Phân tích mô hình phân bố và quan hệ không gian của một số các loài cây rừng lá rộng thường xanh, miền Bắc Việt Nam

Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 - 2018 23 BIODIVERSITY, SPATIAL AND ASSOCIATION PATTERNS OF NATURAL TREE SPECIES IN TROPICAL BROADLEAVED FOREST IN NORTHERN VIETNAM Phan Quoc Dung1, Nguyen Hong Hai2 1,2Vietnam National University of Forestry SUMMARY Ecological processes in forests can be studied via the spatial distribution of tree species. However, the distribution pattern of a species may be obscured by environmental heterogeneity. In order to answer these questions: What are the prevailing types of intraspecific spatial distributions and interspecific association patterns at tree species in a tropical rain forest? Which ecological processes could structure these patterns? The techniques of point pattern analysis were implemented on mapped two 1-ha forest plots in Ba Vi National Park, Cuc Phuong National Park. We analyzed (i) The effect of environmental heterogeneity on tree distributions; (ii) Intraspecific associations and (iii) Interspecific associations. Our analyses showed that: (i) Environmental conditions were homogeneous at all two plots. (ii) In two plots, almost dominant species were aggregated at various scales up to 50 m due to the limited distribution of each species while the rest was random distribution. (iii) Attraction and independence in two plots are remarkably higher than repulsion pattern of tree species. Overall, spatial aggregation of a species can be induced by limited seed dispersal or patchy habitat conditions while random distributions were effected by competitive relations or even human activities. The repulsive interactions between some tree species are explained by negative interactions of tree species. Keywords: Environmental homogeneity, Northern Vietnam, spatial point pattern analysis, tropical broad-leaved forest. I. INTRODUCTION Spatial patterns of forest trees result from complex dynamic processes such as establishment, dispersal, mortality, land use and climate (Franklin et al., 2010), especially in tropical forests which were known as the world’s most species-rich terrestrial ecosystems. An important question for all scientists in researching of forest ecology is how to understand the processes and mechanisms that control species coexistence and community structure, especially at various spatial scales. Studies on species-rich tropical forests produced numerous hypotheses on species co-existence, these relevant issues have been addressed in numerous studies (Chesson, 2000; Wright, 2002; Volkov et al., 2005). Barot (2004) highlighted the impact of both exogenous and endogenous factors on the spatial and temporal distributions of tree species. Other studies investigated dispersal limitation (Hubbell, 1979), intra- and inter- specific interactions (Callaway and Walker, 1997; Bruno et al., 2003), negative density dependence (Wright, 2002), or habitat preference (Condit et al., 2000). Tilman (2004) emphasized that in the processes of dispersal and competition, environmental niche effects and trade-offs among species are two main factors that made a big difference in spatial patterns of trees. Environmental heterogeneity (such as different soil types, rock outcrops or streams) makes spatial pattern analysis more complicated because it confounds biotic and abiotic effects (Li and Reynolds, 1995; Wiens, 2000). Getzin et al. (2008) found that plant ecology in terms of plant population dynamics and pattern formation may differ between homogeneous and heterogeneous sites, beyond the purely statistical effects of heterogeneity. Dispersal limitation is emphasized as a potential mechanism for separating species in space and reducing competitive exclusion (Seidler and Plotkin, 2006). Besides that, a patchy distribution of trees can also be caused by habitat preference where demographic processes and limiting resources may simultaneously influence spatial patterns (Wagner and Fortin, 2005; Getzin et al., 2008). Thus, spatial aggregation of a species can be Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 24 induced by limited seed dispersal or