**1. Introduction**

Although the main role of agriculture is food production, a part of agricultural land has always been devoted to non-food products, mainly within the framework of emerging technologies. Such uses include the production of bioenergy and various biomaterials [1]. The application of biomass for energy generation and for industry is of great importance and brings benefits to: (i) energy independence, (ii) environmental protection, (iii) economy, and (iv) society [2,3]. Non-food crops cultivation should not compete with food and fodder cultivation on high-yielding, fertile soils of good agricultural quality. On the other hand, the number of available food and fodder species able to produce a satisfactory yield on light, sandy soils of rather poor agricultural quality is not wide [4,5]. The cultivation of tree species using the short rotation coppice technique of lignocellulose biomass production had spread following the first oil crisis. Plants cultivated as short rotation coppices (SRCs) are characterized by a high growth rate, adequate sprouting of the stool bed, and an adaptation to sub-optimal environmental conditions [6]. Plantations for woody biomass production can be adapted depending on planting density and rotation length. Available experimental results indicate that a decrease in stem circumference is a commonly observed response to increasing planting density in poplar. However, studies also showed fast growing hardwoods tree height can increase, decrease, or remain unchanged with increasing planting density [7,8].

Agricultural biodiversity is of great importance nowadays. Plants of different genotype (species, varieties) including those cultivated as SRC, differ in habitat requirements for optimal growth and development and create different habitats for wildlife. Production conditions such as soil quality, water availability, harvest cycle, and technology (i.e., planting density) have an impact on biomass production, but the strength and direction of this impact could vary between species.

The goal of this study was to the determine the grow rate, biometric parameters, and biomass caloric value of six trees species cultivated as SRC depending on planting density.

### **2. Materials and Methods**

#### *2.1. Site Characteristics and Experimental Design*

The experiment was conducted in the experimental farm of Institute of Soil Science and Plant Cultivation—State Research Institute in Osiny, Poland (N: 51◦28 16.37", E: 22◦3 5.11"). The experiment was established in spring 2010 on heavy black soil with a heavy clay granulometric composition. The following trees were included in the experiment:


Trees were planted in April 2010 at 18 ranges of density in a "Nelder wheel design" [9] (see Figure 1). Eighteen concentric rings of trees were planted at radii ranging from 1.5 to 11 m as indicated in Table 1. An additional outer ring was planted as a guard ring to minimize the edge effect. The center of the circle was planted with a small ring to also form a guard. Each of the experimental rings contained 24 trees (six tested species in four replications) planted et equal distances around the circumference, giving a range from 3448 to 51,282 trees per hectare in equivalent planting density (Table 1). Those 24 trees planted as a concentric ring of increasing diameter formed 24 rows of experiment. Each tested species was represented in four rows (replications). Two rows of the same species were always adjacent to each other, while the other two were on the opposite side of the circle—creating an experimental arrangement in the form of a mirror reflection (see Figure 1).


**Table 1.** Planting distances and tree densities.

**Figure 1.** Distribution of species in the experiment design according to Nelder (1962) design (dots of different color represents trees of different species).

#### *2.2. Biomass Analyses*

Trees were harvested after 7 years of growth in February 2017. The harvest was made by hand. Shoot diameter at a height of 15 cm was determined using an electronic caliper with an accuracy of 0.1 mm. Each plant was cut at a height of 5 cm above the ground level. Plant height was determined from the cut place to the top of the plant with the accuracy of 1 cm. Yield of green mass was determined by weighing whole plants immediately after harvesting (green mass included limbs and/or barks). Whole plants were chipped and carefully mixed. Representative biomass samples (in 5 replications) were taken to determine the share of dry mass. In the laboratory, biomass moisture was determined by a drying method at a temperature of 80 ◦C for a period of 14 days. The dry mass yield was determined from the ratio of the green mass yield and its moisture content. Dried samples of trees were burned in calorimeter (KL-12Mn calorimeter, Precyzja-Bit, Bydgoszcz, Poland) in order to assess the higher heating value—HHV (or gross calorific value) of the biomass.

#### *2.3. Statistical Analyses*

Statistical analysis of results was performed using Statistica 10.0 software (StatSoft Polska, Krakow, Poland)). Few clear extreme outliers (observations located outside upper or lower quartile) were removed from the dataset according to the software manual. The level of significance for analysis was set to *p* = 0.05.

