Advertisement

Plant and Soil

, Volume 439, Issue 1–2, pp 487–504 | Cite as

Seasonal patterns of water uptake in Populus tremuloides and Picea glauca on a boreal reclamation site is species specific and modulated by capping soil depth and slope position

  • Morgane MerlinEmail author
  • Simon M. Landhäusser
Regular Article
  • 207 Downloads

Abstract

Aims

Soil water availability is important for tree growth and varies with topographic position and soil depth. We aim to understand how two co-occurring tree species with distinct rooting and physiological characteristics respond to those two variables during two climatically distinct growing seasons.

Methods

Growing season (May to September) sap and transpiration fluxes were monitored using heat ratio method sap flow sensors on Populus tremuloides and Picea glauca in 2014 and 2015 growing along a hillslope with two different soil cover depths providing different rooting spaces.

Results

Across the two growing seasons, a shallow rooting space was the main factor limiting aspen’s leaf area and cumulative sap flux, whereas responses of white spruce were more limited by topographical position. Generally, sap and transpiration fluxes decreased with the season; however, a large precipitation event during the 2015 summer triggered a significant recovery in sap and transpiration fluxes in white spruce, while in aspen this response was more muted.

Conclusions

The two species distinct rooting and physiological characteristics produced contrasting water uptake and water use dynamics in response to rooting space, soil water availability and climate, inviting a more detailed exploration of sap flux and its interactions with climatic and edaphic variables.

Keywords

Sap flow Rooting depth Precipitation events Soil water availability Mine reclamation 

Introduction

Rooting space characteristics are critical components influencing tree and forest establishment, survival and growth across biomes. Spatially, rooting space (defined as the soil volume accessible to roots) determines the extent and expandability of a root system and controls the pool of root-accessible resources. Tree species across biomes have developed different rooting strategies to cope with rooting space limitations including belowground competition with other individuals and/or species (Brédoire et al. 2016; Schenk and Jackson 2002), potentially leading to segregation of rooting zones. While rooting space can be limited by inter and intraspecific competition for resources, it can also be spatially limited by physical and chemical barriers (Bakker et al. 2006; Sakai et al. 2007). These barriers can restrict growth (Matthes-Sears and Larson 1999), reproduction potential (Boland et al. 2000; Schaffer et al. 1999), and modify growth allocation strategies among plant organs (i.e. leaves, shoots/trunks, and roots) (Bockstette et al. 2017; Jackson et al. 1996). However, rooting space is not only defined spatially, but also by the rooting substrate’s quality, in terms of resource availability, aeration levels and soil biological communities. Quality of the rooting substrate is a decisive factor in determining what can be broadly defined as a site’s carrying capacity, i.e. the plant community characteristics and success at which soil resource consumption is equal to resource supply for a given climatic condition, soil texture and management program (Xia and Shao 2009), even when rooting space quantity is limited.

Soil resource availability is driven by several factors regionally and seasonally, including topography, climate, and disturbance regime. Variability of these factors may alleviate restrictive growth conditions imposed by physical limitations in rooting space. Among the soil resources critical for tree growth, water availability is fundamental, and highly dependent on rooting space characteristics. Soil depth, composition and texture affect water retention and thus available water holding capacity (Huang et al. 2011, 2015). Topographic position is associated with both vertical and lateral water movement, with run-on and run-off processes, thereby altering available soil water content. These factors, together with the temporal and seasonal dynamics of soil water availability, are decisive in biomes such as the boreal forest where the growing season is climatically limited to a very short period (Devito et al. 2012) and growing space can be limited by shallow soils or conditions that limit deep rooting such as low soil temperature (Bonan 1992; Van Cleve and Yarie 1986).

Boreal tree species have adapted to these soil conditions by developing different rooting strategies. Trembling aspen (Populus tremuloides Michx.) and white spruce (Picea glauca (Moench) Voss) are two common boreal forest species in the western Canadian boreal region, which co-occur in the boreal mixedwood forest region on mesic upland sites (Alberta Parks 2015; Cogbill 1985; Nlungu-Kweta et al. 2014) and their relationship is often considered facilitative. Mixtures of both species have shown higher aboveground productivity than either pure stands (MacPherson et al. 2001), as well as a larger root system for spruce (Lawrence et al. 2012). The root systems of aspen and spruce have distinct characteristics suggesting potential adaptations to limitations in the quantity and quality of rooting space and could result in an avoidance of belowground competition (Novoplansky 2009). Trembling aspen has an extensive lateral root system, spanning from several meters to tens of meters (Snedden 2013) that is supported by deep sinker roots that can, depending on soil conditions, well exceed 200 cm in depth (Blake et al. 1996; Snedden 2013; Strong and LaRoi 1983). In contrast white spruce has a shallow root system, often limited to the first 40 cm of soil (Kalliokoski et al. 2008; Strong and LaRoi 1983). These divergent root system characteristics could contribute to species-specific responses to reduced soil moisture availability and drought tolerance (Bladon et al. 2006).

Addressing the effects of rooting space and water availability on tree success in natural stands is however challenging. Specifically, the complex suite of interactions and legacies related to soil development and stand history complicate the interpretation of responses. Restoration of the forested ecosystems after open pit (surface) mining offers a unique opportunity to investigate these processes without potentially confounding interactions and legacy effects. Restoration of mixedwood stands in the boreal forest region of Alberta after surface mining requires the reconstruction of an entire soil profile and revegetation (Government of Alberta 2017). The reclamation sites are often located on landforms that are constructed with soil materials (overburden) deemed unsuitable for plant growth (Lilles et al. 2012), potentially limiting root development. Therefore, appropriate soil covers are constructed to protect the root zone from limiting conditions such as elevated salinity and sodicity from the underlying overburden material (Huang et al. 2015; Jung et al. 2014). The soil cover type and thickness provide the rooting space and resources necessary to sustain forests, especially considering the limitations of the overburden, the reconstructed soil covers, and the climatic conditions. However, since there are tradeoffs between ensuring appropriate rooting space for establishing and maintaining forest and the costs associated with the construction of the soil covers (also referred to hereafter as caps) on the landscape, there is an added incentive to optimize capping thickness for forest regrowth and sustainability.

This study aimed to gain insight in the effects of rooting space and moisture stress linked with a topographical gradient on the aboveground characteristics and water uptake of trembling aspen and white spruce trees on a 15-year-old reclamation site in northern Alberta, Canada. Specifically, this study sought to answer the following questions: How do rooting space and soil moisture availability affect the rooting depth, leaf area and water uptake and use of aspen and spruce? Do soil capping depth and topography alter water uptake response to seasonal variations in soil water availability and are there differences between the two species? We hypothesized that trembling aspen will be most affected by soil capping depth due to its generally deeper-exploring root system than white spruce, restricting both aspen tree size and water uptake in shallow caps. Furthermore, we hypothesized that white spruce will be more physiologically sensitive to seasonally dry periods during the growing season and in the drier topographical positions due to its rather shallow root system.

Materials and methods

Research area

The research area – South Bison Hills (SBH) study site – is located north of Fort McMurray, Alberta, within Syncrude Canada Ltd.’s Base Open Pit Mine lease. SBH is a landform that was constructed using saline-sodic clay shale overburden material, excavated during the mine establishment, contoured into a shallow sloped hill (15% slope, ~60 m elevation). The overburden material is highly saline, categorizing this substrate as non-suitable for plant growth (Barbour et al. 2007; Fung and Macyk 2000; Kessler et al. 2010). In 1999, the overburden structure was capped with salvaged subsoil and topsoil materials which are considered suitable for plant growth; first, a layer of salvaged subsoil that consisted of glacial till material, capped with a layer of topsoil material that consisted of a mixture of salvaged peat and mineral soil. The wilting points for vegetation (−1.5 MPa) on the three materials used on site are: ~20 cm3water cm−3soil for the peat mineral mixture, ~18 cm3 cm−3 for the glacial till subsoil, and ~20 cm3 cm−3 for the shale overburden (Huang et al. 2015). In the research area, soil caps were constructed with three different thicknesses: a shallow (35 cm cap composed of 20 cm subsoil and 15 cm topsoil), an intermediate (50 cm depth composed of 30 cm subsoil and 20 cm topsoil), and a thick cap (100 cm cap with 80 cm subsoil and 20 cm topsoil). In 2003, the research area was broadcast fertilized (10–30–15-4 N-P-K-S at a rate of 350 kg fertilizer ha−1 (Lanoue 2003)) and seeded with barley (Hordeum spp.) at a density of 250 seeds m−2 for erosion control. In 2004 one-year-old trembling aspen (Populus tremuloides Michx.) and white spruce (Picea glauca (Moench) Voss) seedlings were planted on all capping treatments, each at a density of 800 stems ha−1.

