Plant Soil DOI 10.1007/s11104-017-3356-0

REGULAR ARTICLE

Boreal forest plant species responses to pH: ecological interpretation and application to reclamation Monica Calvo-Polanco & Wenqing Zhang & S. Ellen Macdonald & Jorge Señorans & Janusz J. Zwiazek

Received: 27 March 2017 / Accepted: 20 July 2017 # Springer International Publishing AG 2017

Abstract Aims Reclamation following oil sands mining in northeastern Alberta (Canada) creates adverse reforestation soil conditions, including extreme pH values. We elucidated pH tolerance limits of boreal plant species and how pH affects nutrient uptake in these plants. Methods We measured growth, gas exchange, and foliar nutrient concentration of 15 common northern boreal forest plants after eight weeks exposure to root zone pH ranging from 5.0 to 9.0. Cluster analyses were used to group these species based on their pH responses. Results Based on their growth and gas exchange responses to pH, the 15 plant species could be divided into five groups, each of which contained species that commonly co-occur in particular boreal forest site types. For the foliar nutrient responses to pH, the 15 species could be grouped into only two categories; both showed decreases

Responsible Editor: Michael A. Grusak. Electronic supplementary material The online version of this article (doi:10.1007/s11104-017-3356-0) contains supplementary material, which is available to authorized users. M. Calvo-Polanco : W. Zhang : S. Ellen Macdonald : J. Señorans : J. J. Zwiazek (*) Department of Renewable Resources, University of Alberta, 4-38 Earth Sciences Building, Edmonton, AB T6G 2E3, Canada e-mail: [email protected] Present Address: M. Calvo-Polanco Biochimie et Physiologie Moléculaire des Plantes, SupAgro/ INRA, Place Viala 1, 34060 Montpellier Cedex 2, France

in foliar N, P, Fe and Zn concentration with increasing pH, with a more pronounced effect on the group that included trembling aspen, paper birch and chokecherry. Conclusions The evidence of differential adaptation to pH by habitat type suggests the importance of soil pH as a factor affecting boreal plant species distribution and could be helpful for selection of species suitable for reclamation of sites with altered soil pH. Keywords Boreal plants . Growth . Mineral nutrients . Oil sands reclamation . pH

Introduction Surface mining of oil sands in northeastern Alberta, Canada, has created challenging environmental problems due to soil removal from vast areas of land. The impacts of this development on the environment are mitigated through reclamation, reforestation and mine closure activities, but the environmental consequences of these activities remain unknown and the methods of soil reclamation and landscape ecological restoration are largely unproven (Grant et al. 2008; Audet et al. 2014; Macdonald et al. 2015). Potential differences in pH between reclamation soils compared with undisturbed forest soils can profoundly affect forest productivity in oil sands reclamation areas. Unmanaged boreal forests near oil sands reclamation sites are characterized by luvisolic and brunisolic soils with organic soils on wetter sites; soil pH values are typically below 6.0 with wet peatland sites having the

Plant Soil

lowest values (Natural Regions Committee 2006). In contrast, the soil pH values measured in oil sands reclamation areas frequently exceeds 8.0 (Howat 2000). Although some plants tolerate a relatively wide pH range of 3.5–8.5, the pH optimum may be relatively narrow for other species of plants (Iles 2001; Larcher 2003). There is little information on pH tolerance range and tolerance mechanisms in boreal forest plants since most of the studies considered pH only in combination with other confounding factors such as salinity (Renault et al. 1999; Kopittke and Menzies 2005; Yousfi et al. 2007; Calvo-Polanco et al. 2014). Species tolerant of high soil pH may show a different pattern of nutrient accumulation in leaves and stems, compared to pH-sensitive species. For example, in trefoil (Lotus corniculatus L), which is tolerant of high soil pH, the concentrations of Al and Zn in leaves and stems decreased with increasing pH (5–8), whereas Mg and K concentrations were unaffected, and Ca and P accumulated with increasing pH in the leaves (Kallenbach et al. 1996). Similarly, species differ in their growth responses in soils with high pH. In a high pH-sensitive species of Lupinus, a combination of decreases in overall root growth, root surface area, root elongation, root cell division or cell elongation and volume was observed, which was in contrast to alkaline-tolerant species of Lupinus and Pisum sativum (Tang et al. 1992, 1993). Reduced root growth at high pH was found to be caused by the plant’s inability to maintain an acidified apoplast (Tang et al. 1996). The effects of high pH may be aggravated by other factors that often accompany high soil pH or are used to increase pH in experimental set-ups. Bicarbonate (HCO3−) and Ca have often been used to produce high pH conditions and to examine high pH tolerance in plants (Kerley and Huyghe 2001). However, in high pH-sensitive Lupinus, HCO3− resulted in specific stress effects such as decreased shoot growth, tap root death, and decreased number and density of determinate roots, whereas Ca affected the whole root system, resulting in shorter tap roots, decreased overall root growth and lateral branching (Kerley and Huyghe 2002). It has been proposed that the poor growth of conifer trees that has been reported for calcareous soils is likely due to impaired root growth and effects on gas exchange caused by reduced water uptake (Zhang et al. 2015). In the present study, we used a controlledenvironment hydroponics set-up to examine the shortterm responses of 15 plant species that are commonly

found in the North American boreal forest to root zone pH. The aims of the study were to: i) investigate the effects of pH as one of the principle stress factors on plants, ii) examine variation in pH tolerance among boreal species that are commonly planted in oil sands reclamation areas with altered soil pH, and iii) identify possible mechanisms that may be responsible for any observed differences in pH tolerance among boreal plants. A finding that plant species sharing similar ecological habitat affinities respond similarly to pH treatments would suggest that root zone pH is among the major factors determining site distribution of the different boreal plants. Further, an improved understanding of response to pH among these species will help inform decision about their use in reclamation of sites with different soil pH.