Subsequently, the normality of the distribution was tested using the Shapiro–Wilk and Kolmogorow–Smirnow tests. The vast majority of the examined features were characterized by a non-normal distribution. Data transformation attempts have not changed the data distribution. Therefore, nonparametric Kruskal–Wallis tests for comparison of many groups of independent variables

were used to assess the significance of differences. Because of data distribution, all average values presented in the study are medians.

The green mass, the share of dry mass, dry mass, and the biometric features of the examined trees, depending on the plant density, were characterized by using the trend equation. The criterion for selecting the trend equation was the highest value of the determination coefficient.

In addition, for each tree species correlation coefficient for the tested parameters relationship was calculated.

Woody plants intended for industry are, by definition, grown in long cycles, and therefore, their growth in subsequent years was not studied. Annual yields were not assessed as experimental design was not adapted to it (it would result in a complete failure of the basic methodological assumptions due to invasive nature of such measurements (annual harvesting)). However, potential annual biomass of dry mass was calculated *Ydm* (t ha−<sup>1</sup> year−1). To calculate the potential annual biomass yield of dry mass at given (chosen) density, the measured average dry mass of a single plant (*Pdm*) (kg) was multiplied by actual density of plantings (plants per hectare) (*Dact*). The obtained result (t ha <sup>−</sup>1) was divided by the number of years of cultivation, which was 7.

Potential annual biomass yield of dry mass:

$$\chi\_{dm} = \frac{P\_{dm}D\_{act}}{\mathcal{T}}$$

#### **3. Results and Discussion**

#### *3.1. Green Mass*

The highest green mass of plants (GM) were observed in poplar AF2. On average, a single plant of poplar weighted 29.1 kg (median). Boxelder maple green mass was on average at a level of 9.3 kg per plant. Black alder, white birch, silver maple, and Siberian elm had a GM on a similar level (Table 2). Nevertheless, the lowest GM was recorded for silver maple, and it was only 2.6 kg per plant. GM was positively correlated with plant height and shoot diameter for all tested plants. For Polar AF2 and Siberian elm, a negative correlation between the green mass of the plants and the share of dry mass was noted. The GM was dropping significantly with the increasing planting density for all tested species. The strength of response of species to the increasing stand density showed some differences. The strongest negative response was observed in black alder, while white birch and silver maple were the least sensitive species to increasing planting density. In the case of other species, the strength of dependence was at a similar level (Figure 2, Table 3). Wilkinson et al. [10] showed that yield of green matter of 1-year-old and 3-year-old willow (*Salix* ssp.) was increasing significantly with increasing plant density (density from 10,000 to 25,000 of plants per hectare).

**Table 2.** Median values of tested plants and their biometric features.


\* Data marked with the same letter do not differ significantly between species (α = 0.05).


**3.** Correlation matrix for the studied features.

**Table** 



**Figure 2.** Relationship between green mass of plants (kg plant<sup>−</sup>1) and planting density (plants ha−1).

#### *3.2. Dry Mass*

Very similar relationships were found for the dry mass (DM) of plants as for green matter. This shows that the dry mass/green mass ratio (or in other words—moisture content) is at constant level at different plating densities and, therefore, have little or no effect on yields. The highest dry mass (DM) of plants were observed in poplar AF2. On average, a plant of poplar weighted 13.1 kg (median). Boxelder maple green mass was on average at a level of 4.7 kg per plant. Black alder, white birch, silver maple, and Siberian elm had a dry mass of plant on a similar level (Table 2). Nevertheless, the lowest dry mass of plants was recorded for silver maple, and it was only 1.4 kg per plant. Walle et al. [11] compared the dry mass of 4-year-old SRC of poplar (*Populus trichocarpa* × *deltoids*), birch (*Betula pendula* Roth), and maple *(Acer pseudoplatanus* L.) cultivated with a density of 20,000, 6667, and 6667 plants per hectare, respectively. The authors found that the average dry mass of plants grown under these conditions was 831, 2007, and 738 g, respectively, which was a noticeably different result than in presented study; however, the growing conditions were also different to the plants tested by Walle et al. [11], which were also about 3 years younger than in the presented study. DM in the presented study was positively correlated with plant height and shoot diameter for all tested plants. For Siberian elm, a negative correlation between the dry mass of the plants and the share of dry mass was noted. The dry mass of plants dropped significantly with the increasing planting density for all tested species. The strength of response of species to the increasing stand density showed some differences. The strongest negative response was observed for black alder, while white birch and silver maple were the least sensitive species to increasing plant density. In the case of other species, the strength of dependence was at a similar level (Figure 3, Table 3). Other authors investigated the response of dry mass of plants to increasing density and found out that it was also dropping for willow (*Salix* ssp.) [12] and for oak (*Quercus robur*) [13].

