Dynamics of Understory Shrub Biomass in Six Young Plantations of Southern Subtropical China

Understory shrubs are an important component of forest ecosystems and drive ecosystem processes, such as ecosystem carbon cycling. However, shrub biomass carbon stocks have rarely been reported, which limits our understanding of ecosystem C stock and cycling. In this study, we evaluated carbon accumulation of shrub species using allometric equations based on height and basal diameter in six subtropical plantations at the age of 1, 3, 4 and 6 years. The results showed that plantation type did not significantly affect the total biomass of shrubs, but it significantly affected the biomass of Rhodomyrtus tomentosa, Ilex asprella, Clerodendrum fortunatum and Baeckea frutescens. The biomass of dominant shrub species R. tomentosa, I. asprella, Gardenia jasminoides and Melastoma candidum increased with stand age, while the biomass of C. fortunatum and B. frutescens decreased. The inconsistent biomass-time patterns of different shrub species may be the primary reason for the altered total shrub biomass in each plantation. Consequently, we proposed that R. tomentosa, I. asprella, G. jasminoides and M. candidum could be preferable for understory carbon accumulation and should be maintained or planted because of their important functions in carbon accumulation and high economic values in the young plantations of southern subtropical China.


Introduction
Shrubs are a type of vegetal group with a specific structure [1] and are essential to the biophysical and biodynamic processes of different ecosystems [2,3]. In arid and semi-arid ecosystems, shrubs are conducive to the increase in the number of herb-layer plant species [4]. In subtropical forests, shrubs facilitate the growth of understory tree seedling [5,6]. In reclaimed mined sites, shrubs act as ecosystem engineers [7]. In addition, understory shrub increases litter decomposition and the accumulation of litter-derived carbon (C) and nitrogen (N) in forest soils by altering soil microbial activity and community composition [8][9][10].
Physiological and ecological characteristics of shrubs have been systematically studied [11][12][13]. Palmroth et al. [14] found that an understory shrub had little response to N additions in terms of photosynthetic physiology or growth in boreal coniferous forests, which indicated that the shrub played a vital role in ecosystem C and N cycling. In addition, understory shrubs can respond quickly to thinning of even-aged stands [15]. Currently, increasing studies on shrubs focus on total shrub biomass

Study Sites
This study was conducted at the Heshan National Field Research Station of Forest Ecosystem (112 • 54 E, 22 • 41 N), located in Guangdong province, China. The climate in this region is typical subtropical monsoon, with a distinct wet (April to September) and dry season (October to March), mean annual precipitation of 1700 mm and mean annual temperature of 21.7 • C. The average altitude is 10-80 m. The soil is classified as an Ultisol developed from sandstone. The dynamics of understory shrub biomass were studied in six plantations, including Eucalyptus urophylla monoculture plantation (EU), Acacia crassicarpa monoculture plantation (AC), Castanopsis hystrix monoculture plantation (CH), 10-species mixed plantation (M10), 30-species mixed plantation (M30) and a naturally recovered shrubland (NS). These plantations were established in 2005 on hilly land with similar site characteristics, and the tree seedlings were planted at a spacing of 2 m × 3 m (about 1650 trees per hectare) except for the naturally recovered shrubland where the plants sprouted naturally and were not be planted. Each plantation had three replicate plots, and the area of each plot was 1 hectare. All plots were distributed randomly in the experimental site with an area of 50 hectares. A permanent quadrat plot of 900 m 2 (30 m by 30 m) was established in each plot with an area of 1 hectare. The height of trees in these plantations is shown in Table 1.

Methods
Vegetation inventories were conducted in permanent quadrats in 2006, 2008, 2009 and 2011, corresponding to 1 year, 3 years, 4 years and 6 years, respectively, after permanent quadrat plots were established in these plantations. A permanent quadrat plot was divided into nine subplots (four 5 m × 5 m small subplots for each subplot). In each subplot, we selected two bidiagonal small subplots and then measured the height and basal diameter of all shrubs during each inventory. There were eighteen small subplots in a permanent quadrat plot for each inventory. The biomass of each shrub species was estimated by allometric regression equations in all plantations [32].

