Bird Assemblage and Diversity along Different Habitat ...

Renz Angelo J. Duco1*, Jay S. Fidelino1, Melizar V. Duya1,2, Mimie M. Ledesma2, Perry S. Ong1,2†, and Mariano Roy M. Duya1,2

1Biodiversity Research Laboratory, Institute of Biology 2Diliman Science Research Foundation

University of the Philippines Diliman, Quezon City 1101 Philippines

Bird Assemblage and Diversity along Different Habitat Types in a Karst Forest Area in Bulacan, Luzon Island, Philippines

Keywords: birds, forest fragments, karst forest, limestone quarry, vertical stratification

Little is known on the effects of degradation and disturbance on bird assemblage in lowland karst forests in the Philippines. In this study, we determined diversity patterns and distribution of birds along the vertical strata in two karst forest fragments adjacent to and one reforestation area within an active limestone quarry area in Bulacan province, Luzon island. Surveys were conducted using mist nets set in the understory (0–3 m) and sub-canopy (4–10 m) from November 2013 to October 2016. A total of 617 individuals belonging to 63 species and 13 feeding guilds were recorded from a mist-netting effort of 654,264.8 mist-net hours (m2∙h); of these, 32 are Philippine endemics and six are threatened species. We recorded the highest number of species in the reforestation area, most of which are generalist and disturbance-tolerant species. Results from permutational multivariate analysis of variance (PERMANOVA) and non-metric multidimensional scaling (NMDS) indicate differences in bird assemblage between the three habitat types and between the two vertical strata within habitat types. Meanwhile, similarity percentage analyses and Mann-Whitney U tests showed that species and guilds that contributed to the observed dissimilarity in the two strata have higher capture rates in the understory. These observations provide insights into the effect of disturbance and habitat alteration in the vertical movement of birds, as evidenced by the shift of some arboreal species to the understory layer in more disturbed habitats. Understanding the behavior and habitat use of birds will, thus, help in identifying appropriate conservation measures to ensure proper resource partitioning among the different bird assemblages in fragmented habitats.

*Corresponding Author: [email protected]†Deceased 02 Mar 2019

INTRODUCTIONKarst forests occur in areas underlain by soluble rocks such as limestone (Ford and Williams 2007). It is one of the twelve forest formations present in the Philippines and covers almost 12% (35,000 km²) of the country’s

total land area (Fernando et al. 2008; BMB-DENR 2019). This type of forest contains unique microhabitats and high environmental heterogeneity, which can support a wide range of flora and fauna – some of which are known to be highly restricted in such habitats (Brown et al. 2010; Tang et al. 2011; Rasalan et al. 2015). Moreover, the presence of unique features such as sinkholes and caves in these

399

Philippine Journal of Science150 (S1): 399-414 Special Issue on BiodiversityISSN 0031 - 7683Date Received: 28 Sep 2020

landscapes provide habitat and protection for many species (Clements et al. 2006).

Unfortunately, karst forests are largely understudied and considered one of the most threatened forest types in the Philippines due to mining and conversion of land to agriculture (Fernando et al. 2008; Tolentino et al. 2020). To date, only about 29% of the total karst forest in the country is protected (Restificar et al. 2006). The unique soil properties and environmental conditions of the karst forest are vulnerable to anthropogenic activities because recovery of its vegetation is very slow and often damages are irreversible (Tuyet 2010). Existing laws do not directly address the protection and conservation of this unique forest type at the landscape level, but focus more on the protection of its cave features and development of the area for ecotourism.

Birds are often used as bioindicator taxa in many studies because they are relatively easy to sample and identify and because they are sensitive to habitat disturbance (Furness 1993). Birds may be especially impacted by modification and destruction of karst forest since the landscape features of this habitat – including caves and limestone crags serve as refugia, breeding, and nesting grounds for many bird species (Lugo et al. 2001; Clements et al. 2006; IUCN 2020). For instance, Aerodramus fuciphagus – which have cultural and economic importance – are known to utilize caves as roosting, nesting, and feeding sites (IUCN 2020). Their nests – made of their saliva – are usually harvested and sold at a very expensive price (USD 2000–3000/kg) or made into a bird’s nest soup, which costs around USD 30 (Thorburn 2015). In addition, Napothera crispifrons is known to be restricted to karst forests in mainland Asia, while Pycnonotus hualon and Phylloscopus calciatilis were recently discovered from karst areas in Southeast Asia (Clements et al. 2006; Woxvold et al. 2009; Alström et al. 2010). In the Philippines, karst forests are important habitats for some highly threatened species of birds. For instance, protected karst forests such as Tabunan, Alcoy, and Nug-as on the island of Cebu have been the few remaining known habitats for the Endangered Kittacincla cebuensis and the Critically Endangered Dicaeum quadricolor (Paguntalan and Jakosalem 2008; BLI 2020; IUCN 2020).

In this study, we assessed patterns of diversity in the bird community (species and guild) of three highly modified habitat types present in an active limestone quarry area in Bulacan Province, Luzon Island. We asked three main questions:

1. How do bird species richness, diversity, abundance, and composition differ between two forest fragments and a reforestation area in a karst landscape on central Luzon?

2. How do bird species richness, diversity, abundance, and composition differ between the understory and sub-canopy in these habitats?

3. How do bird feeding guild composition and abundance differ between these habitats?

The ecological services birds provide such as seed dispersal, pollination, and pest control make it important to understand how they use their habitat and their response to anthropogenic disturbances (Sekercioglu 2012). Furthermore, understanding these patterns can aid in the enhancement of forest restoration efforts and protection measures of mining companies operating in karst areas, as well as inform ecotourism management decisions for these unique areas.