Materials and methods Plant material and growth conditions The plant species included in the experiment are commonly found in the northern boreal forest of Alberta, Canada: jack pine (Pinus banksiana), white spruce (Picea glauca), black spruce (Picea mariana), cinquefoil (Dasiphora fruticosa ssp. floribunda), tamarack (Larix laricina), trembling aspen (Populus tremuloides), paper birch (Betula papyrifera), green alder (Alnus viridis), blueberry (Vaccinium myrtilloides), buffalo berry (Shepherdia canadensis), red-osier dogwood (Cornus sericea), Labrador tea (Ledum groenladicum), bearberry (Arctostaphyllos uva-ursi), choke cherry (Prunus virginiana), and saskatoon (Amelanchier alnifolia). Seeds were collected from plants growing in unmanaged boreal forest near Fort McMurray, AB, Canada. No information is available on the pH conditions at the collection sites, but see below for a description of pH in the different habitat types where the species are commonly found. The seeds were germinated and seedlings grown for six months in a mixture of peat and perlite (5:1 by volume) at the Smoky Lake Forest Nursery (Smoky Lake, AB, Canada). Seed stratification of about 3 months was required for buffalo berry. The plants were grown in a greenhouse at 24/ 20 °C, natural day length/light (supplemented with 600 μmol m−2 s−1 photosynthetic photon flux density (PPFD) lights) and 60% humidity. We selected seedlings of similar size, and transferred them into solution

Plant Soil

needles three to six hours after the lights turned on in the growth chamber in the morning. We used an infrared gas analyzer (LCA-4, Analytical Development Company Ltd., Hertfordshire, UK) and an auxiliary light source (1000 μmol m−2 s−1 PPFD, approximately 400 μmol CO2) as previously described in Calvo-Polanco et al. (2012). Plants were exposed to 1000 μmol m−2 s−1 PPFD for 30 min prior to measurement; this allowed them to adjust to the higher light conditions, reducing the chance of photoinhibition. Leaf and needle surface areas were determined with the Sigma Scan 5.0 software following scanning (Jandel Scientific, San Rafael, CA). Following the gas exchange measurements all plants were harvested. Root collar diameter and height were measured and then leaves, stems and roots were separated, oven-dried at 65 °C for two days and weighed to calculate the leaf, stem, root, and total dry weights and shoot:root dry weight ratios. Since these parameters were measured in plants after the same treatment period (8 weeks), we refer to them as growth.

culture in a controlled environment growth chamber at 16 h photoperiod, 24/18 °C day/night temperature, 65% relative humidity and 350–400 μmol m−2 s−1 (PPFD). Plants were grown in a 50% modified Hoagland’s mineral solution (Epstein 1972) for 15 days before the start of pH treatments. The pH of nutrient solution was adjusted to 5.0, 6.0, 7.0, 7.5, 8.0, 8.5, and 9.0, using an automated hydroponics set-up in which pH of the mineral solution was continuously controlled with KOH and H2SO4 for up to 8 weeks (Calvo-Polanco et al. 2014). Three replicated containers were used for each pH level, with a total of 21 containers for the whole system. There were 6 plants per species in each replicate container of each pH treatment for a total of 18 plants per species per pH treatment. The mineral solution was replaced every two weeks. The solution culture set-up for each treatment combination consisted of a 120 L pail (central container) filled with nutrient solution and connected through vinyl tubing (outlet and inlet) to three 45 L containers with plants as earlier explained (Calvo-Polanco et al. 2014; Zhang and Zwiazek 2016). A pump (950GPH, Danner MFG Inc., New York, USA) inside the pail circulated the nutrient solution between the pail and containers with plants. The total volume of solution circulating in each set was about 200 L. The pH in each pail was controlled with an automated system consisting of a pH controller (PHCN-70 Omega Engineering Inc., Laval, Canada) connected to a pH meter (Orion 3-Star pH Portable Meter, ThermoScientific, Toronto, Canada) and a gel-filled combination pH electrode (Orion 9106 BNWP, ThermoScientific, Toronto, Canada). The pH controller opened a one-way valve (Asco series 8016 G/H, ASCO valve Inc., Brantford, Canada) to release the pH-controlling solution to the central container to maintain the desired pH. The pH-controlling solutions were 5% (w/v) KOH for higher pH ranges and 1% (v/v) H2SO4 for the lower pH ranges. Overall, nutrient solution reached the targeted pH values within ten minutes of the start of treatments and pH fluctuations during the experiment were within ±0.1 range.