The climate of the north-eastern region of Alberta is categorized as sub-humid, cycling through wet and dry periods over both seasonal and longer term time periods (Devito et al. 2012). Temperature climate normals (1981–2010) range from – 17.4°C in January to +17.1°C in July (Environment Canada weather station – Fort McMurray airport, 56° 39’ N and 111° 13’ W, (Environment Canada 2013)), reflecting the region’s cold winters and warm summers, with a short growing season (late May to early September). Total annual precipitation averages 418.6 mm, with 287.3 mm falling during the growing season, between May and September (Environment Canada 2013). Despite having a mean annual precipitation similar to the grassland ecoregion further south, the timing of precipitation and the lower temperatures that reduce evaporation provide more available moisture during the growing season.

Study sites and measurements

We explored the effects of rooting space on water uptake and growth on trees growing in the two more extreme soil capping thicknesses – the shallow 35 cm and thick 100 cm caps. Since the reclamation area was located on a large north-facing hill slope and we expected to see higher soil water availability in the lower topographic position, and vice versa, we selected trees in the upper and lower slope sections (Fig. 1). At each slope position, three plots were established, and within these, three trees of both species (trembling aspen and white spruce) were selected. Plots were located 10 to 15 m from each other in the slope direction to ensure independence of tree response but similar edaphic conditions. Tree selection in the plots was done following a biased tree selection process, selecting one dominant tree and two healthy neighboring trees (within a 5 m radius) for each species. In total, 18 aspen trees and 18 spruce trees (9 each for each topographical position) were selected for both the 2014 and 2015 measurement years on the 35 cm cap. On the 100 cm cap, nine trees of each species were selected only on the lower slope position in 2014, and only on the upper slope position in 2015. Comparing the tree selection with mensuration data collected along the whole hillslope revealed that the selection of trees for this study was representative of the average trees growing in the respective treatments and topographical positions (data not shown). Aboveground characteristics of the selected trees – diameter at breast height (DBH), height, and total leaf area (destructively sampled for a subset of the selected trees, see below) – were measured at the end of the growing season for both years. Yearly basal area increments (BAI) were calculated for each felled tree between 2011 and 2015. The selected trees (44 trees each year) were equipped with heat ratio method (HRM) sap flow sensors in both years between May and September recording at 10 min intervals. The HRM sap flow technique was chosen due to its ability to detect and quantify low flows more reliably than other techniques (Burgess et al. 2001). Description of the technique is available in Burgess et al. (2001). SFM1 sap flow meters (ICT International Pty Ltd.) were used in this study. Obvious deformities in the stem were avoided when installing the sensors. In 2014, the probes were installed at breast height (1.3 m) (representing approximately 38% of the spruce crown, and under the full aspen crown) and in 2015 sensors were installed in all species below the crown (20 cm height for white spruce, 1.3 m for aspen). All sensors within each plot (six sensors – one per tree) were daisy chained to each other and connected to an external 12 V deep cycle marine battery with protective case, charged by a 22 W solar panel with BX metal sheathed wire. Since this is a very sensitive system, malfunctions of sensors, solar panels, or batteries did occur and led to some gaps in the continuous data (see sap flow data processing below). Application, set-up and limitations of the system have been described previously (Bleby et al. 2004; Burgess and Bleby 2006; Dawson et al. 2007; Doronila and Forster 2015; Forster 2017; Looker et al. 2016; Pfautsch et al. 2011).
Fig. 1

Overview of the study site showing the two capping treatments (35 cm and 100 cm). The soil moisture station (black star) with the TDR (time domain reflectance) sensors was located mid-slope. Each treatment was divided into a lower slope and an upper slope section. In each section, three tree plots were established. The inset shows an example of the plot layout, with the white spruce and trembling aspen trees selected for the sap flow sensor installation, and the transect used for establishing the root profile

To assess general vegetation rooting patterns (depths and quantity) as well as both species’ maximum rooting depth, soil cores to a depth of 120 cm were collected in October 2015 for each capping thickness. The soil cores were taken along a 2 m long transect between an aspen and a white spruce tree in each plot (Fig. 1): one core was taken 15 cm from the base of both the aspen and spruce tree and a third core was taken half-way between the two trees, roughly 1 m equidistance. The 120 cm deep soil cores (diameter 5 cm) were then divided into 10 cm sections to build a detailed root mass profile. Live roots were picked from each 10 cm increment, with no distinction between tree and understory vegetation species. The root mass profile obtained from the soil cores thus included all roots, from all vegetation on site. To determine maximum rooting depth of trembling aspen and white spruce, total genomic DNA was extracted from dried ground root material in each 10 cm section using CTAB extraction protocol using species-specific primers for aspen and spruce previously developed (see Supplementary Information A). Further refinement on the resolution of fragment length for aspen using FLAP was employed to separate aspen from willow (Salix spp.) and balsam poplar (Populus balsamifera).

Soil moisture content was continuously monitored using Time-Domain Reflectance (TDR) sensors from O’Kane consultants Inc. for Syncrude Ltd. installed mid-slope on the 100 cm cap at 5 and 20 cm depths (Fig. 1). In addition, we monitored soil moisture content between June 9th and September 11th, 2015 in each of the three plots in the 35 cm and 100 cm caps and in the different slope positions at 20 cm depth. Daily precipitation data for 2014 and 2015 were obtained from a weather station installed by Syncrude Ltd. at the top of the slope on the 100 cm cap. Patterns of precipitation and temperature fluctuations between 2014 and 2015 led to strikingly different seasonal dynamics of soil moisture content in the upper 20 cm of the soil profile (Fig. 2). Cumulative precipitation during the growing season amounted to 271.8 mm in 2014 and 238.5 mm in 2015, respectively approximately 15 and 50 mm lower than the 1981–2010 normal. To further explore precipitation as a driver of sap flux, both growing seasons were divided into an initial wet period during spring and early summer, followed by a dry-down period and finally a dry period using piecewise segmented regression to identify the change-points. Throughout both growing seasons, soil moisture content at 5 and 20 cm did not go below the vegetation wilting point of peat mineral mix of 22% (Ojekanmi and Chang 2014, Fig. 2). In 2015, however, a prominent precipitation event (5 days with a total of 57 mm of precipitation) during the otherwise dry period led to a significant recharge of the upper soil layers.
Fig. 2

Soil moisture content (%, black line) measured at 5 cm (dotted line) and 20 cm (solid line) in the soil profile in 2014 (panel a) and 2015 (panel b), combined with daily total precipitation (black bars). The different periods: wet, dry-down, dry and rewet are indicated by shading of grey: dark grey, medium grey, light grey and medium-light grey respectively. Data from the 20 cm depth prior to mid-June 2015 was unavailable. The date at which trembling aspen flushed in 2015 is indicated in a vertical dotted line with the label “Flush POTR”

Sap flux data processing

In September 2015, two monitored aspen and one spruce tree were selected in each plot (total of 18 aspen trees and 9 spruce trees). These trees were cut and detailed measurements of bark thickness, sapwood and heartwood thickness, estimates of sapwood dry weight, sapwood volume and wound diameter from the sensors were made according to Burgess et al. (2001) and ICT International guidelines (SFM1 Sap Flow Meter Manual v4.02014). Thermal diffusivity properties of the sapwood were calculated following Marshall (1958) and Burgess et al. (2001). Raw heat velocities were corrected for wood thermal diffusivity and wounding diameter using individual measurements for the felled trees and their average for the trees left standing. For the following analyses and for each sensor, the thermistor depth with the larger magnitude of velocity – i.e. the most “active” depth – was chosen to avoid redundancy (both thermistors depths were highly correlated, R2 > 0.9) and less functioning sapwood depths (Link et al. 2014). Raw heat pulse velocity was converted to sap flux density for each 10 min interval (Jp, m3 m−2sapwood h−1) using the software Sap Flow Tool (v1.4.1. ICT International Pty Ltd., Armidale, NSW, Australia). Sapwood area for the un-cut trees was estimated using the existing linear correlation between measured sapwood area and DBH on the cut trees (for spruce: sapwood area = 14.26 × DBH − 68.66; for aspen: sapwood area = 11.93 × DBH − 53.44, with DBH in cm and R2 > 0.9 for both species). To reduce variability, the sap flux density data was aggregated into hourly intervals. Inspection of the sap flux density data showed significant gaps of individual sensors in both 2014 and 2015. The following data-filling procedure was applied to the 2014 and 2015 datasets independently. A multivariate imputation via chained equation procedure, implemented in the mice package in R (van Buuren and Groothuis-Oudshoorn 2011) was used, assuming the data were missing at random. Zero sap flow was assumed after leaf fall for aspen in 2014 (29th September 2014 – 15th October 2014) and before leaf flush in 2015 (26th April 2015 – 1st May 2015). Zero sap flow for white spruce was assumed at night (22:00:00 to 04:00:00) during the fall (29th September 2014 – 15th October 2014, 1st September 2015 – 10th September 2015). An offset based on average sap flux density during these zero sap flow periods was implemented. Daily sap flux (Q, m3 d−1) was then calculated, as well as cumulative sap flux (Qc, m3) between June 1st and August 31st of both years. Daily transpiration flux (T, dm3 m−2transpiring leaf area d−1) and cumulative transpiration flux (Tc, dm3 m−2transpiring leaf area) were calculated using the measurements of transpiring leaf area above the sap flow sensor from the cut trees and their estimates based on the DBH of the uncut trees. When taking into account needle shape and stomata distribution, transpiring leaf area of a white spruce is approximately 3.08 times the projected leaf area (Sellin 2000).

