Eight-year monitoring of the height growth and survivorship of seedlings of Pinus thunbergii Parl. planted with sand fence and bush hedge protection in a coastal sandy environment in Korea

ABSTRACT

This study aimed to examine the effect of distance at three levels (5 m, 15 m, and 25 m) of ranges of plots landwards from a sand fence and bush hedge protection margin on the height growth and survival of seedlings of Pinus thunbergii Parl. planted in a coastal sandy environment in Uljin, East Sea area of Korea. The purpose of the study was to investigate height growth and survival with distance from a bush hedge/sand beach boundary of pine seedlings in a coastal forest plantation. Seedling height growth and survivorship were measured annually in nine plots, with three plots each selected from three ranges of plots at distances of 5 m, 15 m, and 25 m landward from the bush hedge/sand beach boundary from 2007 to 2014, 8 years after planting. Annual averages in seedling height growth significantly reduced in the 2nd and 3rd years of growth but appreciably stabilized from the 4th to 8th year, indicating that the 2nd and 3rd years of seedling growth and survival in the study site were the most critical. We found no statistically significant differences in height growth and survivorship of the planted seedlings between the three ranges of plots at distances of 5 m, 15 m, and 25 m landward from the bush hedge/sand beach boundary in the microsite. This suggested that seedling height growth and survivorship were not significantly affected by the distance factor. This was attributed to the protective and stabilization functions of the bush hedges and sand fence barrier, which reduced near-surface wind damage to seedlings and fostered a relatively uniform height growth and survival of the planted seedlings in the microsite. Furthermore, wind speed (P = 0.014) and soil moisture at ≤ 12 cm depth (P = 0.014) were the most significant of the selected factors limiting the height growth of the seedlings in the study area, as they accounted for 94.4% (R² = 0.944) of the insignificant variability associated with the distance factor. The results provide supportive evidence that sand fence and bush hedge micro-windbreaks can be reliably used to protect, stabilize, and sustain relatively uniform height growth and survival of seedlings of P. thunbergii planted in a windy coastal sandy environment and in other locations with similar soil and climatic conditions.

KEYWORDS:

• Seedling height growth
• bush hedges
• distance from bush hedge
• coastal sandy environment
• wind speed

Introduction

The Korean peninsula has three coastal regions and the sand dunes display numerous representative characteristics of highly disturbed ecosystems (Kim 2005). Since the twentieth century, they became areas of focus for large-scale restoration and vegetation protection projects of mostly black pine (Pinus thunbergii Parl.), which is recognized to be hardier than other tree species in coastal environments of Korea (Choi et al. 2013). Coastal environments in Korea are remarkably prone to deleterious hydrometeorological extremes of windstorms and typhoons. Forestation of coastal land boundaries in natural disaster prone Asian countries is worthwhile because of the compelling evidence that vegetation and trees provide the most natural defenses to protect coastal hinterland from strong winds, storm surge propagation, and windblown sand (Tanaka et al. 2007; Irtem et al. 2009; Tanaka 2009).

Therefore, while planting trees to establish conservation coastal disaster control forests is increasingly recommended around the world, seedlings planted on coastal land boundaries are inevitably exposed to high risk of withering, especially at the early stage of growth, due to the deleterious effects of harsh coastal environmental conditions. It is argued that, to successfully establish and grow, trees planted in lacustrine and harsh coastal environments must be able to tolerate nutrient deficiency, low soil moisture, sand accretion, predation, high soil temperature, drought, salinity conditions, and the withering effects of strong winds (Maun 1994; Olafson 1997). Seedlings of P. thunbergii planted in the microsite of the present of study area are considered native to the Korean coast and exhibit naturalized adaptation capacity to the harsh coastal environment of sandy soils, maritime exposure, drought, salinity, and windy conditions. However, although seedlings of P. thunbergii have natural adaptation to these coastal environmental conditions, they invariably still need protection against the damaging effects of occasional strong winds, especially at an early stage of growth. In Korea, seedlings planted in coastal land boundaries are usually protected by sand fences and windbreak hedges.