At the time of harvest foliar N, P, K, Mg, Ca, Zn, Mn, Cu and Fe concentrations were determined in two or three plants per species randomly selected from each container of each pH treatment for a total of seven plants per species per pH treatment (n = 7). For N and P determinations, 0.5–1 g of ground, freeze-dried tissue was extracted using the sulfuric acid - hydrogen peroxide wet digestion method (Benton Jones 2001). The concentrations of N and P in the tissue were determined using a flame technique with a Model AA880 atomic absorption spectrophotometer (Varian Inc., Mississauga, ON, Canada). Concentrations of Mg, Ca, Zn, Mn, Cu, K and Fe were determined in 0.5–1 g of ground freezedried tissue using the nitric acid wet digestion extraction method (Benton Jones 2001). The concentrations of Mg, Ca, Zn, Mn, Cu and Fe were determined using a flame technique with a Model AA880 Atomic Absorption Spectrophotometer (Varian Inc.).

Gas exchange and growth measurements

Data analyses

After 8 weeks of pH treatments, leaf net assimilation and transpiration rates were measured in two plants per species from each of the three replicate containers for each pH treatment (n = 6 per species per pH treatment). Measurements were made on fully developed leaves and

To organize the species into groups according to their responses to the pH levels we used Cluster Analysis. The data into the Cluster Analysis consisted of average values for the response variables by pH level for each species. Those data were first standardized (to a mean of

Foliar ion concentrations

Plant Soil

zero and variance of one) and a matrix of Euclidean Distances was calculated. We compared among several possible agglomerative clustering methods (single linkage, average linkage, complete linkage, centroid, and Ward’s minimum variance) and used co-phenetic correlation and Gower’s method to select the optimal method (the two always agreed). Silhouette Plots were used to select the optimal number of clusters (Borcard et al. 2011). One Cluster Analysis was completed using data for all the growth and physiology parameters (leaf, stem, root and total dry weight, shoot:root ratio, root collar diameter, height, net assimilation and transpiration rates) and a separate Cluster Analysis was conducted using the data on foliar ion concentrations (N, P, K, Mg, Ca, Mn, Fe, Zn). We used an unconstrained ordination (Principle Components Analysis - PCA) on the same datasets (mean values for each response variable at each pH level for each species) to further verify the species groups determined by Cluster Analysis. Responses of the species groups to the pH levels for various response variables were then explored further through the construction of generalized additive models (GAMs) for each response variable (seven growth, two gas exchange, and eight foliar concentration variables). The datasets for these consisted of the average value of a response variable for the individuals of a given species in a replicate container (since they are considered as subsamples). There were, therefore, three replicate measures per species per pH level. We first compared a model with a simple linear relationship of the response variable to pH to a GAM model incorporating a smoother function representing a nonlinear relationship. The GAM model with the smoother was always superior based on the Akaike Information Criterion (AIC). GAMs were constructed to test the significance (α = 0.05) of the smoother function. To determine whether the groups differed in their response to pH we compared a model with one overall smoother versus one with a separate smoother for each species group. For all response variables, based on AIC, the model with a different smoother for each species group was superior. All analyses were conducted in R vers 3.2.2 using the vegan and gam packages (R Core Team 2015).

Results For the Cluster Analysis based on the growth and gas exchange response variables Average Linkage was

chosen as the best method and the Silhouette Plots indicated the optimal number of clusters to be five (Fig. 1a). The species groups were: 1. Dogwood; 2. Chokecherry and paper birch; 3. Black spruce and tamarack; 4. White spruce, trembling aspen and saskatoon; 5. Jack pine, bearberry, green alder, cinquefoil, Labrador tea, blueberry, and buffaloberry (Fig. 1a). These five groups separated clearly in the PCA (Fig. 2a). Group 1, which included only red-osier dogwood, loaded to the low end of PCA axis 1 and was strongly separated from the other four groups (Fig. 2a). Species in Group 2 (chokecherry, birch) loaded to the low end of the PCA axes 1 and 2, which separated them from the other groups (Fig. 2a). Species in Groups 3 and 4 loaded to the high end of the PCA axis 2 and separated from one another along axis 1 (Fig. 2a). The seven species in Group 5 showed greater spread in ordination space than the other groups but were well separated from them by loading towards the high end of axis 1 and to the mid- to lower end of axis 2 (Fig. 2a). For the Cluster Analysis based on the foliar ion concentration data Average Linkage was chosen to be the best method and the Silhouette Plots indicated the optimal number of clusters to be two (Fig. 1b). The species groups were: 1. Trembling aspen, paper birch, and chokecherry; 2. White spruce, black spruce, jack pine, blueberry, bearberry, Labrador tea, Saskatoon, tamarack, buffalo berry, cinquefoil, dogwood, green alder. The two groups separated along the first axis of the PCA (Fig. 2b). The three species in Group 1 loaded towards the low end of axis 1 while the 12 other species all loaded towards the mid- to higher end of axis 1. Within each group, species showed considerable spread across axis 2 (Fig. 2b). For the Generalized Additive Models (GAM), there was a significant effect of Bspecies group^ (as determined by Cluster Analysis) for every response variable except foliar Fe (Table 1). For the GAMs of the growth and gas exchange variables, the five species groups varied in how many and which variables showed a significant relationship to pH (Table 1a). Groups 3, 4 and 5 were less responsive to pH than Groups 1 and 2. For Group 3 (black spruce and tamarack) only the two gas exchange variables had a significant relationship to pH (Table 1a). For Groups 4 (white spruce, trembling aspen, and Saskatoon) and 5 (jack pine, bearberry, green alder, cinquefoil, Labrador tea, blueberry, buffalo berry) only four out of the nine growth and gas exchange variables showed a significant relationship to pH