MATERIALS AND METHODS

Study SitesThe study was carried out in a karst forest area in Bulacan Province, Luzon island with some sites located within an active quarry area operated by a private company (Figure 1). Original forest formation in our sites is forest over limestone or karst forest (Ong et al. 2016). Three different habitat types were identified based on the degree of disturbance for each site:

Forest fragment 1 (FF1). It is located near the edge of the contiguous karst forest at the foothills of the Southern Sierra Madre Mountain Range in Barangay Bayabas, Doña Remedios Trinidad, Bulacan (14.965° N, 121.100° E). We have established three sampling areas inside a ~ 160 ha forest fragment around a 1.2 km long Puning Cave (160 masl). Puning Cave is about 15-km south of Biak-na-Bato Natural Park, a protected karst area (Restificar et al. 2006), and about three kilometers away from the active quarry site in the mining area (Figure 1). Plants from families Malvaceae and Fabaceae are the most dominant and species-rich in the study sites. This includes species typical of karst forests such as Pterocymbium tinctorium (Malvaceae), Archidendron scutiferum (Fabaceae), and Kleinhovia hospita (Malvaceae), with canopy trees at least 10 m tall (DBH: 15–90 cm). Small-scale slash-and-burn (“kaingin”) and local limestone mining activity are present near the cave area. The site is also a local tourist destination because of the vast cave system and a birdwatching site.

Forest fragment 2 (FF2). One sampling area was established inside a 4.6-ha forest fragment (198 masl) (14.934° N, 121.092° E). The site is located in Sitio Turongisan, Barangay Bayabas, Doña Remedios Trinidad, Bulacan and adjacent (about 500 m north) to an active

Duco et al.: Diversity of Birds in Karst Forest AreaPhilippine Journal of ScienceVol. 150 No. S1, Special Issue on Biodiversity

400

quarry area. The area is characterized by rugged terrain with visible sharp-pointed limestone outcrops. Several small caves are also present within and near the forest patch. Dominant plant families in the site include Fabaceae, Malvaceae, and Moraceae. Canopy trees of at least 10 m tall (DBH: 20 to 136 cm) are present, with Pterocymbium tinctorium (“taluto”) and Garuga floribunda (“bogo”), tree species associated with karst forests dominating the upper canopy structure. The

understory layer is mainly formed by dense clumps of Schizostachyum lumampao (Poaceae), along with herbs and shrubs such as Chromolaena odorata and Desmodium gangeticum.

Reforestation area (RA). It is composed of two sampling areas in a mining concession which covers less than 1 ha each (50 masl) (14.899° N, 121.081° E). The sites are located along a roadside and near a blasting site

Figure 1. Location of the sampling sites for each habitat type within and near a limestone quarry area in Bulacan, central Luzon island, Philippines. Land cover data from the European Space Agency Climate Change Initiative (2017). Note that the mosaic nature of the landscape likely led to inconsistency between satellite land cover classification and on-ground vegetation assessment.

Philippine Journal of ScienceVol. 150 No. S1, Special Issue on Biodiversity

Duco et al.: Diversity of Birds in Karst Forest Area

401

for quarry activities. The reforestation activities started between 1996–2000, with the monoculture of exotic tree species such as Gmelina arborea and Mangifera indica. G. arborea trees ranged in height from 5–20 m, with DBH from 10–54 cm; M. indica trees ranged in height from 5–20 m, with DBH from 12–111 cm. Native species such as Pterocarpus indicus were later inserted and through time, pioneer species such as Melanolepis multiglandulosa and Macaranga tanarius have established in the area. However, none of these have reached more than 10 m in height.

Bird SamplingMist netting was done from 2013–2016, covering both wet and dry seasons. Birds were captured using mist nets (12 x 2.6 m with 36 mm mesh); 20 mist nets were set near the ground level while five nets were placed at sub-canopy (4–10 m) to sample both strata. In the sub-canopy, two sets of three and two mist nets were vertically placed above each other, then raised to the sub-canopy using a pulley system described by Ingle (1993). The nets were opened for 12 h (6:00 AM–6:00 PM) for three consecutive days for each sampling site and were checked every 30 m to retrieve any captures.

All captures were identified following Kennedy et al. (2000). The birds were then measured and tagged with uniquely numbered aluminum bird rings prior to release. Furthermore, birds were classified into feeding guilds based on their dietary requirements (insectivore, frugivore, nectarivore, carnivore, or a combination of these) and their foraging layer (terrestrial, understory, arboreal). Published literature works (Gonzalez 1995; Posa and Sodhi 2006) and web sources such as BirdLife International (2020) and IUCN (2020) were used in categorizing the birds to their feeding guilds. Birds were classified into 13 feeding guilds.

Field sampling was covered by Wildlife Gratuitous Permit Nos. III-2013-06 and III-2016-01. We conducted bird capture and handling in accordance with guidelines to the use of wild birds in research published by the Ornithological Council (Fair et al. 2010).

Data AnalysisWe compared the bird assemblage between habitat types and between vertical strata within each habitat (understory and sub-canopy) in terms of species richness, diversity, capture rates, and composition. To compare species richness between habitat types and strata, we calculated the estimated species richness as the mean of four non-parametric species richness estimators (Chao, Jack1, Jack2, and Bootstrap) using the Community Ecology package (“vegan” v. 2.4-1) (Oksanen et al. 2020) in the R statistical software (v. 3.5.0) (R Core Team 2015).

Sampling completeness was then assessed by comparing the actual number of species sampled to the total estimated species richness, with greater than 90% considered sufficient (Mohd-Azlan et al. 2019). We also generated species accumulation curves with a 95% confidence interval based on expected species richness using the same package in R.

In this study, Shannon’s species diversity index (H’) was used as a measure of diversity, also calculated using the vegan package in R. Values were compared between habitat types and between strata using Kruskal-Wallis and Mann-Whitney U tests, respectively (α = 0.05). These tests were performed using IBM SPSS Statistics for Windows v 20.0.

As the total sampling effort varied between habitats, capture rates were expressed as the number of individuals captured per mist net hour (m2∙h) of sampling effort. Kruskal-Wallis tests and Mann-Whitney U tests (α = 0.05) were also used to compare overall capture rates, as well as species and guild capture rates, between habitat types and strata.

We compared differences in species and guild composition between each habitat type and strata through PERMANOVA analysis using the “adonis” function of the “vegan” package in R (α = 0.05). We also performed two-dimensional NMDS based on Bray Curtis similarity matrices, with data log+1 transformed to reduce the contribution of the most abundant species. Moreover, Similarity Percentage (SIMPER) was also used to determine species or guilds that have contributed most to the dissimilarities observed. Both NMDS and SIMPER were performed using PRIMER (v 6.4.7) (Clarke and Gorley 2006).

