Growth and Morphological Patterns of Norway Spruce (Picea abies (L.) Karst.) Juveniles in Response to Light Intensities

Abstract

(1) Background: The growth and morphology of Norway spruce (Picea abies (L.) Karst.) juveniles were observed under different light conditions due to overstory canopy openness. The aim was to investigate the photo-morphological responses of juveniles for the development of a desirable forest structure. The research was located in a higher altitude zone in central Europe. (2) Methods: Light was estimated using hemispherical photographs. Eight different parameters of natural regeneration were measured on individuals within 1 × 1 m around each point on a 3 × 3 m grid. A total of 1214 sapling measurements (from 10 to 431 cm in height) were taken. (3) Results: Light affected the growth and morphological development of juveniles, resulting in variation in the lateral crown growth and crown length. Acclimations manifested as trade-offs between height and lateral crown growth. Similar shifts were found within relative height growth (the terminal length to the total height) and apical dominance ratio (the terminal length to the branch length). The crown length was proven to be highly capable to acclimation. Light influenced the density weakly, however, when a regeneration index (density × median height) was considered, significant relations with light conditions were discovered.

1. Introduction

Forest regeneration and establishment are critical stages during which considerable losses in species diversity can often occur [1]. Natural regeneration (or natural regeneration in conjunction with artificial regeneration) often results in a large initial seedling bank. This is favored by a high proportion of unoccupied growing space and a diversity of microsite characteristics. As the plants increase in size and fill in the available space, competition becomes more intense and the growing conditions become more homogeneous. As the initial seedling bank progressively declines, species diversity may dramatically decrease as less competitive species are eliminated [2]. These competitive relationships must be considered in silvicultural techniques [3,4,5]. When the transformation of pure Norway spruce (Picea abies (L.) Karst.) stands is adopted, natural regeneration of Norway spruce (combined with planting of reintroduced tree species) must be controlled to prevent reintroduced species from being overtopped and out-competed [6]. One of the main silvicultural tools for this purpose is adjusting light intensities through overstory canopy openness. Thus, forest structure and heterogeneity are encouraged [7,8].

Uneven-aged forest management is characterized by silvicultural practices that create or maintain stands made up of trees of different age classes [9,10]. Different terms and definitions are used for the numerous forms of uneven-aged silviculture [10,11]; however, all these silvicultural systems are based on the creation of canopy openings or gaps to allow light to reach the floor of the stand [5,7,8]. The degree of growth heterogeneity, and structure of the understory induced by the variability of light levels, can be as variable and advanced as the shade tolerance of particular tree species is high [12,13]. In similar studies of forest ecology where light is an independent variable, it is important to consider reference levels of other site conditions, such as nutrients and water supply. This is because shade tolerance is shifted by the availability or scarcity of other site factors [14,15,16]. It is therefore advantageous to perform such an investigation on nutrient-poor sites as a basic level of nutrient resources in combination with average water supply, which can then be used as a reference compared to more optimal conditions (such as nutrient-enriched conditions) where shade tolerance is generally increased [14,15].

Tree species with different levels of shade tolerance show different levels of both morphological and physiological plasticity under various light conditions [4,8,17]. The response of crown morphology and height growth depending on shade tolerance or intolerance has been described in different studies [3,18,19]. In general, the conclusion of those studies is that shade-tolerant tree species can adapt their crown morphology more acutely and in more morphological traits (such as crown length, width, and projection or terminal length) along the light gradient compared to shade-intolerant tree species. Shade-tolerant tree species can effectively reduce their growth and survive under low light intensities due to their diminished growth rate and photo-morphological responses resulting in lower mortality rate in shaded conditions; after their release, they react immediately to increased light intensities with an increased growth rate [15,20,21]. However, they do not reach the same growth rate under canopy-open conditions as shade-intolerant tree species [22].

The three following objects of interest related to photo-morphological patterns of Norway spruce juveniles were analyzed in this paper:

• The existence and intensity of morphological plasticity (i.e., the ability to respond to the different availability of resources by acclimation of the morphological traits) as a response to different light intensities;
• The definition of microsites where the morphological plasticity was realized most distinctly;
• The influence of light intensities on the general development of Norway spruce juveniles (including, for instance, their density) and a consideration of all observed patterns for further use in silvicultural guidelines.

2. Materials and Methods

2.1. Research Sites

The research was conducted on acidic sites, i.e., Fageta piceoso-abietina or Fageta abietina-piceosa, due to the fact that a large proportion of these sites were occupied by secondary Norway spruce monocultures [23]. Selected research plots were represented by forest stands with a pure even-aged Norway spruce overstory that was tended by thinnings from below in young and mid-aged growing stages. Then in the growing stage of forest maturity, shelter-wood cuts of different intensities or gaps were created 15–25 years ago with no further silvicultural interventions since that time. The observations took place in the central part of the Czech Republic in the Bohemian–Moravian Highlands within altitudes of ca. 670–770 m above sea level (Figure 1). The soils of the research plots were determined as cambisols (mainly Eutric Cambisol; less often Dystric Cambisol or Areno-dystric Cambisol). The whole area lies in the geological entity of Moldanubikum with the typically prevailing acidic bedrocks of mainly amphibolite with veins of granodiorite or granite [24]. The mean annual air temperature is 4.9 °C; the mean annual precipitation is 850 mm with 55%–60% falling in the growing season [24,25], and the number of days when the snow cover is present usually ranges between 100 and 120 days [26].

2.2. Experiment Design

Three research plots were established for this study (

Table 1. Fundamental characteristics of the research plots.

Plot Name (ID)	Geographical Coordinates	Elevation (m a.s.l.)	Dimension (m)	Area (m2)	Number of Sampled Sub-Plots
Drtič (A)	49°36′50″ N; 15°53′05″ E	672	27 × 42	1134	126
Tornádová (B)	49°40′12″ N; 15°55′56″ E	737	12 × 21	252	28
Odvozní (C)	49°40′17″ N; 15°56′13″ E	766	9 × 12	108	12

), taking into account experimental conditions as follows:

• Overstory, consisting of pure mature even-aged Norway spruce stands (approximately 100 years old);
• A gradient of overstory canopy openness characterized by a significant gradient of light intensity within the plots ranging from complete canopy closure up to open sky conditions of a forest gap; the maximum size of a gap in the study was considered as ca. 16 m in diameter (i.e., ca. 2/3 of overstory mature tree height) and thus ca. 200–220 m² (ca. 2 ares) of the total size; thus, the light conditions ranged from 0.04 to 0.46 of the total site factor (TSF);
• No further silvicultural interventions were carried out within natural regeneration (once it started) or within the mature overstory once the initial regeneration cut (shelter-wood cut or gaps) occurred.

Field-Map technology (IFER, Prague, Czech Republic) was utilized to identify the position of the mature overstory for both tree positions and their crown projections and to locate borders of research plots. All present mature trees within the plot areas and all neighbors whose crown projections intervened in the area of the plot were included. Afterward, the same equipment was used to cover the research plots in a grid of 3 m × 3 m squares. Every segment (3 m × 3 m) of the grid was permanently marked at its central point (centroid) (Figure S1). Centroids were then targeted for subsequent measurements when hemispherical photos were taken here and the number of Norway spruce juveniles was measured within 1 m × 1 m sub-plots around the centroids.

2.3. Natural Regeneration Measurements

Around each centroid (within the 1 × 1 m sub-plot), the number of all present Norway spruce individuals was counted (individuals both below and over 10 cm in height were included), and the morphological parameters were measured for all seedlings and saplings over 10 cm in height (according to [3]).