To assess the differences in sap flow dynamics over time, we used normalized sap fluxes to compare temporal dynamics between sensors with different absolute magnitudes. We normalized the time series of daily sap (Qn, d−1) and transpiration (Tn, d−1) fluxes by their respective maximum value (99.5th percentile daily integral to avoid normalizing by an outlier) (Link et al. 2014). All terms related to sap fluxes and transpiration fluxes are defined following nomenclature recommendations from Lemeur et al. (2008).

Statistical analyses

All data analyses were done using the R statistical software v3.5.1. (R Development Core Team 2018). For all models, transformations were done to meet the normality assumptions and variance was allowed to change along a variable to meet the homogeneity of variance assumption when necessary. Initial models contained variables interactions when appropriate and following a model selection procedure using AIC (Aikaike Information Criterion) the interactions were removed when not significant. Pairwise Tukey post-hoc tests were done to assess the differences between variables levels using the emmeans package (Lenth 2018). When standard errors are presented, the abbreviation “se” is used. All linear mixed effects models were done using the nlme package (Pinheiro et al. 2018). Proportional root mass was calculated as the proportion represented by the root mass in one 10 cm increment of a soil core over the whole soil core. To decrease heteroscedasticity in the model, the proportional root mass data was split into two datasets: 0–50 cm and 60–120 cm, and analyzed for differences between depths, capping treatments and slope positions using beta-regression from the betareg package (Cribari-Neto and Zeileis 2010). Tree characteristics were analyzed for each species separately using linear mixed effects models with the plot as a random effect. Slope position and year effects were analyzed only for the trees growing on the 35 cm cap. Capping treatment effects were analyzed separately only for trees on the lower slope in 2014 and only for trees on the upper slope in 2015. Diameter at breast height, transpiring leaf area, height and sapwood area were analyzed following this procedure. Kruskal-Wallis non-parametric tests were used on the study’s TDR data to assess differences in seasonal average of soil moisture content between slope positions and capping treatments.

Cumulative sap and transpiration fluxes between June 1st and August 31st of each year were analyzed using mixed effects linear models following the same procedure as for the tree characteristics models. Additionally, species and year differences on the 35 cm cap on cumulative sap and transpiration fluxes were tested.

To evaluate the impact of the slope position and capping treatment on the daily sap flux dynamics during a growing season, the analysis was separated between the different moisture-defined periods. Average daily sap and transpiration fluxes were analyzed for each species separately using linear mixed effects model following the same procedure as for the tree characteristics models. The rate of daily sap flux decline during the growing season was analyzed using generalized additive models (gam) with the mgcv package (Wood 2017) for each species and year separately to evaluate the impact of slope position and capping treatment. Normalized daily sap flux data was used for this analysis to eliminate the differences in absolute magnitudes between trees. The initial model was written as follows:
$$ gam\left( Sap\ flow\sim Variable+s\left( Date, by= Variable,k=k\right)+s\left( Tree, bs={}^{"}{re}^{"}\right)+s\left( Date, Tree, bs={}^{"}{re}^{"}\right), method={}^{"}{REML}^{"}\right) $$

With Variable = {Slope position; Capping treatment}, k = {6,10} to avoid over-fitting, Tree: the individual trees identification and “re” = random effect. The residual maximum likelihood method (REML) was used here instead of the generalized cross-validation (GCV) to avoid under-smoothing and penalize overfitting (Wood 2011). Following a model selection procedure using the AIC and deviance criterion with a Chi-square test, the models were simplified and the linear effect of Variable or interaction with the Date smooth function was removed, as well as the individual smooth term for Tree when necessary. Final models are presented in Supplementary Information B, Table B1. Confidence intervals (95%) were calculated and used to determine the dates during which the two Variable levels were significantly different. Finally, the impact of the late summer precipitation event on daily sap flux was assessed for the 2015 growing season using two methods. First, the ratio of average daily sap flux during the 2 days with peak soil moisture content at 5 and 20 cm depths to the average daily sap flux for the 2 days immediately before the precipitation event was calculated for each tree. Linear mixed effects models assessing the impact of slope position, species and capping thickness were run following similar procedure as previous linear mixed effects models. Second, the collective contribution of this precipitation period (cumulative rewetting that happened between July 12th and August 3rd, 2015) to the cumulative sap and transpiration fluxes over the dry period of the growing season was explored further. To do so, a “fake” dataset was constructed, which created sap and transpiration fluxes data that were representative of the same rewetting period, but with no precipitation event occurring. For that, daily sap and transpiration fluxes for each tree during that period were replaced with data for the same tree using the daily sap and transpiration fluxes 10 days before (July 1st– July 11th) and 11 days after (August 4th – August 15th) the precipitation event. Cumulative sap and transpiration fluxes were calculated for both the real 2015 dataset and the “fake” 2015 dataset between May 17th and August 31st, 2015. The difference (Drewetting) and ratio (Rrewetting) between the two were calculated and analyzed the same way as for the first method for both Qc and Tc.

Results

Tree and rooting characteristics in response to capping thickness and slope position

The selected aspen and spruce trees responded differently in their aboveground characteristics to capping thickness and slope position (Table 1). Trembling aspen trees were larger (p = 0.015) and had twice the leaf area (p < 0.01) when growing on the 100 cm cap than when growing on the 35 cm cap, but only at the lower slope position. Aspen did not show a response to slope position on the 35 cm cap in the measured aboveground variables. In contrast, white spruce trees had a smaller DBH on the 100 cm cap than on the 35 cm cap (p = 0.016) and 23% less leaf area (p = 0.016) at the lower slope position, however there were no differences between capping thicknesses in the upper slope position (Table 1). Spruce responded to slope position on the 35 cm cap, where trees at the lower slope position had a higher DBH (p = 0.019) and a 37% greater leaf area (p = 0.029, Table 1) than trees at the upper slope position.
Table 1

Estimated marginal means (standard error) for trembling aspen (P. tremuloides) and white spruce (P. glauca) trees equipped with the sap flow sensors for the following characteristics: diameter at breast height (DBH, cm), transpiring leaf area (equivalent to projected leaf area for trembling aspen, 3.08 × projected leaf area for white spruce, m2), height (m), sapwood area (cm2) and maximum rooting depth (cm)

 

35 cm cap

Lower slope

Upper slope

Lower slope

Upper slope

35 cm cap

100 cm cap

35 cm cap

100 cm cap

P. tremuloides

  DBH (cm)

7.64 (0.35) a

6.89 (0.32) a

7.16 (0.48) y

8.98 (0.43) x

7.28 (0.57) α

8.66 (0.57) α+

  Transpiring leaf area (m2)

8.37 (1.23) a

7.22 (1.10) a

6.49 (1.56) y

13.92 (1.37) x

8.76 (1.63) α

11.62 (2.17) α

  Height (m)

8.51 (0.28) a

8.01 (0.29) a

8.75 (0.35) x

9.28 (0.33) x

7.86 (0.40) α

8.97 (0.40) α+

  Sapwood area (cm2)

42.85 (1.69) a

37.53 (1.75) b

47.68 (2.19) x

42.11 (2.04) x

48.19 (3.43) β

57.10 (3.43) α

  Maximum rooting depth (cm)

93.3 (8.82)

103.3 (3.33)

NA

100.0 (10.0)

NA

105.0 (5.0)

P. glauca

  DBH (cm)

9.71 (0.34) a

7.76 (0.37) b

9.2 (0.31) x

8.01 (0.31) y

7.81 (0.60) α

8.44 (0.60) α

  Transpiring leaf area (m2)

83.94 (4.61) a

61.30 (4.99) b

78.94 (3.85) x

64.07 (3.85) y

61.91 (8.59) α

71.91 (8.59) α

  Height (m)

6.58 (0.26) a

5.66 (0.28) a

6.58 (0.28) x

6.53 (0.28) x

5.47 (0.37) α

5.89 (0.37) α

  Sapwood area (cm2)

107.2 (7.0) a

84.6 (7.0) a

105.36 (10.58) x

115.14 (10.58) x

81.34 (3.31) α

82.94 (3.31) α

  Maximum rooting depth (cm)

93.3 (3.3)

93.3 (3.3)

NA

93.3 (8.8)

NA

100.0 (5.8)

Results for sapwood area on the 35 cm cap for white spruce represent the 2015 data only, as sensors were located at breast height in 2014 (see methods). Statistical models were run separately on the 35 cm cap to evaluate the impact of slope position, and results are represented with the letters {a,b}, on the lower slope in 2014 and upper slope in 2015 to evaluate the impact of the capping thickness, using letters from the two sets {x,y} and {α, β}. Different letters represent a statistically significant difference (p < 0.05), and a + or - sign represents a marginally significant difference (p < 0.1). NA: not available

Trembling aspen’s sapwood area was positively correlated with leaf area (marginal R2 > 0.85 for all models, slope = 3.4 10−4). Sapwood area was larger on the 100 cm cap than on the 35 cm cap at the upper slope only (p = 0.041, Table 1). Aspen trees on the upper slope position on the 35 cm cap had a marginally larger sapwood area than trees growing at the lower slope position (p = 0.094, Table 1). Contrastingly, sapwood area of white spruce was positively correlated with transpiring leaf area in 2015 only at the upper slope position across capping thicknesses (slope = 1.1 10−4) and marginally so across slope positions on the 35 cm cap (slope = 0.5 10−4). The absence of a relationship between sapwood area and transpiring leaf area in white spruce in 2014 may be due to installing the sensors at breast height thus only capturing a fraction of the crown (approximately 38% of the crown). Capping thickness and slope position did not affect sapwood area in spruce (Table 1).

