Next Article in Journal
Effect of Nondestructive Evaluation of Veneers on the Properties of Laminated Veneer Lumber (LVL) from a Tropical Species
Next Article in Special Issue
Genetic Improvement of White Spruce Mechanical Wood Traits—Early Screening by Means of Acoustic Velocity
Previous Article in Journal
Wood Quality: A Perspective from New Zealand

Forests 2013, 4(2), 251-269; doi:10.3390/f4020251

Article
Specific Gravity of Hybrid Poplars in the North-Central Region, USA: Within-Tree Variability and Site × Genotype Effects
William L. Headlee 1, Ronald S. Zalesny Jr. 2,*, Richard B. Hall 1, Edmund O. Bauer 1, Bradford Bender 3, Bruce A. Birr 2, Raymond O. Miller 3, Jesse A. Randall 1 and Adam H. Wiese 2
1
Department of Natural Resource Ecology and Management, Iowa State University, 339 Science II, Ames, IA 50011, USA; E-Mails: wheadlee@iastate.edu (W.L.H.); rbhall@iastate.edu (R.B.H.); ebauer@charter.net (E.O.B.); randallj@iastate.edu (J.A.R.)
2
Institute for Applied Ecosystem Studies, US Forest Service, Northern Research Station, 5985 Highway K, Rhinelander, WI 54501, USA; E-Mails: bbirr@fs.fed.us (B.A.B.); awiese@fs.fed.us (A.H.W.)
3
Forest Biomass Innovation Center, Department of Forestry, Michigan State University, 6005 J Road, Escanaba, MI 49829, USA; E-Mails: benderb@msu.edu (B.B.); rmiller@anr.msu.edu (R.O.M.)
*
Author to whom correspondence should be addressed; E-Mail: rzalesny@fs.fed.us; Tel.: +1-715-362-1132; Fax: +1-715-362-1166.
Received: 18 March 2013; in revised form: 3 April 2013 / Accepted: 11 April 2013 /
Published: 23 April 2013

Abstract

: Specific gravity is an important consideration for traditional uses of hybrid poplars for pulp and solid wood products, as well as for biofuels and bioenergy production. While specific gravity has been shown to be under strong genetic control and subject to within-tree variability, the role of genotype × environment interactions is poorly understood. Most specific gravity reports are for a limited number of locations, resulting in a lack of information about the interactions between clones and sites over a wide range of climate and soil conditions. The objective of the current study was to characterize the effects of bole position, site, clone, and site × clone interactions for twelve hybrid poplar genotypes grown in Iowa, Minnesota, Wisconsin, and Michigan, USA. Observed specific gravities ranged from 0.267 to 0.495 (mean = 0.352 ± 0.001 for 612 samples taken from 204 trees), with bole position and site × clone interactions having significant effects on specific gravity. Further investigation of the site × clone interactions indicated that environmental conditions related to water stress were key predictors of specific gravity. These data are important for informing genotypic selection and silvicultural management decisions associated with growing hybrid poplars.
Keywords:
biomass; bole position; clones; correlation; feedstock quality; Populus; short rotation woody crops; tree growth; water stress; wood density

1. Introduction

Short rotation woody crops (SRWCs) such as Populus species and their hybrids (hereafter referred to as hybrid poplars) can be grown in strategic locations across the landscape to increase ecosystem services and maintain regional ecological sustainability [1]. These purpose-grown feedstock production systems provide regulating, provisioning, cultural, and supporting services as outlined in the Millenium Ecosystem Assessment [2], which are especially important for contributing benefits such as erosion control, soil health maintenance, and environmental remediation (i.e., the regulating services). For example, certain Populus hybrids are used as components of agronomic intercropping systems to stabilize soils and provide critical wildlife habitat [3,4]. Hybrid poplars are also used extensively in phytotechnologies to clean sites contaminated with inorganic and organic constituents [5,6]. Moreover, given their greater productivity levels relative to most other temperate-grown, deciduous trees [7], hybrid poplars are ideal for providing biomass for multiple end uses (i.e., the provisioning services).

Traditional uses of hybrid poplar biomass include fiber for the pulp and paper industry, as well as solid wood and chips for lumber and engineered wood products [8]. Hybrid poplar feedstocks are also critical components of current energy portfolios in North America [9,10,11], and have been used for energy in Europe for quite some time [12]. Hybrid poplar biomass is suitable for conversion to liquid transportation fuels via thermochemical and biochemical processes [13,14] and direct combustion during combined heat and power (CHP) operations [15]. Given the broad genetic variability within the genus Populus [16,17,18], certain genotypes exhibit wood property traits that are ideal for many of these uses, despite the fact that selection for one trait may reduce the quality and/or contribution of another. Physical traits that contribute to feedstock quality and, thus, the economic viability of growing SRWCs include but are not limited to: moisture content, bark:wood ratio, heating value, and specific gravity [19,20].

Specific gravity is often considered the most important feedstock characteristic for traditional products and energy applications [21,22,23,24]. For example, selection for hybrid poplar feedstocks with high densities and strengths may contribute to elevated pulp yields, while medium densities are more appropriate for oriented strand board (OSB) and plywood [8,25,26]. Likewise, high density wood is a requirement for thermochemical conversion and direct combustion technologies, whereby specific gravity and process efficiency are positively correlated [19,20]. Similarly, greater specific gravities increase transportation efficiencies because more biomass per truckload is delivered to processing facilities [22]. Overall, hybrid poplar specific gravity estimates reported in the literature (since 1979) range from 0.26 to 0.50 (Table 1).

Although yield has been the primary focus of most hybrid poplar tree improvement programs [8,27], the aforementioned benefits of specific gravity for multiple end uses have ensured it has remained one of the key traits during breeding. One of the challenges during selection is the relationship between specific gravity and position along the bole of the tree, which has been somewhat inconsistent with tree age, site, and genotype [26,28,29,30]. Likewise, direct selection for specific gravity has at times been counterproductive to yield objectives given an inverse relationship between wood density and growth rate [23,24,25,28,31,32]. Although specific gravity of hybrid poplar feedstocks has been shown to be under strong genetic control [24,25,28,31,33], it is important to understand the role of genotype × environment interactions. Most specific gravity reports include numerous hybrid poplar genotypes but were conducted at a limited number of locations (i.e., ≤4), resulting in a lack of information about the mechanisms and causes of specific gravity differences among genotypes tested across multiple sites differing in climate and soils.

