Introduction 

Reaching up to 1.20 m in diameter and 35 m in height, wild cherry (Prunus avium L.) is a fast-growing, broadleaved tree species with a wide natural distribution in both Europe and Turkey. Due to its significant environmental (e.g., biodiversity) and economic (e.g., high-quality wood) importance, wild cherry is considered a valuable broadleaved tree species in Europe. The demand for its high-quality wood often exceeds the supply, resulting in higher prices in the market. Therefore, there is an increasing incentive to grow wild cherry in both Europe ([38], [18], [41]) and Turkey ([49], [51], [11]).

Wild cherry is found scattered in mixed deciduous and coniferous-deciduous forests in Europe and Turkey, growing as single individuals or in small groups. It prefers deep, moist but well-drained, fertile, and slightly acidic sandy loam to loamy soils ([49], [40], [38], [51], [11], [45]). In northern Turkey, cherry is mostly found on the mesic north- and east-facing aspects of the mixed deciduous forests of the Black Sea Region (BSR). The coastal belt of the BSR, where it occurs most frequently, is characterized by heterogeneous topography, oceanic climate, and high soil productivity. The average annual temperature of the belt is 14-15^oC, and the average annual rainfall is about 1000 mm ([4]).

Similar tree species are associated with wild cherry in natural forests in Europe (especially Germany - [47]) and Turkey. In the western BSR, it grows in the Castanetum phytoclimatic zone at low elevations ([49], [4], [51]). Wild cherry is found heavily mixed with sweet chestnut (Castanea sativa Mill.), maples (Acer spp.), European hornbeam (Carpinus betulus L.), eastern beech (Fagus orientalis Lipsky), and ashes (Fraxinus spp.). Scattered occurrences of wild cherry are frequent in the pure eastern beech stands of the Fagetum zones in the region. It is occasionally found with oaks (Quercus spp.) on drier sites ([49], [4], [51]). Wild cherry is a shade-intolerant species, making it very sensitive to competition from the surrounding trees. Therefore, shade-tolerant eastern beech frequently outcompetes and displaces wild cherry in natural settings, a situation very similar to that of the wild cherry and European beech (F. sylvatica L.) in Europe ([45]).

In addition to high-quality wood production, intensive plantations may help to conserve native forest resources ([8]). The key characteristics of a successful plantation include appropriate site selection and establishment techniques. Intensive silvicultural treatments accelerate stand development, channel the limited site resources to targeted species and individuals, and reduce rotation periods ([33]). Growing wild cherry in intensively managed plantations can also help reduce Turkey’s shortage of quality timber. Although the seed ecology and early seedling growth performances of different seed sources of wild cherry have been studied to a certain extent, no research has been conducted in Turkey so far on the possible survival and growth responses of young wild cherry seedlings to intensive silvicultural treatments ([11], [12], [14], [15]).

Poor seedling stock can dramatically reduce survival and field performance. Unlike with conifers, there has been little research on the key morphological characteristics associated with the performance of broadleaved tree species following out-planting ([21]). Various morphological characteristics, including root-collar diameter, shoot height, and sturdiness ratio (height-to-diameter ratio), are commonly used to evaluate the ability of hardwood seedlings to tolerate environmental and transplanting stresses ([21]). Identifying key morphological characteristics improving the survival of young wild cherry seedlings in the first period of establishment will certainly help ensure a successful plantation.

This study assessed the early effects of various combinations of intensive cultural treatments (including weed control, soil tillage, and fertilization) and seedling types on the survival, growth, and nutrition of one-year-old wild cherry seedlings planted on four different sites in the western Black Sea Region. The study also focused on the relationships between early seedling survival and various morphological characteristics, including initial diameter, height, and sturdiness ratio.