Yearly basal area increments declined from 2011 to 2015 across the site for both species. The decline was particularly pronounced in 2014 and 2015. Trembling aspen’s BAI was marginally higher (p < 0.1) on the 100 cm cap than on the 35 cm cap on the upper slope, and similar across years between the upper and lower slope positions on the 35 cm cap. Similarly to aspen, BAI of white spruce was marginally higher throughout the years on the 100 cm cap compared to the 35 cm cap on the upper slope position (Supplementary Information C Fig. C1 and Fig. C2). White spruce in the upper slope position had consistently smaller BAI from 2013 to 2015 compared to the lower slope position (p < 0.05) on the 35 cm cap.

Both trembling aspen and white spruce roots were found at or below the 100 cm depth in both capping treatments and therefore penetrated well into the overburden layer, indicating no differences between the capping treatments or the species in occupying the rooting space (Table 1). Total root mass (not separated by species) showed no differences between the capping treatments and slope positions (results not shown). However, differences in proportional root mass distribution between capping thicknesses were visible particularly at the upper slope position (Fig. 3). In the upper 20 cm soil layer, proportional root mass was significantly higher on the 35 cm cap compared to the 100 cm cap while in the 50–70 cm soil layers, the opposite was found (p < 0.05). While this response was evident in the upper slope position, the root mass distribution was much more uniform at the lower slope positions for both caps (Fig. 3).
Fig. 3

Proportional root mass distribution in 10 cm increment layers in the soil profile between 0 and 120 cm, expressed as a percentage of total root mass, for the lower slope (left panel) and upper slope (right panel), on both caps (grey: 35 cm and black: 100 cm). Each point represents the proportional root mass averaged across plots and soil cores, with 95% confidence intervals. Statistical differences from beta-regression on the data split into two datasets (0–50 cm and 60–120 cm) are as follows: grey labels represent post-hoc comparisons for which the 35 cm cap has a higher proportional root mass, black labels represent post-hoc comparisons for which the 100 cm cap has a higher proportional root mass. Labels: “***”: p < 0.001, “**”: p < 0.01, “*”: p < 0.05

Soil moisture and cumulative sap flux in response to capping thickness and slope position

Soil water content at 20 cm depth during the 2015 growing season differed between the two capping treatments and the associated slope positions. The 100 cm cap was marginally wetter than the 35 cm cap at the upper slope position (upper slope; 35 cm = 14.5%, 100 cm = 15.3%, (se = 0.6%), Kruskal-Wallis non-parametric test p < 0.06). On the 35 cm cap, the lower slope position had a significantly higher soil moisture content than the upper slope position (35 cm cap; lower slope = 17.4% (se = 0.9%), upper slope = 14.5% (se = 0.6%), Kruskal-Wallis non-parametric test p < 0.001).

Cumulative sap flux (Qc) in trembling aspen was on average 0.90 m3 (± 0.65 m3) on the 35 cm cap, and 1.59 m3 (± 0.91 m3) on the 100 cm cap across both growing seasons. Qc on the 100 cm cap compared to the 35 cm cap was twice as high (p < 0.05) on the lower slope position and only marginally higher in the upper slope position (Fig. 4, Supplementary Information C, Table C1). No differences between slope positions on the 35 cm cap were found for Qc for trembling aspen (Fig. 4, Supplementary Information C, Table C2). Cumulative transpiration fluxes (Tc) in aspen were similar between capping thicknesses and slope positions in both 2014 and 2015; however Tc was higher in 2015 than in 2014 at the lower slope position but only on the 35 cm cap (Supplementary Information C, Table C1 and Table C2).
Fig. 4

Estimated marginal mean and standard error for individual tree cumulative sap flux (Qc, m3) estimated between June 1st and August 31st of each year (2014 and 2015) for P. tremuloides (left) and P. glauca (right) for the 35 cm (circle) and 100 cm (triangle) capping treatments at each slope position (lower slope: black, upper slope: grey). Outliers were excluded from the statistical analysis. Following the analysis described in Materials and Methods, for each species, the results of three statistical models are presented: the effect of slope position within the 35 cm capping treatment (in black), the effect of capping treatment within the lower slope position in 2014 (in black) and upper slope position in 2015 (in grey). Results from the statistical analysis are presented here with the following labels, ns: non-significant, “+”: p < 0.1, “*”: p < 0.05, “***”: p < 0.001

On the 35 cm cap, cumulative sap flux (Qc) of white spruce was 2.05 m3 (± 0.83 m3) in 2014 and 0.91 m3 (± 0.67 m3) in 2015, highlighting a significant drop in Qc between years (p < 0.001). This is despite the 2014 data reflecting only the cumulative sap flux of a partial crown of spruce (measured at breast height and representing on average 38% of total crown area) compared to a full crown measurement (measured at basal height) in 2015. Cumulative sap flux of spruce on the 100 cm cap was on average 1.58 m3 (±1.19 m3) across both years. Qc was not affected by capping thickness at the lower slope; however, at the upper slope Qc of spruce was marginally higher on the 100 cm cap compared to the 35 cm cap (p < 0.07, Fig. 4). On the 35 cm cap, Qc in spruce was significantly lower on the upper slope than on the lower slope for both years (p < 0.05); on average 58% lower (Fig. 4, Supplementary Information C, Table C2). Cumulative transpiration flux in spruce was not different between capping thicknesses or slope positions (Supplementary Information C, Table C1). Overall, Qc of white spruce on the 35 cm cap in 2014 was approximately twice that of trembling aspen (p < 0.001), but both were similar in 2015 (Supplementary Information C, Table C2).

Seasonal sap flux dynamics

Daily sap Q and transpiration T fluxes in trembling aspen were higher in trees growing on the 100 cm cap than in trees on the 35 cm cap, but this difference was only observed at the upper slope position. During the initial wet spring conditions, Q and T of both species were not different between slope positions on the 35 cm cap (Supplementary Information C, Table C3); however, Q was higher in 2014 compared to 2015 (p < 0.07 for white spruce, p < 0.001 for aspen).

While daily sap and transpiration fluxes were relatively constant during the wet spring, they continuously declined during the dry-down and dry soil moisture periods in both species. Normalization of the daily fluxes eliminated the time-invariant differences in absolute magnitudes in fluxes between trees. Daily normalized transpiration fluxes (Tn) declined at different rates between the two capping thicknesses, but these differences were not consistent between years and species. By the end of the dry season of each year, Tn of each species were similar between the two soil caps (Supplementary Information C, Fig. C3). Daily normalized transpiration fluxes declined at different rates between the two slope positions, indicating a faster decline at the upper slope in both aspen and spruce compared to the lower slope position on the 35 cm cap in 2014 (Fig. 5). In 2015, there were small differences in the decline of Tn between slope positions for white spruce, however trembling aspen displayed a similar response to the one observed in 2014 (Fig. 5). For both years and species, Tn of trees in the lower slope position was higher than for trees on the upper slope position at the end of the dry season, albeit not significantly.
Fig. 5

Average normalized daily transpiration flux Tn (dotted line) during the dry-down and dry soil moisture periods and 95% confidence intervals for each slope position (lower slope: black, upper slope: grey) on the 35 cm cap for both species (left: P. tremuloides and right: P. glauca) for 2014 (top panels) and 2015 (bottom panels). Results from the generalized additive models assessing the effect of slope for each species and year independently are presented in solid lines with the respective slope colors. Time periods with a significant difference between the models fit for each slope position are indicated with vertical dotted black lines and a “*” label (p < 0.05)