Statistics Analysis
Two-way ANOVAs were employed to test the effects of plantation type and stand age on the number of shrub species, total biomass of shrubs, and biomass of Rhodomyrtus tomentosa, Ilex asprella, Gardenia jasminoides, Melastoma candidum, Clerodendrum fortunatum, Baeckea frutescens. One-way ANOVA and multiple comparison analysis (LSD) were performed to determine the effect of plantation type on the number of shrub species and total biomass of shrubs at the same stand age, and the effect of stand age on the number of shrub species and the total biomass of shrubs in the same plantation. Data were reciprocally or square-root transformed when required to meet the assumptions of normality and homogeneity of variance. Statistical significance was determined at P < 0.05. All analyses were performed with SPSS software (SPSS. Inc., Chicago, IL, USA).

Dynamics of Shrub Species
The number of shrub species was significantly affected by plantation type (two-way ANOVA, F = 3.18, P = 0.015). It was higher in AC than in other plantations ( Figure 1). The number of shrub species was the highest in AC at the age of 3 and 6 years (about 13.3 species), and it was the lowest in EU at the age of 1 year and in NS at the age of 6 years (about 7 species). However, stand age did not significantly affect the number of shrub species (two-way ANOVA, F = 1.44, P = 0.243). The interaction between plantation type and stand age was not significant (F = 0.44, P = 0.957).
During the period of 1 to 6 years old, 23 shrub species were found in these plantations regardless of plantation type ( Table 2). About 16, 13, 16, 13, 13 and 15 species were found in NS, EU, AC, CH, M10 and M30, respectively, 10 of which were shared by these given plantations regardless of replicated quadrats. Endemic species also present included Glochidion zeylanicum only in NS, Glochidion eriocarpum in AC, Garcinia oligantha in CH, Cassia alata, Symplocos chinensis, and Psychotria rubra in M30 (Table 2).  Table 2. Shrub species in six plantations.

Species Name Plantations NS EU AC CH M10 M30
Effects of plantation type and stand age on the number of shrub species. NS, EU, AC, CH, M10 and M30 represent naturally recovered shrubland, Eucalyptus urophylla, Acacia crassicarpa, Castanopsis hystrix, 10-species mixed plantation and 30-species mixed plantation, respectively. The value is the average ± standard error, n = 3. Table 2. Shrub species in six plantations.

Species Name Plantations
Note: " √ " and " " indicate that the species are present or absent, respectively in this plantation regardless of stand age. For abbreviations, please see Table 1.

Dynamics of the Total Shrub Biomass
Two-way ANOVA indicated that plantation type did not significantly affect the total biomass of shrubs (two-way ANOVA, F = 1.46, P = 0.223). However, to some extent, the total biomass of shrubs was affected by stand age (two-way ANOVA, F = 2.74, P = 0.054; Figure 2). The interaction between plantation type and stand age was not significant (F = 0.60, P = 0.858). In all given plantations, the total biomass of shrubs at the age of 4 and 6 years was significantly higher than that at the age of 1 year (P = 0.025 and P = 0.013, respectively. The average biomass increased from 704.09 kg ha −1 at the age of 1 year to 1571.66 kg ha −1 at the age of 6 years regardless of plantation type ( Figure 2).

Dynamics of the Total Shrub Biomass
Two-way ANOVA indicated that plantation type did not significantly affect the total biomass of shrubs (two-way ANOVA, F = 1.46, P = 0.223). However, to some extent, the total biomass of shrubs was affected by stand age (two-way ANOVA, F = 2.74, P = 0.054; Figure 2). The interaction between plantation type and stand age was not significant (F = 0.60, P = 0.858). In all given plantations, the total biomass of shrubs at the age of 4 and 6 years was significantly higher than that at the age of 1 year (P = 0.025 and P = 0.013, respectively. The average biomass increased from 704.09 kg ha −1 at the age of 1 year to 1571.66 kg ha −1 at the age of 6 years regardless of plantation type (Figure 2).

