INTRODUCTION
Forest and forestry in Vietnam
Vietnam boasts a rich diversity of forest vegetation, attributed to its varied climatic conditions and topographic features, covering 13.9 million hectares, or 39.9% of the country's land area Among this, 10.4 million hectares, representing 75.2%, are classified as natural forests, which can be divided into eight major groups: close-mixed evergreen broad-leaved rainforest, semi-deciduous mixed forest, mixed limestone forest, coniferous and mixed coniferous broad-leaved forest, sparse forest and seasonal deciduous forest, mangrove forest, Melaleuca forest, and bamboo and mixed timber-bamboo forest Reflecting the characteristics of tropical forests, Vietnam's forests are highly diverse, with hundreds of tree species found within a single hectare, resulting in significant biomass and productivity levels.
Selective logging in tropical forests
Selective logging refers to forest management practices that involve the careful harvesting of valuable trees while maintaining overall forest cover (Lobo et al., 2007) Johns (1985) cites Rapera (1977), defining selective logging as the removal of mature, over-mature, and defective trees in a way that preserves a sufficient number of healthy commercial and other tree species, ensuring future timber crops and protecting soil and water Silviculture systems are typically categorized as multi-ring or round based on the frequency of harvesting activities within a forest rotation.
2 system (Smith and Nichols, 2005); Selective mining is a multi-ring system (Lamprecht,
Selective logging is recognized as a more sustainable alternative to clear cutting, as it helps protect forest integrity while optimizing resource use (Gatti et al., 2014) This silvicultural method is favored by growers who aim to preserve immature seedlings for future crops By retaining key structural components of the forest and various ecological niches, selective logging maintains high biodiversity levels and ensures the ecosystem can recover more quickly to its pre-harvest state (Nzogang, 2009).
Selective logging systems often lead to the transformation of logged forests into secondary forests, raising concerns about the long-term sustainability for specific species and timber sizes (Gadow, 2012) The effectiveness of selective harvesting may hinge on the density of commercial trees removed per area (Whitmore, 1984) Additionally, highly mechanized logging operations can cause extensive damage, affecting not only commercially valuable species but also the overall forest ecosystem (Johns, 1985).
Selective logging in Vietnam
Selective logging is the predominant method for commercial timber production in Vietnam, utilized in natural forests since the 1960s under the term "selected cutting with natural regeneration." This silvicultural approach aims to enhance harvesting efficiency, minimize damage to remaining trees, reduce forest degradation, and boost yields for future cutting cycles.
The selected cutting with natural regeneration system (MARD, 1993) can be described as follows:
Applied forests have a standing volume ≥ 130 m 3 per ha;
The minimum diameter at breast height (DBH) for cutting is determined by timber groups; it ranges from 40 to 50 cm DBH;
The cutting cycle is 35 years,
The maximum cutting intensity is 38 % of the total standing volume
In Vietnam's selective logging practice, the focus is primarily on commercial species and the best-quality stems, neglecting non-commercial species despite their maturity, which negatively impacts the quality of remaining trees and overall forest health (Ho, 1999) This harvesting method reduces the number of individuals per hectare and threatens species resilience and genetic diversity, as it disrupts seed dispersal mechanisms (Le, 1996; Ho, 1999) Therefore, it is essential to study the effects of selective logging on forest structures, natural regeneration, and dynamics in Vietnam.
Forest structure and Dynamics
Forest structure refers to the distribution of tree attributes within a forest ecosystem and is influenced by both natural processes and human disturbances It plays a crucial role in driving ecosystem processes and maintaining biological diversity Understanding forest structure is essential for planning harvest events and managing subsequent forest dynamics Without a clear grasp of forest structure and diversity, it becomes challenging to implement effective silvicultural practices and learn from outcomes.
Forest structure, as defined by Oliver and Larson (1990), refers to "the physical distribution and time of the tree in a place." Key attributes of forest structure include the type, size, shape, and spatial distribution of its components (Speis, 1998) These structures can be characterized using inventory data, which encompasses tree abundance, average diameter, average height, basal area, standing mass, and frequency distribution, along with other quantitative metrics (Brodbeck, 2004).
Forest dynamics refers to the changes in forest structure and species composition over time, influenced by both natural and human disturbances Understanding these dynamics is crucial for creating effective management strategies for harvesting and conserving forest resources, as it helps predict future forest structures and development patterns Measuring tree growth through diameter is vital for forest managers, as diameter increments and growth patterns of various species directly relate to forest stand productivity This information is essential for sustainable forest management, making diameter increment a key variable in allometric equations.
The dynamics of natural forests, particularly in tropical regions, are complex and challenging to assess due to the limited representation of many species Changes in species composition are indicative of natural restoration processes, as noted by Okali and Ola-Adams (1987) Early identification of species composition in secondary rainforests can occur during the development of large canopy and emergence species However, our understanding of these changes remains constrained by the scarcity of long-term studies focused on demographic shifts and tropical forest species.
Nearest Neighbor Characteristics
The structural characteristics of forest stands are defined by the distribution of individuals of the same species, typically varying in diameter and age In mixed forests, the population structure of tree species is influenced by interspecific and intraspecific variations in size, species mingling, and distribution patterns These patterns reflect how individuals are arranged in space, which relates to competition and resource utilization among neighboring trees Tree size indicates the maturation level of a population and its competitive edge within the community, impacting survival viability and ecological niches Intraspecific aggregation highlights the isolation of species within a community, closely linked to seed dispersal, regeneration capacity, and growth.
Over the decades, various methods have been developed to describe forest structural attributes, but accurately detailing small-scale structural features has gained increasing significance Recent advancements have introduced individual tree indices, including the uniform angle index and species mingling and dominance, which focus on characterizing the neighborhood of a reference tree through its n-nearest neighbors Utilizing nearest neighbor statistics enables a better understanding of relationships within tree groups based on species and size class at smaller scales This approach offers several advantages over traditional methods that rely on expression frequency, as it highlights greater structural diversity indicated by increased species inhomogeneity and size class homogeneity.
Our primary objective is to analyze the spatial characteristics of neighborhood trees using modern nearest neighbor statistical techniques To enhance our understanding of structural units, we incorporated three key structural metrics for each species in our analysis: mingling-uniform angle index, mingling-dominance, and dominance-uniform angle index.
MATERIALS AND METHODS
Study site
The Kon Ha Nung region in Vietnam's Central Highlands boasts one of the highest forest cover rates, with approximately 50% of its total land area covered by forests, totaling around 126,000 hectares, primarily consisting of moist evergreen forests Recognized as a significant production forest area in Gia Lai Province, this region is celebrated for its exceptional biological and cultural diversity However, recent years have seen a concerning decline in forest cover in the Central Highlands, decreasing from 53.9%.
From 1999 to 2012, the rate of deforestation in Kon Ha Nung has remained alarmingly high, dropping forest cover to 50.9% (FPD, 2013) The growing demand for timber, non-timber products, and firewood is intensifying pressure on the remaining forest areas, despite an increase in secondary forest cover (Le, 2012) This ongoing deforestation has led to a significant decline in overall forest quality, raising urgent concerns about the need to protect and restore the remaining forests (Le, 1996; Ho, 1999; Joern, 2010).
Figure 2.1 The location of the study site in Gia Lai Province, Central Highlands of
In the early 1980s, the Kon-Nung Experimental Station (KES) forests experienced varying intensities of exploitation, ranging from high to low impact (Le, 1996; Ho, 1999) Despite this, research on the effects of logging on forest structure, diversity, and restoration processes in Kon Ha Nung remains limited (Vu, 1985) There is a pressing need to understand the long-term impacts of selective logging on forest structure, plant species composition, and natural regeneration in this region Additionally, the potential ecological consequences of logging in these forests over time are still unclear While current logging practices are deemed sustainable for ensuring sustainable forest management (SFM), further clarity is needed to develop effective silvicultural concepts and identify suitable species for the long-term restoration of the wet evergreen forest in Kon Ha Nung Understanding the forest's structure, regeneration, and dynamics post-logging is essential for informed management practices.
To address the challenges in understanding tropical tree growth rates, initial investigations involved establishing permanent sample plots and conducting repeated tree measurements However, the available data was often insufficient for a comprehensive analysis of tree growth throughout their lifespan (Le, 1996; Ho, 1999) Consequently, it is essential to design and implement long-term experiments that monitor forest responses to varying logging intensities, as this approach will effectively reveal the dynamics of changes within these ecosystems.
This study analyzes long-term permanent plots to assess the structural recovery of forests by observing changes in species composition, mortality, recruitments, and other relevant forest structure variables The findings aim to enhance practical procedures for forest management and support the maintenance of diverse forest functions in Vietnam, with a specific focus on improving sustainable management practices for the forests of Kon Ha Nung.
The study was conducted in the Kon Ha Nung forests located in K’Bang District, Gialai Province, within the Central Highlands of Vietnam This research site is situated in the northeastern part of Gialai Province, specifically between the coordinates 14°00' - 14°30' N and 108°17' - 108°44' E (Ngo Van Cam, 2015).
Topography and soil
The topography in K‟Bang District tends to become gradually lower from west to east and from north to south The district is mainly characterized by ridges of hills around
The study area is characterized by an elevation of 700 meters above sea level, featuring a diverse topography The western region comprises medium to high hills, while the eastern section is marked by smaller hills and some plains The hills are interspersed with narrow valleys, and the slopes typically do not exceed 25 degrees.
Kon Ka Kinh is the highest mountain in the area under scrutiny; it is located in Western K‟Bang District and has an altitude of 1,748 m
According to Le (1996), four main soil types are found in K‟Bang District:
I Haplic nitisols, which is developed from acidic magma parent rock and has medium soil horizons;
II Rhodic ferralsols, developed from neutral to alkaline magma parent rock and with a deep humus layer and a high humus ratio;
III Plinthic ferralsols, formed in valleys as a result of alluvial deposits; and
IV Ferralic acrisols originating from granite parent rock The composition and distribution of soil types varies among areas Other soil types are recorded but in small amounts
The study area, situated in a tropical monsoon zone, has climatic data collected over a 20-year period from 1990 to 2010 at the Vinh Son Meteorology and Hydrology station, approximately 12 km from the site This region experiences two distinct seasons: the rainy season from April to November and the dry season from December to March The mean annual temperature is 23.6°C, with the warmest months typically being June and July, where temperatures can peak at 29.6°C Conversely, the coldest temperatures, around 13.6°C, are recorded in January.
Table 2.2 Climatology information of the study site (Nguyen et al., 2011; NCEF, 2014)
Rainfall (mm) No of days with rainfall
The region experiences an annual rainfall of 2,042 mm, with over 90% of this precipitation occurring during the rainy season October records the highest monthly average rainfall at 318 mm, while February sees significantly lower levels, averaging less than 40 mm Additionally, the average monthly humidity in the area is 82%.
Figure 2.3 Climate diagram of Vinh Son, near the study site, data recorded from 1990 to 2010 (Nguyen et al., 2011)
Forest resources
K'Bang district is home to a vast forest area of 126,000 hectares, representing 68% of its total land, predominantly featuring tropical moist evergreen forests found at elevations of 500-900 meters Additionally, small montane forests exist on peaks and ridges above 900 meters The forest coverage in K'Bang significantly surpasses the national average, with 68% compared to Vietnam's overall 39.9% However, only about 20% of this forest is classified as protected.