Plant Soil

3

buffalo berry

cinquefoil

green alder

bearberry

Jack pine

saskatoon

Labrador tea blueberry

2

white spruce aspen

paper birch black spruce

1

tamarack

dogwood

(A)

chokecherry

Fig. 1 Results of cluster analysis (Average Linkage UPGMA) based on responses of (a) growth (root collar diameter, height, shoot:root ratio, stem, root, leaves, total dry weight) and gas exchange (NA and E); and (b) foliar ion concentrations to each of the seven pH levels. The dotted line shows the chosen break point for the dendrogram and the brackets show the groupings of species and the group number they are referred by

4 Group 5 Average linkage (UPGMA)

3 paper birch

aspen1

7 chokecherry 9 green alder

8 dogwood

12 cinquefoil

6 buffalo berry

14 tamarack

2 bearberry 11 Labrador tea

blueberry 5

10 Jack pine

13 saskatoon

black spruce4

15 white spruce

2

4

6

Height 8

10

12

(B)

Group 1

Group 2

(Table 1a). Groups 1 and 2 were more responsive to pH. For Group 1 (red-osier dogwood), five out of seven of the growth variables showed a significant relationship with pH while for Group 2 (chokecherry and paper birch), seven out of nine growth and gas exchange variables showed a significant relationship with pH (Table 1a). Only Group 4 (white spruce, trembling aspen and Saskatoon) showed a significant relationship between root dry weight and pH. For the GAMs of the foliar nutrient concentrations, both species groups

showed significant relationships with pH for all of the response variables with the exception of foliar Mg for Group 2 (Table 1b). For red-osier dogwood (Group 1), total dry weight peaked at pH 6 and then declined with increasing pH (Fig. 3). For Group 2 (chokecherry and paper birch), total dry weights also peaked at pH 6 and then declined to pH 7 after which it leveled off. There was a minimal, although significant, negative response of total dry weight to pH for Group 4 and no significant relationship

Plant Soil

PCA (growth and gas exchange)

(A)

Tamarack

1.0

White spruce Dogwood

0.5

Black spruce Jack pine

Aspen Saskatoon

Potenlla

0.0

PC2

-0.5

Buffaloberry Alder Blueberry Labrador tea

Chokecherry

-1.0

Fig. 2 Results of Principle Components Analysis based on based on responses of (a) growth (root collar diameter, height, shoot:root ratio, stem, root, leaves, total dry weight) and gas exchange (NA and E); and (b) foliar ion concentration to each of the seven pH levels. Symbols represent the groups determined by Cluster Analysis (see Fig. 1) and are labelled with species names

Bearberry

Birch

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

PCA (ion data)

1.0

(B)

Dogwood Alder Buffaloberry

0.0

Bearberry Blueberry

Potenlla

Jack pine Tamarack

-0.5

PC2

0.5

Labrador tea Chokecherry

Saskatoon

White spruce

-1.0

Birch

Black spruce

Aspen

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

PC1

in Groups 3 and 5 (Fig. 3). Absolute values of dry weight at each pH level by species within each group are provided in Online Resource 1. For all species groups, leaf and stem dry weight and height responses to pH were similar to those observed for total dry weight (Online Resources 2 and 3). Only Group 4 (white spruce, trembling aspen, and Saskatoon) showed a significant relationship of root dry weight with pH, which was fairly similar from pH 5 to 7 and then declined at higher pH values (Online Resource 4). Group 1 showed a fairly steep linear decline in shoot:root ratio with pH while for Group 2, shoot:root ratio was similar at pH from 5 to 6, declined at pH 7.5 and leveled off at the higher pH (Fig. 3). For shoot:root dry weight ratio,

Group 5 showed only a slight decline with pH and Groups 3 and 4 showed no significant relationship (Fig. 3). Only Group 2 showed a significant relationship between the root collar diameter and pH with a peak ratio at pH of 6, followed by a decline at pH up to 7.5 after which it leveled off (Online Resource 4). For Group 1, none of the gas exchange variables showed a significant response to pH (Table 1a). Groups 2, 3 and 5 showed a relatively linear decline in net assimilation rates with increasing pH and the slope was steepest for Group 2 (Fig. 4). For Group 3, net assimilation declined dramatically between pH 5 and 6 and levelled off thereafter (Fig. 4). Transpiration rates showed somewhat similar trends. For Groups 1, 2, and 4, there

Plant Soil Table 1 Results of the generalized additive models for the size and gas exchange response variables (A) and for the foliar ion concentration variables (B) (A) Response variable