patchy habitat conditions and may also be reinforced by both factors (Webb and Peart addition, negative density dependence or self thinning is proposed as a prominent mechanism for regulating population dynamics and facilitating species coexistence 2002). This mechanism has been considered by a negative density of conspecific distance relation in processes of forest dynamics such as recruitment, growth or survival al., 1992; Peters, 2003; Uriarte et al. The goal of this research aims to analyze and evaluate spatial and association patterns of natural tree species in tropical broad forests in Northern Vietnam. Moreover, ecological underlying mechanisms or processes structuring these spatial patterns are inferred which allow to interpret spatial structure of these forest stands. II. RESEARCH METHODOLOGY 2.1. Study sites and data collection Two 1-ha plots are designed in two different tropical broadleaved forests in Northern Vietnam including Ba Vi National Park (21°04'09.5" N and 105°21'36.5" Phuong National Park (20°17'18.9" 105°39'22.3" E). Establishing typical plots in evergreen broad-leaved forest in the core zone of two National Parks (NP) represent for the forest stands in order to Figure 1. Map of studied plots at Ba Vi and Cuc Phuong National Park , 2000). In - (Wright, (Condit et , 2004). -leaved E), Cuc N and . The plots research ecological conditions, community structure and growth status. The area of eac plot is 1 ha (100 m × 100 m). divided into 100 subplots of 100 m m) by wooden poles and nylon strings. All trees (DBH ≥ 2.5 cm) were marked, identif the species name and measured the diameter at breast height at 1.3 m from ground relative position (x, y) of the tree subplot were measured distance measurer Leica Disto D2 with a precision of 0.1 cm and a compass. Ba Vi National Park tropical monsoon climate. The average annual temperature in the region is 23.4 temperatures down to 2.7 temperature up to 42oC. The annual averag rainfall is 2,500 mm, about 70 total precipitation focusing on July humidity of 86.1%. Cuc Phuong National Park (located in Nho Quan district, Ninh Binh province) is surrounded by limeston mean maximum height of 300 covered by tropical evergreen rainforest. In the core zone, mean annual temperature is 20.6°C, but mean temperature in winter is only 9°C. In the buffer zone, mean annual temperature is about 2° hi humidity is 85% and the average annual rainfall is 2,138 mm per - 2018 h The plot is 2 (10 m × 10 ied . The s in the by using the laser is situated in the oC; at lowest oC; highest e - 80% of the - August; e mountains with - 400 m and is gher. Annual mean year. Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 - 2018 25 2.2. Data analysis Important value and diversity indices Importance Value Index (IVI): was a measure of how dominant a species was in a given forest area. Relative density (RD) was the number of individuals per area as a percent of the number of individuals of all species. IVI (%) = (Relative density + relative Basal area)/2 Relative basal area was the total basal area of Species A as a percent of the total basal area of all species. The Shannon-Wiener index was an information statistic index, which means it assumes all species are represented in a sample and that they are randomly sampled. In the Shannon index, p was the proportion (n/N) of individuals of one particular species found (n) divided by the total number of individuals found (N), ln was the natural log, Σ is the sum of the calculations, and s was the number of species. Shannon Wiener Index (H) = The Simpson’s index was a dominance index because it gives more weight to common or dominant species. In this case, a few rare species with only a few representatives will not affect the diversity. In the Simpson index, p was the proportion (n/N) of individuals of one particular species found (n) divided by the total number of individuals found (N), Σ was still the sum of the calculations, and s was the number of species. Simpson′s Index (D) = III. RESULTS 3.1. Species property of tropical forest studied stands Table 1. Forest stand characteristics in Ba Vi plot No. Species N DBH (cm) IVI (%) Properties Shannon- Wiener Simpson 1 E. wightiana 105 9.6 ± 3.9 5.01 Light demanding & fast growing 2 X. noronhianum 99 10.3 ± 4.7 4.98 Light demanding 3 N. baviensis 55 16.8 ± 11.3 4.73 Light demanding 4 Q. bambusifolia 37 22.3 ± 13 4.35 Moderate inclining to light demanding 5 Q. gemelliflora 13 40.2 ± 18.2 3.58 Light demanding 6 C. lenticellata 71 9.5 ± 4.5 3.41 Light demanding & fast growing 3.36 0.97 7 W. laevis 68 9.4 ± 4.9 3.28 Shade tolerance 8 S. baviense 44 14.4 ± 10.8 3.28 Light demanding & fast growing 9 C. zeylanicum 37 17.1 ± 11.2 3.19 Light demanding 10 C. glaucescens 59 11.2 ± 5.2 3.14 Light demanding 11 A. globiflora 49 11.7 ± 5.8 2.71 Light demanding 12 70 other species 830 58.34 In Ba Vi NP plot, a total of 1,467 tree individuals with DBH ≥ 2.5 cm were enumerated in the 1-ha study plot. 81 species were identified and belonged to 26 families; Shannon - Weiner (H’) = 3.36; Simpson (D) = 0.97. In 11 dominant species, there are 10 Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 - 2018 26 species with light demanding, approximately 91% of total. E. wightiana (Myrtaceae) was most abundant with 105 individual ha-1 with the average size is quite small (9.6 ± 3.9 cm). Moreover, depending on IVI there are 11 dominant species: E. wightiana, X. noronhianum, N. baviensis, Q. bambusifolia, Q. gemelliflora, C. lenticellata, W. laevis, S. baviense, C. zeylanicum, C. glaucescens, A. globiflora with total IVI is 41.66%. Only 10 of them except Q. gemelliflora were selected for further spatial pattern analyses. Table 2. Forest stand characteristics in Cuc Phuong plot No. Species N DBH (cm) IVI (%) Properties Shannon- Wiener Simpson 1 S. macrophyllus 392 9.7 ± 7.3 25.72 Shade tolerance & lower storey 2 C. tonkinensis 29 67.1 ± 30.5 18.39 Light demanding & fast growing 3 S. dives 117 18.8 ± 12.7 12.28 Middle storey 2.78 0.82 4 H. kuzii 94 12.7 ± 8.8 7.1 Shade tolerance & middle storey 5 85 other species 374 36.51 In Cuc Phuong NP plot, the density of trees was quite high 1,006 trees/ha (DBH ≥ 2.5 cm). In total, 89 species were identified in this study plot and belonged to 24 families with the diversity indices: Shannon - Weiner (H’) = 2.78; Simpson (D) = 0.82. The average size of S. macrophyllus was small (9.7 ± 7.3 cm). Based on IV (%), it can be seen that S. macrophyllus with 3 other species: C. tonkinensis, S. dives, H. kuzii were eligible to form group of dominant tree species with total IVI was 63.49%. Three of four given species were shade tolerance and tend to grow in middle and lower storeys. As the results from three plots, the study identified 11 species with highest IVI in Ba Vi plot with total IVI was 41.66%, 4 species in Cuc Phuong plot with total IVI was 63.49%. Comparing diversity indices (D of Simpson), Ba Vi plot performed the highest values at 0.97 while Cuc Phuong plot had the lowest one at 0.82. Thus, the levels diversity in Ba Vi plot were strongly higher than Cuc Phuong site. In addition, the values of Shannon-Weiner (H’) of Ba Vi plot and Cuc Phuong plot, were 3.36, 2.78. Therefore, Ba Vi plot was at high level of population balance and richness. 3.2. Spatial patterns analysis Analysis 1: Environmental heterogeneity effects The spatial patterns of all adult trees (dbh ≥ 15 cm) in study plots were contrasted to the CSR null model to find significant departure at large scales. We used both cumulative and non-cumulative advantages of both L-function and g-functions in this analysis, respectively. The g-function showed that adults in all plots were regular at small scales and that could be evidences of strong tree-tree competition (results not shown). Moreover, L-function also showed no deviation from confidence envelopes at larger scales (results not shown). Therefore, no large scale departure from the CSR null model was observed and the hypothesis of environmental homogeneity was accepted in the study plots. Based on this finding, we applied the homogeneous g- function for the further spatial pattern analyses in this study. JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 Analysis 2: Intraspecific spatial distributions Figure 2. Spatial patterns of dominant tree species in function g Black lines are observed patterns; Ba Vi plot analyzed by the pair correlation 11(r) under null model of CSR grey lines are approximate 95% confidence envelopes Silviculture - 2018 27 Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 28 In Ba Vi plot, intraspecific spatial distributions was analyzed correlation function g11(r). E. aggregated at 1 - 4 m and at large scales of 8 15 m (Figure 2a). In contrast, X showed a strong random distribution over the entire range of scales up to 46 m ( baviensis and Q. bambusifolia were at the begging of scales of 0 - 2 m ( and 1 - 4 m (Figure 2d). There was the same clustered distribution of C. lenticellata, and A.globiflora at 0 - 2 m (Fig baviense was clustered at small scales of 3 Figure 3. Spatial patterns of dominant tree species in Cuc Phuong plot correlation function g Black lines are observed patterns; grey lines are approximate 95% confidence envelopes Analysis 3: Interspecific spatial associations As the results were analyzed by by the bivariate pair correlation under null model of random labeling, we performed 90 bivariate point pattern analyzses for all pairs of dominant species plot. Overall, independence occurred more frequently with 53.3% while attraction 28 and repulsion 17.9%. There were 13 significant positive interactions observed by the pair wightiana was - . noronhianum Figure 2b). N. aggregated Figure 2c) W. laevi ure 2e, f, k). S. - 5 m (Figure 2g). C. glaucescens large scales of 1 - 6 m and 7 A. globiflora was random of scales up to 40 m (Figure In Cuc Phuong plot, based on IV 4 species: S. macrophyllus dives, H. kuzii are considered as dominant tree species and spatial distributions figure 3. S. macrophyllus 34 m (Figure 3a). C. tonkinensis also showed clustered distribution at 2 (Figure 3b) and 4 - 7 m (Fig was random at small scales ( analyzed 11(r) under null model of CSR analyzed function g12(r) for Ba Vi .8% between N. baviensis glaucescens - E. wightiana; noronhianum; S. baviense A. globiflora - X. noronhianum; N. baviensis; C. glaucescens lenticellata - Q. bambusifolia; bambusifolia; C. zeylanicum bambusifolia; A. globiflora zeylanicum - W. laevis; baviense; C. glaucescens - 2018 was aggregate at - 22 m (Figure 2i). over the entire range 2h). I, there were , C. tonkinensis, S. were shown in was aggregated at 1 - and S. dives - 12 m ure 3c). C. tonkinensis Figure 3d). by the pair - E. wightiana; C. C. lenticellata - X. - X. noronhianum; S. baviense - - N. baviensis; C. W. laevis - Q. - Q. - C. lenticellata; C. C. glaucescens - S. - C. zeylanicum. JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 Table 3. Spatial associations of dominant tree species in Ba Vi plot No. Species (1) (1) E. wightiana (2) X. noronhianum 0 (3) N. baviensis + (4) Q. bambusifolia - (5) C. lenticellata 0 (6) W. laevis 0 (7) S. baviense - (8) C. zeylanicum - (9) C. glaucescens + (10) A. globiflora 0 Note: 0: independence; +: positive association (attraction); In contrast, repulsion occurred 8 times between Q. bambusifolia - E. wightiana; wightiana; C. zeylanicum - E. wightiana; baviensis - X. noronhianum; Q. bambusifolia noronhianum; W. laevis - X. noronhianum; laevis - N. baviensis; A. globiflora glaucescens. It can be seen that the interactions are mostly independence, for example: noronhianum - E. wightiana; C. lenticellata wightiana; C. zeylanicum - N. baviensis Spatial associations of 4 dominant tree species in Cuc Phuong plot were showed and Figure 4. Association patterns of dominant tree species in Cuc Phuong analyzed by correlation function Black lines are observed patterns; grey lines are approximate 95% (2) (3) (4) (5) (6) (7) 0 + - 0 0 - - - + - + - 0 0 - + - 0 + + 0 + 0 + 0 0 - - + 0 0 + + 0 0 0 0 0 + 0 + 0 0 + 0 0 0 + + 0 0 + 0 0 -: negative association (repulsion) S. baviense - E. N. - X. W. - C. X. - E. . analyzed with the bivariate pair function under null model of random labeling (Figure 4). As the result, 2 pairs showed repulsion and 4 pairs independence. macrophyllus - H. kuzii macrophyllus - C. tonkinensis relpusive associations. S dives (Figure 4a), S. dives 4d), S. dives - C. tonkinensis kuzii - C. tonkinensis independent in species interactions. g12(r) under null model of random labeling confidence envelopes Silviculture - 2018 29 (8) (9) (10) - + 0 0 0 + 0 + 0 + 0 0 0 0 + + 0 0 0 + 0 0 0 0 - 0 - . -correlation S. (Figure 4b), S. (Figure 4c) were . macrophyllus - S. - H. kuzii (Figure (Figure 4e), H. (Figure 4f) were the bivariate pair Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 - 2018 30 The independent interaction between tree species is a very common in tropical forest with high level of diversity as in the study area. This is also explained by the fact that many species have similar ecological characteristics such as the demand of light or nutrition. The repulsive association of tree species is explained by the fact that forest structure, species composition and forest canopy are altered by multiple impacts. This leads to light- demanding and fast-growing species that tend to grow, compete with other species, and dominate the population. A possible explanation is that attraction patterns are the result of facilitation at small scales. Specifically, the local environment is modified by large trees or canopy gaps and facilitates small intra- and inter-specific associations of trees with similar habitat preferences, e.g. with similar light requirements in our case. Suzuki et al. (2012) highlighted that an attraction pattern may result from similarity in habitat preference of spatially associated species. Alternatively, attraction patterns among species could be consistent with the species-herd protection hypothesis which states that hetero-specific neighbors can promote coexistence by preventing the transmission of biotic plant pests (Peters, 2003; Lan et al., 2012). The two study plots are significantly different in tree species structure, species diversity, and spatial patterns. The effects of forest disturbance by human activities were emphasized significantly through forest community structure. The findings can be used as suggestions for silvicultural treatments and biodiversity conservation of tropical rain forests in study regions. IV. DISCUSSION AND CONCLUSION 4.1. Species diversity of studied forest stands The research has been conducted quantitatively to help clarify the characteristics of natural forests in Vietnam. Regarding the characteristics of tree species, the study identified 11 species with highest IVI in Ba Vi plot with total IVI is 41.66%, 4 species in Cuc Phuong plot with total IVI is 63.49%. Based on IVI, it can be seen clearly that there are not predominantly dominant tree species in Ba Vi plot. However, the tree species are on the top of IVI still can associate with each other in order to form group of dominant tree species. Especially, in Cuc Phuong plot, group of dominant species formed with less than 10 species and ∑ IVI ≥ 40%, will be named for whole community. Comparing diversity indices (D of Simpson), Ba Vi plot performed the highest values at 0.97 while Cuc Phuong plot has the lowest one at 0.82. Thus, the levels diversity in Ba Vi plot is strongly higher than Cuc Phuong site. Moreover, the values of Shannon-Weiner (H’) of 2 plots Ba Vi plot, Cuc Phuong plot are 3.36, 2.78. As the result, both values of (H’) and (D) in Ba Vi plot are the highest comparing with the others, so it would be a representative of a diverse and equally distributed community. Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 - 2018 31 4.2. Spatial patterns analysis Environmental heterogeneity effects After using both cumulative and non- cumulative advantages of both L-function and g-functions in this analysis, we can see that no large-scale departure from the CSR null model was observed and the hypothesis of environmental homogeneity was accepted in the study plots. Intraspecific spatial distributions In Ba Vi plot, almost the spacial distributions are aggregation except X. noroniaum and A. globiflora are performed as strong random distribution. In Cuc Phuong plot, only C. tonkinensis was random while the others were clustered. Thus, the cluster distribution is mainly due to the limited distribution of each species. The random distribution of a number of species studied can be controlled by a variety of ecological processes or mechanisms or even human activities but due to the secondary forest status has been affected and the number of individuals of these species is low, so this research cannot find the root causes of this distribution. Interspecific spatial associations In Ba Vi plot, with 90 bivariate point pattern analyzes, the independence occurred more frequently with 53.3% while attraction 28.8% and repulsion 17.9%. In Cuc Phuong plot, with 4 dominant species, the analyzes showed 4 pairs of repulsion and 8 pairs of independence. Under the influence of heterogeneous environmental conditions, spatial relations include repulsion, attraction and independence. However, homogeneous environment, attractive and independent interaction tend to increase. Especially, the repulsive interactions between some tree species Ba Vi plot and Cuc Phuong plot are explained by negative interactions of tree species. This leads to fast-growing, light demanding species that tend to grow, compete with other species, and dominate the population. REFERENCES 1. Barot S. (2004). Mechanisms promoting plant coexistence: can all the proposed processes be reconciled? Oikos, 106(1): 185-192. 2. Chesson, P. (2000). General theory of competitive coexistence in spatially-varying environments. Theoretical Population Biology, 58(3): 211-237. 3. Getzin S., Wiegand T., Wiegand K. , He F. (2008). Heterogeneity influences spatial patterns and demographics in forest stands. Journal Of Ecology, 96(4): 807-820. 4. Harms, K. E., Wright, S. J., Calderon, O., Hernandez, A. & Herre, E. A. (2000). Pervasive density- dependent recruitment enhances seedling diversity in a tropical forest. Nature, 404(6777): 493-495. 5. Peters, H. A. (2003). Neighbour-regulated mortality: the influence of positive and negative density dependence on tree populations in species-rich tropical forests. Ecology Letters, 6(8): 757-765. 6. Seidler TG, Plotkin JB. (2006). Seed dispersal and spatial pattern in tropical trees. Plos Biology, 4(11): 2132-2137. 7. Volkov, I., Banavar, J. R., He, F. L., Hubbell, S. P. & Maritan, A. (2005). Density dependence explains tree species abundance and diversity in tropical forests. Nature, 438(7068): 658-661. 8. Webb CO, Peart DR. (2000). Habitat associations of trees and seedlings in a Bornean rain forest. Journal of Ecology, 88(3): 464-478. Silviculture JOURNAL OF FORESTRY SCIENCE AND TECHNOLOGY NO. 2 - 2018 32 PHÂN TÍCH MÔ HÌNH PHÂN BỐ VÀ QUAN HỆ KHÔNG GIAN CỦA MỘT SỐ CÁC LOÀI CÂY RỪNG LÁ RỘNG THƯỜNG XANH, MIỀN BẮC VIỆT NAM Phan Quốc Dũng1, Nguyễn Hồng Hải2 1,2Trường Đại học Lâm nghiệp TÓM TẮT Các quá trình sinh thái rừng có thể được nghiên cứu thông qua phân bố không gian của các loài cây. Tuy nhiên, mô hình phân bố của một số loài có thể bị ảnh hưởng bởi sự không đồng nhất của môi trường. Để trả lời cho những câu hỏi như: Các kiểu phân bố cây cùng loài và khác loài phổ biến trong rừng mưa nhiệt đới là gì? Những quá trình sinh thái nào ảnh hưởng tới sự cấu trúc và tổ thành đó? Phương pháp phân tích mô hình điểm không gian đã được thực hiện với 2 ô tiêu chuẩn 1 ha tại Vườn Quốc gia Ba Vì và Vườn Quốc gia Cúc Phương. Chúng tôi đã phân tích (i) Tác động của sự không đồng nhất môi trường tới sự phân bố của cây; (ii) Quan hệ cùng loài và (iii) Quan hệ khác loài của các loài cây trong khu vực nghiên cứu. Kết quả nghiên cứu cho thấy: (i) Các điều kiện môi trường là đồng nhất tại cả 2 ô tiêu chuẩn. (ii) Tại 2 ô tiêu chuẩn, hầu hết các loài cây ưu thế có phân bố cụm lên tới 50 m do sự phát tán hạn chế của mỗi loài, trong khi các loài khác lại xuất hiện phân bố ngẫu nhiên. (iii) Quan hệ tương hỗ và quan hệ độc lập giữa các loài cây là phổ biến hơn so với quan hệ cạnh tranh. Nhìn chung, phân bố cụm của một loài có thể do sự phát tán hạt hạn chế hoặc do thiếu hụt các điều kiện sống. Trong khi đó, phân bố ngẫu nhiên có thể được giải thích bởi sự ảnh hưởng từ các mối quan hệ cạnh tranh hoặc do các tác động của con người. Quan hệ cạnh tranh giữa hai loài có thể do nhu cầu ánh sáng và dinh dưỡng của mỗi loài. Từ khóa: Môi trường không đồng nhất, phân tích mô hình điểm không gian, phía Bắc Việt Nam, rừng nhiệt đới lá rộng. Received : 02/01/2018 Revised : 13/3/2018 Accepted : 20/3/2018