Biomass of the Dominant Shrub Species
The biomass of each dominant shrub species in all given plantations was separately evaluated and analyzed (Figure 3). Two-way ANOVA showed that plantation type significantly affected the biomass of R. tomentosa, I. asprella, C. fortunatum and B. frutescens (Table 3). The biomass of R. tomentosa was significantly higher in M30 than in other plantations, it was also higher in M10 than in CH (P < 0.034, and P = 0.027, respectively). Biomass of I. asprella was significantly higher in AC than in CH, M10 and M30 (P < 0.021). Biomass of C. fortunatum was significantly lower in NS than in AC, CH and M10 (P < 0.021), and it was also lower significantly in EU than in M10 (P = 0.031). Stand age significantly affected the biomass of I. asprella, M. candidum, C. fortunatum and B. frutescens. Stand age also had an effect on the biomass of R. tomentosa to some extent (  (Figure 3). The biomass of R. tomentosa and M. candidum was higher in plantations at the age of 4 and 6 years than that at the age of 1 year (P = 0.034, and P = 0.011, respectively), and the biomass of I. asprella was higher in plantations at the age of 3 and 6 years than that at the age of 1 year (P = 0.005, and P = 0.038, respectively). However, the biomass of C. fortunatum and B. frutescens decreased significantly from 72.65, 36.88 kg ha −1 at the age of 1 year to 18.44, 9.19 kg ha −1 at the age of 6 years regardless of plantation type (P < 0.001, and P < 0.001, respectively). Neither plantation type nor stand age affected the biomass of G. jasminoides, but its

Biomass of the Dominant Shrub Species
The biomass of each dominant shrub species in all given plantations was separately evaluated and analyzed ( Figure 3). Two-way ANOVA showed that plantation type significantly affected the biomass of R. tomentosa, I. asprella, C. fortunatum and B. frutescens (Table 3). The biomass of R. tomentosa was significantly higher in M30 than in other plantations, it was also higher in M10 than in CH (P < 0.034, and P = 0.027, respectively). Biomass of I. asprella was significantly higher in AC than in CH, M10 and M30 (P < 0.021). Biomass of C. fortunatum was significantly lower in NS than in AC, CH and M10 (P < 0.021), and it was also lower significantly in EU than in M10 (P = 0.031). Stand age significantly affected the biomass of I. asprella, M. candidum, C. fortunatum and B. frutescens. Stand age also had an effect on the biomass of R. tomentosa to some extent (  (Figure 3). The biomass of R. tomentosa and M. candidum was higher in plantations at the age of 4 and 6 years than that at the age of 1 year (P = 0.034, and P = 0.011, respectively), and the biomass of I. asprella was higher in plantations at the age of 3 and 6 years than that at the age of 1 year (P = 0.005, and P = 0.038, respectively). However, the biomass of C. fortunatum and B. frutescens decreased significantly from 72.65, 36.88 kg ha −1 at the age of 1 year to 18.44, 9.19 kg ha −1 at the age of 6 years regardless of plantation type (P < 0.001, and P < 0.001, respectively). Neither plantation type nor stand age affected the biomass of G. jasminoides, but its average biomass increased to 41.33 kg ha −1 at the age of 6 years relative to 15.79 kg ha −1 at the age of 1 year regardless of plantation type.
average biomass increased to 41.33 kg ha −1 at the age of 6 years relative to 15.79 kg ha −1 at the age of 1 year regardless of plantation type.  Results are from two-way factorial ANOVA. The factors used for the ANOVA were plantation type (levels: naturally recovered shrubland, Eucalyptus urophylla, Acacia crassicarpa, Castanopsis hystrix, 10species mixed plantation and 30-species mixed plantation), and stand age (levels: 1, 3, 4 and 6 years old).

Effects of Plantation Type
Plantation type affected the number of shrub species significantly. The number of shrub species was obviously higher in AC than that in other plantations, which was the highest in AC, and the lowest in NS. This can be ascribed to the understory abiotic and biotic conditions and the characteristics of species [33,34]. Shrub germination may be affected by herbs through competition for nutrients and light; higher herb biomass hinders shrub species germination [35,36]. In these studied plantations, herbs were dominated by Dicranopteris dichotoma [27]. The herb biomass was the lowest in AC, and the highest in NS [32].
Regardless of plantation type and stand age, 23 shrub species were found in the studied plantations. The number of shrub species was very close to the previous research result (13-23 species) [37]. Carneiro et al. [38] found that the average number of understory species was 6-15 in Eucalyptus plantations aged 1 and 10 years old. Luo et al. [39] also found the diversity of shrub species was similar among subtropical plantations in southern China. In our research, some endemic species in the specific plantations were found, yet the possible reasons why these endemic species were only present in specific plantations are still unclear and could need further study.
Plantation type significantly affected the biomass of R. tomentosa, I. asprella, C. fortunatum and B. frutescens. Soil physicochemical properties could make contributions to the difference [40]. However, we previously found that there were no significant differences in soil organic carbon among these plantations [41], and there were no significant differences in soil pH, soil nitrogen, soil available phosphorus among the above six plantations [42]. Moreover, in our experimental sites, the effects of plantation type on mean soil temperature, soil water content, and leaf area index were not significant either [43]. Therefore, we speculate that it could be ascribed to the differences in microenvironment among different plantations. For instance, AC had the highest canopy density (from direct observation); thus, the light in the understory could be the weakest for all studied plantations.

Effects of Stand Age
Stand age did not significantly affect the number of shrub species. This is inconsistent with another study in humid tropical plantations [44]. The diverse climates could be responsible for the differences [45,46]. A previous study suggested that understory richness was driven by loss of early colonists [47]. In our study, these plantations were still young and the understory shrub species may have not developed to the stage of loss of early colonists. Therefore, the effect of stand age on the number of shrub species could be negligible.
The total biomass of shrubs increased from 704.09 kg ha −1 at the age of 1 year to 1571.66 kg ha −1 at the age of 6 years regardless of plantation type, which was higher than the result (158 kg ha −1 ) of the previous study [48]. Plantation type did not significantly influence the total biomass of shrubs in our study. The likely reasons were that all studied plantations were young and the canopies were not completely closed (from direct observation). A large proportion of shrub species received sufficient light for growth [49]. In addition, shrubs can adopt an arborescent shape under a tree canopy to some extent [50]. The change in biomass with stand age was consistent with Li et al. [51]. In these studied plantations, the total biomass of shrubs were mainly from the biomass of R. tomentosa, I. asprella, G. jasminoides, M. candidum, C. fortunatum and B. frutescens. The summed biomass of these six shrub species accounted for about 85.19-98.44% of the total biomass of shrubs. The result from the survey of shrub species also suggested that R. tomentosa, G. jasminoides, C. fortunatum and B. frutescens were dominant understory shrub species in subtropical forests in eastern Guangdong, China [52]. Furthermore, the dominant shrub species R. tomentosa, M. candidum, I. asprella and G. jasminoides have high economic values. For instance, Rhodomyrtone is a natural antibacterial drug candidate from R. tomentosa [53]. The root extracts from I. asprella lessen acute respiratory distress syndrome in mice infected with the influenza virus [54].
Stand age significantly affected the biomass of I. asprella, M. candidum, C. fortunatum and B. frutescens. It also slightly influenced the biomass of R. tomentosa (Table 3). The biomass of R. tomentosa, I. asprella, M. candidum increased with stand age. The increase in biomass resulted from plant height and basal diameter increases. However, the biomass of C. fortunatum and B. frutescens decreased with stand ages. The decrease in biomass was mainly caused by individuals being reduced (Table 4).