Group

Smoother for Group 1

Group 2

Group 3

Group 4

Group 5

Leaf dry weight

<0.001

<0.001

<0.001

0.128

0.002

0.019

Stem dry weight

<0.001

<0.001

<0.001

0.326

0.265

0.483

Root dry weight

<0.001

0.108

0.438

0.813

0.002

0.467

Total dry weight

<0.001

<0.001

<0.001

0.286

0.005

0.125

Shoot:root ratio

<0.001

<0.001

<0.001

0.113

0.609

<0.001

Height

<0.001

<0.001

<0.001

0.564

0.183

0.101

Root collar diameter

<0.001

0.83

<0.001

0.956

0.78

0.915

Net assimilation

<0.001

0.10

0.005

<0.001

0.039

<0.001

Transpiration

<0.001

0.682

0.213

<0.001

0.46

0.004

(B) Response variable

Group

Smoother for Group 1

Group 2

N

<0.001

<0.001

<0.001

P

<0.001

<0.001

<0.001

K

<0.001

<0.001

<0.001

Ca

<0.001

<0.001

0.005

Mg

<0.001

<0.001

0.327

Mn

<0.001

0.003

0.004

Fe

0.54

<0.001

<0.001

Zn

<0.001

<0.001

<0.001

Given is the significance (p values) of species group as a fixed effect; this tested for differences among the five (A) or two (B) species groups in terms of their value for the response variable. Also given is the significance for the smoother function for each species group, indicating whether there was a significant non-linear relationship between the response variable and pH. AIC values were used to compare models with versus without species group as a fixed effect; in all cases the model including species group had a lower AIC value indicating that there were significant differences among species groups in terms of the relationship between the response variable and pH

was no significant relationship of transpiration with pH and for Group 5 there was a slightly negative relationship. For Group 3, transpiration rates showed a relatively steep decline across the tested pH range (Online Resource 5). For foliar nutrient concentrations, Group 1 (trembling aspen, paper birch, and chokecherry) generally responded more strongly to pH than Group 2 (which contained the remaining 12 species). Groups 1 and 2 showed a significantly negative relationship of foliar N and P concentrations with increasing pH, but the slope was steeper for Group 1 (Fig. 5). For both groups, foliar P was similar between pH 5 to 7 and then declined at higher pH. In contrast, foliar K increased with increasing pH for both groups (Fig. 5). Foliar Mg, Ca and Mn increased in Group 1 from pH 5 to 7 and then declined at higher pH while Group 2 showed little response of foliar Mg, Ca, and Mn concentrations to pH (Figs. 5 and 6). Finally, foliar Fe and

Zn were negatively related to pH for both groups. In Group 1, foliar Fe declined from pH 5 to 7.5 and then leveled off at higher pH (Fig. 6). Foliar Zn declined in Group 1 from pH 5 to 6 and was generally lower at the higher pH. Group 2 showed a slightly negative relationship with pH for both variables (Fig. 6). Absolute values of foliar nutrient concentrations at each pH level by species are provided in Online Resource 6.

Discussion Soil pH is a key environmental factor that affects the distribution and diversity of plant species within terrestrial ecosystems (Hayati and Proctor 1990). In the present study, the 15 studied boreal plant species were clustered into five groups based on physiological and

Plant Soil

(A)

-2 -1 0 1 2 3 4

Standardized Shoot:Root Rao

10 20

-20 -10 0

*

Group 3

10 20 -20 -10 0

n.s.

Group 4

-2 -1 0 1 2 3 4

*

Group 5

-20 -10 0

n.s.

5

6

7

8

9

pH

*

-2 -1 0 1 2 3 4

Group 2

10 20 -20 -10 0

Standardized Total Dry Weight

10 20

-20 -10 0

*

Group 1

Group 2

*

-2 -1 0 1 2 3 4

Group 1

Group 3

-2 -1 0 1 2 3 4

10 20

(B)

n.s.

Group 4

n.s.

Group 5

*

5

6

7

8

9

pH

Fig. 3 Portrayal of the Generalized Additive Models showing standardized (a) total dry weight and (b) Shoot:Root ratio as a function of pH for each of the five species groups. The solid line shows the non-linear relationship (the smoother function) while the

dotted lines show the with 95% confidence interval. Bn.s.^ indicates that the smoother function for that group was not significant; * indicates the smoother function for the group was significant. For species included in each group see Fig. 1a. See also Table 1a

growth responses to root zone pH. These groupings largely reflected the habitat affinities of these species, as observed in natural environments; this was despite the fact that the species that grouped together varied

substantially in size at the end of the experiment (Online Resource 1). This suggests some common adaptations to pH by habitat type. Notably, Group 3 included black spruce and tamarack; both species

Plant Soil

10

Group 1

-5

0

5

n.s.