The five-days precipitation event in 2015 (July 12th to July 16th) led to a significant increase in soil moisture content during the dry period (Fig. 2). During peak soil moisture content, daily sap flux (Q) of trembling aspen increased on average by 150% compared to immediately before the precipitation period on both the 35 and 100 cm caps. On the 35 cm cap, aspen trees in the upper slope position had a larger absolute increase in daily sap flux than the trees growing in the lower slope position (p < 0.01, Supplementary Information C, Fig. C4). This increase in Q translated into an increase of in Qc of 50 dm3 on average or 7% (Fig. 6) compared to a situation where the dry period had continued. Daily and cumulative transpiration fluxes showed a similar response, as cumulative transpiration increased by 4.5 dm3 m−2transpiring leaf area (results not shown). Aspen growing on the 100 cm cap marginally benefitted from this rewetting event (p < 0.1, upper slope position only). White spruce showed a much more pronounced response to this precipitation event on both caps, as Q increased on average by 350% compared to the time immediately before the precipitation event. On the 35 cm cap, spruce trees growing in the upper slope position had a larger increase in Q than trees growing in a lower slope position (p < 0.01, Supplementary Information C, Fig. C4). This rewetting event contributed to an increase in Qc by on average 170 dm3 compared to a situation where the dry period had continued. This increased uptake was significantly lower for the lower slope position, representing an increase in water uptake of 22% compared to 45% for the upper slope position (Fig. 6). Daily and cumulative transpiration fluxes showed a similar response, and cumulative transpiration increased by 1.7 (lower slope) to 2.4 (upper slope) dm3 m−2transpiring leaf area (results not shown). Spruce growing on the 100 cm cap showed a smaller response to this precipitation event than when growing on the 35 cm cap (p < 0.05, upper slope position only).
Fig. 6

Estimated marginal mean and 95% confidence interval of the ratio of the rewetting event contribution to the whole tree Qc during the dry-down and dry periods of the 2015 growing season (6th of June 2015 to 31st of August) to the hypothetical Qc during the dry-down and dry periods of 2015 with no rewetting event (Rrewetting, see the methods section for description of the calculations). Results are presented for each species, slope position (lower slope: black, upper slope: grey) and capping thickness (only for the upper slope, grey circles: 35 cm cap, grey triangles: 100 cm cap) for both species P. tremuloides and P. glauca. Significance (p < 0.05) is indicated by different letters for both models, marginal significance (p < 0.1) is indicated by a “+” sign

Discussion

Water use and growth of trembling aspen was limited by a reduction of rooting space (i.e. cover soil capping thickness) on this restoration site, while spruce was more sensitive to the topographical position along the hillslope. This confirms our hypothesis that based on the root system architecture, growth of trembling aspen would be most affected by soil capping depth, while white spruce would be more sensitive to the topographical position. Aspen trees had as much as twice the leaf area and cumulative sap flux on the thicker cap (100 cm), attaining levels similar to natural sites values (Hogg et al. 2000), compared to trees growing on the thinner cap (35 cm); however, transpiration per unit leaf area was similar. This suggests that, beyond soil water availability, other factors such as nutrient supply may have limited aboveground growth and leaf area development in our study in aspen on the 35 cm capping treatment. A cover soil cap of 35 cm led to a greater proportion of roots found in the upper layers, and limited the vertical extension of root systems into the overburden material. Commonly, aspen roots extend vertically well below 100 cm (Strong and LaRoi 1983) and preferentially proliferate in resource-rich soil areas (Bauhus and Messier 1999) while avoiding competitors roots (Messier et al. 2009; Mundell et al. 2007). Although aspen roots were found beyond the first 35 cm layer in the overburden layer, its high salinity and electrical conductivity is known to restrict root growth (Lazorko and Van Rees 2012). Therefore, the resources available for trees at 100 cm depth in the overburden soil profile on the 35 cm cap could have been limited compared to the same depths on the 100 cm cap. This could have resulted in the reduced growth observed over the last 4 years (2011–2015) (see BAI results) on the thinner cap compared to the thicker cap.

Despite differences in Qc between the two capping thicknesses for aspen, cumulative transpiration flux (i.e. Qc per unit of leaf area; Tc) was similar, regardless of tree size. Thus, differences in water uptake in trembling aspen across the site appear to be regulated by leaf area rather than by differences in physiological responses to climatic drivers. Indeed, leaf transpiration in aspen (and by extension sap flux) have been shown to be closely coupled with climatic conditions (Bladon et al. 2006; Blanken 1997; Hogg and Hurdle 1997). When climatic conditions are relatively similar across sampled trees, such as in this study, the development and extent of leaf area thus becomes crucial in determining the bulk of the growing season’s water uptake capacity regardless of capping treatment (see also Tie et al. 2017 for leaf area control on water uptake). Spring conditions (e.g. water availability and air temperature) rather than internal root signaling have been shown to drive the early development of the bulk of leaf area in aspen (DeByle and Winokur 1985; Delbart et al. 2008; Huang et al. 2010; Pollard 1970). This has an impact on aspen’s ability and potential to respond to climatic variability later in the growing season. The development of a large leaf area during a wet early growing season is beneficial in terms of increased photosynthetic yield and ultimately growth and reserve status; however, this holds true only if climatic conditions stay favorable. If water limiting conditions occur later in the season, an adjustment of the transpiration fluxes for this large leaf area may become necessary. Such an adjustment in aspen was notable when comparing trees between 2014 and 2015 located on the lower slope position on the 35 cm cap where, after leaf area differences were considered, Tc was lower in 2015 than in 2014. This potentially reflects a higher physiological sensitivity to drought of aspen trees growing in more favorable and wet conditions (i.e. lower slope position) compared to the more drought-prone trees at the upper slope position, which may have developed greater drought-resistance and avoidance strategies (see Kelln et al. 2008). This is further supported by the higher daily normalized transpiration fluxes early in the dry season of aspen trees growing on the upper slope position in the 35 cm cap compared to those growing in the lower slope position. Ultimately, when down regulation of the transpiration fluxes at the leaf level reaches the physiological limits of aspen, leaf area adjustment during the growing season (i.e. leaf abscission) may be required to prevent major drought damage or mortality. The environmental controls of leaf area expansion and coincidental xylem formation in aspen during spring conditions are therefore important periods in determining (among other factors) the short term (within a growing season) and long term resilience of aspen to climatic variability and ongoing climate change.

Beyond our initial hypotheses, we expected to find potentially some differences in aboveground growth and water uptake and use in aspen between slope positions particularly for the 35 cm cap. However, we detected no differences in aspen’s growth, Qc and Tc between the slope positions. Lateral root length was not measured in our study; however the ability of aspen to produce long lateral surface roots often spanning dozens of meters in length (DeByle and Winokur 1985) likely resulted in the formation of asymmetrical root systems leading downslope (Snedden 2013). These long feeder roots may have enabled aspen trees at upper slope positions to locate, access, and utilize moisture-rich areas further downslope.

The more shallow-rooted white spruce strongly responded to topographical position on the 35 cm cap. Spruce trees on the upper slope position were smaller in size and had less leaf area than trees growing on the lower slope position. This response might be driven by increased water stress, which may have been enough to generally reduce spruce growth over the years as reflected in the reduced BAI (also see Grossnickle 2000; Wiley et al. 2018). The sap flux measurements partly corroborated the results of the aboveground characteristics of spruce. Drier soil moisture conditions on the upper slope on the shallow cap contributed to a reduction of cumulative sap flux by 30% and by close to 60% during the dry year of 2015 when compared to trees at a lower slope position. While tree size differences partially contributed to this difference, daily transpiration fluxes were also lower on the upper slope, most probably caused by lower water availability, which limited daily transpiration flux under high evaporative demand. Similar observations have been made in spruce and other conifers, showing a reduction in water uptake in periods of low soil water availability (Ježík et al. 2015; Nadezhdina et al. 2007; Schuster et al. 2016). Correspondingly, the drier 2015 growing season had also a large impact on water use of white spruce at both slope positions on the shallow cap: Qc dropped between 50% (lower slope) and 70% (upper slope) compared to the previous year. This drop is most likely an underestimation, as in 2014 the sap flow sensors were installed at breast height (i.e. missing about 62% of the total leaf area of the crown). This excluded the water demand of the portion of the crown that was below the sensor, which represents a significant fraction of the total tree water uptake (Herzog et al. 1998).

Early stomatal closure could have contributed to the observed accelerated decline of sap flux in spruce in response to slope positions and climate (2014 versus 2015) on the 35 cm cap, ultimately lowering Qc and Tc. Soil water deficit and sustained high vapor pressure deficit conditions during the growing season have been shown to lead to early stomatal closure in spruce (Grossnickle 2000; Leo et al. 2014; Zweifel et al. 2002). Furthermore, Střelcová et al. (2013) demonstrated that as soil moisture gradually decreases, the dependency of transpiration in spruce on the atmospheric evaporative demand decreased, leading to a significant reduction of transpiration fluxes in spruce under soil drought. Further, it can be speculated that fine root mortality during the drier soil conditions of the surface layer might have played a role in the additional sap flux decline of spruce. It has been shown that soil moisture stress can lead to fine root mortality (Brunner et al. 2015; Gaul et al. 2008), which can result in a negative feedback loop of decreased root hydraulic conductivity and associated tree hydraulic decline (Cuneo et al. 2016). All these factors could have translated into reduced carbon acquisition and radial growth for spruce trees supported by the decreased BAI under water-stressed conditions.