The use of windbreak hedges in forestry practice is an old technique that has gained renewed interest in coastal forest plantation construction in Korea and beyond to encourage tree growth and avoid the development of damaged, fragmented, and irregular forest structure. Therefore, young seedlings planted in coastal land boundaries need some degree of natural protection close to the ground. In this light, we experimented with bush hedges inside sand fences to function as micro-windbreaks and provide maximum protection and stabilization to seedling growth. Plant hedges are installed outside sand fences because wooden sand fences deteriorate after a short period and, if not repaired or replaced, the remnants often become partially buried and lower than the average height of vegetation and young trees, and so cannot provide the required shelter if the main aim is to protect tree height growth (Soto 2012). Earlier studies by Caborn (1957) reported that the application of plant bush hedges or shelterbelts in forestry practice to safeguard the forest against damage by occasional severe winds is fundamental to ensure stability of stand, enhance stand microclimate, and promote the general condition of the forest.

Still, the use of wooden sand fences, for example, is arguably the most popular and effective site stabilization technique used where protection is required from wind and blown sand throughout the world (Philippe 1975; Woodhouse 1978; Watson 1985). Thus, the frequent occurrence of fences on human-altered coasts necessitates their incorporation into standard field investigations (Nordstrom et al. 2007b). Studies (e.g., Khalil 2008; Soto 2012) have mostly reported the effectiveness of mechanical hedge systems on vegetation restoration, growth, and distribution in sandy coastal and arid environments, but specific empirical studies to evaluate the effect of the natural bush hedges method in protecting the height growth of seedlings of P. thunbergii planted in a windy coastal sandy environment are still limited.

Therefore, the use of wooden and plant bush hedges to anchor the landscape prior to our coastal afforestation project was a plausible technique capable of forming the protective margins necessary in the early stage of forest development. Fundamentally, the use of hedges in forestry practice is mainly to create a desirable protective margin on open ground before afforestation is begun. Caborn (1957) suggested that the layout and structure of a shelterbelt should be such that it is moderately penetrable to the wind as this is desirable to encourage the trees to reach the greatest possible height growth by minimizing any disturbance in the general wind flow pattern near the ground. Bush hedge plants require minimal effort to maintain and they grow to form a strong protective barrier to reduce near-surface wind speed, windblown sand, and salt spray inflow into planted seedling microsite. In addition, bush hedge plants are able to modify microsite climate and soil environments by providing a natural and long-term buffer against wind erosion, thus encouraging trees to grow effectively inside them (Caborn 1957; Cleugh 1998).

We experimented with sand fence and bush hedges in the coastal forest construction because the wind blows particularly strongly in the study coastal area. Thus, we expected the installed sand fence and bush hedges to buffer near-surface wind speed in its direction of stronger movement, particularly in the more exposed front ranges of plots at 5 m. This was to ensure that seedling height growth in the front ranges of plots in the microsite were comparatively relatively uniform with those in the less exposed middle and behind ranges of plots at 15 m and 25 m, respectively. Therefore, a plausible approach in forestry practices to evaluate the protective effectiveness of the sand fences and bush hedges against near-ground wind speed was to study its effects on the growth characteristics of P. thunbergii planted in the windy coastal sandy environment. In this study, we have used seedling height growth data to function as a proxy to directly infer and qualify the protective function and stabilization effectiveness of the sand fence and bush hedges against damaging wind speed, to demonstrate its use as an appropriate and effective site preparation technique in black pine afforestation in windy coastal sandy environments.