5

10

Group 2

-5

0

*

10

Group 3

10

-5

0

5

*

Group 4

5

*

10

-5

0

Standardized Net Assimilaon

Group 5

0

5

*

-5

occur on wet forested peatland sites including rich bogs or poor fens with organic soils typified by pH < 5 (corresponds to ecosites g, i, j or k in the forest ecosite classification systems for the boreal mixedwood ecological area (Beckingham and Archibald 1996)). Group 4 included white spruce, trembling aspen and Saskatoon, which are common in mesic upland mixedwood boreal forests; these sites typically have luvisolic soils with pH of 5–6 (corresponds to ecosites d and e in the forest ecosite classification systems for the boreal mixedwood ecological area (Beckingham and Archibald 1996)). Trembling aspen-white spruce mixed stand is a major forest cover type in North America (Perala 1990). Group 5 included jack pine, bearberry, blueberry and Labrador tea, all of which are common on xeric, poorer sites with sandy soils (brunisolic soils on sandy parent material pH ~ 6; a, b and c ecosites in the boreal mixedwood ecological area; Beckingham and Archibald 1996). It was somewhat surprising that red-osier dogwood was the only species in its group and was particularly sensitive to pH, showing steep declines in leaf, stem and total dry weights, and height above pH 6, since it is not known to have very narrow habitat affinities (Burns and Honkala 1990). It was the fastest-growing species in the experiment (Online Resource 1– note difference in total dry weight between pH 9 and pH 5 or 6); thus, it had the greatest possibility to demonstrate a response to pH during the time period of the experiment. Chokecherry and paper birch (Group 2) were also sensitive to pH higher than 6 showing reduced leaf, stem and total dry weights, shoot:root ratio, root collar diameter and height. This suggests that these species would not be good candidates for planting on sites with high soil pH. The other species groups were relatively unresponsive to pH in terms of growth and gas exchange parameters, suggesting that they could be more suitable for revegetation of sites with a wider range of soil pH levels. Interestingly, all of the conifers in this study are in the relatively unresponsive groups. In previous studies, we also found that tamarack and white spruce were more resistant to high root zone pH compared with several other studied deciduous trees, including trembling aspen (Zhang et al. 2013). One limitation of this study is that the results may not indicate the

5

6

7

8

9

pH Fig. 4 Portrayal of the Generalized Additive Models showing standardized net assimilation as a function of pH for each of the five species groups. The solid line shows the non-linear relationship (the smoother function) while the dotted lines show the with 95% confidence interval. Bn.s.^ indicates that the smoother function for that group was not significant; * indicates the smoother function for the group was significant. For species included in each group see Fig. 1a. See also Table 1a

Plant Soil

*

0

1

2

*

*

*

*

-10 -5

0

5

*

1.0

*

n.s.

0

Magnesium

Potassium

10

-2

-1

Nitrogen Phosphorus

Group 2

-1.0

Standardized Foliar Concentraon

-4 -2 0 2 4 6

Group 1

5

6

7

8

pH

9

5

6

7

8

9

pH

Fig. 5 Portrayal of the Generalized Additive Models showing standardized foliar nutrient (N, P, K, Mg) concentrations as a function of pH for each of the five species groups. The solid line shows the non-linear relationship (the smoother function) while the dotted lines show the with 95% confidence interval. Bn.s.^

indicates that the smoother function for that group was not significant; * indicates the smoother function for the group was significant. For species included in each group see Fig. 1a. See also Table 1a

long-term responses of plants to various root zone pH values, not responses in soil conditions. Plants can acidify the rhizosphere and make the surrounding soil more suitable for their growth (Hinsinger

et al. 2003). It is possible that the species which appeared most sensitive to root zone pH in our hydroponics experiment use such a mechanism to tolerate higher pH values when growing in the

Plant Soil

Group 1

0 -40 -20 0 20 40 -20 0 20 40 60

*

*

-10 0 10 20

Manganese Iron Zinc

Standardized Foliar Concentraon

*

*

-5

Calcium

5

Group 2

5

*

*

*

*

6

7

8

pH

9

5

6

7

8

9

pH

Fig. 6 Portrayal of the Generalized Additive Models showing standardized foliar nutrient (Ca, Mn, Fe, Zn) concentrations as a function of pH for each of the five species groups. The solid line shows the non-linear relationship (the smoother function) while the dotted lines show the with 95% confidence interval. Bn.s.^

indicates that the smoother function for that group was not significant; * indicates the smoother function for the group was significant. For species included in each group see Fig. 1a. See also Table 1a

forest. Further, it is unlikely that there was substantial mycorrhizal colonization of roots for these plants after this time period of growth in hydroponics conditions. Thus, the results do not reflect any possible role of mycorrhizae in conferring pH tolerance, as would be found in natural forest soils.

Root growth has been frequently reported to be more affected by high pH than is shoot growth (CalvoPolanco et al. 2009, 2014; Zhang et al. 2013). However, in this study, only Group 4 (white spruce, trembling aspen, and Saskatoon) showed a significant influence of pH on root dry weight, which declined with

Plant Soil

increasing pH. This response could represent a shift of priorities in some plants to increase root biomass and facilitate the uptake of resources such as nutrients and water, which may be reduced under high pH conditions (Zhang et al. 2013). This is a common response observed in plants exposed to different environmental stresses (Ericsson 1995; Shalhevet et al. 1995). All species showed a general declining trend in net assimilation and transpiration rates with increasing pH, and this effect was the strongest in Group 3, which included black spruce and tamarack. Indeed, gas exchange in this group declined above pH 5. The bog and fen species typically grow in low pH soils (Thormann et al. 1999) and their root water transport, leaf transpiration, and stomatal conductance were reported to be strongly inhibited by high pH (Tang et al. 1993; Zhang et al. 2013). High soil pH is commonly associated also with several nutrient deficiencies in plants, because pH is one of the key factors determining nutrient absorption and solubility in soil (Comerford 2005). In this study, we could not determine the effects of pH on foliar K concentrations since KOH was used to increase pH of culture solution and its foliar concentrations increased with increasing pH. Although this is certainly one of the limitations of the study, we found KOH to be the least intrusive of any other agents that have been used for precise pH control and we successfully used it also in several other studies (Calvo-Polanco et al. 2014; Zhang et al. 2015).There is generally relatively little information available on nutrient requirements of boreal plant species. However, we measured similar foliar nutrient concentrations to those previously reported in similar experiments (Zhang et al. 2013, 2015). The Cluster Analyses for the foliar nutrient concentration data resulted in two species groups (Group 1 and 2); in Group 2 the foliar concentrations of N, P, Ca, Mn, Fe and Zn decreased with increasing pH. The responses for Group 1, however, were more complex. Foliar N, P, Fe and Zn generally declined with increasing pH but foliar Mg, Ca, and Mn had a non-linear response in which foliar concentration was highest around pH of 7. The three species in Group 1 (trembling aspen, paper birch, chokecherry) are associated with upland mesic ecosites in mixed boreal forests that typically have slightly higher pH values (5 to 6) than other ecosite types (see above) (Beckingham and Archibald 1996).