Seasonal sap flux dynamics and final cumulative water uptake of both species were affected by seasonal climatic variability, particularly during the dry period of the growing season. As climatic variability is thought to increase with climate change (IPCC 2013), large but stochastic precipitation events may become more vital for trees and their water needs. A species’ adaptation to these changing conditions would be important, as these events would not only provide short-term relief from drought stress, but could also be important for longer-term soil moisture recharge. A large rewetting event in the late dry period of 2015 was a unique opportunity to investigate the trees responses and their ability to take advantage of the temporarily improved conditions. White spruce greatly benefitted from the rewetting period, as the large increase in its daily sap and transpiration fluxes in response to the precipitation event contributed to an increase of Tc and Qc of 33% compared to if it had continued to be dry. This increase was particularly pronounced on the upper slope position of the 35 cm cap, where this recovery represented an increase of approximately 45% of the cumulative water uptake and use, compared to an increase of 7% for aspen. Most likely, the relatively shallow root system of white spruce contributed to the rapid uptake of rain water from the shallow soil layers. Rapid increase of transpiration fluxes after re-watering during a drought has been reported previously for spruce species (Roberts and Dumbroff 1986), and is consistent with the low drought-resistance of the spruce trees (Blake and Li 2003). In a drier future with changed precipitation patterns, boreal forest trees might rely heavily on those large precipitation events for relieving drought stress (Børja et al. 2016). Interestingly, while aspen appears to be less adapted to these sporadic precipitation events, it might allow for greater soil moisture recharge at the stand level, due to reduced crown interception and limited water uptake response (i.e. less leaf area) compared to white spruce, which has more leaf area intercepting precipitation and uses more of the available soil water (Elliott et al. 1998; Herzog et al. 1998).

Trembling aspen and white spruce are two major components in the northern continental boreal forest region of Canada that showed specific responses to soil water availability driven by rooting space and topography. Their distinct rooting characteristics and physiological differences led to contrasting behaviors in terms of aboveground growth, leaf area development and water uptake and use on the reclaimed site. In light of a drier, but more variable climate, the differing responses of aspen and spruce to seasonal variability and stochastic precipitation events will have profound effects on water cycling and how it is associated with forest composition and cover at the stand and landscape level.

Notes

Acknowledgments

We thank Marty Yarmuch and Craig Farnden for their logistical support and all those who provided field and lab support for this project over the years (Pak Chow, Frances Leishman, Simon Bockstette, Robert Hetmanski, Caren Jones, Jeff Kelly, Angeline Letourneau, Mika Little-Devito, Michelle McCutcheon, Shauna Stack) and data processing support (Newton Tran, ICT International Pty Ltd.). We thank Sean Carey and three anonymous reviewers for their comments on the manuscript.

Funding

We would like to acknowledge the funding support provided by the National Science and Engineering Research Council (NSERC), the Canadian Oil Sands Innovation Alliance (COSIA, Syncrude Canada Ltd., Canadian Natural Resources Ltd., Suncor Energy, Imperial Oil Ltd.).

Supplementary material

11104_2019_4029_MOESM1_ESM.docx (884 kb)
ESM 1 (DOCX 884 kb)