Seven different morphological parameters of Norway spruce juveniles were measured, as listed in

Table 2. Parameters of Norway spruce natural regeneration that were measured within the study.

Parameter	Acronym	Juveniles Involved
Total density	DEN	all
Total height	H (cm)	above 10 cm
Live crown length	Cl (cm)	above 10 cm
Terminal leader length	Tl (cm)	above 10 cm
Length of the longest branch	Bl (cm)	above 10 cm
Width of crown in the widest part of crown	CW1 (cm)	above 10 cm
Width of crown perpendicular to the widest part of crown	CW2 (cm)	above 10 cm
Diameter at height of 5 cm above ground	D (mm)	above 10 cm

. The chosen morphological parameters were employed in studies of the photo-morphological plasticity of different tree species in the past [27,28,29].

Stem diameter was taken as an average from two mutually perpendicular measurements, and it was measured using a digital caliper with an accuracy of 0.1 cm. It was measured 5 cm above the ground, due to the possible occurrence of morphological anomalies within the root collar [30,31]. Other morphological parameters were measured using a telescopic tape measure or a Messfix telescopic measuring stick (NEDO, Switzerland) with an attached level and a precision of 1 cm. In total, 1214 juveniles of Norway spruce were measured with heights ranging from 10 to 431 cm within 166 sub-plots on three different research plots.

2.4. Indirect Estimation of Light Intensities

We used color hemispherical photography evaluation as an indirect method to estimate the potential light intensity [5,32]. This method is also fully compatible with instantaneous direct methods, such as photon flux density measurement [5,33]. A self-leveling platform (Régent Instruments Inc., Quebec City, QC, Canada) with a remote-triggered Nikon Coolpix 995 digital camera (Nikon, Tokyo, Japan) with a fine-quality, calibrated fish-eye lens was employed. The whole self-leveling platform was fixed atop a Manfrotto 269 HDBU tripod (Manfrotto, Cassola, Italy) with a maximum height of 8 m; this enabled photos to be taken from 1.7 m (the essential height of the tripod) to 4.5 m for photographing the tallest juveniles. Photographs were taken during windless weather and standard overcast sky conditions—i.e., when the solar disk was completely obscured (according to Čater et al. 2013 [5]). The camera was leveled and the fish-eye lens oriented toward magnetic north using a compass.

The standardized post-processing and evaluation of the photos were performed with WinSCANOPY 2012 (Régent Instruments Inc., Quebec City, QC, Canada). The quality of photos was controlled visually, and batches (of maximally 20) of similar photos (mainly based on features of exposure and contrast) were created. Then the process of distinguishing sky vs. canopy was performed by selecting different colors of a picture and sorting them into one of these categories. Finally, black and white thresholding was tested for several random pictures of the batch to check the accuracy of the thresholding visually. The value of 60° as a zenith angle detecting an evaluated sector of the picture view was used as the most accurate and effective [5]. The growing season was considered from 1 May to 30 September, according to Löf et al. (2007) [21] and Čater et al. (2013) [5]. Other required values, such as the geographical coordinates, the elevation, and the magnetic declination, were considered. Five different characteristics (two classifying the openness of the canopy and three classifying light intensities (

Table 3. Characteristics of light intensities and canopy openness based on hemispherical photography evaluation used in this study.

Entity	Acronym	Description	Open Area Conditions
Gap Fraction	GFr	pixels of “Sky” in two dimensions	100%
Openness	Op	pixels of “Sky” in three dimensions	100%
Direct Site Factor	DSF	sum of direct light of a growing season	1
Indirect Site Factor	ISF	sum of indirect light of a growing season	1
Total Site Factor	TSF	sum of total light of a growing season	1

)) were taken into account for subsequent analysis of this study according to the manual of WinSCANOPY software [34].

2.5. Data Evaluation

2.5.1. The Effect of Light Intensities on the Density of Natural Regeneration

First, all individuals of all sub-plots in the research plots were considered for the density evaluation. Then, to determine if there was a relationship between light intensity and density along the dimension of natural regression (i.e., height), height classes determined by the tallest juvenile within a sub-plot were created and only those sub-plots were subsequently tested. The height classes were as follows: up to 15 cm, 20 cm, 30 cm, 40 cm, 50 cm, 60 cm, 70 cm, 80 cm, 90 cm, 100 cm, 110 cm, 120 cm, 130 cm, and 140 cm.

2.5.2. Morphological Patterns of Juveniles under Size-Independent Evaluation

The influence of light intensities on two measured parameters (H and D (Table 2)) and six calculated indexes (Table 4) was analyzed. The size (i.e., height classes) was not considered here; the exact value of a parameter or index for a sub-plot was calculated as the median of all present juveniles occurring within a particular sub-plot, according to Dai (1996) [35]. However, to make this size-independent evaluation useful for individuals of varying heights, parameters and indexes involving the parameters of total height (H) or terminal length (Tl) were used. For calculation of the regeneration index (RI), the number of juveniles higher than 10 cm was considered as the total number of juveniles in the equation for which the height was being measured.

2.5.3. Morphological Patterns of Juveniles under Size-Dependent Evaluation

Natural regeneration was divided into height classes according to Lundquist and Fridman (1996) [36], Duchesneau et al. (2001) [37], and Chrimes and Nilson (2005) [19]. The regular height classes concerning a range of heights in this study were as follows: 10–50 cm, 51–100 cm, 101–150 cm, and >151 cm.

The exact value of a parameter or an index of a height class was calculated within a sub-plot as a median of values of all juveniles falling into a particular height class. Thus, when more height classes were present within one certain sub-plot, more values of the same parameter (or index) were obtained, but for different height classes (Table 4). In total, for size-dependent evaluation, the influence of light intensities on all seven measured morphological parameters (Table 2) and eight calculated indexes was tested (Table 4).

2.6. Statistical Analysis

All statistical analyses were performed with p < 0.05. William’s test was employed using QC Expert version 3.3 (Trilobyte Statistica Software Ltd., Pardubice, Czech Republic) to detect outlying values. The software STATISTICA version 10.0 (StatSoft Inc., Tulsa, OK, USA) was used to test the significance of models of light intensities and morphological features; quadratic polynomials were used to intersperse the values. When a model was found to be significant, the Spearman correlation coefficient (r_s) was tested (to test the robustness of correlation against the effect of extreme values, nonlinear correlation, and non-normality). For the visualization of certain indexes within the largest research plot, Wafer plots were constructed in STATISTICA software, and then those outputs were overlaid on the schema of the plot with 50% transparency in a freeware piece of graphics software.

Redundancy analysis (RDA) was performed to investigate the effect of the main light components (direct and indirect—DSF and ISF) and the similarities within intensities of photo-morphological plasticity of different observed parameters and indexes. In this analysis, all measured parameters were included, as were indexes calculated for size-independent evaluation (i.e., regardless of height classes); the density was also included. RDA was performed with R software version 3.6.0 [38] in R Studio version 1.2.1335 [39] using the R package vegan [40].

Table 4. Photo-morphological indexes calculated for Norway spruce juveniles considered within two different approaches of data evaluation.