The availability of P, Fe and Zn is commonly reduced in alkaline soils (Marschner 2012), and the activity of enzymes for assimilating N can be impaired (Crawford and Glass 1998). Interestingly, in Group 1, three divalent cations, Mg, Ca and Mn, showed a bell shaped response to pH with the highest foliar nutrient concentration measured at pH 7. The biomass dilution effect (Jarrell and Beverly 1981) could be responsible for this response to pH since plant growth was greater at the lower concentrations up to pH 7 and then declined at higher pH. For site reclamation, any attempts to manage the soil pH need to consider the impact on soil chemistry including nutrient availability to plants. For the boreal plant species that we tested, N or P uptake could be limited at pH higher than 6.5. Considering low availability of these nutrients in soils of the boreal forest (Lupi et al. 2013), this could be a serious limitation to the revegetation efforts on such sites. Similarly, Fe or Zn deficiency would likely be a problem on sites with soil pH > 6. In conclusion, the results of this study have demonstrated that the 15 studied boreal forest plant species, which are commonly used for the reclamation of oil sands areas with altered soil pH characteristics, greatly vary in their responses to root zone pH. Many of the responses were similar among plant species with similar ecological habitat affinities, as indicated by both Cluster and PCA analyses carried out for physiological and growth parameters. The results point to the importance of soil pH as a factor affecting boreal plant species distribution. Red-osier dogwood, chokecherry and paper birch showed strong growth reductions under high root zone pH conditions suggesting that these species may be less suitable for land reclamation of high soil pH sites. On the other hand, high pH treatments had little impact on growth and physiological responses for groups 3, 4 and 5 and thus we expect better performance of species within these groups for the revegetation of oil sands reclamation with high pH, as compared with plants of Group 1 and 2. This includes all the conifer species (jack pine, white spruce, black spruce, tamarack) and trembling aspen, which is the dominant tree species across millions of hectares in the boreal mixedwood region (Beckingham and Archibald 1996). The negative effects of high pH on foliar concentrations of key

Plant Soil

nutrients such as N and P suggests that high pH could pose a limitation for establishment of these species on oil sands reclamation sites and that specific nutrient supply strategies might be needed. Acknowledgements We gratefully acknowledge research funding for the study provided through the Syncrude Canada Ltd., Suncor Energy Ltd., Albian Sands Ltd., and NSERC Collaborative Research and Development grants to J.J.Z.

References Audet P, Pinno BD, Thiffault E (2014) Reclamation of boreal forest after oil sands mining: anticipating novel challenges in novel environments. Can J For Res 45:364–371 Beckingham JD, Archibald JH (1996) Field guide to ecosites of northern Alberta. Natural Resources Canada, Canadian Forest Service, northwest region, northern forestry Centre, Edmonton, AB., special report 5 Borcard D, Gillete F, Legendre P (2011) Numerical ecology with R. Springer, New York, p 306 Burns RM, Honkala BH (1990). Silvics of North America, Vol. 2– Hardwoods, Agricultural Handbook 654. Forest Service, Unites States Department of Agriculture, Washington, DC Calvo-Polanco M, Jones MD, Zwiazek JJ (2009) Effects of pH on NaCl tolerance of American elm (Ulmus Americana) seedlings inoculated with Hebeloma crustuliniforme and Laccaria bicolor. Acta Physiol Plant 3:515–522 Calvo-Polanco M, Senorans J, Zwiazek JJ (2012) Role of adventitious roots in water relations of tamarack (Larix laricina) seedlings exposed to flooding. BMC Plant Biol 12:99–107 Calvo-Polanco M, Equiza MA, Señorans J, Zwiazek JJ (2014) Responses of rat root (Raf.) plants to salinity and pH conditions. J Environ Qual 43:578–586 Comerford NB (2005) Soil factors affecting nutrient bioavailability. In: BassiriRad H (ed) Nutrient acquisition by plants – an ecological perspective. Ecological studies, Vol. 181. Springer, Berlin, pp 1–14 Crawford NM, Glass AD (1998) Molecular and physiological aspects of nitrate uptake in plants. Trends Plant Sci 3:389–395 Epstein E (1972) Mineral nutrition of plants: principles and perspectives. Wiley, New York Ericsson T (1995) Growth and shoot: root ratio of seedlings in relation to nutrient availability. In: nutrient uptake and cycling in forest ecosystems (pp. 205-214). Springer, Netherlands Grant J, Dyer S, Woynillowicz D (2008) Fact or fiction: Oil Sands reclamation (p 75). The Pembina Institute, Edmonton Hayati AA, Proctor MCF (1990) Plant distribution in relation to mineral nutrient availability and uptake on a wet-heath site in south-west England. J Ecol 78:134–151 Hinsinger P, Plassard C, Tang C, Jaillard B (2003) Origins of rootmediated pH changes in the rhizosphere and their responses to environmental constraints: a review. Plant Soil 248:43–59 Howat D (2000) Acceptable salinity, sodicity and pH values for boreal forest reclamation. Alberta environment,