**Figure 3.** Relationship between dry mass of plant biomass (kg plant<sup>−</sup>1) and planting density (plants ha−1).

#### *3.3. Share of Dry Mass*

The highest dry to green mass (DM to GM) ratio was observed for Siberian elm (58.5%), while the lowest was observed for poplar AF2 (42.8%). White birch, silver maple, boxelder maple, and black alder DM to GM ratio was on a similar level (52.4, 51.8, 50.8, and 48.2%, respectively) (Table 2). DM to GM ratio varied depending on the stand density, but also on the tested species. Share of dry mass was negatively correlated with green mass of plants for three tested species (poplar AF2, Siberian elm and boxelder maple). Share of dry mass also negatively correlated with plant height for poplar AF2 and with higher heating value for white birch (Table 3). Poplar AF2 and silver maple showed no reaction of DM to GM ratio to increasing density of plants, while Siberian elm, black alder, and white birch showed moderately positive increase in DM to GM ratio with increasing plant density. Boxelder maple showed negative response of DM to GM ratio to increasing plant density (Figure 4). Wilkinson et al. [10] found dry mass of 1-year-old and 3-year-old willow (Salix ssp.) not dependent on density of plantings (similar to poplar AF2 and silver maple reaction in presented study). In addition, Stolarski et al. [14] and Kulig et al. [15] found no effect of planting density on the fresh–dry matter ratio of willow. This was also confirmed by Elfeel and Elmagboul [16] for other woody species—*leucaena leucocephala.* On the other hand, Achinelli et al. [17] found that dry to fresh matter content ratio in willow was higher in more dense stands (similar effect to Siberian elm, black alder, and white birch in the discussed study).

**Figure 4.** Relationship between dry mass share (%) and planting density (plants ha<sup>−</sup>1).

#### *3.4. Potential Yield of Dry Mass*

Poplar AF2 had the highest calculated potential of dry mass yield (after 7 years) of about 15 t ha−<sup>1</sup> for all tested planting densities. Boxelder maple was able to match yielding potential that AF2 poplar only at planting density of about 40,000 plants per hectare. Silver maple reached about 50% of this potential (about 7.5–8 t ha−1), while other tested species reached about 5 t ha−<sup>1</sup> of dry mass yield. (Figure 3). The calculated annual yield of dry mass was positively correlated with plant height for poplar AF2, Siberian elm, and silver maple. There was also a positive correlation with plant density for boxelder maple and negative for black alder. In the case of white birch, a strong positive correlation was found with green and dry mass of plants (Table 3). Despite the fact that the dry mass of individual trees decreased with increasing density, the calculated annual yield of dry matter of Siberian elm, boxelder maple, and silver maple was increasing with increasing density of planting (Figure 5). The same was also observed by Geyer, Argent, and Walawender [18] for 7-year-old Siberian elm. Authors found that annual yields of dry matter of Siberian elm were at a level of 4.7 t ha−<sup>1</sup> with stand density of 1400 plants per hectare, while they increased significantly to 9.8 t ha−<sup>1</sup> with stand density of 7000 plants per hectare. In the presented study, calculated annual dry mass yields of Siberian elm varied between 1.8 t ha−<sup>1</sup> for the lowest density and 7.4 t ha−<sup>1</sup> for the highest density. Geyer and Walawender [19] reported that annual dry mass yield of 7-year-old silver maple was increasing from 5.3 t ha <sup>−</sup><sup>1</sup> at 1400 plants per hectare to 11.2 t ha−<sup>1</sup> at 7000 plants per hectare. In addition, other authors found that for some species such as willow (*Salix* L.) [10] and black locust (*Robinia pseudoacacia*) [20] annual yield of dry mass was increasing with increasing planting density. Niemczyk et al. [21] found that annual yields of a 7-year-old poplar can reach up to 8 t ha<sup>−</sup>1. Truax et al. [22] also found a positive correlation of planting density and yield of 8-year-old hybrid poplar. Stolarski et al. [23] found that annual dry mass yield

of willow planted with densities from 12,000 to 96,000 was increasing from 12,000 to 24,000 (optimal density) and decreasing with increasing density from 24,000 to 96,000. Similar reaction was found in presented study silver maple (optimal planting density of about 25,000–30,000 plants per hectare).

**Figure 5.** Relationship between potential yield of dry mass per year (t ha−<sup>1</sup> r<sup>−</sup>1) and planting density (plants ha<sup>−</sup>1).