Indexes Used for Testing of Photo-Morphology of Norway Spruce Juveniles under Size-Independent Evaluation *
Title of Index	Acronym	Equation
Relative height growth	RHG	Tl/H
Apical dominance ratio	ADR	Tl/Bl
Relative crown length	RCL	Cl/H
Relative crown width	RCW	(CW1 + CW2)/2 × H
Height–diameter ratio	HDR	H/D
Regeneration index **	RI (cm/m2)	juveniles’ number * median height
Indexes used for testing of photo-morphology of Norway spruce juveniles under size-dependent evaluation ***
Title of index	Acronym	Equation
Height–diameter ratio	HDR	H/D
Relative crown length	RCL	Cl/H
Relative crown width	RCW	(CW1 + CW2)/2 × H
Crown width–length ratio	CWLR	(CW1 + CW2)/2 × Cl
Crown projection	CP (cm2)	((CW1 + CW2)/4)2 × π
Index of spatial distribution	ISD	(CW1 + CW2)/2 × D
Crown volume according to equilateral cone	CVEC (cm3)	π/12 × (CW1 + CW2)2/4 × Cl
Crown volume according to quadratic paraboloid	CVQP (cm3)	π/8 × (CW1 + CW2)2/4 × Cl

H: total height (cm); Cl: live crown length (cm); Tl: terminal leader length (cm); Bl: length of the longest branch (cm); CW1: width of crown in the widest part of crown (cm); CW2: width of crown perpendicular to the widest part of crown (cm); D: diameter at height of 5 cm above ground (mm); * Indexes were calculated according to Messier et al. (1999) [27], Duchesneau et al. (2001) [37], Robakowski et al. (2004) [41], Grassi and Giannini (2005) [18], and Szymura (2005) [29]. ** The number of individuals in this equation was considered as the total number of juveniles that were higher than 10 cm for which the total height (H) was measured and then used to calculate a median of heights. *** Indexes were calculated according to Šebík and Polák (1990) [42], Takahashi (1996) [43], Messier et al. (1999) [27], Claveau et al. (2002) [44], Duchesneau et al. (2001) [37], Grassi and Giannini (2005) [18], and Szymura (2005) [29].

Table 4. Photo-morphological indexes calculated for Norway spruce juveniles considered within two different approaches of data evaluation.

Indexes Used for Testing of Photo-Morphology of Norway Spruce Juveniles under Size-Independent Evaluation *
Title of Index	Acronym	Equation
Relative height growth	RHG	Tl/H
Apical dominance ratio	ADR	Tl/Bl
Relative crown length	RCL	Cl/H
Relative crown width	RCW	(CW1 + CW2)/2 × H
Height–diameter ratio	HDR	H/D
Regeneration index **	RI (cm/m2)	juveniles’ number * median height
Indexes used for testing of photo-morphology of Norway spruce juveniles under size-dependent evaluation ***
Title of index	Acronym	Equation
Height–diameter ratio	HDR	H/D
Relative crown length	RCL	Cl/H
Relative crown width	RCW	(CW1 + CW2)/2 × H
Crown width–length ratio	CWLR	(CW1 + CW2)/2 × Cl
Crown projection	CP (cm2)	((CW1 + CW2)/4)2 × π
Index of spatial distribution	ISD	(CW1 + CW2)/2 × D
Crown volume according to equilateral cone	CVEC (cm3)	π/12 × (CW1 + CW2)2/4 × Cl
Crown volume according to quadratic paraboloid	CVQP (cm3)	π/8 × (CW1 + CW2)2/4 × Cl

H: total height (cm); Cl: live crown length (cm); Tl: terminal leader length (cm); Bl: length of the longest branch (cm); CW1: width of crown in the widest part of crown (cm); CW2: width of crown perpendicular to the widest part of crown (cm); D: diameter at height of 5 cm above ground (mm); * Indexes were calculated according to Messier et al. (1999) [27], Duchesneau et al. (2001) [37], Robakowski et al. (2004) [41], Grassi and Giannini (2005) [18], and Szymura (2005) [29]. ** The number of individuals in this equation was considered as the total number of juveniles that were higher than 10 cm for which the total height (H) was measured and then used to calculate a median of heights. *** Indexes were calculated according to Šebík and Polák (1990) [42], Takahashi (1996) [43], Messier et al. (1999) [27], Claveau et al. (2002) [44], Duchesneau et al. (2001) [37], Grassi and Giannini (2005) [18], and Szymura (2005) [29].

3. Results

3.1. The Effect of Light Intensities on the Regeneration Density

Only a weak relationship between light intensities and the density of Norway spruce natural regeneration was found. No significant effect was demonstrated when gap fraction (GFr), openness (Op), or indirect site factor (ISF) was considered through all tested classes of the maximal height. However, when direct site factor (DSF) or total site factor (TSF) was contemplated, a few significant relations were shown. For both light variables, significant (p < 0.05) relations were found when height classes up to 30, 40, 50, and 70 cm were investigated. Significance was also proven for DSF when height classes up to 60 and 100 cm were tested (

Table A1. Significant regression models of Norway spruce natural regeneration density and light intensities.

Variable (x)	Height up to (cm)	Equation; DEN (No. of Juveniles/m2)	R2	p	rs
DSF	30	DEN = 16.5597 − 71.5915 × x + 91.067 × x2	0.09	0.0334 *	−0.0786 NS
	40	DEN = 14.2696 − 42.8019 × x + 39.1018 × x2	0.06	0.0366 *	−0.0826 NS
	50	DEN = 15.5823 − 46.297 × x + 38.3294 × x2	0.07	0.0145 *	−0.1196 NS
	60	DEN = 12.6875 − 12.4916 × x − 25.0481 × x2	0.04	0.0464 *	−0.0804 NS
	70	DEN = 13.5962 − 20.4831 × x − 11.5114 × x2	0.05	0.0257 *	−0.0858 NS
	100	DEN = 12.4144 + 8.674 × x − 71.9446 × x2	0.04	0.0419 *	−0.0830 NS
TSF	30	DEN = 16.9492 − 72.669 × x + 91.9606 × x2	0.09	0.0334 *	−0.0546 NS
	40	DEN = 14.3836 − 42.0798 × x + 36.6371 × x2	0.06	0.0423 *	−0.0878 NS
	50	DEN = 15.2769 − 41.107 × x + 26.4127 × x2	0.06	0.0193 *	−0.0937 NS
	70	DEN = 13.33 − 16.5955 × x − 20.0969x2	0.04	0.0367 *	−0.0679 NS

Note: DEN: total density of all juveniles regardless of their height (pieces/m²); DSF: direct site factor; ISF: indirect site factor; TSF: total site factor; p < 0.05 (*), p ˃ 0.05 (^NS).

in Appendix A). However, none of those significant relationships were established by the significance of the Spearman correlation coefficient (r_s). In addition, the significance of relations between DSF and the density of Norway spruce up to 80 and 90 cm was narrowly rejected with p-values of 0.0529 and 0.0562, respectively. All significant models showed a gentle decrease in density with increasing light intensities (i.e., negative correlation). In general, the effect of light conditions on the density of Norway spruce juveniles was found to be very weak and limited; when a relationship was found, it occurred only for juveniles up to 1 m in height.

3.2. General Morphological Patterns of Juveniles under Size-Independent Evaluation

3.2.1. Height Growth

A significant relationship between light intensities and relative height was established for all five light or canopy openness variables. Moreover, all of the discovered models were highly significant (p < 0.001), and their significance was proven to be robust against negative influences by the significance of the Spearman correlation coefficient (

Table 5. Significantly proven regression models of light intensities and photo-morphological acclimations of Norway spruce natural regeneration under size-independent evaluation.