Table Table 1. Summary of published studies testing specific gravity of Populus, since 1979.

Click here to display table

Table 1. Summary of published studies testing specific gravity of Populus, since 1979.
Study LocationAge (yrs)Genomic Group 1Clones (#)Specific GravityReference
Gorgan, Iran22D × N10.33–0.36Kord et al. [26]
Quebec, Canada10, 12D40.37–0.39Pliura et al. [24]
T × D30.34–0.36
D × N40.32–0.34
M × B40.29–0.31
B × N40.31–0.34
Overall190.29–0.45
Quebec, Canada10D30.36–0.37Pliura et al. [23]
T × D30.33–0.34
D × N30.32–0.35
M × B30.28–0.32
Overall120.26–0.38
New York, USA3D × N50.33–0.37Tharakan et al. [22]
N × M20.34–0.36
Quebec, Canada3Overall 2210.27–0.48Zhang et al. [33]
Washington, USA9T10.31–0.46DeBell et al. [30]
T × D20.29–0.47
Iowa, USA4A × A10.32–0.42Semen et al. [34]
A × G10.29–0.35
A × Tr10.36–0.37
Kansas, USA4D90.34–0.40 3Geyer et al. [29]
D × N20.35–0.37
Quebec, Canada9D × N100.30–0.40Hernández et al. [32]
Hungary10, 15D × N30.30–0.36Mátyás and Peszlen [21]
Ontario, Canada12B900.29–0.41Ivkovich [31]
Quebec, Canada9D × N100.28–0.41Beaudoin et al. [28]
Pennsylvania, USA4M × T10.39–0.46Blankenhorn et al. [35]
Mississippi Valley, USA3D750.27–0.39Olson et al. [25]
Pennsylvania, USA4M10.41–0.48Murphey et al. [36]
T10.40–0.50
M × T10.39–0.49

1 Populus species comprising the genomic groups were: (A) P. alba L.; (B) P. balsamifera L.; (D) P. deltoides Bartr. ex. Marsh; (G) P. grandidentata Michx.; (M) P. maximowiczii A. Henry (now considered a subspecies of P. suaveolens Fischer); (N) P. nigra L.; (T) P. trichocarpa Torr. & Gray; (Tr) P. tremula L; 2 A total of 21 clones were tested from the following genomic groups (individual values for each were not explicitly stated): D (1 clone); T × D (3); D × N (10); M × B (4); D × M (2); N × M (1); 3 29 clones were tested; these values are from 11 clones reported in Table 5.

The overarching objective of the current study was to characterize specific gravity among hybrid poplar genotypes grown in Iowa, Minnesota, Wisconsin, and Michigan, USA. The plantation network consisted of ten genotypes grown at four sites for ten years, two genotypes grown at eleven sites for twenty years, and one genotype grown at two sites for fifteen years (Figure 1). Specific objectives for each set of plantations included: (i) testing for differences in specific gravity among three bole positions (diameter at breast height; 1/3 tree height; 2/3 tree height) to account for variability in age of the wood; (ii) testing for differences in specific gravity among genotypes, sites, and their interactions; and (iii) assessing the effects of climate and soils on specific gravity differences observed in (ii). These data are important for informing genotypic selection and silvicultural management decisions associated with growing hybrid poplars to increase ecosystem services across the landscape.

2. Experimental Section

2.1. Site and Clone Selection

Fifteen of the seventeen sites for the current study were selected from two regional networks of hybrid poplar clone and yield trials established in the north-central United States during 2000 to 2001 (10-year-old plantations) [27,37] and 1988 to 1991 (20-year-old plantations) [38,39,40]. The remaining two sites (15-year-old plantations) were established in 1995 as part of the woody biomass feedstock development program at Iowa State University (Figure 1). Table 2 highlights main characteristics for each site, including climate and soils information. Site latitude and longitude (decimal degrees) were determined from GPS coordinates taken at the sites. Soil texture, available soil water holding capacity (ASW; cm) for the top 100 cm, and depth to water table (WT; cm) were determined for each site using the Natural Resources Conservation Service (NRCS) Web Soil Survey [41]. Mean annual precipitation (P; mm) and mean growing season (April to October) temperature maximum (Tmax; °C) and minimum (Tmin; °C) were determined from the National Oceanic and Atmospheric Administration (NOAA) National Climatic Data Center using 30-year climate averages (1981 to 2010) from the weather station nearest to each site [42]. From Tmax and Tmin, average temperature (Tavg) and temperature differential (Tdiff) were calculated (Tavg = (Tmax + Tmin)/2; Tdiff = TmaxTmin).

Forests 04 00251 g001 1024
Figure 1. Plantation networks in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA.

Click here to enlarge figure

Figure 1. Plantation networks in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA.
Forests 04 00251 g001 1024
Table Table 2. Characteristics of plantations in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA.

Click here to display table

Table 2. Characteristics of plantations in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA.
State 1SiteYear PlantedHeight (m) 2Lat (°N) 3Long (°W) 3Soil TextureASW (cm) 4WT (cm) 4P (mm) 5Tavg (°C) 5Tdiff (°C) 5
IAAmes200015.1 ± 0.442.0593.66Fine Sandy Loam17.36122880.617.012.6
IAKanawha199518.7 ± 0.342.9393.80Clay Loam18.650849.415.811.9
IASutherland199518.5 ± 0.642.9395.54Silty Clay Loam20.400780.316.313.9
MIEscanaba200112.5 ± 0.245.7787.20Fine Sandy Loam15.01>200728.212.611.1
MNBelgrade199017.2 ± 0.445.6795.11Loam18.2076653.015.314.3
MNBemidji198817.9 ± 0.347.5894.93Loamy Sand10.39>200676.412.711.8
MNFairmont198818.7 ± 0.243.6994.35Clay Loam18.3515830.816.511.4
MNGranite Falls198721.4 ± 0.844.8095.52Loam19.0975726.715.312.8
MNLamberton198818.7 ± 0.944.2595.29Clay Loam18.0015709.715.613.3
MNMilaca198919.4 ± 0.345.7893.63Silt Loam18.3915748.314.112.9
MNUlen198914.4 ± 0.547.0996.18Loam 16.9015628.114.313.2
MNWarren198920.8 ± 0.748.1496.65Fine Loamy Sand9.7360547.613.414.3
MNWaseca200016.3 ± 0.444.0693.54Clay Loam18.3645907.315.911.9
WIArlington200018.0 ± 0.343.2989.37Silt Loam21.6461869.214.713.9
WILancaster199122.5 ± 1.342.8290.79Silt Loam15.23>200898.415.510.9
WIMondovi198819.3 ± 0.344.5291.65Silt Loam21.12>200881.115.415.0
WIRhinelander198821.5 ± 0.645.6389.46Loamy Sand7.4361675.413.012.9