The main aim of the study was to examine the effect of distance at three levels (5 m, 15 m, and 25 m) of ranges of plots landward from the sand fence and bush hedges on the height growth, root collar diameter (RCD), and survivorship of seedlings of P. thunbergii planted in a coastal sandy environment. The study also aimed to determine the most significant environmental variables affecting seedling height growth and to examine the effect of distance from sand fence and bush hedges protective margin on selected environmental factors in our study microsite. We hypothesized that there were no statistically significant differences in the height growth, RCD, and survivorship of the planted seedlings between the three ranges of plots at distances of 5 m, 15 m, and 25 m landward from the installed sand fence and bush hedges in the microsite.

Materials and methods

Study area

The study area is located in Uljin-gun, Gyeongsangbuk-do province in the East Sea area of South Korea (Figure 1). It falls within latitude 36^º38′00″ to 37^º06′04″ N and longitude 129^º04′30″ to 129^º29′30″ E. The study area has a mean annual rainfall of about 1274 mm and mean annual temperature of about 13 °C.

Figure 1. Geographical location of the investigation site.

Experimental design

In this coastal study area, a microsite was carved out and installed with bush hedge mats and sand fences as protective structures. The sand stabilization fence poles were 1.8 m in height and 4–5 cm or 6–7 cm in diameter and the material used was bamboo (Figure 2, right). The bamboo windproof fence was fortified with small round steel poles 1.5 m high at 4 m intervals along the sand fence/beach boundary to provide adequate stability. The size of the bush hedges plots was 5 × 5 m and 40 cm in height above the surface. A total of 92 plots with bush hedge protection were installed inside the sand fences and the bush hedge material was made of dry mats (Figure 2, left). A total of 25 4 year old seedlings of P. thunbergii were hand-planted at approximately 1 m intervals in each bush hedge plot in 2006 to establish a coastal disaster control forests loam. The sand fences were installed outside the bush hedges protection with adequate air space between poles to provide a comprehensive protection system and to guarantee seedling growth and survival. Live shrubs were planted outside the sand fence boundary (Figure 2) to grow and provide further protection to the planted seedlings in their later stage of growth and development, particularly when the sand fences and bush hedge eventually deteriorated. A detailed description of the planting materials and methods used for the construction of the bush hedges and sand fences is found in Kim et al. (2012). Bush hedges function as near-surface micro-windbreaks, reducing surface sand erosion by wind, windblown sand, and salt spray, as well as physical damage such as cuttings caused by strong winds to young planted seedlings. The sand fence material was made of bamboo and installed with spaces between to allow air flow between poles, offering a protective margin inclined to the windward direction of stronger movement as illustrated in Figure 3. Bush hedges, though shorter in height, are tightly spaced and earthed to form internal strips to significantly disallow air flow and reduce soil moisture dry-off caused by near-surface winds, hence providing stronger protection to seedlings’ growth and survival. The planted shrubs was expected to provide long term protection to the coastal forest development as the sand fence and bush hedges will eventually deteriorate and become remnants within the forest floor, providing surface roughness against wind erosion.

Figure 2. External representation of the bush hedges (left) and sand fence (right) from beach boundary constructed in 2006 before seedlings were planted.

Figure 3. Schematic presentation of the selected sample plots inside the sand fence and bush hedges and direction of stronger wind movement in the microsite.

Pinus thunbergii was the most suitable tree species because it is considered native to the Korean coast and exhibits naturalized adaptation capacity to the harsh coastal environment of poor soils, maritime exposure, drought, salinity, and windy conditions. Before afforestation, bush hedges were installed at a right angle direction to strong sea winds to reduce wind damage and excessive sand movement into seedling plots. Because P. thunbergii tolerates acidic and nutrient deficient soils intrinsic in our study area, neutralization and fertilization of the soil was unnecessary and the effect of soil nutrient properties was not considered in this study.