  Materials and methods  

Site description

The study was carried out on four different sites in the western Black Sea Region (BSR) of Turkey (Tab. 1). Three sites, the Karadeniz Ereglisi (Eregli), Bendere, and Akçakoca (Deredibi) sites, are in recently clear-cut, natural, pure or mixed eastern beech stands located in the vicinity of the Black Sea coast ([1], [2], [52], [53]). The overstorey is primarily a closed beech canopy. Sweet chestnut, maples, and wild cherry make up < 10% of the forest canopy ([1], [2], [52], [53]). The understorey is overgrown with purple-flowered rhododendron (Rhododendron ponticum L.). The average yearly temperature and precipitation are 13°C and 1100 mm, respectively ([1], [2], [52], [53]). The fourth location, Cumayeri, is inland and was formerly a degraded oak (Quercus spp.) site. Its average yearly temperature and precipitation are 13°C and 840 mm, respectively ([3]).

The soils of the three coastal sites are comparatively fertile and well-drained, and vary from sandy loam to clay, whereas the inland site has heavy, clayey soil with low drainage ([1], [2], [3], [52], [53]). The standard treatments employed by the General Directorate of Forestry prior to plantings were applied to all the sites in preparation for this study. In the fall of 2007, the sites were first raked in a broadcast manner to remove the existing woody vegetation (rhododendrons and oaks). Then, using a bulldozer carrying a brush rake or soil ripper, the sites were ripped down to the first one-meter soil depth in order to promote root growth.

Plant material

In the late summer of 2006, the Eregli Forest Management Directorate collected seeds from naturally open-grown mature (40 to 50 years-old) wild cherry trees. These trees were scattered throughout Halli and Gümeli (41^°05’09”N; 31^°28’00”E) between 400-800 m a.s.l. in the sub-province of Karadeniz Ereglisi, Zonguldak. Seedlings were grown from seeds under standard nursery practices at the Zonguldak Devrek Forest Nursery (41^°13’30”N; 31°57’35”E) during 2007. Seedbeds were irrigated as needed and fertilized monthly between May and August with 18-46-0 diammonium phosphate at the rate of 45 kg N and 115 kg P ha^-1. Herbaceous weeds growing in the nursery beds were removed by hand monthly during the growing season.

At the end of the growing season, a group of the cherry seedlings were lifted, root-pruned, and then transplanted into 4-L plastic pots. The seedlings were irrigated on an as-needed basis. A month later, a second group of bare-root seedlings were lifted and root-pruned similar to the first group. The roots of these were covered with burlap and kept moist until they were planted. By the end of 2007, all of the potted and bare-root seedlings had been carefully transferred to the experimental sites and planted.

Cultural treatments

Four treatments, including different combinations of weed control (WC), tillage (T) and fertilization (F), and finally the control (no treatment), were used for the study. For the weed-control treatment (WC), in the early spring of two consecutive years (2008 and 2009), herbaceous vegetation growing within a 50-cm radius of a given seedling stem was completely removed to the bare ground using a hand sickle. The treatment was repeated within the same growing season on an as-needed basis to control weed regrowth.

For the second treatment (WC + T), the soil within a 50-cm radius of seedlings was hoed by hand in the early spring for two growing seasons. This treatment basically removed the competing vegetation and tilled the soil around the seedlings in one step. As in the WC treatment, weeds that regrew within the same growing season were manually controlled when deemed necessary.

For the last and most intensive treatment (WC + T + F), seedlings received the same WC + T treatment; in addition, subsequent to hoeing, 15-15-15 NPK and triple super phosphate (TSP) fertilizers were applied once by hand to the tilled soil around the seedlings at 275 kg ha^-1 and 138 kg ha^-1 rates, respectively. Successive weed competition that occurred within the same growing season was eliminated manually.

Experimental design and statistical analysis

A factorial design within a randomized complete block design (RCBD) with four blocks (sites) was used for the experiment. The first experimental factor was the seedling type, occurring at two levels (potted and bare-root), whereas the cultural treatment was the second factor, occurring at four levels initially. However, all seedlings in the control treatment were killed off one year after planting due to severe herbaceous competition. To prevent this fact from swamping any treatment differences between the non-control treatments, the data were analyzed without the control treatment, except for the leaf nutrient analysis.

The potted and bare-root wild cherry seedlings were separately planted within two adjacent rows, constituting a row pair (Fig. 1). There were four pairs of seedling rows on each experimental site. The order of the seedling type (bare-root or potted) was randomly determined for each row pair prior to planting. Each row constituted an experimental unit for this study, containing 22-24 potted or bare- root seedlings planted with 3 x 3 m spacing. In total, 368 potted and 372 bare- root seedlings were planted for the study.