1 IA, Iowa; MI, Michigan; MN, Minnesota; WI, Wisconsin; 2 Mean height (± one standard error) of trees tested at each site (n = 4 to 39); 3 Lat, latitude; Long, longitude; 4 ASW, maximum available soil water for the top 100 cm; WT, depth to water table. Soil data (including texture) were obtained from the Natural Resources Conservation Service (NRCS) Web Soil Survey [41]; 5 P, mean annual precipitation; Tavg, mean growing season temperature (April to October); Tdiff, difference between maximum and minimum growing season temperatures (Tmax and Tmin, not reported in the table). Climate data were obtained from the National Oceanic and Atmospheric Administration (NOAA) National Climatic Data Center using 30-year climate averages from 1981 to 2010 [42].

In total, twelve hybrid poplar clones belonging to five genomic groups were tested (Table 3). Ten clones were selected from the 10-year-old network based on availability at two or more of the four sites during initial surveys; however, some clones were not available at the time of harvest. Two clones were selected from the 20-year-old plantations based on tree health surveys conducted by Zalesny et al. [1]. The single hybrid aspen clone (i.e., Crandon) from the 15-year-old sites was chosen based on current interest of assessing specific gravity of this genomic group for energy conversion and previous work by the research team on kraft pulp [34]. Table 3 also lists genomic groups and clones tested for each plantation network.

Table Table 3. Hybrid poplar genomic groups and clones sampled at each site in a study testing location and genotype effects on specific gravity in the north-central region, USA 1. Site × clone combinations shaded in black were included in the original study but did not have requisite sample sizes for the current analyses.

Click here to display table

Table 3. Hybrid poplar genomic groups and clones sampled at each site in a study testing location and genotype effects on specific gravity in the north-central region, USA 1. Site × clone combinations shaded in black were included in the original study but did not have requisite sample sizes for the current analyses.
Forests 04 00251 i001

1 See Section 2.4 for a description of color designations; 2 See Table 1 for species authorities.

2.2. Field and Laboratory Sampling

In the field, a maximum of four dominant trees per clone were felled and total tree height was determined to the nearest 0.1 m at each site. Cross-sectional disks were harvested from each bole at 1.4-m height (i.e., diameter at breast height, DBH), 1/3 height of the tree, and 2/3 height of the tree. Disks were sealed in plastic bags to avoid potential shrinkage, before transporting them for processing to the analytical laboratory at the Institute for Applied Ecosystem Studies in Rhinelander, WI, USA. Diameter to the nearest 0.1 cm and fresh mass to the nearest 0.1 g were determined for each disk. The disks were then oven-dried at 55 °C until constant mass, which was recorded with the same precision as fresh mass. One cross-sectional area of each disk was sanded, wetted, and imaged for ongoing studies. The sanded disks were then cut in half along a plane extending through the pith. A wedge, free of bark and defects, was cut from the half-disk beginning at the pith and extending to its outside edge.

2.3. Specific Gravity Measurements

Specific gravity was determined from each wedge according to the ratio between oven dry mass and green (saturated) volume. Specifically, each wedge was immersed in water for 24 to 36 h and placed into a water-filled desiccator under vacuum until they sank. After sinking, excess moisture was removed from all surfaces of each wedge, which was then placed into a beaker of water atop an analytical balance, and the displacement volume was determined to the nearest 0.1 g. Following green volume determination, wedges were oven dried at 105 °C until constant mass, which was determined to the nearest 0.1 g.

2.4. Experimental Design and Data Analysis

Specific gravity data from 204 trees were assigned to one of three datasets: (i) 10-year-olds having a complete factorial design for sites and clones (depicted as light green with down-diagonal line in Table 3); (ii) 20-year-olds having a complete factorial design for sites and clones (depicted as dark green in Table 3); and (iii) a mixed dataset of 10-, 15-, and 20-year-olds having an incomplete factorial design (i.e., lacking at least one site × clone combination required for a complete factorial of sites and clones; depicted as gray with up-diagonal line in Table 3). A summary of the datasets is shown in Table 4. For all datasets, only sufficiently represented clones (≥2 trees at a given site) were included in the analyses.

Table Table 4. Number of sites, clones, and site × clone combinations represented in the 10- and 20-year-old factorials, and mixed (10-, 15-, and 20-year-old) non-factorial analyses. Each site × clone combination was represented by two to four trees.

Click here to display table

Table 4. Number of sites, clones, and site × clone combinations represented in the 10- and 20-year-old factorials, and mixed (10-, 15-, and 20-year-old) non-factorial analyses. Each site × clone combination was represented by two to four trees.
DatasetNumber of:
TreesSites 1Clones 1Site × Clone Combinations
10-year-old factorial594416
20-year-old factorial719218
Mixed (10-, 15-, and 20-year-old) non-factorial 2748721

1 See Table 3 for specific site × clone combinations; 2 For the mixed dataset, each clone was missing from 1 or more sites, and therefore no factorial analyses of site and clone were conducted. Thus, only differences between site × clone combinations were evaluated.

All three datasets were analyzed as split-plot designs. Site, clone and site × clone effects were evaluated at the main-plot level (with tree as the experimental unit), and bole position was evaluated at the split-plot level (with wedge as the experimental unit), for the 10-year-old and 20-year-old datasets. Site and clone were not included as treatment factors for the mixed dataset due to each clone being absent from at least one site; instead, the data were analyzed for differences among existing site × clone combinations at the main-plot level, with position being evaluated at the split-plot level. The data were analyzed using PROC MIXED (method = type 3) in SAS (SAS Institute, Inc., Cary, NC, USA), with denominator degrees of freedom calculated via the Kenwood-Rogers method [43]. When significant main effects and interactions were found (p < 0.05), multiple comparisons analyses with Tukey’s adjustment [43] were conducted to identify significant differences (p < 0.05) among least-squares means. In the event a significant interaction containing a significant main effect was found, the interaction was evaluated rather than the main effect.