Index	Regression Model	Variable (x)	R2	p	rs
RHG	RHG = 0.1904 − 0.0151 × x + 0.0005 × x2	GFr	0.26	0.0000 ***	0.3698 *
	RHG = 0.1911 − 0.015 × x + 0.0005 × x2	Op	0.26	0.0000 ***	0.3577 *
	RHG = 0.1104 − 0.1592 × x + 0.8724 × x2	DSF	0.10	0.0004 ***	0.1949 *
	RHG = 0.1813 − 1.1899 × x + 3.5281 × x2	ISF	0.25	0.0000 ***	0.2878 *
	RHG = 0.1105 − 0.1999 × x + 1.0346 × x2	TSF	0.12	0.0002 ***	0.2097 *
ADR	ADR = 0.3433 − 0.0285 × x + 0.001 × x2	GFr	0.24	0.0000 ***	0.4362 *
	ADR = 0.347 − 0.0287 × x + 0.001 × x2	Op	0.25	0.0000 ***	0.4218 *
	ADR = 0.1865 − 0.0733 × x + 1.1619 × x2	DSF	0.08	0.0015 **	0.2257 *
	ADR = 0.3402 − 2.4533 × x + 7.4787 × x2	ISF	0.25	0.0000 ***	0.3784 *
	ADR = 0.179 − 0.0943 × x + 1.374 × x2	TSF	0.09	0.0006 ***	0.2442 *
RCL	RCL = 0.5603 − 0.0024 × x + 0.0002 × x2	GFr	0.12	0.0001 ***	0.4242 *
	RCL = 0.559 − 0.0023 × x + 0.0002 × x2	Op	0.12	0.0001 ***	0.4301 *
	RCL = 0.5272 + 0.6574 × x − 0.53 × x2	DSF	0.11	0.0001 ***	0.3337 *
	RCL = 0.5723 − 0.3511 × x + 2.2008 × x2	ISF	0.14	0.0000 ***	0.4395 *
	RCL = 0.5113 + 0.7212 × x − 0.5476 × x2	TSF	0.12	0.0000 ***	0.3560 *
RI	RI = 116.5526 − 10.4147 × x + 0.798 × x2	GFr	0.09	0.0006 ***	0.2430 *
	RI = 126.1057 − 12.5477 × x + 0.8617 × x2	Op	0.10	0.0003 ***	0.2568 *
	RI = 215.7408 − 2436.291 × x + 10498.1164 × x2	ISF	0.16	0.0000 ***	0.3567 *

Note: RHG: relative height growth; ADR: apical dominance ratio; RCL: relative crown length; RI: regeneration index (cm/m²); GFr: gap fraction; Op: canopy openness; DSF: direct site factor; ISF: indirect site factor; TSF: total site factor; p < 0.05 (*), p < 0.01 (**), p < 0.001 (***).

). Relative height growth (RHG) increased moderately with increasing low light intensities; the increment became progressive with approximately 20% of the light intensity of open area conditions. This means the terminal length related to total height was higher in high light intensities. A comparison of RHG and the apical dominance ratio (ADR) is stated below (see Section 3.2.2) and visualized in Figure 2A. Parameter H was not found to significantly correlate with light intensities.

3.2.2. Lateral Crown Growth

The apical dominance ratio was significantly influenced by light intensities for all five light or canopy openness variables (p < 0.01). Moreover, all five regression models were proven to be robust against negative influences by the significance of the Spearman correlation coefficient (Table 5). For all five regression models, a positive correlation was observed when ADR was increasing with light intensity. This means the terminal length increased under higher light intensities while the lateral growth was slower. In low light intensities, lateral growth prevailed, and the annual height increment was suppressed. ADR, comprising both branch length (Bl) and terminal length (Tl) and thus expressing shifts within the level of height increment and lateral growth, was shown to be a suitable index for observation of this phenomenon. When considering only lateral growth expressed by relative crown width (RCW) within such a large gradient of juvenile dimensions (although H was the denominator for RCW), no significant correlation with light intensities was found.

In addition, changes within both RHG and ADR along the light gradient were reasonably similar. When the ranges of median values of sub-plots were compared, RHG was found to range within 15.2 times (the ratio of the highest median value of the sub-plot to the smallest median value of the sub-plot), while the same approach showed a range of 23.3 times for ADR. However, when all sets of juveniles were compared, the range was much wider and represented approximately 68 times and 131 times for RHG and ADR, respectively. In addition, the visualization of both indexes was performed (Figure 2A) on the largest research plot, Drtič (A). This demonstrated a similar rate of photo-morphological acclimations for both indexes. The same positive correlation of both ADR and RHG was also confirmed by RDA, which showed a similar level of effect of both DSF and ISF (i.e., direct and indirect components of light (Figure 3)).

3.2.3. Relative Crown Length

A significant positive correlation between light intensities and relative crown length (RCL) was established for all five light or canopy openness variables. All of the models were highly significant (p < 0.001), and their significance was proven to be robust against negative influences by the significance of the Spearman correlation coefficient (Table 5). RCL increased with light intensities, which proved the ability of Norway spruce juveniles to reduce crown length under low light intensity. The range of photo-plasticity of this index was lower compared to both RHG and ADR when within the comparison of median values of sub-plots (the highest median value of sub-plot/the smallest median value of sub-plot); the range was a multiple of 4.3 (when the whole set of juveniles was investigated, the range was a multiple of approximately 6.3). The variation of RCL within the largest research plot (A) is shown in Figure 2B. Within the RDA analysis, huge similarities were found between the effects of DSF and ISF on RCL, RHG, and ADR (Figure 3).

3.2.4. Regeneration Index—RI

A significant positive correlation between the regeneration index (RI; RI = the number of individuals × median of the height) and three variables (GFr, Op, ISF) was established. For those three regression models, a high level of significance was found (p < 0.001), and the significance of the Spearman correlation coefficient was shown for all three (Table 5). However, when both DSF and TSF were tested, no significant influence was revealed.

This means that if dimension was not considered, the density of natural regeneration weakly correlated with light conditions within a limited range of its height (maximally up to 1 m (see Section 3.1)). However, when RI was tested, significance was proven. In addition, the common density (pieces/m²) negatively correlated with light conditions (albeit poorly), while the regeneration index (cm/m²) correlated positively. This shows the fundamental importance of regeneration height within the regeneration index. The total height (and other parameters or indexes expressing the height growth) correlated positively with light in this study as well. Furthermore, based on the RDA, a positive correlation of RI with light intensity can also be concluded (Figure 3).

3.2.5. Stem Diameter

Statistical analyses did not prove the effect of light intensities on the thickness of the stem when diameter (D) did not correlate with light intensities. Moreover, no significant influence of light intensities was found on the height–diameter ratio (HDR) where D is included as a denominator (when H alone was also not found to significantly correlate with light intensities in this general approach of size-independent evaluation).

3.3. Morphological Patterns of Juveniles under Size-Dependent Evaluation

3.3.1. Height Growth

Terminal length (Tl) and total height (H) were tested, and significant regression models for both were established with a positive correlation between height growth and light intensities. The Tl was more dependent on light intensities; it was significantly influenced by light in three height classes (10–50 cm, 51–100 cm, >151 cm) (

Table A2. Significantly proved regression models of light intensities and morphological parameters expressing height growth of Norway spruce natural regeneration under size-dependent evaluation.