environmental sciences division, Edmonton Alberta. Report # ESD/LM/00-2 Iles J (2001) Community trees: community tree steward program. Iowa State University, Ames, Requirements for plant growth. Dept. Horticulture, p 4 Jarrell WM, Beverly RB (1981) The dilution effect in plant nutrition studies. Adv Agron 34:197–224 Jones JB Jr (2001) Laboratory guide for conducting soil tests and plant analysis. CRC Press, Boca Raton Kallenbach RL, McGraw RL, Beuselinck PR (1996) Soil pH effects on growth and mineral concentration of birdsfoot trefoil. Can J Plant Sci 76:263–267 Kerley SJ, Huyghe C (2001) Comparison of acid and alkaline soil and liquid culture growth systems for studies of shoot and root characteristics of white lupin (Lupinus albus L.) genotypes. Plant Soil 236:275–286 Kerley SJ, Huyghe C (2002) Stress-induced changes in the root architecture of white lupin (Lupinus albus) in response to pH, bicarbonate, and calcium in liquid culture. Ann Appl Biol 141:171–181 Kopittke PM, Menzies NW (2005) Effect of pH on Na induced ca deficiency. Plant Soil 269:119–129. doi:10.1007/s11104004-0395-0 Larcher W (2003) Physiological plant ecology, 4th edn. SpringerVerlag, Berlin, 450 pp Lupi C, Morin H, Deslauriers A, Rossi S, Houle D (2013) Role of soil nitrogen for the conifers of the boreal forest: a critical review. Intern J Plant Soil Sci 2:155–189 Macdonald SE, Landhäusser SM, Skousen J, Franklin J, Frouz J, Hall S, Douglass FJ, Quideau S (2015) Forest restoration following surface mining disturbance: challenges and solutions. New Forest 46:703–732 Marschner P (2012) Marschner's mineral nutrition of higher plants. Elsevier Ltd., Waltham Natural Regions Committee (2006) Natural regions and subregions of Alberta. Compiled by D.J.Downing and W.W. Pettapiece. Government of Alberta. Pub. No. T/852 Perala DA (1990) Populus tremuloides Michx. In: burns RM, Honkala BH (eds). Silvics of North America: volume 2. Hardwoods. Agriculture handbook 654. USDA, Forest Service, Washington, DC R Core Team (2015) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria https://www.R-project.org/ Renault S, Paton E, Nilsson G, Zwiazek J, MacKinnon M (1999) Responses of boreal plants to high salinity oil sands tailings water. J Env Qual 28:1957–1962 Shalhevet J, Huck MG, Schroeder BP (1995) Root and shoot growth responses to salinity in maize and soybean. Agron J 87:512–516 Tang C, Longnecker NE, Thomson CJ, Greenway H, Robson AD (1992) Lupin (Lupinus angustifolius L.) and pea (Pisum sativum L.) roots differ in their sensitivity to pH above 6.0. J Plant Physiol 140:715–719 Tang C, Cobley BT, Mokhtara S, Wilson CE, Greenway H (1993) High pH in the nutrient solution impairs water uptake in Lupinus angustifolius L. Plant Soil 155-156:517–519 Tang C, Longnecker NE, Greenway H, Robson AD (1996) Reduced root elongation of Lupinus angustifolius L. by high pH is not due to decreased membrane integrity of cortical cells or low proton production by the roots. Ann Bot 78:409–414

Plant Soil Thormann MN, Szumigalski AR, Bayley SE (1999) Aboveground peat and carbon accumulation potentials along a bog-fenmarsh wetland gradient in southern boreal Alberta, Canada. Wetlands 19:305–317 Yousfi S, Wissal M, Mahmoudi H, Abdelly C, Gharsalli M (2007) Effect of salt on physiological responses of barley to iron deficiency. Plant Physiol Biochem 45: 309–314 Zhang W, Zwiazek JJ (2016) Effects of root medium pH on root water transport and apoplastic pH in red-osier dogwood

(Cornus sericea) and paper birch (Betula papyrifera) seedlings. Plant Biol 18:1001–1007. doi:10.1111/plb.12483 Zhang W, Calvo-Polanco M, Chen ZC, Zwiazek JJ (2013) Growth and physiological responses of trembling aspen (Populus tremuloides), white spruce (Picea glauca) and tamarack (Larix laricina) seedlings to root zone pH. Plant Soil 373: 775–786 Zhang W, Xu F, Zwiazek JJ (2015) Responses of jack pine (Pinus banksiana) seedlings to root zone pH and calcium. Environ Exp Bot 111:32–41