3. Results and Discussion

3.1. ANOVA Results

For the 10-year-old factorial, specific gravity was significantly influenced by position, clone, and site × clone interactions (Table 5). Multiple comparisons analysis of the position effect showed no significant difference between DBH (0.327 ± 0.003) and 1/3 height (0.329 ± 0.003), while 2/3 height was significantly higher than both of these (0.344 ± 0.003). Multiple comparisons analysis of site × clone effects showed the specific gravities for clones NC13563 and C916000 were consistently lowest and second-lowest, respectively, across all sites (Figure 2). Clone C918001 had the highest specific gravity at Escanaba and Arlington (where it was significantly higher than NC13563 and C916000), while C916400 was highest at Ames (where it was significantly higher than NC13563 and C916000) and Waseca (where it was significantly higher than all the other clones).

Table Table 5. Probability values from analyses of variance comparing specific gravity of hybrid poplar clones grown at 17 sites throughout the north-central region, USA. Three bole positions were also tested. See Experimental Section for descriptions of all fixed effects. Significant values are in bold.

Click here to display table

Table 5. Probability values from analyses of variance comparing specific gravity of hybrid poplar clones grown at 17 sites throughout the north-central region, USA. Three bole positions were also tested. See Experimental Section for descriptions of all fixed effects. Significant values are in bold.
Source of Variationp-Value
------- 10-year-old factorial -------
Site0.4011
Clone<0.0001
Site × Clone0.0078
Position0.0002
Position × Site0.1293
Position × Clone0.5963
Position × Site × Clone0.5212
------- 20-year-old factorial -------
Site<0.0001
Clone0.4529
Site × Clone0.0430
Position<0.0001
Position × Site0.0010
Position × Clone0.0183
Position × Site × Clone0.1918
------- Mixed (non-factorial) -------
Site × Clone<0.0001
Position<0.0001
Position × Site × Clone0.1213
Forests 04 00251 g002 1024
Figure 2. Specific gravity of four hybrid poplar clones grown at four sites in the north-central region, USA. Each bar represents the mean of two to four trees with one standard error, according to the factorial analysis of 10-year-old plantations. Bars labeled with different letters are different at p < 0.05.

Click here to enlarge figure

Figure 2. Specific gravity of four hybrid poplar clones grown at four sites in the north-central region, USA. Each bar represents the mean of two to four trees with one standard error, according to the factorial analysis of 10-year-old plantations. Bars labeled with different letters are different at p < 0.05.
Forests 04 00251 g002 1024

For the 20-year-old factorial, specific gravity was significantly influenced by position, position × clone, position × site, site, and site × clone (Table 5). Multiple comparisons analysis of the position × clone effect showed all positions were significantly different for DN34 and increased from DBH (0.353 ± 0.003) to 1/3 height (0.364 ± 0.003) to 2/3 height (0.385 ± 0.003). Similarly, all positions were significantly different for DN182 and increased from DBH (0.346 ± 0.003) to 1/3 height (0.367 ± 0.003) to 2/3 height (0.396 ± 0.003). The differences between the two clones were not significant at DBH or 1/3 height, but were significant at 2/3 height. For the position × site effect, multiple comparisons analysis showed that specific gravity at 2/3 height was significantly greater than DBH (with 1/3 height being intermediate) for eight of the nine sites (Table 6). The nature of the intermediate specific gravities at 1/3 height varied by site and can be summarized as follows: (i) significantly lower than at 2/3 height but not significantly different than at DBH (Bemidji, Granite Falls, Mondovi, Rhinelander); (ii) not significantly different than at 2/3 height but significantly higher than at DBH (Fairmont, Lamberton); (iii) significantly lower than at 2/3 height while also significantly higher than at DBH (Belgrade, Milaca). One site showed no significant position effects (Warren). Multiple comparisons analysis of site × clone effects showed the specific gravities for clones DN34 and DN182 at Mondovi were significantly lower than DN34 at Belgrade and DN182 at Fairmont and Lamberton (Figure 3). Clones DN182 at Granite Falls and DN34 at Rhinelander were also significantly lower than DN182 at Fairmont.

For the mixed non-factorial analysis, specific gravity was significantly influenced by position and site × clone interactions (Table 5). Multiple comparisons analysis of the position effect showed no significant difference between DBH (0.335 ± 0.003) and 1/3 height (0.338 ± 0.003), with 2/3 height being significantly higher than both of these (0.366 ± 0.003). For site × clone effects, multiple comparisons analysis showed a general trend of Crandon and DN34 having the highest specific gravities across available sites, with NC13649 and NC13624 exhibiting the lowest specific gravities, and the remaining clones (NC14018, NM2, and NM6) being intermediate (Figure 4).

Table Table 6. Least-squares means of specific gravity (± one standard error) for each combination of site and bole position from the 20-year-old factorial analysis of variance in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA. Combinations with different letters are significantly different at p < 0.05.

Click here to display table

Table 6. Least-squares means of specific gravity (± one standard error) for each combination of site and bole position from the 20-year-old factorial analysis of variance in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA. Combinations with different letters are significantly different at p < 0.05.
SitePosition 1Specific Gravity
Belgrade2/30.412 ± 0.007 a
Fairmont2/30.405 ± 0.007 ab
Milaca2/30.400 ± 0.007 abc
Fairmont1/30.397 ± 0.007 abcd
Bemidji2/30.396 ± 0.007 abcde
Rhinelander2/30.395 ± 0.007 abcde
Lamberton2/30.387 ± 0.007 bcdef
Granite Falls2/30.380 ± 0.008 cdefg
Lamberton1/30.379 ± 0.007 defg
Warren2/30.377 ± 0.007 efgh
WarrenDBH0.372 ± 0.007 fghi
Belgrade1/30.371 ± 0.007 fghi
Milaca1/30.370 ± 0.007 fghij
Warren1/30.366 ± 0.007 ghijk
Mondovi2/30.364 ± 0.007 ghijkl
Bemidji1/30.362 ± 0.007 ghijklm
LambertonDBH0.359 ± 0.007 hijklmn
FairmontDBH0.358 ± 0.007 hijklmn
Granite Falls1/30.353 ± 0.008 ijklmno
Rhinelander1/30.353 ± 0.007 ijklmno
BemidjiDBH0.351 ± 0.007 jklmno
BelgradeDBH0.346 ± 0.007 klmno
MilacaDBH0.345 ± 0.007 lmno
Granite FallsDBH0.342 ± 0.008 mno
RhinelanderDBH0.340 ± 0.007 no
Mondovi1/30.339 ± 0.007 no
MondoviDBH0.333 ± 0.007 o

1 DBH, diameter at breast height (1.4 m); 1/3, one-third height of the tree; 2/3, two-thirds height of the tree.

Forests 04 00251 g003 1024
Figure 3. Specific gravity of hybrid poplar clones DN34 and DN182 grown at nine sites in the north-central region, USA. Each bar represents the mean of two to four trees with one standard error, according to the factorial analysis of 20-year-old plantations. Bars labeled with different letters are different at p < 0.05.