The forest classification reveals that 36% (approximately 48,000 hectares) of the area is designated as "special use," while 10% (around 13,000 hectares) is categorized as protection forest The majority, comprising 54% (about 72,000 hectares), is classified as "production" forest, indicating a diverse range of forest types from rich to poor and including plantations (PPC, 2008; Ngo Van Cam, 2015).
K’Bang District is renowned for its remarkable biological and cultural diversity, earning recognition from numerous national and international conservation organizations as a biodiversity hotspot in Vietnam (Pollard, 2005).
The Kon Ha Nung forests, located in Binh Dinh Province, showcase a rich biodiversity that integrates characteristics from four distinct floristic regions: Indian, Malaysian, Sino-Himalayan, and Indochinese Research conducted by Ngo Van Cam in 2015 revealed that these forests are home to 546 vascular plant species across 376 genera and 122 families Among these, 201 species are woody, 120 are recognized for their medicinal properties, and 48 species are categorized as "other," which includes plants suitable for ornamental use.
In K'Bang district, the majority of forest areas are managed by local government agencies, which include eight state forest enterprises, national parks, nature reserves, and a forest test station The provincial governments are implementing local forest management strategies in alignment with the plans approved by the Ministry of Agriculture and Rural Development (MARD) Similar to other natural forests in the Central Highlands, the forests in K'Bang are subject to these management practices.
K'Bang district is a key area for timber production, particularly for medium and large trees, and serves as one of the primary production forests in Gialai province The volume of timber harvested peaked in 1988 at over 123,000 m³, but has since declined due to a logging ban imposed in 1997 on nearly all natural forests in Vietnam From 2000 to 2010, the annual timber harvest was estimated at just 20,000 m³, with current logging permits being limited to specific small areas.
History of silvicultural and forest management practices in the study site
Established in 1980, the Kon Ha Nung Experiment Station (KES) serves as a key site for researching tropical and forestry ecosystems, particularly focusing on silvicultural practices Managed by the Tropical Forest Research Center (TFRC) of the Forest Science Institute of Vietnam (VAFS), KES encompasses 1,400 hectares of forest Prior to logging in 1980, the study areas exhibited uniform forest formations, with the silvicultural approach utilized being "selected cutting with natural regeneration" (Ho, 1999) The elevation variation within KES is minimal, at less than 20 meters, resulting in consistent topographical conditions across the research site.
In 1980, logging activities with varying intensities were implemented in a section of the KES, categorized as low (less than 30% of standing volume extracted) and high (30% - 50% of standing volume extracted), while approximately 100 hectares remained unlogged The forests have been well-preserved post-logging, with minimal harvesting or significant damage, apart from occasional collection of non-timber forest products by local residents.
Logging operations begin with marking trees using consecutive numbers on yellow plastic cards, followed by cutting them down with a chainsaw Deltas are primarily managed and transported along slopes for easier access Additionally, loggers utilize bulldozers to build roads, side paths, timber piers, and to facilitate the transportation of timber away from the site.
14 system by winches After logging, no other treatments (egg, cutting, climbing, grafting of non-commercially valuable species, and removal of the hairs of logged logs) are carried out.
Data collection and analysis
This study, initiated in 2004, evaluates the species composition, forest structures, natural regeneration, and dynamics of key forest ecosystems in Vietnam It aims to assess the impacts of different silvicultural treatments on previously logged natural forests Utilizing a stratified random sampling method alongside field observations, the research encompasses a comprehensive range of forest stands, categorizing three forest types based on their levels of past disturbance.
1999) and can be described as follows:
Low-impact Logging (LIL), where the standing volume extracted was less than 30% of the total standing volume
High-impact logging (HIL), where logging ranged from 30% to 38% of the total standing volume
For each of three forest types UF, LIL, HIL one-ha permanent sample plots (PSPs) were established in 2004
In each forest type, a one-hectare sample plot (100 x 100 m) was established and divided into 25 subplots measuring 20 m x 20 m each These plots are oriented north, with corners clearly marked using concrete beams (15 cm x 15 cm x 75 cm) and documented on a site map via Garmin GPSMAP-76 CSx Reference posts were utilized to accurately lay out a 20 by 20 m grid across the plot, with each thread carrier beginning on a compass bearing to ensure precise alignment.
15 post 25m ahead by an individual The procedures for establishing, measuring, and maintaining such study plots were recommended by Alder and Synnot (1992)
Figure 2.5 Layout of a 1 ha sample plot
Data collection took place in designated sample plots, where all trees with a diameter at breast height (DBH) of 10 cm or more were numbered and marked with red paint The location of each tree was mapped and integrated into a GIS database using MapInfo Professional 11.0 Fieldwork adhered to specific guidelines to ensure accuracy and consistency.
Field identification of species was conducted, and for any unidentifiable stems, samples including leaves, flowers, fruits, and bark were collected for further analysis by taxonomists in the laboratory.
- Total tree height was measured by a mechanical-optical device (Blume- Leiss altimeter)
- Tree diameter was measured at 1.3 m height above ground by using tape
The data were offered by Dr Ngo Van Cam, Tropical Forest research Center, Pleicu, Gialai
Importance Value Index (IVI): is a measure of how dominant a species is in a given forest area
IVI (%) = (Relative density + relative Basal area)/2
Relative density (RD) is the number of individuals per area as a percent of the number of individuals of all species
Relative basal area is the total basal area of Species A as a percent of the total basal area of all species
The Shannon-Wiener index is a statistical tool used to measure biodiversity, assuming that all species in a sample are represented and randomly sampled It is calculated using the formula where 'p' represents the proportion of individuals of a specific species (n) to the total number of individuals (N), with 'ln' denoting the natural logarithm, 'Σ' indicating the sum of calculations, and 's' representing the total number of species.
The Simpson‟s index is a dominance index because it gives more weight to common or dominant species
The presence of a few rare species with limited representatives does not significantly impact overall biodiversity In the Simpson index, the proportion (p) of individuals of a specific species is calculated by dividing the number of individuals of that species (n) by the total number of individuals (N), while Σ represents the sum of these calculations, and s denotes the total number of species.
Species evenness measures the relative abundance of different species within an environment, indicating how closely balanced their populations are It is quantified using a diversity index, which assesses the numerical equality of species in a community Pielou's evenness index is commonly used to represent this concept mathematically, providing insights into the biodiversity of an ecosystem.
Species mingling (M) refers to the composition and spatial arrangement of tree species within a forest It is quantified by the ratio of the nearest neighbors that belong to different species compared to a reference tree.
∑ v j = 1 if neighbor j is not the same species as reference tree i, otherwise v j = 0
Dominance (U): describes the size differentiation between a reference tree and its four nearest neighbors It is defined as the proportion of n nearest neighbors that are smaller than reference tree (Figure.2.6b)
∑ vj = 0 if neighbor j is smaller than reference tree i, otherwise vj = 1
Uniform angle index (W): describes the degree of regularity for the four nearest neighbors as reference tree It is defined as the proportion of angle () smaller than the standard angle
Figure 2.6: Definition of the spatial parameters: Mingling (a), Dominance (b) and
The methods described above were implemented by using softwares Past 3.0
To ensure accurate paleontological statistics using Microsoft Excel and Crancord, we implemented the nearest neighbor edge correction method, as proposed by Pommerening & Stoyan (2006), to mitigate the edge effects in the calculations of Mi, Wi, and Ui.
RESULTS AND DISCUSSION
Tree species diversity and composition
Figure 3.1 Simpson diversity indices of three forest types from 2004-2012
Comparing three forest types shows that Simpson diversity indices tend to same with the levels of disturbance (UL, LIL, and HIL) from D= 0.9
Compare for each forest type through time, we can see UL has no change in Simpson index for 3 years (2004-2012), D=0.9
LIL has no change in Simpson index for 3 years (2004-2012), D=0.9
HIL has no change in Simpson index for 3 years (2004-2012), D=0.9
Figure 3.2 Shannon diversity indices of three forest types from 2004-2012
Comparing three forest types shows that Shannon indices tend to increase with the levels of disturbance (UL, LIL, and HIL) from H= 3.7 to 3.9
Compare for each forest type through time, we can see UL has no change in Shannon index for 3 years (2004-2012), H= 3.8
In LIL, Shannon indices tend to increase from 2004-2008 (from H= 3.7 to 3.8) after that it is no change with H= 3.8
HIL has Shannon index higher than 2 different forest types (UL= 3.8, LIL= 3.8) but, it has tend to same for three years (2004-2012) with H= 3.9
Figure 3.3 Evenness diversity indices of three forest types from 2004- 2012
Comparing three forest types shows that Evenness indices tend to increase with the levels of disturbance (UL, LIL, and HIL) from J= 0.57 to 0.61
Compare for each forest type through time, we can see UL tend to increase for 3 years (2004-2012), J= 0.58 to 0.6
In LIL, Shannon indices tend to same from 2004-2008 (from J= 0.59) after that it decrease with J= 0.59 to 0.57
HIL has Evenness index higher than 2 different forest types (UL= 0.6, LIL= 0.59) but, it tend to decrease from 2008- 2012 with J= 0.61 to 0.58
Figure 3.4 Species richness of three forest types from 2004 - 2012
Comparing three forest types shows that species richness tend to increase with the levels of disturbance (UL, LIL, and HIL) from 74 to 93(number of species)
Compare for each forest type through time, we can see UL change in species richness for 3 years (2004-2012), species = 77 to 79(number of species)
In LIL, species richness tend to increase from 2004-2012 with species = 74 to 79(number of species)
HIL has species richness higher than 2 different forest types (UL= 79, LIL= 79) from species = 89 to 93(number of species)
Figure 3.5 Tree abundance of three forest types from 2004- 2012
A comparison of three forest types reveals that tree abundance generally increases with disturbance levels, ranging from 503 to 674 trees Analyzing the changes in tree abundance over time for each forest type, the UL forest shows a slight increase from 529 to 537 trees between 2004 and 2012.
In LIL, Tree abundance tend to decrease from 2004-2008 (from number of trees = 626 to 619) after that it tend to increase from 2008-2012 with number of trees= 619 to 672(number of trees)
HIL has Tree abundance tend to decrease from 2004-2008 (from number of trees = 649 to 641) after that it tend to increase from 2008-2012 with number of trees= 641 to 674(number of trees)
2004: 17.27AG + 9.81GP + 5.01BH + 4.85WP + 0.10 DF + 0.10PB + 0.10SL
2008: 17.49AG + 10.75GP + 4.99BH + 4.55WP + 0.11DR + 0.10SL + 0.10DF
2012: 17.14AG + 10.25GP + 4.72 BH + 4.32 WP +0.11PN + 0.103PB + 0.101LD
AG- Aglaia gigantea; GP- Garuga pierii; BH- Baccaurea harmadii; WP- Wendlandia paniculata; DF-Dimocarpus fumatus; DR- Diospyros rubra; PN- Polyalthia nemoralis;
SL- Symplocos lucida; LD- Lansium domesticum, PB -Podocaropus brevifolius
Almost dominant trees species composition does not change over time
2004: 7.13MB + 5.18DC + 5.17Ms + 4.65Qs + 4.32NB + 4.03As + 3.78Ss + 2.99Ds +
2008: 7.24MB + 5.42DC + 4.87Qs + 4.59Ms + 4.37NB + 3.82Ss + 3.48As + 2.99Ds
2012: 7.32MB + 5.57DC + 4.83Qs + 4.56Ms + 4.48NB+ 4.1Ss + 3.32SJ + 3.31As
The article discusses various plant species, including MB (Michelia braianensis), DC (Dialium cochinechinensis), Ms (Machilus sp), Qs (Quercus sp), NBP (Nephelium bacsasense), As (Aphanamixis sp), Ss (Symplocos sp), Ds (Duchesnea sp), AP (Acronychia pedunculata), MA (Melia azedarach), Ns (Nephelium sp), SJ (Syzygium jambos), EM (Evodia meliaefolia), and MI (Mangifera indica) Each of these species contributes to biodiversity and ecological balance, showcasing the importance of preserving plant varieties for environmental health and sustainability.
Almost dominant trees species composition does not change over time But from 2004-
2008 Ms, As was replaced Qs, Ss
2004: 7.43Ds + 6.96MB + 6.49WP+ 5.1VP + 4.86CP + 4.75Ms + 3.49AG + 2.92SJ +
2008: 7.43Ds + 7.16MB + 6.43WP + 5.03VP + 4.97Ms + 4.86CP + 3.53AG +
2012: 7.09MB + 6.98Ds + 6.39WP + 4.82Ms + 4.59SH + 4.4VP + 3.47AG + 2.88MI
Ds- Duchesnea sp, MB- Michelia braianensis, WP- Wendlandia paniculata, VP- Vitese parviflora, CP- Chisochenton paniculatus, Ms- Machilus sp, AG- Antidesma ghasembilla,
SJ- Syzygium jambos, MI- Millettia ichthyochtona, SM- Sapindus mukorossi, CA- Canarium album, PB- Podocaropus brevifolius, Ss2- Syzygium sp2, SH- Syzygium hancei,
GM- Gonocaryum maclurei, KC- Knema corticosa, SR- Shorea roxburghii
Dominant trees species composition change over time from 2004- 2008 is Ds but 2012 is
VP was replaced Ms thus, LIL tend to change dominant trees species more than (UL, HIL)
Figure 3.6 Change diameter of three forest types from 2004- 2012
Comparing three forest types shows that DBH tend to increase with the levels of disturbance (UL, LIL, and HIL) DBH = 22.22 to 27.41(cm)
Compare for each forest type through time, we can see UL has change in DBH for 3 years (2004-2012), DBH = 25.68 to 26.74(cm)
In LIL, DBH tend to increase from 2004-2008 (from DBH = 23.82 to 24.99) after that it tend to decrease from 2008-2012 with DBH = 24.99 to 24.83(cm)
HIL has DBH tend to decrease from 2004-2008 (from DBH = 22.22 to 23.45) after that it no change from 2008-2012 with DBH = 23.45(cm)
Figure 3.7 Change High of three forest types through time series
Comparing three forest types shows that Tree height dynamics tend to increase with the levels of disturbance (UL, LIL, and HIL) High = 17.8 to 26.7(m)
Compare for each forest type through time, we can see UL has change in Tree height dynamics for 3 years (2004-2012), High = 18.3 to 26.7(m)
In LIL, Tree height dynamics tend to increase from 2004-2008 (from High = 18.3 to 19.2) after that it tend to decrease from 2008-2012 with High = 19.2 to 18.9(m)
HIL has Tree height dynamics tend to decrease from 2004-2008 (from High = 18.3 to 17.8) after that it change from 2008-2012 with High = 17.8 to 18.5(m)
The estimation of mortality and recruitment is an elementary descriptor of population in tropical forests; selective logging is sustainable only when tree mortalities are balanced by recruitment and growth
Figure 3.8 Mortality of three forest types through time series
Comparing three forest types shows that mortality tend to increase with the levels of disturbance (UL, LIL, and HIL) Mortality= 17 to 71(number of tree)
Compare for each forest type through time, we can see UL has change in mortality for
2 years (2008-2012), mortality= 37 to 43 (number of tree)
In LIL, mortality tend to increase from (2008-2012) Mortality= 22 to 32
HIL has mortality tend to increase more than (UL, LIL) from 2008-2012 mortality= 17 to 71(number of tree)
Recruitment is closely linked to logging intensity, peaking in the High Intensity Forest (HIF) due to improved conditions for regeneration, such as soil and light disturbance Conversely, the lower recruitment levels observed in the Low Intensity Forest (LIF) may be attributed to selective logging practices that retained only a limited number of strong trees.
Figure 3.9 Recruitment of three forest types through time series
Comparing three forest types shows that Recruitment tend to increase with the levels of disturbance (UL, LIL, and HIL) recruitment = 9 to 66 (number of tree)
Compare for each forest type through time, we can see UL has change in recruitment for 2 years (2008-2012), recruitment = 17 to 57 (number of tree)
In LIL, recruitment tend to increase from (2008-2012) recruitment = 22 to 32 (number of tree)
HIL has recruitment tend to increase more than (UL, LIL) from 2008-2012 recruitment
We used only data recorded in 2012 to analyze with nearest neighbor characteristics due to recruitment and mortality did not chance significantly from 2004-
Figure 3.10 Mingling characteristics of all trees in three study plots
The analysis of the Kon Ha Nung plot, illustrated in Figure 3.10, reveals that species mixture (mingling) is concentrated at high levels, ranging from a mixture index of 0.1 to 0.89 This evidence indicates that the dominant species are highly integrated with other tree species in their vicinity.
Figure 3.11 DBH characteristics of all trees in three study plots
The analysis of diameter at breast height (DBH) dominance in relation to nearest neighbors revealed that the species examined exhibited lower DBH values compared to their closest counterparts The DBH measurements were categorized into three time series: unlogged, low impact logged, and high impact logged, with values ranging from 0.19 to 0.24.
Figure 3.12 Uniform Angle Index characteristics of all trees in three study plots
The Uniform Angle Index illustrates the spatial distribution of reference individuals in relation to their nearest neighbors The dominant species exhibited patterns ranging from regular to clumped, with values of W between 0.01 and 0.63 These findings indicate that the Uniform Angle Index reflects a random arrangement among other tree species in the vicinity.
CONCLUSION
Tree species diversity and composition
High tree species richness and diversity tend to be a general property of the moist evergreen forest in Kon Ha Nung in comparison with the world‟s other tropical forests
In 2012, a study recorded 100 species across 46 families in the Kon Ha Nung forests, with species richness ranging from 70 to 86 species per hectare The Shannon-Wiener index varied between 3.7 and 3.9, while the Simpson index stood at 0.9, and the Evenness index ranged from 0.57 to 0.61 The research indicated an increase in species richness from 74 to 93 and tree abundance from 503 to 674 trees Additionally, the diameter at breast height (DBH) increased from 22.22 to 27.41 cm, and tree height dynamics showed growth from 17.8 to 26.7 cm The dominant families in these forests included Myrtaceae, Lauraceae, Magnoliaceae, Sapindaceae, Burseraceae, Euphorbiaceae, and Fagaceae.
The moist evergreen forests of Kon Ha Nung are characterized by a diverse mix of tree species, including Paramichelia braianensis, Dialium cochinchinensis, Machilus odoratissimus, and others This diversity indicates that the forest does not consist of a single dominant species, but rather a combination of various species such as Dacryodes dungii, Baccaurea harmadii, and Castanopsis poilanei, among others.
Selective logging at varying intensities showed minimal impact on tree diversity and species composition, with logged forests exhibiting a 90% resemblance to unlogged forests (UF) Although slight differences were observed at both family and species levels, diversity indices such as Shannon-Wiener (H), Simpson's (D), and Evenness (J) indicated no significant differences among the three forest types Overall, selective logging resulted in negligible damage to forest diversity.
Thirty-one forests exhibited diversity values similar to those of the UF, indicating a potential for the tree community to revert to a state resembling its former condition over time However, the presence of persistent invasive species poses a significant threat, suggesting that the forest may never completely regain its original diversity.
Tree dynamics
The study reveals minimal differences in forest-related changes such as species composition, diversity, structure, and growth, all influenced by logging intensity Notably, undisturbed (UL) forests exhibit a more stable species composition, with the top ten UL species remaining unchanged over time Conversely, species like Pometia lecomtei and Polyalthia laui show a decline in importance in both high and low impact logged sites These declining species are being supplanted by less vibrant, intermittently occurring species, such as Paramichelia braianensis, suggesting that logged sites are gradually shifting towards the species composition characteristic of undisturbed forests.
Different logging intensities resulted in varying stem dynamics across three forest types, with high logging grades showing a greater proportion of tree recruitment compared to lower intensities or unlogged areas This phenomenon was attributed to the removal of large stems, which created larger gaps and increased resource availability for new recruits However, mortality rates did not vary among the different forest types.
All forest stands exhibited a consistent trend of increasing basal area and stand volume Between 2004 and 2012, there were no significant differences in mean basal area growth rates across stands subjected to varying logging intensities However, diameter increments were notably affected by these logging intensities, as demonstrated by the observed trends in diameter growth over time among the three stands.
32 forest types Overall, annual diameter increments varied between 0.2 – 0.3 mm Median diameter increments in both the unlogged and low impact logged sites were lower than those in the high impact logged.
Nearest neighbor characteristics
Spatial structure is one of the key parameters to describe standing forest structure
We calculated and described structural parameters such as Mingling, DBH dominance and Uniform Angle Index by using Crancod and Microsoft Excel softwares The results showed that: most of studied species were found highly mixed with other species, M= 0.1-0.89 These evidences shown that these dominant species were highly mixed with other tree species in adjacent neighbors In DBH dominance analysis, DBH= 0.19- 0.24 the results showed that these species were less advantage in DBH comparing to their nearest neighbors Uniform Angle Index shows spatial distribution of reference individuals to their nearest neighbors, Dominant species were regular to clumped pattern with W= 0.01-0.63 These evidences shown that these uniform angle index were random with other tree species in adjacent neighbors
1 Ngo Van Cam, 2015.Long-term effects of logging on structures and dynamics of moist evergreen forests in Kon Ha Nung, Central Highlands of VietnamThai,
2 Brodbeck, F., 2004 Structure and Processes in Traditional Forest Gardens of Central Sulawesi, Indonesia Cuvillier Verlag Goettingen, Goettingen, Germany
3 Pommerening A & Stoyan D (2006) Edge-correction needs in estimating indices of spatial forest structure Canadian Journal of Forest Research 36(7): 1723-1739
4 Alder, D., Synnott, T.J., 1992 Permanent sample plot techniques for mixed tropical forest Department of Plant Sciences, University of Oxford, Oxford
5 PPC, 2008 Degree No 53/QD-UBND dated 04 Feb 2008 reclassified the Gia Lai forest into 3 categories: special use, production and protection forest (in Vietnamese) In: Forestry, D.o (Ed.) People'commitee of Gia Lai province, Pleiku
6 Sub-FiPi, 1999 Planning of 3 forest types and forest land in Central Highlands (technical report, in Vietnamese) In Sub-Forest Inventory & Planning Institute, Quynhon
7 Pollard, E.H., 2005 Preliminary high conservation value forest assessment for
Ha Nung State Forest Enterprise In WWF Greater Mekong programme, Ha Noi
8 Okali, D., Ola-Adams, B., 1987 Tree population changes in treated rain forest at Omo Forest Reserve, south-western Nigeria Journal of tropical ecology 3, 291-313
9 Speis, P.A., 1998 Forest Structure: A key to the ecosystem In: MacKinnon, J.A.T.a.A (Ed.), Structure, Process, and Diversity in Successional Forests of
Coastal British Columbia Natural Resources Canada, Victoria, British Columbia, Canada
10 Swaine, M., Lieberman, D., Putz, F., 1987b The dynamics of tree populations in tropical forest: a review Journal of tropical ecology 3, 359-366
11 Oliver, C.D., Larson, B.C., 1990 Forest stand dynamics McGraw-Hill, Inc., New York
12 Joern, S., 2010 Assessment of status of natural forest resources at the SFCs Ha Nung and So Pai In WWF Greater Mekong Programme, Hanoi
13 Smith, R., Nichols, J., 2005 Patterns of basal area increment, mortality and recruitment were related to logging intensity in subtropical rainforest in Australia over 35 years Forest Ecology and Management 218, 319-328
14 Nguyen, B.N., Nguyen, Q.T., William, D.S., Yurdi, Y., 2009 Forest and Poverty Data in Vietnam: Status, Gaps, and Potential Uses RECOFTC, Rights and Resources Initiative, Vietnam Forestry University, Hanoi
15 Vu, T.P., Nguyen, T.M.L., Nguyen, N.L., Do, D.S., 2011 Final report on forest ecological stratification in Vietnam In, United Nation program on reducing emission from deforestation and degradation Research Center on Forest ecology and Environment, Ha Noi
16 Nzogang, A., 2009 Tropical forest dynamics after logging-natural regeneration and growth of commercial tree species - in southeast Cameroon In, Dissertation Faculty of Forest and Environmental Sciences, Albert-Ludwigs- Universitọt, p 170
17 USAID, 2013 Vietnam tropical forest and biodiversity assessment In Sun Mountain International and the Cadmus Group, Inc Hanoi
18 FPD, 2013 Forest Condition Change of 2011 In Forest Protection Department, Ha Noi
19 MARD, 2006 Handbook for Forestry Sector In: Ministry of Agriculture and Rural Development, M (Ed.) Transportation Publisher, Ha Noi
20 MARD, 2005a Regulations on harvesting timber and forest products (Decision number 40/2005/QĐ-BNN) In: Development, M.o.A.a.R (Ed.), Ha Noi
21 MARD, 1993 Principles for silviculture techniques applied for forests that provide Timber and bamboo In: Forestry, M.o (Ed.), QPN 14-92 Agriculture Publishing House, Ha Noi
22 Vu, D.P., 1985 Study on growth of pure and mixed forests for sustainable forest management in Vietnam In Vietnamese Academy of Forest Sciences,
23 Whitmore, T.C., 1990 An introduction to tropical rain forests Oxford University Press, USA
24 Whitmore, T.C., 1984 Tropical Rain Forests of the Far East Clarendon Press, Oxford
25 Richards, P.W., 1996 The tropical rain forest: an ecological study Cambridge University Press, New York
26 Regeneration changes in tree species abundance, diversity and structure in logged and unlogged subtropical rainforest over a 36-year period Forest Ecology and management
27 Vanclay, J.K., 1994 Modelling forest growth and yield: applications to mixed tropical forests School of Environmental Science and Management
28 Vanclay, J.K., 1991a Aggregating tree species to develop diameter increment equations for tropical rainforests Forest Ecology and Management
29 De Jong, W., Do Dinh Sam, Trieu Van Hung, 2006 Forest rehabilitation in Vietnam: histories, realities and future Center for International Forestry Research (CIFOR), Bogor, Indonesia
30 Gunter, S., Weber, M., Stimm, B., Mosandl, R., 2011 Silviculture in the Tropics Springer Berlin Heidelberg
31 Lobo, J., Barrantes, G., Castillo, M., Quesada, R., Maldonado, T., Fuchs, E.J., Solas, S., Quesada, M., 2007 Effects of selective logging on the abundance, regeneration and short-term survival of Caryocar costaricense (Caryocaceae) and Peltogyne purpurea (Caesalpinaceae), two endemic timber species of southern Central America Forest Ecology and Management
32 Rapera, R., 1977 Effects of logging on residual stands BIOTROP Spec
33 Johns, A.D., 1985 Selective logging and wildlife conservation in tropical rain- forest: problems and recommendations Biological Conservation
34 Smith, R., Nichols, J., 2005 Patterns of basal area increment, mortality and recruitment were related to logging intensity in subtropical rainforest in Australia over 35 years Forest Ecology and Management 218, 319-328
35 Smith, D.M., Larson, B.C., Kelty, M.J., Ashton, P.M.S., 1997 The practice of silviculture: applied forest ecology John Wiley and Sons, Inc
36 Smith, D.M., Hawley, R.C., 1986 The practice of silviculture Wiley, New York, USA
37 Lamprecht, H., 1989 Silviculture in the tropics: tropical forest ecosystems and their tree species - possibilities and methods for their long-term utilization Technical cooperation-Federal Republic of Germany (GTZ), Eschborn
38 Gatti, R.C., Castaldi, S., Lindsell, J.A., Coomes, D.A., Marchetti, M., Maesano, M., Di Paola, A., Paparella, F., Valentini, R., 2014 The impact of selective logging and clearcutting on forest structure, tree diversity and above-ground biomass of African tropical forests Ecological Research,
39 Nzogang, A., 2009 Tropical forest dynamics after logging-natural regeneration and growth of commercial tree species - in southeast Cameroon In,
Dissertation Faculty of Forest and Environmental Sciences, Albert-Ludwigs- Universitọt,
40 Gadow, K.e.l., 2012 Forest Structure and Diversity In, Continuous Cover Forestry Springer Netherlands, Goettingen,
41 Gadow, K., Nagel, J.r., Saborowski, J., 2002 Continuous cover forestry: assessment, analysis, scenarios Springer Netherlands
42 Gadow, k., Hui, G., 1999 Modelling forest development In Kluwer Academic Publishers, Dordrecht
43 Whitmore, T.C., 1990 An introduction to tropical rain forests Oxford University Press, USA
44 Whitmore, T.C., 1984 Tropical Rain Forests of the Far East Clarendon Press, Oxford
45 Le, Q.H., 2012 Growth, demography and stand structure of Scaphium macropodum in differently managed forests in Vietnam In, Dissertation Utrecht University, Utrecht, Netherlands, p 168
46 Le, S., 1996 Research on forest structure and proposal new selecting cutting system in Kon Ha Nung - Central Highlands (in Vietnamese) In, Dissertation Forestry University of Vietnam, Hanoi, p 129
47 Ho, D.S., 1999 Researching on technical measure of forest habilitation and maintenance applied to natural broadleaf forest in Northern Central Highland (in Vietnamese) In: report, S (Ed.) Forest Science Institute of Vietnam, Hanoi
List of tree species in all three plots:
Number Vietnamese name Scientific name Family
1 Ba bét đỏ Mallotus metcalfianus Euphorbiaceae
2 Ba bét trắng Mallotus apelta Euphorbiaceae
3 Bạc lá Croton argyratus Euphorbiaceae
4 Bồ hòn Sapindus mukorossi Sapindaceae
5 Bơ rừng Persea sp Lauraceae
6 Bời lời cam bốt Litsea cambodiana Lauraceae
7 Bưởi bung Acronychia pendunculata Rutaceae
8 Cẳng gà Pteris semipinnata Pteridaceae
9 Chân chim Schefflera octophylla Araliaceae
10 Chẹo tía Engeldharta chrysolepis Juglandaceae
11 Chìa vôi Enddlela asiatica Enddleiaceae
12 Chòi mòi Antidesma corriaceum Euphorbiaceae
13 Chôm chôm Nephelium bacsasense Sapindaceae
14 Cò ke Microcos paniculata Tiliaceae
15 Cóc đá Dacryodes dungii Burseraceae
16 Côm lá nhỏ Elaeocarpus lanceifolius Elaeocarpaceae
17 Côm lá to Elaeocarpus apiculatus Elaeocarpaceae
18 Côm tầng Elaeocarpus dubius Elaeocarpaceae
19 Cứt ngựa Archidendron robisonii Mimosaceae
20 Dạ hương Disoxylum loureiri Meliaceae
21 Dâu gia đất Lansium domesticum Meliaceae
24 Dung lá nhỏ Symplocos sp2 Symplocaceae
25 Dung lá to Symplocos sp3 Symplocaceae
26 Dung sành Symplocos lucida Symplocaceae
27 Dung trứng Symplocos sp5 Symplocaceae
29 Giổi bà Michelia balansae Magnoliaceae
30 Giổi nhung Paramichelia braianensis Magnoliaceae
31 Gội nếp Aglaia gigantea Meliaceae
32 Gội tẻ Aglaia silvestris Meliaceae
33 Hoa khế Nuihonia sclerantha Ericaceae
34 Hoắc quang Wendlandia paniculata Rubiaceae
35 Hồng rừng Diospyros tonkinensis Ebenaceae
37 Kháo lá nhỏ Machilus thunbergii Lauraceae
38 Kháo lá to Machilus parviflora Lauraceae
39 Kháo lông Machilus robusta Lauraceae
40 Khế rừng Averrhoa sp Oxalidaceae
41 Kim giao Podocarpus fleuryi Podocarpaceae
42 Lát xoan Choerospondias axillaris Anacardiaceae
43 Lèo heo Polyalthia nemoralis Annonaceae
44 Lim xẹt Peltophorum ferrugineume Caesalpiniaceae
45 Lòng mang Pterospermum heterophyllum Sterculiaceae
46 Mặt cắt Rapanea neriiflolia Myrsinaceae
47 Máu chó Knema corticosa Myristicaceae
48 Máu chó lá to Knema pierrei Myristicaceae
49 Mít nài Artocarpus asperula Moraceae
51 Nhãn rừng Dimocarpus fumatus Sapindaceae
52 Nhọ nồi Diospyros pilosella Ebenaceae
56 Re bầu Cinnamomum obtusifolium sp Lauraceae
57 Re gừng Cinnamomum obtusifolium Lauraceae
58 Re lá nhỏ Cinnamomum albiflorum Lauraceae
59 Re lá to Cinnamomum sp Lauraceae
60 Săng đá Linociera sangda Oleaceae
61 Săng máu Horsfieldia sp Myristicaceae
62 Sao cát Anisoptera robusta Dipterocarpaceae
63 Sến đất Sinosideroxylon sp Sapotaceae
64 Sến mủ Shorea roxburghii Dipterocarpaceae
65 Sồi phảng Castanopsis cerebrina Fagaceae
68 Thạch đảm Tristania sp Sp
69 Thanh thất Ailanthus malabarica Simaroubaceae
70 Thị rừng Diospyros silvatica Ebenaceae
71 Thôi ba Alangium chinense Alanginacea
72 Thôi chanh Marlea begoniaefolia Cornaceae
73 Thông nàng Podocaropus imbricatus Podocarpaceae
74 Thông tre Podocaropus brevifolius Podocarpaceae
77 Trám chim Canarium tonkinensis Burseraceae
78 Trâm đỏ Syzygium zeylanicum Myrtaceae
79 Trám hồng Canarium subulatum Burseraceae
80 Trâm lá nhỏ Syzygium sp1 Myrtaceae
81 Trâm lá to Syzygium sp2 Myrtaceae
82 Trâm móc Syzygium jambos Myrtaceae
83 Trâm nước Syzygium ripicola Myrtaceae
84 Trâm quả to Syzygium hancei Myrtaceae