RESULTS

Bird CommunityAfter 654,264 mist-net hours (m2∙h), we recorded a total of 617 bird individuals belonging to 63 species and 13 feeding guilds (Table 2). Species richness was highest in the RA with 41 species, while 30 species and 29 species were recorded in FF1 and FF2, respectively. A high proportion of bird species were recorded in only one of the three habitat types (Table 2). Twenty-three (23) species were caught exclusively in the RA, nine species exclusively in FF1, and six species exclusively in FF2. Only 12 species were recorded across all three habitats. Most of the species captured were insectivores – understory insectivores (UIs) (12 species), terrestrial insectivores (nine species), and arboreal insectivores (AIs) (eight species).


402

Of the 63 species recorded, 32 are Philippine endemics – four of which are endemic to Luzon island (Table 2). The number of endemics was highest in FF1 (21 species). We also recorded one species categorized as Vulnerable in the IUCN Red List (2020) and six species listed under different threatened categories in the Philippine Red List of Threatened Wild Fauna (BCSP 2020).

Species Richness, Diversity, and AbundanceSpecies richness was highest in the RA, although sampling effort was different for each habitat type (Table 1). Moreover, species accumulation curves for the three habitat types and both vertical strata did not reach an asymptote, and it is likely that additional species will be captured with additional sampling effort (Figure 2). This is also supported by the calculated sampling completeness, which ranged from 50–70% (Table 1).

Mean estimated species richness was 70 for RA, 42 for FF1, and 38 for FF2 (Table 1), although Kruskal-Wallis test found no significant difference in the estimated species richness between habitat types (H = 2.747, p = 0.253). Shannon index values and mean capture rates also did not differ significantly between habitat types (H = 1.965, p = 0.374; H = 3.741, p = 0.152, respectively).

Species Assemblage and CompositionBird species assemblages differed significantly between the three habitat types based on PERMANOVA (F = 3.50, p = 0.001). Pairwise comparisons revealed significant differences in assemblage between FF2 and the RA (F = 2.52, p = 0.002), and between FF1 and the RA (F = 3.78, p = 0.001), but not between the two forest fragments (F = 1.52, p = 0.126). These results are also evident in the

Species Guilda Residence statusb

Endemicity IUCN (2020)c

BCSP (2020)c

FF1 FF2 RA

Acrocephalus stentoreus FGI R LC 1 0 0Actenoides lindsayi UI E Philippines LC 14 8 0Aerodramus amelis AI E Philippines LC 0 11 2Alcedo atthis UC M LC 0 0 2Anthus gustavi TI M LC 0 0 1Anthus rufulus TI R LC 0 0 2Bolbopsittacus lunulatus AF E Philippines LC 3 0 0Cacomantis sepulcralis UI R LC 0 2 0Calliope calliope TI M LC 0 1 0Caprimulgus manillensis AI E Philippines LC 2 1 6Ceyx cyanopectus UIC E Philippines LC CR 0 0 3Chalcophaps indica TF R LC 10 1 10Cinnyris jugularis UIN R LC 1 1 0Collocalia marginata AI E Philippines LC 8 6 0Collocalia troglodytes AI E Philippines LC 7 2 3Copsychus luzoniensis TI E Philippines LC VU 10 11 0Copsychus mindanensis TI E Philippines LC 1 2 6Dasylophus cumingi AFI E Luzon (greater) LC 0 0 1Dasylophus superciliosus AFI E Luzon (greater) LC 0 2 0

Table 2. Species list of birds recorded including their assigned feeding guilds, residence status, endemicity, conservation status based on IUCN (2020) and BCSP (2020), and the number of individuals mist-netted in each habitat type.

Table 1. Observed and mean estimated species richness (± SE), total abundance, mean capture rates (± SE), Shannon diversity index (± SE), total sampling effort, and sampling completeness based on mist-netting of birds in three habitat types in a limestone quarry area on central Luzon island, Philippines.

FF1 FF2 RA Total

Number of Philippine endemics 21 17 19 32

Total observed species richness 30 29 41 63

Mean estimated species richness 41.82 ± 6.65 38.03 ± 6.16 70.28 ± 9.26 96.98 ± 15.0

Total number of individuals 150 150 317 617

Mean capture rate (capture/m2∙h) 0.0009 ± 0.008 0.0017 ± 0.000 0.0008 ± 0.006 0.0015 ± 0.006

Shannon-Wiener index (H') 1.22 ± 0.17 1.48 ± 0.28 1.05 ± 0.17 1.21 ± 0.11

Total sampling effort (m2∙h) 243,360 136,843 274,061 654,264

Sampling completeness (%) 71.73 76.25 58.33 64.96



403

Species Guilda Residence statusb

Endemicity IUCN (2020)c

BCSP (2020)c

FF1 FF2 RA

Dicaeum australe AF E Philippines LC 0 0 2Dicrurus balicassius AI E Philippines LC 4 0 1Erythropitta erythrogaster TI NE Philippines LC 3 0 0Ficedula narcissina UI M LC 0 3 7Gallirallus torquatus TFI R LC 0 0 1Geokichla cinerea TI E Philippines VU VU 2 1 0Geopelia striata TF R LC 0 0 12Halcyon smyrnensis UI R LC 3 1 4Hierococcyx pectoralis UI E Philippines LC 1 1 1Hypothymis azurea UI R LC 1 4 0Hypsipetes philippinus AFI E Philippines LC 36 22 39Lalage nigra AI R LC 0 0 2Lanius cristatus UI M LC 13 20 57Locustella lanceolata TFI M LC 0 0 1Lonchura atricapilla TF R LC 0 0 2Lonchura leucogastra TF R LC 0 0 2Lonchura punctulata TF R LC 0 0 1Loriculus philippensis AF E Philippines LC CR 1 0 0Macropygia tenuirostris AF NE Philippines LC 0 0 1Merops philippinus AI R LC 0 0 1Merops viridis AI R LC 0 11 4Monticola solitarius TFI R,M LC 0 0 1Ninox philippensis UC E Philippines LC 3 2 0Oriolus chinensis AFI R 4 0 0Orthotomus chloronotus UI E Luzon LC 0 2 0Otus megalotis UC E Luzon (Greater) LC 6 2 5Passer montanus TIG I LC 0 0 1Pericrocotus divaricatus FGI M LC 0 0 1Periparus elegans UI E Philippines LC 3 10 1Phapitreron amethystinus AF E Philippines LC CR 1 0 0Phapitreron leucotis AF E Philippines LC 0 0 3Phylloscopus borealis UI M Philippines LC 0 1 0Phylloscopus cebuensis UI E LC 0 0 1Phylloscopus sp. UI 0 1 0Pitta sordida TI R LC 1 0 0Pycnonotus goiavier AFI R LC 4 9 106Pycnonotus urostictus AFI E Philippines LC 0 8 1Sarcops calvus AFI NE Philippines LC 4 0 1Saxicola caprata TI R LC 0 0 1Terpsiphone cinnamomea SAL NE Philippines LC 1 0 0Todiramphus chloris UIC R LC 0 0 19Treron axillaris AF E Philippines LC VU 1 0 0Turnix ocellatus TFI E Philippines LC 1 4 1Zosterops meyeni AFI NE Philippines LC 0 0 1aAF – arboreal frugivore, AFI – arboreal frugivore-insectivore, AI – arboreal insectivore, FGI – frugivore-granivore-insectivore, SAL – sallying insectivore, TF – terrestrial frugivore, TFI – terrestrial frugivore-insectivore, TI – terrestrial insectivore, TIG – terrestrial insectivore-granivore, UC – understory carnivore, UI – understory insectivore, UIC – understory insectivore-carnivore, UIN – understory insectivore-nectarivorebE – endemic, I – introduced, M – migrant, N – near-endemic, R – resident, R,M – resident, migrantcCR – Critically Endangered, DD – Data deficient, LC – Least Concern; VU – Vulnerable

