Structural Elements and Morphological Characteristics of Pedunculate Oak ( Quercus robur L . ) in Young Even-Aged Stands of Spačva Forest

Morphological characteristics of trees play an important role in trees’ defence and adaptation to changing habitat conditions in the context of global climate change. Developed crown indicates a more developed root system [1, 2] and therefore a greater degree of adaptation to habitat and climate changes [3]. Vajda [4] determined that in the same habitat conditions the proportion of dead pedunculate oak (Quercus robur L.) trees is smaller in the group with large diameter at breast height and welldeveloped crowns, while Shifley et al. [5] identified the AbsTRACT


INTRODUCTION
Morphological characteristics of trees play an important role in trees' defence and adaptation to changing habitat conditions in the context of global climate change.Developed crown indicates a more developed root system [1,2] and therefore a greater degree of adaptation to habitat and climate changes [3].Vajda [4] determined that in the same habitat conditions the proportion of dead pedunculate oak (Quercus robur L.) trees is smaller in the group with large diameter at breast height and welldeveloped crowns, while Shifley et al. [5] identified the social position of trees as one of the significant factors that increase the likelihood of the decline of North American oaks from the red oak group.Dekanić [6] considered the morphological features of pedunculate oak trees to be an extremely important preparatory factor in the complexity of individual tree dieback, while size, diameter increment and basal area increment are a good indicator of its vitality [7][8][9] and an indicator of habitat change [10].
Previous research shows that in the process of dieback of penduculate oak, trees with larger crowns, i.e. greater assimilation surface, are more resistant, while their crowns become more vital as their increment increases.It is known that in general pedunculate oak trees with underdeveloped crowns are more likely of dieback.One of the main causes of such crown development and intensive dieback is, among other things, the insufficiently intensive thinning which failed to be conducted in young stands.Thinning is known to be a necessary and essential silvicultural operation, while absence of tending in young stands causes a high degree of tree legginess and small reduced broom-like crowns which are not resistant to biotic and abiotic factors.Such crowns indicate small root systems that have difficulties to adapt even to smaller (ten year) oscillations in water-air regime in the soil.Spačva Forest (Figure 1) comprises of diverse lowland habitats in which forests of pedunculate oak management class occupy 96% of the total area which amounts to almost 40,000 hectares.This makes it one of the largest coherent complexes of lowland pedunculate oak forests in Europe whose value cannot be measured monetarily because its environmental effect, primarily the anti-erosion and hydrological effect, is much more important than the economic one [11].Today's oak groves of Spačva are remnants of old-growth forest structures which were mostly clear-cut between 1880 and 1914, reducing the forest cover of Slavonija from 60% to 35% [12].Almost 75% of Spačva Forest consists of stands older than 80 years [13], while analysis of surface structure of pedunculate oak stands in Spačva Forest and its projection area shows that in the next thirty years regeneration will be very intense and consequently lead to a significant increase of areas with the stands of age-classes I and II.
Bearing in mind these facts and previous research on the structure and dynamics of the harvest of dead and declining pedunculate oak trees in Spačva Forest [14], as well as the dynamics of the decline of pedunculate oak trees depending on phytosociology and age [15], a real problem has appeared in that same areas in terms of financial and human resources and how to carry out all necessary work during both the regeneration phase and tending phase of the stand.Tending operations conducted in young stands are essential to their structure, quality and productivity; the intensity and methods of tending have to be harmonized with natural processes prevalent in the stand and with biological characteristics, environmental requirements and silvicultural traits of the tended species [16].
In light of these findings Croatian Forest Research Institute and Croatian Forests Ltd.Zagreb, in 2010 jointly proposed a scientific experiment on permanent experimental plots called "The Impact of the Intensity of Silvicultural Tending on Pedunculate Oak Dieback".The experiment was conducted in the area of Forest Administration Vinkovci, Forest Office Vinkovci, Management Unit Kunjevci.
Since very little research has been done on the morphology of young even-aged stands of pedunculate oak both in Croatia and globally, and since there is little data available in the literature, except on the features of growth and the development of crowns in the youngest stands of age-classes I, II and III [17][18][19][20], the purpose of this article is to contribute to the knowledge about those values.Therefore, along with other data on this study, the article includes rarely measured and published data on tree crown radius, first branch height and other features, all with the aim of better knowledge and understanding of relations within young pedunculate oak stands.

MATERIALs AND METHODs
The process of creating surface structure projection area started with the analysis of data from the management plan.Based on the management plan data 13 management units with predominant pedunculate oak management class were selected, covering the area of Spačva Forest managed by Forest Administration Vinkovci (Table 1), and were then used for further analyses.
The results from the data analysis were used for creating surface structure projection area for the studied management units for the period of the next 140 years (2011-2150) (Figure 2).
This projection has shown that with current methods of natural regeneration of pedunculate oak stands under the crown canopy of old trees already by the year 2020 the surface area of young stands is expected to increase; the area of the age-class I will reach maximum value in 2044 on a total of 18,000 ha or 43% of the studied area, while the surface of age-classes I + II will reach maximum value in 2050 on a total of 27,000 ha or 64% of the studied area.
As  3).All the plots were established in a typical pedunculate oak-hornbeam forest (Carpino betuli-Quercetum roboris typicum Rauš 1971).Preliminary selection of experimental plot locations in each subcompartment was made by terrain reconnaissance along with the digital terrain model created for the purpose of removing the effect of the microrelief.First the plots were temporarily staked out by using the already established network of silvicultural lines and paths and their position was recorded with a GPS device.After importing the preliminary locations of plots into the ArcMap program, the plots were arranged so that all plots are within 50 cm of the altitude range of the terrain.The final locations of plots were permanently marked by wooden stakes containing repetition marks and the number of the plot (Figure 4).The size of each of the 20 staked out plots is 33 x 35 m (Figure 5).The area of each plot is 0.12 ha (Table 2).
On each plot, 30 pedunculate oak trees were permanently marked with orange plastic tags for log classification (Figure 6).In total 600 trees were marked.The selection of trees was done according to the phenotypic and spatial criteria.Phenotypically best trees were chosen which are completely healthy and without noticeable damage (mechanical, game damage, barkpeeling damage, bitten off shoots).The trunks of those trees had to be of good quality (straight, without rowlocks or thick branches), while the trees had to be the highest in the vertical structure.They also had to be grown from seeds and had to have well-developed crowns (crown tops in the upper third of the crowns).The older trees that represented pre-growth were not selected.By using the spatial criterion 10 pedunculate oak trees, evenly distributed on the surface with triangular spacing between the selected trees, were chosen on each line.Depending on the field conditions the distance was 2-3 meters.
In each subcompartment two repetitions composed of three experimental plots called "Standard practice", "Treatment" and "Control" were set up, with the exception of subcompartment 32A in which an additional plot called "Reduction" was set up as well.
On "Standard practice" plots, procedures in accordance with current standards in practice were carried out.In this process it is assumed that the number of oak tree trees was not reduced or that it was reduced only minimally.
On "Treatment" plots the number of oak tree trees was reduced to about 1000 future trees per ha with spacing On "Control" plots no treatment was conducted and the plots were left to develop naturally.The measured values will be used for comparison with other plots.
In subcompartment 32A the "Reduction" plot was established on which high quality pedunculate oak trees were left with spacing of about 1.5 m, while all competitors were removed by cleaning, carried out on the entire surface of the line.All other pedunculate oak trees were removed, regardless of whether or not they obstructed the selected trees.
After the establishment and marking of permanent experimental plots, the selection of future trees and the conducted cleaning, measurements were carried out.The measurements were carried out in the spring of 2011.On each of the three inner lines two cross-sectional diameters at breast height, trunk length (height of the beginning of the crown), the height of the widest part of the crown, two crown radius and the total height of the tree were measured on marked pedunculate oak trees.
On two subplots of 5x5 m in size, the total measurement of trees was conducted by tree calliper gauge millimetre.Heights (with a bar for measuring height in cm) and diameters at breast height were also measured on the subsample of pedunculate oak and common hornbeam trees for the purpose of creating height curves.The diameter of trees which did not outgrow their breast height was measured at the half of tree height (0.5 H).The obtained data were analysed using Microsoft Office 2010 package and STATISTICA 7.1 [21].In order to eliminate the impact of outliers and extreme values the median was taken as the mean value.

REsULTs WITH DIsCUssION
The analysis and processing of filed data provided basic elements of the structure which are shown in Tables 3 and 4 which offer interesting conclusions.
The number of pedunculate oak trees as compared to hornbeam (Table 3) is not optimal; in young stands of pedunculate oak and common hornbeam (3-10 years) the total number of plants ranges from 35,000 to 40,000 per ha, out of which pedunculate oak trees should amount to 24,000-32,000 trees per ha and common hornbeam to about 18,000 trees per ha, along with other accompanying plant species [22].
Novotny et al. [23] have found that for the middle of first age-class there are 2175 trees per ha of pedunculate oak with a basal area of 9.47 m 2 •ha -1 , 2850 trees per ha of common hornbeam with a basal area of 8.23 m 2 •ha -1 (in total 5750 trees per ha and 20.34 m 2 •ha -1 ), while according to the guidelines for creating ecological-management types [24] in the first age-class there are 3495 trees per ha of pedunculate oak with a basal area of 9.4 m 2 •ha -1 , and 2460 trees per ha of common hornbeam with a basal area of 4.9 m 2 •ha -1 (in total 5955 trees per ha and 14.3 m 2 •ha -1 ).If we compare this data with the results from Table 3, we can clearly see big difference which indicates a significantly disrupted natural structure according to the number of pedunculate oak trees and their basal area.The situation is in a way satisfactory in subcompartment 32A, while in subcompartments 34A and 38A the values are above the optimum.Such results show the problematic future of these stands in terms of their stability and productivity.The value of the crown length in relation to the total height of pedunculate oak trees in subcompartment 32A is 74.3%, in 34A 53.5% and in 38A 54.3%.These values are higher than the results of previous research; Dubravac [25] and Hren and Krejči [26] determined the value of crown length to be 45% of the total tree height.
Figure 7 represents a box-and-whisker chart.Based on this figure, while taking into account the subcompartment age, it can be concluded that trunk length, i.e. trunk purity increases with age; in subcompartment 32A its average value is 1.35 m, in 34A 3.28 m and in 38A 4.85 m.
Crown length of the tree also increases with age (Figure 8).The mean values range from 2.78 m in subcompartment 32A to 5.66 m in subcompartment 38A.Dubravac [19] stated that the crown length of the age-class I amounts to 2.93 cm, but it should be emphasized that this referred a small sample (1 experimental plot, 18 trees).The same author [18], also on    9) the values for crown radius are 0.73 m for subcompartment 32A, 1.07 m for 36A and 1.30 m for 38A.From the created height curves (Figure 10) it is apparent that the height gain is intense and that the curve shifts "up and to the right".At the time of intensive height growth of the even-aged stands, a significant shift of the height curve is achieved [16], and comparison of the height curves for subcompartment 34A and 38A shows that in the older stand the trees of the same diameters have a higher height.Also, there is greater diameter dispersion with the increase of the stand's age.

CONCLUsION
The initial measurement on permanent experimental plots shows unsatisfactory number of pedunculate oak trees along with the high number of common hornbeam trees, which is in direct link with improper implementation of tending, thus endangering the future and the stability of these stands.
The overall basal area has a tendency of continuous growth in relation to the age of experimental plots.The initial increase is large, from 9.79 m 2 •ha -1 to 24.59, followed by a significant increase in the number of trees and an increase in diameter at breast height.The increase is then reduced, amounting to 28.76 m 2 •ha -1 on the oldest experimental plot with a significant reduction in the number of trees and the increase in diameter at breast height.
Trunk purity of young pedunculate oak trees was measured from the ground to the first branch height and is expressed in meters.Based on the measured data it can be concluded that with the increase in age the height at which the first branches appear also increases.In other words, trunk purity increases with age, therefore amounting to an average of 1.35 m on the youngest experimental plot, while on the oldest plot it amounts to an average of 4.85 m.
After conducting periodic surveys of the established plots by the year 2020 a sufficient amount of data should be collected whose processing, analysis and interpretation would provide guidelines for improving the future management of young pedunculate oak stands.

FIGURE 1 .
FIGURE 1. "Spačva" forest and investigated area on the LANDSAT satellite image from August 20, 2000.

FIGURE 5 .
FIGURE 5. Structure of permanent experimental plot

FIGURE
FIGURE 6. Marking future trees of pedunculate oak

TAbLE 1 .
part of the project "The Impact of the Intensity of Silvicultural Tending on Pedunculate Oak Dieback" 20 permanent experimental plots were established in the area of Forest Administration Vinkovci, Forest Office Vinkovci, FIGURE 2. Graphical overview of trending projection of I. and II.age class for the next 140 years List of managment units and number of subcompartments for pedunculate oak managment class used for analysis Management Unit Kunjevci, in subcompartments 32A, 34A and 38A (Figure

6 .
Marking future trees of pedunculate oak *Stand age during initial measurement in 2011

TAbLE 2 .
Basic information about subcompartments in which experimental plots are established.

TABLE 3 .
Number of trees and basal area in researched area the sample of one experimental plot, determined the mean value of 1.39 m for crown radius in the age-class I.In our research (Figure

TABLE 4 .
Morphological traits of pedunculate oak for researched area