We systematically selected three plots each at distances of 5 m, 15 m, and 25 m forming the selected three ranges of plots landward from the bush hedges/beach boundary in the microsite, making a total of nine sample plots measured and considered for data collection (Figure 3). Within each range of plots, the three surveyed plots were 10 m or four plots width apart (Figure 3). The study design involved annual monitoring and measurements of seedling height growth, RCD, and survival from 2007 to 2014. In addition, we measured selected environmental factors including wind speed, soil moisture at depths of 7.6 cm, 12 cm, and 20 cm, soil electrical conductivity (EC), and soil acidity (pH) in 2010 and 2014.

Data collection

We used a measurement staff leveler to measure the height of individual trees in each of the selected nine sample plots divided into three ranges of plots. Wind speed was measured in each plot using a Kestrel 1000 pocket wind meter (Nielsen Kellerman, Boothwyn, PA, USA) at distances of 5 m, 15 m, and 25 m, landward from the bush hedge/beach boundary by holding the wind gauge up at a height of about 2 m above sea level, facing different directions, including the direction of stronger wind movement. We measured soil moisture content in each plot at depths of 7.6 cm, 12 cm, and 20 cm using TDR 100 soil moisture probe for field scout (USA). We also measured soil electrical conductivity and soil temperature using EC Testr II EUTECH (Singapore), and soil acidity using pH 3000 WINDAUS (Germany). All the selected environmental variables were measured in situ five times each at the center positions of each plot and the values were recorded after each measurement in 2010 and 2014. Average values from the five measurements for each environmental variable were used for our statistical analysis.

Data analysis

We used ANOVA to explore differences in the mean values of seedling height growth, RCD, and survival with distance at three levels (5 m, 15 m, and 25 m) landward from the bush hedge. Significant differences between means of dependent variables based on distance treatment groups were determined using Tukey multiple comparisons post-hoc tests. We used simple regression analysis to investigate the effects of distance from the bush hedge on seedling height growth. We also used multiple regression analysis with distance at three levels (5 m, 15 m, and 25 m) landward from the bush hedge protective margin as the covariate to examine its effect on the measured environmental factors. We used IBM SPSS v21 for all our statistical data analysis and a significance level of P = 0.05.

Results

Seedling height growth, RCD, and survivorship in relation to distance from sand fence and bush hedges and selected soil factors and wind speed

Data from field measurements showed appreciable variation in annual means of seedling height growth increments from 2007 to 2014 and the results are presented in Figure 4. Annual means in seedling height growth increment became relatively stable from the 4th year after planting but less so in the 2nd and 3rd years after planting (Figure 4).

Figure 4. Annual means of seedling height growth increments for whole microsite. Whiskers represent error bars with standard deviation.

Means of seedling height growth, RCD, and survivorship 8 years after planting for the sampled individual plots representing the three ranges of plots at distances of 5 m, 15 m, and 25 m landward from the bush hedge/coast boundary are presented in Table 1.

Table 1. Effect of distance at three levels of ranges of plots on the height growth, RCD, and survivorship of Pinus thunbergii seedlings 8 years after planting.

Distance (m)	Plot	Height growth (m)	RCD (cm)	Survivorship (%)
	1	5.1	12	96
5	4	5.3	10	92
	7	5.0	12	76
(Front range plots)	Mean ± SD	5.13 ± 0.15a	11.3 ± 1.15a	88.0 ± 10.5a
	2	4.0	6	100
15	5	4.9	10	100
	8	5.0	10	6.0
(Middle range plots)	Mean ± SD	4.63 ± 0.55a	8.68 ± 2.30a	86.0 ± 23.0a
	3	5.5	10	92
25	6	3.9	8	76
	9	5.1	12	84
(Behind range plots)	Mean ± SD	4.83 ± 0.83a	10.0 ± 2.0a	84.0 ± 8.0a
	Total mean ± SD	4.86 ± 0.55	10.0 ± 2.0	86.2 ± 13.4
	P-value	>0.05	>0.05	>0.05

Notes: Means ± SD were measured in 2014. The same letters in each column indicate insignificant differences according to Tukey multiple comparisons post-hoc test (P < 0.05) in a three-way ANOVA.