generated NMDS plot showing clustering between FF1 and FF2 samples, separated from the RA (Figure 5).

SIMPER analysis revealed species that contributed most to the difference in bird assemblage between the three habitat types. For instance, Pycnonotus goiavier contributed most to the dissimilarity observed between the habitat types, being more abundant in the RA (Figure 6). Moreover, Geopelia striata and Todiramphus chloris were caught exclusively in the RA, adding to the observed dissimilarity in bird assemblages between habitat types.

Meanwhile, the endemic birds Actenoides lindsayi and Pycnonotus urostictus also contributed to the differences in bird assemblage observed, being more abundant in FF1 and FF2, respectively.

Guild StructureThirteen (13) feeding guilds were identified during the study. However, four guilds represented by only one representative species and/or one or two individuals were excluded from the analysis to avoid skewing the results


404

due to these low sample sizes. The UI guild had the most representative species (12 species), while arboreal frugivore-insectivores (AFI) comprised 39% (n = 238) of the total number of individuals captured.

Guild assemblage was found to be significantly different between habitat types (F = 3.71, p = 0.001). Species richness of three guilds – AI, TF (terrestrial frugivore), and UIC (understory insectivore-carnivore) – showed a significant difference between habitat types (Figure 3): AI was more species-rich in FF2 while TF and UIC are significantly more species-rich in the RA. These three guilds also showed a significant difference in their capture rates between habitat types. Pairwise analyses showed that AI was significantly more abundant in FF2 than

either FF1 or RA while TF was more abundant in the RA than in FF2 (Figure 4). UIC, meanwhile, was exclusively captured in the RA.

Vertical StratificationPERMANOVA showed a significant difference in bird assemblage between vertical strata within each habitat type (FF1: F = 2.70, p = 0.006; FF2: F = 2.66, p = 0.017; RA: F = 2.90, p = 0.001). These results are also evident from the NMDS ordinations, showing a clear separation between understory and sub-canopy assemblages (Figure 7).

Additionally, more species were captured and Shannon index showed species diversity to be higher in the understory than in the sub-canopy for all habitat types

Figure 2. Species accumulation curves based on mist-netting of birds in a) three habitat types and b) two forest strata in a limestone quarry area on central Luzon island, Philippines. Dashed lines indicate 95% confidence intervals.



405

(FF1: U = 6.50, p = 0.001; FF2: U = 0.00, p = 0.008; RA: U = 9.00, p = 0.001). Mann-Whitney U-tests showed that mean capture rate was higher in the understory than the sub-canopy in the RA (U = 4.00, p = 0.001). On the other hand, capture rates between the two strata did not differ significantly between the two forest fragments (FF1: U = 42.00, p = 0.920; FF2: U = 10.00, p = 0.691) (Figure 8).

DISCUSSIONOur mist-netting record of 63 bird species is 9% of the total number of bird species recorded in the Philippines (WBCP 2019). Moreover, we recorded 32 Philippine endemic species, which accounts for about 13% of the total number of endemic birds in the country (n = 241). Of these, four are Luzon endemics – about 33% of the total number (n = 12)

Figure 3. Box-and-whiskers plots of differences in mean species richness of feeding guilds based on mist-netting of birds between two forest fragments and a reforestation area in a limestone quarry area on central Luzon island, Philippines. Asterisks denote significant pairwise differences. Filled points are outliers. FF1 – forest fragment 1, FF2 – forest fragment 2, RA – reforestation area; AF – arboreal frugivore, AFI – arboreal frugivore-insectivore, AI – arboreal insectivore, TF – terrestrial frugivore, TFI – terrestrial frugivore-insectivore, TI – terrestrial insectivore, UC – understory carnivore, UI – understory insectivore, UIC – understory insectivore-carnivore.


406

Figure 4. Box-and-whiskers plots of differences in mean capture rates of feeding guilds based on mist-netting of birds between two forest fragments and a reforestation area in a limestone quarry area on central Luzon island, Philippines. Asterisks denote significant pairwise differences. Filled points are outliers. FF1 – forest fragment 1, FF2 – forest fragment 2, RA – reforestation area; AF – arboreal frugivore, AFI – arboreal frugivore-insectivore, AI – arboreal insectivore, TF – terrestrial frugivore, TFI – terrestrial frugivore-insectivore, TI – terrestrial insectivore, UC – understory carnivore, UI – understory insectivore, UIC – understory insectivore-carnivore.

of endemic bird species in Greater Luzon. We recorded the highest number of endemic species in FF1, six of which were not recorded in the other habitat types surveyed.