References

  1. Alberta Parks (2015) Natural Regions and Subregions of Alberta. A Framework for Alberta’s Parks. Alberta Tourism, Parks and Recreation., Edmonton, AlbertaGoogle Scholar
  2. Bakker MR, Augusto L, Achat DL (2006) Fine root distribution of trees and understory in mature stands of maritime pine (Pinus pinaster) on dry and humid sites. Plant Soil 286:37–51.  https://doi.org/10.1007/s11104-006-9024-4 CrossRefGoogle Scholar
  3. Barbour L, Chanasyk D, Hendry MJ, Leskiw L, Macyk T, Mendoza C, Naeth A, Nichol C, O’Kane M, Purdy B, Qualizza C, Quideau S, Welham C (2007) Soil capping research in the Athabasca Oil Sands region. Technology Synthesis, vol. 1. Syncrude Canada Ltd. - Internal Publication.Google Scholar
  4. Bauhus J, Messier C (1999) Soil exploitation strategies of fine roots in different tree species of the southern boreal forest of eastern Canada. Can J For Res 29:260–273.  https://doi.org/10.1139/x98-206 CrossRefGoogle Scholar
  5. Bladon KD, Silins U, Landhäusser SM, Lieffers VJ (2006) Differential transpiration by three boreal tree species in response to increased evaporative demand after variable retention harvesting. Agric For Meteorol 138:104–119.  https://doi.org/10.1016/J.AGRFORMET.2006.03.015 CrossRefGoogle Scholar
  6. Blake TJ, Li J (2003) Hydraulic adjustment in jack pine and black spruce seedlings under controlled cycles of dehydration and rehydration. Physiol Plant 117:532–539.  https://doi.org/10.1034/j.1399-3054.2003.00059.x CrossRefPubMedGoogle Scholar
  7. Blake TJ, Sperry JS, Tschaplinski TJ, Wang SS (1996) Water relations. In: Stettler RF, Bradshaw HDJ, Heilman PE, Hinckley TM (eds) Biology of Populus and its implications for management and conservation. NRC Research Press, National Research Council of Canada, Ottawa, ON, pp 401–422Google Scholar
  8. Blanken PD (1997) Evaporation within and above a boreal aspen forest. PhD thesis. University of British ColumbiaGoogle Scholar
  9. Bleby TM, Burgess SSO, Adams MA (2004) A validation, comparison and error analysis of two heat-pulse methods for measuring sap flow in Eucalyptus marginata saplings. Funct Plant Biol 31:645.  https://doi.org/10.1071/FP04013 CrossRefGoogle Scholar
  10. Bockstette SW, Pinno BD, Dyck MF, Landhäusser SM (2017) Root competition, not soil compaction, restricts access to soil resources for aspen on a reclaimed mine soil. Botany 95:685–695.  https://doi.org/10.1139/cjb-2016-0301 CrossRefGoogle Scholar
  11. Boland AM, Jerie PH, Mitchell PD, Goodwin I, Connor DJ (2000) Long-term effects of restricted root volume and regulated deficit irrigation on peach: II. Productivity and water use. J Am Soc Hortic Sci 125:143–148CrossRefGoogle Scholar
  12. Bonan GB (1992) Soil temperature as an ecological factor in boreal forests. In: Shugart HH, Leemans R, Bonan GB (eds) A systems analysis of the global boreal Forest. Cambridge University Press, Cambridge, pp 126–143.  https://doi.org/10.1017/CBO9780511565489.005 CrossRefGoogle Scholar
  13. Børja I, Svĕtlík J, Nadezhdin V, Čermák J, Rosner S, Nadezhdina N (2016) Sap flux – a real time assessment of health status in Norway spruce. Scand J For Res 31:450–457.  https://doi.org/10.1080/02827581.2015.1130851 CrossRefGoogle Scholar
  14. Brédoire F, Nikitich P, Barsukov PA, Derrien D, Litvinov A, Rieckh H, Rusalimova O, Zeller B, Bakker MR (2016) Distributions of fine root length and mass with soil depth in natural ecosystems of southwestern Siberia. Plant Soil 400:315–335.  https://doi.org/10.1007/s11104-015-2717-9 CrossRefGoogle Scholar
  15. Brunner I, Herzog C, Dawes MA, Arend M, Sperisen C (2015) How tree roots respond to drought. Front Plant Sci 6:547.  https://doi.org/10.3389/fpls.2015.00547 CrossRefPubMedPubMedCentralGoogle Scholar
  16. Burgess SSO, Bleby TM (2006) Redistribution of soil water by lateral roots mediated by stem tissues. J Exp Bot 57:3283–3291.  https://doi.org/10.1093/jxb/erl085 CrossRefPubMedGoogle Scholar
  17. Burgess SSO, Adams MA, Turner NC, Beverly CR, Ong CK, Khan AAH, Bleby TM (2001) An improved heat pulse method to measure low and reverse rates of sap flow in woody plants. Tree Physiol 21:589–598.  https://doi.org/10.1093/treephys/21.9.589 CrossRefPubMedGoogle Scholar
  18. Cogbill CV (1985) Dynamics of the boreal forests of the Laurentian Highlands, Canada. Can J For Res 15:252–261.  https://doi.org/10.1139/x85-043 CrossRefGoogle Scholar
  19. Cribari-Neto F, Zeileis A (2010) Beta Regression in R. J Stat Softw 34:1–24.  https://doi.org/10.18637/jss.v034.i02 CrossRefGoogle Scholar
  20. Cuneo IF, Knipfer T, Brodersen CR, McElrone AJ (2016) Mechanical failure of fine root cortical cells initiates plant hydraulic decline during drought. Plant Physiol 172:1669–1678.  https://doi.org/10.1104/pp.16.00923 CrossRefPubMedPubMedCentralGoogle Scholar
  21. Dawson TE, Burgess SSO, Tu KP, Oliveira RS, Santiago LS, Fisher JB, Simonin KA, Ambrose AR (2007) Nighttime transpiration in woody plants from contrasting ecosystems. Tree Physiol 27:561–575.  https://doi.org/10.1093/TREEPHYS/27.4.561 CrossRefPubMedGoogle Scholar
  22. DeByle NV, Winokur RP (1985) Aspen: ecology and management in the western United States, USDA Forest Service General Technical Report RM-119. Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colo. 283 p.  https://doi.org/10.2737/RM-GTR-119
  23. Delbart N, Picard G, Le Toan T, Kergoat L, Quegan S, Woodward I, Dye D, Fedotova V (2008) Spring phenology in boreal Eurasia over a nearly century time scale. Glob Chang Biol 14:603–614.  https://doi.org/10.1111/j.1365-2486.2007.01505.x CrossRefGoogle Scholar
  24. Devito K, Mendoza C, Qualizza C (2012) Conceptualizing water movement in the Boreal Plains. Implications for watershed reconstruction. Synthesis report prepared for the Canadian Oil Sands Network for Research and Development, Environmental and Reclamation Research Group, p 164 Google Scholar
  25. Doronila AI, Forster MA (2015) Performance measurement via sap flow monitoring of three Eucalyptus species for mine site and dryland salinity phytoremediation. Int J Phytoremediation 17:101–108.  https://doi.org/10.1080/15226514.2013.850466 CrossRefPubMedGoogle Scholar
  26. Elliott JA, Toth BM, Granger RJ, Pomeroy JW (1998) Soil moisture storage in mature and replanted sub-humid boreal forest stands. Can J Soil Sci 78:17–27.  https://doi.org/10.4141/S97-021 CrossRefGoogle Scholar
  27. Environment Canada (2013) Canadian Climate Normals 1981–2010 Station Data - Climate - Environment and Climate Change Canada - Environment and Climate Change Canada [WWW Document]. URL http://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?stnID=2519&dispBack=0&month1=1&month2=9. Accessed 7.10.18
  28. Forster MA (2017) How reliable are heat pulse velocity methods for estimating tree transpiration? Forests 8:350.  https://doi.org/10.3390/f8090350 CrossRefGoogle Scholar
  29. Fung MYP, Macyk TM (2000) Reclamation of oil sands mining areas. In: Barnhisel RI, Darmody RG, Daniels WL (eds) Reclamation of drastically disturbed lands, ASA, CSSA and SSSA. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, pp 755–774.  https://doi.org/10.2134/agronmonogr41.c30 CrossRefGoogle Scholar
  30. Gaul D, Hertel D, Borken W, Matzner E, Leuschner C (2008) Effects of experimental drought on the fine root system of mature Norway spruce. For Ecol Manag 256:1151–1159.  https://doi.org/10.1016/J.FORECO.2008.06.016 CrossRefGoogle Scholar
  31. Government of Alberta (2017) Environmental protection and enhancement act. Revised Statutes of Alberta 2000 Chapter E-12. http://www.qp.alberta.ca/documents/acts/e12.pdf. Accessed 20 Apr 2018
  32. Grossnickle SC (2000) Ecophysiology of northern spruce species : the performance of planted seedlings. NRC Research Press, Ottawa, 409Google Scholar
  33. Herzog K, Thum R, Kronfus G, Heldstab H-J, Hasler R (1998) Patterns and mechanisms of transpiration in a large subalpine Norway spruce (Picea abies (L.) Karst.). Ecol Res 13:105–116.  https://doi.org/10.1046/j.1440-1703.1998.00250.x CrossRefGoogle Scholar
  34. Hogg EH, Hurdle PA (1997) Sap flow in trembling aspen: implications for stomatal responses to vapor pressure deficit. Tree Physiol 17:501–509.  https://doi.org/10.1093/treephys/17.8-9.501 CrossRefPubMedGoogle Scholar
  35. Hogg EH, Saugier B, Pontailler J-Y, Black TA, Chen W, Hurdle PA, Wu A (2000) Responses of trembling aspen and hazelnut to vapor pressure deficit in a boreal deciduous forest. Tree Physiol 20:725–734.  https://doi.org/10.1093/treephys/20.11.725 CrossRefPubMedGoogle Scholar
  36. Huang J, Tardif JC, Bergeron Y, Denneler B, Berninger F, Girardin MP (2010) Radial growth response of four dominant boreal tree species to climate along a latitudinal gradient in the eastern Canadian boreal forest. Glob Chang Biol 16:711–731.  https://doi.org/10.1111/j.1365-2486.2009.01990.x CrossRefGoogle Scholar
  37. Huang M, Lee Barbour S, Elshorbagy A, Zettl J, Cheng Si B (2011) Water availability and forest growth in coarse-textured soils. Can J Soil Sci 91:199–210.  https://doi.org/10.4141/cjss10012 CrossRefGoogle Scholar
  38. Huang M, Barbour SL, Carey SK (2015) The impact of reclamation cover depth on the performance of reclaimed shale overburden at an oil sands mine in northern Alberta, Canada. Hydrol Process 29:2840–2854.  https://doi.org/10.1002/hyp.10229 CrossRefGoogle Scholar
  39. IPCC (2013) In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Climate Change 2013: The Physical Science Basis. Contribution of Working Group 1 to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New YorkGoogle Scholar
  40. Jackson RB, Canadell J, Ehleringer JR, Mooney HA, Sala OE, Schulze ED (1996) A global analysis of root distributions for terrestrial biomes. Oecologia 108:389–411.  https://doi.org/10.1007/BF00333714 CrossRefPubMedGoogle Scholar
  41. Ježík M, Blaženec M, Letts MG, Ditmarová Ľ, Sitková Z, Střelcová K (2015) Assessing seasonal drought stress response in Norway spruce (Picea abies (L.) Karst.) by monitoring stem circumference and sap flow. Ecohydrology 8:378–386.  https://doi.org/10.1002/eco.1536 CrossRefGoogle Scholar