The mean height growth of seedlings from field measurement was higher in the front range of plots at 5 m (5.13 ± 0.15) than in the middle range of plots at 15 m (4.63 ± 0.55) and behind range of plots at 25 m (4.83 ± 0.83), but the observed field differences were not statistically significant (P >0.05) as revealed by our three-way ANOVA and Tukey multiple comparison post-hoc tests (Table 1). Also, the observed differences in the means of RCD and survivorship of the planted seedlings between the three ranges of plots at distances of 5 m, 15 m, and 25 m in the microsite were not statistically significant (P >0.05) (Table 1).

Furthermore, results from a simple linear regression analysis between environmental factors as the independent variables and seedling height growth as the dependent variable are presented in Table 2. The results revealed that all the measured soil variables of soil moisture at 7.6 cm and 12 cm and soil pH, in addition to wind speed were statistically significant factors affecting seedling height growth in our study area (P <0.05); the effect of soil EC was only significant at the 0.1 level (Table 2).

Table 2. Regression coefficients for measured seedling height growth and selected soil factors and wind speed.

	Unstandardized coefficients		Standardized coefficients
Variables	B	Std. error	Beta	t	Sig.
Constant	11.748	1.471		7.986	0.004
Wind speed (m/mps)	–0.818	0.159	–0.951	–5.136	0.014
Soil moisture at 7.6 cm (%)	–1.788	0.342	–49.162	–5.229	0.014
Soil moisture at 12 cm (%)	2.419	0.468	48.651	5.171	0.014
Soil acidity (pH)	0.107	0.025	0.793	4.263	0.024
Soil electrical conductivity (µS/cm)	1.743	0.733	0.402	2.377	0.098
Distance (m)	0.006	0.005	0.254	1.229	0.344

Notes: Dependent variable: height growth; significance level P < 0.05, R² = 0.944; adjusted R² = 0.85.

The effect of distance on the height growth of the planted seedlings was not statistically significant (Table 2). The predictive model factors accounted for 94.4% of the variability in seedling height growth and was statistically significant (P = 0.043).

Results of a Tukey multiple comparisons post-hoc test to determine significant differences in the means of each of the measured environmental factors between the three ranges of plots at distances of 5 m, 15 m, and 25 m is presented in Table 3. The results revealed that the measured environmental factors did not significantly vary between the three ranges of plots at distances of 5 m, 15 m, and 25m from the protective bush hedge in the microsite as evident by our Turkey multiple comparisons post-hoc test in multiple ANOVA (P >0.05) (Table 3). In addition, results of a multiple regression analysis to determine the effect of distance at three levels of 5 m, 15 m, and 25 m from bush hedges on the measured environmental factors is presented in Table 4. The effect of distance on the environmental condition of the three ranges of plots was not significantly different (P >0.05) and this is presented in Table 4.

Table 3. Effect of distance at three levels of ranges of plots on selected soil factors and wind speed 8 years after seedlings of Pinus thunbergii were planted.

Distance (m)	Wind speed (m/mps)	Soil moisture at 7.6 cm (%)	Soil moisture at 12 cm (%)	Soil moisture at 20 cm (%)	Soil acidity (pH)	Soil electrical conductivity (µS/cm)
5	0.97 ± 0.18a	52.28 ± 2.69a	35.17 ± 2.05a	24.87 ± 1.60a	8.03 ± 1.98a	0.14 ± 0.07a
15	0.70 ± 0.17a	52.44 ± 5.84a	35.28 ± 4.31a	24.96 ± 3.24a	5.88 ± 0.53a	0.14 ± 0.02a
25	0.91 ± 0.34a	46.97 ± 8.04a	31.30 ± 5.84a	21.92 ± 4.51a	5.54 ± 0.33a	0.10 ± 0.04a
Total	0.86 ± 0.24	50.56 ± 5.81	33.92 ± 4.25	23.92 ± 3.25	6.48 ± 1.56	0.12 ± 0.04
P-value	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05

Notes: Means ± SD were measured in 2014. The same letters in each column indicate insignificant differences according to Tukey multiple comparisons post hoc test (P < 0.05) in a multiple ANOVA.

