005. Conflict of interest Leaf hydraulic capacity in ferns, conifers, and angiosperms: impacts on photosynthetic maxima. New Phytol. 165:839–846. None declared. Cˇermák, J., J. Úlehla, J. Kucˇera and M. Penka. 1982. Sap flow rate and transpiration dynamics in the full-grown oak (Quercus robus L.) in floodplain forest exposed to seasonal floods as related to potential References evapotranspiration and tree dimensions. Biol. Plant. 24:446–460. Clearwater, M.J., F.C. Meinzer, J.L. Andrade, G. Goldstein and Akeroyd, M.D., S.D. Tyerman, G.R. Walker and I.D. Jolly. 1998. Impact N.M. Holbrook. 1999. Potential errors in measurement of nonuniform of flooding on the water use of semi-arid riparian eucalypts. J. Hydrol. sap flow using heat dissipation probes. Tree Physiol. 19:681–687. 206:10 4 –117. Cowan, I.R. and G.D. Farquhar. 1977. Stomatal function in relation to leaf Alsheimer, M., B. Köstner, E. Falge and J.D. Tenhunen. 1998. Temporal metabolism and environment. Symp. Soc. Exp. Biol. 31:471–505. and spatial variation in transpiration of Norway spruce stands within Davidson, E.A. and I.A. Janssens. 2006. Temperature sensitivity of soil a forested catchment of the Fichtelgebirge, Germany. Ann. For. Sci. carbon decomposition and feedbacks to climate change. Nature 55:103–123. 440:165–173.

Tree Physiology Online at http://www.treephys.oxfordjournals.org 610 Angstmann et al.

Dawson, T.E. 1996. Determining water use by trees and forests from of black spruce (Picea mariana) found in interior Alaska. Can. J. For. isotopic, energy balance, and transpiration analyses: the roles of Res. 13:834–839. tree size and hydraulic lift. Tree Physiol. 16:263–272. Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological Dawson, T.E., S.S.O. Burgess, K.P. Tu, R.S. Oliveria, L.S. Santiago, field experiments. Ecol. Monogr. 54:187–211. J.B. Fisher, K.A. Simonin and A.R. Ambrose. 2007. Nighttime transpi- Ise, T., A.L. Dunn, S.C. Wofsy and P.R. Moorcroft. 2008. High sensitiv- ration in woody plants from contrasting ecosystems. Tree Physiol. ity of peat decomposition to climate change through water-table 27:561–575. feedback. Nat. Geosci. 1:763–766. Delzon, S. and D. Loustau. 2005. Age-related decline in stand water Islam, M.A. and S.E. MacDonald. 2004. Ecophysiological adaptations use: sap flow and transpiration in a pine forest chronosequence. of black spruce (Picea mariana) and tamrack (Larix laricina) seed- Agric. For. Metereol. 129:105–119. lings to flooding. Trees 18:35–42. Dickman, D.I., J.G. Isebrands, J. Eckenwalder and J. Richardson. 2001. Islam, M.A., S.E. MacDonald and J.J. Zwaizek. 2003. Responses of Poplar culture in North America. National Research Council Research black spruce (Picea mariana) and tamarack (Larix laricina) to flood- Press, Ottawa, 397 p. ing and ethylene. Tree Physiol. 23:545–552. Drushka, K. 2003. Canada’s forests: a history. Forest History Society Jackson, R.B., E.G. Jobbágy, R. Avissar, et al. 2005. Trading water for Issues Series, McGill-Queen’s University Press, Montreal, 150 p. carbon with biological carbon sequestration. Science Environment Canada. 2009. National Climate Data and Information 310:1944–1947. Downloaded from Archive. Government of Canada, 11 November 2008. http://climate. Kasischke, E.S. and J.F. Johnstone. 2005. Variation in postfire organic weatheroffice.ec.gc.ca. layer thickness in a black spruce forest complex in interior Alaska Ewers, B.E., R. Oren, T.J. Albaugh and P.M. Dougherty. 1999. Carry- and its effects on soil temperature and moisture. Can. J. For. Res. over effects of water supply on water use of Pinus taeda. Ecol. Appl. 35:2164–2177. 9:513–525. Kolb, T.E. and J.E. Stone. 2000. Differences in leaf gas exchange and