Fig. 1 - An illustration of the experimental layout of bare-root (B) and potted (C) one-year-old seedlings of wild cherry out-planted on four different sites in the western Black Sea Region of Turkey.

  Enlarge/Shrink   Download   Full Width  Open in Viewer

The four experimental treatments were randomly assigned to the four seedling row pairs for each experimental site (Fig. 1). The effects of the seedling type and cultural treatments on seedling survival, growth, and nutrition were analyzed with the two-way analysis of variance (ANOVA). P values < 0.05 were considered significant. Data were analyzed using the SAS package ([39]).

Measurements

For each treatment, the seedlings were measured for initial height and root-collar diameter (hereafter termed diameter) at the beginning of the experiment, and re-measured at the end of each growing season for two years. The percent of seedling survival was also determined for each treatment for each growing season. The relative growth rate of the seedlings in each treatment was measured using a formula for the first and second growing seasons ([36] - eqn. 1):

where RGR is the relative growth rate of a seedling from time 1 to time 2; V₁ is the seedling diameter (mm) or height (cm) at the beginning of the experiment; V₂ is the seedling diameter (mm) or height (cm) at the end of the first or second growing season.

For nutrient analysis in the first growing season, 15 seedlings on each seedling row (i.e., experimental unit) were randomly chosen in July 2008 to determine the treatment effects on seedling nutrition. However, only five to eight seedlings could be used for leaf sampling of the control treatment group due to low seedling survival at the time of the sampling. Eight to ten leaves from different crown positions (upper, middle and lower) were collected from each sample seedling. The leaf samples were air-dried, later ground with a coffee grinder. After grinding, leaf samples were dried at 80^oC and weighed in 100-200-mg aliquots for total C analysis, and 500-mg aliquots for analysis of N ([23], [53]). Leaf C and N concentrations were determined using a dry combustion method in a LECO CNS 2000 Carbon Analyzer (LECO Corp., St. Joseph, MI - [32], [53]). For nutrient analysis, plant tissue samples were digested with a mixture of nitric and perchloric acids ([23], [53]). Phosphorus concentrations were determined with a Spectronic Colorimeter. K and Ca were determined with a Jenway Flame Photometer ([43], [53]).

  Results 

Seedling survival and growth

One and two years after treatment (YAT), no significant interactions were detected between the cultural treatment and seedling type. Seedling survival, diameter, height, and relative growth rate 1 and 2 YAT did not significantly differ among the cultural treatments (Tab. 2). However, seedling type significantly affected the mean survival rate of the seedlings 1 and 2 YAT (Tab. 3), with the potted seedlings showing almost a 12% greater survival rate than the bare-root seedlings 1 and 2 YAT. The two seedling types did not dramatically differ in growth variables 1 and 2 YAT, except for second-year diameter. The potted seedlings were almost 35% greater in diameter than the bare-root seedlings 2 YAT (Tab. 3).

Leaf nutrient analysis

Similar to the findings of the survival and growth data, no significant interactions between the cultural treatment and seedling type were found for leaf nutrient analysis (Tab. 4). The seedling type had no significant effect on the concentrations of the leaf nutrients analyzed. Also, the effects of various cultural treatments, including the control, on leaf C, P, K, and Ca concentrations were not significantly different. However, the seedlings with the most intensive cultural treatment (WC+T+F) had a significantly greater (28%) leaf N concentration compared to the seedlings of the WC and control group (Tab. 4). The seedlings with the WC+T+F treatment had the lowest C:N ratio among those of all treatments. The mean leaf C:N ratios of the control and WC seedlings were significantly greater (33% and 44%, respectively) than those of the WC+T and WC+T+F seedlings (Tab. 4).