  42. Jung K, Duan M, House J, Chang SX (2014) Textural interfaces affected the distribution of roots, water, and nutrients in some reconstructed forest soils in the Athabasca oil sands region. Ecol Eng 64:240–249.  https://doi.org/10.1016/j.ecoleng.2013.12.037 CrossRefGoogle Scholar
  43. Kalliokoski T, Nygren P, Sievänen R (2008) Coarse root architecture of three boreal tree species growing in mixed stands. Silva Fenn 42.  https://doi.org/10.14214/sf.252
  44. Kelln C, Barbour SL, Qualizza C (2008) Controls on the spatial distribution of soil moisture and solute transport in a sloping reclamation cover. Can Geotech J 45:351–366.  https://doi.org/10.1139/T07-099 CrossRefGoogle Scholar
  45. Kessler S, Barbour SL, van Rees KCJ, Dobchuk BS (2010) Salinization of soil over saline-sodic overburden from the oil sands in Alberta. Can J Soil Sci 90:637–647.  https://doi.org/10.4141/cjss10019 CrossRefGoogle Scholar
  46. Lanoue AVL (2003) Phosphorus content and accumulation of carbon and nitrogen in boreal forest soils. M.Sc. Thesis. University of Alberta, Edmonton, ABGoogle Scholar
  47. Lawrence DJ, Luckai N, Meyer WL, Shahi C, Fazekas AJ, Kesanakurti P, Newmaster S (2012) Distribution of white spruce lateral fine roots as affected by the presence of trembling aspen: root mapping using simple sequence repeat DNA profiling. Can J For Res 42:1566–1576.  https://doi.org/10.1139/x2012-082 CrossRefGoogle Scholar
  48. Lazorko H, Van Rees KCJ (2012) Root distributions of planted boreal mixedwood species on reclaimed saline–sodic overburden. Water Air Soil Pollut 223:215–231.  https://doi.org/10.1007/s11270-011-0852-3 CrossRefGoogle Scholar
  49. Lemeur R, Fernández JE, Steppe K (2008) Symbols, SI units and physical quantities within the scope of sap flow studies. In: Fernández E, Diaz-Espejo A (eds) Acta Horticulturae. International Society for Horticultural Science (ISHS), Leuven, pp 21–32Google Scholar
  50. Lenth RV (2018) emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.2.3. https://CRAN.R-project.org/package=emmeans. Accessed 20 July 2018
  51. Leo M, Oberhuber W, Schuster R, Grams TEE, Matyssek R, Wieser G (2014) Evaluating the effect of plant water availability on inner alpine coniferous trees based on sap flow measurements. Eur J For Res 133:691–698.  https://doi.org/10.1007/s10342-013-0697-y CrossRefGoogle Scholar
  52. Lilles EB, Purdy BG, Macdonald SE, Chang SX (2012) Growth of aspen and white spruce on naturally saline sites in northern Alberta: implications for development of boreal forest vegetation on reclaimed saline soils. Can J Soil Sci 92:213–227.  https://doi.org/10.4141/cjss2010-032 CrossRefGoogle Scholar
  53. Link P, Simonin K, Maness H, Oshun J, Dawson T, Fung I (2014) Species differences in the seasonality of evergreen tree transpiration in a Mediterranean climate: analysis of multiyear, half-hourly sap flow observations. Water Resour Res 50:1869–1894.  https://doi.org/10.1002/2013WR014023 CrossRefGoogle Scholar
  54. Looker N, Martin J, Jencso K, Hu J (2016) Contribution of sapwood traits to uncertainty in conifer sap flow as estimated with the heat-ratio method. Agric For Meteorol 223:60–71.  https://doi.org/10.1016/J.AGRFORMET.2016.03.014 CrossRefGoogle Scholar
  55. MacPherson DM, Lieffers VJ, Blenis PV (2001) Productivity of aspen stands with and without a spruce understory in Alberta’s boreal mixedwood forests. For Chron 77:351–356.  https://doi.org/10.5558/tfc77351-2 CrossRefGoogle Scholar
  56. Marshall DC (1958) Measurement of sap flow in conifers by heat transport. Plant Physiol 33:385–396.  https://doi.org/10.1104/PP.33.6.385 CrossRefPubMedPubMedCentralGoogle Scholar
  57. Matthes-Sears U, Larson DW (1999) Limitations to seedling growth and survival by the quantity and quality of rooting space: implications for the establishment of Thuja occidentalis on cliff faces. Int J Plant Sci 160:122–128.  https://doi.org/10.1086/314105 CrossRefGoogle Scholar
  58. Messier C, Coll L, Poitras-Larivière A, Bélanger N, Brisson J (2009) Resource and non-resource root competition effects of grasses on early- versus late-successional trees. J Ecol 97:548–554.  https://doi.org/10.1111/j.1365-2745.2009.01500.x CrossRefGoogle Scholar
  59. Mundell TL, Landhäusser SM, Lieffers VJ (2007) Effects of Corylus cornuta stem density on root suckering and rooting depth of Populus tremuloides this article is one of a selection of papers published in the special issue on poplar research in Canada. Can J Bot 85:1041–1045.  https://doi.org/10.1139/B07-089 CrossRefGoogle Scholar
  60. Nadezhdina N, Cermák J, Meiresonne L, Ceulemans R (2007) Transpiration of Scots pine in Flanders growing on soil with irregular substratum. For Ecol Manag 243:1–9.  https://doi.org/10.1016/J.FORECO.2007.01.089 CrossRefGoogle Scholar
  61. Nlungu-Kweta P, Leduc A, Bergeron Y (2014) Conifer recruitment in trembling aspen (Populus tremuloides Michx.) stands along an east-west gradient in the boreal mixedwoods of Canada. Forests 5:2905–2928.  https://doi.org/10.3390/f5112905 CrossRefGoogle Scholar
  62. Novoplansky A (2009) Picking battles wisely: plant behaviour under competition. Plant Cell Environ 32:726–741.  https://doi.org/10.1111/j.1365-3040.2009.01979.x CrossRefPubMedGoogle Scholar
  63. Ojekanmi AA, Chang SX (2014) Soil quality assessment for peat–mineral mix cover soil used in oil sands reclamation. J Environ Qual 43:1566.  https://doi.org/10.2134/jeq2014.02.0061 CrossRefPubMedGoogle Scholar
  64. Pfautsch S, Keitel C, Turnbull TL, Braimbridge MJ, Wright TE, Simpson RR, O’Brien JA, Adams MA (2011) Diurnal patterns of water use in Eucalyptus victrix indicate pronounced desiccation-rehydration cycles despite unlimited water supply. Tree Physiol 31:1041–1051.  https://doi.org/10.1093/treephys/tpr082 CrossRefPubMedGoogle Scholar
  65. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team (2018) _nlme: Linear and nonlinear mixed effects models_. R package version 3.1–137. https://CRAN.R-project.org/package=nlme. Accessed 20 Apr 2018
  66. Pollard DFW (1970) Leaf area development on different shoot types in a young aspen stand and its effect upon production. Can J Bot 48:1801–1804.  https://doi.org/10.1139/b70-264 CrossRefGoogle Scholar
  67. R Development Core Team (2018) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/. Accessed 20 Apr 2018
  68. Roberts DR, Dumbroff EB (1986) Relationships among drought resistance, transpiration rates, and abscisic acid levels in three northern conifers. Tree Physiol 1:161–167.  https://doi.org/10.1093/treephys/1.2.161 CrossRefPubMedGoogle Scholar
  69. Sakai Y, Takahashi M, Tanaka N (2007) Root biomass and distribution of a Picea—Abies stand and a Larix—Betula stand in pumiceous Entisols in Japan. J For Res 12:120–125.  https://doi.org/10.1007/s10310-006-0270-3 CrossRefGoogle Scholar
  70. Schaffer B, Whiley AW, Searle C (1999) Atmospheric CO2 enrichment, root restriction, photosynthesis, and dry-matter partitioning in subtropical and tropical fruit crops. Hortic Sci 34:1033–1037Google Scholar
  71. Schenk HJ, Jackson RB (2002) The global biogeography of roots. Ecol Monogr 72:311–328. https://doi.org/10.1890/0012-9615(2002)072[0311:TGBOR]2.0.CO;2Google Scholar
  72. Schuster R, Oberhuber W, Gruber A, Wieser G (2016) Soil drought decreases water-use of pine and spruce but not of larch in a dry inner alpine valley. Aust J Forensic Sci:1–17Google Scholar
  73. Sellin A (2000) Estimating the needle area from geometric measurements: application of different calculation methods to Norway spruce. Trees 14:215–222.  https://doi.org/10.1007/PL00009765 CrossRefGoogle Scholar
  74. Snedden JE (2013) The root distribution, architecture, transpiration and root sapflow dynamics of mature trembling aspen (Populus tremuloides) growing along a hillslope. M. Sc. Thesis. University of AlbertaGoogle Scholar
  75. Střelcová K, Kurjak D, Leštianska A, Kovalčíková D, Ditmarová Ľ, Škvarenina J, Ahmed Y (2013) Differences in transpiration of Norway spruce drought stressed trees and trees well supplied with water. Biologia 68:1118–1122.  https://doi.org/10.2478/s11756-013-0257-4 CrossRefGoogle Scholar
  76. Strong WL, LaRoi GH (1983) Root-system morphology of common boreal forest trees in Alberta, Canada. Can J For Res 13:1164–1173.  https://doi.org/10.1139/x83-155 CrossRefGoogle Scholar
  77. Tie Q, Hu H, Tian F, Guan H, Lin H (2017) Environmental and physiological controls on sap flow in a subhumid mountainous catchment in North China. Agric For Meteorol 240–241:46–57.  https://doi.org/10.1016/J.AGRFORMET.2017.03.018 CrossRefGoogle Scholar
  78. van Buuren S, Groothuis-Oudshoorn K (2011) Mice: multivariate imputation by chained equations in R. J Stat Softw 45:1–67.  https://doi.org/10.18637/jss.v045.i03 CrossRefGoogle Scholar
  79. Van Cleve K, Yarie J (1986) Interaction of temperature, moisture, and soil chemistry in controlling nutrient cycling and ecosystem development in the Taiga of Alaska. In: Van Cleve K, Chapin FS, Flanagan PW, Viereck LA, Dyrness CT (eds) Forest Ecosystems in the Alaskan Taiga. Ecological Studies (Analysis and Synthesis), vol. 57. Springer, New York, pp 160–189.  https://doi.org/10.1007/978-1-4612-4902-3_12
  80. Wiley E, Rogers BJ, Griesbauer HP, Landhäusser SM (2018) Spruce shows greater sensitivity to recent warming than Douglas-fir in Central British Columbia. Ecosphere 9:e02221.  https://doi.org/10.1002/ecs2.2221 CrossRefGoogle Scholar
  81. Wood SN (2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc Ser B Stat Methodol 73:3–36.  https://doi.org/10.1111/j.1467-9868.2010.00749.x CrossRefGoogle Scholar
  82. Wood SN (2017) Generalized additive models : an introduction with R, 2nd edn. Chapman & Hall/CRC, Boca RatonCrossRefGoogle Scholar
  83. Xia YQ, Shao MA (2009) Evaluation of soil water-carrying capacity for vegetation: the concept and the model. Acta Agric Scand Sect B Soil Plant Sci 59:342–348.  https://doi.org/10.1080/09064710802203537 CrossRefGoogle Scholar
  84. Zweifel R, Bohm JP, Hasler R (2002) Midday stomatal closure in Norway spruce--reactions in the upper and lower crown. Tree Physiol 22:1125–1136.  https://doi.org/10.1093/treephys/22.15-16.1125 CrossRefPubMedGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Renewable ResourcesUniversity of AlbertaEdmontonCanada

Personalised recommendations