Ewers, B.E., D.S. Mackay, S.T. Gower, D.E. Ahl, S.N. Burrows and water relations among species and tree sizes in an Arizona pine-oak http://treephys.oxfordjournals.org/ S.S. Samanta. 2002. Tree species effects on stand transpiration in forest. Tree Physiol. 20:1–12. northern Wisconsin. Water Resour. Res. 38:1103–1114. Konishi, S. and G. Kitagawa. 1996. Generalised information criteria in Ewers, B.E., S.T. Gower, B. Bond-Lamberty and C.K. Wang. 2005. model selection. Biometrika 83:875–890. Effects of stand age and tree species on canopy transpiration and Kozlowski, T.T. 1997. Responses of woody plants to flooding and salin- average stomatal conductance of boreal forests. Plant Cell Environ. ity. Tree Physiol. Monogr. 1:1–29. 28:660–678. Kreuzwieser, J., S. Fuˇrniss and H. Rennenberg. 2002. Impact of water- Ewers, B.E., D.S. Mackay and S. Samanta. 2007. Interannual consis- logging on the N metabolism of flood-tolerant and non-tolerant tree tency in canopy stomatal conductance control of leaf water potential species. Plant Cell Environ. 25:1039–1049. across seven species. Tree Physiol. 27:11–24. Lamhamedi, M.S. and P.Y. Bernier. 1994. Ecophysiology and field per- Fougnies, L., S. Renciot, F. Muller, et al. 2007. Arbuscular mycorrhizal formance of black spruce (Picea mariana): a review. Ann. Sci. For. colonization and nodulation improve flooding tolerance in 51:529–551. at University of Wyoming Libraries on June 27, 2016 Pterocarpus officinalis Jacq. seedlings. Mycorrhiza 17:159–166. Lavoie, M., D. Paré, N. Fenton, A. Groot and K. Taylor. 2005. Franks, P.J. 2004. Stomatal control and hydraulic conductance, with Paludification and management of forested peatlands in Canada: a special reference to tall trees. Tree Physiol. 24:865–878. literature review. Environ. Rev. 13:21–50. Franks, P.J., I.R. Cowan and G.D. Farquhar. 1997. The apparent feedfor- LeBarron, R.K. 1945. Adjustment of black spruce root systems to ward response of stomata to air vapour pressure deficit: information increasing depth of peat. Ecology 26:309–311. revealed by difference experimental procedures with two rainforest Manabe, S. and R.T. Wetherald. 1980. On the distribution of climate