Our species accumulation curves did not reach an asymptote, indicating that additional sampling is needed

to fully represent the bird community in the sites. This is supported by our species records from transects conducted concurrently with our mist net sampling. From a combined total of 134 species recorded from transects and mist-netting, 70 were sampled exclusively in transects. These



407

included several species that are more active in the higher canopy and open-air, such as raptors and aerial foraging birds. These species are difficult to sample using mist nets, especially since our sampling of the canopy layer was limited to just five nets per site. Our mist nets also likely undersampled owls, egrets, and sunbirds – several species of which have also been recorded in our transects.

Figure 5. NMDS ordination plots showing bird assemblage based on mist netting of birds between two forest fragments and a reforestation area in a limestone quarry area on central Luzon island, Philippines.

Figure 6. Mean capture rate (individuals/m2h) of species contributing to differences in assemblage between habitat types. Error bars represent 95% confidence interval while asterisks indicate significant differences: *p < 0.05, **p < 0.01, ***p < 0.001.

Ground birds such as waterhens and buttonquails were also observed in transects but not captured in mist nets, as these species are known to be elusive and are usually able to avoid mist nets. Moreover, the lower shelf of our mist nets does not touch the ground and is usually elevated when set up and thus are not ideal for capturing ground birds (MacArthur and MacArthur 1974). Additional


408

Figure 7. NMDS ordination plots of the understory and sub-canopy bird captures based on mist-netting of birds between two forest fragments and a reforestation area in a limestone quarry area on central Luzon island, Philippines.

Figure 8. Comparison of estimated species richness, mean capture rate (individuals/ m2∙h), and diversity (Shannon index) between understory and sub-canopy captures based on mist-netting of birds between two forest fragments and a reforestation area in a limestone quarry area on central Luzon island, Philippines. Error bars indicate a 95% confidence interval.



409

sampling effort may result in the sampling of these species. Such undersampling may also be attributed to the inherent limitations of mist netting as a bird sampling technique. These include net avoidance after increased netting in the same site and comparatively low efficiency in terms of detection rate per unit effort, among others (Ralph et al. 2004). Thus, the combination of mist netting and transects will likely lead to more complete documentation of the bird community in any particular area. Nonetheless, evaluation studies have shown that bird abundances derived from mist-netting showed good correspondence with independently derived data, such as point counts and transects at the same locations (Dunn and Ralph 2004).

Bird Assemblage and Distribution across Habitat TypesBird assemblages differed significantly between habitat types, which may be attributed to differences in their floristic composition, vegetation structure, and levels of disturbance in these habitats. A recent study done in our sites by Galindon et al. (2018) found a significant difference in the tree and herbaceous species composition between forest fragments and exotic species-based reforested areas. Based on our results, our two forest fragment sites did not vary in terms of bird species composition, sharing most of the species present in these two habitat types. Their similar floristic composition and vegetation structure may have resulted in a higher resemblance in bird assemblage. These similarities include the presence of at least 10-m high canopy trees, emergent layer dominated by families Malvaceae and Euphorbiaceae, understory formed mostly by Schizostachyum lumampao (Poaceae), and presence of climbing bamboos.

The differences in resource availability and environmental conditions between forest fragments and RAs may have influenced the composition of the bird community that could thrive in them (Azman et al. 2011). For example, P. goiavier was found to be more abundant in RAs because this species has a high tolerance for disturbance and occupies a wide range of habitats, but is rarely seen in good quality forests (Kennedy et al. 2000). In addition, G. striata and T. chloris – both of which are known to be disturbance-tolerant and common in urban areas – were caught exclusively in the RA. Meanwhile, endemic species such as P. urostictus and Aerodramus amelis characterized FF2, although the abundance of the latter may be due to proximity to caves in the area, which serves as their nesting sites. Moreover, Copsychus luzoniensis and Actenoides lindsayi were abundant in both FF1 and FF2 but never captured in the RA. These species are more commonly found in the understory and edge of lowland forests and are more sensitive to anthropogenic disturbance (Kennedy et al. 2000; Posa and Sodhi 2006). The distribution of these species within the study area

provides additional support for the degree of tolerance and habitat preference of some of the abundant species occurring in karst forests in Luzon.

Bird Assemblage and Distribution across Vertical StrataWithin each habitat type, we observed a significant difference between the two vertical strata in terms of species richness, diversity, capture rate, and species composition. Our results revealed that the understory had more species and higher capture rates and diversity index values compared to the sub-canopy.

We found that some arboreal species (mostly bulbuls) exhibit a downward shift from the higher forest strata to the understory in both RA and FF2. For instance, P. goiavier and Hypsipetes philippinus, which contributed 38% to the observed dissimilarity between strata in the RA, have significantly higher capture rates in the understory. In FF2, P. goiavier and P. urostictus also have significantly higher capture rates in the lower strata. In FF1 meanwhile, our species-level analysis showed that birds were generally found in the strata we expected them to prefer based on their designated feeding guilds.

The vertical shift towards the understory we observed in these bird species may be related to the availability of resources and complexity of the habitat. In the RA, for example, the understory is mostly dominated by the invasive shrub Lantana camara – the fruits of which are commonly consumed by generalist frugivores such as bulbuls, flowerpeckers, babblers (Ramaswami et al. 2017). The abundance of this shrub may have drawn these arboreal species to the lower stratum, along with the availability of other understory plants such as Oplismenus compositus, Piper sp., and grasses (Poaceae) – which birds may utilize as food or materials to build a nest. Moreover, the simplification in the vertical vegetation structure of the RA caused by the planting of exotic tree species may have favored the increase of specific trophic guilds and generalist species. A study by Munro et al. (2011) has shown that reforested areas maintained to have complex vegetation structures could still harbor high numbers of forest-dependent species and support other trophic guilds. Thus, management practices that aim to increase habitat and vegetation structure variability by planting native understory and canopy plant species should be considered when implementing reforestation plans in an area. Such practices may facilitate recolonization by birds, as well as other native wildlife such as fruit bats, amphibians, and reptiles.

Similarly, the reduced canopy cover and denser understory vegetation in FF2 may have apparent edge effects on some of the bird species. The modification of the habitat may


410

have influenced the movement of some species to shift downwards during midday to avoid the heat while the dense foliage in the understory layer could have provided the substrate for some insectivore guilds to glean their prey and use to perch for aerial sallying (Pearson 1971; Wunderle et al. 2005).