Click here to enlarge figure

Figure 3. Specific gravity of hybrid poplar clones DN34 and DN182 grown at nine sites in the north-central region, USA. Each bar represents the mean of two to four trees with one standard error, according to the factorial analysis of 20-year-old plantations. Bars labeled with different letters are different at p < 0.05.
Forests 04 00251 g003 1024
Forests 04 00251 g004 1024
Figure 4. Specific gravity of seven hybrid poplar clones grown at eight sites in the north-central region, USA. Each bar represents the mean of two to four trees with one standard error, according to the mixed (nested) analysis of 10-, 15-, and 20-year-old plantations. Bars labeled with different letters are different at p < 0.05. Bars are shaded according to genomic group, as defined in Table 3.

Click here to enlarge figure

Figure 4. Specific gravity of seven hybrid poplar clones grown at eight sites in the north-central region, USA. Each bar represents the mean of two to four trees with one standard error, according to the mixed (nested) analysis of 10-, 15-, and 20-year-old plantations. Bars labeled with different letters are different at p < 0.05. Bars are shaded according to genomic group, as defined in Table 3.
Forests 04 00251 g004 1024

3.2. Interpretation of Position Effects

The effects of bole position on specific gravity vary widely in the literature, likely due in part to differences in the relative bole positions sampled in different studies. However, some general observations can be made. For example, a trend of specific gravity decreasing with increasing height has been reported [26]. Others have reported decreasing specific gravity in mid-heights but similar densities at low and high bole positions [28,29,30]. Geyer et al. [29] found that branch wood (analogous to the 2/3 height position in this study) had significantly higher specific gravity than DBH wood, which corroborates the general trend observed for all three datasets in this study. Specifically, the 10-year-old and mixed datasets showed significantly higher specific gravities at 2/3 height than at DBH across all sites and clones, and the position × site effect for the 20-year-old dataset showed the same trend at eight of the nine sites evaluated. Previous research with hybrid poplars found that wood near the pith had higher specific gravity than subsequent growth rings [28]. Thus, the higher proportion of near-pith wood at 2/3 height (relative to that at DBH) likely explains the higher specific gravities observed in this study.

3.3. Interpretation of Site × Clone Interactions Using Site Covariates

To better understand the significant site × clone effects observed in this study, readily-available site variables were evaluated as covariates for the least-squares means of specific gravity: latitude (L), available soil water holding capacity (ASW), water table depth (WT), annual precipitation (P), growing season average temperature (Tavg), and growing season temperature differential (Tdiff); see Table 2 for descriptions and values. In addition, mean growth rate (cm·year−1) based on outside bark diameter (OBD) at breast height for each clone at each site was evaluated as a covariate. Using PROC GLM in SAS, the effects of clone, each covariate, and their interactions were analyzed for statistical significance across all 204 trees from all three datasets. Non-significant factors were removed in a stepwise fashion until only significant factors (p < 0.05) remained. The resulting model indicated significant effects of clone (p < 0.0001), Tdiff (p = 0.0272), and clone × WT (p = 0.0494) on specific gravity, as well as a strong model fit (R2 = 0.89).

The coefficient for Tdiff was negative (−0.00366), indicating that the trees produced wood of lower specific gravity under higher temperature differentials. This is consistent with previous research showing lower specific gravities of wood produced under sub-optimal conditions [24] and particularly conditions of water stress [44], presumably due to reduced cell wall thickness. This water stress hypothesis is further supported by analysis of the slopes associated with the clone × WT interaction (Table 7). For clones C916400, DN182, and DN34 (10-year-olds) the slopes are all negative and significantly different from 0, which indicates that deeper water tables (and presumably greater water stress) are associated with lower specific gravities for these clones. For the remaining clones, the slopes were found to be not significantly different from 0, meaning the clones were not significantly affected by water table depth. Interestingly, the slope for the 10-year-old DN34 was significantly different from 0, whereas that for the 20-year-old DN34 was not. This suggests that the effects of WT on some clones may decrease with age, possibly as the trees grow larger and develop root systems capable of reaching deeper water tables. Thus, genetic and/or developmental differences in rooting depth may explain why some clones appear more sensitive to WT than others. However, further study is recommended to specifically test these hypotheses.

Table Table 7. Probability values from analyses of variance testing the hypothesis that slopes are equal to zero for the clone × depth to water table (WT) interaction, by clone, in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA. Significant values are in bold.

Click here to display table

Table 7. Probability values from analyses of variance testing the hypothesis that slopes are equal to zero for the clone × depth to water table (WT) interaction, by clone, in a study testing site and genotype effects on specific gravity of hybrid poplars in the north-central region, USA. Significant values are in bold.
CloneAge (y)Slope(Clone × WT)P(Slope = 0)
C916000100.0000150.8636
C91640010−0.0001860.0394
C91800110−0.0000380.6669
DN34 10−0.0002340.0411
DN3420−0.0000700.1084
DN18220−0.0001380.0105
NC13563100.0001240.1625
NC1362410−0.0000620.5159
NC1364910−0.0000580.5440
NC1401810−0.0001390.1216
NM2100.0000310.7426
NM610−0.0000390.6836

In the ANOVA of the 10-year-old dataset (Figure 2), the effects of Tdiff, and clone × WT are apparent when comparing C916400 (a WT-sensitive genotype) with the remaining (non-WT-sensitive) clones. The non-WT-sensitive clones had their highest (NC13563, C916000) or second-highest (C918001) specific gravities at Escanaba, which had the lowest Tdiff (11.1 °C) of the 10-year-old sites (Table 2). In contrast, the WT-sensitive clone C916400 had its highest specific gravities at Waseca (which had the shallowest WT at 45 cm) and its lowest values at Escanaba (which had the deepest WT at >200 cm). Similarly, in the ANOVA of the 20-year-old dataset (Figure 3), two of the three highest specific gravities were observed for DN182 (a WT-sensitive genotype) at Fairmont and Lamberton (tied for shallowest WT at 15 cm). The lowest specific gravities were observed for DN182 and DN34 at Mondovi, which had both the deepest WT (tied with Bemidji at >200 cm) and the highest Tdiff (15.0 °C). For the mixed dataset, the only WT-sensitive genotype (10-year-old DN34) had lower specific gravities at Escanaba (WT >200 cm) than at Arlington (WT = 61 cm) and Ames (WT = 122 cm).