trees. Plant Cell Environ. 20:142–145. change resulting from an increase in CO2 content of the atmosphere. Gorham, E. 1991. Northern peatlands: role in the carbon cycle and J. Atmos. Sci. 37:99–118. probable responses to climatic warming. Ecol. Appl. 1:182–195. McCulloh, K.A., K. Winter, F.C. Meinzer, M. Garcia, J. Aranda and Goulden, M.L. and C.B. Field. 1994. Three methods for monitoring the B. Lachenbruch. 2007. A comparison of daily water use estimates gas exchange of individual tree canopies: ventilated-chamber, sap- derived from constant-heat sap-flow probe values and gravimetric flow and Penman-Monteith measurements on evergreen oaks. measurements in pot-grown saplings. Tree Physiol. Funct. Ecol. 8:125–135. 27:1355 –1360. Goulden, M.L., S.C. Wofsy, J.W. Harden, et al. 1998. Sensitivity of McDowell, N., H. Barnard, B. Bond, et al. 2002. The relationship boreal forest carbon balance to soil thaw. Science 279:214–218. between tree height and leaf area: sapwood area ratio. Oecologia Granier, A. 1987. Evaluation of transpiration in a Douglas-fir stand by 132:12–20. means of sap flow measurements. Tree Physiol. 3:309–320. Mediavilla, S. and A. Escudero. 2003. Photosynthetic capacity, inte- Grant, R.F. 2004. Modeling topographic effects on net ecosystem pro- grated over the lifetime of a leaf, is predicted to be independent of ductivity of boreal black spruce forests. Tree Physiol. 24:1–18. leaf longevity in some tree species. New Phytol. 159:203–211. Hatton, T.J. and H.-I. Wu. 1995. Scaling theory to extrapolate individual Meinzer, F.C. and D.A. Grantz. 1991. Coordination of stomatal, hydrau- tree water use to stand water use. Hydrol. Proc. 9:527–540. lic, and canopy boundary-layer properties—do stomata balance Heimann, M. and M. Reichstein. 2008. Terrestrial ecosystem carbon conductances by measuring transpiration? Physiol. Plant. dynamics and climate feedbacks. Nature 451:289–292. 83:324–329. Herrera, A., W. Tezara, E. Rengifo and S. Flores. 2008. Changes with Meinzer, F.C., B.J. Bond, J.M. Warren, and D.R. Woodruff. 2005. Does seasonal flooding in sap flow of the tropical flood-tolerant tree spe- water transport scale universally with tree size? Funct. Ecol. cies, Campsiandra laurifolia. Trees 22:551–558. 19:558–565. Hinckley, T.M., J.P. Lassoie and S.W. Running. 1978. Temporal and spa- Mimikou, M.A. and E.A. Baltas. 1997. Climate change impacts on the tial variations in the water status of forest trees. For. Sci. reliability of hydroelectric energy production. Hydrol. Sci. J. 20:a0001–z0001(1). 42:661–678. Hom, J.L. and W.C. Oechel. 1983. The photosynthetic capacity, nutrient Monteith, J.L. 1995. A reinterpretation of stomatal responses to humid- content and nutrient use efficiency of different needle age-classes ity. Plant Cell Environ. 18:357–364.

Tree Physiology Volume 32, 2012 Boreal forest tree transpiration 611

Moore, G.W., B.J. Bond, J.A. Jones, N. Phillips and F.C. Meinzer. 2004. Schuur, E.A.G., J. Bockheim, J.G. Canadell, et al. 2008. Vulnerability of Structural and compositional controls on transpiration in 40- and permafrost carbon to climate change: implications for the global car- 450-year-old riparian forests in western Oregon, USA. Tree Physiol. bon cycle. Bioscience 58:701–714. 24:481–491. Simard, M., N. Lecomte, Y. Bergeron, P.Y. Bernier and D. Paré. 2007. Mott, K.A. and D.A. Parkhurst. 1991. Stomatal response to humidity in Forest productivity decline caused by successional paludification of air and helox. Plant Cell Environ. 14:509–515. boreal soils. Ecol. Appl. 17:1619–1637. Oishi, A.C., R. Oren and P.C. Stoy. 2008. Estimating components of Sobrado, M.A. 1994. Leaf age effects on photosynthetic rate, transpi- forest evapotranspiration: a footprint approach for scaling sap flux ration rate and nitrogen content in a tropical dry forest. Physiol. measurements. Agric. For. Meteorol. 148:1719–1732. Plant. 90:210–215. Oleksyn, J., M.G. Tjoelker, G. Lorenc-Plucin´ska, A. Konwin´ska, Stocks, B.J., M.A. Fosberg, T.T. Lynham, et al. 1998. Climate change ´ R. Zytkowiak, P. Karolewski and P. Reich. 1997. Needle CO2 exchange, and forest fire potential in Russian and Canadian boreal forests. structure and defense traits in relation to needle age in Pinus held- Clim. Change 38:1–13. reichii Christ—a relict of Tertiary flora. Trees 12:82–89. Tretiach, M. 1993. Photosynthesis and transpiration of evergreen Oren, R., N. Phillips, B.E. Ewers, D.E. Pataki and J.P. Megonigal. 1999. Mediterranean and deciduous trees in an ecotone during a growth Sap-flux scaled transpiration responses to light, vapor pressure defi- season. Acta Oecol. 14:341–360. cit, and leaf area reduction in a flooded Taxodium distichum forest. Turner, N.C. and M.J. Long. 1980. Errors arising from rapid water loss Downloaded from Tree Physiol. 19:337–347. in the measurement of leaf water potential by the pressure chamber Parker, J. 1950. The effects of flooding on the transpiration and sur- technique. Aust. J. Plant Physiol. 7:527–537. vival of some southeastern forest tree species. Plant Physiol. Vartapetian, B.B. and M.B. Jackson. 1997. Plant adaptations to anaero- 25:453–460. bic stress. Ann. Bot. 79(Suppl. A):3–20. Pataki, D.E., R. Oren and W.K. Smith. 2000. Sap flux of co-occurring Vertessy, R.A., R.G. Benyon, S.K. O’Sullivan and P.R. Gribben. 1995.

species in a western subalpine forest during seasonal soil drought. Relationships between stem diameter, sapwood area, leaf area and http://treephys.oxfordjournals.org/ Ecology 81:2557–2566. transpiration in a young mountain ash forest. Tree Physiol. Pezeshki, S.R., J.H. Pardue and R.D. DeLaune. 1996. Leaf gas exchange 15:559–567. and growth of flood tolerant and flood-sensitive tree species under Vertessy, R.A., F.G.R. Watson and S.K. O’Sullivan. 2001. Factors deter- low soil redox conditions. Tree Physiol. 16:453–458. mining relations between stand age and catchment water balance in Phillips, N. and R. Oren. 1998. A comparison of daily representations mountain ash forests. For. Ecol. Manage. 143:13–26. of canopy conductance based on two conditional time-averaging Wang, C., B. Bond-Lamberty and S.T. Gower. 2003a. Carbon distribu- methods and the dependence of daily conductance on environmen- tion of a well- and poorly drained black spruce fire chronosequence. tal factors. Ann. For. Sci. 55:217–235. Glob. Change Biol. 9:1066–1079.

Roy, V., P. Bernier, A.P. Plamondon and J. Ruel. 1999. Effect of drainage Wang, C., B. Bond-Lamberty and S.T. Gower. 2003b. Soil surface CO2 and microtopography in forested wetlands on the microenvironment flux in a boreal black spruce fire chronosequence. J. Geophys. Res. and growth of planted black spruce seedlings. Can. J. For. Res. 107:8224–8231. at University of Wyoming Libraries on June 27, 2016 29:563–574. Waring, R.H. and B.D. Cleary. 1967. Plant moisture stress: evaluation Ryan, M.G. and B.J. Yoder. 1997. Hydraulic limits to tree height and tree by pressure bomb. Science 155:1248–1254. growth. BioScience 47:235–242. Whitehead, D. and P.G. Jarvis. 1981. Coniferous forests and planta- Ryan, M.G., B.J. Bond, B.E. Law, R.M. Hubbard, D. Woodruff, E. Cienciala tions. In Water Deficits and Plant Growth. Ed. T.T. Kozlowski. and J. Kucera. 2000. Transpiration and whole-tree conductance Academic Press, New York, NY, pp 49–152. in ponderosa pine trees of different heights. Oecologia Wolken, J.M., S.M. Landhäusser, V.J. Lieffers and U. Silins. 2011. 124:553–560. Seedling growth and water use of boreal conifers across different Scholander, P.F., H.T. Hammel, E.A. Hemmingsen and E.D. Bradstreet. temperatures and near-flooded soil conditions. Can. J. For. Res. 1964. Hydrostatic pressure and osmotic potential in leaves of 41:2292–2300. mangroves and some other plants. Proc. Natl Acad. Sci. USA Zaerr, J.B. 1983. Short-term flooding and net photosynthesis in seed- 52:119 –125. lings of three conifers. For. Sci. 29:71–78.

Tree Physiology Online at http://www.treephys.oxfordjournals.org