No shifts in stratification were observed in FF1. This may be due to the relatively lower degree of disturbance experienced by the site, owing to local policies to protect the cave and its vicinity for ecotourism activities. Moreover, large trees – which are required by some avian groups such as parrots and large-bodied frugivores for foraging, perching, and nesting – are relatively more abundant in the area.

Feeding Guild StructureBird assemblage and diversity patterns among the three habitat types were further assessed by considering the dietary niche of the identified bird species. Frugivores and insectivores, considered to be the two major dietary groups in birds, are well-represented in all three habitat types in our study areas and may provide essential ecological services to these degraded habitats (Sekercioglu 2012). Insectivores (AI, TI, UI) were the dominant guild within each habitat type and in our study area, with 29 insectivores out of the 63 species sampled. Although insectivores are commonly observed to dominate tropical forests in terms of species richness and abundance, they are very sensitive to habitat modification (Sekercioglu et al. 2002). This is especially the case for those insectivores situated in the understory, which usually become the first groups to disappear when a disturbance occurs within their habitat.

UI and TI were least represented in the RA (10 species) and made up mostly of migratory (Lanius cristatus, Ficedula narcissina) and generalist species (Copsychus mindanensis), whereas resident Philippine endemic forest-dependent species such as Actenoides lindsayi and Copsychus luzoniensis represent the insectivores in the FF1 (11 species) and FF2 (15 species).

More frugivorous birds were recorded in the RA (eight species) than either FF1 (five species) or FF2 (one species). The diversity and distribution of frugivorous guilds could be largely affected by both abundance and richness of their food and the vegetation structure of habitat types (Moegenburg and Levey 2003). For instance, most of the frugivorous birds encountered in the RA are disturbance-tolerant species (G. striata, Lonchura atricapilla, Lonchura leucogastra, and Lonchura punctulata) that commonly feed near the ground on grains from plant sources such as rice and grasses. Meanwhile, all the frugivore species present in FF1 are forest-dependent – four of which (Bolbopsittacus lunulatus, Loriculus philippensis, Phapitreron amethystinus, and

Treron axillaris) are Philippine endemic arboreal feeders. The availability of fruits from tall trees may explain the abundance of these arboreal frugivores (AFs) in FF1. On the other hand, frugivore guilds were represented in FF2 by just a single species, Chalcophaps indica. This is surprising, as we expected the same AFs captured in FF1 to also be present in FF2. Indeed, we observed the same species in FF2 from our transects on the site. It is possible that in FF2, these species forage in even higher strata than the sub-canopy covered by our netting and, thus, may have eluded capture. Hunting pressure has also been observed to cause the decline of frugivores in forests on Luzon island (Posa and Sodhi 2006; Española et al. 2013). However, it is unlikely that hunting had a significant influence in the reduced number of frugivores in FF2 as the area was managed by a mining company, and security personnel is present at the site. Thus, we saw that – in general – TFs were more abundant in the RA while AFs were more abundant in the forest fragments, likely due to differences in the kinds of plant resources they consume (i.e. seeds and grains vs. fruits from trees).

We identified 18 species belonging to several omnivore guilds (AFI, FGI, TFI, TIG, UIC, UIN). Omnivores are particularly dominant in the RA with 14 species recorded while seven and six species each were recorded for FF1 and FF2, respectively. The abundance of generalist species with wide dietary breadths in more disturbed habitats is a common trend in fragmented landscapes of tropical forests (Devictor et al. 2008). These omnivorous guilds are dominated by bulbuls (Pycnonotidae), which are habitat and dietary generalists, and are known for their importance in forest restoration in lowland forests (Corlett 2017). We also recorded seven Philippine endemic species that belong to these omnivorous guilds in the reforestation sites. The presence of these groups in the area underlines the value of reforestation in mined-out areas, particularly for endemic species. The RAs in the mining site were first established in 1996 and 2000 and were initially planted with Gmelina arborea, a fast-growing exotic tree species (Galindon et al. 2018). Native species were inserted later and other pioneer and naturally occurring species re-established through time. Studies suggest that bird species richness and abundance in areas dominated by exotic plant species are significantly lower compared with sites with native vegetation (Braithwaite et al. 1989; Fleishman et al. 2003; Isacch et al. 2005; Flanders et al. 2006). However, a study by Thompson and Thompson (2004) showed that bird composition changes with the reforestation age and biodiversity recovery in the RAs planted with exotic species is still possible by re-establishing the floral component using native species. Thus, it is also important to monitor reforestation sites’ recovery and evaluate the re-colonization of the original wildlife and native plant species in the area.



411

CONSERVATION IMPLICATIONSAlthough we recorded the highest number of species in the RA, the simplification in the vertical vegetation structure and high level of disturbance in these exotic species-based reforestation sites resulted in the dominance of generalist and disturbance-tolerant species of lower conservation value. Thus, priority should be given to protect existing forests and increase the habitat and vertical complexity of these sites to attract forest-dependent and sensitive bird species of high conservation priority. Planting of native and indigenous tree species are also recommended for reforestation projects since studies suggest that reforestation sites composed of native trees and maintained to have complex vegetation structure facilitates recolonization of original wildlife in the area (Thompson and Thompson 2004; Munro et al. 2011). RAs should also be established near remaining forest fragments to increase habitat availability and maintain habitat diversity in fragmented landscapes. Green belts or forest corridors should also be established to maintain landscape connectivity between fragmented habitats and reduce the impacts of fragmentation in wildlife (Mullu 2016). This will allow greater mobility of sensitive species, such as UIs, between fragments. Furthermore, long-term programs on reforestation maintenance and protection are also needed to allow natural forest regeneration.

The abundance of endemic and forest-dependent birds, as well as the presence of threatened species in FF1 highlight the conservation value of karst habitats. In the Philippines, existing policies and laws do not directly address the protection of karst areas but instead focused mainly on the protection of caves and cave features to boost tourism (Tolentino et al. 2020). However, the high level of endemism and presence of threatened species in karst habitats underlines the importance of protecting the forests surrounding the caves for these species to continually thrive in the area. Local management policies should also consider reducing further development near caves and forested areas, as well as the strict enforcement of slash-and-burn activities for agricultural purposes, to maintain the diversity of birds on the site.

ACKNOWLEDGMENTSWe acknowledge the support given by Holcim Philippines Inc. and Holcim Mining Development Corporation – member and associate member, respectively – of the LafargeHolcim Group to conduct this research in their area of operation in Bulacan. We also wish to thank the local government units in each study site for providing logistical assistance and local guides. We acknowledge the following for their assistance in data gathering and helping

us throughout the duration of fieldwork: Willardo Reyes, Primitivo Aznar III, Ricardo Buenviaje, Joel Sarmiento, Jameson Reynon, Nonito Antoque, Rey Antoque, and Jerry Cantil. We thank the Department of Environment and Natural Resources and their respective provincial and regional offices where the study was conducted for granting the necessary permits to conduct the research. We are thankful to the Diliman Science Research Foundation for managing the grant.

REFERENCESALSTRÖM P, DAVIDSON P, DUCKWORTH JW,

EAMES JC, LE TT, NGUYEN C, TIMMINS R. 2010. Description of a new species of Phylloscopus Warbler from Vietnam and Laos. Ibis 152(1): 145–168.

AZMAN NM, LATIP NSA, SAH SAM, AKIL MAMM, SHAFIE NJ, KHAIRUDDIN NL. 2011. Avian diver-sity and feeding guilds in a secondary forest, an oil palm plantation and a paddy field in riparian areas of the Kerian River Basin, Perak, Malaysia. Tropical Life Sciences Research 22(2): 45–64.

[BCSP] Biodiversity Conservation Society of the Philip-pines. 2020. Philippine Red List of Threatened Wild Fauna. Department of Environment and Natural Resources Biodiversity Management Bureau (DENR-BMB), Quezon City, Philippines. 120p.

[BLI] BirdLife International. 2020. Important Bird Areas factsheet: Nug-as and Mount Lantoy. Retrieved on 09 Jun 2020 from http://www.birdlife.org

[BMB-DENR] Biodiversity Management Bureau–Depart-ment of Environment and Natural Resources. 2019. Protection and Conservation Strategy and Action Plan. Retrieved from http://www.bmb.gov.ph/bmb/CAWED/Proposed_Policy/CMPCSAP_2019-2028.pdf?cv=1

BRAITHWAITE RW, LONSDALE WM, ESTBERGS JA. 1989. Alien vegetation and native biota in tropi-cal Australia: the impact of Mimosa pigra. Biological Conservation 48: 189–210.

BROWN RM, DE LAYOLA LA, LORENZO A, DIES-MOS ML, DIESMOS AC. 2015. A new species of limestone karst inhabiting forest frog, genus Platyman-tis (Amphibia: Anura: Ceratobatrachidae: subgenus Lupacolus) from southern Luzon Island, Philippines. Zootaxa 4048 (2): 191–210.

CLARKE KR, GORLEY RN. 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth.

CLEMENTS R, SODHI NS, SCHILTHUIZEN M, NG PKL. 2006. Limestone karsts of southeast Asia: im-


412

periled arks of biodiversity. Bioscience 56: 733–742.

CORLETT RT. 2017. Frugivory and seed dispersal by vertebrates in tropical and subtropical Asia: an update. Global Ecology and Conservation 17: 1–22.

DEVICTOR V, JULLIARD R, JIGUET F. 2008. Distribu-tion of Specialist and Generalist Species along Spatial Gradients of Habitat Disturbance and Fragmentation. Oikos 117(4): 507–514.

DUNN EH, RALPH CJ. 2004. Use of mist nets as a tool for bird population monitoring. Stud Avian Biol 29: 1–6.

[ESA] European Space Agency. 2017. Land Cover CCI Product User Guide Version 2. Tech. Rep. Retrieved from maps.elie.ucl.ac.be/CCI/viewer/download/ESACCI-LC-Ph2-PUGv2_2.0.pdf

ESPAÑOLA CP, COLLAR NJ, MARSDEN SJ. 2013. Are populations of large‐bodied avian frugivores on Luzon, Philippines, facing imminent collapse? Animal Conservation 16: 467–479.

FAIR JM, PAUL E, JONES J ed. 2010. Guidelines to the use of wild birds in research, 3rd ed. Washington DC: Ornithological Council. 216p.

FERNANDO ES, SUH MH, LEE J, LEE DK. 2008. Forest Formations of the Philippines. ASEAN Korea Environmental Cooperation Unit (AKECU), National Seoul University, Seoul, Korea. 232p.

FLANDERS AA, KUVLESKY WP, RUTHVEN DC, ZAIGLIN RE, BINGHAM RL, FULBRIGHT TE, HERNANDEZ F, BRENNAN LA. 2006. Effects of invasive exotic grasses on South Texas rangeland breeding birds. The Auk 123: 171–182.

FLEISHMAN E, MCDONAL N, MAC NALLY R, MUR-PHY DD, WALTERS J, FLOYD T. 2003. Effects of floristics, physiognomy and non-native vegetation on riparian bird communities in a Mojave Desert water-shed. Journal of Animal Ecology 72: 484–490.

FORD DC, WILLIAMS PW. 2007. Karst Hydrogeology and Geomorphology. John Wiley and Sons Ltd. 562p.

FURNESS RW, GREENWOOD JJD, JARVIS PJ. 1993. Can birds be used to monitor the environment? In: Birds as Monitors of Environmental Change. Furness RW, Greenwood JJD eds. Chapman and Hall, London. p. 1–41.

GALINDON JM, PASION B, TONGCO MD, FIDELINO J, DUYA MR, ONG P. 2018. Plant diversity patterns in remnant forests and exotic tree species-based refor-estation in active limestones quarries in the Luzon and Mindanao biogeographic sub-regions in the Philip-pines. Ecological Research 33: 63–72.

GONZALEZ JC. 1995. Status of Birds at the U.P. La-guna Land Grant, Luzon, Philippines. Journal of the Yamashina Institute for Ornithology 27: 12–29.

INGLE NR. 1993. Vertical Flight Stratification of Bats in a Philippine Rainforest. Asia Life Science 2: 215–222.

ISACCH JP, MACEIRA NO, BO MS, DEMARIA MR, PELUC S. 2005. Bird-habitat relationship in semi-arid natural grasslands and exotic pastures in the west pampas of Argentina. Journal of Arid Environments 62: 267–283.