Seedling morphological characteristics

Among the three non-control cultural treatments, there were no statistically significant variations for seedling survival and growth 1 and 2 YAT. The relationships between the morphological characteristics and the survival of the bare-root seedlings were assessed using additional correlation and regression analysis. Relationships of the first- and second-year survival of the bare-root seedlings to the initial diameter, height, and height-to-diameter ratios were determined using simple regression and correlation analysis. Second-order regressions described the curvilinear relations of the initial diameter, height, and sturdiness ratio of the bare-root wild cherry seedlings with seedling survival 1 and 2 YAP (Fig. 2, Fig. 3).

Fig. 2 - Relationship between mean seedling survival rate (%) and initial seedling diameter (mm) for one-year-old wild cherry seedlings planted on four sites in the western Black Sea Region of Turkey one and two years after planting with regression equation, Pearson correlation coefficient, and significance level.

  Enlarge/Shrink   Download   Full Width  Open in Viewer

Fig. 3 - Relationship between mean seedling survival rate (%) and mean initial seedling height (cm) for one-year-old wild cherry seedlings planted on four sites in the western Black Sea Region of Turkey one and two years after planting with regression equation, Pearson correlation coefficient, and significance level.

  Enlarge/Shrink   Download   Full Width  Open in Viewer

The seedling diameters demonstrated a significant positive relationship with survival 1 and 2 YAP, with relatively high correlation coefficients (Fig. 2). First- and second-year seedling survival rates increased almost linearly with increasing seedling diameter up to diameters of 7-8 mm, yet tended to decline above this diameter range (Fig. 2).

Similarly to the first-year results for diameter, seedling height demonstrated a significant relationship with seedling survival 1 YAP (Fig. 3). Seedling survival increased with increasing height up to 60-70 cm, after which the relationship gradually turned negative. However, these two variables had no significant relationship 2 YAP (Fig. 3). Finally, the relationships between the sturdiness ratio and survival for both one- and two-year-old seedlings were not significant.

  Discussion and conclusions 

Wild cherry has a high demand for light, soil water, and nutrients; therefore, the species is highly sensitive to herbaceous weed competition in the first three years of establishment ([40], [24], [28], [13]). Unwanted vegetation substantially reduces the growth and survival of young cherry seedlings during establishment ([28], [30], [13]). This was clearly confirmed by the present study, which had total seedling mortality in the control treatment group (Tab. 2). Elimination of competing vegetation is therefore an important requisite to producing wild cherry individuals with diameters of 50-60 cm in a 50- to 60-year rotation ([24], [34]). However, the present study has shown that neither fertilization nor soil cultivation, nor a combination of the two performed as well as weed control in terms of additional survival and growth in the first two years (Tab. 2); therefore, they are not recommended. Similar results were obtained in Latvia ([6]) and in the Czech Republic ([9]), where various intensive cultural treatments, including mechanical and chemical weed control and tillage, did not produce significant growth differences for young wild cherry seedlings in the first three and five years following planting.

There are contradictory results in the previous studies of the effects of fertilization on the seedlings of broadleaved tree species ([22]). One group of these studies states that fertilization increases the seedling growth of broadleaved tree species ([5], [21], [22], [42]). However, another group reported that fertilization during the early establishment period does not have a major sustainable effect on tree seedling growth ([10], [29], [22], [8]) and is actually detrimental in some cases ([22]). The results of the present study were consistent with the findings of the second group.

The foliar nutrient levels of the cherry seedlings in the present study are similar to those of young eastern beech seedlings reported in previous studies that were carried out in the mesic, coastal part of the western BSR ([52], [53]). Based on foliar analysis of the present study, as well as the survival and growth data, we found no substantial evidence that added nutrients had a significant effect on the cherry seedlings, except for N (Tab. 2, Tab. 4), and this was not substantiated by the survival and growth data. The lack of effects of fertilization on cherry survival, growth, and nutrition suggests that the productivity of the experimental site was adequate. The mesic and coastal sites of the western BSR, where three-fourths of this experiment were carried out, are well known for their relatively high productivity ([4], [53]). The present case demonstrates that costly site operations, including fertilization, should be thoroughly justified before being applied to cherry planting sites. However, one should remember that these are early assessments and may change with future long-term data.