Table 4. Multiple regression analysis of selected soil factors, wind speed, and distance factor.

Covariate	Dependent variable	Type III sum of squares	d.f.	Mean square	F	Sig.
Distance (m)	Wind speed (m/mps)	0.124	2	0.062	1.030	0.412
	Soil moisture at 7.6 cm (%)	58.214	2	29.107	0.823	0.483
	Soil moisture at 12 cm (%)	30.909	2	15.455	0.814	0.487
	Soil moisture at 20 cm (%)	17.891	2	8.946	0.801	0.492
	Soil acidity (pH)	10.957	2	5.479	3.777	0.087
	Soil electrical conductivity (µS/cm)	0.002	2	0.001	0.395	0.690

Note: Significance level P < 0.05.

Discussion

Relating seedling height growth to distance from sand fence and bush hedges

We prepared our site condition by planting bush hedges inside sand fences to protect seedlings from damage and mortality caused by near-surface strong winds from the direction of the coast to landward development. The bush hedge was expected to significantly reduce probable differences in the height growth, RCD, and survivorship of the planted seedlings among three ranges of plots and thus prevent the development of irregular forest structure at the study microsite. The mean total seedling height growth and RCD were 4.86 m and 10 cm, respectively, and the mean total survival was 86.2% for the entire microsite 8 years after planting in 2006, indicating appreciably high seedling height growth, RCD, and survivorship in the microsite as at 2014.

An interesting result was that there were no significant differences in the height growth, RCD, and survivorship of the planted seedlings between the three ranges of plots at distances of 5m, 15m, and 25m from the coast/bush hedge boundary as revealed by three-way ANOVA and verified by Tukey multiple comparison post-hoc tesst. We attributed this outcome to the protective and stabilization functions of the bush hedge micro-windbreak, which sustained relatively uniform seedling growth characteristics across the microsite.

According to previous studies, micro-windbreak protective margins are able to enhance microsite climate and soil conditions by providing a natural and long-term buffer against wind erosion, and encouraging trees to grow stably inside them (Caborn 1957; Cleugh 1998). In addition, other studies (e.g., Philippe 1975; Woodhouse 1978; Watson 1985; Olafson 1997; Khalil 2008) have reported that micro-windbreak techniques such as wooden sand fences have been effective in the restoration and afforestation of coastal and arid sandy environments. Therefore, the results on seedling height growth and survivorship provide demonstrative evidence of the protection and stabilization effectiveness of the installed bush hedges and sand fence barrier to deter occasional strong winds and prevent significant differences in seedling height growth, RCD, and survival among the three ranges of plots in our study microsite.

Relating seedling height growth to environmental factors

Besides inferring the protective effectiveness of sand fences and bush hedges from data on the height growth, RCD, and survivorship of the seedlings of P. thunbergii, we also aimed to determine the effect of environmental factors on solely the height growth of the planted seedlings and the most significant environmental factor affecting seedling height growth at our study site. The findings suggest that seedling height growth was most significantly affected by both wind speed and soil moisture at ≤12 cm, (P = 0.014) at our study site. In addition, soil acidity and soil EC were also important predictive factors according to the regression analysis. Although our study did not consider the effect of soil nutrients, the results are in line with those of a previous study which reported that nutrient deficiency, lack of soil moisture, predation, salt spray, sand accretion, and strong winds are probably the most important limiting factors on the growth and survival of seedlings in coastal and lacustrine sand dune systems (Maun 1994). In addition, other studies (e.g., Palik et al. 1997; Rodriguez-Trejo et al. 2003; Knapp et al. 2008) have reported that soil moisture is an important limiting factor in the growth of longleaf pine species even in well-drained soils such as the sandy soils inherent in the coastal environment where our study was undertaken. The soil moisture limiting condition in our study site stems from the fact that sand generally has high infiltration rates, lower water holding capacity, and low capillarity, and releases available water easily for consumption by trees, causing limited moisture content usually in the dry top layers (≤20 cm) of the sand dunes while ensuring the conservation of moisture at lower depths in the young coastal plantation. The limiting effect of occasional strong winds on the growth of young planted seedlings in coastal land boundaries is uncontestable.

