Assessing the diversity and density of birds at pine forest in tam dao national park

G OALS

This research aims to offer essential insights into the diversity and density of bird populations, which are vital for effective management and conservation efforts of biodiversity in the pine forests of Tam Dao National Park.

O BJECTIVES

- To assess the diversity of bird of pine forest in Tam Dao national park

- To estimate the density of bird of pine forest in Tam Dao national park

S TUDY AREA

The study was conducted in pine forest in Tam Dao national park with coordinates 21°21’- 21°42’ degrees north latitude, 105°23’- 105°44’ degrees east longitude

Tam Dao National Park spans an area of 34,995 hectares across Vinh Phuc, Thai Nguyen, and Tuyen Quang provinces, featuring the Tam Dao mountain range that stretches over 80 kilometers from Son Duong District to Phuc Yen Town The park is home to numerous mountains exceeding 1,300 meters, with the highest peak being Tam Dao Bac at 1,592 meters Of the total area, 26,163 hectares are covered by natural moist evergreen forests, accounting for 70% of the park's landscape Various forest types are present, including subtropical moist low mountain forests, pygmy forests, bamboo forests, and restoration areas post-mining However, human activities such as harvesting, cultivation, and forest fires have reduced forest coverage, particularly below 700 meters, where secondary shrubs and pine plantations have taken over previously forested land.

Study area is pine forest This is a plantation of Tam Dao national park with area approximate to 2900 ha In below map, study area is represented by violet.( Figure 3.1)

Source: Tam Dao national park Figure 3.1: Location of pine forest in TDNP

D ATA COLLECTION

C OLLECTING DATA

A bird survey was conducted in the pine forest of Tam Dao National Park from June to August 2016, utilizing data from a previous survey by Vu Tien Thinh et al (2012) for comprehensive analysis.

Equipments for survey are binoculars, cameras, data sheet, and map

Data collection was conducted using the line transect method to survey bird density and diversity This approach is particularly effective for small-sized animals and in terrains that are not conducive to movement during the survey process.

In a study conducted within a pine forest, 12 line transects were established at elevations ranging from 200m to 600m, randomly distributed throughout the area Bird surveys were conducted during peak activity times, from 5:30 AM to noon and from 3:00 PM to 6:00 PM, with each transect measuring 500 meters in length To minimize edge effects and the impact of status changes on detection probability, transects were positioned at least 70 meters away from status boundaries and spaced a minimum of 100 meters apart to ensure independence Observers walked along the transects at a speed of approximately 0.5 km every 40 minutes, recording the presence and number of birds by sight and sound, while also estimating the distance from the birds to the transect Each transect was surveyed six times to ensure comprehensive data collection.

Individuals are detected to be recorded as y 1, y 2 ,…yi in which y i is the distance from line transect to individual i

The collected data is recorded in data sheet as table below (Table 3.1)

Table 3.1: Field data sheet used to collect information

Site: ……… Observer (s): ………… Line transect number: … Time start/finish: ……

D ATA ANALYSIS

To estimate the density of objectives, surveyors must assess detection probability using investigative data This probability is determined based on the frequency distribution at various distances from the transect or observer A widely used software for analyzing detection probability is DISTANCE 6.0 (Thomas et al.).

Before estimating detection probability, surveyors must define detection probability function Four basic functions which are used to estimate detection probability by distance are:

With: g(y) is the detection probability of an object with distance to line transect is y (length unit); w, , b are the estimated parameters;

In addition to the four primary functions, three expansion series—Cosine, Simple Polynomials, and Hermite Polynomials—can effectively alter detection probability curves and enhance the simulation of detection probability variations The integration of standard functions with these expansion series is represented by the equation: g(y) = basic function(y) + expansion series(y).

In some cases, this combination can indicate the relationships between detection probability and distance better

The Akaike Information Criterion (AIC) is utilized to identify the most effective function for estimating detection probability, following the principle of balancing standard deviation and variance (Anderson, 2007) The function with the lowest AIC value is selected for this purpose Additionally, the Distance method provides χ2 values to assess the fit of the function to the empirical frequency distribution Once the optimal function is identified, detection probability (Pa) is calculated by dividing the area under the upper curve of the function by the rectangle area and multiplying it by the transect width.

The estimation of density with distance sampling cannot exactly unless five fundamental assumptions have to be satisfied:

Objects positioned directly on the line are detected with absolute certainty, achieving a detection probability of 100% (g(0) = 1) However, as the distance from the line increases, the likelihood of detection significantly diminishes (Buckland et al 2001).

 Objects do not move All measurements are made from the objects’ initial location, before it was affected by the observer (Buckland et al 2001)

 Measurements are accurate All angles, distance, objects, sex and other necessary measurements are measured with accuracy without any errors (Buckland et al

For animal species that exist in groups, cluster sizes can be recorded accurately However, while estimates may be precise near the observation point, they can become less reliable at greater distances To mitigate bias from size-biased sampling, it is recommended to apply the regression correction, which is the default setting for clustered data in the software Distance (Buckland et al 2001).

The sampled plots, whether circles or strips, should accurately represent the entire survey region While this assumption is typically inherent in appropriately randomized designs, it becomes crucial when dealing with non-random plots.

Although other assumptions are made, generally only the above five have any practical significance

In each transect measuring L meters, n objects are detected, and their perpendicular distances to the transect line are recorded Once all objects along the line are accurately identified, the object density in the survey, denoted as D, is estimated according to the method outlined by Buckland et al (2001).

With: n – Total objects is detected in the surveys ̂ - Detection probability ≤ 1 a = 2Lw (w – width of line transect, L – total line length)

Variance of density estimation is calculated by the formula: ̂ ( ̂) = ̂² { ̂( ) + ̂( ̂ )

( ) } Variance of density estimation show fluctuations the number of individuals detected in transect (E(s) is group size)

The total number of individual in each species (population size) is calculated by time density ( )̂ with area (A): ̂ ̂

S PECIES DIVERSITY AT PINE FOREST IN TDNP

In a survey of 12 line transects conducted in the pine forest of Tam Dao National Park, a total of sixteen bird species from eight families were identified Notable species included the Sooty-headed Bulbul (Pycnonotus aurigaster), Red-vented Bulbul (Pycnonotus cafer), Common Tailorbird (Orthotomus sutorius), and Ashy Drongo (Dicrurus leucophaeus) as detailed in Table 4.1.

Table 4.1: Bird diversity at pine forest in TDNP

No Family Common name Latin name Number of individual

Striped tit-babbler Macronous gularis 43

Laughingthrush Garrulax pectoralis 38 Timaliidae Puff-throated babbler Pellorneum ruficeps 21

Buff-breasted babbler Pellorneum tickelli 38

Red-whiskered Bulbul Pycnonotus jocosus 32

Pycnonotidae Red-vented bulbul Pycnonotus cafer 109

2 Sooty-headed Bulbul Pycnonotus aurigaster 172

Puff-throated Bulbul Alophoixus pallidus 71

3 Laniidae Rufous-backed shrike Lanius schach 24

4 Sylviidae Common tailorbird Orthotomus sutorius 141

5 Dicruridae Ashy Drongo Dicrurus leucophaeus 140

6 Cisticolidae Rufescent prinia Prinia rufescens 31

7 Corvidae Gray Treepie Dendrocitta formosae 16

Red-billed blue magpie Urocissa erythrorhyncha 34

8 Paridae Great tit Parus major 28

Table 4.2: Number and percentage of Observation and Hearing in detection species

Table 4.2 refers the number and percentage of Observation and Hearing in detection species:

In a study of 16 species, 63% of groups (601) were identified through observation, while 37% (353) were detected by sound Notably, as the distance from the transect increased, the proportion of groups detected by hearing rose, while the proportion identified through observation declined.

D ENSITY OF BIRD SPECIES AT PINE FOREST

D ESCRIBING SURVEY DATA

During the survey, 16 bird species groups were identified, but due to time constraints, we focused on four specific species: the Red-whiskered Bulbul (Pycnonotus jocosus), Red-vented Bulbul (Pycnonotus cafer), Sooty-headed Bulbul (Pycnonotus aurigaster), and Rufous-backed Shrike (Lanius schach) to estimate their density Most groups were detected within 20 meters of the transect lines To model detection as distance increased, the distance data was categorized into five groups, as detailed in the accompanying table.

M ODELING DETECTION PROBABILITY BY DISTANCE METHOD

The analysis of detection probability across four data sets reveals a consistent trend: as the distance from transect lines increases, the likelihood of detection decreases This trend is visually represented in detection probability histograms generated by the distance method By applying this method, the most appropriate functions for each species were identified, and χ2 values were calculated to assess the fit of these functions against the empirical frequency distribution A χ2 value exceeding 0.05 indicates that the selected function effectively models the variation in detection probability by distance The following section presents the distance data analysis, including histograms of the functions and tables of pertinent parameters.

Table 4.4: Red-whiskered Bulbul’ parameters

The Half-normal function, with the lowest AIC value of 101.1, is identified as the best fit model for representing the detection probability of Red-whiskered Bulbul (Pycnonotus jocosus) density Following this, the Uniform, Hazard-rate, and Negative exponential models exhibit higher AIC values Additionally, the χ2 value is 5.24, with 3 degrees of freedom and a p-value of 0.68, indicating a strong model for analyzing the fluctuations in detection probability by distance.

Figure 4.2: Detection probability functions for Red-vented bulbul

Table 4.5: Red-vented bulbul’ parameters

The Uniform function, with the lowest AIC value of 339.77, is identified as the most suitable model for accurately representing the detection probability of Red-vented bulbul (Pycnonotus cafer) density Following this, the Half-normal, Hazard-rate, and Negative exponential models have higher AIC values Additionally, the χ2 value is 1.63 with 3 degrees of freedom, resulting in a p-value of 0.67, which is greater than 0.05.

Figure 4.3: Detection probability functions for Sooty-headed Bulbul

Table 4.6: Sooty-headed Bulbul’ parameters

The Hazard-rate function, with the lowest AIC value of 446.61, is the most effective model for representing the detection probability of Sooty-headed Bulbul (Pycnonotus aurigaster) density Following this model are the Uniform, Half-normal, and Negative exponential models, which exhibit higher AIC values Additionally, the χ2 value is 3.59 with 1 degree of freedom and a p-value of 0.36, indicating a good fit for modeling detection probability fluctuations by distance.

Table 4.7: Rufous-backed shrike’ parameters

The Negative Exponential function, with the lowest AIC value of 73.06, is the most effective model for illustrating the detection probability of Rufous-backed shrike (Lanius schach) density Following this, the Uniform, Half-Normal, and Hazard-Rate models exhibit higher AIC values Additionally, the χ2 value is 0.77, with 3 degrees of freedom and a p-value of 0.45 (>0.05), indicating that these parameters are also suitable for modeling the variations in detection probability by distance.

E STIMATING DENSITY OF BIRD SPECIES

The below tables show transects’ density of each data set of each specie

Table 4.8: Density of each specie in each line transect

According to table 4.8, we see that there are big differences between density of transects of all four species That means all four species

Changes in distance can significantly affect detection probability and the likelihood of observation or hearing As distance increases, the number of detected groups tends to decrease, leading to a gradual decline in detection probability However, this trend does not hold true in all cases.

The small body lengths of the four bird species (11-15 cm) and the dense primary forest environment limit visibility, making it challenging for surveyors to spot them from a distance Consequently, the likelihood of observing bird groups decreases with increasing distance, leading to a reliance on their sounds for detection However, due to the quiet nature of their songs, only a few groups are identified through acoustic signals at greater distances.

The population densities of various bird species in the studied area include the Red-whiskered Bulbul at 0.278 birds per hectare, the Red-vented Bulbul at 0.495 birds per hectare, the Sooty-headed Bulbul at 1.59 birds per hectare, and the Rufous-backed Shrike at 0.22 birds per hectare.

The Sooty-headed Bulbul is the most abundant among the four species studied, yet all exhibit uneven distribution, as shown by significant density differences across transects (Table 4.8) For instance, the Red-whiskered Bulbul's density varies by a factor of 14 between its minimum and maximum transects This disparity can be attributed to the location and environmental conditions of the transect lines Transects situated near human habitation and influenced by human activities tend to report fewer bird groups compared to those in more restricted areas Furthermore, transects with lower tree density facilitate easier detection, while those offering favorable living conditions—such as ample food and shelter—result in higher bird detection rates.

The volume of bird sound significantly impacts detection probability, with Red-vented Bulbul songs and calls accounting for 59.6% of detections The Uniform function model indicates that detection probability decreases as distance increases, highlighting that these birds' calls are loud and easily recognizable from afar In contrast, the weaker calls of Red-whiskered Bulbul, Sooty-headed Bulbul, and Rufous-backed Shrike make them difficult to hear, resulting in lower detection probabilities for these species.

Distance sampling surveys offer the flexibility to be conducted year-round, unaffected by visibility variations In comparison to traditional methods, density estimates from distance sampling are typically higher, particularly as the width of the transect decreases To achieve more accurate results, surveyors should focus on smaller areas, although this approach can be time-consuming and costly Consequently, distance sampling is recommended over traditional methods for more effective wildlife density estimation.

Distance sampling is an effective method for estimating bird detection probability and density; however, biases can still arise To minimize these biases and ensure high accuracy in data collection, surveyors must exercise greater care in counting and estimating distances during field investigations.

A total of 954 groups were analyzed, revealing 16 species across 8 bird families As distance increases, the number of detected groups decreases, leading to a decline in detection probability At greater distances, groups are primarily identified through acoustic signals rather than visual observation, resulting in a shift where the proportion of auditory detections rises while visual observations fall.

The detection probability of Red-whiskered Bulbul (Pycnonotus jocosus), Red- vented bulbul (Pycnonotus cafer), Sooty-headed Bulbul (Pycnonotus aurigaster) and

Rufous-backed shrike (Lanius schach) respectively are 0.123, 0.08, 0.09 and 0.47 These values are affected by distance and the call volume of birds

Using the distance sampling method, we identified the best fit models for various bird species The Red-whiskered Bulbul has an estimated density of 0.278 (CI: 0.156-0.31) with a Half-normal model, while the Red-vented Bulbul shows a density of 0.495 (CI: 0.42-0.58) using a Uniform model The Sooty-headed Bulbul, with a density of 1.59 (CI: 1.35-1.89) based on a Hazard-rate model, is the most abundant species among the four Lastly, the Rufous-backed Shrike has an estimated density of 0.22 (CI: 0.125-0.388) with a Negative exponential model.

The study reveals that the four species examined exhibit an uneven distribution across transect lines, attributed to variations in location and natural conditions of each line.

Distance sampling demonstrates greater accuracy than traditional methods due to its higher density estimated results, making it a preferred choice for bird and wildlife surveys Therefore, the implementation of distance sampling in fauna surveys should be increased.

The density survey results of bird species in Tam Dao National Park significantly aid in bird conservation efforts By analyzing bird abundance data, we can track changes in population sizes and evaluate the effects of habitat loss, pollution, and harvesting, as well as determine the viability of isolated populations However, there is a lack of specific studies focused on bird species within the pine forests of Tam Dao National Park Therefore, it is essential to conduct further research in these areas to better understand the diversity and density of both specific and overall avian species.

1 Rachel E McCaffrey, 2005 Using Citizen Science in Urban Bird Studies In Urban habitats: An electronic journal on the biology of urban areas around the world (Center for Urban Restoration Ecology), Arizona, US

2 J M Marzluff, R Bowman, and R Donelly, eds., 2001 Avian Ecology and

Conservation in an Urbanizing World 1st edition, Springer Science and Business Media New York, New York, 585 pages

3 Krizler C Tanalgo; JohnArislynPineda; MaricelAgrvante&AmerolZabide, 2015 Bird

Diversity and Structure in Different Land-use types in Lowland south Central Mindanao, Philippines Tropical life sciences research 26( 2): 85-103

4 Buckland, S.T , 2016 Point transect surveys for songbirds : Robust Methodologies The

5 Nguyen Hai Tuat, Tran Quang Bao ,Vu Tien Thinh, 2011 Ứng dụng một số phương pháp định lượng trong nghiên cứu sinh thái rừng Nhà xuất bản Nông nghiệp, Hà Nội

6 Thomas, L., S.T Buckland, E.A Rexstad, J.L Laake, S Strindberg, S.L Hedley, J.R.B

Bishop & T.A Marques, 2009 Distance software: design and analysis of distance sampling surveys for estimating population size In Journal of Applied

Ecology (Marc Cadotte, Jos Barlow, Nathalie Pettorelli and Philip Stephens and Martin Nuủez), UK, pp 5-14

7 Tordoff, A W et al, 2003 Directory of important bird areas in Vietnam: key sites for conservation BirdLife International in Indochina and the Institute of Ecology and Biological Resources, Hà Nội, 233 pages

8 Nguyen Cu, Le Trong Trai, Karen Phillips, 2000 Bird Vietnam Nhà xuất bản lao động- xã hội, Hà Nội, 250 pages

9 Anderson, D R., K P Burnham, G C White, and D L Otis, 1983 Density estimation of small-mammal populations using a trapping web and distance sampling methods In Ecology Ecological society of America, USA, pp 674-680

10 Buckland S.T, Anderson D.R, Burnham K.P, Laake J.L, Borchers D.L and Thomas

L, 2001: Introduction to Distance Sampling: Estimating Abundance of Biological Populations 1st edition, Oxford University Press, Oxford, UK, 448 pages

11 Craig Robson, 2005 Birds of Southeast Asia Princeton University Press, Princeton,

12 Vo Quy, 1981 Chim Viet Nam: hình thái và phân loại Nhà xuất bản khoa học và kỹ thuật, Hà Nội

13 Vo Quy and Nguyen Cu, 1995 Danh luc chim Viet Nam Nhà xuất bản nông nghiệp,

14 Vo Quy, 1993 A catalogue of the birds of Vietnam Thanh Pho publishing house

15 Vu Tien Thinh, Paul F Doherty and Kathryn P Huyvaert, 2012 Avian conservation value of pine plantation forests in northern Vietnam Bird Conservation International

Table : List of bird species recorded at each line transect at pine forest

Transect Common name Latin name Number

Streak-breasted Scimitar Babbler Pomatorhinus ruficollis 1

Rufous-backed shrike Lanius schach 2

Red-vented bulbul Pycnonotus cafer 6

Sooty-headed Bulbul Pycnonotus aurigaster 24

Puff-throated babbler Pellorneum ruficeps 1

7 Red-vented bulbul Pycnonotus cafer 4

11 Red-vented bulbul Pycnonotus cafer 31

13 Red-whiskered Bulbul Pycnonotus jocosus 1

Định dạng
Số trang	38
Dung lượng	841,64 KB