In the present study, the Cumayeri site was characterized by two extreme edaphic conditions. The soil was mostly waterlogged during the winter, whereas the water table fell quickly during the summer, leaving very dry and hard-to-penetrate soil. The lowest seedling survival and growth occurred on this site (data not shown), suggesting that soil moisture and aeration are essential and even more critical than soil nutrients for wild cherry survival and growth ([40], [19], [38]). These findings present the opportunity to test the effect of plowing as an option for site preparation. Previous experiments have demonstrated that plowing may enhance soil drainage, weed control, and root growth of tree seedlings, enabling the establishment of broadleaved tree plantations in clay soil ([25], [35]).

Care should be taken with total plantations of wild cherry, since they are more susceptible to diseases than mixed plantations ([44]). Mixing wild cherry with other broadleaved tree species, including ash (F. excelsior L.), is in fact recommended for enhanced productivity and disease control ([26]).

The superiority of potted seedlings over bare-root seedlings for tree seedling survival and growth is well documented ([50], [16]). The results of the present study were consistent with this finding (Tab. 4). The bare-root seedlings of broadleaved tree species commonly undergo a transplant shock, mostly due to drought and nutrient deficiency experienced following planting ([46]). The soil column surrounding the root system of potted seedlings protects them from environmental stresses (e.g., drought, freezing-thawing cycles, and transplant shock) and physical stresses (e.g., abrasion, crushing, and root stripping). Additionally, in the spring, the constant contact of the root with the soil might physiologically activate potted seedlings earlier than the bare-root seedlings, improving survival and growth ([37], [21], [22], [16]).

Using high-quality seedlings is an important prerequisite for successful plantation. The initial diameter has been reported to be the trait that was most significantly and positively related to early field survival and growth performance for many broadleaved tree species ([7], [21]). For example, in one Canadian study ([7]), one-year-old red oak (Quercus rubra L.) seedlings with large initial diameters (> 8-10 mm) exhibited greater growth than those with smaller diameters. The present study corroborated this for wild cherry, yet with a size limit (Fig. 2). The diameter corresponds closely with above- and below-ground features, including root volume, area, and biomass, that are correlated with the success of seedlings after outplanting ([7], [21]). Also, greater diameters indicate greater carbon storage and energy for broadleaved seedlings, thus enhancing survival until seedlings begin harvesting resources from the soil following outplanting ([20]). In the present study, the tendency of seedling survival to decline above a certain diameter range (7-8 mm) for wild cherry (Fig. 2) might suggest “the lack of balance in larger seedlings” ([48]).

Seedling height defines the photosynthetic and transpiration capability of seedlings and their competitiveness against weeds, and thus correlates well with seedling growth ([20], [16]). Kupka ([27]) stated that the initial height is a good estimate of wild cherry seedling survival during establishment. As they are sensitive to weed competition, young wild cherry seedlings that are taller can gain a substantial advantage over competing vegetation ([27]). Similar to findings of the present study (Fig. 3), the curvilinear relation of the initial height with survival has been reported for Q. serrata Murray and Q. acutissima Carruth. ([31], [17], respectively). Hashizume & Han ([17]) found that the survival of oak seedlings increased with initial height up to 100 cm, yet gradually declined for seedlings taller than 150 cm.

For successful wild cherry plantations, the quality and type of seedlings are important. Potted seedlings are to be preferred to bare-root seedlings for enhanced early survival and growth. Initial seedling diameter and height are effective indicators of early seedling survival. Selecting seedlings for planting of approximately 8 mm in diameter and 70 cm in height is therefore recommended for greater survival of bare-root seedlings of this broadleaved tree species.

Finally this study has been focused on the early development and on a short period of time; further work and long-term data are needed to confirm the results.

  Acknowledgments 

This work was supported by the Scientific and Technical Research Council of Turkey (TÜBITAK - grant number TOVAG COST 106O817). We thank the Bolu and Zonguldak Regional Directorates of Forestry of the General Directorate of Forestry, the Turkish Ministry of Environment and Forestry, for access to research sites and for their other support in this work. We also thank Mrs. K. C. Hollandsworth and Nuriye Peachy for editing this paper for English. Lastly, we express appreciation to all of the anonymous reviewers who took part in the revision process of this manuscript for making important contributions.

  References