Height Class	Regression Model	Variable (x)	R2	p	rs
10–50 cm	Tl = 3.5544 − 0.2377 × x + 0.009 × x2	GFr	0.18	0.0001 ***	0.4052 *
10–50 cm	Tl = 3.4941 − 0.2304 × x + 0.0088 × x2	Op	0.18	0.0001 ***	0.4020 *
10–50 cm	Tl = 3.458 − 20.8439 × x + 72.033 × x2	ISF	0.22	0.0000 ***	0.4187 *
10–50 cm	Tl = 1.9048 + 5.5023 × x + 0.4773 × x2	TSF	0.06	0.0059 **	0.1944 *
51–100 cm	Tl = 9.7984 − 48.9895 × x + 151.7929 × x2	ISF	0.09	0.0275 *	0.3180 *
51–100 cm	Tl = 8.1192 − 17.2793 × x + 73.168 × x2	TSF	0.08	0.0469 *	0.2650 *
˃151 cm	Tl = 17.7294 − 90.283 × x + 356.3486 × x2	DSF	0.48	0.0001 ***	0.5342 *
˃151 cm	Tl = −0.6218 + 46.0722 × x + 41.9031 × x2	ISF	0.14	0.0363 *	0.4228 *
˃151 cm	Tl = 21.1318 − 126.4246 × x + 439.5598 × x2	TSF	0.48	0.0001 ***	0.5471 *
˃151 cm	H = 20.712 + 1827.4301 × x − 3017.8096 × x2	DSF	0.21	0.0139 *	0.4374 *
˃151 cm	H = −68.8182 + 2527.9261 × x − 4391.4594 × x2	TSF	0.22	0.0134 *	0.4194 *

Note: Tl: terminal length (cm); H: total height (cm); GFr: gap fraction; Op: canopy openness; DSF: direct site factor; ISF: indirect site factor; TSF: total site factor; p < 0.05 (*), p < 0.01 (**), p < 0.001 (***).

in Appendix A). By contrast, total height was significantly correlated only with light intensities in the tallest height class (>151 cm) (Table A2 in Appendix A). A distinctive effect of light on Tl was found within the smallest height class (10–50 cm), where it significantly correlated with four variables of light (GFr, Op, ISF, TSF) with a high level of significance (p < 0.01).

3.3.2. Lateral Crown Growth

To observe morphological development at the level of crown lateral growth, three parameters (branch length (Bl), crown width 1 (CW1), crown width 2 (CW2)) and three indexes (relative crown width (RCW), crown projection (CP), index of spatial distribution (ISD)) were tested. All of them were significantly correlated with light conditions within observed height classes (

Table A3. Significantly proved regression models of light intensities and morphological parameters and indexes expressing lateral growth of Norway spruce natural regeneration under size-dependent evaluation.

Height Class	Regression Model	Variable (x)	R2	p	rs
51–100 cm	Bl = 19.2155 + 228.5272 × x − 540.4043 × x2	ISF	0.11	0.0256 *	−0.3692 *
51–100 cm	CW1 = 34.4871 + 408.7886 × x − 978.2429 × x2	ISF	0.13	0.0124 *	−0.4541 *
51–100 cm	CW2 = 76.9358 − 0.3667 × x − 0.0132 × x2	GFr	0.09	0.0266 *	−0.3917 *
51–100 cm	CW2 = 79.166 − 0.5299 × x − 0.01 × x2	Op	0.09	0.0235 *	−0.3987 *
51–100 cm	CW2 = 59.6086 + 102.3799 × x − 358.9949 × x2	ISF	0.11	0.0124 *	−0.4040 *
51–100 cm	RCW = 1.8688 − 0.051 × x + 0.0006 × x2	GFr	0.26	0.0001 ***	−0.5712 *
51–100 cm	RCW = 1.9707 − 0.0579 × x + 0.0007 × x2	Op	0.28	0.0000 ***	−0.6031 *
51–100 cm	RCW = 1.8489 − 4.0591 × x + 3.2983 × x2	ISF	0.37	0.0000 ***	−0.6643 *
51–100 cm	CP = 672.2795 + 33124.752 × x − 80280.9727 × x2	ISF	0.12	0.0182 *	−0.4297 *
101–150 cm	Bl = 561.0136 − 3200.0795 × x + 5035.5044 × x2	ISF	0.59	0.0193 *	−0.5000 NS
101–150 cm	CW1 = 640.0183 − 3323.1949 × x + 5071.6685 × x2	ISF	0.48	0.0331 *	−0.5091 NS
101–150 cm	CW2 = 645.9867 − 3606.0909 × x + 5750.5151 × x2	ISF	0.59	0.0251 *	−0.4455 NS
101–150 cm	RCW = 6.4675 − 36.6654 × x + 58.6871 × x2	ISF	0.71	0.0120 *	−0.4727 NS
101–150 cm	CP = 1.0569 × 105 − 6.2668 × 105 × x + 9.8709 × 105 × x2	ISF	0.61	0.0180 *	−0.5000 NS
˃151 cm	RCW = 3.6175 − 0.2223 × x + 0.0042 × x2	Op	0.18	0.0409 *	−0.5015 *
˃151 cm	RCW = 1.0578 − 1.698 × x + 0.336 × x2	DSF	0.55	0.0000 ***	−0.7449 *
˃151 cm	RCW = 1.6269 − 4.1525 × x + 3.1664 × x2	ISF	0.29	0.0014 **	−0.5806 *
˃151 cm	RCW = 1.129 − 2.0758 × x + 0.8032 × x2	TSF	0.56	0.0000 ***	−0.7423 *
˃151 cm	ISD = 5.5782 − 5.3475 × x − 2.0348 × x2	DSF	0.32	0.0008 ***	−0.5711 *
˃151 cm	ISD = 10.0399 − 32.7416 × x + 44.103 × x2	ISF	0.16	0.0267 *	−0.4280 *
˃151 cm	ISD = 5.9229 − 7.584 × x + 1.7462 × x2	TSF	0.32	0.0008 ***	−0.5583 *

Note: Bl: branch length (cm); CW1: crown width 1; CW2: crown width 2; RCW: relative crown length; CP: crown projection (cm²); ISD: index of spatial distribution; GFr: gap fraction; Op: canopy openness; DSF: direct site factor; ISF: indirect site factor; TSF: total site factor; p < 0.05 (*), p < 0.01 (**), p < 0.001 (***); p ˃ 0.05 (^NS).

in Appendix A). In general, all observed parameters and indexes (where crown width was situated as a numerator within indexes) demonstrated a strong negative correlation with light intensities that expressed an initiation of the lateral growth under low light and its suppression under high light. The parameters Bl and CW1 and the index CP were correlated with light intensities only in two cases each (throughout the whole gradient of height classes), and thus they were rarely correlated with light compared to other observed parameters and indexes of lateral crown growth. This significance was established for all three in height classes 51–100 cm and 101–150 cm. For a similar range of natural regeneration (height ranging from 51 to 150 cm), the correlation of CW2 was also proven; however, four regression models were proven for this parameter. The ISD significantly correlated with light conditions only when the highest class of juveniles was considered (>151 cm); however, for this height class, the correlation was proven by three significant regression models. The highest level of correlation was demonstrated by RCW, as eight significant regression models were proven for this index, which was correlated with light conditions within a wide range of dimensions—for juveniles taller than 51 cm, in particular (i.e., the height classes 51–100, 101–150, and >151 cm). This index was especially appropriate for the tallest juveniles (height class >151 cm), where four significant regression models were found.

These results clearly demonstrate that photo-morphological plasticity at the level of the lateral growth of the crown has not yet developed within the smallest height class (10–50 cm) of Norway spruce natural regeneration. In this height class, no significant influence of light conditions on observed parameters and indexes of the lateral crown growth was established.

3.3.3. Relative Crown Length and General Shape of the Crown

The number of established regression models that covered all observed height classes showed the highest capability to acclimation at the level of crown length and general crown shape. Cl and RCL were shown to be correlated with light intensities within three height classes (10–50 cm, 51–100 cm, >151 cm) through 8 and 10 significant regression models, respectively. The correlation of CWLR was demonstrated within all four height classes through 16 significant regression models. Since Cl was a parameter and RCL an index where the crown length was situated as a numerator of the index, both showed the same trend within the light gradient. Thus, the crown length was positively correlated with light intensities, which means it was suppressed under low light and enhanced under high light. Compared to the photo-morphological plasticity of lateral growth that was not found within the smallest dimension of juveniles of 10–50 cm (see Section 3.3.2), the photo-plasticity of the crown length was shown even in this initial stage of the natural regeneration. In particular, RCL was frequently found to be significantly affected by light intensities (five regression models) in the smallest height class. Furthermore, three models both for Cl and CWLR were also proven for the smallest juveniles (

Table A4. Significantly proved regression models of light intensities and morphological parameters and indexes expressing the crown length of Norway spruce natural regeneration under size-dependent evaluation.

Height Class	Regression Model	Variable (x)	R2	p	rs
10–50 cm	RCL = 0.5913 − 0.0065 × x + 0.0003 × x2	GFr	0.08	0.0051 **	0.3511 *
10–50 cm	RCL = 0.5899 − 0.0063 × x + 0.0003 × x2	Op	0.08	0.0046 **	0.3528 *
10–50 cm	RCL = 0.5333 + 0.4538 × x − 0.304 × x2	DSF	0.05	0.0122 *	0.2499 *
10–50 cm	RCL = 0.5936 − 0.549 × x + 2.2822 × x2	ISF	0.08	0.0051 **	0.3372 *
10–50 cm	RCL = 0.5244 + 0.4743 × x − 0.2612 × x2	TSF	0.06	0.0084 **	0.2685 *
10–50 cm	CWLR = 1.1184 + 0.0573 × x − 0.0019 × x2	GFr	0.06	0.0455 *	−0.2854 *
10–50 cm	CWLR = 1.1229 + 0.0563 × x − 0.0018 × x2	Op	0.06	0.0436 *	−0.2899 *
10–50 cm	CWLR = 1.1199 + 4.9663 × x − 14.5085 × x2	ISF	0.07	0.0447 *	−0.2868 *
10–50 cm	Cl = 12.868 − 0.1274 × x + 0.0082 × x2	GFr	0.04	0.0409 *	0.2121 *
10–50 cm	Cl = 12.7175 − 0.1159 × x + 0.008 × x2	Op	0.04	0.0354 *	0.2149 *
10–50 cm	Cl = 12.7194 − 11.4916 × x + 68.5436 × x2	ISF	0.05	0.0195 *	0.2264 *
51–100 cm	CWLR = 2.6967 − 0.0429 × x − 0.0001 × x2	GFr	0.14	0.0041 **	−0.5547 *
51–100 cm	CWLR = 2.8315 − 0.0496 × x − 6.3876 × 10−5 × x2	Op	0.16	0.0018 **	−0.5969 *
51–100 cm	CWLR = 1.9634 − 2.0859 × x + 0.6197 × x2	DSF	0.08	0.0341 *	−0.3290 *
51–100 cm	CWLR = 2.2819 + 0.6117 × x − 11.3368 × x2	ISF	0.26	0.0001 ***	−0.6968 *
51–100 cm	CWLR = 2.2373 − 3.9869 × x + 3.5742 × x2	TSF	0.10	0.0180 *	−0.3747 *
51–100 cm	RCL = 0.5197 + 0.9095 × x − 0.897 × x2	DSF	0.08	0.0331 *	0.2753 *
51–100 cm	RCL = 0.8403 − 2.2948 × x + 5.8313 × x2	ISF	0.11	0.0181 *	0.3168 *
51–100 cm	RCL = 0.4589 + 1.3188 × x − 1.5389 × x2	TSF	0.10	0.0247 *	0.2902 *
51–100 cm	Cl = 41.4323 − 29.6034 × x + 180.7043 × x2	DSF	0.14	0.0065 **	0.2624 *
51–100 cm	Cl = 12.3012 + 155.6449 × x − 143.2971 × x2	ISF	0.07	0.0449 *	0.2665 *
51–100 cm	Cl = 38.9685 − 14.9096 × x + 162.9465 × x2	TSF	0.14	0.0053 **	0.2773 *
101–150 cm	CWLR = 13.8824 − 0.8997 × x + 0.0157 × x2	GFr	0.58	0.0322 *	−0.4182 NS
101–150 cm	CWLR = 15.3481 − 1.0121 × x + 0.0178 × x2	Op	0.60	0.0292 *	−0.4364 NS
101–150 cm	CWLR = 10.3003 − 56.1753 × x + 84.8293 × x2	ISF	0.62	0.0087 **	−0.5546 NS
˃151 cm	CWLR = 6.399 − 0.397 × x + 0.0072 × x2	GFr	0.34	0.0013 **	−0.6320 *
˃151 cm	CWLR = 6.1167 − 0.3702 × x + 0.0066 × x2	Op	0.33	0.0014 **	−0.6173 *
˃151 cm	CWLR = 1.8672 − 5.1136 × x + 5.0869 × x2	DSF	0.62	0.0000 ***	−0.7537 *
˃151 cm	CWLR = 2.3706 − 4.9823 × x + 0.7249 × x2	ISF	0.30	0.0012 **	−0.5762 *
˃151 cm	CWLR = 2.0488 − 6.1746 × x + 6.5934 × x2	TSF	0.63	0.0000 ***	−0.7665 *
˃151 cm	RCL = 0.2823 + 0.0178 × x + 1.239 × 10−5 × x2	GFr	0.19	0.0119 *	0.4051 *
151 cm +	RCL = 0.189 + 0.0264 × x − 0.0002 × x2	Op	0.17	0.0206 *	0.3589 *
˃151 cm	Cl = −14.4542 + 1462.7692 × x − 2300.0565 × x2	DSF	0.30	0.0022 **	0.5230 *
˃151 cm	Cl = −83.1842 + 1988.695 × x − 3314.3855 × x2	TSF	0.30	0.0022 **	0.5171 *

Note: Cl: crown length (cm); RCL: relative crown length; CWLR: crown width–length ratio; GFr: gap fraction; Op: canopy openness; DSF: direct site factor; ISF: indirect site factor; TSF: total site factor; p < 0.05 (*), p < 0.01 (**), p < 0.001 (***); p ˃ 0.05 (^NS).

in Appendix A).

CWLR, which aggregated both the lateral growth of the crown and its length, showed particular importance—it expressed the general shape of the crown. CWLR highly correlated with light intensities as 16 significant regression models through all dimensions of juveniles were proven (Table A4 in Appendix A). The lateral growth was situated as a numerator and the crown length as a denominator within the index; CWLR demonstrated a significantly negative correlation with light intensities (when the lateral growth alone demonstrated the same negative correlation while the crown length exhibited a positive correlation). This means that under low light intensities, juveniles were creating umbrella-shaped crowns (i.e., a high value of CWLR) through an acceleration of the lateral growth and shortening of the crown length. An exception was seen in this pattern only within the smallest juveniles (10–50 cm) where the CWLR changes were created through crown length acclimations. This was due to a lack of significant photo-plasticity in the lateral growth found for the smallest juveniles (see Section 3.3.2), while the acclimations of crown length within the smallest height class were broadly demonstrated by the significant plasticity of both Cl and RCL.

3.3.4. Stem Diameter and Height–Diameter Ratio

A significant relation between light intensities and D within all observed height classes was not established, fully proving previous findings of no relation of this natural regeneration parameter on light (see Section 3.2). However, while no effect of light intensities on HDR was found within general evaluation (under size-independent evaluation), these analyses showed the effect of the light on HDR in three height classes (

Table A5. Significantly proved regression models of light intensities and the height–diameter ratio of Norway spruce natural regeneration under size-dependent evaluation.

Height Class	Regression Model	Variable (x)	R2	p	rs
51–100 cm	HDR = 4.7679 + 0.0172 × x + 0.0018 × x2	GFr	0.13	0.0054 **	0.4334 *
51–100 cm	HDR = 3.9528 + 0.0839 × x + 0.0004 × x2	Op	0.14	0.0045 **	0.4442 *
51–100 cm	HDR = 3.6928 + 9.0674 × x + 1.4289 × x2	ISF	0.17	0.0015 **	0.4481 *
101–150 cm	HDR = −31.7919 + 253.2491 × x − 414.6227 × x2	ISF	0.65	0.0400 *	0.3091 NS
˃151 cm	HDR = −2.9996 + 0.6657 × x − 0.0117 × x2	GFr	0.20	0.0151 *	0.4995 *
˃151 cm	HDR = −1.6448 + 0.5431 × x − 0.009 × x2	Op	0.20	0.0121 *	0.4875 *
˃151 cm	HDR = 5.4822 + 1.3029 × x + 7.8746 × x2	DSF	0.23	0.0065 **	0.4172 *
˃151 cm	HDR = 7.439 − 18.9085 × x + 50.8065 × x2	ISF	0.16	0.0254 *	0.3861 *
˃151 cm	HDR = 5.5556 − 0.2287 × x + 12.4244 × x2	TSF	0.23	0.0057 **	0.4322 *

Note: HDR: height–diameter ratio; GFr: gap fraction; Op: canopy openness; DSF: direct site factor; ISF: indirect site factor; TSF: total site factor; p < 0.05 (*), p < 0.01 (**); p ˃ 0.05 (^NS).

in Appendix A). No relationship between light and HDR was found in the smallest height class (10–50 cm), while the most distinctive relationship was found for the tallest height class (˃151 cm). This finding clearly confirms the correlation of total height (H) and light intensities which was shown for the same height class (˃ 151 cm), while within other height classes the correlation of the H and the light was missing and was represented (within the height growth) through a distinctive correlation of Tl (see Section 3.3.1). In general, regardless of particular parameters or indexes, height growth was clearly correlated with light conditions. There was no correlation between stem diameter in this or other analyses. Thus, these proven relations between HDR and light conditions took place through acclimations of the H to light intensities while D remained unchanged throughout the changing light gradient. This was confirmed by a positive correlation between HDR and light when all significant regression models exhibited an increment in HDR with the increment in light. This shows the same trend of correlation as with height growth.

3.3.5. Crown Volume

No effect of light intensities on the crown volume was found regardless of which algorithm for the crown volume calculation was used for the testing (crown volume according to equilateral cone or crown volume according to quadratic paraboloid).

4. Discussion

Other authors have established the same positive correlation between light conditions and height growth of Norway spruce natural regeneration [19,45]. The lower correlation of total height (H) compared to the terminal length (Tl) can be explained in that the response of height growth to particular light conditions is not immediate. On the contrary, it takes time from when the height growth is again in full compliance with light conditions, as has been established by Nilson and Lundqvist (2001) [46], Robakowski et al. (2004) [41], and Chrimes and Nilson (2005) [19]. This is in compliance with the results of Diaci and Firm (2011) [3], who demonstrated a significant correlation between Tl and light intensity both for Norway spruce and silver fir (Abies alba Mill.) seedlings, but did not find a significant correlation with H. Tl has a higher probability of being adaptive to current light conditions, while H reflects the history of height growth through changing light conditions. Although light conditions were reasonably stable for a long time (at least for 15 years or more) in this study, some changes did occur (due to overstory growth and overstory canopy development). A similar situation can be concluded for the index RHG, as RHG plasticity was more carried out through Tl changes than H changes. The parameter of Tl was also proven by Jaloviar et al. (2013) [16] as being associated with light intensities both for Norway spruce natural regeneration and silver fir underplantings. By contrast, they found no significant effect of light on the RHG of either tree species [16].

A high level of photo-morphological plasticity in this study was found for lateral crown growth. However, within the evaluation involving height classes, none of those adaptations (at the level of parameters Bl, CW1, and CW2 or indexes RCW, CP, and ISD) were employed in the smallest height class (10–50 cm). Similar results were established by Duchesneau et al. (2001) [37] for balsam fir (Abies balsamea (L.) Mill.) when they found a generally higher level of correlations of morphological plasticity and light for the height class of 100–200 cm compared to the height class of 50–100 cm. Similar findings were confirmed for other conifer shade-tolerant tree species [28,44]. However, a continual photo-morphological acclimation through a gradient of the dimension of natural regeneration can be limited, because as Wright et al. (1998) [15] found, a height of 6 m can be a threshold for many different tree species when the described trend starts to weaken or even disappears.

It was widely established that the plasticity of crown morphology plays a key role in the adaptation ability to survive under a wide range of light intensities, and thus this feature is uniquely distinctive for shade-tolerant tree species [12,13,47,48]. Crown morphology has a crucial effect on light use efficiency and thus on the competitive strength of an individual [44,48,49,50]. This competitiveness within natural regeneration can increase if light availability becomes limiting due to overstory shelter and can vary according to the ratio of diffuse and direct light in understory microsites [8]. Consequently, shade-tolerant tree species have the ability to adapt their growth just through acclimations via photo-morphological plasticity of the crown, shoots, and needles or leaves [51,52]. Thus, shade-tolerant conifer tree species crown morphology varies from conical under high light conditions to umbrella-shaped under low light conditions [18,53,54]. This morphological development consists of acclimations via a reduction in height growth and crown length [29,55] and initiation of crown lateral growth to maximize carbon gain under shade [41,54,56]. In addition, the described pattern is aimed at reducing self-shading within the crown [18,55,57]. As light intensities increase, height increment is initiated, making the light more available for the lowest whorls, which initiates the establishment of a continual conical-shaped crown [58]. Acclimation ability (expressed as different increment allocation patterns resulting in different architecture) varies between tree species and plays a key role in photosynthetic efficiency (essential under low light conditions) and thus offers a competitive advantage to particular tree species [22,28,49,52].

Within the 10–50 cm height class, juveniles invest in vertical growth (proven significant correlation of Tl and light) to reach the best social position through height increment with no lateral crown growth having yet taken place. However, to optimize increment allocation (i.e., carbon gain) and to avoid self-shading, the ability to reduce crown length is already fully developed even in this early stage of natural regeneration (broadly proven by significant correlations of Cl, RCL, and CWLR). As Takahashi (1996) [43] stated, the whole process is a balance between investment into height growth to reach future profit and investment into surviving when the current total height is preserved, as is broadly shown in other studies [18,22,29,41]. Although the index CWLR integrates both lateral crown growth and crown length, it significantly correlates with light. This takes place through morphological acclimation of crown length and can be concluded from the established plasticity of both Cl and RCL in the smallest height class when no proof of morphological plasticity for lateral crown growth was found. The two described patterns of photo-morphological acclimations (height growth and crown length) were also present in juveniles taller than 51 cm when the ability to employ photo-morphological acclimations to crown lateral growth was supplemented.