3.4. Correlations between Specific Gravity and Growth Rate

One of the most challenging aspects of selection for specific gravity is its negative relationship with growth rate, with correlations ranging from −0.66 to −0.28 [23,24,25,28,31,32]; however, decades of breeding for improved genotypes have resulted in some progress towards closing this gap. For example, DeBell et al. [30] reported growth rate did not influence wood density for three hybrid poplar clones in western Washington, USA. Likewise, Zhang et al. [33] reported wood density was not significantly correlated with growth for 21 hybrid poplar clones at two sites in southern Quebec, Canada. As alluded to above, OBD growth rate was not found to be a significant covariate for specific gravity in this study. Similarly, the correlation between OBD and specific gravity was found to be weak (though statistically significant) across all datasets (r = −0.41; p < 0.0001), as well as for individual analyses: (i) 10-year-old dataset (r = −0.24; p = 0.0016); (ii) 20-year-old dataset (r = −0.35; p < 0.0001); (iii) mixed dataset (r = −0.34; p < 0.0001). The weak nature of this relationship may logically be attributable to some of the factors which influence growth rate having opposing effects on specific gravity. For instance, both decreased water stress and increased nitrogen availability would be expected to increase growth rate; however, the former would be expected to result in higher specific gravities (as demonstrated in this study) and the latter would be expected to result in lower specific gravities (as demonstrated by Pitre et al. [45] and Hacke et al. [46]). Thus, the specific factors dictating growth rate at a given site (e.g., climate, soils, nutrient management, etc.) should be expected to be more informative for explaining differences in specific gravities than growth rate itself.

3.5. Genetic Trends in Specific Gravity

In general, genetic control of specific gravity was moderate to high in the reported literature, with broad sense heritabilities or repeatabilities ranging from 0.45 to 0.92 [24,25,28,31,33]. More specifically, common trends existed across taxonomic sections and genomic groups, with Aigeiros genotypes (P. deltoides and P. nigra parentage) exhibiting similar densities to those belonging to the Populus section (P. alba, P. grandidentata, P. tremula), which were both approximately 9% less than the Tacamahaca species (P. balsamifera, P. trichocarpa, P. maximowiczii) (Table 1). One exception was reported at the species level [23,24], whereby genomic groups with P. balsamifera parentage exhibited uncharacteristically low maximum specific gravities (≤0.32) relative to their Tacamahaca counterparts (≤0.42). Although the design of the current study did not warrant heritability estimation or specific testing of these observations, trends associating specific gravity with taxonomic sections and genomic groups existed. In contrast to previous studies, our genotypes with Tacamahaca parentage exhibited the lowest overall specific gravities (~0.359), while Aigeiros (~0.372) and Populus (~0.386) differed observationally from Tacamahaca and one another. These results are likely due to the fact that the number of genotypes per genomic group was limited in the current study. For example, the high ranking of the Populus section was attributed to one clone, Crandon (P. alba × P. grandidentata). Other hybrids within the section Populus have exhibited much lower densities [34]. In addition, the Tacamahaca genotypes tested were not ideal for these comparisons. The first two clones, NM2 and NM6, were typically the most productive in the north-central region for traditional applications [27,37]; therefore, the negative correlation between specific gravity and growth rate reported above may have accounted for the majority of the responses observed (i.e., fast growth contributed to low density wood). The second Tacamahaca genomic group ((P. trichocarpa × P. deltoides) × P. deltoids) contained, at most, 25% of its alleles from P. trichocarpa, which may not be a true representation of the species as reported elsewhere [30,36]. Nevertheless, given the use of molecular genetics technologies, detailed studies with requisite designs for testing these relationships would be very beneficial for hybrid poplar production systems.

4. Conclusions

Specific gravity of individual trees in the current study ranged from 0.267 to 0.495, with a mean of 0.352 ± 0.001 (n = 612 samples from 204 trees). These values corroborated previous reports for hybrid poplars (Table 1). The variation in specific gravity in the current study was primarily attributed to three factors. First, within-tree variability associated with bole position from which stem disks were harvested showed that specific gravity increased with increasing tree heights. Second, genotype × environment interactions significantly influenced specific gravity for 10-, 15-, and 20-year-old plantations. In particular, site characteristics associated with increasing water stress resulted in decreased specific gravities. The two most important factors controlling this response were depth to water table and the differential between mean maximum and minimum temperature for the growing season (April to October). Third, specific gravity was tightly related to taxonomic sections, species, and genomic groups. In contrast to previous studies, however, Tacamahaca genotypes exhibited lower densities relative to Aigeiros or Populus, with the latter having the highest specific gravities. Trees in the current study also exhibited a weak, inverse relationship between specific gravity and growth rate. Overall, these results are important for informing tree improvement decisions related to the choice of parental material and subsequent selection for yield versus wood quality traits, especially for provisioning ecosystem services such as woody biomass for pulp and paper, solid wood products, and biofuels/bioenergy.

Acknowledgments

This study was funded by the U.S. Forest Service Research and Development Washington Office Woody Biomass, Bioenergy, and Bioproducts Program, as well as the U.S. Forest Service Northern Research Station Climate Change Science Council and the Institute for Applied Ecosystem Studies (RWU-NRS-13). We thank the following research collaborators for access to their field sites: Gregg Johnson (Waseca; University of Minnesota), Glen Stanosz (Arlington; University of Wisconsin), Jeff Strock (Lamberton; University of Minnesota), and Tim Wood (Lancaster; University of Wisconsin). In addition, we are grateful to the private landowners who let us harvest their trees on the remaining sites. We also thank Sue Lietz for creating Figure 1; Kricket Koehn for laboratory assistance; David Coyle and J. Y. Zhu for review of earlier versions of the manuscript.