[IUCN] International Union for the Conservation of Nature. 2020. The IUCN Red List of Threatened Spe-cies. Version 2020-2. Available from http://wwww.iucnredlist.org

KENNEDY RS. 2000. A guide to the birds of the Philip-pines. Oxford: Oxford University Press.

LUGO AE, CASTRO LM, VALE A, LÓPEZ T, PRIETO EH, MARTINÓ AG, PUENTE ROLÓN ARP, TOS-SAS AG, MCFARLANE DA, MILLER T, RODRÍ-GUEZ A, LUNDBERG J, THOMLINSON J, COLON J, SCHELLEKENS JH, RAMOS O, HELMER E. 2001. Puerto Rican Karst—A Vital Resource: U.S. Department of Agriculture, Forest Service, General Technical Report WO-65. 100p.

MACARTHUR R, MACARTHUR A. 1974. On the Use of Mist Nets for Population Studies of Birds. Proceed-ings of the National Academy of Sciences of the United States of America 71(8): 3230–3233.

MOEGENBURG SM, LEVEY DJ. 2003. Do frugivores respond to fruit harvest? An experimental study of short-term responses. Ecology 84(10): 2600–2612.

MOHD-AZLAN J, FANG VAM, KAICHEEN SS, LOK L, LAWES MJ. 2019. The diversity of understorey birds in forest fragments and oil palm plantation, Sarawak, Borneo. Journal of Oil Palm Research 31: 437–447.

MULLU D. 2016. A Review on the Effect of Habitat Fragmentation on Ecosystem. Journal of Natural Sci-ences Research 6(15): 1–15.

MUNRO NT, FISCHER J, BARRETT G, WOOD J, LEAVESLEY A, LINDENMAYER DB. 2011. Bird’s response to revegetation of different structure and floristics – are “Restoration Plantings” restoring bird communities? Restoration Ecology 19: 223–235.

OKSANEN J, BLANCHET FG, KINDT R, LEGENDRE P, MINCHIN PR, O’HARA RB, SIMPSON GL, SOLYMOS P, STEVENS MHH, WAGNER H. 2020. Package vegan: Community Ecology Package. Re-trieved from https://cran.r-project.org/web/packages/vegan/vegan.pdf



413

ONG PS, DUYA MRM, LEDESMA M. 2016. Biodi-versity baseline assessment of Holcim Quarry Sites: Central Luzon, Philippines. In: Proceedings of the International Conference of the 5th VNU-HCM In-ternational Conference on Environment and Natural Resources (2016 ICENR) and the 11th International Long Term Ecological Research-East Asia-Pacific Regional Network Conference (2016 ILTER EAP), Vietnam National University, Ho Chi Minh City, Vietnam. p. 114–122.

PAGUNTALAN LM, JAKOSALEM PG. 2008. Signifi-cant records of birds in forests on Cebu island, central Philippines. FORKTAIL 24: 48–56.

PEARSON DL. 1971. Vertical stratification of birds in a tropical dry forest. Condor 73: 46–55.

POSA MRC, SODHI NS. 2006. Effects of anthropogenic land use on forest birds and butterflies in Subic Bay, Philippines. Biological Conservation 129(2): 256–270.

RALPH CJ, DUNN EH, PEACH WJ, HANDEL CM. 2004. Recommendations for the use of mist nets for inventory and monitoring of bird populations. Stud Avian Biol 29: 187–196.

RAMASWAMI G, SOMNATH P, QUADER S. 2017. Plant-disperser mutualisms in a semi-arid habitat invaded by Lantana camara L. Plant Ecology 218: 935–946.

RASALAN JB, BARRION-DUPO AL, BICALDO PR, SOTTO MP. 2015. Spider Assemblages of Puting Bato Cave 3–4 and Surrounding Karst Forest Environs, with Additional Notes on the Cave-dwelling Nature of Phlogiellus kwebaburdeos. Museum Publications in Natural History 4.

R CORE TEAM. 2015. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Retrieved from https://www.R-project.org

RESTIFICAR SDF, DAY MJ, URICH PB. 2006. Protec-tion of karst in the Philippines. Acta Carsologica 35: 121–130.

SEKERCIOGLU CH. 2012. Bird functional diversity and ecosystem services in tropical forests, agroforests and agricultural areas. Journal of Ornithology 153(1): 153–161.

SEKERCIOGLU CH, EHRLICH PR, DAILY GC, AY-GEN D, GOEHRING D, SANDI RF. 2002. Disap-pearance of insectivorous birds from tropical forest fragments. Ecology 99: 263–267.

TANG JW, LÜ XT, YIN JX, QI JF. 2011. Diversity, composition, and physical structure of tropical forest over limestone in Xishuangbanna, South-West China. Journal of Tropical Forest Science 23: 425–433.

THOMPSON SA, THOMPSON GG. 2004. Adequacy of rehabilitation monitoring practices in the Western Australian mining industry. Ecological Management and Restoration 5: 30–33.

THORBURN CC. 2015. The edible nest swiftlet industry in Southeast Asia: Capitalism meets commensalism. Human Ecology 43(1): 179–184.

TOLENTINO PJS, NAVIDAD JRL, ANGELES MD, FERNANDEZ AP, VILLANUEVA ELC, OBENA RDR, BUOT JR IE. 2020. Review: Biodiversity of forests over limestone in Southeast Asia with emphasis on the Philippines. Biodiversitas 21: 1597–1613.

TUYET D. 2010. Characteristics of Karst Ecosystems of Vietnam and Their Vulnerability to Human Impact. Acta Geologica Sinica 75: 325–329.

[WBCP] Wild Bird Club of the Philippines. 2019. Check-list of the Birds of the Philippines 2019. Retrieved on 15 Jul 2020 from www.birdwatch.ph

WOXVOLD IA, DUCKWORTH JW, TIMMINS RJ. 2009. An unusual new bulbul (Passeriformes: Pycnon-otidae) from the limestone karst of Lao PDR. Forktail 25(25): 1–12.

WUNDERLE JM, WILLIG MR, HENRIQUES LMP. 2005. Avian distribution in treefall gaps and understo-rey of terra firme forest in the lowland Amazon. Ibis 147: 109–129.


414

Bird Assemblage and Diversity along Different Habitat ...

Documents

Transcript of Bird Assemblage and Diversity along Different Habitat ...