The environmental conditions inside the microsite were not significantly different between the three ranges of plots at distances of 5 m, 15 m, and 25 m away from bush hedges towards landward development. The significantly stable and uniform environmental conditions in the microsite could partly explain the statistically insignificant differences detected in the height growth, RCD, and survival of the seedlings between the three ranges of plots at distances of 5 m, 15 m, and 25 m away from bush hedges towards landward development. Previous studies have reported that bush hedge plants forming shelterbelts’ protective margins are able to modify microsite climate and soil conditions and encourage stable tree growth by providing a natural and long-term buffer against wind damage in the microsite (Caborn 1957; Cleugh 1998). We could not find previous empirical studies examining the effect of distance from bush hedges or other hedge systems on seedling growth and survival in coastal sandy environments to lend sufficient supporting evidence for our study, but our results provide persuasive evidence of the importance of using bush hedge micro-windbreak protection during coastal forest plantation construction in windy coastal sandy environments. We concede that the thinning of the plantation in subsequent years of seedling growth may also have contributed measurably to improving the height growth, RCD, and survival of planted seedlings, but this was not quantified in this study. Soil chemical properties such as nitrogen and organic matter were not considered as variables for this study because P. thunbergii plant species is naturally well adapted to the chemical nutrient deficient soils of Korea and these factors are thought to have negligible impact on seedling growth and survival in comparison to the wind and soil moisture limiting factors at the study site.

Conclusions

In conclusion, understanding the importance and protective function of bush hedges installed inside sand fences as a micro-windbreak shield to enhance and foster the growth of P. thunbergii seedlings planted in a windy coastal sandy environment can significantly increase success in coastal forest plantation construction and potentially reduce the impact of coastal hydrometeorological disasters around the world. This study's main aim was to examine the effect of distance at three levels (5 m, 15 m, and 25 m) of ranges of plots from the sand fence and bush hedges on the height growth and survivorship of P. thunbergii seedlings planted in a coastal sandy environment. In effect, there were no significant differences in the height growth and survivorship of the planted seedlings and in the measured environmental factors between the three ranges of plots landward from the sand fence and bush hedge/beach boundary. We attribute the insignificant differences in the height growth and survivorship of the planted seedlings between the three ranges of plots in the microsite to the protective and stabilization functions of the sand fence and bush hedges micro-windbreak. Seedling height growth fluctuated significantly in the first 3 years after planting but appreciably stabilized from the 4th year. In particular, the 2nd and 3rd years of seedling growth in the study site were the most critical. Wind speed and soil moisture at ≤ 12 cm depth were the two most significant predictive factors of seedling height growth in the microsite. The results suggest that the bush hedges and sand fence provide protection and stabilization, helping to sustain the height growth and survivorship of P. thunbergii seedlings planted in a windy and coastal sandy environment. This is particularly important in the early years when seedling growth is naturally retarded, and when withering and physical damage caused by drought stress and strong wind speed conditions, respectively, are most deleterious. The study results provide baseline data and persuasive evidence to guide forest construction in windy coastal sandy environments, and in other locations with similar soil and climatic conditions.

Acknowledgements

We thank Lee Kee Whan, Jung Yu-Gyeong, and Park Ki-Young for their assistance in the field monitoring work.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the 2013 Yeungnam University Research Grant.