Conflict of Interest

The authors declare no conflict of interest.

References

  1. Zalesny, R.S., Jr.; Donner, D.M.; Coyle, D.R.; Headlee, W.L. An approach for siting poplar energy production systems to increase productivity and associated ecosystem services. For. Ecol. Manag. 2012, 284, 45–58. [Google Scholar] [CrossRef]
  2. Millennium Ecosystem Assessment (MEA). In Ecosystems and Human Well-Being: Synthesis; Island Press: Washington, DC, USA, 2005; p. 155.
  3. Thelemann, R.; Johnson, G.; Sheaffer, C.; Banerjee, S.; Cai, H.; Wyse, D. The effect of landscape position on biomass crop yield. Agron. J. 2010, 102, 513–522. [Google Scholar] [CrossRef]
  4. Headlee, W.L. Productivity and Biometry of Hybrid Poplars with Respect to Establishment, Regeneration, Regional Modeling, and Utilization of Bio-Energy Byproducts. Ph.D. Thesis, Iowa State University, Ames, IA, USA, 2012. [Google Scholar]
  5. Burken, J.G.; Schnoor, J.L. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 1998, 32, 3379–3385. [Google Scholar] [CrossRef]
  6. Zalesny, J.A.; Zalesny, R.S., Jr.; Wiese, A.H.; Sexton, B.T.; Hall, R.B. Uptake of macro- and micro-nutrients into leaf, woody, and root tissue of Populus after irrigation with landfill leachate. J. Sust. For. 2008, 27, 303–327. [Google Scholar] [CrossRef]
  7. Headlee, W.L.; Zalesny, R.S., Jr.; Donner, D.M.; Hall, R.B. Using a process-based model (3-PG) to predict and map hybrid poplar biomass productivity in Minnesota and Wisconsin, USA. Bioenergy Res. 2013, 6, 196–210. [Google Scholar] [CrossRef]
  8. Stanton, B.; Eaton, J.; Johnson, J.; Rice, D.; Schuette, B.; Moser, B. Hybrid poplar in the Pacific Northwest: The effects of a market-driven management. J. For. 2002, 100, 28–33. [Google Scholar]
  9. Johnson, J.M.F.; Coleman, M.D.; Gesch, R.; Jaradat, A.; Mitchell, R.; Reicosky, D.; Wilhelm, W.W. Biomass-bioenergy crops in the United States: A changing paradigm. Am. J. Plant Sci. Biotechnol. 2007, 1, 1–28. [Google Scholar]
  10. U.S. Department of Energy. U.S. Billion-Ton Update: Biomass Supply for a Bioenergy and Bioproducts Industry; Perlack, R.D., Stokes, B.J., Eds.; Oak Ridge National Laboratory: Oak Ridge, TN, USA, 2011; p. 227.
  11. Langholtz, M.; Graham, R.; Eaton, L.; Perlack, R.; Hellwinkel, C.; de la Torre Ugarte, D. Price projections of feedstocks for biofuels and biopower in the U.S. Energy Pol. 2012, 41, 484–493. [Google Scholar] [CrossRef]
  12. Berndes, G.; Hoogwijk, M.; van den Broek, R. The contribution of biomass in the future global energy supply: A review of 17 studies. Biomass Bioenergy 2003, 25, 1–28. [Google Scholar] [CrossRef]
  13. Zhu, J.Y.; Pan, X.; Zalesny, R.S., Jr. Pretreatment of woody biomass for biofuel production: Energy efficiency, technologies, and recalcitrance. Appl. Microbiol. Biotechnol. 2010, 87, 847–857. [Google Scholar] [CrossRef]
  14. Wang, Z.J.; Zhu, J.Y.; Zalesny, R.S., Jr.; Chen, K.F. Ethanol production potential from poplar wood through enzymatic saccharification and fermentation. Fuel 2012, 95, 606–614. [Google Scholar] [CrossRef]
  15. Jenkins, B.M.; Baxter, L.L.; Miles, T.R., Jr.; Miles, T.R. Combustion properties of biomass. Fuel Proc. Technol. 1998, 54, 17–46. [Google Scholar] [CrossRef]
  16. Weber, J.C.; Stettler, R.F.; Heilman, P.E. Genetic variation and productivity of Populus trichocarpa and its hybrids. I. Morphology and phenology of 50 native clones. Can. J. For. Res. 1985, 15, 376–383. [Google Scholar] [CrossRef]
  17. Rajora, O.P.; Zsuffa, L. Allozyme divergence and evolutionary relationships among Populus deltoides, P. nigra, and P. maximowiczii. Genome 1990, 33, 44–49. [Google Scholar]
  18. Eckenwalder, J.E. Systematics and Evolution of Populus. In Biology of Populus and Its Implications for Management and Conservation; Stettler, R.F., Bradshaw, H.D., Jr., Heilman, P.E., Hinckley, T.M., Eds.; NRC Research Press, National Research Council of Canada: Ottawa, ON, Canada, 1996; Chapter 1. pp. 7–32. [Google Scholar]
  19. Kenney, W.A.; Sennerby-Forsse, L.; Layton, P. A review of biomass quality research relevant to the use of poplar and willow for energy conversion. Biomass 1990, 21, 163–188. [Google Scholar] [CrossRef]
  20. Kenney, W.A.; Gambles, R.L.; Sennerby-Forsse, L. Feedstock Characteristic and Quality. In Ecophysiology of Short Rotation Forest Crops; Mitchell, C.P., Ford-Robertson, J.B., Hinckley, T., Sennerby-Forsse, L., Eds.; Elsevier: London, UK, 1992; Chapter 10. pp. 267–284. [Google Scholar]
  21. Mátyás, C.; Peszlen, I. Effect of age on selected wood quality traits of poplar clones. Silvae Gen. 1997, 46, 64–72. [Google Scholar]
  22. Tharakan, P.J.; Volk, T.A.; Abrahamson, L.P.; White, E.H. Energy feedstock characteristics of willow and hybrid poplar clones at harvest age. Biomass Bioenergy 2003, 25, 571–580. [Google Scholar] [CrossRef]
  23. Pliura, A.; Yu, Q.Y.; Zhang, S.Y.; MacKay, J.; Périnet, P.; Bousquet, J. Variation in wood density and shrinkage and their relationship to growth of selected young poplar hybrid crosses. For. Sci. 2005, 51, 472–482. [Google Scholar]