As a consequence, other studies considered both ADR and RCL as indexes for expressing the level of shade tolerance, and deducing the light intensities under which juvenile shade-tolerant trees have been growing [29,41].

The crown length is considered the essential photo-morphological characteristic for distinguishing between shade tolerance and semi-tolerance, since an absence of this acclimation for shade semi-tolerant tree species, such as eastern white pine (Pinus strobus L.), has been shown [27,44,55].

Our finding of the correlation between HDR and light verified the classification of Norway spruce as a shade-tolerant tree species because this level of plasticity is also considered a typical ability of shade-tolerant tree species [59] and is absent in shade semi-tolerant tree species [27,44]. Although we did not prove a significant correlation between HDR and light conditions in the general evaluation, it was established for juveniles taller than 51 cm when evaluated according to height classes.

Thus, only the initial stage of natural regeneration did not exhibit acclimation of HDR, whereas a correlation of HDR with all observed light variables for the highest stratum of natural regeneration (height class >151 cm) was established. The observed plasticity of HDR was created through acclimations of height growth when stem thickness was not related to light intensities. No correlation between D and light can be explained by intense intraspecific competition within natural regeneration. To some extent, the increment is invested into height growth to reach a favorable position in relation to neighboring trees. To a greater extent, it is invested into adaptations of the crown to make photosynthesis under shade as effective as possible to reach a high probability of survival [18,43,58] while investment into stem radial growth is reduced and affected by the density of juveniles. In such conditions of dense natural regeneration, the intraspecific competition was a critical factor when compared to the results of Coates and Burton (1999) [60], who established the significant influence of light on D for five different conifer tree species five years after planting. It is clear that in young planted seedlings, competition was substantially lower compared to natural regeneration in our study. Furthermore, the crucial influence of density on D was also described by Chrimes and Nilson (2005) [19], as well as Petritan et al. (2012) [61], who established the significance of this finding for conifers rather than broadleaf species. Thus, the correlation of stem radial increment and light intensities is fully revoked when intraspecific (or interspecific) competition appears; only the pattern of height growth and its dependency on light remains. Consequently, HDR plasticity, fully arranged through height growth plasticity, occurs. This also confirmed the finding of the independence of D on light intensities in the natural regeneration of other conifers, as Kučeravá et al. (2012) [62] found the same for silver fir and Claveau et al. (2002) [44] observed the same for four conifer species from a group of five. On the other hand, in this study, only D was tested when there were no available data about radial increment. Radial increment can be more sensitive to certain light conditions, as was established by Duchesneau et al. (2001) [37] for balsam fir or Greis and Kellomäki (1981) [63] for Norway spruce.

This study established only a weak relationship between regeneration density and light. Lunquist and Fridman (1996) [36] and Hasenauer and Kindermann (2002) [64] also did not find a correlation between Norway spruce regeneration density and the basal area of the overstory. Similarly, Szewczyk and Szwagrzyk (2010) [65] did not find a correlation for Norway spruce either, while for silver fir, they found a significant correlation. Jaduď et al. (2014) [66], who evaluated the influence of light intensity on the density of seedlings after directseeding (using the same amount of seeds within various treatments), found no correlation between light intensities and density for Norway spruce or silver fir. We established that there was a fragile relationship for natural regeneration up to a maximum of 1 m in height, which corresponds well with the findings of Roženbergar et al. (2007) [67]. They did not find a significant correlation between light and the density of silver fir and observed a continual decrease in its density with its increasing height until a height of 110 cm when density developed asymptotically to axis x. Thus, the height of approximately 1 m can work as a general threshold when the (intraspecific) competition (resulting to autoreduction) starts to dominate, and the influence of light (which slightly affects the density of juveniles up to 1 m and is reflected in height growth as well) is retracted. By contrast, the regeneration index (RI) where dimension was considered was significantly correlated with light in this study, which was confirmed by Grassi et al. (2004) [68] both for Norway spruce and silver fir. The importance of dimension (height) on the development of density has been clearly demonstrated. Furthermore, the importance of microclimates in microsites is an important factor, as shown by Kupferschmid and Bugmann (2005) [69], Hunziker and Brang (2005) [70], and Kathke and Bruelheide (2010) [71]. More studies focused on the complex mechanisms of various factors affecting Norway spruce natural regeneration density should be conducted in the future.

5. Conclusions

The results of our research indicate that Norway spruce juveniles have immense photo-morphological plasticity in lateral crown growth, crown length, and height growth. The ranges of plasticity (expressed by the maximal and minimal values of RHG and ADR) for height growth and crown lateral growth acclimated to light conditions to a similar extent. However, the plasticity of lateral crown growth was absent in the smallest height class (10–50 cm). The capability for acclimations in crown length was established robustly throughout the whole range of natural regeneration dimensions. HDR showed a significant positive correlation with light intensities for juveniles from 51 cm in height. This shows the positive impact of overstory shelter on the mechanical stability of juveniles (making the HDR value smaller due to the shade of the overstory shelter) and is a representative example of nature’s automation (also known as biological automation) principles in tending in young forest stands. However, this means that when the overstory canopy is intensively opened (or even removed) by silvicultural intervention or by disturbance, the risk of mechanical lability of Norway spruce understory will occur if no further tending is applied in this natural regeneration.

The existence and intensity of photo-morphological acclimations firmly established Norway spruce as a shade-tolerant species. Thus, light conditions in the forest stand understory (which can be easily influenced by silvicultural techniques aimed at the forest stand overstory) can effectively encourage forest stand structural heterogeneity as Norway spruce juveniles follow light conditions with their growth. Moreover, through the use of appropriate silviculture techniques, it is possible to take full control of the growth vigor of Norway spruce juveniles. This can be effectively used for the transformation of Norway spruce monocultures to uneven-aged mixed stands, once it is deemed appropriate for Norway spruce to be one of the species in the future species composition of the forest stand. For such silviculture practice, two main growth regimes can be generally distinguished: suppressed and enhanced. These regimes are not distinguished by distinct thresholds but can be found at the level of ca. 0.2 TSF. From this intensity, both RHG and ADR progressively increase, whereas until ca. 0.15 TSF they very slightly increase with light (RCL increases constantly with TSF within the whole range). When it is required to suppress the vigor of Norway spruce juveniles (e.g., the target being to encourage advanced artificial regeneration of reintroduced tree species during Norway spruce monoculture transformation), light intensity should not exceed this threshold by too much (to support reintroduced species). When Norway spruce is to be encouraged to reach advanced growing stages earlier, the light intensity should be over 0.2 TSF. However, regardless of these two main growth regimes, the growing response to different light conditions is continual after the threshold is exceeded, and therefore the encouraged heterogeneity of the overstory canopy openness leads to high growth diversity and also a high structure of Norway spruce regeneration. Therefore, variability of patches of overstory canopy created by a combination of sections with a closed canopy, mid-open canopy, and small gaps can be favorable for this purpose. Regeneration density did not correlate well with light intensities (and if some weak trends were found, they demonstrated a negative correlation with light). These results demonstrated that the regeneration establishment is more an issue of appropriate forest microclimate, where perhaps both temperature and humidity regimes play key roles and light is the further additional variable, which indicates that a shelter-wood cut can be more favorable for regeneration establishment than the creation of gaps. And after it reaches certain growing stage of about 1 m in height, intraspecific competition and autoreduction start to be driving forces of its density.