  24. Pliura, A.; Zhang, S.Y.; MacKay, J.; Bousquet, J. Genotypic variation in wood density and growth traits of poplar hybrids at four clonal trials. For. Ecol. Manag. 2007, 238, 92–106. [Google Scholar] [CrossRef]
  25. Olson, J.R.; Jourdain, C.J.; Rousseau, R.J. Selection for cellulose content, specific gravity, and volume in young Populus deltoides clones. Can. J. For. Res. 1985, 15, 393–396. [Google Scholar] [CrossRef]
  26. Kord, B.; Kialashaki, A.; Kord, B. The within-tree variation in wood density and shrinkage, and their relationship in Populus euramericana. Turk. J. Agric. For. 2010, 34, 121–126. [Google Scholar]
  27. Riemenschneider, D.E.; Berguson, W.E.; Dickmann, D.I.; Hall, R.B.; Isebrands, J.G.; Mohn, C.A.; Stanosz, G.R.; Tuskan, G.A. Poplar breeding and testing strategies in the north-central US: Demonstration of potential yield and consideration of future research needs. For. Chron. 2001, 77, 245–253. [Google Scholar]
  28. Beaudoin, M.; Hernández, R.E.; Koubaa, A.; Poliquin, J. Interclonal, intraclonal and within-tree variation in wood density of poplar hybrid clones. Wood Fiber Sci. 1992, 24, 147–153. [Google Scholar]
  29. Geyer, W.A.; DeWyke, J.; Walawender, W.P. Biomass and gasification properties of young Populus clones. Wood Fiber Sci. 2000, 32, 375–384. [Google Scholar]
  30. DeBell, D.S.; Singleton, R.; Harrington, C.A.; Gartner, B.L. Wood density and fiber length in young Populus stems: Relation to clone, age, growth rate, and pruning. Wood Fiber Sci. 2002, 34, 529–539. [Google Scholar]
  31. Ivkovich, M. Genetic variation of wood properties in balsam poplar (Populus balsamifera L.). Silvae Gen. 1996, 45, 119–124. [Google Scholar]
  32. Hernández, R.E.; Koubaa, A.; Beaudoin, M.; Fortin, Y. Selected mechanical properties of fast-growing poplar hybrid clones. Wood Fiber Sci. 1998, 30, 138–147. [Google Scholar]
  33. Zhang, S.Y.; Yu, Q.; Chauret, G.; Koubaa, A. Selection for both growth and wood properties in hybrid poplar clones. For. Sci. 2003, 49, 1–8. [Google Scholar]
  34. Semen, E.; Kuo, M.; Su, Y.C.; Hall, R.B.; Stokke, D.D. Physical properties of kraft pulp from four-year-old aspen hybrids and crosses. Wood Fiber Sci. 2001, 33, 140–147. [Google Scholar]
  35. Blankenhorn, P.R.; Bowersox, T.W.; Strauss, C.H.; Kessler, K.R.; Stover, L.R.; Kilmer, W.R.; DiCola, M.L. Effects of management strategy and site on specific gravity of a Populus hybrid clone. Wood Fiber Sci. 1992, 24, 274–279. [Google Scholar]
  36. Murphey, W.K.; Bowersox, T.W.; Blankenhorn, P.R. Selected wood properties of young Populus hybrids. Wood Sci. 1979, 11, 263–267. [Google Scholar]
  37. Zalesny, R.S., Jr.; Hall, R.B.; Zalesny, J.A.; Berguson, W.E.; McMahon, B.G.; Stanosz, G.R. Biomass and genotype × environment interactions of Populus energy crops in the Midwestern United States. Bioenergy Res. 2009, 2, 106–122. [Google Scholar] [CrossRef]
  38. Hansen, E. Mid-Rotation Yields of Biomass Plantations in the North Central U.S.; Res. Pap. NC-309; U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station: St. Paul, MN, USA, 1992; p. 8. [Google Scholar]
  39. Hansen, E.A.; Ostry, M.E.; Johnson, W.D.; Tolsted, D.N.; Netzer, D.A.; Berguson, W.E.; Hall, R.B. Field Performance of Populus in Short-Rotation Intensive Culture Plantations in the North-Central U.S.; Res. Paper NC-320; U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station: St. Paul, MN, USA, 1994; p. 13. [Google Scholar]
  40. Netzer, D.A.; Tolsted, D.; Ostry, M.E.; Isebrands, J.G.; Riemenschneider, D.E.; Ward, K.T. Growth, Yield, and Disease Resistance of 7- to 12-Year-Old Poplar Clones in the North Central United States; General Technical Report NC-229; U.S. Department of Agriculture, Forest Service, North Central Research Station: St. Paul, MN, USA, 2002; p. 31. [Google Scholar]
  41. Web Soil Survey. Available online: http://websoilsurvey.nrcs.usda.gov (accessed on 21 February 2013).
  42. National Climatic Data Center (NCDC) monthly summaries. Available online: http://www.ncdc.noaa.gov/oa/ncdc.html (accessed on 21 February 2013).
  43. SAS Institute Inc. SAS/STAT® 9.3 User’s Guide; SAS Institute Inc: Cary, NC, USA, 2011.
  44. Farmer, R.E. Variation and inheritance of eastern cottonwood growth and properties under two soil moisture regimes. Silvae Genet. 1970, 19, 5–8. [Google Scholar]
  45. Pitre, F.E.; Pollet, B.; Lafarguette, F.; Cooke, J.E.K.; MacKay, J.J.; Lapierre, C. Effects of increased nitrogen supply on the lignification of poplar wood. J. Agric. Food Chem. 2007, 55, 10306–10314. [Google Scholar] [CrossRef]
  46. Hacke, U.G.; Plavcová, L.; Almeida-Rodriguez, A.; King-Jones, S.; Zhou, W.; Cooke, J.E.K. Influence of nitrogen fertilization on xylem traits and aquaporin expression in stems of hybrid poplar. Tree Physiol. 2010, 30, 1016–1025. [Google Scholar] [CrossRef]
Forests EISSN 1999-4907 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert